Leet ON ms, AS Aaa EOS CORN Sateen oo 0 ee ee Tithe ir » A > ay ak ‘ ae Crk flats ah pees atti ST paca erase teh Ree Reece pies eee Ath peg 4 Cee ne Rac Sal as RU Ror NET eo a Cinetcsatas ae Sat Riarneerea siete : S vy yl Yi Gyiniit) CNA Uta awe ry Peas ay het larwlea th | | ie f Aenea 4 i tt Hake path cra Ho (iG GAA Ati in i aes ri Mii AP oan, nin A YEAR OF COSTA RICAN NATURAL HISTORY THE MACMILLAN COMPANY NEW YORK « BOSTON - CHICAGO - DALLAS ATLANTA + SAN FRANCISCO MACMILLAN & CO., Liarep LONDON - BOMBAY + CALCUTTA MELBOURNE THE MACMILLAN CO. OF CANADA, Li. TORONTO A YEAR OF COSTA RICAN NATURAL HISTORY yt! ; BY ce AMELIA SMITH CALVERT SOMETIME FELLOW IN BIOLOGY, BRYN MAWR COLLEGE AND , . PHILIP POWELL CALVERT PROFESSOR OF ZOOLOGY, UNIVERSITY OF PENNSYLVANIA EDITOR OF ENTOMOLOGICAL NEWS WITH MAPS AND ILLUSTRATIONS Nem York THE MACMILLAN COMPANY 1917 All rights reserved EV, h.3e1714 CopyRIGHT, 1917 By THE MACMILLAN COMPANY Set up and printed. Published February, ror7 PREFACE THE completion of the Panama Canal and its use by the ships of all the world will have a profound effect on the countries near the Isthmus. Changes in the external rela- tions, industries and occupations of the neighboring peoples are to be expected; these may even be accompanied by changes in the face of nature herself. Costa Rica, lying immediately north of Panama, with her high mountains, her rushing rivers, her great variety of climate and of natural products, will share in these transformations. The natural- ist may regret some of them but he may at least leave for the future a picture of what the past contained. To con- tribute to such a record, and at the same time to introduce some of the features of tropical life to immediate visitors to Costa Rica, are the aims of this book. This little re- public is so readily accessible, it is so easy for foreigners to travel there and it offers such wonderful inducements to naturalists and entomologists (for many of whom. the time and expense involved in visiting most portions of the Ameri- can tropics are absolutely prohibitive) that it certainly should be much better known than it is at present. The information here brought together would have been wel- come and time-saving to us, had it been at hand when we first entered the country, and while our descriptions re- late primarily to Costa Rica, we believe that they apply also to conditions existing throughout much of tropical America. Our primary concern in Costa Rica was a study of the dragonflies with reference to their seasonal distribution, which necessitated visits to the same localities at different vi PREFACE times throughout the year, and our movements within the country were mainly for the purpose of carrying out that plan. Our investigations have not yet been completed and we have little to say in these pages on that technical subject. What we here set forth are chiefly our more incidental ob- servations recorded in our diary. To make this record of the greatest possible value we have sought the aid of specialists for the identification of our specimens and are indebted to Senor Adolfo Tonduz, botanist of the Museo Nacional, San José, Costa Rica (while we were still in that country), and Prof. Henry Pit- tier (after leaving it) for the names of our plants. M. P. de Beauchamp has identified our planarians, Dr. H. A. Pilsbry the molluscs, Prof. J. P. Moore the annelids, Dr. R. V. Chamberlin the myriopods, Mr. H. W. Fowler the crustacea and lower vertebrates, Mr. N. Banks the arachnids and neuropteroids (exclusive of odonata), Mr. J. A. G. Rehn the orthoptera, Mr. G. C. Champion many of the coleoptera: and heteroptera, Prof. W. M. Wheeler the ants, Prof. T. D. A. Cockerell the bees, Mr. W. J. Fox the fossorial hymenop- tera, Dr. Henry Skinner the lepidoptera, Dr. W. T. M. Forbes the caterpillars, Messrs..E. 'T. Cresson, Jr., J. R. Malloch and C. P. Alexander the diptera, Messrs. F. Knab, E. A. Schwarz, O. Heidemann and Dr. H. G. Dyar.certain insects from bromeliads, and Prof. A. P. Brown the rock specimens. Profs. J. F. Tristan and A. Alfaro and Mr. C. H. Lankester, with whom we made various excursions in Costa Rica, aided us in identifying many animals, Mr. Lankester especially the birds. We ourselves are responsi- ble for the determination of the odonata and some of the larger coleoptera and hemiptera. Such success as may have attended our studies in Costa Rica was largely due to the many kind friends resident in the country: Prof. J. F. Tristan and members of his family, PREFACE vil Mr. and Mrs. J. B. Clark, Prof. A. Alfaro, Sefior J. C. Zele- don, Mr. and Mrs. C. H. Lankester, Sefior and Senorita Bonnefil, Mr. John M. Keith and the United Fruit Company, all of whom we again thank heartily for their assistance. Our appreciation of the kindly and graceful acts of the liberal and enlightened Government of Costa Rica is ex- pressed in the first and twentieth chapters, but we wish to emphasize it in this place also. A. &. C. P. P. C. TABLE OF CONTENTS List or ILLUSTRATIONS. . CuarTer I. Puerto Lindy. . sisi Arrival at Bini Plea Streets and “nuildioge=Vulburee—Hanana trade, handling of fruit—Railways out of Limén—Jamaican negroes— Hospital—Coral reefs—Crabs—Swampy shores—Rainfall. CuarTer II. Costa Rican TopocraPuy AND RaILROADS. . ‘ Mountain groups, chief Cordilleras—River gy steens Ristntell af At- lantic and Pacific Slopes—Costa Rica and Northern Railways, sta- tions and elevations—Railroad equipment—Routes into the interior— Banana country—Pineapple station—Outlook on Reventazon valley —Over the Continental Divide—San José and Alajuela—The Pacific Railroad, stations and elevations—Guanacaste trees—Rio Grande de Tarcoles—An odd meal station—Slides at Las Lapas—Collecting from the train. Cuapter III. San José.. ‘ San José in 1840—In 1910—Plan of city: saviaeertaden Panis baila ings in different quarters—Method of selling postage stamps—Schools ~—National Museum—Instituto Fisico-Geogrifico—Girls’ College— Liceo—Brade’s garden. Cuarter IV. Tue Town or Cartaco ann Its Lire...........-.--- Cartago our Headquarters—Situation and Plan of town—Houses, streets, water-supply—Transportation, ox-carts—Hotel de Cartago— Food, vegetables, fruits, milk, butter, cheese—Market—Tortillas and piedras—Plaza—Central American Court of Justice, “‘Peace Palace” —Districts of Cartago—Churches, San Francisco, a Turno General— Nuestra Sefiora de los Angeles and its shrine—Cemetery—Religious festivals of Holy Week, Corpus Christi, Christmas—Secular celebra- tions—Presidential elections of 1909. CuapTer V. Natura History Work 1n CaRTAGO.. : . Rearing of Odonata (Dragonflies)—Food of the iene d in anivte= Blood-worms—Cartago dragonflies of special interest—Moths and beetles attracted to the electric lights—Squirrels, armadillos—The common town sparrow, familiar birds—Vultures, the black vulture bathing—The lanes in the southern border of Cartago—Color changes in a tree-frog—‘Shady Lane” and its changes throughout the year— PAGE 1-7 8-23 24-31 32-53 54-89 x TABLE OF CONTENTS Butterflies with brushes of hair at the tail and on the wings—Clear- winged butterflies—A brilliant hover fly—Animals of the “Bloody Ditch”—Blackfly (Simulium) bites—Collecting microscopic insects— Common plants and weeds—Living fences—Flowers in the plaza. Cuapter VI. West anp Norts or CarTaco.. Continental divide, El Alto—Laguna de Gchosiss Thee table farmer—On La Carpintera—Tres Rios—Scissor-tail flycatchers—The Reventado River—Clear-winged butterflies in shade and sun—Mim- icking butterflies—Butterfly and milkweed—Lanes on Irazi—Straw- berry farm at San Rafael—Tierra Blanca, a typical house and meals— Peaches—Roadside flowers, moths and humming-birds—Laguna del Reventado—Sabanilla, Indian relics in graves—Cot and its waterfall. Cuarrer VII. Tue Vorcano Irazé.. ae : ase Contour and southern dlepes—Hraption of ayaa —Hiscarteat in rainy season —Jiménez lecheria—Sunrise—Oak forest—Cinder cone—“ Islands of vegetation”—Dead trees—Conspicuous plants on cinders—Craters, flowers blooming in them—Forest in rain—Ascent in dry season— Camp in crater—Summit of Irazi—Circuit of crater—Plan of crater, daughter and granddaughter craters—Elevations—Flowers blooming in dry season—Indian rattle. Cuarter VIII. Lower Nercupors or Cartaco.. ‘ Concepcion de Cartago—San Isidro del Tejar, “becliive road” —Tobost —Butterfly migration—Native clay-pits—Ticks—“Bad cattle”— Agua Caliente—Jocosal—Rio Navarro—Dulce Nombre—Orosi— Beautiful bugs—Paraiso—Cachi—Pefia Blanca—Rio Oro—Banks of the Reventazon—Orosi falls. Cuarter IX. Juan ViNas—Arounp THE Lacuna.. ; Juan Viiias a favorable locality for the natucalice—Different altitudes easily accessible—Our cabin and its immediate surroundings—Lumi- nous insects—The laguna and its flora—A lichen-like katydid— Solitary wasp and katydid—Yellow-tailed orioles (oropéndolas) and their nests—Birds around the laguna—Cecropia trees—“ Eye of the ox”—December wild flowers—Leaf-cutting ants. Cuarrer X. Juan ViNas—ABove THE Lacuna... Ascending the road to Juan Viiias village—A snide read, its ts butteriiies and blossoms—A stretched-out Ascalaphid—Place-fascination for a dragonfly—Hellgrammites’ eggs—A grand mountain view—The Messrs. Cochenour’s beneficio—Sugar-cane plantation—Messrs. Schaus and Barnes’ work on Costa Rican Lepidoptera—Village of Juan Vifias—E] Sitio. PAGE 90-115 » II6-141 . 142-168 - 169-183 - 184-190 TABLE OF CONTENTS Xl PAGE Cuapter XI. Juan Vinas—Tue WatTeERFALLS. . ites veweces IQI-208 Waterfalls along the railroad, nearer and facthes—-Violent discharge of pollen—Waterfall-inhabiting dragonflies and their larve—Thaumat- oneura and its male dimorphism—Insects as distributors of diatoms— An insect like a winged seed—Crickets and other Orthoptera by the waterfalls—Spiders and fresh-water crabs—The peculiar larve of the dragonfly Cora—Quebrada Honda and pirates of the seventeenth cen- tury—A lizard hunting insects—Juan Vinas in February—Landslide at the nearer waterfall. Cuapter XII. Juan ViNas—Tue Reventaz6n VALLEY BELOW THE Lacuna....... + 209-229 Rio Narento= Government “Road—Old éeaile—Tndian hit=adian family, maize—Slides—Climbing the cafion side—Forest in river bottom—Seasonal changes in foliage—Begonias—Conspicuous flowers —Kites—A vain lizard—A handsome snail—Insect seasons—Morphos and Owl butterflies—Boulder-haunting moths—Odd caterpillars— Noticeable beetles—Wasp and spider—Wasps’ nests—Katydids and walking-sticks—Shade-loving dragonflies—Very long dragonflies. : Cuaprer XIII. Juan Vinas—Tue TENANTS OF BROMELIADS......... 230-243 Bromeliads and their distribution—Animals in bromeliads: Big-jawed ants, Dragonfly larve, Pseudoscorpions—Beetles—Earthworm— Scorpion, Snails—Earwigs—Planarians—Rearing and transformation of the dragonfly Mecistogaster modestus—Possible origin of its brome- liadicolous habit and long body—Fauna of one bromeliad—Elements of bromeliad fauna, their inter-relations. CuapTer XIV. TuRRIALBA AND PERALTA........ . 244-267 The town and volcano of Turdalba—Paccero on che banks of he Rio Turrialba, earthworms in bromeliads, Peripatus, jicara, manzana rosa, constricted trees, cacao, basilisk—Rio Azul—Snakes, yuca, a reason for collecting butterflies—Potreros of the Rio Colorado, parrakeets, loro, a shade-loving dragonfly—A dense forest, ants’ nests on low grass, an ethereal butterfly. The railroad station at Peralta, a gay-legged bug, a dragonfly in the rain, Costa Rican toucans, Hetaerina dragonflies—The Laguna, shade- loving dragonflies and butterfly, the mistake of a walking-stick, horse- flies and paint, a striking dragonfly, disintegrating rock, heliconias and mosquitoes breeding in them—The forest, monkeys, Megaloprepus and armadillo—The Rio Reventazén, floods, cable-car, basilisks, bobos —Rio Chiriqui, a migrating moth, lianas, Megaloprepus again, copper and blue dragonflies, Costa Rican deer—Toucans and parrakeets con- trasted—Good Friday on the railroad—Blood-sucking flies. xii TABLE OF CONTENTS PAGE Cuapter XV. Santa Ciara.. ae . 268-297 Plains of Santa Clare —Joudaey't to Gudeiles; la Janta, erapendolae— Salvador—Collecting in the forest—The largest living dragonfly—A bold butterfly—Forest dragonflies, relations of intermingled species— Brilliant moths—Bobo—Florida road—Unexpected dragonfly— Beetles at light—Brightly colored flowers—La Emilia—Bees in orchid baskets—Snakes in bromeliads—Guacimo—Banana_ cutting—Dis- tribution of dragonflies in banana fields—Procession of “white ants” —Butterfly swarms—Summary of Santa Clara collections. Cuaptrer XVI. Tue Banana River Country. . veweeees 298-322 Philadelphia South—Cockroach and woo Teaveliitg by * hesclbeg?? — Cieniguita River, Iguana, Egg-laying dragonflies—Dragonflies in flocks—Agua Buena Creek—Foraging ants—Abundance of lizards and frogs—Bearesem East—On the Banana River—Oropéndolas— Westfalia Laguna, water hyacinths—Wary dragonflies—Eye-size and habits of insects—Caimans—Thorny acacia—Rubber trees—Plants and animals common to Banana River and Cartago. CuarTer XVII. Atayueta AND THE VoLCANO Pods......... - 323-359 Heredia—Barba—Alajuela—Beneficio El Beast Costa: Rican Ocabes trees—Preparation of coffee for market and export—Fine discrimina- tion of coffees—Trees near Alajuela, Guayabo, Jocote, Vainilla, Bull’s horn thorn—Hoja de Pato—Clicking butterflies—Dragonflies of shade and of bright light—Social wasps—Orb-weaving spiders—Roads— El Cerro—Extracting sugar—Waterfalls of the Poas and the Caracha —Motmots—Manzana rosa—Feast of the Immaculate Conception— Vegetation in the wet and in the dry seasons—Madera negra—Guavos and Cuajiniquil—Shedding of leaves at different times in the year— Weather predictions, the pintas—Carrizal—Upland pastures—Mon- key’s comb—Eruption of Poas—Visit to the volcano—San Pedro de Poas—Crater of Poas. Cuarrer XVIII. Turrtcares anp THE Rio GranpE DE TARCOLES. ... 360-385 “Diana Club,” the Custodio’s house—Making of tortillas—Meals— Scorpions—Familiar weeds—Bull’s horn thorns—Peripatus—Names for dragonflies—Butterfly swarms—Fossil shells—Guinea corn— Dragonflies with gaily colored eyes—Rio Tizate—Roble tree—Caimito and nance, popular names—Peculiar pods—Flying earwigs—Black boulders—Pord-pord trees—Ant-lions—Quebrada de Salas—Atenas —Yellow cortess—Cebadilla—Marafion—Slight earthquake. CuarterR XIX. Surupres anp OroTina........ ‘ - 386-401 Journey to Bonnefil farm—Farmhouse and surfoutidines-Suneeatniuits and traditions—Guapinol—Cacao—Jicaras and guacals—Path to TABLE OF CONTENTS xili PAGE Surubres River—Animals in Heliconia leaf bases—Otter—Rare drag- onflies, larve in sand—Trapiche—Charral, thorny plants—Wine palm, pejibaye palm—Return ride—Orotina. CuarTer XX. GuanacastE—Puntarenas TO LIBERIA..........-.- 402-442 Commission to examine teachers—Termites indoors at Orotina—Ride to Esparta—Railroad to Puntarenas—Town of Puntarenas—Gulf of Nicoya—Collecting near Puntarenas—Steamboat “Cariari”—As- cending the Rio Tempisque, Crabs’ burrows, Mangroves, Egrets and Crocodiles, Puerto Humo, Mud banks—Estero de Bolson—Puerto Ballena—Ride to Filadelfia, open forest, howling monkeys, white- faced monkeys—Filadelfia, marimba, dances—Journey to Liberia, villages, country houses, crossing the Rio Tempisque, Llano Grande— Town of Liberia, Schoolhouse, Cocoanuts, Quelite trees, White streets —Rio Liberia—Spiders preying on dragonflies—Weeds and flowers— A large grasshopper—Termites’ nests—Papaya and tiste—Potter wasps— Dragonflies in flocks—Quebrada de Panteon—Enormous pods —Quebrada Clara—Jays—“ Vegetable brain” —Examination of teach- ers—Ride to Hacienda Guachipelin, Carretera Nacional, Rio Colorado ~—Hacienda Guachipelin—Rincon de la Vieja—Collecting along the roadside, grasshoppers—Northeast wind and rain—Solfataras of Cerro Guachipelin—Von Seebach’s ascent of Rincon de la Vieja—Venison— Isolation of farm—Return to Liberia. Cuarrer XXI. GuanacastE—Santa Cruz........ » 443-474 Journey to Filadelfia, Llano Grnde—Biledelta, Rio Tempianue—Ride to Santa Cruz, collecting en route—Santa Cruz, Schoolhouse, water supply, large roach, assassin bug—Rio Diria—Dragonfly flocks— Cerro las Pilas—Large thorns—Phalangid swarm—Climbing the Pilas—Abundance of lizards—Bulls’ horn thorn—Costa Rican dance —Padre Velasco—Spearing catfish (barbudos)—Fiestas in Santa Cruz, procession, bull fights, drinking—Pantedn—Low temperature— House of Rollo Jaen—Rare dragonflies in dry forests—Congo mon- keys—Laguna Garzal, warm water—Collecting around the Laguna Garzal—Santa Barbara de Santa Cruz—Excavating Indian graves, artefacts—Drawing water from stream margins—Rio de las Caias— Spoonbills and herons—Snail shells—Maps of Guanacaste—Ride to Bolsén—Pisote—Bols6n, flood mark in house, schoolhouse, library —Steamboat “Castro,” voyage to Puntarenas—Porpoises and croco- diles—Peculiarities of the Province of Guanacaste. Cuaprter XXII. Carrnaco Deteta Est....... - 475-491 Early record of earthquakes—Records of Instituto Hisicd.Geogrifico— Shocks of April 13, 1910—Rumors and fantasies—Damage by shocks xiv TABLE OF CONTENTS PAGE of April 13—Shocks of May 4—Destruction of Town—Departure from Cartago. Apprenpix I. Autuors’ IT1nERARyY, wits Notes on WEATHER IN PLACES OTHER THAN CARTAGO... 2.2... cece cece cece ee cece cece ce se eee es 493-500 Appenpix II. Carraco WeaTHER Recorps, May 8, 1909-May 5, 1910 501-511 Appenp1x III. Papers Basep 1n WHOLE oR IN ParT ON THE COLLEC- TIONS MADE BY THE AUTHORS IN CosTA RICA......... 0000 ee ee eee SI2Z-515 Aprenpix IV. A List or SELECTED LiTERATURE RELATING CHIEFLY TO THE Natura History or Costa Rica, Exc.usive or THaT CITED INVAPPENDIZ ITD sic cgeecses se eats’ eee ecien vere west Gade aneee ees SIOSSAE Systematic List or Pants anp ANIMALS MENTIONED...........-.-+ 543-554 FINDER ojo. dsc ceesig See ae 0h 8 AA Se RNR HE ae apes EROS cece © Jere ncsile sheen sis 08 SS S~SIT LIST OF ILLUSTRATIONS (Except where otherwise noted, the illustrations are from photographs by Amelia S. Calvert. In many of those of plants a 20-centimeter scale is shown.) Some Costa Rican Insects (Direct from specimens by the Phototype Engraving Co. Incolors.) .. . Frontispiece Relief Map of Costa Rica, by J. Rudin-Hefti . . . . facing page 8 Reventazon Valley, East from Juan Vilas. . . . . i. “9 Reventazon Valley, West from Juan Vifas. . . . . # “16 Volcano Irazfi, from a field near Cartago . . ... ee “« 16 Poré Fence, near Cartago . . 1. ww ee ee oe Te: Fruits of the Pord . . . ye a Hoag es samen 67 Cartago from the Tierra nen: Road oS Be es ss «Ba Towers of Nuestra Sefiora delos Angeles . . . . . 7 “32 A Street in Cartago. . . <2 # a 8 wow eG ee ae 333 A Corner of the Park, Cortaee Gl as Oa? Gad Set Ged nel gah de “33 Branch of Higueron. . . 2. 2. 1. 1 1 ee ee ne “40 Flowering Sprays of Targda . . . 1... ww - “40 Echeveria australis 2. 6 1 ee ee ee ee ee “ 4l Chayotes and Sapayo. . . . . 2. 1 we ee ee es «Al A Street Corner in Cartago . . . .... es ee “48 Church of Guadalupe, Cartago. . . . . ...- a «48 Good Friday Procession, Cartago . . . .... “49 Palm Sunday Procession, Cartago. . . e “49 Wings of Dragonfly, Anisagrion (Phot. by H. x Walters) “61 Scarabeid Beetles (Phot. by H. A. Walters) . . . . facing “ 64 Scarabzid Beetles (from life). . . . . 1... s - ie 65 The Bottom of the “Shady Lane”... 1... “s 92 Slaughterhouse and Zopilotes, Cartago . . . .. .- rf “72 Xylosma salamannt, in flower. . .. - 2 2 e+ ee i eS a8 Carboncillo in flower. . . . es “ 73 Anal Brushes of Butterfly, Lycoreaa aieiuatle (Diawine by PEPsGy) 3) 6 eg es Se oak, de 48: sae to we “ 78 xvi LIST OF ILLUSTRATIONS Guitite in flower . Terciopelo de Santa Waa Clavel del Monte, in flower . Lengua de Vaca in flower. Zorillo real, in flower Pavoncillo in flower Churristate in flower. Guijarro, Flower and fruit Mimetic Butterflies (Phot. by Phototype Engraving Co. ) Siphocampylus in flower 3 a % Salvia involucrata in flower F Stone Image from Indian Grave (Phot. by H. A. Walters) Geranium mexicanum in flower . mes 3 Botrychium obliquum Spirea argentea in flower . Castilleja irasuensis in flower Chionolena lavendulacea Gunnera insignis . Oaks on Irazi. Dead Trees on Irazt Craters of Irazi (Photo. by Rudd) Mother Crater of Iraza é Granddaughter Craters of Iraza Sketch Plan of Craters of Iraza (Drawing by P. P. C). Trazti ‘‘Knife-edge”’ ‘ Granddaughter Craters of ae A Natural Garden in the Irazi Crater Senecio erstedianus in flower . Senecio multivenius in flower . Myrtus erstedii in flower . Bidens trazuensis in flower Pernettya coriacea in flower ke Limnanthemum humboldtianum in flower . Schoolhouse in Taras Reventazon River at Orosi Sketch Map of Jocosal (Drawing bys A. s. C. *). Constrictor Trees and vulnhyies Purple-leaved Urera Telanthera and Epiphytic Ficus . Juan Vinas Crater facing page 80 “ce “c “ “cc ce oe “ec oc ee “cc ii9 80 81 81 88 88 89 89 96 112 112 113 120 120 120 120 120 121 128 128 128 128 129 131 136 136 136 136 136 137 137 144 144 145 145 152 160 161 161 168 LIST OF ILLUSTRATIONS Nests of Oropéndolas . Ojo de Buey . Cecropia Trees Trail of Leaf-cutting Anes ‘ Pieces of Leaves gathered by Tea tevtters j Openings of Nest of Leaf-cutters Leaf-cutting Ants (Phot. by H. A. Walters). Army Ants (Phot. by H. A. Walters). ; Larva of Dragonfly, Cora (Phot. by H. A. Walters) é The “Farther Waterfall” at Juan Vinas. ‘ The “Nearer Waterfall” at Juan Vinas . Larve of Dragonfly, Thaumatoneura (Phot. by H. A. Walters) . Adults of osama Tiswmawwans (Phot. bp H. A. Walters) . Pilea. Myriocarpa Grasshopper, Dellia aseatats Croat cut lent be Mead Nat. Sci. of Phila.) . Katydid, Anabropsis costaricensis (iiom anid), Flowers of Jorquito . ‘ Megalopygid caterpillar tiem life). Walking Stick, Bacteria (from life). Euchroma goliath (from life) . Harlequin Beetle (from life). . . Beetle from Bromeliad (Phot. by H. A. Walters). Antenna Cleaner of Same Beetle (Phot. ey H. A. Walters) . B Walking Stick, Ceroys bisthiet tfon life) Caterpillar from Bromeliad (Phot. by H. A. Walters) . Transformation of Dragonfly, Mecistogaster modestus. Bromeliads on a Branch of Pord Bromeliadicolous Ants (Phot. a H. A. Walters) Trees with Epiphytes . : Inflorescence of Wild Ginger . ‘ Inflorescences of Vijagua and of Helicadial The Banana River Country (from a United Fruit Co. Map) . ‘ ‘i Volcano Barba and Coffee Patios : Covering Coffee Xvil » Saring page 169 - facing oe ri3 ce “ after . facing “ce 169 176 176 177 177 184 185 185 192 193 193 200 201 201 203 203 216 217 217 224 225 232 232° 232 232 232 233 233 256 257 257 299 328 328 XVili LIST OF ILLUSTRATIONS Guava Tree and Fruit . Vainilla in flower Leaves of Guanacaste, Jocot aad Guachindin Manzana Rosa Church at Santa Bastian Volcano Pods . , Active Crater of Pods . Eruption of Pods Bull’s Horn Thorn . Ants from Bull’s Horn Thorn (Phot. by H. A. A. Walters) Guanacaste Tree and Pods fo ts Pods of Mimosa asperata . Poré-poré Fruits . Fruits of Anacardium chinowatent? aaa a neat Farmhouse at Surubres. Coyol Palms . Surubres River ‘ ‘ A Fumarole of Cerro @uachiseltn ‘Phot. by a F. Tristan) . . Quebrada de Panteea: Takeda (Phot. by J. F. Tristan). Laguna Garzal (Phot. by J. F. Tristan) . p - Rio Tempisque at Filadelfia (Phot. by J. F. Tristan) School Children at Santa Cruz (Phot. by J. F. Tristan) Spoonbills and Nest (from life, phot. by J. F. Tristan) . “La Giganta” (Phot. by J. F. Tristan) . Bo, Ch Veiled Dance (Phot. by J. F. Tristan). “ House at Santa Barbara de Santa Cruz (Phot. = I. F. Tristan) . At the Water-hole, Rio Soils Bastar (Phot. iy J. F. Tristan) . Plaza and Sehnigliaten, Bold. (Phot, ing J. F. Tristan) House at Bolsén (Phot. by J. F. Tristan) ak Shop “ El Irazi,” Cartago Ame “El Iraza” on Nag 5, 1910 . El Calle Real, Cartago, looking west . El Calle Real on May 5, 1910 Church of San Francisco, Cartago . San Francisco on May 5, 1910 (Phot. by. Pasater Bree: ) Overturned Tower of Carmen Church, Cartago Wrecked Schoolhouse, Cartago . « facing page 329 “c after “ ce “ce facing “ce * 336 336 337 337 352 352 353 368 368 369 376 376 377 392 392 393 393 424 424 425 425 456 456 457 457 464 465 465 480 480 480 480 480 480 481 481 LIST OF ILLUSTRATIONS xix The Central American Court of Justice after May 4, 1910 (Photo. by Paynter Bros.) . . . . . . . facing page 488 The Panteén at Cartago. . . 8 489 Fragment of Pottery from Indian Grave Phot by H. A. Walters). . . ae ae as “489 Map of Costa Rica (after H. Pittier). - +. . 2 « « at back of book A YEAR OF COSTA RICAN NATURAL HISTORY A YEAR OF COSTA RICAN NATURAL HISTORY CHAPTER I PUERTO LIMON Limon, the Atlantic seaport of Costa Rica, is situated at 10° of 30” North Latitude. It was not opened to foreign commerce until 1867 and did not attain much importance until work was begun on the railroad to the interior in the early seventies. Previous to that time travelers entering Costa Rica on the Atlantic side either ascended the San Juan River from Greytown and then its tributary the Sara- piqui for about twenty miles, thence on horseback through the pass of Desengafio, to San José, or they went up the Rio Matina from its mouth, fourteen miles northwest of Limon, and then overland and up the valley of the Reventazon River, with considerable difficulty. We arrived at Limon on the Hamburg-American steamer ‘‘Sarnia” from New York on Saturday, May 1, 1909, and left for the United States by the ‘‘ Prinz Joachim” of the same line, on Monday, May I0, 1910. Between these two dates is comprised our year of Costa Rican Natural History. The “‘Sarnia’s”’ port of call before Limon was Greytown, Nicaragua, where there is no pier. We watched the lighter rising and falling as it came out to our steamer, a mile off shore, and looked at the low-lying coast with interest in- spired by the opening chapter of Thomas Belt’s 4 Naturalist in Nicaragua. Belt landed in Greytown and his fascinating 2 A YEAR OF COSTA RICAN NATURAL HISTORY book was our introduction to many things which we saw in Costa Rica. Six hours later, in the early morning, we reached Limon. The doctor’s inspection was over by nine but the steamer did not dock until noon. The dock room was limited by the breaking of the largest pier a few weeks previously and two steamers were already tied up, obliging us to wait until one of them could move. At the custom house, thanks to the arrangements made for us by the late Senor Calvo, then Costa Rican Minister at Washington, our luggage was not even opened, a courtesy which we greatly appre- ciated. All the luggage—five little trunks, a canvas carry- all, a wooden box of bottles and a large parcel—was handled by one porter or “faquin” (not “cargador” as in Mexico) and all at the same time, on a small freight truck. The pier at which we landed extended some hundreds of feet outward from the shore with the custom house near its sea end. At its land end was the station of the railroad to San José and the interior, and opposite the station stood the building of the United Fruit Company. Immediately north of this last was the principal plaza or public square, planted with many kinds of trees and shrubs. As usual in Costa Rican plazas, the outermost trees were nearly all figs or “‘higuerones,” here of great size and beauty, which formed impressive aisles with their long, rather slender, compressed and subdivided branches. Their foliage is very dark green and produces a deep shadow. ‘The inedible figs had fallen to the ground under some of the trees and were half an inch in diameter, pale yellow in color and stuffed with small pale fruits. One peculiarity was the abundance of aerial roots, as thick as whipcords, dark in color except the pale yellowish tips. These roots hung in profusion from many of the branches, often at heights of twenty or thirty feet above the ground and were sometimes more than a yard PUERTO LIMON 3 long. They were the more easily seen as nearly all the foli- age was at the very tops of the trees, producing a flat crown, with no leaves to conceal these roots. Royal and cocoanut palms were most conspicuous in the center of the plaza, which was the resort of many small birds and lizards. The nu- merous flower beds were outlined with red and pink amaryl- lids and contained great bushes of Coleus and many varieties of crotons as large as lilac bushes, with gay variegated leaves of unusual shapes. , Some blocks south of the railroad station and as many inland from the sea, was another small plaza with a neat market house in the center. On the south side of this square stood the comfortable and clean Grande Hotel, where we stayed when in Limon. Still farther inland were high banks covered with trees showing how much the land had been cut down to make the present level of the streets. The town was clean, well-drained, well-paved, with both streets and sidewalks of good width. In spite of its clean- liness there were hundreds of black vultures or “‘zopilotes” flying about, perching on trees, squabbling over tidbits in the scavengers’ carts, or walking in the streets with their curious rheumatic, jerky gait. Limon when we first saw it was built largely of wood, although its roofs were of gal- vanized iron. The ground floors were used as shops; the upper residence stories were provided with balconies where it was pleasanter to sit. There were few gardens around the houses and the town was rather closely built up. During the following year several fires occurred in different parts of Limon and the areas so bared were extensive. In May, 1910, some of these were already being built on with brick walls instead of wood so that the end result will be beneficial. All of the bananas and much of the coffee produced in Costa Rica are shipped from Limon. They are the chief exports of the country, 9,365,690 bunches of the fruit being 4 A YEAR OF COSTA RICAN NATURAL HISTORY sent out in 1909. On May 9, 1910, from 10 A. M. to 8 P. M. bananas were loaded on the “Prinz Joachim.” A banana train of ordinary box freight cars was run on the covered pier and within a few yards of the ship’s side. Negroes standing within each car handed out the bunches of green fruit singly to other negroes without, each of whom laid his bunch on his left shoulder, which was padded with a burlap bag to prevent bruising the bananas, and carried it to the end of a loading machine. Here another negro took the bunch and placed it carefully on an endless belt by which it was carried into the ship. The belt—the chief part of the loading machine—was about thirty feet long and was operated in an almost horizontal plane by a small portable engine at its side. The entire machine was covered by a canvas awning and a larger awning was stretched over the ship end of the machine and the hatch into which the fruit was passed from the belt by hand. The edges of the hatch were padded with burlap-covered cushions, so that in all much care was taken to avoid bruising the fruit or exposing it to undue heat or moisture. Once we timed the number of bunches passed on to the loading machine in one minute, finding it to be twenty-four, but at other times the number was greater. Railroad tracks left Limon from both its northern and its southern ends. Of those to the south, some follow the coast line to the Banana River, while others soon turn westward to the interior. These railroad beds are the only roads leading out of Limon and of course are not available for wheeled vehicles but only for equestrians or pedestrians. Walking southward on the tracks leading toward Banana River and keeping as near the sea as the wet and muddy ground would permit, P. had his first near view of a pelican at large, when one alighted in the shallow water of the beach 1See the map in Chapter XVI. PUERTO LIMON 5 for a moment before flying off again. We had already seen them from a distance while lying off Greytown and while waiting to dock at Limon, large, low-flying birds whose heads and bodies formed a single horizontal line as they moved over the water. The great majority of the inhabitants of Limon were Eng- lish-speaking negroes many of whom came from Jamaica. Their huts and shanties were much in evidence, chiefly be- tween the tracks and the beach. Some of them bore names as ‘‘Rest Cottage,” “‘Seaman’s Providence”’; the sign of the “Jamaican Hotel” kept by Nicolo Bartoli, on the main street, announced ‘“‘Beds with matrimony 42”. A small river, which the railroad crossed by an iron bridge, empties into the sea just at the southern end of Limon, and a sign by the side of the stream read “Sharks. Dangerous. Bathing Prohibited.” A very wet spot where the river joined the sea was occupied by scores of zopilotes, some of them getting a few pickings from a fish-net spread out to dry. Beyond the bridge the railroad ran among cocoanut palms but was fenced in on each side with barbed wire. Striped lizards (Cnemidophorus or the like) ran over the ties and here and there leaf-cutting ants carried their burdens. We called on Dr. Emilio Echeverria, who was in charge of the hospital located on a little exposed point at the north end of the town. It was the joint property of the United Fruit Company and the Costa Rican Government, had a capacity of 117 beds, and was opened in 1905. It was perfectly screened and Dr. Echeverria told us there had been no yellow fever in Lim6n for three years. He himself had lived here five years and although he did not consider himself immune had never had a chill or any symptom of malarial fever. But as he said, he took no risks and never sat out of doors in the evening. Some of the upper balconies of the hospital commanded fine views of Limén and the mountain barrier 6 A YEAR OF COSTA RICAN NATURAL HISTORY behind the low coastal plain. The buildings were sur- rounded by beautifully kept and luxuriant gardens, from which the Doctor gave A. an armful of flowers. Among them were pink and white oleanders, camellias, a pretty flower new to us which Dr. Echeverria called “Jupiter,” cotton, hibiscus and a euphorbiaceous plant resembling Poinsettia but with some of the bracts wholly red while others were red at the base only, the rest of the bract being bright green. Close to the hospital were the towers of the wireless tele- graph, while west of it were railroad tracks which follow the shore line north and west. The coast here consists of a coral rock which forms cliffs, reaching a height of a hundred feet in some places, and flat reefs which are washed by the tide. Dr. Echeverria thought the reefs were still forming, as a cer- tain reef now visible, although awash, was not so when he first came to the hospital. Professor Pittier mentions the ex- istence of corals in the swamps separating the Blanco and Cuba Rivers in the neighborhood of Moin, five or six kilo- meters from the sea; “‘in the midst of an extraordinary vege- tation rise, here and there, pillars of the same material, naked and branching, like one sees along the reefs of the coast and which appear to have been washed by the waves only yester- day.” ‘These various observations indicate a gradual eleva- tion of the coast in this region. The rock forming the cliffs was in some places compact and solid, in others loose, in the form of irregular fragments one to six inches in diameter em- bedded in a yellow clay. In the tidal pools, on May 8, 1910, were Neritina snails, small sea-urchin shells (I saw no living urchins) and crabs with quadrangular carapaces. On shore also crabs were the most abundant animals, particularly a species with a squar- ish slaty-blue carapace about four inches wide. Their burrows were everywhere in the holes of the rock or in the PUERTO LIMON 7 clay which covered it in many places. There were many small streams, mostly of clear water, coming down to the sea in the three miles traversed on foot this day. One of them—of dirty water, however—about a mile and a quarter from Limon occupied a gorge with rather bare sides. The fall of this stream was considerable. Many of these streams formed pools or swampy spots where they met the railroad embankment and around these the crabs were especially abundant. In some places the railroad occupied a ledge cut into the coralline cliffs, in others, where the cliffs were lower or farther inland, it occupied an embankment which had been reénforced, here and there, by dumping odds and ends of railroad iron on the sea side. The coral cliffs offer a strong contrast with the country south of Limon along the Banana River line, which is swampy and flat. Still farther north of Limon, beyond the coral cliffs, the country is again swampy, as at Swamp Mouth for example, before the Rio Matina is reached. A number of tall deciduous trees remained throughout these three miles, seeming to indicate that formerly a forest ex- tended to the sea. Most of the vegetation was second growth or had been still further cleared for bananas and yuca. There were many cocoanut palms, fan palms, heli- conias, wild gingers, pipers and breadfruit trees. I did not see many butterflies. A big Morpho fluttered by me twice but I could not catch it. Unfavorable weather conditions no doubt diminished the number of insects seen, as it was cloudy after nine o’clock in the morning and rained at fre- quent intervals, although not hard. The entire Atlantic slope of Costa Rica has a heavy precipitation, Limon having an annual rainfall of 127 inches (3224 mm.); the minimal months are February, September and October. CHAPTER II COSTA RICAN TOPOGRAPHY AND RAILROADS From Limén to Puntarenas, the chief Pacific port of Costa Rica, the air-line measures nearly 125 miles (200 kilometers). A railroad now connects them but its windings and its ascent to the continental divide at a height of 5000 feet (1524 meters) increases this distance to 180 miles (289 kilometers). The divide is formed by a mountain chain, the main Cor- dillera, extending from west-northwest to east-southeast, beginning south of the Lake and Isthmus of Nicaragua and continued into Panama. Near Latitude 10° North, the. Cordillera is separated by the river valleys of the Reventa- zon (Atlantic side) and Grande de Tarcoles (Pacific side) into a northern and a southern portion, the first termed Cor- dillera de Guanacaste, the latter Cordillera de Talamanca. A little north of the tenth parallel, the Cordillera de Guana- caste curves to the east, forming three groups of volcanoes . known collectively as the Cordillera Central. These groups, from east to west, are: 1, Turrialba (11,224 feet, 3421 meters) and Irazt (11,325 feet, 3452 meters); 2, Barba (9524 feet, 2903 meters); and 3, Poas, formerly called de los Votos (8786 feet, 2678 meters). Irazii is separated from Barba by the pass of La Palma (5098 feet, 1554 meters), Barba from Poas by that of the Desengafio (6115 feet, 1864 meters). The principal members of the Cordillera de Guanacaste, going northward, are Tenorio (4700 feet, 1432 meters). Miravalles (5675 feet, 1730 meters), Rincon de la Vieja or Cuipilapa [4500 (?) feet, 1371 (?) meters], and Orosi (5155 feet, 1541 meters). These also are volcanic in character and 3 05 oF 08 of OF ° Sarin ° wy ow 08 9 ob oz ° JLAaH-NIGNY TAS W/o ee 28s @) 2) 40 dW 43173U Ny b 25 To face* ° ‘SeUIA UeNf{ Woy ysey ‘AdT[TeA UOZeUdADY sp * To face p. 9 TOPOGRAPHY AND RAILROADS 9 their contours from most points of view, like those of the Cordillera Central with the exception of Turrialba, rise gradually and gently to their summits. The Cordillera de Talamanca is more continuous than the northern Cordillera and is much less known. It begins with the Cerros of Candelaria, or Escazti (7956 feet, 2425 meters) to the south and southwest of San José, and with Carpintera (5705 feet, 1739 meters) to the east. The latter connects with Irazi and forms the watershed between the Rio Reventazon and the Rio Grande de Tarcoles. Com- ponent parts of the Cordillera de Talamanca, proceeding southward, are the Cerro de las Vueltas (10,127 feet, 3087 meters), Buena Vista (11,614 feet, 3540 meters), Chirripo Grande (12,467 feet, 3800 meters), the highest point in Costa Rica, Cruz del Obispo (g100 feet, 2775 meters), Kamuk or Pico Blanco (11,794 feet, 3595 meters) and Cerro Pando, beyond which, in the Republic of Panama, is the Volcano of Chiriqui. The Cordilleras de Guanacaste and Talamanca are nearer to the Pacific than to the Atlantic Ocean, so that the Atlantic slope is much the larger. It is divided by the Cordillera Central into a northern and a southern portion, most of the northern portion draining directly or indirectly into the San Juan River, the outlet of the Lake of Nicaragua, while the rivers of the southern portion and a part of the northern empty directly into the Caribbean Sea. The prevailing winds are the warm easterly trades. They become saturated with moisture as they pass over the Carib- bean and the heavily forested low Atlantic coast. Farther westward, as they meet the high mountains of the interior, their moisture is precipitated so that after they pass these elevations they are much drier. As a consequence the At- lantic slopes usually have a heavy rainfall throughout the year and there is little or no distinction between wet and dry 10 A YEAR OF COSTA RICAN NATURAL HISTORY seasons except in places which lie in “precipitation shadows” so to speak, like Cartago sheltered by Irazi. On the Pacific side, on the contrary, there is often a well-defined division of the year into a dry season (‘‘verano”’) from December to May and a wet season (“‘invierno”) from June to November, the latter due to westerly winds bringing moisture from the Pacific. These topographical and climatic conditions largely de- termine the location and operation of Costa Rican railroads, for with every rain the soil and rock—in many places loose and friable—slide in large quantities down the slopes. The erosion so produced has formed a vast number of valleys, canons and channels of all sizes, whose positions, contours and other characters are constantly changing. The capital of Costa Rica, San José, is on the Pacific slope at an altitude of 3800 feet (1160 meters). The construction of a railroad connecting it with the Atlantic coast conse- quently involved the crossing of the continental divide and the traversing of the moist and ever-changing Atlantic slope. As originally planned, this railroad was to run on the north side of the volcanoes Turrialba and Irazi, crossing approxi- mately at right angles the rivers which descend from these peaks, ascend the pass of La Palma and so reach the Meseta Central, or central plateau, in which San José lies. The tracks were actually laid on this line from Lim6n to Carrillo (1226 feet, 374 meters) at the north side of the pass, between 4872 and 1882, and communication between Carrillo and San José was established by means of a carretera, or road suitable for ox-carts, about twenty miles long. It was im- possible to maintain this line in operation, however. The numerous rivers crossed by the road are subject to so many violent floods that the stream-beds are constantly shifting and railroad bridges were left high and dry over an empty channel while the river flowed in an entirely new bed. At TOPOGRAPHY AND RAILROADS II the time of our visit to Costa Rica trains ran westward over this line—known as “Linea Vieja” or “Old Line” —only as far as Guapiles. The town of Carrillo had disappeared, having had nothing but the railroad to maintain it; there was nothing there but a telephone guard station on the wire running to San José, which followed the line originally sur- veyed for the railroad. A trail connects Guapiles with Car- rillo but was said to be extremely rough and the fording of the rivers, particularly the Toro Amarillo and the Sucio (so called because of its muddy waters), very dangerous. Another railroad route to the capital therefore became nec- essary and this was constructed by an American, Minor C. Keith, between 1884 and 1890. It diverges from the Old Line at La Junta, on the left bank of the Rio Reventazén, 38.6 miles from Limon, and reaches the interior by follow- ing the great cafion of that river, on the south side of Tur- rialba and Irazi. It ascends to El Alto, where between Irazi and La Carpintera, it crosses the continental divide, descends the Pacific slope to San José and continues to Alajuela. The railroads from Lim6n to Carrillo, Alajuela to Cartago and from Esparta to Puntarenas were constructed when General Tomas Guardia was supreme in Costa Rica (1870- 1882), by means of two loans obtained in England. When Mr. Keith undertook the building of the line from La Junta to Cartago, the first two sections just named as built by President Guardia were ceded to him. He too had recourse to foreign capital and the entire road from Lim6n to Ala- juela when completed became the property of the Costa Rica Railway, a British corporation. The United Fruit Company of Boston, organized in 1899, began the building of the Northern Railway running from Limon to the banana districts of Zent and Matina, which was opened for traffic in 1902. In 1905 the Costa Rica Railway Company leased 12 A YEAR OF COSTA RICAN NATURAL HISTORY its railway to the Northern Railway Company and the latter took over the operation and management of the line from July 1 of that year. The Atlantic railroads being of English and American con- struction have their distances estimated in miles, a series of posts often only a quarter of a mile apart alongside the tracks giving the distances from Limon. ‘The distances and grades may be appreciated from this list:— Distance From Lim6n ALTITUDE PrinciPAL STATIONS MILES FEET METERS TANTO Hiss aaa tars eoe aadens Gdaee AWE GO Lotsa eee BE mea Bind Moin Junction............... Bos Wi ok alia dudniesiie aXe. Lent: unos 0 gies acde Sve are DOA ares ts daaeheen we seen MATT Sacatrencoatae diesem’ sist oon DT AO ss eiteci mrs SSiPovee EZs Madre de Dios............... DB Peck cis asativibnacsoiin eles Indiana Junction............. Boni Pelnasekeemas aut tess SIGUITIEB. c acsun end dip ata denmcuen'e BO Fic nscwlnnkceg 196.8.... 60. La: Juntais wena as caged aaa os BOO Auiecianin daneeeen TS87's cue SF On the “Old Line” to Gu4piles, technically the Santa Clara Division, are Tae PACA esas patna enlteceareatoaeanar cee BS Ohiicgeaiteransiste at ne 187s sora 57% Cairo Junction............... BOO... eventos nts ce Baca (Gera RIA a ca nkistcace Gases apes BD Be sis casiiscecer and DestierrOs pices eis gare anse aici AAs. g. oth erica kets GuAcimo Junction............ IS TI aac pecans ses ATT CRSZE cares tds esas bars alae anlar GSS eal a leay ale v8 stent Guapiless: 2 cetnodieadennwsitcds I Silos ivere maneeeeie 984. .... 300. On the “New Line” to San José and Alajuela are dea: JUN tais of oats eutanse sale deans BSH Os icistac tain suns WSs. dacog (6% IGT a as. cone siperoncdice ated nentees BB. ea ere mans daas uae Las Lomas.................. WG Dis ices atetanet ted 879. .... 268. Peralta gvopsen senate ee es BA cBiin Baud sieves TOS S' goes’ G22 Turrialb a... aa cccioven eine ses Os Aisne cs wien nace 2037. .... 621 Tucurrique.................,. OB ake: echere nweey silo Jian Vans oe cavemen gic edna PDOs iuas resi ots 3286. ....1002. Santiago... cciaenedevenrs FOAL so siecl Gi ase. 3536. ....1078. PaTaISOl cdc. (isin dereoeren gender Soc Ay emt tle whu 4392. ....1339. Carta gy ne wsna ee cacseacen oa BOW wake Series 4760. ....1451. TST GO out ig. cee ta aiaeaira Rte ores OB a hea tite 5137. ....1566. TOPOGRAPHY AND RAILROADS 13 DisTAaNcE FRoM LIMON ALTITUDE PRINCIPAL STATIONS MILES FEET METERS Tres: Riosissnuyaved sea iewnve + QOS0: dcleneencd 4363. ....1330. San José... ...... retebebee eet 102. Posi arenes 3868. ....1179. FAC red ta she ercoele ele Gubne win aler TOS) Tiss aet-an cous 37805. saan TE§2, San Joaquin................. TRO Pours Gagacen ahs moe BechéVerrian i xterra claurcwe-agl POvOs oy waveex es ih Alajuclatys naccicaweancaaceniceen LEG Ga oaviateneten S29 T ie. i tna G85): The gauge of all the railroads in Costa Rica, on both the Atlantic and the Pacific sides, is the same, three feet, six inches (1.0668 meters). The passenger coaches, at the time of our visit, were of the American type, and of two classes, differing chiefly in the character of the seats; in the first class these were covered with cane or leather, in the second were simply uncovered wood. Some coaches were divided into a first and a second class section by a partition with a door in the center. On the daily train each way between Limon and San José there was also a “‘chair car” at the end of the train. In this the seats were comfortable wicker arm chairs which the passengers could move about or carry to the large observation platform at the rear of the car, which allowed a fine view of the country for the whole way. There were no return, excursion or commutation tickets on the At- lantic railroads; the rate of fare was about four cents (U. S.) a mile, first class, and anyone failing to purchase a ticket before boarding a train at any place where tickets were sold was fined 50 centimos (= 23-++cents, U. S.) which fine was not refunded. An extra charge was made for the chair car, which made the total fare between Limon and San José nearly six dollars. When we made our first journey from Limon to San José on May 3, 1909, our baggage was divided into two lots; one weighing 142 kilos was checked to Car- tago, the other to San José. After deducting an allowance of 20 kilos free on each of our tickets, we paid 17 colones, 40 14 A YEAR OF COSTA RICAN NATURAL HISTORY centimos (or, as exchange was in Limon just then, about $7.75 U.S.) for the transportation of our baggage. We made this trip without incident, leaving Limon at 10.30 A. M. and arriving in San José about 5.15 P. M., only fifteen minutes late. Trains between Limén and Zent Junction could take one of two routes, either plunging almost immediately into the interior by way of the main line of the Northern Railway via Castro and Zent, or following the sea beach for about eleven miles north of Limoén to Swamp Mouth and then turning westward on the main line of the Costa Rica Rail- way. On the first of these we traveled on May 3. For the first thirty-eight miles or so the road ran through almost un- interrupted banana farms, the names of which were posted here and there, particularly over the gateways to the main farmhouses. ‘These were usually two-storied frame build- ings surrounded by porches. The cabins occupied by the negro workmen clustered around the stations, sometimes only two or three together, at other places forming little villages. Each cabin contained only one or two rooms; the outside was usually whitewashed, the roofs either of palm thatch or corrugated galvanized iron. Many of the farm- houses and cabins were enlivened by surroundings of gay and variously colored crotons, especially the farmhouse of Monte Verde. At Gute Hoffnung (25 miles from Limén) the whole country, forest and open, was flooded in early No- vember; in some places the water was over the tracks, con- cealing them completely and here the train slowed down very perceptibly. The cabins of the negroes were built on short piles, the necessity for which was made evident by this flood for many cabins stood within little lakes. There were almost no roads and the paths leading to the cabins from the railroad bed were also lifted on piles out of the water. TOPOGRAPHY AND RAILROADS 15 Near each cabin grew as a rule a group of breadfruit trees (Artocarpus communis) with their deeply-divided large glossy leaves. In early November the green fruit was nine inches in length and somewhat eggplant-shaped. In late November we bought a breadfruit at Siquirres, for ten cen- timos, and had it baked for us at Cartago. In taste it re- sembles a sweet potato which is not sweet—butter and salt were required, we thought. The fruit is very solid; the rind thin and marked off in little polygons, corresponding to the individual fruits composing it, for the breadfruit is a multiple fruit allied to the mulberry and the fig. We never saw it in the market at Cartago, and the tree is confined to the low warm sections. Its native home is the South Sea Islands; in Costa Rica it is known as “‘arbol de pan,” and the fruit “‘fruta de pan.” Scattered through the banana plantations stood giant trees whose great trunks, without a branch up to a height of fifty feet or more, proclaimed them to be remains of a for- est which formerly covered all this area. Lianas hung from their branches or bromeliads with long narrow green leaves and bright red flowers grew on the trunks even of dead trees. Especially near Siquirres, huge festoons of the gray moss-like bromeliad Tillandsia depended from among the proper foliage. There were also rubber trees (Castilloa sp.) here and there—but not those cultivated under that name in the United States as ornamental plants, which are properly figs (Ficus sp.). The alternative rail route from Limén to Zent Junction, the main line of the Costa Rica Railway, ran in sight of the surf to Swamp Mouth, then bent to the west and entered a forest consisting in some places exclusively of “swamp palms,” in others of the same palms mixed with exogenous trees. These palms have leaves resembling those of ‘the cocoanut palm but rather coarser, and differing in that each 16 A YEAR OF COSTA RICAN NATURAL HISTORY leaf—twenty to thirty feet long—apparently springs from, or very near, the ground, instead of from the top of a dis- tinct trunk. The number of great leaves is such as to hide the trunk within. This forest extended to a point between Waterloo and Estrada stations, sixteen and twenty miles respectively from Limon. Swamp palms also occurred along the railroad leading from Limon southward to the Banana River. The banana planters say, ‘‘Wherever swamp palms are growing, it’s no use to plant bananas there.” Siquirres, the starting point and terminus of trains on the Santa Clara division, was an important shifting station for the railroad and, in the early months of our year, was head- quarters for engineers and others engaged in the replace- ment of the bridge over the Reventazon River, but is not an attractive place. In May and June, 1909, the railroad crossed the Pacuare and Reventazon Rivers on temporary wooden bridges. We saw the ironwork of former bridges, swept away by floods in the preceding December and Jan- uary, lying here and there in the stream-beds, but before November new steel bridges had been put in place. At Las Lomas the banks of the Reventazon, along which the railway is cut, are particularly soft and wash much with every heavy rain, both above and below the tracks. These “slides” are a constant source of expense due to the fre- quent repairs they entail and unfortunately they are by no means limited to the vicinity of Las Lomas. Every now and then they cause delays in the running of trains lasting from a few hours to several days. Along the Reventazén from La Junta nearly to Turrialba the railroad lies in the river valley and the tracks are not very far above the water level. They wind with the wind- ings of the rushing stream, which with the high trees and the fern-grown banks contribute to make this part of the ride The Reventazon Valley, West from Juan Vifias. The Volcano Irazi, from a field near Cartago. To face p. 16 *"WOLPUdPO]]VL09 DULLYINAT ‘o10g aq jo sq}inif *(wosso]q UT anq ssaqeey) osvqied Ieou 99U0 7 o10g Vv 7 To face p. 1 TOPOGRAPHY AND RAILROADS 17 very beautiful. On the rocks in the river black cormorants (Phalacrocorax vigua vigua) stood upright watching for prey. A short distance northeast of Peralta the train passed through a tunnel and crossed the Chiriqui River where it empties into the boiling, surging Reventazén. As Tur- rialba station is approached the railroad is higher above the river and farther from it, and much of the land was cleared, grass-covered and used for grazing. The “Banana Country” has been left far behind and with it most of the negroes, for few were seen beyond Turrialba. From now on there are more and more extensive views of the country, with hill and dale, mountain and valley and often magnifi- cent distances. Turrialba was looked forward to eagerly as the “pineapple station,” just as Irapuato on the Mexican Central was known as the “strawberry station.” ‘The negroes, chiefly women, brought to the passenger coaches both whole pineapples and also slices thereof. The whole fruit cost 30 centavos, a single slice 5 centavos. ‘They were thoroughly ripe, sweet and juicy and exceedingly refreshing, for the train had just risen from the hot lowlands. Everyone thrust his head from the car windows to devour the luscious dripping fruit, presenting rather an amusing spectacle as one glanced along the train. Near Juan Vinas station, which in our year was the break- fast (almuerzo) stop at 11 A. M. for the daily train down from San José to Limon, the tracks in several places are on a shelf cut into the side wall of the cafion of the Reventazon and so close to the edge of the shelf that one can look from the car window down on the river 800 feet below. Here are some of the most glorious views of the whole ride, for here are spread out the whole breadth and depth of the great ‘Reventazon valley, stretching east and west for many miles. Between Juan Vinas and Santiago the barranca or 18 A YEAR OF COSTA RICAN NATURAL HISTORY valley of the Rio Birris is crossed on a high bridge. The railroad now diverges widely from the river. Cartago, the capital of the country until 1823 and until May, 1910, the second largest town, lies in the midst of cattle pastures, while to the north rises the huge and gently sloping mass of Irazi. El Alto, at the continental divide, was not a settle- ment; the highest point on the line was marked by a stone cross set up by the side of the wagon road which is parallel and close to the tracks. South of this road is a stretch of swamp and open water, the Laguna de Ochomogo. The railroad now descends the Pacific slope and the dif- ference in the amount of moisture received during the year is almost immediately indicated in the greatly diminished number of epiphytic plants on the trunks and branches of trees, as contrasted with the conditions visible on the Atlantic side. Many more coffee plantations (cafetales) are now visible from the car windows than earlier in the journey and between Tres Rios and the capital they border the tracks on both sides. This train ran no further than San José, but from the same station other trains ran westward to Alajuela. The train for the Pacific coast left from another station in the lower, southern, part of San José, although the Atlantic and Pacific stations were connected by rail. The Pacific Railroad (Ferrocarril al Pacifico) is State- owned and was constructed between September, 1897, and November, 1910. It now connects San José with Puntarenas, but from 1902 to the end of Igro it was in operation from the capital only as far west as Santo Domingo, now known as Orotina. During the whole of our stay in Costa Rica, trains from San José ran only as far as this place. To reach Puntarenas it was then necessary to take horse or mule 14 miles to Esparta, whence another short railroad carried one to the port. The principal stations on the Pacific road are as follows: TOPOGRAPHY AND RAILROADS 1g Distance FRom San José ALTITUDE STATIONS MILES KILOMETERS FEET METERS DA JOSE ikanmenieneeweye Idwedeee cade, 3800..... 1158 Sabananciciceiacsmuedswenee Mivakeue eoaayes Las;PaVasecenieacewicnireiuies: seldhcd Aagiwda 3516..... 1072 San Antonio de Belen.......... «1... eiiat i BIT... 9st OO de Apt as ils sacmntaminels gavgsie: eaveieeees 2850.... 869 Nu@strG:AMGsss itemangagwaiin: ova BeBEG geead Ciruelasy cuca yeee awe es bias mines, saeco tah 2794..... 852 BUPEUCATESs 5 4.) uct haannieen 1Or8ie F632) wus 2280.0544 680 Cebadilla..............0..0... Wawa olmeear ~ dees ATEN AG: iewieawi nelle cacemabba: -@acetunbe: anmede’ La Balsa. . a ied Beales Gnas, » saarbunaatass TET 2 ce ccs 461 Esobalve a o.vaewiean seven werenecien es nike, = eas 1348..... Atl GONCEPCIBtecicacgseeswweradas weaves aise gs 1040..... 317 Orotina (Santo Domingo). ...46. .. .. 74... Ce 260 Coyolar ic: sectnce tnt. Snide “Webaa bee: sy Puntarenas................... O8> con yen OO; wemences TOndvxs 3 On leaving San José the road passes through rolling coun- try lying between the volcanoes of Barba and Pods on the north and the cerros de Candelaria on the south. Some distance to the north of the railroad are the roofs of the towns of Heredia and Alajuela. The chief crops in cultiva- tion at the time of our visit were coffee, corn (maize) and rice. One of the most characteristic trees seen from this railroad, below 3000 feet, is the ‘““Guanacaste” (Enterolo- bium cyclocarpum). ‘They often stand alone and are very handsome trees with trunks three to four feet in diameter and an immense spread of branches. The head is as large as our spreading oaks but the spray is more graceful and lacks the stiff ruggedness of the oak. The guanacaste is a Mimosa with a much divided leaf as fine and delicate as a maidenhair fern. Beyond Pavas the line crosses the deep valleys of the Rios Torres, Virilla and Bermudez. San Antonio de Belen has an historical interest for the naturalist for in 1847 it was. 20 A YEAR OF COSTA RICAN NATURAL HISTORY the headquarters of the Danish botanist, Anders Sandoe Oersted, to whom some of the earliest scientific exploration of Costa Rica is due. At Ciruelas a branch road runs off to Alajuela. Beyond Cebadilla Station the Rio Grande de Tarcoles is crossed by the longest bridge in Costa Rica, a cantilever 452 feet in length and 300 feet above the water. Tablets at the end of the bridge state that it was designed and built by Milliken Bros. of New York. Alongside the single track is a walk about three feet wide for pedestrians. The bridge commands some fine views of Barba and Pois to the northeast, the Aguacate Mountains to the north and northwest and the canon of the river. To the southwest is a deep notch in the hills marking where the cafion opens to the Pacific but the ocean itself is not visible. Atenas was the meal station and as we saw it in October, 1909, was an odd place. Under an open shed stood tables on which all the comestibles were spread out to view. There were no plates or forks. Everything was served on “‘tor- tillas” or maize cakes;—broiled chicken joints, stewed beef, sausages, “‘picadillo,” cheese, omelettes—each portion was laid on a tortilla and the customer picked up what he wanted and finished the meal by eating his “plate.” Most of the things were excellent. For drinks there were milk, and coffee “sin 6 con leche”? but cooked with dulce or brewn sugar, in the Costa Rican style. The town of Atenas is five miles (eight kilometers) distant from the station. Beyond Atenas Station the train ran around the southern base of the bare and steep Aguacate Mountains, wherein are the prin- cipal gold mines of Costa Rica. In the wet season the track of the Pacific railroad was often in very bad condition and trains were frequently de- layed for hours owing to landslides. The worst place was at Las Lapas, where the track ran on a shelf cut in a soft de- composed rock which slipped and slid and carried the whole TOPOGRAPHY AND RAILROADS 21 roadbed with it. When we went down to Orotina on Oc- tober 15, 1909, we were glad to learn that the “bad part of the road was now good”—for this day only perhaps, but sufficient unto the day are the slides thereof. We ap- proached Las Lapas slowly and cautiously. The most dangerous part was a very sharp narrow bend around a small side valley. Here there was a Y-shaped track, our train approaching along one arm of the Y toward its stem. The engine had previously been put on the rear end of the train, and we were now pushed on to the stem of the Y far enough for one car to be over the switch and uncoupled. It was held back by some twenty peons while the engine pulled the rest of the train away over the same arm leaving the one car on the stem. Then the switch was thrown open for the other arm (which was three to four feet lower than the first arm), the car went down by gravity and was stopped by its own handbrakes in the middle of a none-too-safe-look- ing timber bridge. This operation was repeated for each car and the engine, until the train was made up again, each car reversed, on the bridge and the engine pushed us on ahead of itself. The track was extremely uneven and shaky. As we crawled along the cars swayed and lurched from side to side and endwise, the tracks groaned and brakes screamed, but we did somehow stick to the tracks. For half a mile each side of this place the track looked as if it would slip away any minute. There were many marks of recent slides, the most impressive being a basket on a cable, stretched from one arm of the Y to the other arm, to con- vey passengers when the road was really impassable. In January, 1910, the passing from one branch of the Y-track to the other at Las Lapas was accomplished in a much shorter time, for the stem of the Y had been lengthened by cutting into the hillside, so that the entire train could be run onto the stem at once and then off to the other arm. 22 A YEAR OF COSTA RICAN NATURAL HISTORY Beyond Las Lapas we wound about the deep side valley of Rio Concepcién, a northern tributary of the Rio Grande de Tarcoles. The track twists and turns in a most extraor- dinary fashion so that there were many places where one could look across and see another section of it running paral- lel to that on which the train stood, and every turn opened a new and beautiful vista. West of Turricares the grade is very steep; going in that direction the main thing is to hold the train from running too fast down hill, but traveling east is another story. On the return trip in the same month of October, the train pulled out of Orotina at nine o’clock— “‘a la punta” plus some minutes. The distance from Oro- tina to San José is 46 miles, the difference in elevation about 3000 feet and our running time was twelve hours. The day was bright and beautiful until almost four in the after- noon and we had little rain at any time. The scenery was magnificent and the ride should have been a pleasure, but fatigue made it tiresome before we reached San José. In addition to the dangerous condition of the roadbed there was no coal, whereby the locomotive was forced to burn wood instead. The consequence was that it could run only a few miles before the steam was exhausted and it was nec- essary to stand still and wait until the pressure rose high enough. These delays could sometimes be utilized. At Rio Con- cepcién we stopped fully half an hour. A stream of clear water flowed alongside the track from a brook serving as a water supply. There were six or seven species of dragon- flies flying over this water so P. rigged up the net and caught specimens of all but the two most common ones. No one had ever collected these insects here before—and it gave some passengers something to look at while they waited. In such wise we made the trip, running fifteen or twenty minutes, then standing three-quarters of an hour to get up TOPOGRAPHY AND RAILROADS 23 steam again so that it was almost nine o’clock when we reached San José. Yet three days earlier the train went from Orotina to San José in three hours; but on October 22 it was so delayed that it only reached San José at 5 A. M. of the 23d! CHAPTER III SAN JOSE The American traveler Stephens wrote of the capital of Costa Rica as he saw it in February, 1840: “San José is, I believe, the only city that has grown up or even improved since the independence of Central Amer- ica... . The buildings in San José are all republican; there is not one of any grandeur or architectural beauty; and the churches are inferior to many erected by the Spaniards in the smallest villages. Nevertheless it exhibited a development of resources and an appearance of business unusual in this lethargic country; and there was one house in the plaza which showed that the owner had been abroad, and had returned with his mind so liberalized as to adopt the im- provements of other countries, and build differently from the custom of his fathers and the taste of his neighbors. . In San José, by-the-way, all the ladies were what might be called good business-men, kept stores, bought and sold goods, looked out for bargains, and were particularly know- ing in the article of coffee.” During the following seventy years the “development of resources” brought considerable grandeur which has been described in English many times by visitors of very different abilities. Owing to this wealth of description, to the exist- ence of several albums of photographic views (such as that of Zamora) and to the character of the present volume, we shall devote but a very few pages to the capital. On Dec. 31, 1910, San José had a population of 30,854. Its plan was that of a chess-board with perfectly straight SAN JOSE 25 streets, “Calles’”” running north and south, ‘“Avenidas” running east and west. The two main thoroughfares, the Calle Central and the Avenida Central, divided the city into four quarters; the Calles were numbered 1, 2, 3, etc., east or west from the Calle Central, correspondingly the Avenidas I, 2, 3, etc., north or south from the Avenida Central. The streets and many buildings were lighted by electricity. The Calle Central and the Avenida Central contained many of the shops and other important edifices and in 1909-10 were traversed by lines of electric cars with overhead wires. At the same time the ox-cart was still almost exclusively in use for all kinds of hauling, including the carriage of the trav- eler’s trunks to his hotel and formed a striking contrast to the bustling “trolley” as they passed each other. Coming from Cartago to San José one day in February, P. found the street by the railroad station, from the opposite curb to the adjoining freight station, filled with ox-carts taking away or bringing goods. Never had he seen so many. In addition to the electric cars there were many two-horse carriages for hire in the streets, which were much patronized. In the northeastern quarter were the station of the Atlantic railroad, the custom house (Aduana), the National Park with its group of statuary commemorating the defeat of the American filibuster Walker by Costa Ricans in 1857 (an event which assumes large proportions in the national his- tories), the tall iron building occupied in our time by the Escuelas Superiores, and the Parque Morazan. All these lay within three blocks to the north of the Avenida Central. In the outskirts of the northwestern quadrant was the imposing and finely situated penitentiary; in the portion nearer the center of the city the various Government offices and the Post Office (Correo). The method of distributing postage stamps and postal cards was quite a novelty to the traveler from the north. The Government did not issue or 26 A YEAR OF COSTA RICAN NATURAL HISTORY sell stamps, etc., but gave a concession to do this to some agency, at that time the Banco de Costa Rica, formerly to the Banco Anglo-Costaricense. The Bank made a con- tract with engravers in England or in the United States to print certain quantities of stamps of different denominations as approved by the Government. A percentage of the face value of the stamps was paid by the Bank to the Govern- ment in return for the concession and this sum largely paid the expenses of the Post Office Department. The Bank sold the stamps to anyone who wanted them, a discount of 6 per cent being given on sales amounting to 25 colones ($11.62+) or more, so that a person buying 25 colones’ worth of stamps from the Bank paid only 23) colones for them. Many storekeepers throughout the country bought stamps at this discount and retailed them, at face value of course, making a little more than 6 per cent on the transaction. It was not part of the duty of postmasters to sell stamps, although they might undertake it like other shopkeepers if they chose to do so; before we knew this we were astonished at being unable at times to buy stamps at the Cartago Post Office. Periodicals (newspapers, magazines, etc.) could be mailed free to points on railroads in Costa Rica, as part of a policy to extend knowledge and diffuse information. In the southwestern quadrant was the Plaza Central, whose border of fig-trees, represented in many pictures, had been removed previous to our visit, giving the square a barer and more open appearance than is usually to be found in such places. On the east side of this Plaza was the Cathedral, on the north side the principal barracks. This quadrant con- tains also several hotels of which the Imperial was most fre- quented by foreigners, the Banco de Costa Rica and the Gothic church of La Merced with a single spire over the main (west) entrance and another over the junction of nave and transepts. Nearby,were the large hospital of San Juan de SAN Josh 27 Dios and the Escuelas Mauro Fernandez, named for the recent Minister of Education whose portrait appears on the 5 centimo postage stamps of this period. These Escuelas were two-storied with a basement under part of them. The Palacio de Justicia, a one-storied building with a small patio planted in roses and containing courts of “Casacion”’ and “Apelacion” and a law library, and farther west on the south side of the Avenida Central the almshouse and the Asilo Chapui for the insane also belong to this quarter. The west-bound electric cars turned to the south at the west end of the Avenida Central and skirted the southern edge of the Sabana, a common forming part of the land bequeathed to the ‘“‘sons”’ of San José by the prelado presbitero don Manuel Antonio Chapui de Torres in 1777. On various sides of the Sabana and facing it were some handsome private residences, gardens and club buildings; there was also a Sabana station on the Pacific Railroad here. In the southeastern section was the magnificent Teatro Nacional, the chief architectural monument in the country, which was largely constructed of white and yellow Italian marbles and has often been described. Here too was the one-storied Museo Nacional, which, under the administra- tion of Senores don Anastasio Alfaro and don Juan J. Ferraz, has, with limited means, accomplished much for the explora- tion of the country and its natural history. It was founded by the Government May 4, 1887, as ‘“‘a public establishment where are deposited and classified all the natural and artistic products which may serve as a basis for the study of the riches and culture of the country.” At the same time don Anastasio Alfaro was named administrating Secretary, to act under the direction of a Committee, appointed Jan- uary 28, 1888, consisting of Sefiores don José C. Zeledon, don Manuel Carazo Peralta, don Juan Rojas, don Juan Francisco Echeverria, don Enrique Pittier and don Pablo Biolley. 28 4A YEAR OF COSTA RICAN NATURAL HISTORY The Museo contains public collections of the animals, woods, native pottery and prehistoric goldsmith work, maps, arms and armor of Costa Rica, an extensive herbarium, a scientific library and study collections in various groups of animals, One of its earliest acquisitions was the important collection of Costa Rican antiquities formed and bequeathed to the Government by don José Ramon Rojas Troyo of Cartago. Connected with the Museo at different times have been the naturalists Paul Biolley (born in Switzerland, 1862, died at San José, 1908), George K. Cherrie, an American (member of Colonel Roosevelt’s Brazilian expedition of 1913-14), Cecil F. Underwood, an Englishman, and J. Fidel Tristan, a Costa Rican. To Director Alfaro and to the botanist of the Museo, M. Adam Tonduz, we were greatly indebted for assistance in our work. The first volume of the Anales of the Museo was published in 1887, since which time nine volumes have appeared, some of them under the joint title of Anales del Instituto Fisico-geografico y del Museo Nacional. The Instituto Fisico-geografico was established by Pres- idential decree April 7, 1888, under the name of the Insti- tuto Meteorologico Nacional, although meteorological obser- vations were begun by its director, Professor Henri Pittier (de Fabrega), a native of Switzerland, on December 12, 1887. Its title was soon changed to that of FisicosGeo- grafico, indicating its enlarged scope. Professor Pittier served as Director of the Instituto until 1904, developing it to such an extent that Dr. Karl Sapper, who visited it in 1899, wrote of it as being entirely on the plane of European institutions of a similar kind. In addition to the Anales above mentioned, the Instituto published a Boletin in 1901- 03. In later years the Instituto has become the ‘meteoro- logical section of the Museo Nacional whose building it immediately adjoins, and the yearly operations of both es- tablishments have been described in the Memorias de Fomento SAN JOSE 29 presentada al Congreso Constitucional by the Sub-secretario de Estado en el Despacho de Fomento. A few blocks from the Museo was the three-storied Co- legio de Sefioritas, the highest educational institution for girls in the republic. Its director, Professor J. Fidél Tris- tan, educated under German professors in the University of Santiago, Chile, and speaking English fluently, was our chief guide and mentor in our explorations and to him more than to any other one person do we owe much of our success. Education receives great attention in Costa Rica and the sums spent for this purpose form a large part of the national expenditures.1. The report of Don Alfredo Volio, Secre- tary of State in charge of Public Instruction for 1909, gave the number of official primary schools in October, 1908, as 357, of which 82 were for boys, 79 for girls and 196 for both sexes. The 357 schools had 887 teachers and 27,452 pupils (14,210 boys, 13,242 girls). The Colegio de Seforitas had a five-year course and gave instruction in mathematics, geog- raphy, physics, chemistry, natural sciences, history, peda- gogical psychology, the practice of teaching, Spanish, French and English, domestic economy, singing, drawing, manual training, calisthenics, morals and religion. The number of students including those in “‘preparatorio” in- creased from 97 in 1899 to 306 in 1908. The roof of the Colegio (like any other of the higher buildings in San José, indeed) commands magnificent views of the vicinity of the capital. Along the northern horizon from east to west are Irazii, the pass of La Palma leading northward to Carrillo, the lesser cerros of Zurqii, the volcano Barba, the pass of Desengano, the volcano Pods and lastly the Aguacate Mountains; on the east, La Carpintera (5700 feet) separates 1 For the figures for 1909 and 1910, respectively, see the Bulletins of the Pan American Union, vols. XXXI, p. 121, and XXXII], pp. 87-88, Washington, D. C., July, 1910, and July, 1911. 30 A YEAR OF COSTA RICAN NATURAL HISTORY this valley from that of Guarco in which Cartago lies; to the south is the imposing ridge of Candelaria (Escazi1), while to the west a gap indicates the way to the Pacific. ‘The earth- quakes of April and May, 1910, rendered the Colegio build- ing unsafe and instruction was transferred temporarily to the one-storied Presidential residence on the north side of the Avenida Central. On the outskirts of the southeastern quarter, facing each other on opposite sides of a wide street, were the Liceo de Costa Rica and the Escuela Normal. The Liceo was two- storied with an oblong patio, where the ceremony of award- ing certificates took place and behind was a separate gym- nasium of ample dimensions. The Liceo and the Escuela were of exactly the same size and shape externally, which seemed a pity, and both were covered with cement or stucco highly colored to represent stone, which spoilt the effect greatly. The Liceo was the highest educational institution for boys in the country and offered, in 1908, a five-year course, a normal course of two years, a commercial course of two years and a one-year technical course. Including 30 students in the “‘preparatorio,” the total in attendance in 1908 was 241. The former University of San Tomas in San José has been discontinued for many years. South of the Liceo and separated from it, in our time, by several open blocks was the gaily-stuccoed abbatoir, and near it the station of the Pacific Railroad. This was also the southern terminus of the electric cars of the Calle Cen- tral, their other terminus being the village of Guadalupe north of San José. Herr Alfred Brade, a German naturalist and collector, had a flower garden, with a few animals in captivity, which formed one of the sights of San José at this time; these were algo in take southeastern quarter. Among the animals were tigres” of different kinds and some “pavones” (Crax pana- SAN JOSE 31 mensis), handsome gallinaceous birds allied to turkeys and peafowl. The male was black with golden yellow wattles, the female brown and yellow, barred like a pheasant. Both had erectile crests of curled feathers on the head and upper neck. They were very tame and when Herr Brade put his hands in the cages and petted the birds, it was a great sight to watch these crests rise as the birds responded and looked eagerly for tidbits.! 1A female pavon belonging to Herr Brade is figured from a photograph and de- scribed by C. Wercklé in Boletin de Fomento, Aiio IV, No. 1, pp. 62-64, San Jose, Jan., 1914. CHAPTER IV THE TOWN OF CARTAGO AND ITS LIFE During the twelve months that we lived in Costa Rica we made Cartago our headquarters, always retaining our room in the hotel and keeping there our rather large equipment. From Cartago we made excursions to many different parts of the country, remaining away one to two weeks but always bringing back our material to be worked on and stored in Cartago until we were ready to send it to Philadelphia. By this method we avoided many of the dangers and discom- forts of a long sojourn in the tropics. Cartago has a cool and bracing climate—it was one of the places where Panama Canal Zone employees were permitted to spend their vaca- tions—, it was almost free from mosquitoes and malaria was not endemic. Cartago lies on the northern slope of the so- called Valley of Guarco, a slope which ascends northward to become the great mass of the volcano of Irazi and south-. ward descends gradually to the Rio Agua Caliente, a tribu- tary of the Rio Reventazon. On the south side of the Abua Caliente the hills rise much more abruptly. The railroad station of Cartago is 4760 feet above sea-level, the Agua Ca- liente at the bath house, two and one-half miles away, is 4460 feet, while Irazi rises to 11,300 feet. The town is laid out with great regularity, its streets running almost exactly north-south and: east-west; many of the north and south streets are continued northward as lanes and tracks which form a maze over the lower slopes of Irazi running between: the stone walls that enclose the potreros or pastures. To the south of the town some of these streets become wide Cartago, from the Tierra Blanca Road. Towers of Nuestra Sefiora de los Angeles and behind them La Carpintera. To face p. 32 A Street in Cartago, looking toward Irazu. A Corner of the Park, Cartago. To face p. 33 THE TOWN OF CARTAGO AND ITS LIFE 33 cartroads leading to the neighboring villages of Dulce Nombre, Concepcion de Cartago, Agua Caliente, San Isi- dro del Tejar, Tobosi and the more distant villages of Pa- raiso and Orosi. Others were lanes cut off by fences after a distance of half a mile or so, and some of these lanes formed our richest collecting fields about Cartago. The town was much spread out, especially along the main east-west street, the Calle Real, and contained six to seven thousand inhabitants. The streets were unusually wide for a Spanish town and there were scarcely any two-storied buildings because of the frequent earthquakes. The houses were of plastered adobe and were often tinted but not usually of a bright color. Our hotel was pale green, the stores across the street a salmon yellow. The roofs were of curved red tiles, the eaves projecting some two feet beyond the walls, and the sidewalks were so narrow that in many cases the eaves completely sheltered them. Only on a part of the Calle Real, which leads to the Panteon, was there a sidewalk as much as five feet wide. With the years, the red tile roofs become coated with green and gray lichens and in the depres- sions spring up ferns and Echeveria australis, a plant re- sembling houseleeks. It has an inflorescence about a foot high and when we arrived in May the old flower-stalks were standing up bare and brown, in great numbers. The rains were bringing out the new buds but as yet they scarcely showed. The windows were almost invariably casements protected within by heavy wooden shutters, and often the house was so little raised above the ground that the window sill was not more than two feet from the level of the sidewalk. Our own was about four feet. Some of the main streets had excellent macadam paving, with deep rounded gutters of solid concrete at each side. Where such streets intersected the gutters were covered with stone slabs at the roadway level. During the dry season’ 34 A YEAR OF COSTA RICAN NATURAL HISTORY a novel method of sprinkling the dust was employed on these streets. By temporarily and partly damming a gutter where it passed under the stone slabs, considerable water was soon accumulated behind the dam and a troop of small boys, straddling the gutter, scooped up this water with tin basins and dippers and threw it out on the roadway until the dusty surface was well dampened. Then the temporary dams of boards and cloths were removed to the lower end of the next square. The poorer streets displayed various degrees of paving from stone blocks to soil, but in these cases there was a gutter down the middle one to two feet wide, well defined by stones and carrying a stream of water several inches deep. Clear water was constantly rushing down these gutters of the north-south streets from the mountain, and in addition the surplus water from the town reservoir, three-quarters of a mile above the town, was turned into the streets every night carrying off any surface drainage. Water for domestic purposes was not taken from these streams but was piped to the houses. Even far out of the town it was common to see a water-pipe, tap and cement sink outside the house door. Neither did Cartago depend upon surface drainage but had an excellent sewage system and during our stay there was planning the installation of a sewage disposal plant. It was a common thing to see women washing clothes in the stream flowing down the middle of the street, rubbing and soaping them on the large stones forming the edge of the channel and later, following the invariable custom, hanging them up to dry on the nearest barbed wire fence. The barbs were a most convenient form of clothespin, always at hand when needed, and as barbed wire has been used in amazing quantities all over Costa Rica the washerwomen are never at a loss and greatly prefer to hang the wash on the fence rather than to follow their own native custom of dry- ing clothes on the ground. THE TOWN OF CARTAGO AND ITS LIFE 35 Cartago was lighted by arc lamps and most of the houses used electric lights. The town was well policed, clean and orderly. We rarely saw the people carrying loads on their heads or backs. ‘There were some pack animals, but the hauling was mostly done by bullock-carts, as the roads about Cartago and San José are numerous and fairly good. The carts often had solid wheels, each one the section of a tree-trunk, but iron tires. They were often painted bright colors, blue, red or occasionally yellow. The oxen were large, powerful, handsome creatures, for the cattle in this grazing country are of good breed. The yoke rests partly on the horns, to which it is strapped, partly on the head, and there is an odd wooden shield over the neck. The driver walks in front of his team carrying the goad, a long slender staff with an iron point, so that the sharp end rests on the yoke. Occa- sionally the goad is laid across the yoke and the driver walks beside his team. There was at least one horse cart in Cartago and a few carriages but most people rode on horse- back. Away from the line of the railroad, indeed, every- one must either walk, travel in ox-carts or ride, for no vehicle less sturdy than the ox-cart could survive the roads. The “Hotel de Cartago”’ where we lived consisted of two excellent native houses thrown together by doorways and passages, and Mr. Weldon, the American proprietor, had introduced sanitary features such as additional plumbing and windows and in every room had removed one pane of glass for which he substituted fine wire netting. The impor- tance of the latter improvement was not apparent to us until we learned—by transgressing—that the police regu- lations of Cartago did not permit first floor windows to be open or unfastened after midnight. These houses were built around patios in the usual style and in each patio there was a large cement sink with fine cold, clear, running water, 36 A YEAR OF COSTA RICAN NATURAL HISTORY while in the kitchen patio there was even a hot water tap— the rarest of luxuries. The main patio, of which the dining- room formed one side, contained some flower beds, a large “Peruvian pine” or Araucaria and was further adorned with two big macaws, the common red and blue and a some- what rarer brilliant red one, and a tame but chained white- faced monkey. The antics of these three creatures provided endless amusement and they had frequent little differences —really a “monkey and a parrot of a time.” The food in Cartago was extraordinarily good. Fowls and pigs were plentiful and the cattle raised on Irazii fur- nished excellent beef. Sheep are not bred in Costa Rica so that mutton is unknown. The vegetable gardens pro- duced the greatest variety of vegetables throughout the year and we enjoyed fresh peas, string beans, young carrots, young beets, cauliflowers, onions, sweet and white potatoes and green corn nearly every month of the year, while there were only a few months in which we did not have fresh asparagus and strawberries. The latter in particular were of exceptionally large size and delicious flavor. These in- troduced plants only grow to such luxuriousness and per- fection in the higher parts of the country. There were also, of course, the native fruits and vegetables, especially bana- nas, plantains, frijoles, or black beans, the tips of the young fronds of a tree-fern, called “‘rabo de mico”’ (monkey’s tail) because of their curved ends, the “manihot” or ‘“yuca” (Manihot utilissima and palmata) and above all “chayotes.” ‘The chayote (Sechium edule) is a cucurbit with a globular or ovoid fruit having but one large seed. It is a most useful plant. The young leaves are eaten either cooked like spinach or. to enrich soups. The young fruits, an inch or so long, are stewed or baked in milk. When ripe the fruits are six or seven inches long, with a firm prickly skin and are then baked whole, or halved and stuffed before baking, or fried THE TOWN OF CARTAGO AND ITS LIFE 37 in slices. There seemed no end to the possibilities of the chayote in cookery. This plant, under the name of “cho- cho,” is already grown in some parts of the southern United States, particularly in Louisiana. In their seasons, anonas, mangoes, grenadillas (or “‘may-pops’”’) and aguacates or “alligator pears” as the Americans call them, were common, while oranges and pineapples were good, plentiful and cheap almost the whole year. Anonas or “custard apples” are particularly delicious tropical fruits. Under a dull green or brownish skin with indistinct overlapping scales is a rich, juicy, finely flavored pulp of the consistency of a firm cus- tard, in which are embedded a number of large, black, shiny seeds about three-quarters of an inch long, the whole fruit being six to eight inches long, three to four across. The flavor suggests both grape and pineapple although it is quite distinct from both. Good milk was obtained from several dairy farms, notably the Lecheria belonging to Don Ricardo Jimenez, situated on the upper slopes of Irazii at about gooo feet elevation. He not only sold milk but also made excellent butter, chiefly for sale to the foreigners. Most Costa Ricans, certainly all the poorer families, do not make or use butter at all. Any surplus milk over the day’s needs is made at once into a soft white cheese, which is very good and often used on tor- tillas as butter is used by those who are accustomed to the latter article. Cartago was a market town and owed some of its impor- tance to that fact, as it had no manufactures. Coffee and cattle were the chief agricultural productions of the surround- ing country. The market was a large enclosure near the station, with corrugated iron roof; stores and a hotel occu- pied the north and south sides, while the east and west ends were merely fenced in. The great weekly market was held early Sunday morning. Then the streets were full of country 38 A YEAR OF COSTA RICAN NATURAL HISTORY folk and ox-carts, the shops were all open, everyone wore his or her gayest and best, and it was one of our pleasures to wander about the market at that time. There were few stalls or booths except those in which shoes, clothing or imported enamelware were sold. Most of the market people squatted on the cement floor with their wares about them— sacks of oranges or lemons, little heaps of vegetables, and fruits, beautiful flowers, unglazed pottery of local manufac- ture, polished cows’ horns and quite artistic combs and pins made of the horn, tanned hides from which sandals could be cut and equipped with thongs all ready for wear, while the customer waited, sacks of green coffee and great quan- tities of beans, maize and rice. Little wheat is grown in Costa Rica and the imported wheat flour is rather costly. Outside of the towns the tortillas made of ground maize are the staple bread stuff and are very good when hot and freshly made. We found them dry and unpalatable when they were cold. In the towns excellent bread is made of wheat flour and is used except by the poorer people. There were also little charcoal stoves over which tortillas were baked, coffee boiled or black beans or rice cooked for sale in the market, and coops of live fowls at times or perhaps one or two held by a string about one leg. No cattle, pigs or horses were sold in this market as there was a weekly cattle market in another part of town, on Thursdays. At festival times there appeared great numbers of shiny tin ornaments, flowers and candle-holders, destined to adorn the shrines of favorite saints. Tortillas being a staple food, the grinding of the maize into meal is a daily occupation of the poorer Costa Rican women. The native instruments are the “piedra” (stone), a large block of the common lava, the upper surface smoothed and slightly hollowed, and the “mano de piedra,” a sort of stone rolling-pin without handles. The piedra may be much THE TOWN OF CARTAGO AND ITS LIFE 39 more elaborate and those exhumed from old Indian graves usually have four short legs and often have animals’ heads carved on the ends, while the mano de piedra has a handle and looks in side view like a modern chopping knife. Some piedras of this more elaborate pattern are still in use, but in the ordinary kitchen the simpler style is the common one. It was interesting to see how this old native implement was being replaced by a modern imported machine of the coffee mill or meat-grinder type, so familiar in most American kitchens, and both the iron grinders and the piedras were sold in the market. The Plaza of Cartago was particularly pretty, a large square shaded around the outer sides by tall, thick, higuer- ones (Ficus sp.) and by a great variety of other trees within. There were many well-tended flower beds, containing a won- derful variety of plants, not masses of one or two kinds only. The number of different plants in bloom at one time was often very striking. The gravelled walks were neatly kept, there were plenty of comfortable benches and the park in every way was an ornament to Cartago and the town was justly proud of it. As in other Costa Rican towns, the band- stand occupied the middle of the park and here the military band played on Tuesday and Thursday evenings and twice on Sundays. The almost invariable existence and good play- ing of the bands even in small villages in Latin America astonish most northern visitors, and certainly a North Amer- ican town of the size of Cartago that had so excellent a band would be unusual. On July fourth the “Star Spangled Banner” was included in the program of the evening con- cert as a compliment to the American visitors. On the north side of the Plaza was the Municipal Palace. On the east side stood the Parroquia, a large unfinished church, begun many years ago and damaged by an earth- quake. before the roof was on. In 1910 it was a big, thick- 40 A YEAR OF COSTA RICAN NATURAL HISTORY walled, roofless, stone building, apparently without foun- dations as there was no excavation within the walls for a cellar so that foundation walls could not be seen. There were the bases of the usual two towers, two chapels on each side, a curved chancel, and down the interior two rows of bases for the columns that will divide the nave from the aisles. Around the inner sides of these towers ran stone steps up which we climbed for the sake of the fine view from the top of the walls. In the southwest part of Cartago a large square was oc- cupied by the “Central American Court of Justice,” erected in great part at Mr. Andrew Carnegie’s expense for the peace- ful arbitration of subjects of dispute among the republics of Central America. When we arrived in Cartago the “Peace Palace,” as it was familiarly called, was up to the level of the first floor and work progressed steadily though slowly during the year we lived there. It was practically finished by May, I910, and was a very handsome building of brick and plaster, with some beautiful carved marble statues. The dedication was planned to take place in June, 1910, but the Peace Palace was absolutely wrecked by the earthquake of May fourth. ““Cartago”’ is the name of one of the five “‘provinces’’ of . Costa Rica, of the “canton central” of that province and of the town which is at once the capital of the province and head of the canton. Every canton is divided into ‘‘dis- tritos.” In this case the chief villages of several distritos are not only adjacent to Cartago but perfectly continuous with it—a geographical fact which is difficult for strangers to appreciate, but which explains the widely different es- timates of the size of Cartago. Thus the villages of Carmen on the north, San Rafael to the northeast, Los Angeles‘ to the east, San Francisco to the south and Guadalupe on the west are to the visitor absolutely undistinguishable from A Branch of Higuerén, Ficus sp. Flowers and Fruits of the Targua, Croton gossypitfolius. To face p. 40 Echeveria australis, from house roof. Two Chayotes and a Sapayo. To face b. ar THE TOWN OF CARTAGO AND ITS LIFE AI Cartago and the latter name was frequently applied to the whole assembly of villages. Each district had its own chief church as well as smaller ones. Cartago proper contained the handsome Gothic church of San Nicolas, the unfinished Parroquia, the Sole- dad and several others including the pretty chapel of the Orphan Asylum at the west end of the Calle Real. The Orphan Asylum had quite extensive buildings and we often saw the boys at work in their large garden. The district of Carmen lay just north of the railroad and the Carmen church was very conspicuous from its extreme width and disproportionately low, squat, square towers. South and east of the Parroquia was the large church and monastery of San Francisco. It was unfinished when we settled in Cartago and on Sunday, the eleventh of June, a big “Turno” was held for its benefit. The town was pla- carded for days with large posters announcing a “Turno General” for the benefit of the church of San Francisco de Cartago and we waited with some interest to learn what a “Turno” might be—the dictionaries vouchsafing no perti- nent information. The event proved to be an occasion on which everyone “turned to” and worked for the church. The market that day was exceedingly crowded and there were all sorts of extra things for sale,—more flowers, more pots, more cooked food, little odds and ends that looked like personal treasures—all to be sold for the good of San Francisco. Outside the market were numbers of temporary booths where young ladies—not market people—sold or more frequently raffled everything imaginable, just like a church fair at home. There were many groups. of pretty little girls in the streets all day selling ‘‘chances” on dolls, toys and soon. About eight o’clock in the morning we saw a long procession of oxen coming down the mountain and when they were nearer we found that each pair dragged a 42 A YEAR OF COSTA RICAN NATURAL HISTORY large log or tree-trunk, pointed at one end so it could be more easily hauled. There were twenty-five or thirty teams in this one file and we followed it to the back of the church. Here a previous procession had deposited almost as many loads of sand, stone, gravel, bricks and other building ma- terials. The gates were open giving us a glimpse into the garden where we could see some monks in their brown robes with hoods and rope girdles. ‘They looked pleased and with reason, for all this labor and material was donated and work on the church progressed rapidly after the Turno. Before we left Costa Rica the church and other buildings had been finished, forming one of the finest ecclesiastical groups in Cartago. Alas! We not only saw it finished but utterly destroyed a short time after its completion. At the extreme east of the town stood the Church of Nuestra Senora de los Angeles, one of the most remarkable of the Cartago churches from the large collection of silver votive offerings and the sacred spring behind the church. The shrine enclosing the spring was a curious eight-sided pavilion; there was no roof but the sides were filled in with iron grillwork and a sign invited alms for the completion of the edifice. The door was locked but one could look through the ironwork of any side. Steps and an inclined path of cement led down to the spring in the center, and this path was bordered on each side by wonderful constructions of ce- ment apparently representing the waves of the sea and a lighthouse. Each wave was coated with sea-shells, and there was besides a tiny full-length figure of the Virgin under a glass case, a large Sacred Heart under another glass case, while a little promontory between two windings of the path was occupied by a tiny garden of real flowers carefully en- closed by chains and wire netting. The inside walls below the grillwork were adorned with paintings of landscapes— hills, a train crossing a high bridge, people walking on a THE TOWN OF CARTAGO AND ITS LIFE 43 boulevard, etc. The diameter of the enclosure was not much over thirty feet; its contents formed a curious mixture. This shrine marks the place where a healing spring mi- raculously appeared in the early eighteenth century, bub- bling through a hole in a round flat rock on which was found a small image of the Virgin. The image was often famil- iarly and affectionately known as “La Negrita’’ because it appeared in a part of the old settlement occupied by the poorer people and such few negro slaves as there were. Both image and spring have always been endowed with miracu- lous healing powers and the fame of Nuestra Senora de los Angeles spread in time to Guatemala and even to Mexico and Peru. Many pilgrims traveled long distances to ar- rive in Cartago on August second, when La Negrita made a formal procession from some strange church (to which she was translated at midnight of August first) back to her own church. The pilgrims joined in the procession escorting the image and many strange vows were performed which had been promised if a cure was effected. Some of these were highly ridiculous such as painting the face strange colors, neglecting to shave, dressing like women or carrying heavy stones on the head. The more usual vow took the form of offering a gold or silver (preferably gold) replica of the afflicted organ. There was a wonderful collection of silver arms, legs, heads, hands, feet, ears, eyes, livers, and almost every part of the human anatomy, in glass cases (well fenced in) and some years ago there were also many gold ones but a German bishop gathered up the best of them and took them to the Berlin Museum. This proces- sion was held annually until a few years ago, when a more intelligent Cura opposed the custom and it was finally abol- ished. The spring still bubbles through the circular stone, and in former years there were many precious bits of stone 44 A YEAR OF COSTA RICAN NATURAL HISTORY sold as healing relics, the stone miraculously lifting itself up and growing whole again overnight. The round stone is of the black volcanic rock so common about Cartago—the inference as to the origin of the fragments is obvious. At the western end of the village of Guadalupe, west of Cartago, stood the very picturesque church of Guadalupe, shaded by a magnificent higueron and across the street was the Panteon or cemetery. The latter was surrounded by a high, thick wall containing many niches each of a size to hold one coffin, which was hoisted into place by a big travel- ing crane. When a body was put into one of these spaces the end was sealed and ornamented with the name and date, done in various styles,—black and gold, plain marble, sky- blue with bright sprays of posies, or whatever the taste of the mourners dictated. There the coffin remained as long as the rent was paid, but it was removed to some general unmarked grave as soon as the payments ceased. In addi- tion to the tiers of niches in the wall there were a number of isolated vaults at the ground level, or groups of three or four, and some monuments and life-size carved statues. A few trees were planted, among them junipers trimmed into ghastly likenesses of crosses, crowns, etc., but there was nothing—in this land of flowers!—approaching a garden effect. The lower part where the poor were buried was the most desolate spot imaginable—bare ground, jagged holes half full of water, new mounds all unmarked, irregular, loosely filled in, with here and there a wooden cross all aslant or a feeble attempt at a flower. The people do not seem to object or they would change it of course, but it looked dreary in a country where trees and flowers run riot if given a chance. On All Saints’ Day we walked out to the Pan- teon to see if there was any particular decoration in honor of the day. Some of the tombs had fresh wreaths and gar- lands, some of fresh flowers, some artificial, and others of THE TOWN OF CARTAGO AND ITS LIFE 45 bright tin, and there were perhaps more people there than usual. Palm Sunday, or Domingo Santo de Ramos, was the beginning of the great Holy Week celebrations. Early mass was celebrated in a number of churches, but the chief event was the procession of Nuestro Sefior del Triunfo through the streets from the church of Guadalupe to the Church of the Soledad. The procession consisted of several hundred men and boys walking on the left side of the street, as many women and girls on the other side, with the band in the mid- dle playing a gay march anything but religious in tone. Behind the band a space was kept clear, then followed three ragged barefoot boys, the middle one carrying a small diamond-shaped purple banner while the outer boys held tall candles. After them came two men leading a donkey on which was a life-size figure of Christ—Nuestro Sefor del Triunfo—holding the.reins in the right hand and a large bunch of palm leaves in the left. The figure was dressed in purple velvet, with a long flowing purple velvet cloak embroidered in gold, a wide lace collar and a big purple hat from under which hung long curls—such a costume as Charles the Second of England wore. It had not been pos- sible to tie the figure on the donkey securely and in spite of men who walked close on each side and held the image in place with their hands it swayed and lurched from side to side in a manner more suggestive of a drunken man than a sacred figure. Behind the donkey came a crowd of men and boys filling the street solidly from curb to curb and extending about half a block behind. All the men walked bare- headed, and most were peons although there were plenty of people of the upper classes waiting in the park to join the procession as it came by and apparently to go to church with it. On Monday afternoon there was a procession from the 46 A YEAR OF COSTA RICAN NATURAL HISTORY Soledad to San Nicolas, escorting a life-size Christ seated on a gold chair. The figure was dressed in red and gold, hat- less but with a large nimbus. Immediately behind walked some priests most elaborately dressed, and back of them the band, the drums beating a strange muffled tattoo when we heard it. The usual crowd of men and boys surged down the street after the procession, the women and girls walking on the sidewalks only. On Tuesday the procession of Jesus of Nazareth left San Nicolas church about 6.45 P. M., went south two blocks, then east to the Soledad and returned to San Nicolas by the main street. There were many men bearing square glass- sided lanterns with one candle inside, mounted on long poles and held high in the air. Each lantern, from a distance, gave a cross-shaped light so marked that when the men came near it was hard to believe the lanterns were simply square-sided boxes. The lanterns walked on each side and behind the figures, but not in front. The police, caps in hand, kept the street clear of men and boys immediately in front of the procession. The first figure was a life-size standing St. John in blue carrying a large cup—shaped like a huge double egg- cup—in his right hand. Next came a life-size standing Christ in red, with bowed head and bound hands; in addition to the carriers, men walking on each side held long red cords that seemed to be fastened to the hands. (In the flickering torchlight it was impossible to see details.) Priests followed the Christ, then the band playing a very mournful and sub- dued air. The third figure was a standing Virgin with eyes and head upraised—also in blue. This figure was carried by young girls. Then followed more priests and the usual crowd. As the swaying figures and twinkling lights passed slowly down the street and around the corner into the church, accompanied by the mournful music, the effect was impres- sive and solemn, almost beautiful. THE TOWN OF CARTAGO AND ITS LIFE 47 About five on Wednesday there was a procession consisting of a life-size Veronica carrying a large cloth on which were painted three thorn-crowned faces; St. John in blue with his cup; a Christ with red velvet robes embroidered in gold, an immense head of curls reaching below the waist and crowned by a huge nimbus, and the Virgin, followed by the band and the usual accompaniment of men, women and children. It was noticeable that in these processions very few hatted women were to be seen. Many of the poorer women do not usually wear hats, although the custom of doing so is grad- ually spreading. During Holy Week, however, all hats seem to be laid aside by them and the rebosa—here either black or some very light color—is draped tightly and closely around the head and shoulders. The women of the upper classes always wear hats but these women do not appear in the processions, only as spectators. Holy Thursday seemed to be the greatest gala day of the year! All day long, country folk in their cleanest and starch- iest, city people in their newest, brightest, gayest, most ballroom-y attire strolled up and down the streets, visited the churches and generally enjoyed themselves. In the evening A. went to see the decorations in some of the nearby churches. All were beautifully and tastefully decorated and all were full of people, particularly the great Franciscan church from which the evening’s procession started. The people were sitting all over the floor so that if we had cared to approach the altar we could not have done so. There was much subdued gossiping and visiting about from group to group and the frankest criticism and discussion of the gor- geous costumes and huge picture hats of some of the Costa Rican young ladies, who did not hesitate—as we did—to force a way through the crowd and rarely ran the risk of soiling their fine dresses by kneeling on the floor. In the afternoon the governor of Cartago and his staff, in black 48 4A YEAR OF COSTA RICAN NATURAL HISTORY clothes and high silk hats, accompanied by the band in full dress uniform with red silk pompons in their caps, visited the churches and shrines. In the evening the figure of Christ was led to trial escorted by Roman soldiers. About eleven o’clock on Good Friday morning the pro- cession of Christ bearing the cross left Carmen church and went to San Nicolas. At that corner it was met by a second procession of Veronica, St. John, Mary Magdalene and the Virgin. As the two processions met, the first halted, the figures in the second bowing to the Christ, until the Virgin was opposite when the Christ and his mother bowed to each other. Then she turned and took up her station beside him. A long pause was made here while an impassioned and fiery sermon was preached from an open-air pulpit. Then the processions moved onward and the Crucifixion followed. The burial took place about four o’clock, when the figure was carried through the. streets in a gold-trimmed glass coffin accompanied by a dozen little girls as angels, holding ribbon streamers and followed by the Virgin in a magnificent black velvet robe spangled with silver. The coffin was put into a space in the altar of San Nicolas. The church bells were of course silent from Good Friday to Sunday, but on Good Friday the small boys carried— and frequently sounded—wooden rattles like an old- fashioned watchman’s rattle, sticks that they clapped and beat together and a noisy device consisting of a board with a number of iron loops or rings fastened at one point only. The board was turned and twisted quickly from side to side making the metal rings flap up and down. These various rattles were the only noises permitted on Good Friday. It should be noted also that no one rides horseback on Good Friday. Costa Ricans never do and for a foreigner to ride on that day is to run counter to deep-seated conventions A Street Corner in Cartago and Church of San Nicolas. Church of Guadalupe, Cartago. To face b. 48 The Good Friday Procession. The Palm Sunday Procession. To face p. 49 THE TOWN OF CARTAGO AND ITS LIFE 49 and lay himself open to rebuke and dislike or stronger measures. Early Easter morning—long before light—the church bells began to ring with a great clamor and the band was heard playing its gayest march as it conducted a procession from church to church. On Sunday, June thirteenth, we walked about the streets to see the celebration of the last day of Corpus Christi. Being market day there were many visitors in town and large crowds around altars erected at the street crossings. These were large wooden platforms each with a high back support- ing a canopy. They were all differently decorated and fin- ished, but all had much tinsel, many beautiful flowers and enormous quantities of white paraffin paper crumpled up over the whole background for clouds. The effect of this background was astonishingly good and pleasing. There were four altars in our part of town, forming the corners of a large rectangle two blocks on a side. Along the streets con- necting the altars men had set up branches of trees so that the streets were lined with green. Many of the houses were decorated with coarse lace curtains hung outside the windows and doors, and looped up with crepe paper festoons, ribbons or banners bearing the words ‘‘Viva Jesus Sacrementado.” The crowd of celebrants passed along the street from one altar to the next, headed by the band. When they reached the altar there was a service and singing, the crowd kneeling in the streets. The altars now were occupied by half a dozen little girls dressed in pale blue, pink or yellow, with huge wings, representing angels, who were disposed among the white paper clouds very prettily. All the men who happened to be near stopped during the service and stood bareheaded until it was over. As soon as the service was ended at an altar the latter was taken down and we saw the little angels being lifted from their resting places among the clouds, before 50 A YEAR OF COSTA RICAN NATURAL HISTORY the procession had reached the next stopping place. The daily rain at that season made it necessary to remove the altars at once. On December fifth a turno was held for the benefit of the Orphan Asylum. It was not as great a function as that for the Church of San Francisco in July. Many articles were sold by chances at booths at the east end of the market house and a number of the guests of the hotel took chances and won small prizes. In the morning a few ox-carts brought gifts of firewood. In December the shops were given over to the most as- tonishing collection of small toys, mostly of tin, for “La Noche Buena” or Christmas Eve—such toys as we had never seen before as well as dolls of a familiar pattern, and many drums and whistles. The children seemed to play with toys very little at other times so that the shop windows were doubly attractive to them. It was the custom to go to confession on Christmas Eve, then attend midnight mass. All evening there were bombs and bells from the churches, increasing each hour. We were rudely awakened about one o’clock by the passage of the pious churchgoers, on their way home from mass. The air was rent with firecrack- ers, bombs, whistles, horns and church bells, with wild savage yells making us think of Indian warwhoops. On Christmas evening we went to the chapel of the Orphan Asylum to see some reputed decorations but found none. Service was in progress and we stayed a few minutes, the orphans singing the responses rather pleasingly. On the way home we went into a house to look at an elaborate “‘por- tal” or representation of the first Christmas. The whole portal occupied a raised platform about five by ten feet, enclosed by a little green-draped fence a foot high. The Holy Family were seated under a sort of altar which was stable as well, for cows and horses were looking through THE TOWN OF CARTAGO AND ITS LIFE 51 the walls and climbing over the top. Three wise men were riding, single file, up an inclined plane leading to the middle of the altar. The rest of the space was taken up by farms, trees, wild animals, a railroad train on little tracks, hunters, horseback riders, a glass lake in a very realistic bower of branches and moss, with swans, ducks and geese swimming on its surface-and giraffes feeding in the underbrush, a wind- mill, etc., etc. All these things were dolls and other toys, cleverly dressed and arranged in the midst of moss, lichens and green cloth, with small tree-trunks, bushes and little toy houses. These Christmas altars were said to be in many houses in Cartago, but this was the only one we saw in our short walk. The Fifteenth of September is the Costa Rican “Inde- pendence Day” and was celebrated in Cartago by the school children singing patriotic songs. The military band played in the Plaza, the first number on the program being “ Himnos Centro-Americanos,”’ or the national anthems of. Costa Rica, Guatemala and so on. The twelfth of October, being the anniversary of the day on which Cristobal ‘Colon desembarca en la isla de S. Salvador,” as the daily calendar in the hotel dining-room put it, was observed as a holiday, stores and schools being closed. The Presidential Election was held in August, 1909, and when we arrived in Costa Rica in May we found the houses and shops everywhere freely placarded with the red signs and posters of Don Rafael Iglesias or the blue signs and post- ers of Don Ricardo Jimenez. The former was the candidate of the Civilista Party and supposed to be favored by the United Fruit Company and almost all the foreigners. The latter was the candidate of the Republicanos and we were told that Don Ricardo was the really popular candidate supported by the large majority of Costa Ricans and particu- larly by the numerous element that disliked all foreigners 52 4 YEAR OF COSTA RICAN NATURAL HISTORY and the Americanos most of all. We were also told that al- though Don Ricardo was the popular candidate his election would never be “permitted” and further that in such a case there would certainly be civil disorders. But it is safest to prophesy after the event. The red posters, usually bearing the words “Viva Igles- ias,”’ vastly outnumbered the blue in Limén and the little settlements through the banana country owned by the Fruit Company. But blue signs with the legend “Viva Jimenez” began to appear as the railroad ascended from the lowlands, and even before Cartago was reached had become more frequent than the red. Cartago was the home of Don Ricardo and naturally the center of his party. Political meetings and discussions occupied much time and attention among the Costa Ricans and sometimes became very heated —phenomena at which no American had any right to wonder. But at no time was there any disorder and although we can- not pretend that we read all the newspapers we certainly did not see in those we did read any such wholesale and un- restrained vituperation or vilification of the opposing candi- date as characterized our own recent campaign. As the elec- tion time approached the excitement of course grew intense. We were in Cartago during the three days of the election. Indeed we particularly planned to be there, so much had we been told about the probability of grave disorders in smaller places—a precaution utterly needless as it turned out. The first day of the election, August 29, was very quiet and peaceful. The returns began to come in about ten o’clock on the second day with reports of overwhelming majorities for Don Ricardo. There was naturally much enthusiasm in Cartago, and by night the town was given over to a screaming, cheering mob who waved blue banners, blue flags or blue rags and blew horns, while the band wildly played each instrument with cheerful disregard of all the THE TOWN OF CARTAGO AND ITS LIFE 53 others. The small boys were of course in their element, and we had not suspected that these sad-looking youngsters could be so hilarious. This enthusiasm was mild compared to the excitement of the third day, as the successive bulletins came in and Don Ricardo’s victory grew. At the last mo- ment before the election, the existing government, which had been counted on absolutely by the Iglesias faction to work for them, turned to Jimenez and orders were issued to chiefs of police, etc., to vote for him. This no doubt ac- counted for his overwhelming success, but it seemed certain that Jimenez was the popular candidate and would have received the most votes. In the evening a really big crowd went to the station to meet the San José train, as Don Ri- cardo was expected home on it. He did not come, however, so the crowd and the band swarmed up and down the streets. There was much shouting of “Viva Jimenez,” “‘Viva Ricardo Jimenez,” etc., but little of the concerted cheering to which we were accustomed. On the contrary each one cheered at random, and much of the yelling bore a strong resemblance to warwhoops and shrieks of the Indians in Buffalo Bill’s show. It sounded very unlike the deep roar of one of our enthusiastic crowds. The San José paper, La Informacion, for September 1 gave the final results as Jimenez 38,000, Iglesias 14,000. This election determined the electors (similar to our elec- toral college) who met in February to cast votes for the presidency—Costa Rica has no vice-president. Contrary to the fears of many foreigners there was no disorder any- where in the country during the election. CHAPTER V NATURAL HISTORY WORK IN CARTAGO One of the chief objects of our expedition being the study of the living dragonflies or Odonata, and of their larve, we very soon began to rear at Cartago such eggs and larve as we could procure. Eggs were obtained from the Zygop- terous species (those having wings of equal width, often called damsel flies) by bringing home and keeping in homeopathic vials the plant tissues in which we had seen the female insert her eggs. The Anisopterous dragonflies (or those having the hind wings wider than the fore wings) could often be induced to oviposit if we dipped the tip of the abdomen repeatedly into water in a vial. When the eggs in these vials hatched, as they usually did sooner or later, the young larve were watched, drawings of several of the early stages were made from life, in many cases under the binocular microscope and some individuals of each stage, with exuvie, preserved for future study. Larger larve, which were collected either near Cartago or brought there from other places, were kept in glass jars, tumblers, or bottles for ob- servation. Most of these larve were kept one in a bottle, for the practical reason that if they did not start one in a bottle they quickly became so. There were a few exceptions to this rule. Thaumatoneura larve, for instance, are slug- gish and not aggressive, and if the difference in size be not too great can be safely kept together but in general it is dangerous to attempt it. Each larva had a stick, branch, clump of water weed or small bromeliad for a resting place, and we soon found it was desirable to allow the stick to pro- NATURAL HISTORY WORK IN CARTAGO 55 ject above the surface of the water even with larve not ready to transform. Every jar was covered with mosquito netting at once to keep out mosquitoes and to keep in the rightful occupant. Each moult was recorded and the exuvia preserved. And first, last and all the time the larve had to be fed. As the number of jars containing larve increased, the task of feeding and caring for such a family was at times quite absorbing. The necessity of looking after them was one reason for our frequent returns to Cartago, and if we wished to be absent more than a fortnight it was essential for one of us to return for a day or two purposely to feed the live stock. We were fortunate in finding suitable food with little difficulty. To the youngest larve to be fed, those which had just used up their egg-yolk, we supplied infusoria, copepods and similar small animals, which we nearly always could be sure of finding in abundance in any grassy pool. For the older ones we gathered mosquito larve, or small Agrionine larve which at times could be literally raked in by dozens and were quite popular. Our main dependence, however, was on the red larvae of the gnat Chironomus, which always could be caught in any desired numbers, by the aid of a little water net, in a certain dirty ditch at the lower end of the town. The existence of these “‘blood-worms” in such © unvarying abundance was really the secret of our success in rearing as many species as we did. They are greedily eaten by all but the extremely young dragonflies and were served up in “assorted sizes,’ tiny worms to tiny dragonfly larve, middle-sized worms to half-grown larve and fat full-grown worms or pupz to the big fellows. They are very juicy with few hard parts to be discarded and have the further advantage of wriggling constantly, so that our charges saw them readily. It was a droll sight (although undoubtedly hard on the blood-worms) to watch a hitherto motionless 56 4 YEAR OF COSTA RICAN NATURAL HISTORY larva “sit up and take notice” when a squirming worm suddenly appeared in his tumbler, to see how he followed its movements until the chance came to strike out for it with his jointed mask with its sharp jaws at the end, then how he held the prey and chewed away until the last wrig- gling bit had been eaten, when he immediately started after a second worm. The large larve sometimes ate nine or ten blood-worms as fast as they could be dropped in and one could see them “swelling wisibly” and trace the red worms in the alimentary tract through the thin body walls. The little dragonfly larve were often turned head over heels by the prey, but they hung on and continued to eat no matter what position they were knocked into by the unfortunate blood-worm. Even Thaumatoneura, which probably feeds on quite different larve in its native waterfalls, ate the blood-worms in confinement if these were made to wriggle over the stone directly in front of its head. I never saw these sluggish larve pursue the blood-worms as other species did. Collecting the blood-worms was easy for all that was necessary was to hold a water net at the narrow end of a cer- tain ditch, and the worms carried by the current were caught and held by the net. We gathered them for months and the supply showed no signs of failing. On February 24 we counted the worms that passed downstream, without an disturbance on our part, the number reaching one hundred and fifty in five minutes, and on March 11 we counted one hundred and eighty in the same space of time. The only unpleasant feature of gathering them was the dirtiness of the water in the ditch—a house drain—but it is the habit of these worms (which are world-wide in their distribution) to live in such water. One day in December we supposed we had gathered plenty of food, chiefly Agrionine larve and copepods, to last for six days, but on examining the dish some hours later, this NATURAL HISTORY WORK IN CARTAGO v4 was found to be a miscalculation. Not only had the larger devoured many of the smaller, but in addition, when the pail of water was poured into a flat dish and left to settle, there appeared incredible numbers of green hydras. These attached themselves to the sides of the dish in such quantities that they looked like green moss; they were equally thick all through the mass of mud that had been gathered in with the larve. Even the larve themselves were adorned with them. I found hydras attached to heads, eyes, antenne, wing-pads, legs, abdomens and gills, sometimes so thickly clustered that I thought the larve were green-striped or spotted. ‘These hydras ate up all the smallest kinds of food, the copepods and daphnids, so that it was soon neccessary to go out for more. If a larva in rearing showed signs of being ready to trans- form we usually removed the twig with the larva upon it to a dry netted tumbler so that the perfect insect could not fall back into the water and drown or soak its wings before they hardened. Occasionally this happened during our absences but we had few accidents of that kind. Owing to the elevation of Cartago we had little difficulty with mould and mildew, while cockroaches and ants did not give us anything like the trouble we would have had at a lower location, so that we were able to preserve our specimens with the minimum of labor. Of the dragonflies which we found in the immediate vi- cinity of Cartago, three species were especially familiar, as they occur also in the eastern United States. Two of these were species of Pantala, flavescens and hymenea, large insects of strong flight, with a wing-spread of almost four inches and the hind wings very broad at the base; the former with a yellowish body and some yellow at the base of the hind wings, the latter predominantly olive and with a rounded brown spot on the hind wing base. Both have an 58 A YEAR OF COSTA RICAN NATURAL HISTORY extensive distribution. P. flavescens is well known as the most widespread of all dragonflies of the world, having been found in all quarters of the globe except Europe, in many an oceanic island, in a Himalayan pass at an elevation of 11,000 feet and was observed to fly in numbers at 11 P. M. of the 11th of April, 1896, on to a Peninsular and Oriental steamer in the Indian Ocean, 290 miles from the nearest land. P. hymenea is confined to the New World, but ranges from New Jersey and South Dakota to Chile and Argentina. It is, therefore, not surprising to find these Pantalas in Costa Rica. The third species is much more remarkable. It is the smallest of all the species known in the United States and bears the disproportionately long name of Anomalagrion hastatum. It is, in one respect, unique among all dragonflies in that the stigma, or opaque spot, is, on the front (but not the hind) wing of the male, set back from the front edge of the wing instead of being bordered by that edge,— whence Anomalagrion. -The body of 4. hastatum is very slender and its delicate wings do not permit of anything but a weak flight. Yet it has been found throughout the United States east of the Rocky Mountains, in Mexico, Central America, the West Indies, Venezuela and even in the Gala- pagos Islands, although, previous to our collecting it, it had not been reported from Costa Rica. Within the last-named country we found it at altitudes between 5000 (EI Alto) and 2200 (Turriicares) feet, and on both the Atlantic and Pacific slopes. As it has been found at sea-level in Guatemala and Honduras, it is likely that it occurs at the similar loca- tion in Costa Rica. One of the unsolved problems of geo- graphical zodlogy is why this feebly-flying insect should have so extensive a distribution when other species of more powerful flight and living in similar situations, both as larve and adults, are confined to much smaller areas. In its larval stages, 4nomalagrion must possess a great range of NATURAL HISTORY WORK IN CARTAGO 59 adaptability to temperature, since it is exposed to freezing weather in the United States while in Costa Rica it has never to contend with that condition. The wide distribution of Anomalagrion hastatum has ap- parently not led to the formation of different races charac- teristic of different parts of its area of occurrence, as in many other species of animals. Thus, the Costa Rican males have an average length of abdomen of 18.3 millimeters and of hind wing of 10.5 millimeters; corresponding figures for in- dividuals chiefly from New Jersey and Pennsylvania are 18.9 and 10.8. The females are slightly larger, longer in abdomen and have distinctly longer wings. Figures cor- responding to the above for the so-called “black” females are: Costa Rica 19.1, 13.2; United States 19.5, 12.9; and for the “orange” females: Costa Rica 19.2, 13.1; United States 19.2, 13. The slight differences here shown may in part at least be due to unequal shrinkage of the abdomens after death, as all these measurements are from dried specimens. The abdomen of the male is yellow, becoming more and more of an orange from its fore to its hind end, with marked black spots on all but the last two or three segments. The size and shape of these black markings vary much, but the tendencies of their variation are the same in specimens from Costa Rica and from the United States. The females for the most part appear in two forms; one, known as the “orange” female, has a large (anterior) part of the abdomen orange, the other, the ‘“‘black”’ female, has the entire upper surface of the abdomen black. Other less conspicuous color differences also exist. The majority of dragonflies as they grow old show on one or more parts of the body an appearance similar to the “bloom” of a plum or dark grape, whence they are said to be “pruinose.” This pruinosity is frequently to be found on the black females of Anomalagrion but not on the orange females. Transitional 60 A YEAR OF COSTA RICAN NATURAL HISTORY females are also found in which the orange area of the abdo- men is faintly or deeply clouded or almost black. Com- bining all this evidence, the conclusion seems justified that, in this species at least, the ‘‘black” females are merely older stages of those which formerly were ‘“‘orange.” There is thus a marked color change accompanying age in the female sex of this species, but not in the male. Even greater color changes of this character may be found in both sexes of another species far more abundant about Cartago—Anisagrion allopterum. The individuals which have just transformed from the larval stage and expanded their wings, so that they are able to fly weakly, have the head and thorax of a pale dull brown (“‘light pinkish cin- namon’’) and the abdomen of a pale yellowish with a dark brown or black cross ring at the hind end of most of its segments. Eventually, when some pruinosity is visible on limited parts of the body, the head has become black above, between the eyes, with four adjacent blue spots, the thorax is of a greenish-blue with a middle black stripe and a black shoulder stripe each side, and the abdomen is black above except for a fine pale bluish ring at the fore end of each of the second to the seventh segments. It is evident, therefore, that a very great difference exists between the colors of the recently metamorphosed individ- uals and those which are partly pruinose. As we found individuals representing many intermediate conditions, fly- ing in the same localities, it seems justifiable to conclude that the difference represents a change accompanying ad- vancing age. This conclusion is strengthened by the fact that those with pale colors possess a weaker and softer chiti- nous covering on their bodies, while the predominantly black ones have a much firmer and harder exoskeleton. The color changes which take place in the different re- gions of the body of this Anisagrion appear to be, to some NATURAL HISTORY WORK IN CARTAGO 61 extent, independent of each other, so that a given stage in the blackening of the abdomen is not necessarily correlated with a certain stage in the blackening of the thorax or of the head. In other words, these changes appear at different times in different parts of the same individual and in differ- ent order in different individuals, just as changes in the color of the hair of different parts of the human body, or of different parts of the human head, take place at different rates in different persons or even in the same person. The females of 4. allopterum become pruinose on a slightly more extensive area of the body than do the males and Right wings of the dragonfly Anisagrion allopterum, male. Enlarged 4 times. this pruinosity appears at an earlier stage in the blacken- ing of the abdomen than in the other sex, but on the other hand the blackening of the thorax and of the head takes place relatively later. In all stages of its winged existence, whatever be its colors, the male of Anisagrion allopterum is easily recognized by a peculiarity of its hind wings which is not to be found in any other dragonfly of Costa Rica. The vein which parallels the front edge of the hind wing near its tip, when it reaches the stigma is distinctly bent away from the front edge and towards the hind edge, but continues to the wing-tip. For some little distance outward from the point where the bend occurs, this vein is markedly thickened and at the wing-tip 62. 4 YEAR OF COSTA RICAN NATURAL HISTORY the cells (or small areas bounded by veins) are distinctly smaller than they are in the corresponding part of the fore wing. In the older males these small cells become partly or completely filled by thickening of chitin (of which the veins are chiefly formed) and in this condition strongly suggest the condition to be found in any dragonfly’s wing which has been injured before, or at, the time of transformation. Twenty-five years ago, this peculiarity of allopterum males might have been seized upon as an example of an inheri- tance of mutilation effects, but it can hardly be regarded as such now. Insignificant as these structural details of the wing of this little insect may seem, they are very constant. Indeed the arrangement of the veins of the wings of these and many other insects is among the most reliable of recog- nition marks, and the trained entomologist can often identify an insect from a detached wing or even, in some cases, a wing-fragment, with perfect certainty. Finally be it added that the hind wing of the female of allopterum shows only a slight tendency toward these features of her partner. Still more abundant about Cartago, perhaps the most abundant dragonfly of the vicinity, was Argia extranea, about 1°/, inches long, */; to ‘/, of an inch longer than Anis- agrion allopterum. The male is bright blue marked with black; the female has the blue of the male replaced by a pale brown or a violet brown. The genus 4rgia is very char- acteristic of tropical America, where it is represented by many species, some of them resplendent in metallic body colors, and a few species reach even to Canada. It is dis- tinguished, among other features, by the length of the stiff hairs or bristles on the legs, which are twice, or more than twice, as long as the intervals between them. Argia ex- tranea lives throughout a large part of Costa Rica. Watch- ing a female laying eggs in a plant at Juan Vinas on June 24, 1909, we brought a bit of the plant in water to Cartago and NATURAL HISTORY WORK IN CARTAGO 63 kept it in one of our tumblers. The eggs hatched between July 21 and August 5 and we were able to rear one of these larve in the tumbler until its final transformation to the winged state on February 23, 1910, the larval period thus lasting about seven months. When we were in Cartago in May, June and July we fre- quently went out in the early evening with cyanide jars, to capture the numerous moths which were then resting on the houses nearest the arc lights. Although the collection of Lepidoptera was entirely a secondary consideration with us, no entomologist could have passed by such opportunities as these walls offered, for the number and variety of species were a constant source of wonder and admiration. During July A. often captured great numbers of moths in the morn- ings, “sleeping” on walls, door-jambs or window frames. None were ever taken on the sunny side of the street and they were most numerous on the main street where the lights were brightest at night. On one occasion we happened to see and to gather a number of large handsome sphinx moths on a Saturday morning, when, being a school holiday, the small boys were out on the streets in force. Their at- tention was naturally attracted and by the time we reached the corner of our hotel street we had enough interested assist- ants around to block the sidewalk completely. Our bottles and hands being full we were glad of the excuse to retreat to our room lest the police should consider it necessary to dis- perse the mob and arrest the ringleaders. Others helped us also and on the evening of December fourth Mr. Turner, who was staying at our hotel, brought a huge moth which he caught near an electric light. It was alive and vigorous. We put it in a tin can with chloroform on cotton, and meas- ured and papered it next day. Its body was 2‘/, inches long, its front wing nearly 6 inches and the total spread of its front wings 12’/, inches. Its wings were colored in differ- 64 A YEAR OF COSTA RICAN NATURAL HISTORY ent shades of gray in zigzag lines above, dark purples with pale spots below. It was Thysania agrippina, the largest known species of the Noctuz, recorded from Mexico to Brazil in the Biologia Centrali-Americana. A striking-looking moth which came to the electric lights on May 13 and June 16 was Manidia lunus, having a wing- spread of about 314 inches, the wings on their upper surface of different shades of brown crossed by nearly twenty wavy lines of white or yellow, the hind wings with “tails” 34 inch long bordered by three black “‘eye-spots”’ with blue “pupils.” In daylight, resting in the corners of rooms in the hotel, were often seen large moths whose wings were of different tints and shades of brown and purplish-brown, crossed by many wavy lines of lighter and darker colors. They were attracted by the electric lights both out-of-doors and indoors. Some of these were Erebus odora, stray specimens of which are frequently observed far north in the United States; it reaches a wing-spread of six inches or more and has on the front wing, not quite halfway out from the base, a comma- shaped mark 4 inch high, in black and blue margined with a buff line. Another and similarly colored was a Letts, a smaller insect (wing-spread 4 inches) with a patch of orange hairs along the fore edge of the upper surface of the hind wings. On the morning of May 18, while out in the street near our hotel, we saw a bright yellow-breasted bird seize a sphinx moth which was resting on the ground, fly with it to an electric light wire and alighting there feed upon the moth, subsequently letting it fall. The moth (Pachylia resumens) had more than the posterior half of its abdomen eaten, but ats legs and wings still fluttered when stimulated. The bird an the meantime found another sphinx moth and flew off with it. These moths were resting near an electric street ight fixture to which they probably had been attracted the ‘snuoynl snsaqosg JO aye I 8 pur ayetuay ‘L *(eteqry wory) afew suspaaxa snouvyg “9° ‘Pq0]2G Sts0jm) JO ae] 5 pue aeway ‘by “s177092g0] sajaksojky Jo ape ‘€ pue a[eway “Z ‘opeul ‘wormkpus vuauy ‘1 “aZIs aft] A]IvaNj—sapleog praqevieag Scarabaeid Beetles—About life size. Megaceros philoctetes, male (from life). Dynastes perseus, male (from life). Vo face p. 65 NATURAL HISTORY WORK IN CARTAGO 65 preceding night. We record this occurrence because it was the only time during our entire year in Costa Rica that we observed a bird feeding on Lepidoptera. We went there very favorably disposed toward the theories of mimicry and of protective resemblance. We expected to see many instances of birds pursuing and feeding upon butterflies, and frequently looked for such, but were utterly unsuccessful. Our experience in this respect entirely coincides with that of Mr. William Schaus in his three years in Costa Rica. Among the conspicuous insects which were attracted to the electric lights in Cartago, as in other parts of the world, were scarabaeid beetles, armed on the upper surface, in the males at least, with prominent horns. Like their allies of the temperate climes, the June bugs and chafers, they are heavy blundering fliers and their size was frequently such as to cause a momentary annoyance if they struck against one’s head. To one interested in the variety of form which Nature displays, the horns offer a field for much speculation. They may be one, two or three in number and variously placed on the head or the prothorax. Xyloryctes lobicollis, a dark brown scarabaeid an inch long, which came to the Cartago lights in May, has a single sharp- pointed rhinoceros-like horn, on the top of his head but no horn on the thorax, although there is a low knob-like swelling on the middle of that region (which is somewhat hollowed out in front of and below the swelling). The female, of the same size and color as her mate, has a shorter pointed cone on the top of her head corresponding to his horn while her prothorax is smooth and convex throughout with no indica- tion of knob or concavity. Enema endymion came to the lights on the same nights with Xyloryctes. He is longer—1}4 inches or more in length—and wider—34 inch as against 5/g—and of a much redder brown; his wing-covers are much smoother and show, 66 A YEAR OF COSTA RICAN NATURAL HISTORY to the naked eye, very few of the longitudinal rows of pits and punctures so easily seen on the preceding species; on the other hand his prothorax has a more roughened surface. He too has but one horn and that on his head, but it is curved more strongly backward so that the interval between it and the prothorax is much less. The latter has no horn or knob but is flattened and smoother on the middle of its forward half. We did not take the female of this species. As de- scribed by M. Chevrolat, she differs by the horn on the head being triangular, its base occupying the entire width of the head, and her thorax more distinctly pitted. Pinotus (or Copris) carolinus, the largest of the dung- beetles of the eastern United States, was also at Cartago, with its broad body, 7-grooved wing-covers and steep fronted prothorax, and his flat head bears a single broad horn notched at its tip. Coelosis biloba did not meet us at Cartago but was at Cachi and at Juan Vifas and may occur at the higher town. He is 11% inches long, of a reddish-brown which becomes darker to blackish on the front of the prothorax and on the head. The latter has a stout horn which is curved upward and then backward through more than a quadrant, so that its pointed tip lies in the interval between the two broad and rather thin prongs of a great forwardly-directed projec- tion from the prothorax. These prongs diverge (as one looks at the beetle from above) and each ends in a tip which is cut squarely across. The female offers a striking contrast owing to the total absence of both horn and thoracic projec- tion. Megaceros philoctetes male, which was at Cartago in March, develops the horns of Coelosis to a greater degree. This is a heavier beetle although of the same length, its color is a blackish-brown, its wing-covers without conspicuous lines or furrows as Coelosis has them. The horn on the head NATURAL HISTORY WORK IN CARTAGO 67 is longer and stouter, not so strongly curved backward so that it by no means reaches the interval between the pro- thoracic prongs and is notched at the tip. The prothorax rises into a projection which at its summit divides into a right and left prong, widely separated from each other by a straight edge. While we found no female of this species, we have seen specimens collected in Panama by Mr. G. C. Champion. Her head has a minute point on its upper sur- face to represent his horn, her prothorax is convex but not elevated and two very small points, near its fore edge, close to the middle line and about '/,, of an inch apart, may cor- respond to his prongs. The male of the giant Perseus beetle (Dynastes perseus) also of this family, one individual of which lived as our boarder, as related on another page, has also developed a cephalic horn and a prothoracic process or horn. The length of the body exclusive of the horns is three inches, the head and prothorax are blackish, the wing-covers green pitted with black. This Central American Perseus has been confused with the South American Hercules which is still larger and has three teeth on the upper surface of the cephalic horn instead of two. We have not seen the female of Perseus but the female Hercules is of a dull brown throughout, has a small point (not a twenty-fifth of an inch high) on the top of the head, no projection of any kind on the prothorax and the surface of her wing-covers is finely netted, quite unlike the elytra of her partner. She is about two and one-half inches long. Quite a different arrangement of the horns exists in Strat- egus julianus which also came to the electric lights of Car- tago in May. This beetle is 134 inches long, its head and prothorax blackish-brown, almost black, the wing-covers a rich deep mahogany-brown, the last smooth and polished but with some discontinuous longitudinal scratches, and a 68 A YEAR OF COSTA RICAN NATURAL HISTORY continuous groove parallel to the suture where the wing- covers meet and less than a sixteenth of an inch distant from it. The head in both sexes is roughened and pitted and bears two small projections too low to be called horns. The male has three great horns on his prothorax; the female has one very low pointed tubercle in the place of the front horn of the male, and still less to represent the two lateral horns. Much speculation has been indulged in as to the use of these horns. Darwin (1871), chiefly from lack of evidence as to any function, came to “The conclusion that the horns have been acquired as ornaments.” We have been able to find no recorded observations on the uses to which the horns are put in any of the Costa Rican species just described, and very few for allied kinds throughout the world. Such as we have brought together are as follows: Mr. G. E. Murrell quoted by Lintner (1889) observed many Dynastes tityus on ash-trees at Coffe, Virginia, “‘never using their horns, so far as I can see, except for fighting.” Dr. F. Ohaus (1911) ‘‘was able to observe in Megasoma hector, a few specimens of which he found near Petropolis, Brazil, that the beetle uses its cephalic horn as a good weapon, with which it can pinch hard, . . . the South American Dynas- tids used this organ chiefly as a weapon.” Dr. Albert L. Bennett (1899), who observed Goliathus druryi in German West Africa and in French Congo, says: ‘“‘the male beetles use their cephalic horns in fighting with one another, as well as for puncturing the bark of vines in order to bring about a flow of the sap upon which they feed.” Mr. J. Doll (1885) found, in Hell’s Cafion, Colorado, a Dynastes which was thought by Dr. G. H. Horn to be D. granti. Doll says: “They are always found near the tips of branches [of the mountain ash], where by means of their projecting thoracic horn they scrape through the soft bark to cause a flow of sap which is very sweet, and of this consists their food.” NATURAL HISTORY WORK IN CARTAGO 69 Prof. R. W. Doane (1913), of Stanford University, ob- served that Oryctes rhinoceros, feeding on the growing heart in the crown of the cocoanut trees in Samoa, has first to pass through a web-like sheath before it reaches the hard wood. “In doing this the head is lowered and the [cephalic] horn thus thrust forward. The horn becomes imbedded in the tissue of the plant and when it is raised serves as an anchor to hold the insect while it pulls or pushes its body forward with its legs, or while it tears the tissue of the plant with its heavy mandibles.” Baron von Higel informed Dr. David Sharp (1899) that in Java he had “‘observed large numbers of Xylotrupes gideon; he noticed that the males sometimes carry the females by the aid of their horns.”! Dr. Ohaus has also kindly informed us that a correspondent of his friend Herr Nagel has observed, in Venezuela, that the males of Dynastes hercules fight very violent battles among each other for the females; that they seize and crush with the cephalic and prothoracic horns, the weaker male often having its thorax and elytra crushed, and that the victorious males take the females between the horns and carry them away. One Sunday, July 25, a peon brought to A. at Cartago, a male Dynastes perseus which he said he had found in the market among sugar-cane which had come from Juan Vinas. He was evidently afraid of it and was glad to sell it. We kept the beetle alive until August 24, most of the time in a 1 As the observations are very much scattered, it may be useful to give here the exact references to the original statements. Darwin, Descent of Man, Part II, Chapter X, rst ed., 1871, New York edition of 1896, p. 297. Lintner, sth Rept. on Injur., etc., Insects, State of New York (42d Rept. N. Y. St. Mus. Nat. Hist.), 1889, p. 231. Ohaus, Deut. Ent. Zeitschr., 1911, p. 230. Bennett, Proc. Ent. Soc., London, 1899, p. xiii. Doll, Bull. Brooklyn Ent. Soc., VII, p. 121, 1885. (See also Lintner, 7th Rept. on Injur., etc., Insects, State of New York, p. 252, footnote f, 1891.) Doane, Science, n. s., XXXVIII, p. 883, 1913. Sharp, Cambridge Nat. Hist., VI, p. 199, 1899. (Cf. also Cunningham, Sexual Dimorphism in the Animal Kingdom, p. 254, 1900.) Andrews, Journ. Exp. Zool. xxi, pp. 435-436, 1916 (“the significance of these male horns [of Dynastes tityus] remains entirely problematical”’). 7O 4 YEAR OF COSTA RICAN NATURAL HISTORY tin vasculum, and supplied him with sugar-cane. He ate readily and his attitude when feeding on the cane is shown in our photograph from life. He was very slow and pon- derous in his movements and never made any effort to fly when out of the box, but twice, at night, we were awakened by the sound of his bumping against the inside of the vas- culum, as if he were trying to fly, accompanied by a hissing. We learned nothing of the uses of his horns. He came to an untimely end by chloroform and the autopsy was performed at once. Two species of mites, a millimeter and six-tenths of a millimeter long respectively, were found living on this giant and were later described by Mr. Banks as new species {of Celenopsis and of Hypoaspis). The squirrels in the Plaza were a great source of amuse- ment. The common one (Sciurus adolphet dorsalis) is a handsome little animal, very dark gray above, reddish- tawny below and on the under surface of the tail, pale gray on the cheeks and with extremely large black eyes. I watched one trying green mangoes by nibbling one, finally selecting and eating another, dropping down the green skin and eating the hard pulp. Before this one was half eaten it too was dropped and the squirrel was off to a third. As the country immediately around Cartago was all under cultivation we did not see there any large wild mammals, but on November 1 as we left the hotel door we saw two boys carrying two small armadillos, about a foot long, which they offered for sale. We stopped to look at the interesting little creatures. The bands, twelve counting head and tail, were still quite soft. Even the large ears were scaly and stood up like a bat’s. The claws were very strong, long and curved. The boys said the armadillos ate earthworms and no doubt they eat insects also. The English sparrow has not reached Costa Rica, the bird most nearly taking its place being Brachyspiza capen- NATURAL HISTORY WORK IN CARTAGO 71 sis peruviana. This sparrow is about the same size as the English sparrow but much handsomer, with a reddish streak over each eye and a greater contrast of light and dark browns on the back. It has a pleasing song and is less aggressive and pugnacious than the English sparrow but is much like it in its tameness and intimacy with man and his dwellings. The notes of the common Costa Rican robin (Planesticus grayt casius) in April, its size and shape and general behavior were so like our own that I never grew ac- customed to seeing its yellowish-brown breast instead of a reddish one. In April we also heard a meadow lark (Stur- nella magna alticola) with a note much like that of our northern bird. Late in December we saw a night hawk, very close to (if not the same) subspecies (Chordeiles vir- ginianus virginianus) that we have in Philadelphia from May to September. It looked quite homelike but we did not hear its familiar cry. Our northern subspecies has been observed in Costa Rica, in May and in September by Mr. Carriker. Although the black vulture (Catharista urubu brasiliensis) is the common one all over Costa Rica we once, on Septem- ber 9, saw a pair of red-headed vultures or “‘zonchiches” (Cathartes aura aura) in a lane in San Rafael. It was the second time we had seen this species; the first occasion was a few days before near Alajuela when we met a pair in com- pany with three or four of the black species, eating the car- cass of a small dog. The black species is entirely slaty- black including the naked skin around the head and neck, the beak and the legs; it is called “zopilote” by the Costa Ricans and “jim crow” by the Americans. The other species has the head more slender, red, beak white, legs reddish-white and the remainder of the body brownish-gray. It is a close relative of the common Turkey Buzzard of North America, but is smaller; the brown borders to the 72 4 YEAR OF COSTA RICAN NATURAL HISTORY feathers of the back are narrower, less well marked, or even lacking, while the color of the upper side of the shafts of the primaries (brown when freshly moulted) soon bleaches to an old ivory, or yellowish-white, instead of remaining perma- nently dusky brown as is usually the case in the northern form (Nelson). At the bottom of the hotel street, perhaps ‘half a mile south of the railroad, was a large white slaughterhouse and in the mornings it was no uncommon thing to see a line of black vultures on the top of the front wall, waiting for their share, while many others were on the ground or perched in nearby trees. These zopilotes are very numerous around Cartago and are of course exceedingly useful scavengers. We sometimes. watched them bathing. Their habit was to fly into the water, take a few runs through it in their usual stiff-legged fashion, then lower the head into the water, and, crouching, walk forward until the feathers were thor- oughly wet. This happened in a short time for the feathers seemed to have very little oil on them. Then up flew the birds into some sunny place and after a few preliminary shakings assumed the “spread eagle” attitude, with wings outstretched, head thrust forward and always with their backs to the sun. Occasionally one would make some hop- ping steps, ruffling all its feathers, then spread out againg A hundred or more of these awkward birds all facing one way and all standing with their great wings spread out as far as possible, presented a ludicrous spectacle. The slaughterhouse stood below and outside the town itself on what we always called the “Slaughterhouse Road” and which we usually followed when going to the numerous lanes south of Cartago. These lanes, though running straight and parallel to each other for fully half a mile south of Slaughterhouse Road, were remarkably individual in character, and one of them, our “Shady Lane,” was much The Bottom of the “Shady Lane,” with Xylosma on each side. The Cartago Slaughterhouse and Zopilotes. To face p. 72 “Vind pue JoMOp ul—vs0yfipuois DLpuvij]0)—O|[t9uoqies ‘tuuburzjos vusojky Jo Avidg BSuuamopy VW To face p. 73 NATURAL HISTORY WORK IN CARTAGO 73 the richest place we found near Cartago and became our favorite collecting ground here. One lane led directly to the town refuse heap—where we went once and having watched and listened to the horde of vultures had no desire to go again. However in justice both to the town and the vultures it should be said that this garbage heap was astonishingly inoffensive, and moreover flies were not a great pest in Cartago. The day we went along this lane was May I0, 1909. The rains of the two preceding days had changed the dusty roads to damp and produced little swampy spots in the pastures. Over one of these swamps a species of dragonfly (Sympetrum illotum virgulum) was swarming. The males have bright red bodies, the females dull brown and many pairs were engaged in lay- ing eggs. Big blue and blackish Aeshna dragonflies flew swiftly about and several smaller species were found here and there by the roadsides. There was an exceedingly handsome frog here (Agalych- nis helene), one of which we took to the hotel and kept for some days.. When “‘asleep” its body, exclusive of legs, was two and one-half inches long. Ordinarily the upper surface of body and legs was a bright pea-green; below, the body was speckled white and reddish-brown, with a band of brilliant beautiful blue on each side of the abdomen. There were great color changes, however. When first caught it was pale green; when we took it out of the vasculum at home it was a dark dirty green with pale spots on the back. While we photographed the frog it grew light again with paler spots over the back. After nightfall it was again very dark, but although it spent the night in a dark cupboard it was pale green when we first looked at it in the morning and again at 4.30 P. M. The tips of all the toes (four on the front, five on the hind feet) are expanded into large fleshy discs with which the frog climbs readily. It was able to cling to the 74 A YEAR OF COSTA RICAN NATURAL HISTORY vertical side of our glass graduate without other support than its toes. When caught it secreted a quantity of sticky mucus having a powerful and disagreeable odor, which it was difficult to remove from the hands. The morning of May 11 was not so bright as that of the preceding day, although the sun was shining. We revisited these same swampy spots but not a single individual of the bright red species of dragonfly nor one of the peculiar frogs was to be seen there, although we did find one of these frogs later in the morning in the bottom of a nearby ditch. Other lanes were bordered only by the living fences of poro. Such fence-rows are very pretty, especially when the conspicuous red blossoms were seen in relief against the deep blue sky, and the shade afforded by the trees—often fifteen to twenty feet high—is most grateful but it is neither dense enough nor constant enough to encourage a shadow- loving fauna. But our “Shady Lane,” really a continuation of the Calle de las Monjas, although bordered primarily by the pord fences, contained so many other trees that it was like a little strip of woodland with a grassy path running through it. At its upper end, the side vegetation consisted of young Anona trees, thickets of blackberry, Xylosma, lantanas, with scattered higuerones. We once pulled off a few twigs from these wild figs, finding the milky juice quite sticky. The nearly spherical, brownish-black figs were about half an inch long and filled with small dry fruits. There was no pulp and no marked taste to the little fruits or to the figs as a whole and they are not used as human food, but bats are extremely fond of them. As the lane descended, the hhiguerones were larger, real giants three to four feet in diameter, some of them, and more closely clustered and as the shade grew denser the trees were more thickly covered with epiphytic plants such as bromeliads, orchids and Colum- NATURAL HISTORY WORK IN CARTAGO 75 neas. A tiny stream flowed down this part of the lane, Selaginellas and thick green mosses covered the stones and tree-trunks, and it was often hard to realize, while standing in the cool deep shadows of these trees, that less than fifty feet away the open potreros were panting under the brilliant tropical sunlight. Still lower, there were few trees, but hedges of Xylosma, many clumps of wild roses with pale pink, very double flowers and scattered through the grass numerous boulders of andesite lavas, each one the center of a little thicket of roses or the creeping sensitive plant. The lane ended in a grassy cross lane (that is, east-west) along one side of which ran the “bloody ditch” a stream that carried the refuse from the slaughterhouse and this cross lane finally led to the main road to Agua Caliente,—but as it lost in interest as it neared the high road we rarely followed it so far. When we first traveled down the Shady Lane in May, 1909, it contained no water except the little stream in the lower part, but the conditions varied greatly during the year. In late June there were many pools but we could pick a path through the lane. On August 20 we found the greater part of the lane a pond with hummocks of higher grass as islands and the smaller plants almost entirely sub- merged. Many dragonflies—or “planchitas” as we once heard an Indian call them here—were darting about laying eggs and from the place once a dry grassy path we dredged up quantities of dragonfly larva and a tadpole two inches long. In this condition the lane remained for about two months, a most useful source of food and material to us and always ‘interesting to visit. On October 14, therefore, we were greatly disappointed to find that it had been utterly ruined for our purposes of larva-hunting, by having a number of cattle driven through it. Their hoofs trampled the smaller plants and converted large grassy watery stretches into 76 4 YEAR OF COSTA RICAN NATURAL HISTORY black mud so that we were driven to look for larve in other pools, particularly one in the lane to the east, the Calle de P. Peralta. A lull in the rains caused the pools to shrink considerably about the middle of November, but on Novem- ber 29 the pools were full again and all the potreros soaking wet. On February 19 we found both these lanes much drier than they had been for months. Even the deeper pools in both which served as homes for aquatic animals since August had now no water and hardly any spots of wet mud. By February 24 the Shady Lane was dry and beginning to resume the conditions it presented when we first knew it. The bottom of what had been the deepest pool was now covered thickly with yellow-flowered Jussiaeas. P. dug down into some pockets of soft wet mud still remaining to see if any traces could be found of the dragonfly larve so common here in the wet season, but without success. A similar trial in the dried pool in the next lane east, a prolific source of larve in that period, was also a failure, although the empty shells of snails, former inhabitants of the pool, were there. By March 11, 1910, the conditions of dryness approached those found on our first visit to the Shady Lane nearly a year before. On May 27, 1909, when returning through the Shady Lane, I caught a butterfly which to my great astonishment, as I was taking it out of the net, thrust out from each side of the hind end of the abdomen a brush of blackish hairs looking like an ordinary camel’s hair brush. I immediately thought of the two brushes as possibly odor-producing organs to diffuse some disagreeable smell which would act as a means of defense, but I could not detect any odor whatever. I took the butterfly home alive and kept it so until the after- noon of the following day. As I took it from the envelope some hours after catching it the butterfly voluntarily thrust out its brushes once, and again on the next day when I began NATURAL HISTORY WORK IN CARTAGO 77 to work on it, but at no other time even when the abdomen was stroked, rubbed, or its tip slightly irritated. By pressing on the hind part of the abdomen with thumb and forefinger most of the brush could be made to extrude and the process completed by gently pulling with a small forceps. At no time could either of us detect any odor connected with the apparatus and a colorless or slightly yellow liquid that some- times passed out from the abdomen was tasteless to me. However in a specimen taken on May 29 above Cartago there was a distinct odor resembling cedar shavings. This butterfly (Lycorea atergatis) had the body 30 mil- limeters long and a wing expanse of go millimeters (334 inches). Its wings were tawny-brown, black, yellow and white, black predominating on the front wings, tawny-brown on the hind wings. The black on the front wings forms three streaks from the base outward which unite together at half the wing’s length; the black outer half bears six or seven mostly elongated yellow spots and some small whitish dots near the wing margin. On the hind wings the black forms a border containing thirteen small rounded whitish spots; there are two other blackish-brown stripes united at their outer ends not reaching to the black border. The head is black with some small white spots, the thorax and abdomen are striped with black and dull yellowish. The antenne are 18 millimeters long, blackish, but the last 7 millimeters yellowish. When the brushes were thrust out each appeared through a vertical slit. Each brush consisted of a stalk and the bundle of hairs. When at rest within the abdomen the hairs lie inside the whitish stalk through whose translucent walls they can be indistinctly seen. By the lengthening and turn- ing inside out of the stalk the hairs were gradually brought into view but were all closely pressed against each other forming a straight, parallel-sided brush, but when the stalk 78 4A YEAR OF COSTA RICAN NATURAL HISTORY had completely lengthened and brought even the bases of the hairs into view the hairs were directed every way and the brush became spherical. The brushes of right and left side were thrust out together, in our ex- perience, and when fully expanded gave an extreme width of about 24 . millimeters or 1 inch to the appar- [| atus at the tail end, which of course was in striking contrast to the same end when the brushes were with- | 1 J drawn. Each stalk was about the W/Z same length as the last abdominal s Y at <= segment and on its inner side was , ‘S ~ a flesh-colored elongated tubercle about half as long as the stalk. Abdomen of butterfly, Lyee Each brush was composed of some orea atergatis, male, showing hundreds of hairs, each hair was brushes fully extended. Drawn about 6 millimeters long. Under from life. Enlarged 1.7 times. i ‘ the compound microscope the hairs are pale yellow for a short distance at their bases and entire in outline, but for most of its length the hair is black, opaque, and its outline is saw-toothed. The teeth are sharper as the tip of the hair is approached, and are apparently little scale-like protuberances directed toward the tip of the hair.! Lycorea is a representative of the Danaine butterflies (of which the Monarch or Milkweed butterfly is the best- 1In an African Danaine butterfly, Amauris psyttalea, Dr. G. D. H. Carpenter made the observation that the male fluttered about four inches over the female’s head, quickly protruding the brush at the hind end of the abdomen and as quickly drawing it in again and adds “it was impossible to doubt that the male was en- deavouring to excite the female” (Proceedings, Ent. Soc., London, 1914, p. cxi). Scent organs of male Danaine butterflies have recently been discussed by Dr. H. Eltringham in the Transactions of the same Society for 1915, pp. 152-176, through which further references to the literature on the subject may be obtained. NATURAL HISTORY WORK IN CARTAGO 79 known member in the United States) a group characterized among other features by the possession, in the males only, of the pair of anal hair tufts. Lycorea atergatis is found from the State of Vera Cruz, Mexico, to Peru. Off to one side of the Shady Lane, close to the fence where the ground was higher than in the center, was a tiny by-path running under the big trees and vines. Here we were in a deep arbor of shade, very narrow to be sure but thick enough to be quite dark, and here were often flying numbers of the beautiful clear-winged Ithomiine butterflies. These have very few scales, only along the wing margins and on some of the principal veins, the rest of the wing being as transparent as a fly’s. They are exceedingly hard to follow as they fly in these dark places. Occasionally they issued into the clearer sunlight but were far more common in the shade. The Ithomiine butterflies, like the Danaine, are Nympha- lids but their males, instead of possessing the anal hair pencils, have one or more tufts on the upper surface of each hind wing near its front edge. On June 17th, three species of these Ithomiine were taken in this little stretch of the lane—Dircenna klugii (2'/, to 234 inches spread) with the margins and bands of the front wing and the margins of the hind wing pale blackish-brown, the transparent membranes, especially of the hind wings, with a reddish tinge; Episcada apuleia (one and seven-eighths inches spread) with the mem- brane very clear, the margins and veins narrowly blackish- or reddish-brown, a milky spot at the front edge of the hind wings; and Ithomia heraldica (two and one-eighth to two and one-fourth inches spread) a non-transparent species, whose front wings are reddish-brown at base, the rest blackish- brown with yellow spots, the hind wings semi-transparent pale reddish-brown with narrow blackish-brown margins. In June, when the morning sunlight was particularly 80 4 YEAR OF COSTA RICAN NATURAL HISTORY brilliant in contrast to the afternoon mist or rain, we found many insects flying over the pools and submerged parts of this lane. Beautiful Syrphids hovered over the flowers, their gorgeous colors looking wonderfully gem-like. They are bright green above, the green melting into a vivid dark blue while beneath, the green turns in certain lights to a rich copper-red. As the insect hovers in one place for some minutes, then darts off to another spot and hovers again, the wings, like a humming-bird’s, vibrating so rapidly that the fly seems to hang motionless, all these colors shine and glow so that the little creature is fairly dazzling. It is Volucella obesa, a species found from Florida and New Mexico to Chile, and in the East Indies; its body is 11 millimeters long, its wing-spread 24 millimeters. The wings have a brown spot on the front edge at a little beyond mid-length, not reaching halfway across the wing and a brown dot also on the front edge a little before the wing-tip. The “bloody ditch” harbored quite a varied fauna. On June 17 P. dredged up some snails, a few small fishes (not over two inches), leeches, a large number of black or sand-fly larve (Simuliide), dragonfly larve, and many individuals of the bug Deinostoma dilatatum an inch long. These last are interesting on account of the habit of carrying the eggs—to the number of twenty to a hundred—on the back. The eggs stand side by side and are attached by their tail ends, for when the time of hatching approaches two dark spots indicating the eyes of the young may be seen under the shell of that end of the egg which is toward the observer. We took one of the egg-bearing bugs indoors and kept it in a glass of water; some of the eggs hatched and the young at once swam actively around in the water, but the parent died before more than seven or eight of the eggs had hatched. On other days we found two kinds of frogs in our net. We wondered whether the leeches were able iE Ltg ( so es all 2s poo a pent to i ‘. BE Guitite—Acnistus arborescens—in flower. Terciopelo de Santa Maria—Miconia @ruginosa. To face p. 80 Clavel del Monte—/ussiza sp. Lengua de Vaca—Conostegia lanceolata. To face p. 81 NATURAL HISTORY WORK IN CARTAGO 81 to take blood directly from the water or whether they took it from the muzzles of cattle when these chanced to drink at this stream or some of its affluents higher up, but when the stomach of a well-preserved specimen was examined its contents were found to be sand-fly larve only. These leeches were identified by Professor J. P. Moore as probably Erpobdella triannulata, a species described from Lake Ama- titlan, Guatemala, a lnioogh larger, some being 3 inches long. There it is said to “abound, seeking concealment by day beneath stones and similar objects in the shallow waters and at night becoming active in the pursuit of small worms, insect larve and even the smaller members of its own species, which constitute its chief food. Having no toothed jaws it is not an habitual blood sucker, though it doubtless, like related species, attacks abraded surfaces of higher ani- mals when opportunity offers.” The adult sand-flies were naturally quite abundant and at times troublesome. The bite produces a characteristic mark ontheskin. At first there is a red dot surrounded by a paler red circular area which may be one-eighth inch in diam- eter. The next day the surrounding area becomes bluish while the central dot remains red, becoming darker and darker with succeeding days until itis at last almost black and although small (one-fiftieth inch) is quite conspicuous and raised more or less on the top of a little lump. Meanwhile the surrounding area has resumed its usual color, and these stages may occupy a week or more. The pain and irritation produced vary much in different individuals. Although dragonflies were our chief objects, we also col- lected Microdiptera regularly at each locality we visited. This was the easiest possible collecting, for all that was nec- essary was to sweep a special form of net (devised by Mr. E. T. Cresson, Jr., for this purpose) over the grass and low herbage, emptying the contents into a cyanide jar. Later, 82 4 YEAR OF COSTA RICAN NATURAL HISTORY we poured the catch into some convenient flat receptacle and sorted out the insects from the grass seeds and other vegetable debris. Of course all sorts of small insects beside microscopic flies were gathered, in such sweepings, and look- ing them over gives a faint idea of the astounding variety of shapes and colors existing among insects less than an eighth of an inch in length. In this sort of collecting we ob- tained a really wonderful number of genera and species. The common plants, wild flowers and weeds around Car- tago were a constant joy to us, they were so numerous and so beautiful. Some showed no periodicity at all that we could observe, being in bloom every month in the year. Among these were the lantanas (Lantana camara), the pretty double wild roses, the “Hoja del Aire” (Bryophyllum caly- cinum), the “Guijarro” (Stemmadenia bignonieflora), the common red and yellow milkweed or “Viborrano” (4s- clepias curassavica) the yellow Bidens (B. pilosa) and the showy yellow “Clavel del Monte” (Jussiga sp.). Others were rarely without a few flowers but produced the great mass of their blooms at one time. Such were the poro and the magnificent ‘‘Reine de la Noche” (Datura arborea). The pord (Erythrina corallodendron and costaricensts) is a tree much used as a “living fence,” the trunks a foot or less apart and connected by three or four strands of barbed wire. It endures an indefinite amount of pruning and branches out again with great rapidity. Whether pruned in the fence-row or allowed to grow more naturally in the cafetales, where it is sometimes planted for shade, the pord is a straggling tree that seems never to have made up its mind whether it will really flower or not. The ternate leaves are large and glossy, the leaflets much like the leaves of a balsam poplar. The papilionaceous flowers are a brilliant beautiful red, the keel enormously elongated (two to three inches long) while the other petals are extremely reduced. NATURAL HISTORY WORK IN CARTAGO 83 Both in producing new leaves and in flowering it is as irregu- lar, as full of caprices and vagaries, as plant could be. It flowered sporadically from May to December, the greatest number of blooms being noticed in November when the trees were most nearly leafless. At that period the color effect of a number of porés together was astonishingly like that produced by red maples in bloom, both in the rich yet misty red color and the stiff straggling spray. At such times, one branch might be leafless and loaded with red blossoms, another on the same tree bare of flowers but full of new leaves. In March, however, it was really full of leaves and fruits. The latter are long, narrow, brown pods, constricted markedly between the beans; they open while on the tree, showing the bright coral-red beans attached to the margins of the pods. The pord is of the same genus as the “palosa- bre” described by Belt in his Naturalist in Nicaragua. The Reine de la Noche grew freely among trees along the roadsides and borders of streams, reaching a height of ten to fifteen feet. Its white trumpets were ten to twelve inches long, and in April at the height of its season covered the plant so thickly that the foliage was almost entirely hidden. A stream-bank bearing a thicket of these plants in full bloom is a sight never to be forgotten. ‘ Still other species have a very marked season, such as the Poinsettia or ‘“‘Pastora” (Euphorbia pulcherrima) which blooms at Christmas time and in its full glory is a wonderful sight with its clusters of brilliant red leaves and little red and yellow flowers. The castor-oil (Ricinus communis) and the common blue sage (Salvia polystachya) flowered most freely in December. This sage, although a troublesome weed, is a beautiful plant, the flowers being the deepest, richest blue imaginable and borne in great profusion, so that a field infested with it presents a solid sheet of color. Many of the numerous mimosas and acacias also had a seasonal blooming. 84 4 YEAR OF COSTA RICAN NATURAL HISTORY One of the prettiest was the “Carboncillo” (Calliandra grandiflora), a bush often twelve feet high with an extremely compound feathery leaf, a stem handsomely striped with dark red, and large conspicuous flower clusters, of which the long dark red filaments were the showiest part. Its flat, dark red pods, borne on the outer ends. of twigs, were also pretty. We found it flowering from October to February. This plant grows by choice on the edges of banks or cliffs or stream ravines, often overhanging the water and never far from it. The “Guitite” (Acnistus arborescens), a shrub or small tree reaching fifteen feet in height, bloomed in May and June; its small, exceedingly sweet-scented, white flowers were thickly clustered along the stem and were highly attractive to many kinds of insects and therefore to humming-birds. The fruits were ripe in August, pretty little orange-colored berries which many birds greedily devoured and which we ourselves liked very much when we cautiously tasted them. We found them deliciously sweet and juicy, reminding us of currants or gooseberries. As the Guitite is a Solanaceous plant we thought it wise not to eat too freely of the berries since the Costa Ricans do not seem to use them as food. Melastomes were numerous both in species and individuals and varied greatly in habit. An herbaceous species with particularly handsome foliage, very common along fences and hedgerows, was the “Terciopelo de Santa Maria” (Miconia aeruginosa). Its leaves, often six inches long, were a rich dark green but covered with a beautiful velvety pile of long, soft, dark red hairs. Its flowers were inconspicuous and we found them borne most freely from April to Septem- ber. Another common Melastome, a woody species, ten to fifteen feet high, growing in fields and potreros, was the “Lengua de Vaca” (Conostegia lanceolata). It had narrow, stiff leaves but its pretty white flowers, which we saw in NATURAL HISTORY WORK IN CARTAGO 85 July and August, made the tree both conspicuous and at- tractive. On wooded banks and roadsides we often found the delicate Tibouchina bourgeauana as well as the beautiful Arthrostemma campanulare, the latter bearing large pink flowers more than an inch across. Acalyphas (such as 4. leptopoda) were common in wet places, and crotons were also numerous, both small orna- mental species and trees such as the flat-topped “Targua”’ (Croton gossypitfolius), which grows thickly in rough or waste land. Their blooming extended over some months but we found the targua fruiting in August. In the parks, crotons were to be found blooming all the year round, practically. From August to November the “Alacrancillo” (Dalea alope- curoides) was exceedingly abundant. It is a leguminous weed, troublesome in the pastures but very pretty neverthe- less. It grew two to three feet high with racemes of small lavender to white flowers, the inflorescence having a delicate gray or pearl color. The commonest solanaceous plants were the “‘Berengenas” (Solanum sp.), straggling woody shrubs with small blue or white flowers, large coarse leaves and clusters of. green berries. They grew by the roadsides and in poor, neglected pastures, much as Jamestown-weed does in the eastern United States. They were unattractive plants to us but the chosen food of many potato beetles and we often hunted over the leaves for them. These beetles (Leptinotarsa undecimlineata) much resemble the common “potato bug” of the United States (L. decemlineata) but differ in having the legs, except the pads, and the under sur- face of the body black, while in the latter these parts are chiefly reddish or reddish-yellow. Around houses, in poor pastures and on the open roadsides were great quantities of a most troublesome malvaceous weed, “Escobilla” (Sida rhombifolia). It formed thick clumps covered with pretty little buttercup-yellow blossoms half an inch across. These 86 4 YEAR OF COSTA RICAN NATURAL HISTORY only opened when the sun was hot and full upon them and having rather woody stems and stiffish leaves, both leaves and flowers wilted almost as soon as plucked and never re- vived in water. Consequently the only way to photograph this plant was at mid-day in situ and a hot and trying task it was! The flower beds in the Plaza at Cartago contained not only a great variety of so-called garden flowers and plants but also many of the common native plants allowed to reach large size as individuals—a practice we highly approved. It was in the Plaza we first saw the huge Gunnera insignis, which is not a native of the Cartago neighborhood but grows in the sheltered ravines and pockets on the volcanoes, the “Guacamayo” (Bocconia frutescens) the “Ortiga” (Wi- gandia urens), the handsome red Salvia involucrata. We made lists on several occasions of the plants blooming in the Plaza. On May 16A. noted the following: Agapanthus, Ageratum, Amaryllis, asters, balsams, calla lilies, campanulas (blue and white), cannas, Cassia, chrysanthemums, cosmos, Dahlia, daisies, evening primrose, geraniums, gladioli, Hi- biscus, Lantana, larkspur, lilies (red and yellow), lobelias, Nigella, Oleandra, poppies, roses, Scabiosa, squills, sweet williams, Stephanotis, Tradescantia, tropeolums, Zephyranthes and among foliage plants were bamboo, bananas, crotons, Coleus, numerous ferns, and palms. On Christmas Day we found the Plaza a riot of flowers and noted the following: Abutilon, Agapanthus, Agave (2 sp.), Amaryllis, asters, balsams, begonias, borage (blue), Bougainvillea, large bromeliad, Browallia, cannas, chrysan- themums, Cleome, Cobea scandens, Coleus, yellow composite resembling Lactuca, cosmos, daisies, Dahlia, Datura arborea, geraniums, Gladiolus, Heliconta, heliotrope, Hibiscus, honey- suckle, jasmine, Lantana, larkspur, lilies (orange yellow, red-green, tiger) lupine, maize, manzana rosa, marigolds, NATURAL HISTORY WORK IN CARTAGO 87 Mimosa, morning-glories, @nothera, olive, purple orchids, ortiga, palm, papaya, pastora, Phlox, poppies, roses, Scabi- osa, Sedum, Senna, Spanish paint brushes, spirzeas, Ste- phanotis, sweet alyssum, Tradescantia, tropeolums, yellow umbellifer, Verbena, violets, Vinca, zinnias, and ten others entirely new to us. Where hedgerows were allowed to grow there was a won- derful tangle of bushes and vines. The hedge itself was usually composed of Xylosma salzmanni, a plant looking somewhat like the California privet but armed at the tip of each twig with a pair of sharp strong thorns. In May and June it bears solitary white flowers. Blackberries and wild roses climbed through and about the Xylosma, while melastomes, large-flowered Jussizas, the showy ‘“‘Zorillo real” (Hamelia patens) with its red and orange tubular flowers, “‘Pavoncillo” (Cestrum aurantiacum), yellow Bidens, lantanas and ageratums grew in its shelter. Occasionally, where a hedgerow was backed by poro or other trees, there was now and then a solitary guijarro (Stemmadenia big- nonieflora); this is an apocynaceous tree with large glossy leaves and magnificent tubular flowers, creamy-white on the corolla lobes and shading to deep sulphur yellow in the throat. These are followed by highly colored twin fruits. The follicles have a greenish rind over an orange flesh and when they open show masses of shiny black seeds each embedded in a vividly red pulp. Like so many members of this order, the juice is quite poisonous and it is said that to use the ashes of its wood in preparing maize for tortillas causes serious sickness. In spite of its beauty, therefore, it is rarely used as a decorative plant, but we always enjoyed seeing it. The “Santa Lucia” or ageratum (Ageratum conyzoides) is by no means confined to hedgerows, for it is a troublesome weed almost ubiquitous on the plateau. Its color ranges 88 4 YEAR OF COSTA RICAN NATURAL HISTORY from pinkish white through pale or sky-blue to beautiful lavender, and like the more intensely blue salvia often covers the fields with its exquisite color. On the higher parts of its range, as on the Cerro Jocosal, Carrizal and Poas, the blue tends to become darker. In rainy weather we often found the railroad track with its banks and ditches the best or indeed the only available path for pedestrians. East of Cartago the line crossed a number of small streams on low iron trestles, and usually a little footpath led down to the water so that we could get into the pastures easily. Boulders and masses of andesite lavas or tuffs are scattered thickly over these potreros on the slopes of Irazt and the superficial likeness to a granite- strewn New England pasture is striking. A second glance shows that the trees along the stream are melastomes, cro- tons and mimosas, highly characteristic tropical American groups, instead of birches and maples, while the boulders are encircled with sensitive plants, wild roses as double as chrysanthemums, or masses of bryophyllum, and the pas- tures are enclosed with living cerclos of poré, their crimson blossoms flaming against the deep blue sky. No, it was not New England, but a far richer land pulsating with life all the year round! On the railroad banks many flowers were abundant. Sometimes a pretty little composite (Zexmenia longipes) studded them. Almost always the delicate lavender flowers and halberd-shaped leaves of the “‘Churristate” (Anoda hastata) could be seen among the stronger milkweeds, agera- tums, lantanas and coarser composites. Here and there we were sure to see the long graceful panicles of Jresine paniculata, one of the Amarantacee. This plant was com- mon not only near Cartago but also at Juan Vifias, in both places growing on roadside or railway banks. On November 30, P. made a list of the conspicuous plants Zorillo real—Hamelia patens. Pavoncillo—Cestrum aurantiacum. To face p. 88 22 EES RRR, Churristate—Anoda hastata. Flower (and fruit on a smaller scale) of Guijarro—Stemmadenia bignonieflora. : To face p. 89 NATURAL HISTORY WORK IN CARTAGO 89 that were in bloom along the track east of Cartago. It included a species of Commelina; Bryophyllum calycinum; the leguminous Dalea alopecuroides, pord, Indigofera anit (vine with purple flowers), another unidentified vine with yellow flowers, Calliandra grandiflora, a creeping sensitive plant with spherical pink inflorescences and a few large paired leaflets, and a mimosaceous shrub with white flowers; an unidentified euphorb; Sida rhombifolia and Anoda hastata of the Malvacez; a species of Jussiaea; the guayabo (Psidium guayava) with ripe fruits on the ground beneath; the common milkweed; Lantana camara; Salvia polystachya; Eryngium carling; Pavoncillo (Cestrum auriantiacum) and three species of Solanum; and three composites, Ageratum conyzoides, Bidens pilosa and Zexmenia longipes—in all twenty-five species. The fence trees here carried quite a number of epiphytic bromeliads on their trunks, all of very small size, only two or three inches long. One of the commonest was Catopsis fulgens. They contained little or no water and their fauna consisted of a few earwigs, cockroaches, spiders, isopods and often ants. When the latter were abundant the others were less likely to be present. The roads that climbed up the volcano, especially the narrower and less traveled paths, were often fragrant with the perfume of the sweet pale violets, growing in profusion among the clumps of Bryophyllum, ferns and mossy stones. They were well concealed and their fragrance was usually the first indication of their presence. A very different road- side plant, preferring sunny bare places, was a showy species of Siphocampylus, growing two to three feet high with long racemose inflorescences crowded with tubular red and yellow flowers. We met it from the level of Tierra Blanca to the Reventazon River at Juan Vifias, always on sunny exposed banks. CHAPTER VI WEST AND NORTH OF CARTAGO The Continental divide in the central portion of Costa Rica is formed by the mountain or mountains called La Carpintera lying about three miles west of Cartago. The greatest altitude of these mountains is given as 5695 feet by Pittier. The railroad crosses the divide through a pass be- tween La Carpintera and Irazi which also accommodates the Carretera Nacional from Cartago to San José, the high- est point being El] Alto. To the south side of the tracks, is a flat swampy place among the hills, with some standing water filled with grasses, knot-weed, rushes, etc. It is known as the Laguna de Ochomogo and apparently occupies the floor of an old crater with very irregular and broken-down walls. It is the site of the first battle of the first civil war of Costa Rica, when, on April 5, 1823, adherents of the Mex- ican empire of Iturbide, from Cartago and Heredia, met in arms the republicans of San José and Alajuela. Padre Fran- cisco Quintana, crucifix in hand, induced the combatants to separate, but the republicans subsequently occupied Car- tago and the capital of the country, hitherto at that town, was removed to San José. In the second civil war in 1835, Ochomogo was again the scene of a Carthaginian defeat by the forces of President Braulio Carrillo. P. first visited the laguna on July 7, 1909. It ought to be a good place for dragonflies but careful search along its edges discovered only a single species, the tiny Anomalagrion has- tatum, which we have described in connection with Cartago. A strong wind was blowing from the east during most of this WEST AND NORTH OF CARTAGO gI day and the blustriness was rather disagreeable in certain exposed spots. On September 25, Professor Tristan and P. were at Ocho- mogo. The laguna is in the midst of a pasture enclosed by barbed wire fences and on entering we received permission from a man working there to go around the laguna and catch insects. We employed ourselves in this fashion from eight o'clock until noon, when the sky became very threatening and more rain was falling than previously during the day. There were a great many frogs all around the edge of the swampy area, a number of swallows skimmed swiftly over the center; once a large brown heron (?) was disturbed by our presence and we saw a few wild ducks. The fields bor- dering the pasture contained chiefly maize, some of it very high, perhaps ten to twelve feet, and in tassel. As we left the field bordering the laguna, we paused by a house and as it was raining and three hours before the next train for Cartago and no shelter at the “‘station,”’ Professor Tristan asked permission to wait at the farmhouse. This re- quest led to an exciting dialogue. The farmer at once accused us of “‘disregarding the rights of the proprietor”; when told that his brother had given us permission to pass around the laguna he admitted that, but said we had no permission to hunt ducks. Professor Tristan replied that we had not hunted ducks and that if we had any we would pay for them. The farmer retorted that we would pay for them by being shot, while Professor Tristan called out that if he shot us we would send him to San Lucas (the penal island). The farmer ran indoors and brought out a double-barreled shot-gun and came near us threateningly, Professor Tristan explaining calmly what we were doing; but the farmer dis- believed our story, saying that although we had no guns we were hunting ducks with our insect nets! We offered to open our boxes and show him that no ducks were there and th? 92 \4 YEAR OF COSTA RICAN NATURAL HISTORY man then lowered his gun and walked back disgustedly to his house without accepting our offer. He did not seem drunk; perhaps his conduct can be explained as due to igno- rance. He was an exception to all the Costa Ricans I met in his treatment of us. Professor Tristan told me that in Costa Rica everyone is free to go into a pasture or coffee field, but not into a man’s house or the enclosure surrounding it. Tres Rios or La Union, a town of 1500 inhabitants (1910), lying six miles west of Cartago, is in the midst of plantations producing some of, if not the, best Costa Rican coffee. The numerous streams in its neighborhood flow ultimately into the Pacific. On the fourth of December P. after exploring along the southern and eastern edges of Tres Rios found a pretty road which finally led high up on Car- pintera. It crossed several streams along which were num- bers of a velvety dark blue moth (4rdonea morio) with a wing-spread of one and three-eighths inches. Not having planned to climb Carpintera to-day I had no breakfast, but the weather was fine and clear and the trail fairly good so that I concluded to go up as far as the trail led and be hungry, unless I found a chance of buying tortillas at some peon’s cabin. Human residences, people and even cattle were few, however, and no food was to be had. The trail, winding, steep and in some places slippery, eventually terminated near the top of a spur lying next east of the highest peak. This spur had been cleared of trees and was in part planted in maize, but was mostly potrero which had become charral (overgrown with bushes) to such an extent that the bushes were as high or higher than one’s head; they grew close to- gether and were often bound together by vines so that go- ing through them was slow work and it was hardly possible to gain any spot from which one could have a view in all directions. From the north of this spur, a little below its top, I had a grand view to the north bounded by the moun- WEST AND NORTH OF CARTAGO 93 tains Irazi, Zurqii, Barba and Pods, terminating westward at the Aguacate Mountains while in the valley enclosed by these volcanoes I could plainly see the towns of Tres Rios, San José, Heredia and Alajuela, the last-named distant about eighteen miles. Between Irazi and Zurqii the pass of La Palma gave me a glimpse of low-lying clouds evidently overhanging the plains of Santa Clara while above the clouds was a perfectly clear horizontal line which suggested the Atlantic Ocean but must have been some sort of mirage. Crossing over this spur gave a fine view of Cartago and its surroundings, although I could not certainly make out Pa- raiso. To the west was the highest point of La Carpintera, about a mile away; it was apparently covered with forest to the top so that it seemed as if no one point there was suffi- ciently open to afford a view in all directions. There were no Alpine plants on this spur of Carpintera. I noted the familiar targtia, escobilla, guacamayo, the blue sage, a pokeberry (Phytolacca sp.) as well as the beautiful “San Miguel” (the melastome Blakea gracilis) and Monnina costaricensis. La Carpintera has many pretty spots along the trail fol- lowed, which does not pass through any old forest; small flowing streams begin high up on this spur. Descending by the way I had come until near the bottom, I found some in- sects new to me in the shape of brilliant metallic green Chrysomelid beetles, butterflies, and my first female Libel- lula foliata (a dragonfly). Then instead of returning to Tres Rios I turned off eastward hoping to reach E] Alto in time to catch a train, due about 3.30, for Cartago. I might have done so if I had not stopped to stalk (and eventually catch) a couple of dragonflies of two species (Brechmorhoga pertinax eurysema and Macrothemis pseudimitans), which I had previously taken only at lower levels. Similarly unex- pected was the sight of four green parrakeets (Conurus) 94 A YEAR OF COSTA RICAN NATURAL HISTORY which flew screaming over my head. In time I came out on the Carretera Nacional which I followed for the rest of my journey, but long before I reached El Alto I could see the train ahead of me. So I reached Weldon’s Hotel on foot about five o’clock, decidedly tired. On March 17, 1910, we started on horseback from Car- tago, along the Carretera Nacional. Before reaching Tres Rios itself we turned south into a road running toward Car- pintera. The carretera was not particularly attractive, being hot and dusty, but this side road was pretty, winding about among coffee fields and in places grassy and shady. After a pause for breakfast we rode on up until the road became so steep, rocky and slippery that we thought it in- advisable to take the horses farther and therefore tied them in a grassy place and finished the climb on foot. The road was still prettier in the upper part, where it was always shady and in places thickly overgrown. At intervals were beautiful clumps of large white begonias in flower and early fruit, with less frequent pink begonias having narrow pointed leaves. On one of the trees we saw among many other epiphytes a large and curious orchid with long narrow white petals. A pretty violet mallow with flowers over an inch across grew by the fences and there were numerous fruits we did not know at all—clusters of oval red berries and of stiff papery balloons resembling the fruits of a “‘balloon-vine,” but this plant was a shrub. The savagely thorny “Horquetilla” (Xylosma salzmanni) used in Cartago as a hedge plant, had been planted for the same purpose along this road but had been allowed to grow up into trees, pretty to look at but not always easy to “duck” under where they hung low. After climbing about an hour we crossed a gate and entered an open potrero at 5730 feet, near the top of the spur visited on December 4. (This altitude, furnished by our aneroid, WEST AND NORTH OF CARTAGO 95 is probably a little too high.) One of the reasons for making this particular trip was the magnificent view of the mountains from this potrero, for which we carried the whole photo- graphic outfit. In this we were grievously disappointed, for even when we started the mountains were not perfectly clear and we could see, in our successive glimpses, how they were misting over. By the time we reached the potrero the Aguacate Mountains had disappeared, Pods was almost invisible and Barba so faint that I could not focus on it in the camera screen, while even the pass of La Palma was dis- appointingly misty. On the return we had some fine opportunities to watch the curious scissor-tail fly-catchers (Muscivora forficata) which perched on the telegraph wires, and made short erratic flights after insects. Both perching and flying, their ex- travagantly long forked tails were opened and shut con- stantly with a jerky motion and also moved up and down in the ordinary fashion of birds in balancing. Through the western part of Cartago flows a small stream, the Reventado, which arises high up on Irazt in the old crater or Laguna del Reventado, and eventually empties into the Agua Caliente River in the bottom of the valley. For much of its course it is a rapid, rocky, brawling little stream, its bed full of boulders. The part near and above Cartago we always called the ‘“‘Paltothemis River” because the showy, swift-flying Paltothemis lineatipes was very common there and seemed to rule the stream. This hand- some dragonfly has a wing-spread of three and a half inches and a conspicuous red body one and seven-eighths inches long, and the habit of each individual’s patrolling a certain limited bit of territory was well marked and easily observed. It was often amusing to see how quickly an interloper was attacked and driven off by the dragonfly in possession. A smaller dragonfly, Heterina cruentata, was also abundant 96 4 YEAR OF COSTA RICAN NATURAL HISTORY here and we obtained its larve clinging to the submerged Stems of plants growing on the bank. We visited many different parts of the Reventado, par- ticularly above the town on the slopes of Irazi. As the stream had cut little ravines and gorges for itself in many places, it was not always possible to keep at the water’s edge so that sometimes we followed its course at the top of the ravine, sometimes on its banks. In one of the potreros above the Reventado we took Gomphoides ambigua on De- cember 15; this is a Gomphine dragonfly not previously known from Costa Rica, although we also obtained it from San José and from the peninsula of Nicoya. The little river was often heavily shaded with big trees, with undergrowth of ferns and caladiums, tradescantias and mosses. A partic- ularly charming bit was evidently the bed of an extremely old crater, roughly elliptical in shape with high steep walls except on the south side, and containing many fine large trees. The current here was swift and the stream was poor in dragonflies. On May 24 A. obtained a stone-fly and two stone-fly exuviz, small may-fly larve, a couple of species of caddis-worms, a galgulid (‘‘toad-shaped bug’’)—all similar in general appearance to forms in the eastern United States. On the same day some Ithomiine butterflies, Hymenitis ‘oto (with clear transparent wings spreading two and one- eighth inches, whose margins and veins are blackish, and with an oblique milky-white band and milky-white spots ‘on the front wings) and Dircenna klugii (already mentioned in Chapter V) were found only in dark shady spots in these woods. The transparency of their wings made them much more difficult to see, whether in flight or at rest, than other ‘butterflies whose wings were of the normal style—colored and not transparent. Dr. Sharp, in the Cambridge Natural History (Vol. VI, p. 346), quotes Wallace briefly as making a :similar observation on these butterflies. However on May ‘a[eUl “ooIM. vEULayT . ‘ a[eWloa} ‘orury Ijaz sniuooya zy ‘a[BUl ‘u0DIL0% Saplany We x) Solpdeqing onewiyy ‘aeul ‘5170384910 DaLOIKT ‘apeulay ‘snssksop situnyoa yy ‘apeulay ‘soutxoug viydsomsiqT Tofacej < WEST AND NORTH OF CARTAGO 97 29, while Ithomiines (Jthomia heraldica, see page 79, Dircenna klugi, Episcada salvinia, very similar to E. apuleia described on page 79 but a little larger, and Mechanitis doryssus) were taken in shady lanes near this part of the stream, a male of D. klugii was caught in bright sunshine and two males and a female of this species were taken in bright sunlight at Cartago on August 21. Ithomiines (the species not recorded in our notes nor represented in our specimens) were seen flying in bright sunshine both in the morning (11 A. M.) and about 1 P. M. of May 30, in the same general locality, and in bright sunshine on a road above Tierra Blanca on July 12. In this last case the country was open for many rods on both sides of the road and the wind was blowing strongly, which may account for this occurrence. Of D. klugit we have four specimens, three males and one female, expressly labeled as having been taken in bright sunlight and four (three males, one female) labeled as having been caught in the shade; on placing these in two parallel rows for comparison, I could find no constant difference be- tween the two lots. We also have two males of [thomia hippocrenis, a clear transparent-winged species with mark- ings similar to those of Hymenitis oto, taken in sunshine, on flowers, in a new road in the forest at Guacimo, on June 7; our other specimens of hippocrenis from Guacimo and Guapiles, are labeled “‘Forest,”’ with no mention of shade or sunlight. On December 4, on La Carpintera, the clear- winged Episcada apuleia and salvinia were observed flying both in shadow and in bright sunlight. We may conclude that while the Ithomiine butterflies are chiefly to be found in the shade they are not restricted to that condition. Three of the species of butterflies collected in the Reven- tado region above Cartago on May 29—Lycorea atergatis, Mechanitis doryssus, and Melin@a imitata—are so exceed- ingly alike in colors and color-pattern, and not so unlike in 98 A YEAR OF COSTA RICAN NATURAL HISTORY size or in shape of wings, as to furnish a very typical case of mimicry. Messrs. Godman and Salvin in their volumes on the butterflies in the Biologia Centrali-Americana mention four other species namely: Eresia mechanitis, Eueides zorcaon, Heliconius telchinia and the female of Dismorphia praxinoe, as having colors and patterns very similar to the first three mentioned. The last four also occur in the same localities with the first three and although we did not meet them to- gether some of our readers may make their acquaintance in nature and wish some means of distinguishing them apart, for it will be noted that not merely seven species but seven genera are involved in this case of resemblance. One of these genera, Dismorphia, has the usual number of functional legs of insects, six, and each of the two claws at the tip of each leg split lengthwise or toothed, features which place it in the family Pieride. The other six genera have only four legs used for walking, the first pair being unfitted for this action, and in consequence belong to the family Nymphalide. This is divided into several sub- families, of which we are here concerned only with the Dan- aine (Lycorea), Ithomiine (Mechanitis, Melinea), Heli- conine (Heliconius, Eueides) and Nymphaline (Eresia). Danainez, Ithomiine and Heliconine have the central basal area of the hind wings known as “‘the cell’ enclosed on all sides, while in many Nymphaline, including Eresta, the cell is open at its outer end. The Danaine and the Ithomiine have the last (hindmost) vein of the front wing apparently formed by two veins uniting near the wing-base; in the Heliconine on the other hand this same vein arises singly. One distinction between Danaine and Ithomiine has already been mentioned on page 79, the males of the former having two retractible tufts of hairs at the hind end of the abdomen, but not along the front edge of the hind wings, while the males of the Ithomiine lack the abdominal WEST AND NORTH OF CARTAGO 99 tufts but have a patch of long hairs along this wing edge. The Danaine females have the last section (tarsus) of the first pair of legs more shortened than in the Ithomiine fe- males. The two Ithomiine genera mentioned, Mechanitis and Melinea, differ in a number of structural details of which the most readily seen, is that the first two veins (costal and subcostal) of the hind wings are united, or very close to- gether, in the basal half of their course in Mechanitis, while these two veins are widely separated, except at the extreme base, in Melinga. Finally the two Heliconine genera con- cerned differ from each other in that Heliconius has each antenna more than half the length of the front edge of the front wing, with only a slight club-shaped thickening at the tip, while Eweides has each antenna less than half the length of the front edge of the front wing and ending in a distinct club. As differences of secondary importance but more easily seen, Melinea, Mechanitis and Lycorea have the cell of the hind wing longer than half of the wing, while Hel1- contus and Eueides have it shorter than half of the hind wing. By way of summary, the relationships and classification of these seven butterflies may be shown thus: Family Pieridz Dismorphia praxinoe Subfamily Danainz Lycorea atergatis : vs Melinga imitata ly Ith we Family SEs tail anne: Mechanitis doryssus Nymphalide Subfamily Heliconine Pus iene Subfamily Nymphaline Eresta mechanitis While there is an undoubtedly close external resemblance between all these seven butterflies, it is quite another ques- tion whether this (unconscious) mimicry, to use the technical term by which it is known to zodlogists, is of importance as a means of protection to any of the species concerned. 100) =6 YEAR OF COSTA RICAN NATURAL HISTORY As stated on page 79 the male Lycorea atergatis can emit an odor like cedar shavings, but we were unable to make any observations tending to show that this odor, or the resem- blance of the other butterflies to the odoriferous Lycorea, has a defensive value against birds or other animals. Over these slopes of Irazti,—as all around Cartago—in- deed over all Costa Rica down to sea-level—we frequently saw that commonest of butterflies, Anartia fatima. It is a Nymphaline and has a wing-spread of one and seven- eighths inches. Its wings are dark brown above, darker at the tips than at the bases. The front pair have a cross-band of yellow at about mid-length, three yellow spots in a row at four-fifths’ way to tip and two smaller yellow spots between the hindmost of the first three spots and the yellow cross- band. The hind wings have a yellow band of nearly the same width as that on the front wings and almost in continu- ation of it, but lying much nearer to the outer edge of the wing; a short distance on the basal side of this band are four red spots. The edges of all the wings have narrow whitish scallops here and there. The undersides of the wings are a much paler brown, the yellows and reds of the upper surface are repeated and are of greater extent but paler. Professor Tristan has found that it is the most frequent butterfly visitor of the common milkweed (Asclepias curassavica) and the chief agent in its pollination. The pollen masses are carried from inflorescence to inflorescence attached to the claws of the insect’s feet. The whole southern side of the great mass of Irazi below the forest was covered with pastures or big fields of maize or potatoes, each enclosed in a loosely built stone wall topped by a strand or two of barbed wire. The maize harvest was all over by March. On March 15, 1910, we rode up Irazi toward the village of Llano Grande, going up to 7000 feet. Everywhere, as high up as we could see, the corn was cut WEST AND NORTH OF CARTAGO IOL and the ears piled up ready for the ox-carts or else the field was already plowed or plowing. There were some small collections of houses here and there on the mountain as well as the villages of Cot, Tierra Blanca and Llano Grande. An intricate network of lanes and roads followed the outlines of the potreros and we fre- quently explored the smaller byways. Well-traveled roads led to the villages, the one to Tierra Blanca in particular continuing as an excellent cartroad all the way up to the Jimenez Lecheria (at gooo feet) and thence toward the east, and these main roads kept their own direction so that one knew where one would arrive. But the smaller lanes—often shaded so that one did not see readily down the mountain— had a habit of running up hill then turning and running down again and it was impossible to tell without trial whether any lane would lead i in the desired direction or not. They were delightful places for walking, nevertheless. On a bright, breezy morning in May many species of butterflies flew or were blown into or across these lanes. Scarabeid beetles were abundant, at least as individuals. We saw a great buzzing metallic green thing suggesting Allorhina, a species looking like our own rose-bettle (Macrodactylus) and one or two others of the same size or a little larger, all but the first-named falling on one’s hat or clothing almost con- stantly. The rose-beetle-like species were also taken at Cartago on June 17 and July 5. Compared with our rose-beetle of the eastern United States (Macrodactylus subspinosus) they are slightly larger, being about three-eighths of an inch long, the head and thorax a more vivid green which is also metallic in its reflections, while the wing-covers are a deeper and redder brown. In one species (M. suavis) each elytron bears two distinct rows of very short whitish hairs similar to those which cover much of the upper surface and the lower sur- 102) A YEAR OF COSTA RICAN NATURAL HISTORY face of the entire body. This species has the tarsi of all the legs and the apices of the hinder four tibiz black, the rest of the legs red. The other species (M. sericinus) lacks these hairs on the wing-covers; instead a little more than the rear half of these organs is metallic green; the tarsi of all the legs and all of the hinder four tibie are black. In both species the females may be distinguished by having a slender proc- ess on the prosternum just behind the first coxe and by having one or two pale bristles (in addition to the numerous short whitish hairs) on each side of the ventral surface of the abdominal segments. In April, small blue butterflies (Lycena isola) were very abundant on the eastern road to Tierra Blanca at an eleva- tion of 5000 to 6000 feet, particularly over damp spots and on horse-droppings. This species, which has a wing-spread of three-quarter inch, has been found as far north as IIli- nois. Below Cartago the Reventado ran through coffee fields too thoroughly cultivated for the stream to have much in- terest for us or was simply an unshaded brook in open pas- tures. San Rafael was a good-sized village northeast of Cartago and practically continuous with it. North of San Rafael was the berry farm of Juan Rudin, a Swiss who had been ‘in Costa Rica some twenty years. He had about two man- zanas of beautiful fertile land in strawberries, the finest berries and most perfect plants imaginable. We found this farm on March 29, Ig10, after some inquiries, one of the things by which we were to recognize it from the road being a large and beautiful rosebush in the patio. After a little preliminary conversation under the rose tree—which fully justified its reputation in this land of fine roses, for it grew ten feet high, spread out in both directions the whole length of the house and was covered with magnificent pinkish- WEST AND NORTH OF CARTAGO 103 white roses—we were conducted into the fields and made free of the berries. Mr. Rudin seemed much gratified at our delight and indeed it would be hard to find a prettier field. The berries were in perfect order, the rows as straight and plants as clean as if just arranged for a prize contest. Each plant was about a foot across, covered with tiny buds, open flowers, young and ripe fruit. They were allowed to bear fruit for ten months in the year only, otherwise they would exhaust themselves. The berries were large, sweet, juicy and tender. At first we felt shy about eating the fruit but Mr. Rudin’s urging soon induced us to try a few and the more we picked the more clumps of berries we saw at a short distance that called to us. So we wandered over the field eating strawberries as we had never dreamed of doing, and whenever we stopped picking for a moment Mr. Rudin gathered a handful on their long stems and urged them upon us. Here and there among the berries were tall rose trees, seven to eight feet high, with large, not very double shell-pink roses of exquisite shape, color, texture and perfume. We gathered a few, which caused Mr. Rudin to send for his shears and cut a huge bunch of roses and carna- tions, for he also had some rows of these flowers. Orange trees were full of flowers and ripe fruit, and on a little sunny bank were masses of long-stemmed sweet violets. When we had eaten all we thought we could we moved out toward the gate, where we untied our horses and pre- pared to take some pictures before we left. We coaxed the babies—five little children all remarkably pretty—into po- sition, photographed them and then started to say good-by. But no. We must sit down in the corridor a minute. So we were conducted to a table in the corner of the corridor and there beheld a huge dish of the biggest and finest straw- berries—and we were expected to eat them all! We had only intended to make a short stay at the berry- 104 A YEAR OF COSTA RICAN NATURAL HISTORY farm but we were there for two hours and it was almost breakfast time when we rode away. However no one wanted breakfast after our strawberry feast so we sent the flowers and a message home with a boy and started on our delayed ride. From San Rafael we went east along a wind- ing, shady road until we came in sight of a neat-looking two- story house whose inhabitants were all out on an upper bal- cony watching us. The man of the family seemed to think we were coming to call—which had not been our intention— and came down the road to meet us, so we rode in and tried to buy maize for the horses. He fed the whole five (we were riding with a party from the hotel), gave us milk and bread and would not consider our paying for the refreshment as we had hoped to do. He told me the place was called ‘‘Mata- mora” and that he and his family were spending a few months there, Cartago being their home. From Matamora we rode on, the road quite boggy at times but always pretty and shady, until we issued on a better traveled road and turning to the right soon came in sight of Paraiso, quite a little below us, and then returned to Cartago by the main road from Cot through San Rafael. To anyone looking up the huge slopes of Irazti from Car- tago, the village of Tierra Blanca, and the light or whitish clay from which it took its name, were conspicuous land- marks. ‘Tierra Blanca was over 2000 feet higher than Car- tago and the several roads leading up to it vied with each other in difficulty; the eastern road in particular was a miracle of steepness and slipperiness. ‘This village was a sort of halfway house to those making the volcano trip— though it was by no means halfway up. We passed through it a number of times and also made several excursions with Tierra Blanca as the objective point, while in July, 1909, P. spent some days in the Telegraph Office as the guest of Senior Federico Tristan, who was then acting telegraph oper- WEST AND NORTH OF CARTAGO 105 ator there. On the tenth of July, as we sat on the steps of the office, we saw many individuals of the migrating butter- fly Timetes chiron. ‘They were flying from north to south and consequently down the mountain, and although we made no actual count it seemed a case of true migration. On July 12 P. took up his quarters in the office, which ac- cording to his aneroid was situated at an elevation of about 6875 feet, consequently about 2100 feet higher than Cartago. It was a small one-story frame structure divided into a nar- row passage leading to the little window where telegrams and postal matter could be received and distributed (for the post office was also located here) and four rooms communi- cating with each other, while a fifth compartment was at this time devoted to ducks and chickens. The building was raised a couple of feet above the ground but this space below was not enclosed. Each room had one window; the sashes in the window slid up and down in a country where most windows are casements, but had no sash cords and the upper sash was fast. My bed was a canvas stretcher cov- ered with a straw matting. Sefor Tristan lent me a pillow but had warned me to bring covers for my cot as he had only his own, but the wind blew hard those nights and so much came through the cracks between the boards of the walls and floor that it was decidedly chilly. Besides the ducks in the fifth room there was a recently hatched duckling and its ‘‘hatcher,” an elderly hen, who spent the nights in Senor Tristan’s bedroom and had the run of the four rooms during the twenty-four hours if they so desired, for there were no doors to separate the rooms from each other. It was just as well, therefore, that there were no floor coverings, other- wise hen and duckling would not have improved them. Tierra Blanca at this time had 137 males over twenty-one years of age and its total population was about 500. Most of its houses were on the eastern road from Cartago, which 106 4 YEAR OF COSTA RICAN NATURAL HISTORY was here so steep that the highest house was fully 500 feet above the lowest. At one of the houses on this road, that of Eduviges and Ramona Brenes, we had our meals. It was situated on a raised bank or terrace on the east side of the road, the terrace having a stone retaining wall bor- dered at one end with clumps of flags (/ris) and purple lilies. This terrace was nearly level on top, the bare earth hardened and flattened by long use. On this terrace, at perhaps twenty feet from the retaining wall, stood the two- roomed, one-storied house with thick whitewashed adobe walls. It had the usual tile roof, gabled at each short side and the eaves projected over the front or west wall of the house nearly four feet. At one end under the projecting roof were two cement sinks fitted with brass spigots and supplied with water piped (in galvanized iron pipes) from the mountain above. Each of the two rooms had one door but no windows and no ceiling, for, looking up when inside, one saw the slanting roof beams which supported the tile roof. The floor was the same hard beaten earth as the ter- race and on the same level. The room nearest the sink was the kitchen; it had no chimney—these houses rarely do. The smoke finds its way out by such holes, accidental or otherwise, as may exist near the roof and incidentally filling the room and blackening the interior. All the cooking was done with wood of course. I saw an iron coffee mill, used not only for grinding coffee but also maize for the making of tortillas. ‘The darkness, the smoke and my own disinclination to intrude, however, prevented my seeing many details of this room, besides which the kitchen was more or less occupied by the persons of Sefiora Brenes herself—a large motherly woman—others of the family and one or more female neighbors. The other room was bedroom and everything else, fur- nished with beds and narrow, backless wooden benches. On WEST AND NORTH OF CARTAGO 107 the whitewashed wall of this room, opposite the door, a large picture frame was outlined in red and blue; in front of this stood a table holding several pictures of saints, including two which evidently had the Immaculate Conception of Murillo as their foundation. Our meals were served on a table placed in front of the house under the projecting eaves and covered with a muslin tablecloth. We had white china dishes, except that some of the coffee cups were colored or bore inscriptions, and the labels on the knives indicated German manufacture. The food was good, clean and well-cooked; for my four days’ board, twelve meals, I paid 3 colones, 90 centavos ($1.80). At seven or so we had “‘café,” consisting mainly of sweet- ened coffee (that is, boiled with dulce) without milk, and tortillas, but in addition I had a couple of “huevos paredis por agua”’—soft boiled eggs. At eleven was “‘breakfast”’ —rice, black beans, a combination of eggs with some vege- table such as potatoes or sapayo (one of the squash-cucum- ber family), boiled or stewed beef, boiled potatoes, yuca, chayote, plantain, etc., finishing with coffee. Dinner at five was much the same except that it began with soup and the coffee was preceded by a “dulce,” usually in the form of stewed fruit, such as peaches. Peaches are grown to a considerable extent around Tierra Blanca and brought for sale to the market at Cartago, but always green or at least more or less unripe. Professor Pittier says in his recent book on the common plants of this country that it is rare to see a ripe peach, the people seeming to prefer them unripe. There is, moreover, great difficulty in ripening them on the trees owing to the fondness for them shown by birds and by wasps and other insects. Even when not destroyed by wasps the fruit rots on the trees when ripening, which seems to indicate that the peach is not adapted to this moist climate. 108 A YEAR OF COSTA RICAN NATURAL HISTORY Tortillas and bread, the latter not home-made but bought at the store, were also served at breakfast and at dinner. The bread is almost always in the form of narrow loaves of varying size and shape but pointed at both ends, more like elongated rolls with hard crust on all sides. Once at the Brenes’ we had a sort of corn bread, at least Sefior Tristan said it was made from maize; it was thicker than the tor- tillas, not yellow, so apparently it contained no eggs and rather heavy—I did not care for it. Butter there was none, but a substitute was often present in a fresh white cheese resembling Neufchatel, in the form of a fairly solid cake with a little whey. It was unsalted, but a piece of it on a tortilla with a little pinch of salt improved the tortilla very much in my opinion. . The most novel feature of our meals at the Brenes’ was the animal surroundings. The beasts belonging to the estab- lishment consisted of two cows, each with a frolicsome calf, a young bull, a couple of young pigs, a dog (and another which usually came from across the road) a cat, three ducks and numerous chickens. All of these gamboled or strayed on the terrace under the projecting roof and not infrequently into the rooms of the house. Usually when we were eating, some one was continually employed in chasing the live stock away. ‘The two cows were not present often but once they were milked under the eaves just previous to our meal. Two larger pigs were doubtless only prevented from joining our party around the table by being too big to push between the barbed wires which bounded their pen back of the house. A priest and his brother took meals at the same table for part of my stay, and I havea distinct recollection of one even- ing, when the priest sat at one end of the table while eight inches from the back of his chair were the flanks of the young bull whose head was through the kitchen doorway. There was so much mud in Tierra Blanca owing to the WEST AND NORTH OF CARTAGO 109 frequent rains, the sticky nature of the soil and the degree to which it was washed on account of the steepness of the roads, that it was no wonder that the clothes of the inhabi- tants were usually stained. The existence of the cement sinks at the Brenes’ produced an improvement over the usual method of washing clothes, for here it was done in one of the sinks while ordinarily it was done on stones on the banks of some stream irrespective of the cleanliness (or muddiness) of the water thereof. All washing of clothes was of course in cold water only. Senor Tristan had a broom of the usual civilized type in the telegraph office, but at the Brenes’ a bundle of leafy twigs did duty for sweeping, an operation which was usually performed by the man of the house during my presence there. After dinner we returned to the office until seven o’clock, then went back to the Brenes’ where Sefior Tristan had a phonograph and for two hours American, Spanish and other songs were produced in the sleeping room for the gratification of the family and a few neighbors. Sefiora Brenes joined us later and indulged in her cigarette, sitting on the edge of the bed in which the little girl was sleeping. The illumina- tion was furnished by a single candle. The phonograph provided Sefor Tristan’s usual evening amusement and must have added greatly to the pleasures of these people. The road which led to the top of the village went on to the craters of Irazi. The highest house was 7150 feet above sea-level. Looking down on the village gave a picturesque view of the red roofs and white walls of the houses nestling among the green vegetation, while beyond was the valley of Cartago bounded on the south by range after range of moun- tains. From this point, on some days, San José was also visible. After climbing the Tierra Blanca hill, the road keeps nearly to a level for a couple of miles, dividing then 110. =A YEAR OF COSTA RICAN NATURAL HISTORY into two forks, the right going to the highest point of the volcano eventually while the left led to the lower western parts of the mountain. The land on each side of the road was all open, either as pastures or fields of maize; there was con- sequently nothing to obstruct the view and one could see far in nearly all directions. It is a beautiful country and would be well worth a visit of several weeks. A number of bright-colored butterflies were flying about this road on July 12. At places were many guitite trees with clusters of tiny pale greenish-white flowers along the smaller branches. These flowers were very attractive to bees, butterflies and some day-flying moths. Among the latter was a handsome species (Uranidia fulgens) with wings shaped as in the Papilio butterflies and long tails to the hind pair of wings. The wings are black crossed by bars of me- tallic green, and the tails just mentioned are edged with white near their tips. Each front wing is one and one-half inches long, each hind one, including the half-inch tail, two inches. ‘The appearance and manner of flight of this moth are so butterfly-like that for long after we came to Cartago I was deceived by it and thought that some of them must be butterflies. There were also humming-birds of several species around these guitites, doubtless feasting on insects attracted by the flowers. Sweet violets, purple and mixed purple and white, were numerous on the roadside banks. In hedges between fields and along the road near Tierra Blanca were a number of shrubs fifteen to twenty feet high, having rather dull green, cordate leaves about six inches long and in April bearing large clusters of soft orange berries one- eighth inch in diameter. The plant was called “‘Ortiga” in Tierra Blanca, probably the ortiga referred to Urera cara- casana by Pittier. By following the left or western fork of the road one reached the Rio Reventado, at that point a small stream WEST AND NORTH OF CARTAGO III flowing through a grove of oaks (encinos) and other trees. In fact this place was the lower limit of the oak forest here, of which more hereafter, but the oaks must have reached still lower down in former years for on another day I no- ticed a solitary oak in a field between Tierra Blanca and Cot at 6800 feet. This oak has leaves two to three inches long and with entire edges, but the acorns are one and a quarter inches long in some cases. Associated with the oaks here at the Rio Reventado were at least two species of the characteristic tropical American family, the Melastomes —one a small-leaved tree (Conostegia) ten to fifteen feet high, the other an herb (Miconia) with larger and softer leaves. Grasshoppers of familiar forms and colors (Chortophaga meridionalis, Schistocerca zapoteca, Dichroplus morosus) were not rare along the road on July 12, but nowhere could I find a single dragonfly, not even at the apparently favor- able crossing of the Reventado. Beyond this crossing, the road ascends steeply for some miles, winding around the hills which form the mountain slope. Indian graves have been found in this vicinity and at Sabanilla, where there are two or three small houses. Sefior Tristan pointed out a hole in the road-side bank from which two gold ornaments had been taken in the construction of the road. On July 13 I went a mile beyond Sabanilla to look for the Laguna del Reventado or Laguna del Dirumbo. Some bars were stretched across the end of the road and climbing over these I entered a field with trees and bushes, ascended to a ridge marked with tall oaks and then descended into a flat-bottomed valley between two great divisions of the mountain (Iraza). This valley is an extinct crater of Irazi, although the Danish naturalist Oersted regarded it as a dis- tinct volcano, that of Reventado. Even from Cartago one can see an immense bare place on the southwestern side of the highest mass of Irazi, and now that I was in the val- 112. A YEAR OF COSTA RICAN NATURAL HISTORY ley I was almost at the very foot of this huge slide—an al- most vertical face of yellow rock perhaps three hundred feet high and a hundred wide, not reaching as far down as the valley floor. On one side of the floor was the Laguna del Dirumbo (also called Laguna del Reventado and Laguna de los Derrumbaderos), the source of the little Rio Reven- tado. The laguna was about eighty feet wide and one hun- dred feet long, of clear cold shallow water. Here and there beds of grass dotted its area, and its undefined and swampy sides were bordered by trees and bushes. Around the la- guna were many oaks both alive and dead, the whitened trunks of the latter giving a curious and mysterious effect, and among the oaks some clumps of bamboo and much grass. Below the laguna cattle were feeding, the part oc- cupied by them being separated by a barrier or fence of dead logs and branches. There were several smaller boggy streams in addition to the Reventado and a second smaller swampy spot. The altitude of this place was 9100 feet by my aneroid. There were very few insects in this valley and to my astonishment and disappointment not a dragonfly could I see. In July, 1906, I had a similar experience at Lake Moraine on Pike’s Peak in Colorado, at a not very different elevation. Indeed the highest altitude at which we personally met dragonflies in Costa Rica was below Tierra Blanca at about 6450 feet. Prof. J. Fidel Tristan visited the Laguna del Reventado in July, 1910, and soon after wrote me that he had seen “three big dragonflies flying on the laguna. They were of a dark color and I tried to catch them but my efforts were not enough.” In April, 1911, he was more successful and took a single specimen in a rivulet near the laguna. This specimen, a male of Argia terira, a small slender blue and black species, furnishes the present record for the highest altitude for Odonata in Central America. *DIDLINIJOBUL 018]0$ “ds snjhduvsoydrg To face Stone Image from an Indian Grave near Sabanilla, one-half actual size. To face b n> WEST AND NORTH OF CARTAGO 113 On one of his later visits (in July, 1913) to Sabanilla and the Laguna del Reventado, Prof. Tristan found on a rotten tree a dead specimen of the brilliant silvery-gold beetle, Plusiotis batest, which he sent tous. This resplendent insect, one and three-sixteenths inches long, is a member of a genus of Lamellicorn beetles “highly characteristic of the Central American region” and always attracting attention on ac- count of their brilliant metallic colors. We did not meet it in our own experiences in Costa Rica. Formerly the Laguna del Dirumbo was much larger, but in October, 1891, the barrier which held back the water gave way and a great flood poured down the Reventado affecting even Cartago; no lives were lost as the sound of the ap- proaching water was heard in time to give warning to people living near the river. A larger, prehistoric flood is supposed to have broken down the south wall of the crater to its present level, and to have carried and scattered on its course the numerous boulders and rocks now so striking a feature of the landscape around Cartago. Romanes, however, who studied the Cartago region in 1910, thinks that the boulders of lava, especially to the east of the town, represent a single lava flow which has weathered out in place and that this lava flow once formed a dam which held back a large but shallow Cartago Lake. The sun was shining while I was in the valley of the laguna (11 A. M.) but every quarter hour or so a light cloud swept down from over the summit of the mountain. These clouds increased in density and as I went back to Sabanilla it rained at intervals. After looking in a nearby potato field to see if the plants were attacked by potato “bugs” (Leptinotarsa) and finding none, I walked back to Tierra Blanca, much of the way through rain which continued through the afternoon. In April, 1910, M. Jaime Lacroix, a Frenchman who had lived in Costa Rica for some years and made a business of 114 A YEAR OF COSTA RICAN NATURAL HISTORY digging for Indian relics, brought us a stone image, a piece of pottery and a small jade celt from graves which he had recently excavated near Sabanilla. The image is of a stand- ing man, broken off above the knees, the lower parts of the legs having evidently been lost long ago and not in the recent excavation. The head bears a sort of double-crowned head- dress and has a muzzle like a bat’s. The arms are akimbo, the hands being placed on the front of the belly and between _ them is an ornament represented as suspended from a cord which passes around the neck. Around the hips is carved some sort of garment, possibly feather-work. The piece of pottery is a shallow circular dish supported on three legs, each of which probably represents a tapir’s head, and is hollow; a small loose ball of clay inside makes each leg a rattle. The bowl of the dish is decorated in a crude symmet- rical pattern of red and yellow and has a single animal’s head in high relief on the outside. Cot, a village a few miles to the southeastward of Tierra Blanca, was connected with it by a single strand of telegraph wire which furnished an easily followed guide. Going this way on July 14, I came to a promising-looking small stream, which I explored for dragonflies but without success, ob- taining only a few little Diptera. A little farther on, the road © was bordered with flowering guitites and here too were many humming-birds. Across the fields to the northeast was a large house of a style of architecture very un-Costa Rican, having a tall round tower at one end and some other eleva- tion at the other. It was the residence and dairy farm of Don Leopoldo Peralta; the house had a fine situation and must command a beautiful view. The sky was very threat- ening on this day and relatively few insects were flying, but among them was a handsome yellow and reddish-brown bumble bee (Bombus ephippiatus); it was called here el rey de las abejas, king of the bees. WEST AND NORTH OF CARTAGO II5 In October we visited Cot from Cartago, a distance of seven miles. It is a straggling village except the streets bordering the grassy plaza. The church had no tower, but under a central peak in the roof were two niches in which the church bells hung. ‘This village is an old one, going back to Indian times, while Tierra Blanca is modern, but there is nothing in the outward appearance of Cot to indicate any greater age than belongs to its neighbor. The roads near Cot command extensive views of the Reventazén Valley. Half a mile east of Cot is a fine waterfall about one hundred feet high, locally called “Ojo de agua” (eye of water) as waterfalls and springs commonly are termed in Costa Rica. The stream is not usually large but the drop is almost sheer and the whole place very pretty. There must be a great volume of water immediately after heavy rains, as the pool at the bottom showed that it had been much larger a few hours before our visit. About the waterfall was much red: earthy rock, a volcanic (andesite) ash, discolored by weather- ing, containing plagioclase, hornblende, glass, iron hydrox- ide and clay. The vegetation around the fall is quite differ- ent from that of the Juan Vifias waterfalls. The mosses and Selaginellas so characteristic of the latter were absent from Cot and instead there were many Equisetums, comme- linas (C. virginica), Browallia demissa, Cupheas (C. in- fundibulum, C. wrightii) and phytolaccas (P. octandra). The fall was shaded by a handsome tree-composite with bright yellow flowers. Along the roadsides even below Cot we saw the pretty Geranium mexicanum, which we found to be so common in the cinders of the crater of Irazi but it was much rarer at this lower level. CHAPTER VII THE VOLCANO IRAZU The volcano of Iraza, highest of the four great volcanoes (Turrialba, Irazi, Barba and Pods) of the central plateau, and, with the exception of Buena Vista and Chirripo Grande, the highest mountain in Costa Rica, reaches an altitude of 11,322 feet (Pittier). It is a huge, “sprawling” mountain with gradually sloping sides and an exceedingly irregular skyline, and with no typical volcanic cone visible from Cartago. The volcano dominated the country and the land- scape from San José to Juan Vifias, and a year’s residence at its foot caused us to share the mingled love and awe with which the Cartaginenses regarded “E] Volcan.” The south- ern slopes, up to 9050 feet, were in 1910 covered with maize and potato fields or cattle farms. Above that level—alas! the line is constantly rising!—are the remnants of the vast oak forests that formerly covered the whole mountain. Above the oaks again are the cinders of the volcano proper. The center of volcanic activity has moved steadily from south to north over the crest of the mountain and there are a number of abandoned, worn-out craters on the south side. A very well-marked one is that occupied by the Laguna del Reventado, described in the preceding chapter. The last great eruption of Irazi, leaving out of account the steam from the solfataras on the north side, was in 1723. A contemporary writer, Don Diego de la Haya, the governor of Cartago, has an account of it which has been reprinted by Don Cleto Gonzalez-Viquez. The eruption began on the sixteenth of February, with feathery clouds of smoke THE VOLCANO IRAZU 117 over the summit, and continued intermittently until Decem- ber of the same year. The smoke was followed by rumblings, flames were visible at night and there were many eruptions of ashes, stones and other burning particles, covering the volcano and its surroundings. The ashes fell upon the streets, roofs and patios of Cartago, the fields and trees of the vi- cinity and, at times, the finer particles, entering the eyes, nostrils and mouths of the inhabitants, caused sneezing and coughing. On March 3 the governor sent an expedition of ten persons to visit the summit—the first recorded ascent of the mountain. They reported that in the midst of an opening having almost two leagues’ circumference was a fire like a pan of tar and “that it ejected from time to time ashes, sand and small stones, the quantity of these that had fallen on all the neighborhood, as well as of larger rocks, be- ing so great that they could load a hundred ships of large size.” In the absence of any exact statement as to the size of these “‘large rocks”’ it is not possible to determine whether any of the boulders that now litter the slopes of Irazi were ejected in 1723 or (if due to any eruption at all) are to be ascribed to an earlier, pre-Spanish eruption. De la Haya’s account mentions the masses of ashes that sifted over the city, while a bombardment of such boulders as these would inevitably have crushed the houses and could hardly have failed of special mention. There was a series of earthquakes accompanying this eruption, the most violent lasting from September, 1723, to February, 1724, although they did not cause any very great damage. The ascent of Irazti offered no particular difficulty, as it was possible to ride the entire distance, but it was a fairly long and expensive trip if one wished to do more than make a flying visit of a few hours, and the mountain was so fre- quently cloud-capped that we did not attempt it as often 118 A YEAR OF COSTA RICAN NATURAL HISTORY as we would have liked. In July, while staying in Tierra Blanca, P. went up on foot, but the day proved so misty that not only was there no view from the summit, but he was really unable to make out more than a few features of the crater. He only found the trail by accidentally meeting a youth who was hunting a strayed mule and was able to guide him, and the mist was so dense that it would have been too dangerous to wander far from the trail. Thanks to this chance encounter P. saw the edge of a deep daughter-crater yawning in the floor of the largest crater and learned that there were many such pits, but he could not see more than a few feet ahead and therefore returned to Tierra Blanca after waiting awhile for the unlikely chance of the mist’s blowing out of the crater. While returning over the crater and the cinder cone, the only guide through the mist was his own footsteps in the damp cinders, but he was able to follow these without much difficulty to the path in the forest. On September 22 we set out for the crater on horseback. Although in the midst of the rainy season, which is not usually a favorable time, the summit had been beautifully clear for several days so that we felt encouraged to make the trip then. Starting from Cartago about seven in the morn- ing, we reached Tierra Blanca at nine, enjoying brilliant sunshine and beautiful views. After the customary baiting of the horses there we pushed on to a point some six hundred © feet above the village, where a long stretch of relatively level road follows the terrifically steep hills of Tierra Blanca, and where the road is joined by another ascending from the valley of Cartago but east of Tierra Blanca. This latter was usually known as the ‘‘milk road” because it was the one followed each morning by the train of pack-mules from the Jiménez Lecheria, and although considerably longer it was far easier to travel, as it skirted the Tierra Blanca hills. We ate our breakfast here, under some roadside guitite trees THE VOLCANO IRAZU 11g laden with fruit, and then resumed the journey. Our first plan had been to visit the crater twice on this trip, going up directly from Cartago so that we would have the afternoon at the summit, coming down to the Lecheria for the night, revisiting the crater at sunrise and then returning to Cartago in the afternoon of the second day. The Lecheria was the only place where one could stay so long with horses, as the last water and pasture were 1500 feet and several miles below the crater. The plan would have been quite feasible if we had had fair weather, but by breakfast time the clouds were gathering rapidly and before we reached the fork of the road, the left leading to the forest and the other to the Lecheria, the rain had begun. We tried the forest a short distance but the road soon became extremely slippery and the clouds were now so thick .on the summit that it was useless to go farther, so we turned back to the Lecheria. This Lecheria was a large dairy farm belonging to Don Ricardo Jiménez, the successful candidate for the presidency. It was situated at the lower limit of the oak forest, at an alti- tude of 8850 feet, in a beautiful nook between two hills. Many magnificent oaks were still standing near the house. The oaks of this level bore a wonderfully gorgeous parasite, perhaps Loranthus (Psittacanthus) schiedeanus, which Oersted collected on Irazi at 9500 feet elevation in January, 1847. It had large opposite leaves of a dull dark green, and immense tufts of red, red and yellow, or orange flowers. The sep- arate tubular flowers are about three inches long and very slender, one-eighth to one-quarter of an inch in diameter, but the mass of bloom on one tree was often as large as a small house and even in the gray light of cloud and rain was most brilliant. At the Lecheria we rode into an open, covered milking- shed, having a haymow above and a galvanized roof, and were glad to be under shelter. The “boss” was not present, 120 A YEAR OF COSTA RICAN NATURAL HISTORY but we made ourselves at home to the extent of tying up our horses, loosening girths, taking off our wet coats and hanging them up to dry, while a number of men looked on with appreciative grins at our rather draggled appearance. The pack-mules with the empty milk cans had passed us on the way up and were waiting in this shed to be unsaddled. Presently the manager appeared and gave us permission to have food for the horses and to sleep in the haymow, and one of the workmen now unsaddled our beasts and threw down what looked to our eyes enough hay for four horses. As a matter of fact our steeds ate steadily at it for two hours and a half, but they had surrounded it all by that time. Then they were watered and given a second similar feeding at which they worked all night. Having arranged the matter of staying, we looked about us while it rained. Our particular shed also sheltered some ox-carts and every now and then an ox or mule wandered in out of the rain, or to pass the time of day with the visiting horses. There were three milking-sheds in all, two of them quite large. We noticed also some American cultivators, harrows, etc., and two huge imported draught horses which made the native horses look like toys. The sheds themselves and their surroundings were very like such places at home where the owner has not been stirred by the modern ideals of cleanliness—neither worse nor better than hundreds of barnyards in the eastern United States. The house was a curious mixture of boards, bamboo, adobe, galvanized iron and other scraps. It seemed to have three rooms, a bedroom on the right, a recessed room in the center having a little covered, earth-floored porch before it and apparently con- taining sacks, seeds and other supplies and on the left a large kitchen. We reached the farm about 1.30 P. M. At 2.30 the cows began to come in from the fields, large handsome creatures ‘mnnbiygo mniyrkog “ULNUDIINIUL UNIUDLIY) 8 ~ mu 2 & x “sisuanspar vlayqysory ‘pajuasiv vesigs oy ae SR fr” sy XX Flowers and Entire Plant of Chionolena lavendulacea. Leaf, Inflorescence and Entire Plant of Gunnera insignis. (White cloth in upper view is 35 inches wide.) To face p. 121 THE VOLCANO IRAZU 121 that looked as if they could really give an appreciable quan- tity of milk. (Most of the cows here give very little, and if the calf dies, or is removed, the cow goes dry immediately.) We did not make an exact count but there were about sixty. The cows were fastened by stanchions and fed with hay and then the milking began. The milk was strained through cloths and then dripped over pipes, through which cold water was constantly running, directly into the cans in which it was carried to Cartago. The milk cans were of the familiar type with a copper label on the shoulder stating that they were made in Bellows Falls, Vermont. Some time after milking the cows were set free again and returned to the pastures, being milked only once a day. The rain stopping about four o’clock, we walked around a little admiring the views, the huge oaks and beautiful flowers and enjoying the antics of a number of frisky calves. A little after five we were invited into the kitchen to have some hot coffee and warm ourselves by the fire, an invitation we were glad to accept. The kitchen was a large room with a door and one small window in a corner. Its walls were board on one side, mud and bamboo on the others, while the floor was earthen. For furniture it had two or three benches against the walls, a long rude table, a rack of hooks holding white enamel cups, another rack full of white enamel plates, a swinging shelf on which some edibles were kept— I saw the cook take down one of the usual cakes of brown dulce—a tall wooden coffee strainer, some pots and kettles and in the center a modern cookstove, which however had no chimney connections. But although the stove was mod- ern, the cook was not and the entire front and sides of the stove were wide open and obviously warped and rusted be- yond the possibility of ever shutting again. The sticks of wood were twice as long as the fire box and were pushed in as the ends burned. From a kettle on the stove the cook 122 A YEAR OF COSTA RICAN NATURAL HISTORY ladled up a big dish of rice for us and also lent us one tin spoon with which we took turns; later we had each a cup of coffee and although we had brought plenty of provisions we were pleased to have these hot things. We sat on one side on one of the long benches and in other corners were half a dozen peons eating black beans, rice and tortillas. There was little light from outside and that rapidly waned, while within was only the firelight, so that the dark room with its dusky figures and the cook moving about among the shadows made a picture suggestive of Rembrandt. As soon as we had finished our supper, we retired to the haymow, spread out our shawls and blankets and went to sleep. The nights here are often very cold but fortunately for us there was no wind that night so we slept comfortably in the open-sided hay shed. Below us, horses and mules stamped and munched at what seemed to us an endless supply of hay. The peons climbed into the other end of the haymow and settled down after some giggling and the whole Lecheria was in repose shortly after dark. Some- where in the very small hours an alarm clock buzzed and after some sleepy groans and rustlings we heard the pack train, each mule with two big milk cans, start down the mountain for Cartago. With the first glimmer of light we climbed down with our outfit, saddled our horses and were off to the summit. The morning star still glimmered, the sun was just touching the highest peaks when we rode away from the farm and anything so beautiful as that morning ride I have never seen. As the sun rose and the mists dis- solved, range after range of hills and mountains appeared and soon when we came to a clear place in the road, we could see the valley containing Heredia and Alajuela, and still more to the west the Gulf of Nicoya and the Pacific Ocean, the islands in the Gulf and the hills of Nicoya beyond. South of us rose the Cordillera of Talamanca, looking immensely THE VOLCANO IRAZU 123 high. Other turns in the road showed us Cartago far be- low, the white fronts of its houses making conspicuous white lines as the sun shone on them. A mile or so after leaving the farm we entered the forest and here began the really hard part of the climb for the horses, as the trail is steep and there are many projecting roots to climb over—or stumble over if the horse is careless. It is a singularly impressive forest. The oaks composing it are of great size, very spreading and are covered with gray- green lichen so that they look almost unreal. In places the shade was dense, in others the sunlight filtered through and its warmth was quite welcome. ‘There were many beautiful flowers. Along the roadsides from the Lecheria up to the upper timber line we saw numbers of a pretty blue flag-like plant and great quantities of tall forget-me-nots with the usual blue and pink flowers. In the forest shade we par- ticularly noticed the handsome Siphocampylus regelit and Cavendishia veraguensis, both having conspicuous red blos- soms. The air was sweet with the fragrance of flowers and the glorious freshness of the mountain air. Many birds were singing, their notes nearly all different from any we heard at Cartago. After about a mile and a half of this forest path we came out abruptly on the cinders of the cone. There is no grad- ual diminution in size in the oaks; on the contrary they end suddenly so that it is possible to draw a line on one side of which are large trees, on the other none. This line is of course very sinuous but there is no difficulty in tracing it. In the forest the soil is black volcanic ash, while on the cin- der cone the particles are much coarser. The trails over the cinder cone look like artificial cinder paths, winding through the bushes and flowers. The vegetation growing on the cinders, while quite abundant, is not uniformly dis- tributed but grows in patches or “islands,” so that Irazu 124. A YEAR OF COSTA RICAN NATURAL HISTORY here shows an intermediate stage between the bare cinders produced by an eruption and the fully clothed summit of a long-extinct volcano. Such vegetable islands are due to the accidental lodging of some individual of the pioneer species by whose roots the cinders are held together so that other and different seeds can sprout and obtain a foothold in the shelter of the first. Standing in the cinders and among the growing plants are a great many small dead trees, all about the same size and all bleached quite white. These dead trees are not oaks; they were seen in January, 1847, by Oersted who mentioned them as Comarostylis rubes- cens (now termed Arctostaphylis ledifolia), one of the heaths. One of the most common and, because of its num- bers, conspicuous plants of these cinder gardens was a pretty little geranium (Geranium mexicanum). Its flowers are one-half to three-quarters of an inch across, from a pure white to a deep pink in color, daintily veined in purple. It starred the slopes wherever grasses had obtained foothold and then it reminded us strongly of our own “spring beau- ties.” It grew through the huckleberries and myrtles and opened its delicate blossoms among the glossy leaves of the stronger plants. Another common plant, growing several feet high, was Spirea argentea, with stiff, thick racemes of small, inconspicuous but very fragrant white flowers. A curious little fern (Botrychium obliquum) was growing among the geraniums. Dr. H. Christ, of Basle, wrote of our speci- men of this fern, “that it does not differ in any respect from those from the north of the United States and that this species plays in Costa Rica the réle of a glacial relict (“reli- quat glaciaire’’). At the very summit nothing grew—probably the wind was too strong. It took us a long time to reach this point as the trailing in the cinders is slow and tiresome, but at last we turned down into the crater itself. Looking north from THE VOLCANO IRAZU 125 the summit we saw the blue-green expanse of the forest- covered plains of Santa Clara with the silver line of the Rio Sucio winding towards the Sarapiqui. The Atlantic Ocean we could not see at any time and now the clouds in the val- leys cut off any view of the Pacific. We were less fortunate than Stephens in early February, 1840, or Hoffmann, on May 7, 1855, who record having seen both oceans from this volcano. Clouds were gathering near Irazi even before we made the summit, faint thin wisps of stratus clouds that were not promising. We rode over a wall some four hundred feet high, down into the largest crater and dismounting there at 8.45, tethered the horses to a bush. The floor of this crater is almost perfectly level, of fine ash and cinders strewn with huge boulders and in it were growing a few bushes, a pale yellow Gnaphalium (G. attenu- atum) and many plants and patches of the showy Castil- leja irasuensis, its red leaves and flowers looking even brighter against the gray cinders. The cinders and small stones were to a great extent covered with a gray, thick powdery lichen, which gave a strange and deceptive appear- ance of softness to the crater floor; there was also a quantity of Chionolena lavandulacea, a low, much-branching Com- posite plant with gray scale-like leaves and an immense root system. In a few spots we saw a small Alpine saxifrage. There were many patches of the huge Gunnera insignis es- tablished especially on the walls of the steep-sided secon- dary craters. The red-veined, roughly circular leaves of this plant measured over a yard across and were borne on thick stout petioles two to three feet long. The stem, ribs, and to some extent the whole surface were covered with short, sharp, recurved spines. The inflorescence was a showy, crowded spike of little red flowers, eighteen inches long. Although the conditions were highly unfavorable, we 126 A YEAR OF COSTA RICAN NATURAL HISTORY photographed a leaf and flower stalk in the crater, as they were much too big to carry down to Cartago. There were several kinds of birds flying in and around the crater. Insects were present, in small numbers; we saw a green lace-wing fly, a few small Diptera, one large fly, a small fly or two and one or two specimens of a small species of may-fly. In the mists of the crater on July 15, P. saw a few moths of fair size but was unable to catch one, for they were speedily swept out of sight by the wind. Their wings seemed to be of a dark color with a pale patch on the middle of the front pair. The stratus clouds that promised so badly were unfor- tunately true to their promise. Shortly after nine o’clock the crater began to fill with mist which blew in from the east. It lifted three or four times, so that we were able to secure some photographs in spite of the moisture on the lens, but each time returned as a denser cloud, cutting off the view of the craters completely and rendering any cir- cuit of them quite useless. We knew by this time that after the summit was once thoroughly cloud-capped it was al- most certain to remain so for the rest of that day, and so with great reluctance, we started on the return journey about eleven. Before we were halfway across the cinders actual rain commenced and was soon very heavy, continu- ing so until we were far below the forest. The clouds were low, and the dead trees and gray cinders looked pale and misty but we had no difficulty in retracing our path, and ab- sence of wind made the low temperatures not uncomfort- able. If the forest had looked unreal in the bright sunlight it was positively unearthly in the gray clouds, the lichen- covered trunks standing up like the ghosts of trees. The rain made the trails very slippery, in places really danger- ous, and all of us—horses and riders—were thankful when we passed into the clearer, wider high road. THE VOLCANO IRAZU 127 The rain stopping for a short time, we turned aside at a branch road 1400 feet above Tierra Blanca, where there is a little stream, and watered and fed the horses and refreshed ourselves. But all the rest of the way down there was a steady and exceedingly hard downpour, and although we followed the longer milk road to avoid the Tierra Blanca hills, the road was steep and so slippery with the rain that even our surefooted little horses could do no more than crawl. It was consequently a long and tiresome ride down to Cartago, but we managed to keep the cameras and plates dry and in spite of the rain we were glad we had made one ascent in the wet season. Our next visit to the volcano was in the dry season, when we carried a tent and camped two days in the crater itself, and in spite of some unfavorable weather were able to make a much more complete examination of the volcano. Leaving Cartago in the morning of March 31, with a guide and a packhorse, we rode directly up to the crater, only pausing in Tierra Blanca a few minutes to speak with our friend Senora Brenes. In great contrast to our previous visits to the top of Irazi in July and in September was the dryness of the road for the whole distance, even through the forest. This made our ride easier, speedier and more agreeable. In the forest, we were sorry to see that a number of large oaks by the roadside had been felled or burned since our last visit, a destruction which has doubtless been in progress for many years. About halfway in the forest we turned aside to fill our water cans at a small and beautiful spring known as Chicoa. Its altitude is probably near 9700 feet and it is the nearest water to the crater, distant three or four miles. Not until we were about to begin the descent into the cra- ter did the first clouds roll over us, producing intermittent mist and rain. We selected a level spot, on the floor of the 128 A YEAR OF COSTA RICAN NATURAL HISTORY great crater near its eastern edge, for our tent, where it would be protected by low trees and bushes from the prevailing easterly and north-easterly winds. At this place there is a wide stretch of smooth, bare, black volcanic sand bordered with gray and reddish mosses on the edge toward the inner slope of the crater wall. We pitched the tent where sand and moss met and had no reason at any time to regret our choice of a location. The altitude here was about 10,890 feet. The tent went up readily, all three taking part in its erection. As soon as it was up all things brought with us were placed inside, including the saddles of our two horses and of the packhorse; these formed an unexpected but welcome addi- tion to our “furniture” while the saddle cloths served as pillow and foot-coverings at night. The guide then mounted his horse and started on his return to Cartago, driving the other three horses before him. So at two o’clock in the afternoon we two were sitting in the tent eating our mid- day meal, alone in the great crater, six or seven miles from the nearest human habitation. It rained or misted at intervals, while we arranged things in our tent, and gathered firewood, stacking it inside in the hope that it might dry a little. While doing so we noticed various kinds of footprints in the sand near the tent, chiefly of four-footed creatures. There were small prints in groups of fours of a size suggestive of skunks and larger ones which were probably made by a big cat such as a puma, but al- though we especially feared the visits of skunks we were not troubled by any sort of animal at any time during our stay, nor did we see any quadrupeds even when we went out of the tent in the night. Some small noises we heard but they soon passed into the stillness that reigned whenever the wind ceased blowing in the trees above us. Shortly after five the clouds cleared away and we went to the edge of the large daughter-crater north of our tent to Oaks at the Upper Limit of the Forest, Irazt. Dead Trees on the Inner Slope of Crater, Irazi. To face p. ~-° Cppra 4q ‘o10yd) ‘a ‘H ‘d—a ‘g ‘£) sioqeID Jovysneppuricy sulmoys ‘yynos ay} WoOIf NzZeIT Jo (IID dayeig dojysneqd usia}s9 MM “D2ZBAT JO (J) JoIeI_ JayIOYAY ayI Jo 100],q YLT, ‘nzesy jo “yy ‘q— a ‘g stoye1Z Joqysnep-puriy . 129 To face THE VOLCANO IRAZU 129 look down into its depths at its own daughter-craters within. We could look far north of Irazii also and see the windings of the Rio Sucio through the forest plain, and to the west parts of the volcanoes Barba and Poas were visible. Our bed was made by laying the tentsack and newspapers on the soft dry moss, upon them the rubber poncho, on that an old shawl, then ourselves in winter clothing and rolled up in blankets, two each. We were cold at times during the night but we slept, even if not continuously. The lightness of the tent astonished us even before the moon rose, and at times when the stars were temporarily covered with clouds—at all times during the nights it was almost light enough to read our watches. The tent was fourteen feet long; it had belonged to Mr. Schaus and had been especially waterproofed by having a coat of lard, paint and linseed oil rubbed into it and then while still moist treated with white paint and it proved perfectly water tight. At five the next morning the moon and stars were shining most brilliantly and we felt that one of the reasons for camping up here was fully justified, for the early morning is usually the clearest part of the whole day. We took a bite and set off at once westward from our tent across the main crater, which is nearly a mile in diameter, and then climbed up a ridge of ashes and cinders to the summit of Ir- azu, which our aneroid made 11,300 feet. As we climbed a very strong wind blew from the east, but from time to time we passed a low tree or small shrub and although the branches and foliage of these were not dense they afforded an astonishing amount of shelter from the cold wind. On reaching the summit we saw a thick layer of cloud all around the mountain several thousand feet lower down, blotting out Cartago, San José and the other towns. Only the ex- treme upper tips of Barba and Pods (which are 9524 and 130 A YEAR OF COSTA RICAN NATURAL HISTORY 8786 feet high, respectively) were above the clouds and the same was true of the highest peaks which we could see to the south, including presumably the highest mountains of all Costa Rica. To the northeast Turrialba stood up splen- didly and was about as much free from cloud as Irazi— Turrialba being 11,224 feet high (Pittier). Westwardly we could look down the valley of the Rio Grande de Tarcoles, which was not cloud-obscured until near its mouth, but to the north we saw only great masses of white clouds, beauti- ful to look upon as the rising sun shone on them. As there seemed no reason for thinking that the cloud masses below us would clear away we left the summit at 6.30 and returned to the tent for some breakfast. About eight o’clock we set out to make the round of the daughter- craters. The volcano proper of Irazii consists of a large incomplete crater, here called the mother-crater (I of the sketch-plan) which has a diameter of the greater part of a mile. Within this are two daughter-craters whose combined east-west diameters are nearly as great as that of the mother-crater but whose north-south diameters are less, so that on the south side a considerable area of the floor of the mother- crater is left between its south wall and the south rims of the daughter-craters. The southern rim or wall of. the mother-crater is of unequal height from point to point, one of these peaks being the highest part of all Irazti, while there is another high point at the eastern side of this mother rim. ‘The two daughter-craters are in plan somewhat the shape of a figure eight, the smaller loop or crater being east, the larger west. The smaller eastern crater (II of the sketch) contains no secondary or ‘‘granddaughter”’ craters, the larger western daughter-crater (III) contains eight distinct granddaughter craters (A-J of same sketch) but two of these latter (D and E) might be called twins. These eight grand- THE VOLCANO IRAZU 131 daughter craters are of different depths; thus the floors of J and A appeared to us higher than those of all the others, while G and H seemed the deepest. North East Summit South SkeTcu-PLaNn oF THE Craters oF Irazu, April 2, 1910 from point S on the edge of the rim where the o0-shaped “daughter- craters” meet. T, location of our tent on the floor of the “‘mother-crater,”’ 10,880 feet above sea-level. I, “Mother-crater,” II, Eastern ‘“daughter-crater,”’ III, Western “daughter-crater,” A-J, “granddaughter craters” in the larger, western daughter-crater. Dotted line shows our route of April 1, 1910. vr, Bare faces of vertical rock are exposed here. lr, Laminated rock, the layers more or less horizontal. A piece of this rock collected in April, 1910, was identified as tuff from a hardened mud- flow of the volcanic ash (probably). Floor of mother-crater, especially in its eastern part near T of fine black sand, and on this floor at:least the yellow grass-like plant is chiefly associated with fine sand. The western part of this floor consists of coarser cinders interspersed with large and small boulders of angular outline. A subcubical block about 3.5 centimeters on an edge, brought 132. A YEAR OF COSTA RICAN NATURAL HISTORY from this place July 15, 1909, proved to be porphyritic basalt lava with small plagioclase phenocrysts, while other fragments, one centimeter and less in diameter, from the same place and date, are volcanic lapilli ‘of scoriaceous lavas, basalt and pyroxene andesite apparently. The Arctostaphylos, lupine, myrtle, and Ericad are more partial to the coarser parts of the floor. In craters G and H there are bare vertical rock faces, but with these and the exceptions noted above under or and /r, the walls and floors of the craters are of loose cinders of varying size. The only other published sketch-plan of the craters of Iraztii of whose existence we have been able to learn is that of Dr. Karl Sapper (1901). As may be seen by comparing his much more detailed plan with ours, the two differ considerably, yet are sufficiently alike to enable one to iden- tify the various craters represented. No doubt Dr. Sapper’s sketch is more accurate than ours, in spite of his description of his own as “eine approximative Veranschaulichung.” We have left our sketch as it was originally drawn except that we have changed our lettering of the different raters to conform with his. Dr. Sapper considers granddaughter crater J to be really a part of the floor of the western daughter-crater (III), grand- daughter crater E to be younger than D, and G and H to be the youngest of the granddaughters. Two sections through the craters of Irazii have been published by Prof, Tristan in his 1910 paper on the volcano. We add a list of approximate altitudes from Dr. Sapper’s paper, “ Die sudlichste Vulkane Mittelamerikas ” (1901). Altitude Designation Diameter Altitude of Rim of Bottom I (Urkrater)............ 1200 3414 and less 3320 II (Ostkrater)... ..- 500X400 3230-3380 3185 TU (drittétin cura rnnns 800X700 3230-3320 3185 Hauptkrater) Me cthtecicaaettexcs chica Guns Renee cing eee ner 3170 3160 Bio. -anctkeyataese aay Soees Sed es 3160 3150 Cui. ciiiadealmaabet ea cacertaae womnae 3170 3160 Aer, eed aconlbaaye dines nt ecbeteactsaneeale anced ean <3160 | enclosed 3150 by a rim Eis Sek Sesh nee aeangon eked AGO MOE 3150 | 3160 3140 BF) eigen onsug Ged tha ainctaiandes acne EN yd <3190 3180 CGuckee seyhdseukeasee ae sone 3200 3140 Teese pated enn eeeariiboge ete 3200 3140 THE VOLCANO IRAZU 133 Altitude Designation Diameter _— Altitude of Rim of Bottom [isaseieasdantesecd ene Gaisaidan cine arate omit: 3170 3150 TDs ese Ava acta ea haa ary gat a RR 3160 <3160 (PARTE 6d 86 Visi bcdsewex cen asne 3330 C and L are still smaller craters not shown on our sketch-plan. All dimensions are given in meters. The course which we followed is shown by the dotted line on the sketch-plan. From time to time the mist blew in so heavily as to blot out everything from view. At such times we sat down in the lee of a bush, covered ourselves and es- pecially the cameras with the poncho and waited, some- times half an hour, until the mist cleared away and then resumed our walk. In spite of these drawbacks we were able to get some photographs while making this circuit. All of the craters, mother, daughters and granddaughters, have more or less vegetation on their walls, floors or both. A plant looking like marsh grass, growing in tufts, is char- acteristic of the crater floors and the dead yellowish stems gave a sulphurous appearance to them when seen from a distance; some of this grass too has a decidedly salty odor, which increases the superficial resemblance to a salt marsh. Several species of birds were not uncommon in the craters —a thrush not unlike a robin, some small sweet songsters and humming-birds. Little gray spiders (Pardosa sp.) ran over the ground, a handsome red and yellow bumble bee, apparently of the same species (Bombus ephippiatus) as occurs also around Tierra Blanca, occasionally flew by, small gray moths and flesh flies were abundant. We stopped to photograph a group of beautiful Bromeliads growing on the ground beneath the branches of small trees and now in flower; the inflorescence was red and green, the leaves were streaked and spotted with red. On pulling one of these apart we found, in the water between the leaf-bases, live mosquito larve (Aedes near Ae. aureostriatus) and other 134. A YEAR OF COSTA RICAN NATURAL HISTORY dipterous larve (Chironomids allied to Tanypus and Culi- coides), a weevil (Anchonus coarctatus) previously recorded only from Cerro Zunil in Guatemala, and small planarian worms (probably the young of Rhynchodemus bromelicola). The northern rim of crater III falls away on the north side perfectly vertically, to a depth of one hundred and fifty to two hundred feet, to a valley filled with luxuriant vegetation. So sheer is the precipice and so narrow is the rim that it is difficult to see how the wall has been retained on this side. Crater G is one of the latest of these vents of Irazi. We saw a slight quantity of steam or vapor issue from it. Its walls are very steep and we did not risk a fall into it in order to see the deepest part. Some sulphur coats part of its walls and floor, as is the case with the north wall of crater IT also. Professor Pittier reported sulphur vapors issuing from G or H in 1888, but both Dr. Sapper and Professor Tristan assert that since that time no activity has been observed in any of the craters. Our observation just mentioned, of steam or vapor issuing from G, stands in contradiction to these statements. The really active part of Irazi, in recent years, is on the north slope of the mountain, at an elevation of about 9250 feet, outside of all the craters. In December, 1888; and January, 1889, Professor Pittier found the entire slope in activity with blue clouds of sulphurous smoke and hot water flowing. Dr. Sapper, on February 28, 1889, found only a few fumaroles but no hot water springs. This locality, termed ‘“‘boca nueva” or “volcancillo” by the Costa Ricans, we did not visit. We returned to the tent level by climbing the trail through the cinders on the inner (eastern) slope of the western rim of daughter-crater III. This western rim seems to belong to both daughter and mother-craters or else the western rim THE VOLCANO IRAZU 135 of the mother-crater has disappeared. But to whichever it belongs, it forms a very narrow sharp ridge several hundred feet high, which falls away steeply both to east and west, so that it fully justifies the name of “knife-edge” we gave it both now and in our September visit. Ascending this trail and walking on the top of the knife-edge we were again ex- posed to the strong easterly wind which, bringing mist and rain once more, made traveling very difficult and wet us with uncomfortable thoroughness before we reached our tent again, about 1.30 P. M. The rain and mist continued all afternoon and more or less all night; whenever we woke we heard the roaring of the wind over the crater top, but very little reached us in our sheltered nook. It was dark and rainy in the morning and we stayed rolled up in our blankets, as the warmest place, until nearly seven o’clock. At intervals the mist lifted a little, occasionally even leaving the craters quite clear for a few minutes. We employed these intervals in hunting for beetles under stones in the black sand, in making a sketch of the various craters from the rim we visited the first night and in gathering plants for identification. The beetles found proved to be three small species of Carabidae, all chiefly black in color. The largest, one-third of an inch long, Amara costaricensis, is shining, with eight strie on each elytron, antenne and legs reddish-brown, the tibiz of the first pair each with two distal spines which are longer and stronger than those on the other legs. An- chomenus 10-punctatus, one-quarter of an inch in length, has a satin-like luster above but is shining below. Its elytra have each eight strie but the legs are black except for the reddish-brown coxe. Bembidium rogersi was the smallest of the three, one-sixth of an inch long, shining, also with eight striz on each elytron but the four striz of each wing-cover nearer the middle line of the body are more 136 A YEAR OF COSTA RICAN NATURAL HISTORY strongly marked than any of the striz of the two preceding species; the legs and the first antennal joint are reddish- brown, the remainder of the antenne darker. The crater was almost as much of a flower garden now as in September. Immediately about the tent there were few plants on the crater floor except the yellowish grass-like plant, which was ripened into a sort of standing hay, some moss and the gray-leaved Chionolena lavandulacea so com- mon over the whole crater. Its pretty white flowers were now abundant. On the slope of the crater wall behind us were small trees and bushes of Spirea argentea, not now in bloom, also many young and medium-sized bushes of Arctostaphylos iedifolia, a plant with stiff leathery leaves that forms the greater part of the woody vegetation on the cinders both inside and outside the crater and is growing up around the dead trees as if replacing them. The dead trees are found on the outside of the cone in fine sand where there is much living vegetation and in coarse cinders where there is little; they are as common on the inside of the mother- crater rim where they are half buried in young trees. They are of uniform height (about fifteen feet) and shape, very widespread for their height and with stiff gnarled spray. It looks as if they had been killed all at once, possibly by some gaseous eruption, and in the mist their bleached skele- tons give a strange and ghostly aspect to the scene. Among the trees and bushes on the inner slope were many plants of Senecio oerstedianus, composites with loose clusters of yellow flowers and large coarse entire leaves whitish below. On parts of the crater floor more distant from our tent. there were a few individuals of Castilleja irasuensis in bloom and many young plants of this species were growing up. In September they were flowering abundantly. Some scat- tered Gnaphaliums were still in flower now also. Where the yellowish grass did not hold sway there were in many The “‘Knife-edge”’ of Irazu, from the Summit. Granddaughter Craters of Irazi G, F, E—D and B, “Knife-edge”’ in Background. ’ To face p. 136 “QZeIT jo ToleiZ 9y2 ul Uspslegy) [vine Ny V ‘ *“SNIUIAITINU 0199UIG “SNUDIPAISLIO O1IIUIG To face p. 137 Bidens irazuensis. Myrtus oerstedit. THE VOLCANO IRAZU 137 places great quantities of an exceedingly pretty little lupine (possibly Lupinus aschenbornii) with dark blue flowers and a compound leaf compressed into a rosette of leaflets. The younger leaves are quite reddish and having a velvety “pile” they gather and hold the mist so that they frequently look as if strewn with jewels, a pretty sight when the sun shines on them. The plant is a compact little hemisphere about a foot across, with a single thick taproot bearing numerous nodules. We saw also a few blossoms of a yellow Hyper- icum; it was mentioned by Oersted as resembling H. brathys, having brilliant yellow flowers, and as being often covered with black fungus (Scorias robinsont). On some of the slopes of the lesser craters, in little pockets more or less protected from the wind, we found veritable flower gardens containing the handsome Bromeliad already mentioned, ferns (Elaphoglossum lingua, E. revolutum, Poly- podium moniliforme), thickets of Pernettya coriacea with its reddish new leaves and beautifully delicate little pink or white bell-shaped flowers; the yellow S. oerstedianus and Myrtus oerstedit. One such garden we photographed under its canopy of dead branches thickly covered with Usnea and other lichens. In some of these gardens we found a plant, probably a Gaultheria, with stiff leathery leaves and sprays three to four inches long of small waxy coral-red blossoms. There were plants of Gunnera insignis in almost all parts of the crater but, except in the green, vertical-walled valley north of the craters, their huge prickly leaves were brown and shrivelled and we saw none in flower. The only human being we saw in the crater was a man who rode up to our tent the first morning to ask if we had seen any cows. We had not, the only traces of cows being old ones, and he rode on past our tent to the east and along a trail that ran up a little gorge between two high parts of the crater-rim, through a green and leafy part of the crater. 138 4 YEAR OF COSTA RICAN NATURAL HISTORY So far as we could see from the tracks in the sand—and they lasted surprisingly long—he did not return past our tent. How he went back we have, no idea. To follow that trail was one of the things we were eager to do, at least far enough to see into what sort of country it led. Certainly there must be some green and attractive pastures or the cattle and horses would not wander there as they seem todo. But for us this trail will be forever shrouded in mystery, for we could not follow it through the thick mist nor would we have been able to see anything if we had attempted it. About eleven o’clock we were pleased to hear the thump- ing of hoofs as our guide and horses approached the tent— pleased because we had supplies for two days only and it was impossible not to speculate on the uncomfortable situ- ation we should be in if for any reason the guide did not re- turn. However, he came, and almost on time, and as soon as he had eaten something we began to pack up. It did not take long to saddle our riding horses and arrange the full saddlebags and blanket rolls. Then we took down the cozy little tent, stuffed it into its coffee sack, loaded the packhorse and started home. During most of this per- formance it rained more or less heavily. P. and I had on raincoats and as soon as the tent was down—the last thing I could help with—I mounted my horse so that my poncho would cover the saddlebags containing the cameras and plates. The guide also had a rubber coat which he did not put on but left carefully spread out over his saddle to keep that dry. He worked with us for an hour in the hard rain, frequently remarking ‘‘Mucha agua”’; then he rode for an hour longer through the mist and rain, still unprotected. Finally he shivered, said “Mucho frio”’ and put on his rain- coat! By that time he looked well drenched and we were much mystified by his proceeding. From the level crater floor we first climbed steeply up the THE VOLCANO IRAZU 139 southwestern wall of the mother-crater and then followed a long zigzag up a steep cinder slope where the path was simply a ledge trampled by horses’ hoofs. This slope, lying outside the crater, was the north side of a ridge running westward from the summit of Irazfi. We were too com- pletely shrouded in mist to see what lay at the bottom of this slope—whether it really had a bottom or continued down directly into wooded slopes. As we rode we looked down on a white sea of clouds some hundreds of feet below. There is little vegetation here; the cinders are coarse and loose with some larger fragments, so that the horses sink to the fetlocks at every step. We now climbed over the crest of this ridge and down on the outside of the cinder cone. We soon began to ride over somewhat finer and harder cinders and as we descended vegetation grew more abun- dant. The Spirzas, Myrtles and Arctostaphylos were not now in flower here, and there were few flowers on the gera- niums, but Hypericums and a white composite resembling Ageratum were blooming freely. Owing to the rain the trail was much more slippery on the way down than it had been on the ascent but even with the cloud the forest was lighter than in September. Then the light was dim and mysterious and the wonderful gray-green trees stood up like ghosts in the dark green gloom. Now it was possible to see long distances through the beautiful forest glades. Part of this difference, particularly near the trail was due to recent cutting, but most was the result of the oaks being in a transitional state between new and old leaves. Many of the old leaves had already fallen or else were dead or shrivelled on the trees, while the new ones were only an inch or so long and a beautiful pink above, whitish below. The mature leaves are dark olive-green, stiff and leathery and about four inches long. We noted only one Cavendishia, so abundant in September, but Sipho- 140 A YEAR OF COSTA RICAN NATURAL HISTORY campylus regelii was still flowering freely. There were num- bers of a pretty plant with glossy dark green leaves and violet-blue flowers an inch and a half long (Cestrum ira- zuense). Here and there, a small delicate composite with pretty feathery leaves, Bidens irazuensis, lifted its single terminal flower head, resembling a yellow (and yellow- centered) daisy. Beginning above the forest and continuing to several hundred feet below it, were draperies of Bomarea, a vine (one of the Amaryllidacez) with pendent clusters of beautiful dark red flowers. A few weeks earlier we saw these vines in the forest near the Orosi falls, where the flowers had the inner sides of petals and sepals richly blotched with black, while those we examined on Irazt showed only clear red. In clearings in the lower part of the forest and on roadsides below it were clumps of a truly magnificent composite, Senecio multiventus, which reminded us of our own golden- rods. It grows eight or ten feet high, and bears a golden- yellow mass of flowers more than a foot across—the indi- vidual heads are small but the aggregate is immense—which have a sweet and delicate perfume. When we were some miles below the forest the rain ceased, the mists gradually lifted and the roads from this level down to Cartago were much drier. Above Tierra Blanca, as we were jogging along a piece of road with high banks of clay, washed by rains and cut by road menders into irregular pockets and gullies, our guide, who had been watching the banks carefully, suddenly pulled up, put his horse half up the bank and began to dig in the friable soil with his fingers. In a minute or two he loosened a curious little object of red pottery—a flattened ball about two inches in diameter with a ring-like handle on one side. The ball is hollow, con- taining some small hard objects which rattle when it is shaken, and on the flatter side opposite the handle is a number of irregular little holes such as might have been THE VOLCANO IRAZU 141 made. with a pointed stick when the clay was soft. Our guide, who was used to looking for buried pottery, noticed the rounded end sticking out of the bank. Of course we have no clue to its age; although the use made of it cannot be certainly stated it is probably a ceremonial rattle. The guide presented it to me and it forms an interesting little souvenir of our volcano trip, as well as an Indian relic of whose genuineness we need have no doubt. CHAPTER VIII LOWER NEIGHBORS OF CARTAGO Cartago lies near the northwest corner of the valley of Guarco. West and north of the town where the land rises rapidly into the flanks of Carpintera and Irazi, it was devoted to maize or potatoes, or to pastures in which great numbers of cattle were raised. In other directions the (down- ward) slope is more gradual and although there were large pastures below the town much of the lower land was occupied by coffee. The valley drains into the Agua Caliente River which flows eastward close to the hills forming the southern wall. The attractive little village of Concepcién de Cartago lay about a mile south southwest of Cartago. Its plaza was a triangular grassy open space bounded on the east by the road from Cartago to San Isidro del Tejar, on the west by the road to Tobosi, these two roads diverging at Concep- cion. The yellow-plastered church of Concepcién de Cartago was particularly neat and pretty, one of its two towers hav- ing a good clock, while the priest’s house next to the church had a well-tended flower garden. Throughout Costa Rica the effort was made to have the public schoolhouse the best building in each little hamlet or village and here in Concep- cidn de Cartago it was exceptionally attractive, well built of adobe and plaster, neatly painted and in good repair. We rode and walked around this plaza often, on our way to the more distant villages of San Isidro and Tobosi, always with pleasure. San Isidro del Tejar lay at the extreme south edge of the valley on the Tejar River, one of the many tributaries of LOWER NEIGHBORS OF CARTAGO 143 the Agua Caliente, and was some four miles from Cartago The San Isidro road had the valuable peculiarity of being sandy (most of the others near Cartago being of clay) and consequently remained in good condition when the clay roads were sometimes girth-deep in mud. In some of the fields bordering this road—locally known to the foreign residents as the ‘Beehive’ road—were depressions filled with rich little pools of water throughout the rainy season and which furnished us a number of larve and exuvie. In September we particularly noted the abundance of a legu- minous plant with yellow flowers and curious appressed stipules—Zornia diphylla—and a pretty Hypericum (H. fastigiatum) in these wet spots. San Isidro itself was but a tiny hamlet, but its surroundings were charming. The hills of the south side of the valley were close to the Tejar and after fording the stream the road climbed at once and forked into two branches. The left fork descended after a few miles to the river level, reforded the Tejar and led back again, through wide grassy lanes bordered with rivulets that proved good collecting ground, to Concepcién de Car- tago. The right fork led directly into the hills. Tobosi, another small village, lies in the extreme south- west corner of the valley where the hills close in on all sides except the east. On one side of the road thither a laguna (at times only a swampy place) furnished us another collect- ing ground. While walking to Tobosi on June 20, 1909, P. saw one of those flights or migrations of butterflies so frequently observed in tropical America. I had already noticed the great abundance of a certain butterfly (Timetes chiron) of about two inches wing-spread, having on the upper sides of the wings alternate stripes of dark and light brown parallel to the body. The light brown is replaced on the under side by silvery-white, pale violet and light rusty-red. Each hind wing has two tails on its 144. A YEAR OF COSTA RICAN NATURAL HISTORY hind edge, an outer, one-half to five-eighths inch long, an inner one-eighth inch. I had often seen the species before but it was not until I reached this laguna about noon that I appreciated that numbers of this butterfly were traveling all in the same direction. I was in a grassy lane about forty feet wide, so I sat down to count the individuals which should pass where I was, within this lane only, in a given time. Between 12.07 and 12.12, I counted one hundred and forty- four and I may easily have missed some. They came, not continuously and at regular intervals but singly or in groups of two, three, four, five, six, or almost any number up to fifteen. They flew along at varying distances above the ground, but all in the same direction—generally eastward— and against the wind, which, when it blew a little stronger now and then, gave them much trouble to beat against. They did not stop to visit flowers or gather any other kind of food but continued unceasingly their undulating flight to the east. A few individuals of other species, as a bright orange Callidryas or ally of Callidryas, seemed to accompany the migrating species occasionally. To be able to identify the species in the future, I caught three specimens, but the first two had badly battered wings and I released them. When set free they seemed disconcerted and started off in new directions but within a minute or two changed their flight to eastward. The condition of their wings implies that the insects had not recently transformed from the chrysalid state. As the butterflies flew toward me, when I faced westward, it was impossible to foretell the exact path which would be taken by any one individual, so great was the lateral undulation in flight. When I left the spot where I had made the count, an hour and a half later, there seemed to be no decrease in the num- ber of the flying butterflies, so that 4320 of them must have passed through that narrow lane while I was in its neighbor- Pernettya coriacea. Limnanthemum humboldtianum. To face p. 144 Schoolhouse in Taras, near Cartago. The Reventazon River at Orosi. To face p. 145 LOWER NEIGHBORS OF CARTAGO 145 hood. As I walked back to Cartago, every now and then, looking over wide open fields, I could see the same species in apparently equal numbers flying toward the east, none returning. How wide the area of flight was I do not know, but it was certainly not less than half a mile and may have been much more, as J have no reason to think that the narrow lane was specially favored by them. Half a mile and the count I made would give 285,120 butterflies for the hour and a half during which I specially noticed them, and they were flying both before and after this time, for my route back to Cartago led across their path to some extent, until I reached coffee plantations surrounded by high hedges so that further observation was impossible. On June 23 we went to Juan Vifias and almost every day that we were there we saw large numbers of the same species of butterfly. On June 30 they were particularly numerous between 11 A. M. and 2 P. M. although not confined to . those hours. At Juan Viias also they flew eastward against a wind (which was strong enough at times to give the butter- flies much difficulty to make their way against it) and went on without stopping for flowers or other objects on the way. At times they were so numerous that they resembled brown autumn leaves swept along by wind. By actual count 394 of this one species passed an imaginary line between our cabin and the railroad station during the five minutes be- tween 12.30 and 12.35 P. M., the distance between these two points being about one hundred feet. This means 14,184 butterflies for the three hours alone, within this one hundred feet, as we have no data on the total area over which the insects were flying. We have already mentioned, page 105, seeing this same species in migration at Tierra Blanca on July 10. On most of the roads leading southward from Cartago we saw many brickyards. The clay was mixed in a circular 146 A YEAR OF COSTA RICAN NATURAL HISTORY pit some ten feet across and one to two feet deep. The actual mixing was done by two horses (occasionally oxen) which walked round and round in the pasty, sticky mass of mud and water, trampling it until it reached the proper consist- ency. The beasts were tied together, the second to the tail of the first, and led by the brickmaker. In addition to mak- ing the square flat tiles used as paving bricks, the long curved roofing tiles, the ollas and smaller pots for domestic purposes, there was a thriving industry (so we were told) in the manufacture of “antique” pottery to be sold later to tourists as relics from Indian graves. Here and there was a square brick oven, eight feet high and five feet on a side, for burning the tiles. These ovens were usually under sheds. In the southwestern part of the valley little coffee was grown, most of the land being in large pastures where horses and more particularly cattle were raised. Many of these fields were badly infested with ticks. These creatures have the habit of resting on grass, leaves or twigs with the first pair of legs extended (there are three pairs in the young, four pairs in the adult) ready to grapple any rough surface that brushes the plant. In this way they get upon their normal hosts, the cattle grazing in the field, and they can equally well attach themselves to human beings. On one occasion in May after walking through a field where cattle often grazed, A. looking down on her khaki skirt and canvas leggings found hundreds of the young of a species of tick (probably Amblyomma cajennense) crawling over them. By scraping the ticks off at once with a knife and later bathing with carbolic soap disagreeable consequences were avoided. But no one who has not lived in the tropics can fully appre- ciate what a pest the ticks may be and the precautions we took against them, such as always wearing high shoes and never going out into the fields without leggings, were fully LOWER NEIGHBORS OF CARTAGO 147 justified by the resulting freedom from annoyance. How abundant the adult ticks may be on cattle is shown by our count of ninety on one side only of one ox in a Cartago street on July 11; many were as big as marrowfat peas, gorged and swollen with blood, and dark brown in color. A fair-sized black bird, the “‘tijo-tijo,” “garrapatero” or “zopilotillo” (Crotophaga sulcirostris) is frequently seen about cattle in pastures. It has a stout bill, with which it picks off and de- vours the ticks (‘‘garrapatas”’) infesting these animals. It is a member of the cuckoo family. In general Costa Rican cattle were amiable and rarely resented the intrusion of two foreigners with insect nets into their fields. But a few days after our settling in Cartago, as A. was walking alone on the Beehive road, a man suddenly clutched her by the arm and with much Spanish and excited gesticulation led her to a gate in the wall of an exceed- ingly dirty little yard. An old woman, toothless, combless and dirty, hurried down to the gate and added her ex- citement to his. I looked in bewilderment from one to another, for my Spanish was unequal to this sudden demand upon it. Evidently I was desired to go into the yard, but why? The sky was clear, the road looked as usual, no more people seemed to be out, what could be wrong? Meanwhile the woman lowered the bars and the man almost pushed me over and the bars were up again before an English voice said “bad cattle coming.’’? And in another minute they came, half a dozen huge bulls which objected to my um- brella even across the fence so that I retreated to the house and closed the offending sunshade. A very small dog at once leaped over the lowest bar and challenged the threaten- ing bull (which could have taken him in like a pill if he had so desired) and the monster moved on. Some men with coils of rope were driving the beasts, while others walked ahead to warn unsuspecting travelers like myself and two 148 A YEAR OF COSTA RICAN NATURAL HISTORY horsemen, who flattened themselves ignominiously against an angle of the fence opposite to me to allow the cattle to pass. About two and a half miles south of Cartago and three hundred feet lower, lay the straggling little village of Agua Caliente, on the north side of the Agua Caliente River. The church had but a single tower and the two bells, large and small, hung not in this tower but over a wooden platform built between two trees by the side of the church. Much coffee was grown in the neighborhood of Agua Caliente and there were also numbers of clay-pits and ovens for burning the tiles. Across the river, quite close to its south or right bank, were the hot springs that gave the place its name. A large wooden bath house had been built over them, equipped with seven or eight enamel-lined tubs piped for hot and cold water, and at one time the whole place had evidently been well fitted up. When we knew it, however, the establish- ment had been much neglected, cold water pipes were out of commission, the tubs chipped, the window glass broken and furniture of the scantiest. Nevertheless there was a caretaker of a sort and many people, both Costa Ricans and foreigners, resorted to Agua Caliente for the hot baths, the water having the reputation of possessing great remedial powers. As we saw it, the water was very brown and was usually extremely hot, and in the absence of cold water to mix with it the bather had either to parboil himself or wait until the bath cooled down to a bearable temperature, which con- sumed much time. Consequently we did not bathe here often. The water of this spring, or Hervidero de Agua Caliente, was analyzed by Luciano Platt and a report presented to the President of Costa Rica in 1865. It has recently been reprinted by Don Cleto Gonzalez Viquez in his Temblores, LOWER NEIGHBORS OF CARTAGO 149 etc. Senor Platt found that the water as it issued from the calcareous rock had a constant temperature of 50 de- grees Centigrade and contained in each ounce of water two grains of solid material, consisting of chloride of sodium and potassium; carbonate and sulphate of lime; carbonate of iron; chloride, carbonate and sulphate of magnesia, arranged in order from greatest to least in quantity. The gases which are set free are principally carbonic acid. A more elaborate result was obtained in 1887 by Dr. C. F. Chandler (quoted in Biolley’s Costa Rica and her Future, page 17), the one and one-half grains of solid material from each ounce of water being composed of sodium chloride, bicarbonate of lime, sodium sulphate, bicarbonates of sodium and magne- sium (in the order named from greater to less) with smaller quantities of other salts. Behind the bath house the hills rose abruptly; about half a mile downstream they lie so close to the river that there was no room for a path or road, which was therefore carried over the hill. The Agua Caliente River at the bridge is pretty, with little islands here, overarching trees festooned with Tillandsia there, with stretches of smooth water al- ternating with boulders and ripples. It was never very pro- ductive of dragonflies, however, so that we spent little time by its banks. By turning south into the hills we came upon a charming grassy lane crossed by a rivulet trickling from a swampy spot. Higher up, this path grew steep and in wet weather slippery, through red soil with masses of disinte- grating feldspar, and higher still entered low second-growth woods, where the first Gomphine of the expedition (Zpigom- phus subobtusus) was captured on May 20. The general group of hills lying south and east of the bath house on the right bank of the Agua Caliente River was called the Jocosal. There was quite a network of roads over the Jocosal, most of them surprisingly good in spite of their 150 4 YEAR OF COSTA RICAN NATURAL HISTORY steepness, and the upper parts of these commanded mag- nificent views. On a morning in the rainy season, before the daily clouds rolled up, the panorama included the hills south and west with Barba showing through a gap between Irazi and Carpintera, the huge expanse of Irazt itself, the Car- tago slope and at our feet the Agua Caliente Valley with its shining river, while to the east lay the beginning of the Re- ventazon Valley with Cachi nestling among the coffee farms. The roads themselves, particularly through the second- growth woods, were exceedingly pretty because of the abun- dance of flowering trees and bushes. On February 11 we especially noted red and pink Melastomes, tree composites with large yellow flowers resembling sunflowers, many other less woody species of the same family, the beautiful blue sage or “ Jalacate,” the red and orange orchid (Epidendrum radicans) and lantanas similarly colored, Bryophyllums and long sprays of blackberries just turning from red to black. At this time many of the trees had young leaves or buds most brilliantly colored in red and yellow-bronze, quite equal to our autumn tints. Some of the fields on the Jocosal seemed to be abandoned entirely to the Santa Lucia in its most exquisite colors. On February 28 we rode over the highest part of the Jo- cosal (according to our aneroid 5250 feet while Agua Ca- liente was 4460), following a winding road which was now a wide and well-beaten track, then a narrow footpath and in general grew less distinct as we descended on the other side. It was soon evident that the horses were on unfamiliar ground and P.’s decidedly objected to the descent, for the road degenerated into a slippery zigzag down a steep hill- side. We knew there must be a river at the bottom and presently we heard it, but it was a long time before we could see it, for the second-growth woods were thick on both sides of the trail. On the lower parts we saw tree-ferns and here LOWER NEIGHBORS OF CARTAGO ISI and there tall Tillandsia-draped trees evidently left from the original forest covering. At last, slipping and scram- bling, we came out on a slope, quite cleared of brush and planted in places with cane, which fell away sharply to the level river bottom some hundred and fifty feet below. In almost all these valleys the change from the flat river bottom to the hill slope is extremely sudden and abrupt. Almost at our feet lay the Navarro River, with some meadows, a hacienda and outbuildings and plantations of cane. Our progress down this bare slope and across the meadow to the stream to water our horses aroused much interest in the people of the house and there was quite a gathering when we rode up to the gate to inquire about the road to Orosi. They courteously invited us to enter the house and rest, but this we had not time to accept, so after obtaining the necessary directions we rode on through a small stream and across the Navarro on a rude log bridge and then followed that river until we reached a place on the road where there was food and water for the horses and a shady wall for us to sit on. It was a pretty spot, where a small stream crossed the road and passed over a thickly shaded, rocky bed to join the foaming, roaring Navarro. Our aneroid made this place 3950 feet high. After our breakfast we went on towards Orosi. In general the road followed the river although at times it climbed up over a hill and down again, where there was no room for a road at the river level. When we had ridden several miles we saw again the valley of the Agua Caliente, now enlarged by the Navarro. Its flat and fertile bottom was about half a mile wide and contained many coffee plantations and some oranges. As we first rode over the little crest that hid the valley from us, we saw spread out at our feet the thou- sands of young orange and grapefruit trees of “Navarro,” an experimental farm of the United Fruit Company, with 152. A YEAR OF COSTA RICAN NATURAL HISTORY range after range of hills beyond the river and its valley. Indeed we lingered many times along this road and on the Jocosal to admire the lovely views. Past “Navarro” we rode on to the iron bridge spanning the united Navarro and Agua Caliente Rivers on the left fork of the road while the right fork goes on to Orosi. On April 12, 1910, with two guests from the hotel, A. rode over the Jocosal by another road, which followed a little N To Cartago via Dulce Nombre Mss) s To Cartago via s Rio A bg the Paraiso road TL ee ala ( : Se Sz ~ Orosé & Caché Sketch Map of Junction of Navarro and Agua Caliente Rivers. side valley and commanded even finer views. In one shel- tered “pocket” was a perfect thicket of tree-ferns, the most luxurious growth of them that I saw anywhere in Costa Rica. This was a particularly clear brilliant day, even the tops of the distant Las Cruces mountains being visible for a while. Great numbers of the dragonfly Heterina cruentata were flying in the road, their glittering metallic bodies and the blood-red spots on the wings flashing in the sunshine. There were swarms of them, in places, and as we rode along they lit upon the horses’ manes and ears as fearlessly as flies. At about the highest part of this road we came upon a LOWER NEIGHBORS OF CARTAGO 153 man resting by the roadside, who was taking two large sacks of oranges to market. He seemed delighted to lessen his load by some sixteen oranges, at a good price (nine cents) and we were equally delighted to have the fruit. After an hour’s further riding I suddenly found ourselves on the road over which P. and I had gone a few weeks previously and it was not long before we began the steep, narrow zigzag down to the farm lying by the Navarro River. We reached this farmhouse at 1.30 P. M. and rejoiced in the prospect of food for ourselves and our horses for we had brought no breakfast. There was no trouble about the first—the old woman was glad to sell us beans, tortillas and coffee—but there proved to be a difficulty about the beasts. Corn there was none. Cane was growing close to the house, so I asked next if we could buy enough cane for our horses, their ordinary food in the stables being chopped sugar-cane. “Si, Si, con mucho gusto,” and the boy was sent for “enough.” But before he started to cut, a fourth rider appeared and put his horse in the trapiche shed along with ours but at a manger where chopped cane was ready. And that ended our prospect of buying cane. The farmer now told us that he could not sell us any as it was not his own cane but the patron’s. Yes, he had some himself but it grew very far off, too far to send the boy and he—the farmer—was sick, as indeed he looked. If the patron was, as we strongly suspected, that fourth man, we could not see why the farmer did not ask his permission to sell cane then and there, but evidently there was more to it than appeared on the surface. At all events the poor horses had only such grass as they could pick up by the roadsides during our numerous halts, and as a result were not anxious to travel fast; in fact we had difficulty in getting them above a walk most of the way home. Although I knew the place perfectly well, I did not recog- 154 4 YEAR OF COSTA RICAN NATURAL HISTORY nize the man, for we had only stopped to ask the way on our previous visit and I had not noticed the farmer particularly. It was evident, however, that he had noticed me. I saw him studying me long and carefully, then looking at Mr. M. Finally he said to me, “You were here two months ago, no?” “Yes.” “Yes, you are the same, but the man is dif- ferent!”? We all laughed when I translated that and Mrs. M. exclaimed, “That is my man!” Of course the farmer did not understand the words but her gestures were unmis- takable and he said at once, “‘O, su esposo”’ with an air of enlightenment. Then he looked at me again in a way that plainly said, “But what the dickens are you doing here?” A longer and much prettier route than the straight road ‘to Agua Caliente ran through the neat and attractive little village of Dulce Nombre, south and east of Cartago. It was quite closely built along the line of the road, with the begin- nings of a brick church which promised to be very fine if it ever reached completion. Dulce Nombre was on the little Rio Toyogares, and the banks of this stream bore quite the largest thicket of Datura arborea that we saw anywhere. From Dulce Nombre one road led west to Agua Caliente. Another ran east and south through the region known as “Las Concavas” and past the hacienda of the same name, descended by zigzags to the level of the Agua Caliente River crossed this river on a suspension bridge and at “‘La Flor” joined the road from the Jocosal and ran on past ‘“‘ Navarro” to Orosi and Cachi. The usual route to the village of Orosi, lying in the valley of the Agua Caliente east of its junction with the Navarro, was to follow the Carretera east through the barren red gravel of the Paraiso country almost to that town, where a road branched south into the valley. There was a small laguna by the side of the latter road which looked promising, but we never found many species of dragonflies in or near it. LOWER NEIGHBORS OF CARTAGO 155 On February 28 Utricularias and Limnanthemum humboldtia- num were blooming freely in the laguna and “cirujanos”’ were flying over it. ‘‘Cirujano” (surgeon) is the Costa Rican name of one of the rail-like birds (Jacana spinosa or Asarcia variabilis) about eight inches in length, with glossy greenish-black head and neck, reddish-brown body acquir- ing something of a purplish tinge below, and pale lemon wings edged with brown. Each long black leg has four toes, each of which is two and three-quarters inches long. The most remarkable external feature is a sharp-pointed, pale yellow, horny spine (carpal spur) one-quarter inch long, on the front edge of each wing near its junction with the body. Such spines are possessed by a number of birds and are re- garded as weapons,! but we know of no observations on their use by our “‘surgeons.” Orosi itself was a rambling, unkempt village with an unusually large, grassy, weedy plaza and a single-towered, blue-washed church, which is said to be the second oldest in Costa Rica and to date from early colonial times. It lies in a rich coffee district with several beneficios near it, and the surrounding country is most beautiful, but did not prove productive for our purposes, so that we did not visit it regu- larly or frequently. On October 26, 1909, A. rode there with a party from the hotel, chiefly to see the great cataract, which was not seen, however, as the guide proved entirely ignorant of the road. There are in fact two large cataracts near Orosi, one of which we saw on February 28 from the top of the Jocosal and which we afterwards located as being on the road from Paraiso to Cachi. The other, much larger, lies deep in the forest several hours south of Orosi. The small boy who acted as guide had apparently never heard of either of them. I noticed on this day that great quantities 1In the article by F. A. Lucas in the Report of the U. S. National Museum for 1893, pp. 653 et seq. 156 A YEAR OF COSTA RICAN NATURAL HISTORY of the wickedly thorny four-sided vine called ‘Rabo de la Iguana” (Mimosa velloziana) were growing in the valley. On the way down we all noticed a colony of large insects on a Croton leaf, our attention being attracted to them by the brilliant metallic reflections as the wind shook the leaf in the sunlight. As we returned we stopped and secured the leaf. The insects proved to be real “bugs” (Hemiptera) about half an inch long and three-eighths wide, with long beaks stuck into the leaf, on which they were packed as closely as possible. They were beautiful creatures. The legs were metallic blue, the bodies metallic bronze-green, bronze-blue, bright blue and red, all glittering and sparkling in the bright sunshine, and the effect of the whole colony was dazzling. They had as yet but the rudiments of wings and belonged to the family Pentatomide; they may possibly be the young stages of Arocera splendens, the adult of which we took at electric lights in Cartago on July 3. Four miles east of Cartago, just south of the railroad, lay a pretty boggy pool, in part thickly overgrown with a sedge (Eleocharis sp.) in which the dragonfly 4shna lutei- pennis laid its eggs, in other places covered from October to June with the lovely little gentian, Limnanthemum hum- boldtianum. The heart-shaped leaves of this plant float like those of water-lilies, while the starry white flowers with their delicately fringed petals are lifted above the water. Although it occurs throughout Central America and the Antilles, L. humboldtianum had not been reported in Costa Rica before our collecting it, except in the district of Tala- manca. The subsoil about this “‘ Limnanthemum pool” was a red gravel and in April great quantities of the common red and yellow terrestrial orchid Epidendrum radicans, were growing in it. This handsome orchid seems to prefer red gravel and exposed, eroded banks, and near Cartago at least we always found it in such situations. LOWER NEIGHBORS OF CARTAGO 157 A little beyond the Limnanthemum pool the railroad en- tered Paraiso, the first village and station east of Cartago. We rode to Paraiso on horseback several times, the road being a fairly good one, but we found little of interest as it ran through pasture and farm land on rather poor soil. On one of these rides, on December 6, we noticed many hawks, large and small. Paraiso was a much smaller town than Cartago, its streets paved roughly with stone. It was chiefly remarkable for the large church on the Plaza, with a single very tall tower, which unusual feature made it a landmark for miles. In March, 1910, we spent a week with Mr. and Mrs. C. H. Lankester at Cachi, a small village in the midst of coffee farms and beneficios on the south side of the Rio Reventazén; Cachi is three miles northeast of Orosi and three miles east of Paraiso—in air-lines. It is most easily reached from Par- aiso, whither we went by train on March 3. Mr. Lankester sent horses to meet us and we started down at once, for Cachi is about “‘an hour and a half” from Paraiso in good weather. We followed a road, whose windings over the bare reddish hills of Paraiso we had often watched from the train, descending steadily, as Cachi lies 1000 feet lower than Paraiso. The ride was both beautiful and interesting. On the last hill we passed the ruins of the old stone church of Ujarras, which is said to date from the sixteenth century and to be the oldest Spanish church now remaining in Costa Rica. Cachi is now a small hamlet, but before the railroad was built it was of some importance and had a large native population. Much coffee is grown in the main and side valleys about Cachi and in shipping season long trains of ox-carts wind up the hilly road and Paraiso is a busy station. There is no other occupation, however. The village consists of two or three roads with lines of houses straggling along 158 A YEAR OF COSTA RICAN NATURAL HISTORY one side. There was a post office and telegraph station, a neat though small one-story wooden schoolhouse and a church consisting of a galvanized iron roof supported on wooden uprights, two tower roofs likewise supported in front, no walls to towers or to “‘nave” except at the back of the “chancel,” and a plain board-walled diminutive “arm”’ of a “transept” on each side. The wooden pulpit was in full view from the road and the cows wandered to its steps, nibbling at the sod. Behind Mr. Lankester’s house, that is, south and south- east, the chief road led to Pefa Blanca. The path we usually followed led successively through a coffee field, a potrero with some deserted houses, and then into a second potrero growing up again into woods. This clearing was bordered by thick tall timber and through it ran a charming brook, a branch of the Sordi, which was often beautifully shaded and full of rocks and deep nooks. It seemed a highly favor- able place for dragonflies of certain kinds, but we saw almost none at all. Here in fruit were many trees of “Vainilla” Cassia spectabilis which we had previously known only in flower, at Alajuela. The fruits are slender pods some twelve inches long, packed with numerous close-set beans flattened against each other like gum-drops. Many composites were also blooming in this potrero. A curious and conspicuqus plant had fruits about an inch long of a deep purple-black, borne on vividly red, fleshy receptacles two to three inches in length. There were numerous ‘“‘Guacamayos,” papa- veraceous plants with large irregular leaves and a big con- spicuous fruit cluster which hangs lopsidedly out of the apex of the plant. Croton trees were abundant, both xalapensis and gossypitfolius. The main road to Pena Blanca ran up the valley of the Naranjo, of which the Sordi is a tributary. There was much swamp land in this valley bottom, which was perhaps LOWER NEIGHBORS OF CARTAGO 159 a quarter of a mile wide and was hemmed in by abrupt cliffs and hills. Pena Blanca is a group of particularly high cliffs rising several hundred feet above the meadows at their bases, with steep often vertical faces but broken into ledges here and there giving foothold to a luxurious vegetation. The cliffs are quite white in color and are most picturesque in their bold outlines and the contrast between the bare gleaming rock and the dark green of the forest capping them and clinging to every ledge. There is a little amphitheater hemmed in by such cliffs. Beyond Pena Blanca the road dwindles to a trail leading into the hills, past other cliffs of darker rock. On the right (east) side of the Naranjo a cart track led up into the wooded hills not far distant and on March 9 we spent the morning there. A tiny stream tumbling out of the woods and shaded by some crotons offered an inviting resting-place when we tired of fruitlessly searching after dragonflies. In the more level meadows nearer the stream were many large trees loaded with epiphytes, which had been left standing—probably to give partial shade to the cattle—when the forest was cleared away. We carefully explored some large and accessible bromeliads growing in great numbers on a fine /mga with glossy leaves—probably “cuajiniquil.” This did not prove fruitful of dragonfly larve, a result we had indeed expected, for the tree stood in the open sunlight because of the cutting of the surround- ing forest. There were enough brown cockroaches (Blat- ella bruneriana), spiders (Cupiennius coccineus, Selenops mexicana), and earwigs to add excitement to the search. In the cleared fields at the upper end tree-ferns ten or more feet high were standing isolated in full sunshine. One afternoon was spent in the valley of the Rio Oro a short distance west of the village. We rode up between cafetales and potreros, then leaving the horses we descended 160 A YEAR OF COSTA RICAN NATURAL HISTORY into a potrero from which a path climbed up into the forest closely paralleling one of the many small branches of this river. Scattered in this forest were tree-ferns of at least two distinct kinds, one with a trunk three to four feet high (Athyrium sp.), the other with a trunk fifteen to sixteen feet high (Cyathea basilevis), together with many Philoden- drons, Monsteras and other Aroids, but few Bromeliads. Insects were scarce, partly because of the general cloudiness of the afternoon, perhaps. Mr. Lankester’s house was beautifully situated half a mile from the Reventazon River and about a hundred and fifty feet above it. It was not a typical Costa Rican resi- dence for although built of adobe it was two-storied and had no patio. A wide veranda or “corridor” ran across the entire front both on the first and second floors, charming places, for they were set with plants and hung with baskets of orchids, some of which were in full flower when we were there. The veranda was draped with a magnificent ‘“Ve- gissima,” a vine bearing masses of pink flowers; at the time of our visit it was in fruit. From the upper corridor espe- cially the house commanded an extensive view of Irazii and Turrialba, although the latter was frequently cloud-cov- ered as early as eight o’clock in the morning at this time of the year. The river was not visible from the house because of the many trees. It could be reached either by following the tributary Rio Zapote through the potreros or by a road through the cafetales which led to a place where an iron bridge had spanned the Reventazén. This bridge had been carried away in one of the periodical floods the preceding winter and as a consequence the carts carrying coffee were obliged to go some distance upstream to the next bridge, greatly increasing the hauling. Fording with carts was im- possible, for the river is full of boulders here and the current swift and strong. A temporary suspension bridge for one = oO x oO v 3 tL o 2 4 g 9 N os v a oO > Oo ~% o a fs) a ° ae a ws ao} o G +e g ° a vo - > a ie Qe oa a a S o Yr n es u a os na oO o a Bb he 8 oO a p n S 3° Oo To face p. 160 Purple-leaved Urera, Telanthera mexicana and Epiphytic Ficus sp., on the Reventazin banks, Cachi. To face p. 161 ie LOWER NEIGHBORS OF CARTAGO 161 or two passengers was formed by a box hung from a pulley sliding on iron cables. The box ran to the lowest part of the cable by gravity and the passengers pulled themselves the rest of the way by a rope working over pulleys on shore. The bridge was rebuilt more strongly and with a longer span in IQII. There are three well-marked river terraces at this part of the Reventazén. In the potreros of the upper terraces we saw many trees, mostly pord, which had been felled to kill the mistletoe—matapalo—with which they were infested. This was not to save the porés remaining but because the birds eat the mistletoe berries greedily and so distribute the seeds to the coffee trees, to which the parasitic mistletoe does great damage. The lowest terrace is a fringe of beau- tiful flat meadows not much above the usual river level, which extend half a mile or more along the right bank. They contained numbers of fine large old trees, which were amazingly rich in epiphytes. Some of these we had not seen before, such as Phyllocacti, a small bromeliad with beautiful pink flowers but wickedly barbed and spiny leaves, new begonias, mistletoes and an amarantaceous plant (Telanthera mexicana), which although not truly a vine grew among other stronger plants and dangled its long, weak, slender stems with their little heads of whitish flowers ‘high above the ground. Although we examined a number of bromeliads for dragonfly larve we found none here. Mr. Lankester told us the trees on this terrace were mostly legumes of various species, but they were thickly overgrown with many kinds of figs (Ficus sp.) which have the “con- strictor” habit. The seeds of such plants germinate some- where on the host tree and their roots and branches grow about the host so closely that the latter is strangled and ul- timately dies. In time the original tree-trunk rots away and the fig, a marvelous tangle of thick roots and stems, 162 A YEAR OF COSTA RICAN NATURAL HISTORY stands alone. Most of the trees in these meadows were in the intermediate stage, with the original tree still living. Growing in patches near the water was a striking Urtica- ceous plant (Urera sp.) having stems as thickly set with prickles as they could be placed and large, spiny, deeply in- cised leaves whose upper sides were a very dark, shiny green, the under sides a rich and beautiful purple. Walking downstream we came to a small patch of to- bacco with the grower’s hut in the middle, went through a cane-brake where the wild cane was several times higher than our heads and finally reached a little back water left by the river in the last flood. Here a few dragonflies (eight species) were flying and were captured. Several times we stopped at some little sandy beaches on the main stream to dredge and were fortunate enough to find six of the curious flattened sand-grubbing larve of a Gomphine. Some of the back channels of the river contained incred- ible numbers of black tadpoles about an inch long; there were many places a foot wide and two or three yards long that were literally covered with them as thickly as they could be packed, looking like black cloth on the gray-white sand of the bottom. Farther upstream, in place of these enchanting meadow strips, the river winds through a stretch of cobblestones and sand, in several channels. Some of these seemed likely places for dragonflies, for the channels contained but little water which flowed much more slowly than that of the main stream. I explored them carefully and minutely, but in vain, although grasshoppers (Heliastus venezuele) colored like the sand except for their red hind wings and red tibiz, were present here. A little farther back from the water and five or six feet higher, were patches of tobacco, and be- tween these and the cafetal was a boggy spot which yielded some insects of several groups. Many small streams cross LOWER NEIGHBORS OF CARTAGO 163 the road and at one of these, constantly alighting on the muddy border, were beautiful butterflies (Timetes marcella) with a wing-spread of two and one-eighth inches. The fore wings are chiefly orange, dark brown at tip, paler brown at base, the latter brown crossed by four or five darker brown lines parallel to the body. Each hind wing has on its hind edge two tails, the outer longer (one-half inch as compared with less phan one-eighth inch for the inner tail). The hind wings are chiefly brown of nearly the same intens- ity as that of the base of the fore wings, bounded with orange on the anterior half of the outer wing-margin, and having a patch of pale purple or reddish-violet in the middle third of the wing, almost from wing-base to near the base of the tails. The under sides of the wings are chiefly buff crossed by lines or narrow stripes of white parallel with the body. Much coffee, sugar-cane and tobacco are now grown in the bottom and side of the main Reventazén valley and of its side valleys, in the neighborhood of Cachi. Mr. Lankes- ter said that practically all the cultivated land around the village was forest twenty-five years earlier. The opposite, north or left bank of the river is the site of an older settle- ment known as Ujarras. Cachi has been a favorite spot for entomological collectors. Here in 1877 came H. Rogers gathering specimens for Messrs. Godman and Salvin’s Biologia Centrali-Americana, but the exact date, in months, of his visit is unknown. Mr. Lankester told me that none of the present inhabitants of Cachi remember him. The late Prof. Paul Biolley collected at Cachi in May, 1905. Messrs. Schaus and Barnes were here at least twice, their latest visit being in September and October, 1909. Owing to the larger extent of forest at the time of Rogers’ sojourn here, the insect and other life of the region must have been quite different. Our meager results, so far as dragonflies 164 A YEAR OF COSTA RICAN NATURAL HISTORY were concerned, may have been due to the season of the year, or to the amount of cloud and rain. On March 6, with Mr. and Mrs. Lankester, we left “Casa Grande” about ten o’clock for Orosi and if possible Orosi Falls. Our road led west along the valley of the Rio Grande or Reventaz6én, crossed the Rio Macho on the long iron suspension bridge and south through Orosi village with its picturesque old church. About half a mile from the vil- lage the road passed a small hot spring. The water, which. was hot enough to steam vigorously and had a strong sul- phurous smell, bubbled up in a circular walled-in enclosure and ran away down the roadside ditch. It is supposed to have some medicinal properties. This seems to be the “Her- videro de Orosi’”’ whose waters were analyzed in 1865 by Luciano Platt, whose report has been mentioned on page 148. We continued south, taking in general the left branch of each fork and passing many coffee farms. By the roadside we saw a huge rock fifteen to twenty feet high, which was flat-topped and bore a most wonderful mass of ferns and the red and orange orchid (E. radicans) often referred to in these pages. The road finally led us to a bridge closed at the far end by a gate and we were then in the property of Senor Tomaso Gutierrez. There was a number of peon houses, chiefly on the right side of the road, and some men passed us carrying a large rodent two feet long with brown and white spotted fur and a short tail. This was the “tepez- cuintle” (Coelogenys paca) which is greatly esteemed as a game animal and considered delicious eating. The hacienda stood on a steep hill near these houses and Mr. Lankester rode up to ask permission to pass through and for directions to the falls, as he had never been all the way himself. Per- mission was readily granted, but reports of the road were not encouraging as the trail was said to be completely overgrown and impossible to find or follow unless we could get a peon LOWER NEIGHBORS OF CARTAGO 165 from the last settlement to cut it out for us. We rode on through a boggy potrero with fine trees here and there. Among these were some magnificent fig-trees which instead of growing on other trees had sprouted on the tops of large boulders, sending their roots down around the boulder as constrictor figs do around their hosts and completely embrac- ing it. The effect of such a tree standing so high above the general level of the field was extraordinary. The potrero bordered the Rio Macho and gradually be- came more wooded as the valley grew narrower. There were many beautiful flowers; a short distance within the timber we gathered a splendid Columnea draping a big tree. This vine has small fleshy opposite leaves and long, tubular, bright red blossoms. In another place we passed bignonia- ceous trees with large trumpet-shaped flowers, in color creamy white shading gradually to rich lemon yellow and exquisitely perfumed, the delight of bees and humming- birds. Here were large-leaved melastomes with peach-pink blossoms an inch across, there small-leaved melastomes with clusters of delicate waxy rose and white blossoms. On the trees we saw the usual epiphytes and new ones with rosy pink blossoms and strange begonias. Looking at the hillside across the valley, the trees were astonishingly like an early autumn hillside at home, some being dark green, some light green, some yellow and others almost orange, but here the colors other than green were due to the new leaves, not to the dying foliage. Certain yellow trees with well-rounded heads were identified by Mr. Lankester as “‘tirra,”’ a tree whose botanical relations have not been determined. They are often left standing when other trees of the forest have been cut away because the wood is so hard and tough and difficult to work. Through some bars we entered a little strip of close woods arching over a road leading to a brook with an awkward 166 A YEAR OF COSTA RICAN NATURAL HISTORY ford; then it passed the last houses on the way where we tried to get a peon to cut the trail. The man refused and probably thought he would force us to turn back. We went on nevertheless and he must have thought better of it, for when we left our horses to begin the climb on foot he had caught up with us and was ready to show us the way. From these houses we rode over the shoulder of a hill with sides sloping sharply down to the river. The road here occupied a ledge cut on the hillside and often wear or slides had nar- rowed it, sometimes to a mere foot-track. It was quite steep, and our horses being fine vigorous animals objected to slow climbing and on every hill broke into a gallop and went tearing up at a pace that took one’s breath away until one grew accustomed to their method of hill-climbing. Beyond this hill an unexpectedly new and good gate let us through the last fence and we soon reached the foot of the first really bad bit of road, where we left the horses tied under a beauti- ful “San Miguel” tree. Starting on foot at one o’clock we climbed first up a very steep, rocky, clayey road, through rather scattered and open woods. There were many crotons and figs and in sunny spots ageratums and yellow thistle-like composites, in others ferns and caladiums. After a hard pull up this hill of about half a mile we came on a more level boggy place where some cattle were feeding. It was here that we found the gor- geous amaryllid (Bomarea sp.) with long pendent clusters of black-spotted red flowers each one and a half to two inches long. After this the forest grew denser, the trees constantly bigger and grander, with the most luxuriant epiphytic drapery and festoons of lianas. There were many fine tree-ferns also, some with the fronds uncurled, others in varying stages of unfolding. The roar of the falls was now constantly in our ears, filling the dark forest with its thunder. LOWER NEIGHBORS OF CARTAGO 167 We soon climbed over a small ridge where the guide’s work with the machete was indispensable, for the trail wound about among huge clumps of “Hoja de Pato,” an unde- scribed Aroid with gigantic arrow- or heart-shaped leaves and strong, partly recumbent stems three to six inches thick. These had overgrown the trail completely and had to be cut away at each step so that we could follow. The rich black soil was soft and deep, often excessively slippery and the trail was a difficult one to travel. At last, at three o’clock, we came out upon a ledge commanding a fine view of the upper part of the falls and a few minutes later reached a large rock near its foot, from which we could see the whole of the cataract. The aneroid recorded 4950 feet at this ‘point. The waterfall was magnificent and it was singularly im- pressive to come upon it in the midst of the solemn grandeur of the tropical forest. The drop was almost perpendicular over sheer rock faces and about one hundred and fifty feet high. In the rainy season it must carry an immense volume of water, for the ‘winter’? channels were visible beside the present cataract, and even now in the dry season it was a large stream. It fell into a deep pool, which showed that it was much larger in “winter” and, turning at a sharp angle, rushed away down a steep, rocky bed in a series of rapids. A constant cloud of spray rose from the falls and the force of the water was too great and too direct to allow of vegetation clinging to the rock walls. These were polished and smooth in many places and bare wherever the water ran. The spray and the rain, which had been intermittent ever since we left the horses, added considerably to the pho- tographer’s difficulties. The water felt intensely cold and we saw no dragonflies near the fall nor indeed in this forest. In sheltered places among the rocks near the falls were patches of Gunnera insignis, the spiny plant with huge cir- 168 A YEAR OF COSTA RICAN NATURAL HISTORY cular plaited leaves which we saw in the crater of Irazit. The constant spray and the evaporation from the cold water no doubt keep the temperature of this place much lower than that of the surrounding forest, enabling such a moun- tain plant as Gunnera to live here. The day was unfortunately cloudy most of the time, with drifting masses of cloud and mist in all the valleys so that we were quite unable to photograph the beautiful vistas of hills and valleys. This we regretted very much as it is unlikely we shall ever be along this road again. It isa difficult trip from Cartago and we would not have seen these falls except for Mr. Lankester’s kindness. The Orosi Falls are well worth any labor required to reach them and our trip thither was one of the finest and most beautiful we had in Costa Rica. “SPUIA uenf 7e UOTILIG perol[ivy pue J9181IZ) plo Jo evunsey euL 4 a ae ot acer ae i To face p. 168 _ ‘azIs [eInjeu yYpem1-su0 jnoqe—vuvisynm vunmnpy—Aang ap ofM ‘seulA uenf ‘sejopugdoig jo sisdN « To face p. 169 CHAPTER IX JUAN VINAS—AROUND THE LAGUNA Juan Viiias was, of all places in Costa Rica, the one at which our most interesting scientific observations were made.. This was partly because of its situation, at about 3300 feet (1100 meters) above sea-level on the Atlantic slope, where bits of the original forest still existed, and partly because we were able to find accommodation here on repeated visits, thanks to Mrs. Clyde Ridgway. Many an excellent locality for the naturalist in Costa Rica is unavail- able because of the difficulty of obtaining shelter and food, unless the traveler has the means to transport such neces- saries to the place where he would be, an undertaking in- volving much expense. We visited Juan Vinas for periods of varying length in June, July-August, September—Octo- ber, December, February, March and April-May, so that we were able to note changes during the year in greater de- tail than in any other locality except Cartago. The canon of the Reventazdon is here quite narrow and about 1500 feet (450 meters) deep, with precipitous sides. The Atlantic Railroad, following the left (north) bank, occupies a shelf or cutting which at Juan Vifias station is 800 feet (245 meters) above the river. The station lay near the southern side of what seemed to be the floor of an old crater. The southern wall of this crater, lying between the railroad and the river, was very low. On all other sides its height was much greater and in places reached 700 feet (213 meters). The north and south diameter of this depression was greater than the east and west diameter and the floor was nar- 170 A YEAR OF COSTA RICAN NATURAL HISTORY rower at the northern than at the southern end. The stretch of railroad track within the crater was rather less than half a mile long. From the station a cart road de- scended by a series of zigzags to an iron bridge spanning the Reventazén and continued downstream to the Indian settlement of Tucurrique. Another road from the station ascended the sloping sides of the crater and, reaching the rim, traversed more level country to the village of Juan Vinas distant about two miles from the railroad. There were thus three easily accessible levels, that of the railroad at 3300 feet, that of the village at 4000 feet (1220 meters) and that of the river bottom at 2500 feet (760 meters). The vegetation of these three levels differed much, the region around the village being largely under cultivation, that of the river bottom most undisturbed, with of course corre- sponding differences in the fauna. It is logical therefore to group our observations in connection with these levels. Alongside the railroad station stood the restaurant man- aged by Mrs. Ridgway, an American, where the train leaving San José each morning for Limon stopped about 11 o'clock for breakfast. Here we too had our meals. We usually slept and worked, when not collecting, in a little cabin across the tracks from the station and some fifty feet above them, on the low southern rim of the crater. It con- tained two tiny rooms, not communicating but having each a door to a little porch, and looked like a cozy doll’s house as it nestled close to a huge tree whose branches partly over- hung it. Behind were two more trees, heavily draped with Tillandsia. These trees had nothing like the spread of branches of some of our oaks or sycamores but they were immensely tall, the stem being bare of branches for many feet above the ground—a condition due to the crowding in the forest and the effort each tree must make to reach light and air. Now that the forest had been cut down around JUAN VINAS—AROUND THE LAGUNA 171 them the trees were too old to send out more branches. The cabin stood in a little garden enclosed by barbed wire. On one side was a patch of plantains mingled with palms, cane and a number of Maranta-like plants. On the other were a few huts made entirely of scraps of corrugated iron, where some peon families lived. In the space between these houses and the railroad was a perfect thicket of the hand- some small milkweed with brilliant red-and-orange blos- soms (Asclepias curassavica) so common in Costa Rica. Our first stay at Juan Vifias in the latter days of June found the ground about our cabin dug up ready for beans to be planted. At the end of July the beans were one to two feet high. By the last of September, the crop was mature, the vines dried up and a negro was at work clearing the ground for a new crop to go in in a week or two, for already there was an amazing growth of weeds and bushes. Gui- tites that had been cut to the ground in June and again in July now showed stalks as high as one’s head and an inch thick, and the guitite fence posts,—really bare posts at our August visit—were bushes almost as high as the cabin. On the evenings of September 30 and October I we no- ticed some luminous insects creeping over the soil imme- diately in front of our cabin, where the beans had been cleared away. We obtained a few specimens each night and kept them alive in a vial for two days or so before pre- serving them. They were beetle larve of the Elater fam- ily—that of the common spring- or click-beetle of the United States. In daylight these larve were pale yellow in color and presented no remarkable difference from other larve of this family, but at night they were very attractive from the glow which they produced. Their lengths ranged from 4 to 22 mm. when resting quietly, but they could stretch out to a little more than these dimensions. Their light-pro- ducing organs were arranged in eleven pairs as follows: 1, 172. A YEAR OF COSTA RICAN NATURAL HISTORY occupying all of the head; 2, the anterior end of the pro- thorax; 3-11, the nine abdominal segments, one pair in each segment, that of the last segment being the largest of the abdominals. Pairs 1 and 2 were much more constantly in activity than the others. The number of the abdominal luminous organs in activity varied from time to time. In some individuals there was only one luminous area on a given abdominal segment, i. e., the right or left organ was absent altogether, functionally at least. Usually when these larve were crawling over the ground only pairs 1 and 2 were luminous; when we attempted to pick up the insect the abdominals also became luminous as the larva ran to escape the fingers. The color of the light produced by all the pairs of luminous organs was the same—green. These larve could run very fast. We could not detect any odor from them nor on the fingers after handling them. The use of this light to the larve is problematical as in this stage of development it cannot serve as a sexual attraction. Unfortunately, we did not attempt experiments to determine whether the larval light might aid in catching prey. Fireflies (Lampyride) were conspicuous also on these evenings at the beginning of October, usually high, or other- wise inaccessible; we noted their presence here in March also. The light they produced was green. In June we had some bright moonlight, the moon being in the second quarter; but several times, between seven and eight o’clock, the mist stole in as a slender finger from the eastern end of the crater, where the railroad entered through a deep cut. Rapidly the mist filled the entire crater, blotting out its edges all around and cutting off the direct moonlight. The floor of the old crater, which our cabin faced, was in great part swampy and had been much wetter in earlier JUAN VINAS—AROUND THE LAGUNA 173 years before any attempts were made to drain it and hence was locally known as the laguna. At the time of our first visit, two ditches had been dug across it. One of these was near its eastern edge, the other crossed it from northwest to southeast approximately, in continuation of a small brook descending the crater slopes. The first ditch was begun about October, 1908, Mr. Cochenour said; at its northern end it received the waters of the Rio Naranjo, which on leaving the crater at,the southeast edge formed a waterfall in a narrqw gorge and rushed down to the Reventazén. Be- tween July and October, 1909, a third ditch bordering the laguna on its southern edge had been dug. The result will, of course, be that the swamp will become dry land, even- tually to be planted in maize and then in coffee, and plant and animal life will be greatly changed. Part of its area was used for grazing cattle during our visits. From the collector’s point of view the ditches made the inner parts of the laguna more accessible and to that extent at least were an improvement. In February and March the south and west ditches were bordered with great quantities of Eleocharis and other water plants, while the water itself contained in places much stone- wort (Chara or Nitella). Toward the center of the laguna was a large patch of handsome mallows (Hibiscus sp., near uncinellus but distinct) with vivid dark red flowers and a spiny stem very evil to handle. March also showed many red-stemmed, pale pink- or white-blossomed begonias, which with some clumps of low spiny-stemmed tree-ferns were perhaps relics of a previously wooded condition. When- ever we went into the laguna we usually found some new object of interest; on October 5 it was the curious plant Monstera with its lacerated leaves, also a forest dweller by nature. In the same visit the edge of the laguna showed the red torches of the gorgeous-flowered Odontonema flagellum, 174. A YEAR OF COSTA RICAN NATURAL HISTORY one of the Acanthacee. At other times we noted quanti- ties of Piper hirsutum and a vine with bright red flowers, Vitis rhombifolia. Along the western edge of the laguna were large groves of guayabo trees (from whose fruit “guava” jelly is made), full of fruit which in July was still un- ripe. In the narrower, northern part of the laguna was a large vegetable garden cultivated by a Cuban, who showed us some pieces of pottery, broken but quaint and interesting, that he had himself dug out of old graves in clearing the land for his garden. Large portions of the sloping crater-walls enclosing the laguna on the eastern, northern and western sides were planted in coffee, the cultivated area on the west side being much less than on the other two. On June 27 we wandered through the western coffee patch, which was on a steep slope of very loose soil so that we had to be constantly on our guard lest we lose our footing and roll down. The coffee here was shaded by plantains and a few large scattered forest trees left standing. P. caught a lichen-imitating katydid (Lichenochrus sp., probably marmoratus), whose markings exactly resembled the mottled green-gray lichen on the tree- trunks. Its body was 18 mm. (three-fourths inches) long; it had projecting knobs on head and thorax and a row of knobs along the middle of the upper surface of the abdomen; the wings were but 3 mm. long, indicating its youthfulness, as the adult insect has the organs of flight well developed.? Later we watched a large red and black solitary wasp work- ing at its burrow. There was a limp green katydid lying outside the burrow when we first saw it and the wasp was fussing around the entrance. She soon went down, head first, and after an interval backed out with a pellet of earth in her jaws. This she passed between her legs from pair to 1See the figure by Rehn in the Proceedings of the Academy of Natural Sciences of Philadelphia for 1905, p. 816. JUAN VINAS—AROUND THE LAGUNA 175 pair to the hindermost when she gave it a hard kick and sent it flying. This excavating was repeated about a dozen times, and sometimes the pellet was kicked a distance of three to four inches. The wasp then seized the katydid in her jaws— I think grasping it by one antenna—and backed in with it. She stayed down several minutes and a great humming went on; probably she laid an egg on the katydid but that we did not see. Coming out head first she ran to the loose earth she had carried up and kicked a little pile toward and into the burrow. Turning round she shoved down with her head such particles as were caught on the edge of the tunnel, then ran in head first and probably rammed the earth hard with her head; at least that is the usual procedure. This she repeated ten or twelve times, then came out and flew away leaving the mouth of the burrow open and unpro- tected. We waited for her nearly half an hour, but in vain, so we shall never know whether she abandoned the burrow or was unlucky in her hunting trip. Among the coffee on the eastern slope of the crater stood a tree with hanging nests of a yellow-tailed oriole or “oropendola.” As we could approach very near to these nests we decided on June 30 to photograph them with— if possible—one or two of the birds themselves on the nests. So everything was made ready and we waited three-quarters of an hour for the birds. But although they sat on nearby trees and finally actually visited the nests, their entrance to the nests was so skillful and their stay so short that our pictures although good of the nests do not show a single bird. There are two commonly found species of oropendolas in Costa Rica. This one at Juan Vinas was the upland and Pacific species (Zarhynchus wagleri wagleri); its average length is fourteen inches in the male, ten and one-half in the female. The lowland oropendola of the Atlantic slope (Gymnostinops montezuma) is larger, nineteen and one-half 176 A YEAR OF COSTA RICAN NATURAL HISTORY inches long in the male, sixteen inches in the female, on the average. Both species are glossy black and dark chestnut with mostly yellow tails. Technically, Zarhynchus has the bill much swollen basally, forming a broad rounded frontal shield whose width is decidedly greater than the distance from the nostril to the tip of the maxilla, while in Gymnosti- nops the basal portion of the bill is less expanded and the frontal shield is less than half as wide as from nostril to tip of maxilla (Ridgway). The chorus of bird-song around our cabin before sunrise each morning was always a pleasant feature and we regretted _ that we were not ornithologists to appreciate it fully. These mornings alone were enough to refute the popular misstate- ment that “birds in the tropics have no song.” In March the drumming of the tree-frogs was incessant after nightfall, accompanied by a two-noted scream which, so far as it went, suggested a whippoorwill and which could be heard at in- tervals all night and after daybreak although it ceased before sunrise. In April we occupied a room in Mrs. Ridgway’s eating-house instead of our cabin. This room had unglazed windows, guarded by solid shutters of galvanized iron with a wooden frame, so that when closed to keep out rain both light and air were also excluded. Above the shutter was an open scroll work designed probably to admit a trifling amount of air when the shutter was closed. One day as we sat inside this room two wrens, after alighting on the top of a shutter, flew to these small openings and peeped in at us. Their little brown heads and round black eyes were very amusing as they were thrust through hole after hole in the scroll work, and the owners thereof scrutinized us carefully first with one eye then the other. The birds continued this for about ten minutes before they finally flew away. A handsome woodpecker with a large crest of flaming red feathers frequented the laguna in the last of July. He had Guarumo Trees (Cecropia sp.), one with wasps’ nest. A Trail of Leaf-cutting Ants on a vertical bank. Pieces of Leaves Collected by Leaf-cutters, with Some of the Ants, Juan Vifias. Openings of a Nest of Leaf-cutters, near Las Concavas. JUAN VINAS—AROUND THE LAGUNA 177 a single sharp note followed by a long flute-like trill on sev- eral lower notes. It was interesting to see him look sharply at the tree-trunk he was standing on, with his head first on one side then the other, then with head drawn back, utter his loud song and as he finished the last note give a few re- sounding taps to the tree. Toucans were not often observed near our cabin but on April 26 one flew across the road close enough for his immense green bill to be seen plainly. As it flew with the grotesque bill held straight out it looked for an instant like a bird which had stuck a banana on its beak end-wise and was flying away with it. We could not get very near, but the bird perched on a bare tree where it showed clearly against the sky. The railroad offered a means of walking about in all seasons of the year, especially in wet weather when the cartroads were deep in mud, and we used it very frequently. It af- forded fine views of the Reventaz6n cafion both up and down- stream after passing out of the crater. Along it, here and there, occurred “‘guarumo” or Cecropia trees, which have rather sprawly, ‘“candelabra-like” branching and whitish trunks, each branch tipped with a cluster of large, deeply divided palmate leaves. They are highly typical of tropical America. The hollow stems are divided by cross partitions at the nodes which show on the outside of the trunk and are often, but not always, inhabited by belligerent ants of the genus Azteca. Although it has been held that the foliage is protected by the presence of these ants, Wheeler (1913) “after examining hundreds of young and old trees” has come to the conclusion of von Ihering and of Fiebrig that the ‘‘Cecropias have no more need of the Aztecas than dogs have of their fleas.” One of these Cecropias, to the west of the crater, had a large pendent paper-covered wasps’ nest on one of its branches and so became a definite land- 178 A YEAR OF COSTA RICAN NATURAL HISTORY mark for us. Noted on our June and August visits, the nest at the end of September was still the home of an active colony of wasps. On September 29 we gathered, along the track, some flower-clusters of a leguminous vine, the “Ojo de buey” (Mucuna mutisiana) which was then in full bloom. The vine clambers high over trees and rocks, dangling its in- florescences like plummets at the end of thread-like peduncles two to three feet long. We had noted these inflorescences on earlier visits when they were younger and the peduncles consequently shorter. The papilionaceous flowers are waxy, greenish-white in color and about two inches long, with a keel very long in proportion to the standard. The pedicels are about the length of the flowers, which are arranged radi- ally owing to the crowding of all the flowers into the last inch or two of the peduncle. Such an inflorescence may be eight or nine inches in diameter, and presents a most curious appearance as it sways in the breeze or hangs motionless in some breathless little nook. The flowers are followed by large flat pods eight to ten inches long and almost two inches wide. These when ripe open along both sutures disclosing a few large round flat beans nearly an inch and a half in diameter, which adhere alternately to one or the other valve. To their shape and rich black color the plant owes its popular name of ‘‘Ojo de buey”—Eye of the ox. Numerous flowers grew along the track. In October we saw blooming Castilleja communis, much less showy than the C. irazwensis we found in the crater of Irazt, a pas- sion flower with curiously truncated leaves (Passiflora lunata) and the red-flowered Russelia sarmentosa. In early December the most abundant wild flowers here were the red- flowered Isoloma wagneri, whose blossoms suggest those of our monkey-flowers (Mimulus), Browallia demissa, a small Solanaceous plant with blue flowers, and a yellow-flowered JUAN VINAS—AROUND THE LAGUNA 179 composite with heads two inches in diameter. The blossoms of the [soloma were attractive to the large black and yellow swallow-tail butterfly Papilio thoas, whose wings at this time were much torn. Both in December and February we found a species of Siphocampylus blooming on the rail- road banks, where also grew an odd little Urticaceous plant (Pilea microphylla) and a handsome small poke (Phyto- lacca decandra) with bright purple flowers. Among the cinders of the roadbed were small Euphorbs (Euphorbia nutans). Nests and paths of the leaf-cutter or “parasol” ants (Atta) were common about the laguna and railroad. These insects derive their English names from the habit of cutting bits from leaves or other parts of plants and carrying them in their jaws, balanced over their heads or bodies, to their underground nests. Thomas Belt, during his four years’ stay in Nicaragua, and independently Fritz Miller in Brazil, ascertained that on these leaf-fragments in the nests grows a fungus, which is used by the ants as food; the leaf-fragments themselves are subsequently abandoned as refuse. Alfred Moeller, working in Brazil, found that the parts of the fungus used by the ants are pale white corpuscles, .25 to I mm. in diameter, which he called “ kohl-rabi clusters ” and which, he believed, are due to culture by the ants, as they are not formed when the fungus is experimentally grown apart from the influence of the ants. On some firmer ground on the west side of the laguna was quite a large nest with many openings and well-defined and hard-trodden ant paths leading to it. On October I we found ants (Atta cephalotes) carrying pellets of refuse material out of several of the openings but very few, almost no, ants carrying leaves or other burdens into the nest. Among the ejecta was an Aphodiid beetle (Atentus sirigicauda) 5 mm. long, shining blackish, although thestrongly knobbed antennz 180 A YEAR OF COSTA RICAN NATURAL HISTORY are reddish; the head and thorax have irregularly arranged punctures, each elytron has nine grooves containing punc- tures. This beetle probably acts as a scavenger. At one large opening were perhaps a dozen big worker or soldier ants, with enormous heads, apparently being dragged out by much smaller workers; others of the soldiers seemed to be fighting among themselves. Something was evidently amiss. We examined the ant-paths both near the nest and at a distance; where in July were hundred of workers each bearing its piece of leaf, fruit or flower, were now only one or two, laden as usual but lonely. Alongside the railroad west of the crater was a wonderful path worn by the tiny ant feet along the face of a steep bank. It could be clearly traced as a hard bare streak for fifteen to twenty feet and was one to two inches wide. The bank was thickly clothed with grasses, ferns and other plants and this bare ledge was plainly visible from the opposite side of the railroad cut. About 8 in the morning of July 22 the ants were only beginning to work and few leaf-fragments were being carried. When we passed about 10.30 the sun was hot and there was tremendous activity among the leaf- cutters and at 3.15 they were still carrying leaves into the nest although it was raining. Across one ant trail there were three columns of small black ants, which caused some interruption and excitement. Leaf-cutters without burdens met and crossed antennz with the blacks, but there was no actual fighting. On September 28 these trails were abandoned and at the mouth of the nest only one leaf-cutter was seen. ‘These ants travel great distances, it is said as much as half a mile, to the plants whose leaves they cut up, and it is quite possible that this large nest sent armies to the coffee plantations on the crater side, thereby drawing upon itself the wrath of the planter. There is constant warfare upon the leaf-cutters by the latter, who fights them JUAN VINAS—AROUND THE LAGUNA 181 with ditches of running water which the ants are unable to cross, by flooding the nests and by exploding them with carbon-bisulphide. The latter is said to be the most suc- cessful method yet devised. On December 2 the leaf- cutters were crossing the railroad west of the crater and another busy column was marching along the horizontal roadway. A third nest was under the road ascending’ from the sta- tion toward the village and had several openings in the road- bed and from these, on August 2, ants were emerging, some carrying pieces of leaves which had been previously taken into the nest. These pieces were carried five or six inches from the openings and dropped on the pellets of earth which were likewise brought up by the ants in much greater num- ber. I examined two of these leaf pieces under my lens but could find no cause for their rejection. Reasons mentioned by Belt in his account of this action (Naturalist in Nica- ragua, 2nd edit., p. 82) did not seem to apply, as no rain had fallen during the day, nor did the pieces appear in the least withered; one had what looked like fungus marks acquired while on the tree, but the other did not. Still another rather small nest was alongside of the rail- road tracks within the crater but east of the station. On passing it on September 29, I noticed two blue-black wasp- like insects running over the top of the nest, one of which entered an opening and entirely disappeared from sight. In a short time it came up, walked about a little and went down again, not to reappear as long as I watched. Two leaf-cutters came out of this opening at intervals after the visitor had gone down the second time and were apparently unconcerned. Later I caught the other of these two blue- black insects, which proved to be a fly (Eristalis scutel- laris), whose manner of walking, of flying and of folding the wings down over the back of the abdomen is like that of a 182 A YEAR OF COSTA RICAN NATURAL HISTORY wasp. It was a female, five-eighths of an inch (16 mm.) long, shining bluish-black with a bright yellow spot on the scutellum (hind part of the thorax) and a pair of yellow spots on each of the second and third abdominal segments. The head was relatively large, hemispherical and chiefly occu- pied by the eyes. The wings were smoky. This species is found also throughout much of South America. As the larve of Eristalis (a genus occurring throughout the world) are known to live “in decaying vegetable matter, manure or in soft mud impregnated with decaying vegetable matter” (Williston), possibly the larve of E. scutellaris may feed on the decaying leaf-fragments gathered by the ants. A few yards from the beginning of the zigzag road down to the Reventaz6n were, in June and late July, two streams of numerous leaf-cutters crossing the road, but on Septem- ber 29 there was only one stream composed of very few in- dividuals. At this time of year, therefore, there seemed to be a marked decrease in the ants’ activities in a number of nests at Juan Viias. Without having given special attention to the distribu- tion of the leaf-cutters in Costa Rica, our notes mention them as occurring not far from a swamp near Paraiso, where were two nests near the railroad track at about 4400 feet elevation. Into one of these, on October 11, the ants were carrying many more bits of plants than was the case at Juan Vinas twelve days earlier. Again, south of the Paraiso road and nearer to Cartago, we came, on October 26, upon a large leaf-cutters’ nest directly in the road, which was in process of being moved around the nest by trails made by horses and mules to avoid it. A.’s horse shied and showed uneasiness until he was safely past, evidently fearing the ground would crumble or his feet sink into holes. This colony was vigorous and flourishing, to judge from its size and the trails leading to it, and was located less than 500 JUAN VINAS—AROUND THE LAGUNA 183 feet lower than Cartago. On March 14 we rode east of Cartago primarily to capture ant pictures, but we saw few active trails except in deep shade or confused with vegeta- tion. On the road near Las Céncavas we photographed a large nest with half a dozen conical openings like little vol- canic craters, over a foot high. We attempted a picture of a bridge composed of sticks and branches by means of which the ants were crossing a little rivulet. But every attempt to clear away intervening vegetation and let in more light caused the ants to slip over to the darker side of the bridge, so that their “bashfulness”—(or negative heliotropism?)— kept most of them out of the pictures. On the Pacific side, a large and active column of leaf-cutters was observed on the carretera nacional between El Alto and Tres Rios at an altitude of 4650 feet, on March 17, and we have specimens from Alajuela dated September 9. These last two lots represent another species, Atta (Acromyrmex) octospinosa, whose differences from cephalotes may be seen very easily from the accompanying photographs. CHAPTER X JUAN VINAS—ABOVE THE LAGUNA The road leading upward from the railroad station to the village of Juan Vifias ran up the west and northwest sides of the crater and gave a good view of the laguna, its vege- table gardens and grazing cattle. The glossy leaves of the coffee on the sloping sides looked very pretty; in June the green berries were about half an inch in diameter. The coffee trees were planted in regular rows five to six feet apart, and the soil kept well cultivated and free from weeds, a peculiar form of machete being used by Costa Rican gar- deners for weeding and “‘hoeing.” Although this road was a ‘‘Government”’ road, it con- sisted of earth and loose stones, and after rains it became deep mud, into which, as our diary of July 29 records, “we frequently slipped to the ankles even in the best spots. It soon began to rain again and worse looking objects than we were by the time we reached our cabin it would be hard to find! Our shoes were so thoroughly soaked that it was tor- ture to get into them next day.” Another memorandum, of December 2, concerning this road, reads, “‘it was liquid mud to a depth of a foot or more, as estimated on the legs of a horse which went plashing through it.” The road crossed several small streams; over the largest, the Rio Naranjo, was an iron bridge on cement piers. It was a picturesque stream ten to twelve feet wide, tumbling over the rocks in little falls. On June 29 we found a rather uncommon insect on a long, freely-hanging vine in the deep shade of this little ravine. It was Colobopterus trivialis, an x i 13 : / Beery ; } AES RS BIS WN ie Leaf-cutting Ants. 1. Workers of Atta octospinosa from near Tres Rios (x2). 2. Workers of Atta cephalotes from Juan Vitias (x2.8). 3. Male of A. cephalotes from Cartago (x2.6). To face >. 184 1, 2. Army Ants, Eetton predator, Agua Buena Creek. x4. 3. The Larva of Dragonfly, Cora, Juan Vinas. x31/3. To face p. 185 JUAN VINAS—ABOVE THE LAGUNA 185 Ascalaphid related to the ant-lions but differing from them in having antennz as long as or longer than the body, and knobbed at the tip. This individual presented a most stretched-out appearance, for it had arranged itself with antenne, body, wings, and hind legs held parallel to the vine to which it clung, the wings being folded over the body in roof-wise fashion. The current of the Naranjo was swift but there were quieter pools in it here and there, owing to the numerous stones in its bed forming partial dams. At one of these pools, about 3 P. M. on August, 2 P. saw a male dragonfly (Brechmorhoga pertinax eurysema) poising with rapidly fluttering wings over a certain limited area. By gradually lowering the net to within a few inches of the insect and then bringing it down into the water as fast as I could, I captured the specimen. No other dragonfly was nearby. About 3.30 in precisely the same spot, another male of the same subspecies was poising in the same place and was caught in the same manner on the same square foot of water. About 4 o’clock a third male of the same subspecies was behaving similarly in the same spot, but this one I failed to take; it was alarmed by the attempt, flew away and I saw it no more. What was there in this spot which had such an attraction for this subspecies that three individuals took up their sta- tion there—not on some stick or stone but hovering in the air? We found three species of Brechmorhoga at Juan Vinas and its vicinity, all of which possess eyes which in life are a beautiful metallic bluish-green or greenish-blue. This color is found in the eyes of many swift-flying dragonflies but is not correlated, as far as I could see, with the habit of life as regards quantity of external light. B. pertinax eurysema on the upper Naranjo was in deep shade, B. vivax on the lower Naranjo near its junction with the Reventazon in less 186 A YEAR OF COSTA RICAN NATURAL HISTORY shade, while at Quebrada Honda, on August 1, another species (B. rapax crocosema) was active in bright unshaded sunshine. At three places along the Naranjo, above, at and be- low the laguna, at altitudes of 3500, 3300 and 2500 feet, respectively, we found in August dead-white plaster-like disks 18 mm. in diameter, which were attached to the surfaces of rocks or trees a few inches to three feet above the water’s surface. Each disk consisted of hundreds of elongated eggs, in contact with each other on their long sides; one end of each egg was applied to the rock or tree-trunk while the free ends were coated over with a layer of the white substance, which served as a protection. We brought one or two of these egg-disks into our cabin and to Cartago, but the eggs hatched and the larve died before we knew— owing probably to the absence of the water into which they would have fallen under normal conditions. These eggs were laid by one of the Sialide and belong to the same genus (Corydalis) as the “hellgrammite fly” of the United States. It was possibly C. crassicornis, which we took around the electric lights in the Cartago streets in May. In May brightly colored butterflies of the tropical Ameri- can groups Heliconide and the genus Ane@a were visiting the flowers on the roadside bushes. In March the whole road was redolent with a faint sweet perfume, suggesting both vanilla and mignonette, produced by a white-blossomed Verbenaceous shrub (Lippia myriocephala) ten to fifteen feet high. At this time the roadside was bordered with Santa Lucia of varying shades of blue, with what appeared to be a white-flowered Ageratum which was always taller than the other (three to four feet as compared with one to two feet), and another white-flowered composite with flowers resembling the Ageratums but with alternate leaves. About half an hour’s easy climb up this road brought one JUAN VINAS—ABOVE THE LAGUNA 187 to the edge of the crater, where the road bent to the left opposite a little cabin which formed a conspicuous land- mark. The country at this upper level was, in elevation and characteristics, not unlike that around Cartago, being used for grazing and agricultural purposes, so we seldom came up here—in fact only four times in all. On three of these visits the clouds obscured the distant panorama, but on the fourth, on March 22, the day was beautifully bright, and when P. reached the rim of the old crater and looked northward the huge masses of Irazi and Turrialba vol- canoes were very little hidden by cloud. It was the first time I had seen them from this place. The whole eastern and southern horizons also were filled with mountains, range after range, the farthest and highest, almost due south, lifting their peaks above a thin layer of white cloud. Seldom have I seen a grander view. In the foreground were the delicate greens of sugar-cane or the darker glossy greens of plantains or coffee; the nearer mountains were clothed with forest of a still darker green, the more distant mountains softening into blue. Beyond the crater’s rim, a short distance before reaching the village, stood a large two-storied wooden house with spacious verandas, enclosed with glass sashes, on each story. It originally belonged to Don Federico Tinoco, but at the time of our visit was occupied by the Messrs. Cochenour, two Canadians who, in codperation with the Messrs. Lindo, were engaged in the raising and preparation of sugar and coffee. We enjoyed their hospitality on two occasions, when they showed us their garden, sugar-mill and coffee beneficio. The garden was begun by Don Federico, who imported a number of trees such as palms and northern willows. One of the handsomest plants was an apocynaceous shrub with glossy leaves and large yellow trumpet-shaped flowers, tinted outside with red-brown. 188 A YEAR OF COSTA RICAN NATURAL HISTORY The sugar-cane plantations occupied the relatively level plateau out of which the deep cafion of the Reventazén and the crater of Juan Vifas station have been cut. They extended to the edge of this cafion, the sloping sides of which were covered with forest and dense undergrowth or, on the crater sides, with coffee and some maize. The sugar-cane was planted in rows five to six feet apart, the plants in each row being only six inches to two feet apart. On June 29 the cane was higher than a man’s head. In appearance sugar-cane is not unlike maize but the main stem is shorter and the leaves are given off much nearer the ground. A so- called wild cane also grows in Costa Rica, reaching a height of ten feet, and is used in the construction of native huts and houses. This is possibly the “‘cafia blanca” (Gynerium saccharoides). It is easily distinguished from sugar-cane because its leaves are definitely two-ranked so that in “end” view the plant looks very flat, while in sugar-cane the leaves are several-ranked, the bases forming a spiral around the main stem, and in no view does the plant appear flat. A small narrow gauge track ran from the sugar-mill through the plantation, so graded that the loaded cars went down by gravity to the mill but were hauled up by mules when empty. On September 30 we saw the cane brought in these cars to the mill, dumped at the head of wide slow- moving belts and fed to the crushers, where the juice ran down into vats while the almost perfectly dry fiber was re- moved to serve as fuel in the preparation of the sugar. The juice was boiled with a little sulphur vapor to purify it and the brown molasses was run into the centrifugal refiners, where we watched it turn paler and paler. First, second and third class sugars were made here—and none of it was beet sugar. All the sugar was used in Costa Rica, freight charges being so heavy and excess production so slight that it did not pay the producers to ship outside. The molas- JUAN VINAS—ABOVE THE LAGUNA 189 ses was all used in the government fabricas, where it was made into alcohol and alcoholic liquors, the making of which was a government monopoly. It was at the Messrs. Cochenour’s house, on May 31, that we first met Mr. William Schaus and Mr. John Barnes, then in their three years of collecting Costa Rican Lepidop- tera. During this period they obtained 1081 species of but- terflies and 4000 species of moths, 409 and 3350 more, re- spectively, than were previously known from Costa Rica. Mr. Schaus regarded Costa Rica as the richest area of all Mexico and Central America, a conclusion to which his pa- pers in the Annals and Magazine of Natural History, describ- ing some hundreds of new species, bear ample testimony. These Lepidopterists were now gathering the nocturnal moths attracted by the electric lights at this house and mill and at E] Sitio, 405 specimens having been the yield of the preced- ing evening. To these two friends we were indebted for many suggestions which guided our own footsteps in Costa Rica. The village of Juan Vifias was invisible from the railroad but could be well seen when one had reached the crater’s rim. It lay on a small hill in the center of a roughly circular valley. P. visited it only once, on March 22. The streets were lighted by incandescent electric lamps with reflectors, the current being supplied from the Cochenour power plant, but were unpaved and had either no sidewalk at all or only a slightly raised earthen path. There was a number of two-story houses, built chiefly of wood and corrugated gal- vanized iron; the two-spired church was also wooden. There were two schools, telegraph and post office, shops, etc. A road led north out of the village toward Turrialba vol- cano, passing first between cafetales and then potreros with cattle; half an hour’s walk brought one to a pretty brook which issued from a woods at the left, and after unit- ing with a similar brook at the road was crossed by a bridge 190 A YEAR OF COSTA RICAN NATURAL HISTORY with stone abutments. I ascended the first brook for a short distance, and spent about two hours (11.30-1.30) searching for dragonflies here, but with little success. When I entered the valley of this rivulet, Turrialba vol- cano was still in full view; when I came out on the road again, its whole upper part was hidden by the clouds. After twenty minutes’ walk this road bent sharply to the northeast, descending into a valley in which lies the settlement of El Sitio, another of the Lindo coffee and sugar estates. I how- ever followed a path running on north, which after some irregularities brought me to the top of a sudden descent of some hundreds of feet, at the bottom of which I could see the whitewashed houses and red roofs of El Sitio. Beyond El Sitio the ground rose again into a high forest-covered hill and beyond that to the slopes of Turrialba itself. At the edge of the hill was the two-storied white house of Senor don Juan Humana and farther to the right, on Turrialba’s slope, the houses and white-towered church of Santa Cruz. We were told that at these houses on the lower slopes of Turrialba the grumblings of the sleeping volcano were frequently heard. CHAPTER XI JUAN VINAS—THE WATERFALLS West of the crater along the railroad the wall of the cafion of the Reventaz6n rose to a height of 700 or more feet above the tracks, in some places perpendicularly, in others in a series of varying slopes and terraces. Brooks from the upper country on their way to the river formed cascades or waterfalls over the cafion sides. Two of these, accessible from the tracks, were often visited and yielded some most interesting insects. In our own conversation and diary we designated them as the “nearer” or “high” and the “farther” waterfall from our cabin. The nearer waterfall, which had the larger volume, slipped over sheer rock faces two to three hundred feet high. Wher- ever there was sufficient slant the adjacent wet rocks were draped with mosses, maidenhair and other ferns, Selaginellas, Commelinas, Tradescantias and a Streptocarpus-like plant. At about a hundred feet above the railroad track (which oc- cupied a narrow ledge cut out of the cliff face), the water usually disappeared beneath a mass of boulders and flowed underground through loose soil to reappear on the slope far below the track. Up over these boulders we climbed to the foot of the actual fall with its vertical rock-face. The amount of water over and among the boulders varied with the rains. On our first visit in June the water did not ex- tend to the rails as a surface stream as it did on December 2, while on February 14 the waterfall was as full, if not fuller, than on December 2. On February 18 there was so much less water coming down that it was again possible to climb 192. A YEAR OF COSTA RICAN NATURAL HISTORY to the foot of the perpendicular fall, and with no more than wet feet and slightly splashed clothing; we could see that the floods of the rainy season had changed the positions of many rocks and logs. On March 23 the quantity of water was less than in February, while on April 26 it had increased again in consequence of a rainfall unusually heavy for this month. These fluctuations, trivial as they may seem, in the volume of a stream which at best was relatively small, had their effect on the aquatic insect and other animal life in- habiting its course, the greater extent of surface water be- low the fall tending to carry such creatures to lower levels than the fall itself and so render them more accessible to us. The farther waterfall was not more than thirty feet high in its last stretch and fell into a little basin five feet above the level of the railroad, whence its waters made their way in a shallow ditch along the tracks for a quarter mile before finding their way toward the Reventazon. Unlike the open nearer fall, this waterfall was enclosed in a little recess well shaded from the sun. Growing around it were begonias, ferns, a delicate little Caryophyll (Stellaria ovata) and Urti- caceous plants including a species of Pilea and a coarse big- leaved Myriocarpa, which we found blooming September 29. The flowers of the Myriocarpa are borne in foot-long catkins in the axils of the leaves. The stamens are curled on them- selves and when ripe straighten out suddenly so that the pollen is violently discharged in a little cloud to a distance of several feet. On climbing to the top of this fall over the steep bank of crumbling soil, one could, with the eye, follow the stream in a series of low cascades. Two brooks in fact united just above and descended through a tiny, dark moss-lined gorge. Here, on August I, a pair of the delicate dragonfly Heterag- rion chrysops was egg-laying in the gloom produced by the The ‘Farther Waterfall” at Juan Viiias. To face p. 192 The ‘“‘Nearer Waterfall” at Juan Viiias. ‘The Larve of Dragonfly, Thaumatoneura. 1, female, x2. 2 and 3, male, x1.8. Ta face b 10% JUAN VINAS—THE WATERFALLS 193 overhanging vegetation. The fluctuations in volume of this fall were very slight. It was in or at the foot of one or other of these two water- falls that we found the transforming larve of certain dragon- flies, —Thaumatoneura, Argia talamanca, Philogenia carril- lica and Palemnema—of whose habits and early stages nothing had been known previously. From our observa- tions here and elsewhere in Costa Rica it appears that Thau- matoneura and this Argia are exclusively waterfall-dwellers, but that Philogenia and Palemnema live in other situations also. Standing at the bottom of the nearer waterfall on June 24, and looking up, I could see the males of two supposed species of Thaumatoneura fluttering through the spray, chasing each other or alighting upon the rocks or plants. One species (tnopinata) had a blackish-brown band across the middle of each wing, the other (pellucida) had uncolored wings. So heedless of the spray were they that it gathered in little drops on the wings or hung suspended from the hind end of the body. On this and following days we observed the mating of Thaumatoneura. The female had not hitherto been known; it has the zips of all the wings black and hence can be easily distinguished from the male even at a distance. Both sexes are large and conspicuous insects, having a wing- spread of four inches. While mating the male held its wings horizontal and expanded, those of the female being vertical and folded together. At the end of ten minutes the male released the female and at the same instant folded his wings together into the vertical position. The female, losing her attachment to the male, assumed a position a little below him on the same twig or rock. Subsequently she flew away to begin egg-laying. We watched supposed egg-laying in three or four cases. 194 A YEAR OF COSTA RICAN NATURAL HISTORY The female made the usual abdominal movements seen in the Zygopterous dragonflies generally, that is, those for in- serting the eggs into plant substances by means of a sharp- pointed ovipositor on the under side of her eighth and ninth abdominal segments. Some of these movements, however, were over hard wet rock so that it seemed unlikely that any eggs were placed on it. At other times the movements were made against moss and roots, the female occasionally bury- ing her abdomen for half its length in the wet mass. We had suspected such a habit even before we had seen the living female, and had carefully examined some of the wet moss, for larve, but found none. The ovipositing female is not accompanied or followed by the male. Indeed once a female flew close to a resting male without attracting any response from him. Males at least often remain on the same spot half an hour or longer without moving. Females seem less persistent, but one timed was ten minutes on the same twig. Except in pairing males, the wings of both sexes, when at rest, are held in the vertical position folded together, but every now and then an individual will open and close its wings once or several times in succession. Owing to the habit of Thaumatoneuras of spending most of their time over these vertical wet rocks, their capture was often only possible by climbing up on the rocks themselvés, or by hanging with one hand to some long dangling root or vine and sweeping the net with the other. Up to July 30 the males which we observed in the pairings of Thaumatoneura were in every case inopinata. On that day P. saw a male pellucida and a female fly to a dead branch on the side of the railroad cutting just west of the farther waterfall and about twenty feet above the tracks. There was no doubt that they were pairing, so it was highly de- sirable to capture them both. The steep bank was com- posed of reddish soil with loose stones and soft rock, wet JUAN VINAS—THE WATERFALLS 195 from recent rains. I climbed up as far as I could, until the steepness was too great to furnish a foothold, but the insects were still beyond the reach of my net handle. I descended and hastily attempted to improvise an additional joint but one stick after another broke, so I used my umbrella and again climbed the bank. But the soft earth gave way and collector, net and umbrella slid swiftly—and separately— down to the tracks. A second attempt had the same result. A third, made more cautiously, was so successful that the net with its handle supported on the umbrella was thrown over the accommodating pair, which had lingered for at least ten minutes on the same branch, and by gently drag- ging it down over the bank IJ at last got the insects safe in my hands. Then I saw that this female was so like the fe- males we had seen and taken pairing with inopinata males that a minute comparison would be necessary to determine whether there was any difference between them or not; yet the males were very different. We saw no other pellucida male pairing at any time during this visit to Juan Vinas, but on October 1, at the same fall, A. spied a pair of pellucida mating, which we were able to secure. The detailed comparison of the single female taken pair- ing with inopinata male and two females taken pairing with pellucida males, with each other and with nine other females taken singly, has failed to reveal any external features by which they may be distinguished. On June 26 a few larve were found at the farther waterfall in among the roots that hung in front of the rock face and were constantly bathed in the water. We suspected them to be Thaumatoneura but not until ten months later did we obtain the proof that they were so. These died very quickly when placed in still water so that we were unable to rear the adults from them. The next day we found one large larva of the same kind but older, crawling up out of the water 196 A YEAR OF COSTA RICAN NATURAL HISTORY on to a big rock. This being a hopeful indication of near transformation we carried it back to the cabin, where we rigged up an ingenious and home-like stream for him in a dish aided by a stone and a bottle. We subsequently took him to Cartago but the ungrateful little beast seemed un- appreciative and died July 25 or 26. In March the supposed Thaumatoneura larve were not rare on the rock face of the nearer waterfall, where there was a thin layer of organic mud subjected to only a moderate amount of spray. No doubt the larve are often carried by the dripping water lower and lower down from the roots in which the eggs are laid. But none of these March larve transformed in our possession. On April 24 at the same fall we found a score of the exuvie or skins of these larve shed at the time of transformation to the adult, while smaller larve than those indicated by the exuviz were crawling over the rocks. On April 26 the long search was crowned with success by finding two larve in the act of transformation into Thaumatoneura adults. One of these larve was hanging on the wet slippery face of the fall in such a position that a small projection above it gave it shelter from the direct spray. The wings, while they hung limp and wet, in fact until they were full size and dry, were green. The body of the dragonfly was brown and green and as the creature hung on the rocks, the body and wings giving the little trembling jerks that usually accompany the growing or “filling out” of a transforming insect, the likeness to a green leaf on a stem trembling in the breeze was very marked. There was much small-leaved Commelina on the face of the fall as well as some delicate grasses, both of a bright pea green, and we ourselves fre- quently confused the Thaumatoneura with one or other of these if we happened to move our eyes from it. It certainly seemed to be a good case of protective coloring. Other full JUAN VINAS—THE WATERFALLS 197 grown larve of Thawmatoneura were found a few days later which, transforming, confirmed the evidence as to their identity. The males of both inopinata and pellucida were obtained from the transformed larve and a minute comparison of the exuvie from which they respectively issued has shown no differences that cannot be considered as due to slight injuries received by each. This fact, together with the lack of observable differences in the females caught pairing with the males of inopinata and pellucida respectively, seems to point to the conclusion that there is but one species of Thaumatoneura after all, a species in which the females are all colored alike, although showing a certain range of variation in size and in the proportions of the hind wings, but with two forms of males whose wings show the very marked difference in coloration which originally led to their being given distinct specific names. These are still useful, of course, in distinguishing the two forms. That this view as to the relations of inopinata and pellucida is correct can only be definitely determined when, from the eggs of known parents, both pellucida and inopinata offspring are reared to maturity. The case of Thaumatoneura, on this interpretation, is relatively rare, but a parallel is afforded by the North Amer- ican butterfly Lycena (or Cyaniris) pseudargiolus, the jump- ing spider Mevia vittata, and certain Lucanid and Staphy- linid beetles. The reverse case, where in a given species, there is but one form of male and several forms of females, occurs more frequently in insects, as for example in the dragonflies of the genera Ischnura, Neurothemis and Ery- throdiplax, and among butterflies of the genera Papilio, Colias and Pamphila. Thaumatoneura, as far as known, is confined to Costa Rica and Panama and appears to have no very close rela- 198 A YEAR OF COSTA RICAN NATURAL HISTORY tions. The other exclusively waterfall-dwelling dragonfly that we found at Juan Viiias, Argia talamanca, belongs to a genus distributed from Canada to Argentina. Many species of Argia occur in Costa Rica, but only talamanca appears to have adopted this peculiar and restricted place of abode throughout its entire life. A. found a pair of talamanca at- tempting to oviposit in the mud on the hard rock face of the nearer waterfall on July 27. On August 3, an exuvia of this species was found on the vegetation to one side of the main stream of the same fall, but on the perpendicular face where there was constant dripping, and immediately above and resting on it, was the male which had just emerged. Single males were found at or near one or the other of these two falls on June 24, September 28 and December 2. Larve which corresponded with the exuvie of August 3 were ob- tained in March. Argia talamanca is a slender dragonfly one and three- eighths inches long, whose clear wings extend about two in- ches. The male has a pale blue thorax with a black line down the middle of its back and two black stripes on the sides, while the abdomen is chiefly black with pale blue at both ends. The female tends to replace the blue with more vio- laceous tints, the middle black line of the thorax is widened into a stripe and the pale blue near the hind end of the ab- domen is limited to one segment (the eighth) instead of covering three. Another species of Argia was met with at these waterfalls but was not restricted to them. This was A. underwoodi, slightly larger than talamanca, resembling it in colors but differing in details of structure and in the male having the middle black stripe on the thorax wider so that black predominates there. We did not obtain the larve of underwoodi, so we know nothing of their habits. The habit of dwelling in and around waterfalls and cas- cades does not appear to have been mentioned for any other JUAN VINAS—THE WATERFALLS 199 dragonflies. Yet it is likely that such a habit will be found to exist in some other members of this group of insects, especially in the tropics of the Old World, where topographic and climatic conditions similar to those of Costa Rica pre- | vail. In countries where frost occurs low temperatures may prevent the survival of Odonate larve in waterfalls of small volume, where the water is spread out in a thin sheet as is the case in those falls which Thaumatoneura inhabits. On the other hand, the force exerted by a large volume of falling water may be too great to permit Odonate larve to dwell therein. In addition to the two waterfalls here described we found Thaumatoneura at only one other—a small cascade in the bottom of the Reventazén valley. We were puzzled by not finding these insects also at other streams under generally similar conditions. An explanation of their absence from one of these, the little Rio Naranjo, is suggested in our diary: “‘At the Naranjo I saw a number of lizards from two in- ches to a foot in length and these possibly keep down the number of insects. At our two waterfalls [i. e., the ‘nearer’ and the ‘farther’] we have seen no lizards.” A structural peculiarity of the larve of Thaumatoneura and of Argia talamanca which greatly interested us when we came to study them after our return to the United States was the condition of the three so-called “‘caudal tracheal gills” with which the abdomen of all members of their suborder (Zygoptera) terminates. In the majority of spe- cies thus far known, these “gills” are thin, expanded and leaf- like; supplied with branching air-tubes, or trachez, they are believed to absorb, through their thin walls, oxygen from the air which is mixed with the water in which they live and the trachez carry this oxygen to all parts of the body. The larvee of Thaumatoneura and of Argia talamanca have these “gills” relatively and absolutely thickened and narrowed to 200 A YEAR OF COSTA RICAN NATURAL HISTORY such an extent as to suggest that they may not be very effi- cient respiratory organs. Examination of the hind part (rectum) of their intestine shows the presence of a few longitudinal folds which thereby increase the surface of this organ. In the other suborder of dragonflies (Anisoptera), the larve breathe by drawing water into the rectum where numerous folds and a rich supply of trachee absorb the needed oxygen. The folds and trachee of our Thaumato- neura and Argia talamanca larve are by no means as highly developed, but their presence is at least suggestive of a sim- ilar, if less intensive, function. While examining the larve and exuvie of Thaumatoneura we found great numbers of diatom shells attached to them. A single leg, which had been detached from an exuvia, was sent to Professor Albert Mann, Custodian in charge of the Diatom Collection of the U. S. National Museum at Wash- ington, who wrote: “I should say that there were at least. 500 diatoms [representing sixteen species and varieties'] on the small leg of this larva. The most of these are forms that grow attached by a transparent jelly-like stipe; but several of them are free and free-moving,—for example the different species of Navicula. ‘The finding of these forms on this aquatic larva is of importance because it helps to throw light upon a difficult problem connected with the djs- tribution of diatoms. It has always been more or less of a mystery how isolated pools and bodies of water of recent origin become stocked with a rather extensive diatom flora and frequently within a very short period of time. Some light has been thrown upon this distribution by supposing these organisms are carried on the feet of wading birds, but this method of transportation is plainly very inadequate. If, however, we add to this transportation by aquatic insects, many of which have the power of flying, we have a means of 1 For list, see Entom. News, Vol. XXVI, p. 299, Philadelphia, July, 1915. 1, 2. Thaumatoneura inopinata, male and female x34. 3, 4. T. pellucida, male and female x34. To face p. 200 Pilea sp. Myriocarpa sp. To face p. 202 JUAN VINAS—THE WATERFALLS 201 transportation that goes a long way to explaining the sudden appearance of large numbers of diatoms in new localities. I therefore look upon this rather abundant flora on the minute leg of this aquatic insect as of some scientific impor- tance. Of course in this case the power to fly does not exist until a later period of development, but many of the water beetles and other insects are doubtless coated with living diatoms in the same way.” Professor Mann’s letter led to the scraping of the legs and bodies of our adult specimens of Thaumatoneura and in this minute quantity of “dust” he found about sixty diatoms, thus confirming his conjecture as to the part insects play in spreading these unicellular plants. Other insects lived at the waterfalls. At one, on June 24, -we secured an extraordinary little creature, something like a leaf-hopper, equipped at its hind end with a long thick brush of iridescent hairs, which could be spread open. The brush was half an inch long, the insect itself only one-quarter inch. When A. first saw it, it had alighted on her knee and exactly resembled a pappus-crowned fruit like that of dandelion or of thistle. Such she thought it was and put out her hand to brush it away when the supposed seed leaped into the air and settled in another place. It proves to be a member of the family Fulgoride. Very likely the brush of hairs (which are really secreted waxy filaments) do serve the same end as the hairs or pappus of the fruits of the Composite—distribution by means of air currents. On the rocks below the waterfalls—and in other rocky streams—were many tiny crickets known as Rhipipteryx. The most common species (R. biolleyi)—and it was very abundant in the situations mentioned on August 3—was about 7 mm. (.28 inch) long, with body and wings black, prothorax and front wings edged with cream-color, some white marks on the head, but the legs orange-red, so that 202 A YEAR OF COSTA RICAN NATURAL HISTORY in spite of its smallness the insect was conspicuous. Its jumping powers were very great, but we never saw it fly. We observed it in August, late September, early October, early December, February, March and late April, on which last occasion it seemed scarcer. On December 2 it was also found in (to us) new surroundings—in little cavities, slightly larger than the crickets themselves, in earth occupied by a nest of small black ants. The crickets were sometimes in holes below the surface of the earth so as to be only partially visible. The ants apparently did not regard them with any hostility. Another, larger, species of Rhipipteryx, R. limbata (11 mm. long) found on rocks below the waterfalls on August 3, but less abundantly, had the legs black with a whitish line on each, the rest of the body colored similarly to biolley1. On February 14 two of them were seen flying, a mode of loco- motion new for them in our experience. After that I tried to make biolleyi fly, but it always jumped, although it has well-developed wings. The legs of Rhipipteryx are curiously modified, the tibiz of the first legs being deeply grooved lengthwise, presumably permitting the tarsus to be folded back into the groove, and the tibia of the third leg gives the appearance of possessing three terminal and two anteter- minal claws. ‘ On the rocks below the falls and also in the dripping vegetation on the rock faces at the side of the waterfall were some somber-hued grouse-locusts or Tettigids (Allo- tettixn peruvianus), whose leaping powers were not as great as those of Rhipipteryx; they were 10.5 mm. long, and were noted in August and in February. On September 29 several brown pink-winged walking sticks (Pseudophasma menius) flew up under our feet, and we took an oddly colored small grasshopper (Dellia minia- tula) 15 mm. or */; inches long, which had much metallic JUAN VINAS—THE WATERFALLS 203 green on its back. In a little clump of vegetation at the foot of the nearer fall we captured, on April 24, a large Sites The grasshopper Dellia miniatula. (From the paper}by Rehn.) The cricket-like katydid Anabropsis costaricensis. (From the paper by Rehn.) cricket-like katydid 42 mm. long (Anabropsis costaricensts); its wings were abbreviated (only 8 mm.), but the length of 204 A YEAR OF COSTA RICAN NATURAL HISTORY its antenne (100 mm.) and maxillary palps (12.5 mm.) was extraordinary. At the farther waterfall were a number of medium-sized “‘spiny-bellied” spiders (Edricus tricuspis, Acrosoma inequalis, Gasteracantha kochi) whose abdomens presented a great variety of odd shapes, armed with rela- tively large spines. A large fresh-water crab, over four inches across the cara- pace and with most formidable pincers, was at the farther waterfall on April 27. He sidled through the stream and up the bank, climbing better than one would expect from so clumsy a creature, and backed into a hole under the roots of trees bared by the water, keeping his pincers out and ready for action. At a little stream still farther on were the very antitheses of this monster—little crabs less than half an inch across. About half a mile west of the farther fall was a third fall or rather cascade reached by a little trail through a bit of exceedingly thick damp woods, full of wild ginger, Heli- conias, ferns and caladiums. The place was extremely pretty and ought to be a favorite haunt of many forest-lov- ing species. Here on May 2 we came upon the body of a fair-sized armadillo some time deceased. After long hunt- ing among the fallen leaves and stones of this brook on April 27, P. found a dragonfly larva with a pair of finger-like gills on the ventral surface of each of the second to the seventh abdominal segments. Nothing like it was known for any member of this order of insects inhabiting the Americas, but since similar conditions had been described thirty years be- fore for certain Indian Calopteryginz, we at once suspected that this larva might also belong to this subfamily and pos- sibly to the genus Cora, the adults of which we knew to live here. The larve found on this day and also on April 29 had lost their three caudal gills, which although forming part of the proper equipment of Zygopterous larvz, seem to JUAN VINAS—THE WATERFALLS 205 be of indifferent use and value. Many larve lose them by accident or by the bite of a brother or other enemy and yet pass their larval existence as tranquilly, and accomplish their transformations as successfully, as their brethren which, like the foxes in the fable, retain their brushes. Three of these peculiar larve found May 2 in the outflow from the farther waterfall possessed their caudal gills which were most odd-looking. They appeared as if cut off straight across at the tip, instead of tapering as usual, the straight edge then scalloped into three points. Each inflated-look- ing gill was one-fifth as long as the short ‘‘pudgy” body, giving the larva a “‘tail-heavy”’ appearance. When first taken into the hand these three larve remained motionless, “playing ’possum” as it were, for a minute or two, and then took to their legs with some speed. In spite of their double equipment of gills these larve did not frequent a different abiding place from others less richly ‘“‘engilled,” their fellow inhabitants being larve of Heterina, Argia and common types of Libellulines. These larve died before transforming but were preserved in alcohol; a microscopic examination of their rudimentary wings showed so close a correspondence with the veins of the wings of the adult dragonfly Cora chirripa, captured at the same waterfall, as to leave no room for doubting that these larve are of that species. It is the presence of the ventral gills that gives these larve their great interest, for breathing organs in such positions are rare in dragonfly larve generally, although common enough in may-fly and Sialid larve. It has been suggested that they correspond to the legs of the hind part of the body of centipedes and their allies, which some consider to represent the ancestors of insects. The adult Cora chirripa has a body one and seven-eighths inches long, the thorax pale greenish or bluish with black lines on its sutures and a broad black stripe on 206 A YEAR OF COSTA RICAN NATURAL HISTORY its middle, the abdomen chiefly black; each wing is one and one-quarter inch in length, slightly yellowish and densely veined. West of Juan Vifas the next station on the railroad was at Quebrada Honda, a side ravine emptying into the Reventa- zon valley on the north. At the time of our first visit here, August I, 1909, there were only a few native huts and one more pretentious house, the Hacienda Maria. That morn- ing was cloudy and threatening. We took the road which led upstream. The hillsides were planted with yuca (mani- hot), maize, squashes, beets, onions or garlic, the neat rows of which often occupied such steep slopes that we wondered how plants and soil retained their positions when it rained hard—as it often did. The land was formerly forest-clad and it still was so at the top of the hills. Here and there some tiny streams descended to swell the volume of the little Rio Quebrada Honda. The valley has some historic interest as being the position occupied by the Spanish gov- ernor Juan Lopez de la Flor in 1665 to resist the attack of French and English pirates and buccaneers under Mansfield and Morgan, who, landing at El Portete, near the site of Limon, advanced as far as Turrialba but retreated on learn- ing of the resoluteness of the governor. Again in 1681, Governor Saenz Vazquez intrenched troops at Quebrada Honda to await the pirates who had disembarked at Ma- tina with the intention of attacking Cartago, but who did not carry out their plans. At the little streams we found a few small leeches (Semis- colex glaber) and planarians, a small fresh-water crab with a carapace one inch across (Pseudothelphusia richmondi?) and some dragonflies. Five slender columns of army ants (Eci- ton hamatum) were crossing the road, but the cloudiness was not favorable to the appearance of many insects. When, later, the sun shone and more insects were on the JUAN VINAS—THE WATERFALLS 207 wing, we were interested in the general similarity in color pattern and in manner of flight shown by two widely dif- ferent dragonflies—Epigomphus subobtusus and Brechmor- hoga rapax crocosema—flying over the same brook. They were of approximately equal size (two inches long), each had the abdomen widened at or near the hind end, the seventh segment thereof bore the most conspicuous pale marking of the entire body (which was in general dark brown with bluish or greenish stripes) and the eyes were blue or bluish-green. Many differences in detail there were but the general resemblance was of the kind that has been called mimicry. There was nothing in this case, how- ever, to suggest that the resemblance offered a means of defense to either species against enemies. As we ate lunch on the steps of the unoccupied Hacienda Maria, we watched a lizard, nearly a foot long, hunting in- sects on the ground below us. Occasionally he would cau- tiously approach a blue dragonfly (Argia extranea) several of which were here and there on the ground, but unsuccess- fully as far as we saw. There were two lines of leaf-cutting ants crossing the range of the lizard’s activities but these he did not molest. Finally we saw him with a fat cricket or katydid in his jaws, nearly choking him, but needless to say it eventually disappeared down his throat. On February 14, 1910, P. went again to Quebrada Honda. Insects of all kinds were scarce, especially dragonflies and butterflies. The exception was Rhipipteryx biolleyi, plenti- ful at the waterfalls and often along the tracks, and a few of the black species (R. limbata) were also in evidence. Leaf-cutting ants were at work again where we had seen their paths on previous visits. Water was standing in the ditches at the sides of the roadbed and flowing in the small streams. The vegetation on the hillsides was everywhere green, in striking and well-known contrast to the condi- 208 A YEAR OF COSTA RICAN NATURAL HISTORY tions prevailing at this season on the Pacific slope from which I had recently come. At Quebrada Honda chayotes were being harvested from the steep slopes. Patches of blue sky began to appear and by four o’clock very few clouds remained, but the bright sunshine brought little in- crease in the insects. Never had Juan Vinas appeared so poor in this respect; only two species of dragonflies, Argia extranea and Anisagrion allopterum, could I find, and these two are abundant at Cartago throughout the year. “By six o’clock clouds were gathering again and at seven a strong easterly wind was blowing, and the candle flames by whose light I write these words, although sheltered within ‘our’ little cabin, flicker uncomfortably.” Our acquaintance with our friend the “‘nearer waterfall” terminated most unexpectedly. On April 30 we went up the railroad tracks to visit it again but found that a landslide had occurred changing the rock bed from the tracks up to the foot of the fall very greatly. We climbed up a short way to reconnoiter, saw that some small trees had fallen across the bed blocking our usual path and that rocks larger than one’s body had been rolled down and many of them left in very unstable positions. In fact the changes since our last visit four days before were far greater than at any time during the ten preceding months in which we had known this fall and we found it impossible to ascend. On inquiry we learned that on April 28 two distinct falls of rock had been heard in the direction of this waterfall, with the crashing of branches and the odor of newly disturbed forest earth. Never again were we able to reach the foot of this waterfall. But we had our Thaumatoneura; and it was our good fortune that we had found so many treasures here before the landslide occurred! CHAPTER XII JUAN VINAS—THE REVENTAZON VALLEY BELOW THE LAGUNA The little Rio Naranjo, arising on the southern slope of the ridge connecting the volcanoes Irazi and Turrialba, made its way to the northwest side of the crater rim of Juan Vinas, descended to the laguna, left it at its southeast side passing under a short, low railroad bridge, and plunged down 200 to 300 feet in a series of falls and cataracts; finally in a bed of gradually decreasing steepness it reached the flood plain of the Reventaz6n and glided gently into that river. The government road from the railroad station to the Reven- tazon left the laguna only a few yards west of the Naranjo cataract, but soon bent sharply westward in the first of a series of zigzags by which alone a cartroad can descend such a steep slope or “cuesta.” The upper zigzags afforded mag- nificent views of the canon and also enabled one to look into the tops of very tall trees whose bases were many feet below. Before this cartroad was constructed an old Indian trail or path led more directly down this slope. Much of it was still distinct and in spite of its steepness often used by pedes- trians because, cutting across the zigzags of the road, it shortened the distance very greatly; it was also in deep shade. Much forest remained on the cajion sides, so that one plunged from the hot sunny cartroad into the grateful protection of tall trees loaded with beautiful epiphytes, and climbed among palms, ferns, arums and many plants allied to bananas and marantas whose gaily colored bracts lit up the green depths with brilliant red or orange flames. The lowest of the zigzags descended from west to east, and 210 JA YEAR OF COSTA RICAN NATURAL HISTORY on its left bank there issued from the hillside a spring of clear water where we often refreshed ourselves, in the shade of the roadside trees and bushes. The descent ended at a ford through the brown waters of the Naranjo where that stream reached the flood plain of the Reventazon. The road then for rather less than half a mile gradually ap- proached the latter river and crossed it by a good iron bridge. The Reventazon here was a stream some seventy-five feet wide, rather smooth just at the bridge but full of boulders and foaming rapids above and below it, and with a very swift cur- rent. It is subject to tremendous floods, as it drains a large mountain territory. Its elevation at this point was 2500 feet by our aneroid. Above the bridge was a well-wooded island a hundred yards or so in length. The greater part of the river flowed on the south side of this island so that the water in the northern arm was shallower and less turbulent. The river banks where they were not steep cliffs or slides were boulder-strewn, and in the crevices grew small palms and maidenhair ferns, while stretching above them were large begonias resembling the “beefsteak”’ variety but having stems four to five feet high. A few yards before reaching the iron bridge there was, on the left side of the road, a waterfall about thirty feet high. Near it a trail led up the steep bank to the stream above the fall. The woods here were very thick and this trail extremely narrow, steep and slippery. In these dark shady nooks we took Palemnema, Cora and other interesting dragonflies, and at the waterfall itself Thawmatoneura. After crossing the Reventazén the road led into a thick plantation of “pejibaye” palms with long feathery leaves and bands of sharp black spines on the trunk. In July, plantains, maize, tobacco, pineapples and yuca were grow- ing with the palms. Yuca (Manihot palmata) resembles JUAN VINAS—THE REVENTAZON VALLEY 211 its near relative the castor oil bean but is feebler in growth and more reddish in color. Tapioca is made from its roots and the root itself is used as a vegetable, boiled like potatoes, and has already been mentioned as forming part of the bill of fare at Tierra Blanca. Under the palms on one side was a hut in a little clearing trampled hard and bare and enclosed in a bamboo fence. The walls of the hut were cane stems set side by side as closely as possible and bound together with plant fibers. The roof, which ran up to a high peak, was thatched with leaves. There was no window or opening for smoke, which crept out through cracks in the thatch at one end. It was a pic- turesque bit, and we usually stopped to enjoy it. One day late in July as we passed, we met an exceedingly handsome Indian woman bearing a huge load of green tobacco leaves on her shoulders and back. Later we met her again return- ing to the river fields for more and had some conversation with her. By this time the clouds were gathering and she told us the rain was coming soon and would be very heavy. As it turned out, however, the rains did not come until after dark, so the native Indian does not invariably read the clouds aright. An Indian family of father, mother, boy and two girls met us in the river bottom on June 28. The first was very friendly, took much interest in our collecting paraphernalia and captures and explained them to the others. We met them again as they were returning from the corn-fields which covered much of the flat land here and indeed some of the steep slopes also, each carrying a heavy sack of husked corn. The father presented us with several ears and would take no money, although he made no objection to our pre- senting each of the children with a little silver piece. The gifts made, the father added a few more ears, shook us each by the hand and we parted with profound ‘“‘Adios.” We 212 A YEAR OF COSTA RICAN NATURAL HISTORY gave the corn to Miss R. who cooked some for us, but it was rather hard and not particularly sweet. Miss R. told us it was impossible to grow sugar corn here because as soon as it came above ground each shoot was cut down and car- ried off by the leaf-cutting ants. The road to the Reventaz6n was much used by ox-carts and pack animals, but traffic was often interrupted by land- slides, especially on the zigzags, and places where carts had passed one day were often impassable for them on the next. We first knew the road in June, but in July, December and February, slides blocked it, often in more than one zigzag. Sometimes the soil slipped down leaving clean rock but so gently that the bushes and young trees were undisturbed, still standing green and flourishing, in their normal vertical positions. Sometimes the destruction was wrought by slipping masses of soft rotten stone which uprooted huge trees and sent them down the slope like battering rams, ploughing through the lower zigzags and tearing up still other trees. On the upper zigzags it was often possible to climb over the displaced material blocking the road, but on the lower ones the whole roadbed was sometimes swept away, leaving gaps that were quite impassable for horses and both difficult and dangerous to cross on foot. Occa- sionally our feet dislodged stones and masses of earth and pebbles that went bounding down the hillside in an unpleas- antly suggestive manner. The worst slide we saw was in Feb- ruary and was between two and three hundred feet high. The road was frequently repaired, however. When we went down on April 27 expecting it would be very bad as a result of rains and earthquakes (and remembering its condition on our preceding visits) we found the road cut through the slides. There had also been much clearing and cutting of trees on the sides so that the road was better than it had been for months previous. JUAN VINAS—THE REVENTAZON VALLEY 213 There were also other trails, leading from the valley bottom up to the railroad east of the laguna. One of these began about two hundred yards west of the iron bridge and led up into fine forest. Still higher it left the woods and passed through low shrubs and tall herbs, at the same time becoming very steep although well-marked, until it suddenly brought me to the edge of a comparatively recent landslide with precipitous walls. However, it was easy to cut a way through the same sort of vegetation up to a point where the forest began again. This forest was for the most part not filled with dense undergrowth but there were ferns and low palms. There was no definite trail here and what looked like men’s paths may have been natural open- ings. However, I had simply to keep going upward and was sure to reach the railroad. Finally after many steep climbs through the woods I came to a bank of strong coarse grass higher than my head, and pulling myself on top of the bank found that I was on the top of a very deep cutting through which the railroad passed. The high grass prevented my seeing well, but soundings with my stick showed that on all sides this bank was almost vertical. After making attempts on three sides and consuming the greater part of an hour I finally reached the tracks. Once I stopped for some purpose and momentarily let go of my stick, which immediately sank out of sight through the grass. I was unable to find it again, and I concluded from what I could feel with my feet and what I could see after I reached the track that the stick must have dropped down the precipitous side of the bank away from the tracks and that I had been on a narrow ridge only a few inches wide at the top and with nearly vertical sides thirty feet above the tracks and much more down to the forest on the other side, but whose precipitous character was concealed by the dense mat of grass. I reached Juan Vifias station just in time to prevent Mrs. Ridgway’s send- 214. A YEAR OF COSTA RICAN NATURAL HISTORY ing the policeman to hunt for me since it was growing late and dark—it was, indeed, no less than 6.45 P. M.! Although there were patches of cultivated ground in the valley bottom, at other places the forest was still untouched and its appearance was very tropical, with its palms, big coarse ferns, wild plantains, lianas hanging from every tree and incredible numbers of perching epiphytes. Many big trees had huge buttresses. In places the earth had washed out from under the supporting roots, which writhed over the ground like monster snakes. There were always many dead leaves on the ground. Our diary for September 29 expressly records that on that day there were yellow leaves on some trees, occasionally even a bright red one, but the great mass of vegetation was strongly green, with many tender green shoots on shrub and tree; the coffee plants showed their glossy dark green foliage and most of their berries were still green though some were reddening. On the other hand, the fields of maize were brown and withered. On February 15 we noted that many of the large trees in the Reventazon cafion, particularly those bearing much Tillandsia, were leafless and putting out tiny young leaf buds. This gave the cafion sides a very different appear- ance from that of six months earlier, and the views of the river were more extensive than we had ever seen them. Both the Naranjo and the Reventazé6n were extremely full and correspondingly noisy. Along the road down the cuesta—indeed everywhere nearby—were wonderful gar- dens of begonias. In places the flowers formed masses yards across so that the cafion side was tinted white, pink or coral red with the profusion of blossoms, a most beautiful and striking floral display. There were at least four species in great abundance and all seemed to be at the height of their flowering at this time. At the end of April the num- JUAN VINAS—THE REVENTAZON VALLEY 215 ber in bloom was noticeably less. Another beautiful Feb- ruary flower was a purple composite (Vernonia sp.) which bore long compound racemes of delicate feathery purple flowers. It grew six to eight feet high and was rather rank and coarse, but the blossoms were as delicate and dainty as could be. Growing in a little wet place near the Naranjo was the same red Hibiscus we had gathered in the laguna. Almost at the bottom of the valley in a patch of forest rich in small palms our attention was attracted by a tree about fifteen feet high (Symphonia globulifera) with curious red flowers. We gathered some and on our return a peon on the road told us the tree was called “jorquito.” The corolla of this flower is thick and fleshy, and conipletely fused; it has the peculiarity of falling off from the ovary as a complete, depressed ring about three-quarters inch across, with an opening of one-quarter inch. The coral-red buds were round and fleshy and looked like berries. There were many birds in the Reventaz6n valley, some of which we heard without seeing. The dropping, three- toned call of the ant thrushes, heard on our first visit in June, became ever associated with this place. In June, and in still greater numbers in late July, swallow-tailed kites (Elanoides forficatus) were conspicuous—we have counted twelve at once. They are known to the Costa Ricans as “gavilan tijerilla” or “tijereta.” These birds were about the size of crows, above mostly black, with a white head, pure white below. Their most striking features were the contrasting blacks and whites and the long, deeply forked, wide-spreading tail. They are consummately graceful birds, and their airy evolutions as they soared, turned, floated upward or downward, or circled in one plane with never a flap of their wings, formed a beautiful sight. The common zopilotes can likewise soar wonderfully but com- pared to these kites they seem heavy and clumsy. In late 216 A YEAR OF COSTA RICAN NATURAL HISTORY September no kites were to be seen anywhere; in their stead were many swifts. On March 20 we saw the kites here again but none in late April. In June and in September there was a handsome black woodpecker (Cephleus lineatus scapularis) with white shoulder straps and a large projecting red crest on its head, tapping on the tree-trunks and apparently finding some food. It was larger than a large pigeon but smaller than a crow. A mottled black and gray lizard (Anolis sp.) was resting on a log by the road to the lower Naranjo on June 28. Its body exclusive of tail was between three and four inches long. On the middle line of the throat was a thin fold of skin which could be inflated and when fully expanded reached at least an inch below the neck. This “dewlap” or gular pouch was bright orange in color and consequently quite conspicuous. ‘The expansion lasted only for a fraction of a minute but was repeated at intervals. I tried to catch the lizard in my net for further examination but failed. For a few minutes I had in my fingers a similar but smaller lizard caught on a nearby stone; it had, however, no such “dewlap.” Perhaps it was a female watching the charms which the male was displaying for her benefit. In late April, usually hanging on damp rocks by the road- side, was a handsome snail (Drymeus sulphureus citronellus) about an inch long. The shell was a bright canary yellow, the foot blue-green, the “horns” dark greenish-brown. Insects were not equally numerous throughout the year in the valley, February being the month when they were least apparent. On first going down to the Reventazon in June the forest was resonant with the loud shrill notes of the cicadas. The insects themselves were not easily seen. The note was not pitched quite as high as that of our annual cicadas in the United States but was sufficiently ear-splitting ‘psafyyngo]3 viuoydmhg ‘ozyinbsof Jo s1aMmoyJ To face p. 216 Megalopygid caterpillar. Wingless Walking Stick, Bacteria sp. x34. (Both from life.) To face p. 217 JUAN VINAS—THE REVENTAZON VALLEY 217 and much more continuous. It did not give the impression of constantly increasing speed but was steady and uniform, and there were so many cicadas that the noise was deafen- ing. In March the penetrating notes of a few cicadas seemed to indicate the waxing of insect life again. Of all the butterflies the most conspicuous were the me- tallic blue species of Morpho. This genus is characteristic of tropical America and extends from Southern Mexico to Argentina. Some authors, like Godman and Salvin, con- sider it to belong to the Nymphalide, forming a special subfamily thereof, the Morphine. Others make of it and its allies a separate family, Morphide. Most of the species have a number of eye-spots, or ocelli, with a diameter of */s inch or less, on the under sides of the wings. Six species and subspecies are considered by Frihstorfer, the latest writer (1912) on the group, to occur in Costa Rica.!' Their wing-expanse is from four to five inches. In Costa Rica Morpho extends from sea-level to an altitude of 3500 feet. We never saw them near Cartago, but occasionally one could be seen along the road leading upward from the station at Juan Vifias to the village. Their flight when undisturbed is an easy sailing with many a rise and fall over the tops of shrubs and bushes, but when alarmed they fly very swiftly. Many of the individuals one sees, except just after they have transformed into the winged condition, have the wings torn and tattered. On May 2, while I sat in partial shade at the little cascade west of the farther waterfall, a great blue Morpho which had several times sailed over the place, alighted on my knee. Its wings were rubbed and their edges ragged so that there was little temptation to 1 Four of these species have the upper surfaces of the wings with more or less metallic blue: M. peleides limpida, M. granadensis polybaptus, M. eypris cypris (with a conspicuous band of yellow across the blue of both pairs) and M. amalonthe centralis. M. theseus aquarius has the wings rich brown, M. polyphemus white with a few black marks. 218 A YEAR OF COSTA RICAN NATURAL HISTORY try to catch it; I sat perfectly still to induce it to linger as long as possible. Presently it flew quietly away and I did not see it return. Bits of the exquisitely blue, iridescent wings of Morpho are often mounted as pendants or lockets in the so-called “butterfly jewelry” and may be familiar to our readers in this form. In places where Morpho was found we sometimes ob- served the great owl butterfly, Caligo. It is a member of the Brassoline subfamily of the Nymphalide and has the same geographical distribution as Morpho. It tends to fly more in the twilight or in shade. The upper surface of its wings (which spread from 4 to 514 inches in the five Costa Rican species) is rich dark brown with slaty blue or buff or orange. The under surface has numerous fine lines and streaks of various tints of brown and of yellow and a num- ber of eye-spots. The eye-spot near the middle of each hind wing is the largest and is from 1% to 5/, inch in diameter, its outermost rim being black, within which is a ring of yellow. It is to these two large eye-spots and their background of color that these insects owe their name of owl-butterflies. Other conspicuous and characteristic butterflies both near the laguna and down the Reventazén road, were the narrow-winged Heliconiines, some members of which have been described in Chapter VI. Such were Heliconius peti- veranus, with black wings, crossed by a broad orange red band on the fore pair, the hind ones having a narrower yel- low band parallel to their front margins, and H. charithonia, whose black wings bear three yellow bands on the front pair, a band and two rows of small spots, on the hind ones. Petiveranus has a wing-spread of 21% inches, charithonia of 344. Four charithonia were sleeping on the roots of a tree, exposed by the washing away of the soil, by the roadside on April 27. Their wings were folded and hanging down, and so dormant were they that A. picked one of the group JUAN VINAS—THE REVENTAZON VALLEY 219 off while the others only fluttered a short distance and re- turned to the same roots and same positions. In April flocks or single individuals of a handsome butter- fly, Anza onophis, were abundant, particularly at wet spots along the road. The upper surface of its wings (which ex- panded 214 inches) was dark blue with pale blue spots near the outer margin. Each hind wing had a slender tail 14 inch long. The under surface was colored brown and much resembled a withered leaf so that the insect was inconspicu- ous when at rest. On June 25 our attention was attracted by small moths which settled on the rocks, both on the river bank and out in the stream, an inch or two above the water surface. There was no apparent cause for their positions, as in many in- stances the rocks were bare and are no doubt frequently submerged. On one rock in the stream about three feet long we counted twenty-five of these moths, facing in differ- ent directions. Their wings were bright orange, the outer margins broadly edged with black, and expanded about 1144 inches. Mr. Champion observed the same stream-fre- quenting habit of this species (Nelo chrysomela) in the Chi- riqui district of Panama. At different times some curious or interesting caterpillars were met with along the Reventazon road. While A. was sitting on a rock at the ford of the Naranjo on June 25 she was startled by seeing walk toward her what looked like a miniature skye terrier about three inches long. It was a caterpillar (family Megalopygide) thickly covered with long silky hairs which on the front end of the body were brown. The hairs of the greater part of the body were whit- ish, beginning on the first abdominal segment and arranged in correspondence with the segments. The head and protho- rax had no long hairs, so that the brown ones above men- tioned were borne by the other two thoracic segments and 220 A YEAR OF COSTA RICAN NATURAL HISTORY these hairs pointed forward as well as upward. The remain- ing white hairs were inclined upward, backward and toward the middle line of the body so as to give a “‘ridge-pole” effect along the back. As the creature walked these hairs sep- arated into cross-rows, corresponding to the segments, and also opened down the middle line, exposing to view shorter, brown hairs lying within the white ones, by which they were concealed when the insect was at rest. In addition each abdominal segment had on each side a bunch of white hairs above and brown below; these bunches curved down to the surface on which the caterpillar walked, simulating legs. The hairs of the upper surface of the body were one-half to three-quarters inches long, of the side bunches shorter; all looked and felt soft and silky, nevertheless an incautious prodding produced a severe stinging in the finger, the pain lasting several hours. Next morning, after a long and tire- some siege, this caterpillar was persuaded to sit for its photograph. Two days later two very curious caterpillars attracted our attention. One, with legs so short as to be invisible from the side, was pale green in color with thirty-four clus- ters of short blue-black spines arranged in four lengthwise rows; each cluster comprised two to twelve spines. Its body was plump and rounded at both ends so as to have the gen- eral appearance, or rather shape, of a pillow (family Me- galopygide). The other caterpillar was covered with hairs and flattened scales—red above the bases of the legs and over the head; above the red they were black for the whole length of the body as well as on the entire upper surface of the thorax; above the black a band of white scales extended along each side for the length of the abdomen, at the front end of which the white bands of the two sides united on the back. Fi- nally the middle of the upper surface of the abdomen was JUAN VINAS—THE REVENTAZON VALLEY 221 black except for the cross-band of white just mentioned. The creature looked almost as if made of variously colored plush and the colors themselves entitled it to be seriously considered as a national German caterpillar—if it had lived in Deutschland. It walked or crawled not continuously as most caterpillars do, but by jerks, with an appreciable stop after each forward jerk. There was a tuft of black scales at the hind end of the body; at the conclusion of one jerk this tuft was up, at the conclusion of the next it was down, so that the progression of this insect was not unlike that of a mechanical toy which is but partly wound up. We kept this caterpillar alive for several days, chiefly in a little wooden match box. On July 3 I had it out on the table while writing these notes and measured its rate of speed over a sheet of paper. In one minute it traveled 18 cm. (7.2 inches). It belonged to the family Eupterotidz. On one of the boulders on the bank of the Reventazén on March 20 was a caterpillar which had probably dropped from a tree overhead. It was not quite an inch long and its upper surface was generally gray. From each side of the body projected three somewhat twisted flat extensions, each nearly half an inch long and an eighth wide. Each one of these extensions was truly a part of the body and was covered with short hairs forming a dense pile. The middle of the upper surface was also covered with a short dense pile of darker blackish hairs. The head of the caterpillar was reddish-brown. When I first saw it I thought it was some sort of bag-worm and that the three extensions of each side were pieces of dry withered leaves which the insect had attached to its bag. Its legs were short as in the slug cater- pillars (Limacodide or Cochliide) to which this species no doubt belongs.!' This description is less accurate than I 1 Our description of this caterpillar and its cocoon suggests the curious North American species Phobetron pithecium. The moth which emerged from the cocoon 222 A YEAR OF COSTA RICAN NATURAL HISTORY would like, as it is based partly on memory; when I came to write it down on returning to our cabin I found that the caterpillar had begun the spinning of a cocoon in the box in which I had placed it. On the following morning it had completed its cocoon, using its own curious outer covering to serve as the outer covering of the cocoon. Bag-worms do exist here, for on April 27 we found a num- ber of small ones on a rock by the side of the Reventazén road. Their cases were about one-third inch long and closely resembled the lichens growing on the same rocks. Each consisted of an outer oval or ellipsoidal case enclosing an inner three-sided prismatic case, the latter containing the minute caterpillar, 4 millimeters in length. Both cases were of silk and both were open at each end; the inner was loosely moored to the outer by a few strands of silk passing from the three edges of the former to the inner surface of the latter. After death at least, the hind end of the inner case projected about one millimeter beyond the outer case. Attached to the outer surfaces of both cases were some fragments of plants such as seed-coats, chaff, an akene with its pappus, and also chitinous parts of insects such as the heads of small ants, prothoraces of beetles, wings and legs. Such cadaver- ous booty is attached to their own silken cases by some neu- ropteroid insects but is not often reported for lepidopterous larve, and it is difficult to imagine that the tiny builders of these bags could overcome ants which must have been as large as themselves. Can they obtain their trophies from the dead which they find in the course of their wanderings? We know nothing of the identity of these caterpillars, but one thinks of the insectivorous habits recorded for some of the tineids, the clothes’ moth family. was found two months later in a badly battered condition. The fragment resembles Ph. pithecium but is not identical; can it be Euryda'variolaris of the Biologia Cen- trali-Americana? JUAN VINAS—THE REVENTAZON VALLEY 223 Of the beetles (Coleoptera) which we found on the cafion side the largest and most striking was the Harlequin beetle (Acrocinus longimanus), well known in tropical America. Its larva, a thick white grub, bores into rubber trees and probably others also. It owes its name of “Harlequin” beetle to the bizarre markings, red, black and gray, of its elytra or wing-covers. Its front legs are immensely long, (14 cm. or 5.6 inches) and also the antennz (15 cm., 6 inches) as compared to the length of the body (6.5 cm., 2.6 inches). Moreover the tibia of these legs is quite sharply bent whereby the grasping power is greatly increased. These peculiarities are alluded to in its specific name, longimanus. We found one lying on its back in the road on July 28 and kept it alive several weeks but were unable to induce it to eat anything. When we picked it up it produced a rasping sound resembling a file drawn over wood. On the middle line of the front part of the mesothorax is a patch of highly polished black chitin, which under the lens shows very fine cross-wise ridges or teeth. The back of the prothorax rubs over these ridges and so produces the rasping. On August 12 we photographed this beetle. During a previous attempt it bit A.’s finger but no ill consequences followed. On both occasions the beetle was exceedingly restless. On August 25 we chlo- roformed it and found many external parasites. Most of these were mites, several hundred in number, representing two species of Uropoda (clavisetosa and bisetosa) described by Mr. Banks as new. They were most abundant between the elytra and wings but were also on the upper surface of the body between the pro- and mesothorax. Two pseudo- scorpions (Chelanops nodulimanus) were on the upper side of the abdomen below the wings, resting on a small silken web as wide as the abdomen, which presumably they had spun. A third pseudoscorpion was found on the cotton hold- ing the chloroform and probably came from the same place. 224 A YEAR OF COSTA RICAN NATURAL HISTORY A beetle of quite unusual form, found in the Reventazén valley on July 28 was Tauroma casta, of the Cassidide (Chrysomelide). It was half an inch (13 mm.) long, bright metallic green in color and had very fine punctures, not visible to the naked eye, on the prothorax and large coarse black punctures on the elytra. Projecting from the anterior outer angle of each elytron was a stout process, one-third as long (4 mm.) as its entire body, curved upward. What can be the function of these elytral processes—if they have any? A familiar beetle was the gray blister beetle Epicauta cinerea, which occurs over the United States east of the Mis- sissippi, through Mexico and Central America to this valley, where we obtained it on June 23. On February 17 a peon who met us on the Reventazin bridge presented us with a large live Buprestid beetle 6.5 cm. long, of metallic green, rose, and purple colors (Euch- roma goliath). We photographed it two days later, the creature proving very sluggish. This species is found from Mexico to Colombia. On April 28 numerous beetles, partic- ularly longicorns and passalids, were flying about although we saw no very large species. One day in July a large reddish spider suddenly ran across the road, followed at a distance of six inches by a black wasp with a yellow spot on its head. The spider doubled and twisted, and hid behind stones; the wasp followed every move, rising on her wings when the spider disappeared. At length she touched the spider with her antennz, when the spider. wheeled. and. struck at the wasp with her front legs, then hurried off in the interval of the wasp’s rising. The chase lasted several minutes, then the spider got under a stone and the wasp, at fault, hovered over the place with quivering antennz, frequently alighting to search the ground but unsuccessfully. Here we interfered by lifting the stone Rn enn a PS Euchroma goliath x1.4 (From life). To face p. 224 “(QP WOIT) 24x -snuvmrsu0] snurz0s9p ‘op 9g urnboze Fy To face p. 225 JUAN VINAS—THE REVENTAZON VALLEY 225 and exposing the spider which happened to be turned upside down. The wasp discovered it as it struggled to turn over and after some cautious feeling about managed to plant its sting into the spider’s ventral surface, almost between the bases of the legs. This disabled the spider and the wasp then stung it repeatedly over almost its entire body. Twice it left the spider and rested on a nearby stone, then returned to sting it again and again. At last the spider was limp enough and the wasp gathered it up between her legs, turn- ing the bulky victim over and over, and then tried to fly off with it. The spider was too heavy to be easily lifted and as the wasp struggled a train of packmules came along obliging us to move; this frightened the wasp away. On the under side of a large rock overhanging the side of the road was a hymenopterous nest covered with a gray papery substance closely resembling the rock in color. This cover was flattened, the nest occupying a concavity in the rock. One July day we stopped to examine it. Cautious lifting of one edge revealed a little stalked comb of several (6-12) cells within and some yellow wasps (Chartergus smithii) who came out to see what the fuss was about. They were little more than one-quarter inch long and had but a short petiole or waist; they belonged to the social wasps (Vespide). They were very polite, giving us no trouble when we lifted the floor of their house. A nest of a different kind was found in April, consisting of mud built around a root exposed by the washing away of the roadside. It was pear-shaped, two and three-quarters inches long and two inches wide at most. In the mud, at unequal distances, were thirty-one cells, some empty, others capped with a cement of different color from the mud. We cut the root across above the nest and brought the latter away with us. On May 16 we found a live black and red wasp half an inch long in our stateroom on the “Prinz Joa- 226 A YEAR OF COSTA RICAN NATURAL HISTORY chim,” which we supposed to have come from this nest. After our return to Philadelphia we took a dead wasp out of one of the cells, similar in size, shape and color pattern to that of May 16 but with yellow replacing the red. Both of these wasps had a long slender waist and belong to the species Zethus histrionicus, of the potter wasp family. The black and yellow one, which certainly came from this nest, closely resembles in size, shape, colors and pattern the social wasp Polybia fasciata, which we obtained in March at Cachi. Zethus is a solitary wasp, that is, there is no cooperation among the adults in building their nests or caring for their young. It is a curious fact that, among the true wasps (those whose fore wings are folded lengthwise) the solitary or social habit appears to be correlated with a structural peculiarity whose relation to these habits is still unper- ceived. In the solitary forms the claws terminating each leg have a tooth before the tip, which does not exist in the social species. We know that certain small parts of the legs of the honey bee perform certain definite functions and it may be that more intensive future study of the habits and modes of work of wasps will show that these teeth have a definite relation to the development or non-development of social life. But for the present, these anatomical details are merely convenient indexes to the nature of the lives of these industrious little artificers. In addition to this differ- ence in the claws which distinguishes the Vespidz (social) from the Eumenidz (potter wasps, solitary), our Zethus has relatively coarse punctures over much of the body, which are absent or but feebly developed in Polybia fasciata. In March an immense paper-wasps’ nest, high up in a rather isolated tree near the Reventazén bridge, harbored an active colony. It was as large as a man’s body, balloon- shaped and with an opening below through which the in- sects were entering and leaving. JUAN VINAS—THE REVENTAZON VALLEY 227 A very large and handsome ichneumon fly (Thyreodon laticinctus) was captured in October, flying in the forest at the bottom of the valley. Its body was more than an inch long, shining black with a bright yellow band across the ab- domen, while the wings were brilliant metallic blue. A curious katydid (Ischnomela impennis) quite fooled us when we found it in the woods within a long, partly folded leaf on our June visit. Its jumping legs, 66 mm. (2.6 inches) long, were stretched straight behind, its first pair, 28 mm. (1.1 inches), straight in front, giving A. as a first impression that the creature was a gigantic member of that spider group which possesses a long narrow abdomen and legs held in similar positions (Tetragnatha). A month later, while striking at a dragonfly in these woods, P. accidentally caught a large green wingless walk- ing stick, a species of Bacteria allied to bicornis. It had a pair of curved yellowish horns on its head, one just behind each antenna, which increased its grotesque appearance. Its body and legs were so soft and flabby that it was almost impossible to handle it without feeling it “give” between the fingers. We thought it full grown, but it moulted in the night, increasing in length to 57/3 inches, but was still wingless. A week later we photographed the creature, which hung like an insect sloth from a twig. The walking stick with beautiful rose pink wings (Pseudophasma mentus), which we occasionally took at the waterfalls, was more com- mon in this valley. Its body was dull brown, two inches (so mm.) long in the male, two and two-thirds inches (67 mm.) in the female, and lacked the cephalic horns. Many times, when we reached the ford of the Naranjo in the valley bottom, instead of crossing it and continuing toward the Reventazon bridge, we would turn off to the right into the forest until we reached a little brook which glided with a hardly perceptible current through very deep 228 A YEAR OF COSTA RICAN NATURAL HISTORY shade. Here came the fairy humming-birds to bathe, flying through the water repeatedly at our very feet and after each dip perching on a branch not a yard away to shake their glittering feathers. Here, too, we often found two species of dragonflies whose habit it is to live along just such deeply shaded streams, Heteragrion erythrogastrum, whose male has a bright red body 45 mm. long and Protoneura remissa with a thread-like body 34 mm. long. In the latter the only visible parts are its red eyes and a small blue spot near its hind end. In this part of the forest and at various points along the Reventaz6n road, we saw, at different times, three species of an extraordinary group of dragonflies, the Anormostig- matini, confined to tropical America. They are recogniz- able by the great length (two to five inches) and extreme slenderness of the abdomen, by the front and hind wings being long, narrow and of equal size and by the stigma (or colored spot on the front edge of each wing near its tip) being divided by cross-veins into several minute areas or “cells,” instead of constituting but one cell as in the great majority of dragonflies. The largest of these three species was Megaloprepus cerulatus, described in Chapter XV. It is chiefly an inhabitant of the lower country, and an individual observed on September 29 near the upper end of this road at about 3000 feet, had reached the highest elevation in Costa Rica at which we ever saw this species. The second in order of size was Mecistogaster ornatus, whose body length was 3 to 4 inches (75-100 mm.) and wing expanse 3°/; to 4*/; inches (90 to 120 mm.). Each wing-tip inward for 5 to 8 mm. was greenish-yellow, bright yellow or orange. Its area of distribution is from Vera Cruz and Durango States, Mexico, to the Amazon valley. The third species was Mecistogaster modestus with a body 3 inches in length and a wing-spread of 3°/; to 3*/; inches. The wings were JUAN VINAS—THE REVENTAZON VALLEY 229 colorless except for the small white, rosy or pale brown stigma. It has been found from the State of Vera Cruz, Mexico, to Panama. When flying, the bodies of these insects are extremely in- conspicuous. An old man, near the Rio Buena Vista in Nicoya, told Professor Tristan that Mecistogasters were not “‘pipilachas” or dragonflies at all but human spirits, be- cause they have not head or body, only wings! When a person dies the spirit (soul) begins to fly for some days in the woods and nobody can disturb it. CHAPTER XIII JUAN VINAS—THE TENANTS OF BROMELIADS Up to the time of our arrival in Costa Rica the larve of these large Anormostigmatine dragonflies, referred to in the closing paragraphs of the preceding chapter, were unknown. Mr. O. W. Barrett, who collected for me in Mexico in 1897 and 1898, made the suggestion} that possibly their larve lived in the large water-retaining bases of bromeliads. Re- membering this we had looked for dragonfly larve in some bromeliads near Cartago, without success. Indeed there is as yet no reason to think that species of Mecistogaster or Megaloprepus exist there, the altitude being perhaps too high. Mr. Frederick Knab, of the U. S. National Museum at Washington, wrote us under date of August 5, 1909, “In the course of my mosquito work in the tropics I investigated the water between the leaves of the epiphyte Bromeliacee and found that besides mosquito larve there was a rich in- sect fauna. The winter of 1907-8 was spent at Cérdoba, Mexico, and there I found that nearly every bromeliid in- vestigated contained dragonfly larve. I was not prepared to rear Odonata but I finally did succeed in breeding out a couple of rather large, unfortunately crippled, Agrionids.” Mr. Knab’s letter determined us to make more persistent search in the bromeliads of the warmer parts of Costa Rica. The pineapple is the most familiar example of this family of plants but many of the other members are epiphytic. Such are the Tillandsias, or Spanish mosses, which form long gray-green festoons both in tropical and sub-tropical 1 Entomological News, XI, p. 601, Philadelphia, November, 1900. JUAN VINAS—TENANTS OF BROMELIADS 231 America. Other epiphytic bromeliads of Costa Rica, which in general appearance are more suggestive of pineapples, are of the genera Androlepis, Aechmea, Billbergia and Pitcair- nia. Collectively they range from sea-level to the craters of such volcanoes as Irazi (11,000 feet), but they are much more abundant in the moister parts of the country. Some- times they are situated close to the ground; frequently they are attached to an unbranched trunk thirty or forty feet from the soil, or may be lodged among the branches at a still greater height. Their narrow leaves, often two feet or more long, taper gradually to near the tip; they are toothed or spined on their edges and in color a bright green or a beau- tiful pink or red. The leaves spring from a very short stalk so that their bases are pressed closely together. As a rule a number of stocks grow side by side on the same host tree, whose trunk or branch they may completely encircle, and as the stocks may also be attached to each other it is not easy to separate one individual plant from its fellows. The leaves diverge from their bases and either stretch up stiffly for their entire length or droop gracefully near the tips; owing to their stiffness and their serrated spiny edges they must be handled carefully to avoid scratched and bleeding hands. On October 3, P. examined some of the more accessible of these plants on the Reventazon road. The first was quite a thick clump, not more than eight feet above the ground with many vines growing about it. I threw a rope, weighted with a stone at one end, over it and by pulling on both ends brought the whole mass to the ground much more easily than I expected, leaving the tree-trunk to which it had been attached quite clean. Some large, black, rather slender ants (Odontomachus hastatus) appeared very soon after the mass fell. They were 15 mm. (°/s) inch long and had long slender jaws bent near the tip; as they ran about 232 A YEAR OF COSTA RICAN NATURAL HISTORY they carried these jaws wide open, at 180°, so that the tips were nearly a quarter inch apart and when they closed them, as they did in a mechanical sort of way with a sudden snap, there was a distinctly audible click. I do not doubt that these ants can inflict a painful bite, although I did not experience it. I cut across each leaf near its base, beginning at the out- side of the clump and working in toward the center, so that I could see whatever might be lodged between the bases of the leaves. When the mass of plants first fell, considerable water ran out from between the leaves, derived of course from the rains. In all the bromeliads examined this day there was also much mud between the leaves, especially the outer ones, as well as dead leaves which had fallen from above or perhaps been carried in by winds. The second bromeliad examined was lying on the ground at the bottom of the Reventazén valley, still attached to its host tree with which it had fallen. The third was near the second on an upright tree, but although I got the rope over it I was unable to dislodge it, pulling with all my strength, and as I shook down a great many of the Odontomachi I had some doubts whether this plant would have yielded many other insects. The fourth and last was in the forest about 200 feet above the river, perched on a tree-trunk ten to twelve feet above ground. The unavoidable shaking, pro- duced by my repeated attempts to throw the rope over it in just the right spot, brought down a few small ants (Aptero- Stigma) but none of the larger Odontomachus, which I took to be a favorable indication. After half an hour’s tugging and pulling the mass, consisting of three plants and weighing surely not less than fifty pounds, fell to the ground. I cut off the leaves as before and soon found a dragonfly larva between two leaf-bases not far from the circumference. In an hour’s further searching no other such larve were 3 . Beetle (Pachyteles seriatoporus) from bromeliad, Juan Vifias, x314. I 2. Fore leg of same showing Antenna-cleaner, x14. 3. Walking Stick (Ceroys bigibbus), Peralta, from life. Nearly life size. 4. Caterpillar (Castnia? sp.) from bromeliad, La Emilia. x234. To face p. 232 O16! ‘2 HOUvAN LS ‘ON Lg "ON 40 VIANXS sg ‘ON 40 VIAQXS “WV 076 | “ol6t ‘p Trtav SS ‘ON VAUVT It KV 50'6 OT6T ‘p Trad 6S ‘ON VAUVT woud ‘KV IS'L KY PPL “IV OFL WY LZ'L WY STL “WY O16 OL6T ‘9 Traav “OI6T ‘} Truav $g (ON VAT 9.48 Am. 9.58 A.M. 10.11 a.m. 10.46 A.M. 10.46 A.M. 9.45 AM. 9.20 a.m. 9.10 Am. The Transformation of Mecistogaster modestus, X1.2. Bromeliads on a Branch of Pord, Cartago. Ants. 1. Queen, 2. Two Workers of Apterostigma calverti, Banana River region. x6. 3. Odontomachus hastatus, Juan Vifias, x3. To face p. 233 JUAN VINAS—TENANTS OF BROMELIADS 233 found and there was so much still to be examined that I left the plants where they had fallen, and resumed the cut- ting off of the leaves the following morning. When all the larger ones had been removed I carried the three stocks, still so firmly united that I was unable to separate them and weighing some fifteen pounds, to the spring farther down the canon side so that I could wash out the mud and make more careful search. At noon—after three hours’ con- stant work—I finished the examination and had found two more dragonfly larve. These larve were kept in bottles with a few bits of bromeliad leaves added to make the sur- roundings homelike. All reached Cartago alive, but the one found October 3 died when attempting to moult on Octo- ber 7; one of the others died from some unknown cause, while the third lived until December 30-31. They were fed, and seen to eat, smaller dragonfly larve and blood-worms. All four bromeliads examined were inhabited by some animals. The fourth, besides containing dragonfly larve, was tenanted by a young scorpion two inches long which had just shed its skin, the latter also found; several species of daddy- longlegs (Phalangids), and of Pseudoscorpions; the latter are like miniature scorpions but without the long slender hind | abdomen. The Pseudoscorpions have a pair of relatively long fore limbs (pedipalps) ending in claws like a crab’s and as the creatures ran or walked over the surfaces of the leaves they held these pedipalps stretched wide apart as if to embrace whatever they met. There were also several species of beetles, both adults and larve; certain of the latter seemed especially well adapted to life between the appressed leaves of the bromeliad, for although the larve were one to one and one-half inches long and a quarter inch or more wide, they were relatively thin, less than one-eighth inch. These are believed to be the larve of Semiotus, one of the click-beetles (Elateride). A caterpillar, perhaps of the moth 234. A YEAR OF COSTA RICAN NATURAL HISTORY Castnia, also lived in this plant, and in the mud between some of the leaves dwelt a fair-sized earthworm (Andiodrilus biolleyi), quick in its movements and with a peculiar smooth proboscis which was frequently thrust out and looked like a spine.? In the first and second bromeliads were also a large scor- pion six inches or so in length, which continually got away from me into the deeper recesses so that I had but momen- tary glimpses of it; round beetle larve of the “June bug” style, a species of snail (two species if the owners of empty shells found also lived here), earwigs, smaller ants, isopods, collembolans and a smaller species of earthworm. Most of December 17 was spent in the Reventazén ex- ploring bromeliads. Some we tried to reach were perched too high, as we discovered when we attempted to throw the rope around them. But we did succeed at last in pulling down a big one, although it took all our combined weight and strength, and were rewarded by finding seven or eight dragonfly larve of different sizes. This bromeliad, among other animals, contained a few of the black Odontomachus, so that their presence does not necessarily preclude the existence of dragonfly larve in the same cluster of plants. We found the bromeliad examined on October 3 lying on the ground where it had fallen and when we tapped upon it, out rushed these ants. They had not been disturbed in their housekeeping by having their house turned upside down. Other bromeliads in this valley which we examined in March contained no dragonfly larve but many planarians (Rhynchodemus bromelicola), flat worms 20 mm. (34 inch) or more in length, striped lengthwise with black and dull yellow. 1 For further details see Entomological News, XXII, p. 405, Philadelphia, Novem- ber, 1911, and the Entomologists’ Monthly Magazine (2) XXIV, pp. 2-5, London, January, 1913. JUAN VINAS—TENANTS OF BROMELIADS 235 Early on the morning of April 4 we noticed that one of our bromeliadicolous dragonfly larve collected on December 17 was climbing up out of its tumbler, which act, together with certain recent changes in the coloring of parts of its body, suggested that it was about to transform. The eyes, the upper surface of the thorax and the tips of the wings had become yellow and a yellow spot had appeared on the middle of the under side of each abdominal segment, the rest of the body remaining a dull brown or reddish-brown. The tumbler contained a little water and a small bromeliad plant with root and leaves entire, about 12 cm. high, gathered near Cartago and replaced from time to time as it withered or decayed, also a little stick on which the larve might climb, as an additional piece of furniture. This larva repeatedly tried to climb up the smooth glass side of the tumbler, apply- ing almost the whole under surface of its wet abdomen to the glass. It would reach to a height of two or three inches and then fall to the bottom of the tumbler but immedi- ately begin climbing on the glass again, even though it touched the stick or the bromeliad. We helped it then by tilting the tumbler so that it reached the netting covering the top and to this it clung with its legs. The netting with the larva still clinging to it was pinned up so that we could photograph the successive changes. The transformation began at 8.15 A. M. and full size was attained by 10.46. The adult dragonfly, a female Mecistogaster modestus, re- mained on the netting until 1.53 P. M., then made its first flight to the window, where it was captured and put into a large paper bag or “‘cage”’ to perfect its colors. A second bromeliadicolous dragonfly larva transformed at Cartago on the morning of April 6, beginning about 7.20, and was similarly treated. From the two larve we obtained the practically complete series shown in the accompanying illustrations. This larva transformed into a male Mecis- 236 A YEAR OF COSTA RICAN NATURAL HISTORY togaster modestus. Both had lost, weeks before, the three conspicuous gills at the hind end of the abdomen, while other larve of the same species apparently about the same age and from the same bromeliad at Juan Vifas still retained the gills. Absence of these organs seemed to make no differ- ence in the transformation of the former two individuals, which accomplished this feat perfectly. This and other bromeliad-dwelling larve had been kept in captivity, under the conditions we have described, for nearly four months. They had been fed with blood-worms, but although we did actually see them swallow some of the worms they never seemed to eat many. Compared with other dragonfly larve, they were very abstemious and we could not see how they obtained enough nourishment to carry them through four months and a successful transfor- mation at the end. In the afternoon of April 26 we walked along the railroad from Juan Vifas station to a fresh clearing planted with beans. Here the ground sloped more gradually than in many parts of the bank below the railroad, and on the edge of the clearing was a tree with some large bromeliads that seemed to be accessible. So thither we went with our rope and big knife, pail and bottles, to secure the vast horde of Mecistogaster larve that ought to be living there. Arrived at the bean patch we found a man busily weeding and asked his permission to go through, which was readily granted. We worked all afternoon on the various clumps of bro- meliads and found a few larve of Mecistogaster, with an accompaniment of mosquito larve, spiders, cockroaches, slugs, planarians, earthworms, and the big-jawed black ants. On April 28 we passed the remains of the bromeliad cluster which yielded the larve from which M. modestus transformed in our room at Cartago on April 4 and 6. A single male of this species was sitting on the tip of a leaf and was easily JUAN VINAS—TENANTS OF BROMELIADS 237 caught. Around bromeliads of a different species two fe- males of M. modestus were fluttering and alighting, and although we did not see them making any motions of ovi- position one of them disappeared into the leaf-bases as if she might be on such an errand. On the outside of one of the yard-long leaves, about six inches from the apex, was a shed larve skin which we secured. On May 1 a female of modestus was transforming on the outside of a leaf on this latter cluster and P. secured both her and her exuvia. These observations show that in spite of unnatural sur- roundings and perhaps lower temperature our larve at Cartago were not retarded in their development as compared with their free relations in their native forest. After our return to the United States Mr. Knab sent us the fragments of the two dragonflies which he bred from arboreal bromeliads. These fragments, of adults which had transformed at Cérdoba, Mexico, in April, 1908, proved to be also Mecistogaster modestus. The origin of the bromeliadicolous habit of the larve of Mecistogaster modestus may possibly be accounted for in the following manner. The majority of the species of Mecistogaster are South American and some of them occur along the Amazon, where also are the headquarters of the Bromeliacee. As is well known, “thousands of miles of forests” along this river are inundated each wet season, so that a person “‘will travel through this forest for days, scraping against tree-trunks and stooping to pass beneath the leaves of prickly palms, now level with the water though raised on stems forty feet high” (Spruce, Wallace). At such periods of high water, epiphytes, whether of the Brome- liaceze or of other families, would often be just at the water’s surface, or only slightly submerged, and would offer to drag- onflies quite ordinary and usual places of oviposition. An association with certain plants might thus be formed by 238 4 YEAR OF COSTA RICAN NATURAL HISTORY Mecistogaster or its ancestors which would persist even when the water-surface was much below the level of the epiphytes. Only such plants as could retain water for long periods of time (weeks and months) would permit the development of essentially aquatic larve and the water must be renewed from time to time. Once the association of this insect with bromeliads or any other suitable plant was formed it might persist with the spread of the insect away from the regions of deep yearly inundation,—the Amazon or wherever we conceive its possible origin to have occurred—to the forests at Juan Vifias where the trees which harbored our larve were far, far above the highest flood marks of the Rio Rev- entazon. M. modestus, M. ornatus and Megaloprepus coeru- latus, as we saw them in Costa Rica, usually fly above the underbrush and when disturbed rise many feet above the ground. Although all dragonflies have the abdomen in the winged stage much longer than in the larva, modestus and, very probably, all Anormostigmatini are extravagant in this regard. In one modestus the fully extended adult abdomen alone was 71 mm. long while the whole body in its larval condition was but 20 mm. The abdomen of the adult modestus is from 514 to 514 times as long as that of the abdomen of the exuvia which it has just left, while in no other Zygopterous species with which we are acquainted is the corresponding figure more than 4?/,. In this latter case (Philogenta carrillica, whose larva is not bromeliadic- olous) the adult abdomen is 42 mm. long as contrasted with a minimum of 63, for our Costa Rican modestus. The excessively long abdomen of the adults of all the Anormostigmatini may be a special adaptation to the life of their offspring in water-containing plants, since the abdo- men of the larva of M. modestus is no longer, proportionally, than in other Agrionine. The space between one leaf of a JUAN VINAS—TENANTS OF BROMELIADS 239 bromeliad and the leaf next without decreases downward and if Mecistogaster’s eggs are deposited in the plant tissue in or near the contained water in accordance with the habit of Zygopterous dragonflies, it would often be necessary for the female to reach far down into crevices possibly too nar- row to admit of the entrance of her thorax and wings. The long abdomen with the ovipositor near its hind end would therefore be of distinct advantage and it will be a matter of great interest to ascertain, by future observations, if the lengths of abdomens seen in various members of the Anor- mostigmatini are correlated with peculiarities in length in the plants or other objects in which they oviposit. The abdomens of the males of the species of Mecistogaster are as long as or longer than those of the females. Their length of course cannot be explained in the way suggested for the females, but is possibly due to the necessary correla- tion in length which must exist between the two sexes to enable them to assume the characteristic mating position. In describing our discovery of these peculiar dragonfly larve we have incidentally mentioned some of the other ani- mals which were living in the bromeliads. In addition to the plants of this family examined at Juan Vinas, La Emilia, Cachi, Cartago and the crater of Irazi also furnished in- teresting bromeliadicoli, although none of the dragonfly group. The favorable character of bromeliads as shelters, or as sources of food, for animals is indicated by the fact that the cluster from one height on a tree at Juan Vifas, which has been designated the “fourth bromeliad” yielded 25 species associated at one and the same time, although one of these species was an external parasite on the body of a beetle. These 25 species were distributed among the following groups: Insects: 18 species (of 15 genera, of 14 fam- ilies, of 7 orders, Orthoptera, Odonata, Heteroptera, Cole- optera in the majority with 11 species, Lepidoptera, Hymen- 240 A YEAR OF COSTA RICAN NATURAL HISTORY optera, Diptera); Arachnids: 6 species (of 6 genera, of 4 families, of 4 orders, Scorpioni, Pseudoscorpioni, Phalangida, Acarina); Annelids: 1 species. Of the beetles which we found in bromeliads at Juan Vinas the most interesting in structure was the carabid Pachyteles seriatoporus, because of its possessing, on its first pair of legs, organs suggesting the antenna cleaner of the bees and ants. In the latter two this organ is formed by the tibia and first section of the tarsus of the first leg, the first tarsal joint bearing a semicircular notch lined with a row of stiff hairs, while at the tip of the tibia is a spine which holds the antenna in place as it is being drawn through the tarsal notch. In Pachyteles seriatoporus the semicircular notch is located at mid-length of the tibia and is bounded by a grooved and pointed process of the under surface; the grooved floor of the notch runs obliquely across the tibia and is smooth and polished on its forward end while its hind edge is guarded with a row of close-set, short hairs. On the under surface of the femur at mid-length is a strong, slightly curved, tapering, sharp spine, so placed that when the tibia is flexed upon the femur the spine of the latter would guide the an- tenna and hold it from slipping out as it was drawn through the tibial notch.1 P. seriatoporus is three-fifths inches (15 mm.) long, uniform shining brown in color, and has an- tennal joints 5-10 almost square, 1-4 and 11 more elon- gated. The Costa Rican naturalist, Sefior C. Picado, has published in Paris (1913) a thorough and detailed study of the epiphy- tic bromeliads and their fauna. He groups the members * Prof. L. C. Miall has recorded in his Natural History of Aquatic Insects (second edition, London, 1903, pp. 376-378) some observations and experiments by Mr. W. F. Baker showing that many Carabidz possess and use a similar antenna-cleaner. No spine on the femur, such as we describe for P. seriatoporus, is mentioned, how- ever. JUAN VINAS—TENANTS OF BROMELIADS 241 of the latter as: 1. Those which attack the plant itself, 2. Those which feed on the vegetable debris and the fungi which develop there, and 3. Predaceous animals. He has brought together a long list of about 250 species included under these three categories. The inter-relations of members of these different elements are interestingly shown by our “fourth bromeliad.” The ants brought from it were described by Prof. W. M. Wheeler as a new species, Apterostigma calverti, of which we had also other specimens (although not from a bromeliad), together with fragments of their nest, from the Banana River dis- trict, in early November. We had suspected these latter to be fungus-growing. Professor Wheeler wrote of them, “As you surmise it is a fungus-growing ant of the most primitive and, at the present time, most interesting genus of Attii. No species of this genus has ever been taken in Bromeliads. All of the known species have been described from cavities in rotten wood where they build a peculiar fungus-garden using caterpillar excrement as a substratum, and enveloping the whole garden in a mycelial web, which is not known to exist in any of the other genera of Attiine ants. Your spe- cies is very dark in color, all the other forms being yellow- ish or brownish. The sculpturing of the thorax is more pronounced and the head of a different shape.” In his technical description of this ant, whose workers are 3.5- 4 mm. and females 4.5 mm. long, he states: “Fragments of one of the fungus-gardens, preserved in alcohol with the ants, seemed to present the same primitive development of the hyphal swellings (‘ambrosia’) growing on a sub-stra- tum of insect (beetle?) excrement as described by Meeller for some of the South American species of Apterostigma.” We have made no direct observations on the nature of the fungus-garden of Apterostigma calverti in the bromeliad; but in view of Professor Wheeler’s statements it is apparent 242 A YEAR OF COSTA RICAN NATURAL HISTORY that the excrement for the growth of the fungus may well be furnished by the caterpillar or by some of the beetles which were found there. This affords, then, a glimpse of the food dependence of a member of the second group on those of Sefior Picado’s first group. The predaceous members, like their non-bromelia-dwell- ing relatives, do not hesitate at cannibalism. The alimen- tary canal of a larva of Mecistogaster modestus taken on April 26 contained a smaller larva of the same species, a mosquito larva and a copepod crustacean. Larve of mosquitoes of the genus Megarhinus have been found both within and without bromeliads and are known to feed not only on other mosquito larve but also on their own species. Of all the elements composing the bromeliad fauna, the first is likely to prove most dependent on the host plant. It remains to be determined how far the members of the sec- ond and third elements are limited to bromeliads. Mr. Knab believes that the beetle larve of the family Cyphon- ide inhabiting bromeliads are distinct species from those which breed in tree-holes and adds, ‘‘My belief, in the mean- time, gains some support from the fact that the Diptera breeding in the Bromeliads have been found to be, almost without exception, confined to this habitat.” Senor Picado has likened the totality of the epiphytic bromeliads to a great interrupted marsh extending through- out tropical America. Noticing the purity of the water re- tained between the leaves where one would expect foulness from the decomposition of the organic material, he made some chemical researches and experiments from which he obtained highly interesting results. The bromeliads pro- duce a gum which has a digestive action on starches and on nitrogenous materials (albuminoids). The products of the digestion of the vegetable and animal detritus retained JUAN VINAS—TENANTS OF BROMELIADS 243 between the leaves, as well as mineral salts, are absorbed by the plant. A bromeliad is thus a veritable dialyzer which constantly removes from the pools formed between its leaves all the products of decomposition. CHAPTER XIV TURRIALBA AND PERALTA In our second chapter, on Costa Rican railroads, we have referred to Turrialba as the “pineapple station.” Our first visit to it was on July 24, 1909, when in company with Professor Tristan P. went there from Juan Vifas. We went first to an acquaintance of Professor Tristan’s, A. Alfaro, whose shop was in the lower story of one of the few two-storied buildings. Turrialba was built entirely of wood; the houses were chiefly shanties or a little better, many of these occupied by Jamaican negroes. One of the cabins was used as a jail in which two men were confined over Sunday and Monday for drunkenness—released on becoming sober and paying a fine or “multa,” part of which went to the support of the school. A better-built but one- storied house was labeled ‘“‘Alcaldia,” the magistrate’s court, while “‘Tesoreria Municipal” was the inscription over a little window like a ticket-seller’s, in the side of Senor Alfaro’s shop. Besides this last there were, I think, only two other two-storied buildings, one of which was occupied by J. Gomez and Brothers, the lower floor as a shop, the upper by sleeping rooms rented out to lodgers. ‘To this we were directed in search of a room for ourselves but, owing to the presence in town of engineers engaged in the extension of the railroad bridge over the Rio Turrialba and of a “Junta Politica” from Cartago to advocate the candidacy of the Republicano party in the Presidential campaign, all rooms were taken and our quarters seemed destined to be the rail- road station. However it was found that the engineers had TURRIALBA AND PERALTA 245 gone away for over Saturday night so that we were given their room; on Sunday, the politicians having departed, we obtained a room for ourselves—with, so far as I was con- cerned, the most uncomfortable, broken-down bed that I have yet occupied. Our meals, which were very good, we had at the little house of a Nicaraguan woman named Si- mona; the fare was Costa Rican but suited me perfectly, and Simona seemed much pleased that I ate her tortillas. Through the little town flowed the Rio Colorado, empty- ing a short distance below into the Rio Turrialba lying to the east. The latter was a very rocky stream with heaps of rounded boulders along its bed. The railroad crossed it on a stout iron bridge, dated as having been built by an English company in 1889. Its ends rested on heavy con- crete piers but the floods of the preceding December and January tilted the eastern end pier so much out of per- pendicular that the bridge was now being lengthened and a new concrete pier sunk in the soil some distance east of the tilted one. Each morning at Turrialba was clear and usually it was possible to see the summit of the Volcano of Turrialba dis- tinctly; later in the day it was apt to be cloud-covered. It is a more imposing mountain than Irazi, whether seen from the northern side as at Guapiles or Guacimo, or from the south as here at Turrialba. Its more imposing appearance is due not only to its greater apparent height, for Turrialba, Guapiles and Guacimo are much lower than Cartago from which we commonly saw Irazi, but also to the more conical shape of its upper portion. Looking at Irazi from Cartago one cannot see the cinders above the tree line, but Volcano Turrialba, seen from Turrialba town, shows the upper limit of the forest clearly and above that, at the western end of the summit, reaches upward a cone whose sides are deeply fissured and ridged, so as to form veritable canons. Pro- 246 A YEAR OF COSTA RICAN NATURAL HISTORY fessor Tristan, who had been to the top, told me that the ascent must be made by one of these cafions, whose steep- ness adds much to the difficulty of the climb, and that some sulphurous vapors are still given off by the volcano. Two eruptions of Turrialba volcano are known. One lasted from August 17, 1864, to the middle of March, 1865; during part of this time there fell a rain of very fine ashes and dust in the valleys of Cartago and San José, carried thence by the prevailing winds. The second and at times more violent eruption began February 1, 1866, and was accompanied by earthquakes which, on May 8, were felt at San José; the column of ashes reached to Puntarenas, a distance of seventy miles in a straight line. On July 24 we crossed the Rio Turrialba on the railroad bridge and walked for a mile or so through a potrero dotted with trees, a coffee-plantation, and a second similar potrero, all bordering the eastern bank of the river; we searched par- ticularly under logs, beneath the bark of decaying trees and in the sheathing bases of the leaves of bromeliads growing upon trees for earwigs, spiders and earthworms. During the course of our stay at Turrialba we found a number of these last in bromeliads seven feet or more above the ground; Professor Tristan supposes that the egg-cases have been brought by birds in the earth on their claws and accidentally dropped so as to enable the worm to develop occasionally in these plants. To me the most interesting find of the afternoon was one of the animals known as Peripatus, a form intermediate in many respects between the earthworm group on the one hand and the centipede group on the other. Peripatus and its immediate allies to the number of fifty species, known collectively as the Onychophora, are found only in the warm regions of the earth—South Africa, Malacca, Sumatra, Australia, New Zealand, South and Central America, Mex- TURRIALBA AND PERALTA 247 ico and the West Indies. Thirty American species have been listed by a recent writer (Bouvier, 1905). In zodlog- ical literature Peripatus is famous as a “connecting link.” We found two specimens this day, the larger one being about three inches long, of a reddish-brown color with yellowish dots and marks; it had a cylindrical body and twenty-nine pairs of legs. Both were found in one of the potreros above mentioned and, as is usual, under decaying logs. The alti- tude was about 2400 feet. I put both specimens in a bottle with some decaying wood and moss in the hope of keeping them alive, as Professor Tristan had kept one under such conditions for nearly a year, and eventually took them to our quarters in Cartago. On August 5 we made a couple of photographs of the larger one but with little success as the creatures became too restless in the amount of light requi- site. In September the smaller one was sent to Philadel- phia by a returning traveler, but did not reach its destina- tion alive. At that time we found two little ones in the bottle, but as these animals live in rotten wood away from the light we supposed we had gathered in more than we knew along with the food. On November 12, 1909, we saw that the large one was dead and mouldy and supposed all was over with the Peripatus family. However, we pulled out the soft wood and to our astonishment found four lively little ones, while everything in the bottle seemed sweet and wholesome except immediately around the dead one. These animals are viviparous and the large one was evidently the mother of the small ones that appeared from time to time. The infants were returned to the bottle and we hoped that they would continue to thrive. Within the next three weeks, however, two small ones were removed from the bottle, dead. We believed that two more small ones still remained, but on December 5 not a trace could be found and we suppose they must have died and disintegrated. A 248 A YEAR OF COSTA RICAN NATURAL HISTORY search for additional specimens of Peripatus in the same potrero at Turrialba on October 2 was unsuccessful. In one of these potreros was a jicara (Crescentia cujete) whose branches were intertwined with those of a manzana rosa (Jambosa vulgaris or Eugenia jambosa). The leaves of the former were rounded at the tip, those of the manzana acutely pointed. The single fruits of the jicara, which were ° still bright green, were as much as six or eight inches long, oval or egg-shaped; they contained many small seeds and the hard rind makes them useful “‘gourds”’ when the seeds are cleaned away, although the actual relationship is with the Bignoniacez, the family of the trumpet vine and the catalpa. The fruits of the manzana rosa resemble the guayabo, both belonging to the Myrtaceze, but are com- monly eaten raw instead of being made into dulces. The flesh has a pleasant flavor, suggesting a rather dry sweet apple—hence the name of ‘‘manzana” or apple. In the pastures were many clumps of trees of two or more species, one of which had produced numerous aerial roots which, running off sidewise from the trunk at right angles, clung to and encircled the trunks of the others, binding them all firmly together. Smaller trees or vines tightly wrapped their branches around the trunks of larger ones and the whole formed an inextricable mass. A common binding plant was the vine Monstera, often cultivated in greenhouses at home and remarkable for the large natural holes in its big deep green leaves. Trees loaded heavily with parasitic plants or epiphytes like Monstera, bromeliads, huge masses of long hanging ferns, great festoons of vine-like aroids with heart-shaped leaves, made this potrero exceedingly inter- esting. Here also, at the highest altitude at which we saw them (about 2400 feet) were a number of cacao trees, with clusters of small flowers and pods directly upon the bark of the trunk; the flowers were cream color, the pods TURRIALBA AND PERALTA 249 green or, when mature, rich brown, and six to eight inches long. On October 2, A. and P. revisited Turrialba and these potreros, when we had a good view of one of the grotesque basilisk lizards. This particular basilisk was on the ground, but on our approach ran ten or twelve feet up a tree and re- mained on the far side even when we walked around the tree. Its body was bright grass green, except that the tail was cross- barred with black. The lizard remained quiet as long as we stood still, but a cautious move preparatory to photograph- ing it caused it to dart instantly to the opposite side of a branch over which it peeped at us warily. On July 27 Professor Tristan and P. walked northeastward along the railroad track, crossing the Rio Turrialba and the Rio Aquiares by the railroad bridges, until we came to a third bridge over the Rio Azul. All three rivers lie in gorges more than fifty feet below the general level of the country and in places the sides of these gorges are steep, in others more gradual. The Rio Azul is much less rocky than the Rio Turrialba, the vegetation reached to the water’s edge and the scene was much prettier. We turned to the right into a charral and across it into the woods which immediately border the stream, but there was not a great variety of ani- mals of the groups in which we were interested. Among other things we looked for fresh-water planarians but found only three, representing possibly two species, one with little earlike lobes on the head. On the morning of July 25 we walked in a general easterly direction from Turrialba along a road which led to the Rio Reventaz6on and across it to San Francisco de Tuis, but did not go as far as the river. We passed a field of yuca; most of the way the road ran through pasture land but finally entered forest, into which we turned aside for a short dis- tance. We saw three snakes this morning, more than we 250 A YEAR OF COSTA RICAN NATURAL HISTORY had previously seen in Costa Rica. One was bright green, about four feet long; it was lying dead in the road. The second, apparently of the same species, glided into the grass; the third was smaller, mottled brown in color. All three were harmless. By the roadsides were many melas- tomaceous trees, lengua de vaca, which also occurred com- monly at Cartago, and on some of the higher trees were nests of the oropéndolas. One man whom we met along the road told Professor Tristan he knew what we were doing—we were collecting butterflies (we collected none this day) to print them upon muslin and thus give the latter its decorative pattern! Sixty years earlier the Brazilians gave the same explanation to Wallace of his actions on the Amazon. North and west of the town the country ascended in gentle swells and low rounded hills, intersected by little streams entering the Rio Colorado. Along these streams were bright green pastures extending high up on the hills, giving the surroundings a soft and billowy appearance, softer, smoother and greener than we saw elsewhere in Costa Rica. The hillsides were formerly forest-covered and here and there single trees had been left standing so that the effect was very park-like. Here in July afternoons before the usual rain, flocks of twenty to thirty green parrakeets (probably Conurus finschi) flew screaming over our heads, resting first on one large tree, then on another, their green color rendering them invisible among the foliage. Their most characteristic feature when in flight is the bluntness, even squareness, of the head in front, owing to the shortness of the bill. Professor Tristan told me that this parrakeet does not learn to talk in captivity. A parrot found on the Pacific side talks readily and for this reason is often kept as a pet— there was one next door to Simona’s and Mrs. Ridgway at Juan Vinas had one. This Pacific, talking, species (4ma- TURRIALBA AND PERALTA 251 zona auropalliata), known as the “loro,” is light green with a yellow spot at the back of the head and a pair of black spots above the base of the bill. The primary wing feathers are chiefly black but are largely concealed by the green plumage when the wings are folded; there is some red on the edge and bend of the wing. Just west of Turrialba the railroad passed through one of these beautiful, green, soft-looking hillside pastures and then bordered a thick forest. Along a brook where it left the forest but still shaded by some trees, were a larger number of a species of Heteragrion dragonfly than I had previously met, although I had found it below Juan Vifias. It had a brilliant patch of gold on the front of the head between the eyes and a one-and-one-half inch long, slender, blood-red abdomen; the colors of the rest of the body were more som- ber, the wings transparent, and in the shadow it was the golden patch and the line of red that were chiefly visible. These colors are possessed only by the male, the female being a sober brown. The species of Heteragrion, as far as I know, are only met with in well-shaded places, never in bright light, and by reason of this habit are often somewhat difficult to see, especially the females. Leaving the stream we entered the forest and attempted to pass through it up the hillside into a pasture above, but there was no path and the ascent was so steep and the vege- tation so dense that after going a little way we gave up our plan. Besides the deciduous (exogenous) trees there were many palms and tree-ferns all bound together with vines and lianas. ‘There were numerous epiphytic ferns which also had a vine-like habit and hung in festoons from the branches of trees. Many trees, lianas and even some ferns had their stems and branches armed with sharp and cruel thorns so that in catching hold of the vegetation it was necessary to look carefully to see what was safe to grasp. 252 A YEAR OF COSTA RICAN NATURAL HISTORY A not uncommon low plant on this hillside was a palm- like grass with leaves divided into two lobes each ending in an elliptical point and known (teste Professor Tristan) as “cola de gallo” (cock’s tail). Two of these cock’s tails, of about three feet in height, had curious nests upon them, attached in each case to a leaf. Both nests were inhabited by small black ants, but the nest-substance resembled that of which termites’ nests are formed and the plan was termite- like, although there were no covered passages leading to them as termite nests have. One leaf was so decayed that, soon after I plucked it, it fell apart. It is difficult to understand why nests should be built on so temporary a foundation as a leaf must afford. Only a few yards from the site of these nests I caught one of the most ethereal of butterflies (Callitera menander). Its wings, spreading two inches, were chiefly transparent and brownish, but each of the hind pair had a blue-black ““eye-spot” on the outer margin and a patch of delicate rose along the hind edge. It is of the Satyrine subfamily of the Nymphalide distinguished by one or more of the veins of the fore wings being swollen at the base. On Saturday, August 7, 1909, P. went to Peralta. On my arrival I presented to Mr. B. M. Hess, the station- master, a letter of introduction from Mr. John M. Keith. Mr. Hess (a native of Lancaster, Pennsylvania) took me to his own room in the second floor of the station and placed a cot at my disposal. I had brought a mosquito net with me, under which I slept each night, as I had been warned that Peralta was malarious. In fact, Mr. Hess said that he was suffering from it; he had a net but was not using it, although when I left he was promising himself to do so. Red ants were abundant in our bedroom. I know I picked many from my cot; perhaps they kept other insects away. Mr. Hess inquired if two other kinds of insects had eaten me up; TURRIALBA AND PERALTA 253 he said they were bad in the next room. I did not see either of these, but I preferred not to look for them and I do not know the actual authors of the numerous bites on my person! Mr. Hess kept a ‘‘commissary,” as general shops were called in this country. He had a Jamaican mulatto woman as cook. She had a brown monkey, a squirrel, two dogs and a cat in the kitchen or the little room next to it,—not the most sanitary arrangement perhaps. For company at meals, besides a young Costa Rican assistant and myself, Mr. Hess usually had—except at our six o’clock ‘“‘coffee”—two or more of the railroad engineers or conductors whose trains might happen to be in Peralta at the time, for Peralta was a station where much shifting of cars, coaling of locomotive tenders and, on Sundays, transfer of passengers took place. One other item of the establishment was the Victor phono- graph kept in the bedroom. This nightly rendered American songs and recitations in my honor while I attended to the day’s catch, and Mr. Hess fell asleep on his bed. A second visit to Peralta was made on March 23, 1910, when I put up as before at the railroad station, Mr. Hess having sent me word that he thought conditions good for collecting. Near noon on August 9 I was standing on the station plat- form when a very curious bug alighting upon it attracted my attention. Its body and folded wings together were about an inch long, its colors yellow, orange and red, as well as I can remember, but its most striking feature was its hind or third pair of legs. Each one of these was one and one- third inches long and very slender except that the tibia had a thin leaf-like expansion half an inch long and one-third of an inch wide, with cross-bands of orange, brownish-red and blackish-brown. I was holding the insect in my hand examining it when most of it flew off, leaving but one of these remarkable legs in my fingers. I was unable to capture 254 A YEAR OF COSTA RICAN NATURAL HISTORY it again and so have nothing but this single member in my possession now, but this disyecta would seem to place the insect in the genus Anisoscelis of the family Coreide. Later that same day it rained and while sitting on the porch of the second story, I watched the dragonfly Orthemis ferruginea fly to and fro over the pool alongside the tracks in spite of the (moderate) rain, although dragonflies are usually fair weather insects. Four or five green parrakeets (Conurus finschi?) flew screaming to a tree on the opposite side of the river, their green color showing well as they flew. By 3.45 P. M. the rain had ceased and I walked north along the track. Not far from the station, a curious bird whose character I could not make out, flew across the track and into the wild cane, of which there was much between the railroad and the river both above and below Peralta station. The cane was now ten to fifteen feet high. I kept looking into it as I approached the place where the bird had entered, hardly expecting to see it again. But when I reached the place, there on a slanting cane, sat my first toucan, big beak and all. J stopped to look at it, it sat and looked at me, giving me the chance to make mental note of its colors be- fore it flew off. It was chiefly black, with a yellow breast bounded with black below, and with an isolated black streak in the middle of the yellow, the rump red both above and below. It was Pieroglossus torquatus, one of the smaller toucans of Costa Rica, this species being about eighteen inches long. Similar to it, and occurring chiefly, if not solely, on the Pacific slope according to Carriker, is Pt. frantzi1, which has a wide crimson band, sometimes mixed with black, across the middle of the yellow belly. Another genus commonly found in Costa Rica has a yellow throat and black breast; this is Rhamphastos. R. tocard, the ‘‘quioro” of Costa Ricans, is twenty-two to twenty-four inches long and has the tip of the maxilla bright yellow. R. brevicari- TURRIALBA AND PERALTA 255 natus, the “curre grande,” with this tip “deep purplish blood-red,” is sixteen to twenty-four inches in length. Not only the toucan thus justified my going out for this anteprandial walk, but also at the first little river north of Peralta station I took the only specimen of Heterina miniata found anywhere in this locality, and probably the highest elevation (1075 feet) from which this species has yet been reported, as it seems to be chiefly a lowlander. Like almost all members of the genus Heterina, the male has a beautiful carmine spot at the base of each of its four narrow, trans- parent wings. This carmine is accompanied by almost no brown along its front edge, in contrast to the condition existing in the ubiquitous Heterina cruentata. Moreover, miniata male has a carmine spot on the tip of each hind wing (where cruentata male is edged with brown) and only one pair of clasping organs at the hind end of the body (the lower pair existing in cruentata being represented merely by stumps). Heterina is an exclusively American dragonfly and is richer in species in the tropics than in the temperate zones, although one member reaches Canada. There were about half a dozen small cabins near Peralta station occupied by Jamaican negroes, and some railroad buildings, but noother human habitations. Behind the latter extended a Y-shaped track to enable locomotives to reverse their heading. At the end of the stem of the Y was a slow- moving stream called simply “laguna.” Mr. Hess having told me of this as a likely place for me, I went there on August 8 about noon. Two women were washing clothes at the laguna, so I went into the low woods just beyond, which consisted of small trees, arums, ferns, heliconias and numerous vines or creepers. I had hardly entered when I caught a medium- sized dragonfly, of an apparently new genus allied to Cora. Nearby I found a pair of Palemnema dragonflies, likewise new, with extremely slender bodies (two inches long in the 256 A YEAR OF COSTA RICAN NATURAL HISTORY male). The egg-laying habits of Palemnema are unknown and I stood in the mud for forty minutes keeping my eyes fixed on the pair, in the hope that I would obtain some information on this point. During this period they shifted their position to five different leaves or twigs but gave no sign of depositing eggs, so, fearing that I might lose them, I caught them in my net. Experience had taught me that the dragonflies which inhabit dark woods are very difficult to keep in view, owing to the transparency of their wings and the prevailing somber tint of their bodies. Often there is, however, a small limited area of a lighter color and, if one can keep his eye on that, he can follow the insect as it flits from plant to plant, but even so the greatest attention is necessary and J have often lost desirable insects. In this Palemnema the most conspicuous light area, but not the only one, was formed by two small parts of the head—the labrum (or upper lip) and the nasus—which were pale blue. A species which I lost in this same piece of woods this day, a Philogenia, had only one or two segments of the hind end of the abdomen pale colored, like the whitish “bloom” sometimes seen on a plum; all the rest of the body was dark and the insect kept to the shadows so closely that it was very difficult to see. Some writers, like Mr. E. B. Williamson, have suggested that these more conspicuous pale areas on dark bodies of dragonflies (not necessarily those confined to dark woods) serve as recognition-marks to the other sex. I naturally looked for additional specimens of the new genus and of the Palemnema, but unsuccessfully, until the increasing cloudiness warned me to seek shelter before the beginning of the usual heavy afternoon rain. Search for them was in vain on the following morning also, when I attempted a grassy road which led up from the laguna into the woods, and which I finally gave up because of the deep ‘eqewiny, iv o1alj0g & ul sayAydidy yim uapry] seer, To face p. 256 Wild Ginger—Costus malortieanus. 1. Vijagua, Calathea insignis, with yellow bracts. 2. Heliconia marie and 3. H. imbricata, both with dark red bracts. To face"p. 257 Oy. TURRIALBA AND PERALTA 257 soft mud beneath the grass. However, I did obtain some species of dragonflies that I had never, or but rarely, taken myself in Costa Rica. Feeding on a patch of morning-glory (Ipomea), known in Costa Rica as ‘“‘churristate’’—the same name is applied also to a mallow, Anoda hastata,— near the laguna were some uniformly dark metallic violet- blue Chrysomelid beetles (Haltica amethystina). Each one was one-third of an inch long and had many fine punctures over its dorsal surface but not arranged in lines. In the afternoon of August 10 I went again to the end of the Y-track and into the woods beyond the laguna, where I had taken the new species two days before. I did not see them again but there were other interesting things. Both here and along the Rio Chiriqui trail where I had been that morning were a number of those transparent, pale-brown- winged butterflies with an eye-spot and rose color on the hind pair (Callitera menander), one of which I had taken in the woods at Turrialba on July 26 (see page 252). I tried to capture some of them but did not find it easy, for unlike most butterflies, they did not rise when alarmed but flew horizontally close to the ground, through or beneath the underbrush, and the dead or living stems and branches prevented me from getting the net over the insects quickly enough to catch them. I saw them only in dark shady places with plenty of undergrowth. Lying on the glossy green leaf of an aroid about a foot above the ground, in a fairly shady but hardly dark place, were a dead leaf and apparently a twig three inches long and one-fifth of an inch thick. Both looked as if they might have fallen by chance upon the living leaf. But on second glance I noticed that from the apparent twig there extended two pairs of slender outgrowths, the members of each pair placed so exactly opposite each other as to suggest at once that they were insect’s legs—and so they were. The appar- 258 A YEAR OF COSTA RICAN NATURAL HISTORY ent twig was a wingless walking stick (Ceroys bigibbus) whose color and roughness of body-surface were perfect imitations of the color and surface of a broken twig. The antenne and the first pair of legs were held close together and straight forward in prolongation of the body and did not betray their possessor as did the oppositeness of the re- maining legs, which projected at marked angles. The most interesting feature of the observation to me was that the insect was resting not on a woody stem or branch, where the protection derived would have been greater, but upon the contrasting background of the green leaf. It is not uncom- mon for veritable twigs to fall upon such leaves, but the occurrence of this insect upon the leaf certainly indicated an absence of the power of selecting an advantageous resting place where its “protective coloration” would protect. The walking stick was taken to Cartago and photographed while still alive. Coming out of the woods, I stopped at the laguna to dredge for dragonfly larve, but without success. My brown tin collecting case was lying in the sun on the ground and some patches of black paint, which had been put on five months previously and were thoroughly dried, attracted a number of large brown-winged horse-flies (Pangonia pyrausta) with probosces nearly as long (one-half inch) as their bodies, which were five-eighths inches long. I paid little attention to them at first, but when one of the insects flew again to the box after it was placed in the shade thirty feet or more from its former position, I caught the fly for future identifi- cation. The paint had not softened, so I do not see what the attraction for the insect was. On March 23, 1910, I visited the laguna again, crossed it and went into the thicket nearby to the spot where in August I had taken Palemnema, the new genus, etc. A small clear- ing had been made here since then and a miscellaneous lot TURRIALBA AND PERALTA 259 of vegetables and plaintains planted. This, with the drier weather prevailing, had so changed the character of the spot that I could hardly recognize it. After retracing my way nearly to the laguna I followed a road which in August I gave up on account of the deep mud. Here I found a dragonfly which I had never seen alive be- fore and which no one had previously taken in Costa Rica— Gynacantha tibiata. It is two and three-quarters inches long and quite striking when flying about because it was bright green at the head including the large contiguous eyes and bright orange at the tail. This road brought me to a swift brook and I spent some time following it up and down until the skies grew clouded and rain began about four o’clock. The rain continued at intervals through the night. During the preceding twelve or fourteen days it had not rained at all, according to the record of Mr. Hess’s rain gauge in front of the station. Two days later I went up the bed of this brook again and examined its banks. Many of the stones which formed the latter were breaking up into clay and one could see all the stages of the disintegration— some were split up into irregular masses so that a kick easily separated the whole, each fragment, however, retaining its shape; others were evidently dissolving rapidly under the action of rain. The vicinity of Peralta was rich in wild cane and in heli- conias (wild plantains), Zingiberacee and Marantacez, the shoots of these plants frequently reaching a height of fifteen feet. Since in the latter orders the apparent stem is really composed of tightly wrapped leaf-sheaths, this height means that a single leaf and its stalk together may be fifteen feet long. The species blooming in March were different from those blossoming in August. The flower clusters of the Heliconias were large and enclosed in bright red bracts, the conspicuously colored portions of the inflorescences. 260 A YEAR OF COSTA RICAN NATURAL HISTORY These bracts are so shaped that they can hold an ounce or so of water. When I examined several inflorescences on March 25, 1910, I found in each bract two or more mosquito larve or pup and in some other quite different dipterous larve also. Some of these mosquito larve were eventually carried to Philadelphia alive in a vial and, although nothing was given them for food but merely the supply of water maintained, they transformed to the adult stage on May 31, 1910. Some of these larve and a pupa were identified by Mr. Knab as Wyeomyia pantoia. They therefore belong to the same genus as the mosquito (W. smithi1) whose larve breed in the water contained in the pitchers of the pitcher plant (Sarracenia) of the United States. The larve of W. smithi are known to be able to withstand fasting for long periods. Larve of other species of Wyeomyia live in the water between the leaves of bromeliads or in bamboo joints in tropical America. The adult W. smithiz is believed not to suck blood, but of W. grayit of St. Lucia and other West Indian islands, Theobald reports that it “bites with great readiness and pertinacity.”” Busck, who studied mosquitoes in Panama, says, “The species of the genus Wyeomyia . . are among the few day-biting mosquitoes and are decidedly noxious where they abound . . . are very persistent biters during the day-time.” Heliconias, bromeliads and bamboo, therefore, offer a breeding place to mosquitoes which annoy man. Behind Peralta station (that is, west) the forest began within a few hundred feet of the railroad. An ascending road led into it, which I followed for some distance on August 7, and then turned aside on the first path to the right. Im- mediately I was in the midst of a thick forest of very tall ex- ogenous trees and many palms, from both of which hung clumps of ferns and lianas binding the trunks and branches together. The soil was damp and, in places, deep mud. TURRIALBA AND PERALTA 261 Although the sun was shining, the foliage was dense enough to prevent strong light from reaching the ground and in consequence the dark woods had a rather mysterious aspect. An onion-like odor made me suspect peccaries. I stole along the path as quietly as possible, hoping that one of the big Staurophlebia dragonflies, such as I had seen at Guapiles and at Guacimo, would come along the path to meet me and my net, but neither of these hopes was gratified. Instead, somewhere far above, was the sound of break- ing branches and a little later a short bark something like a dog’s. Still I crept on cautiously, looking constantly up- ward into the tree-tops; in another minute I saw, in an opening amongst the leaves which permitted of a clear view of his whole person—my first wild monkey! It was the common white-faced or capuchin monkey (Cebus capucinus), such a one as lived chained to a pole in the patio of Weldon’s Hotel at Cartago. This free fellow looked down at me from his bough for a few minutes while I looked up at him and then, although I made no movement to alarm him, he turned and disappeared among the leaves. Near him were two or three others which I could not see as well. Ten or fifteen minutes later, and at a short distance away, I had another and still clearer view of one of these monkeys as he ran along an almost bare branch of a great tree and stopped midway to watch me. Between the two views of the monkeys I saw a Megalo- prepus dragonfly, the largest of all living species, mentioned in more detail in the chapter on Santa Clara. It alighted first on one and then on another twig, seven to ten feet above a little stream whose waters were arched over and partly covered by vegetation. The interest of this observa- tion lies in the fact that in the Biologia Centrali-Americana under this species is quoted the remark of an experienced collector who spent four years in Central America, that he 262 A YEAR OF COSTA RICAN NATURAL HISTORY had always found this species in deep woods and far from water; here the woods were deep, but—. The next morning I went to these woods again but al- though I spent two hours in the immediate neighborhood, I saw neither monkeys nor Megaloprepus, nor indeed much of any kind of animals and no dragonfly larve in the streams. So I returned to the road, which continued to ascend through the forest; here and there in it were pools of rain-water, occa- sionally supplemented from little running streams which preceded the human-made road or have taken advantage of it. I had caught a dragonfly at one of these pools and was putting it away when, looking ahead, I saw an arma- dillo (Dasypus novemcinctus fenestratus) crossing the road some twenty feet away and disappear into the shrubbery. My next visit thither was on March 24, 1910, a cloudy day with fitful gleams of sunshine and I could find few dragon- flies, no monkeys and no armadillos. Mr. H. H. Smith, when collecting in Brazil, observed that certain dragonflies were only or chiefly to be found just after sunset along river banks. With this in mind, I went after dinner on the first day at Peralta (August 7, 1909) to the Rio Reventazén about one hundred and fifty feet away. Mr. Hess called a Jamaican negro to guide me through some garden patches and cane-brakes, after which we crawled around and over numerous boulders and so came out on a flat, treeless and rocky beach. The Reventazén is moder- ately rough here between rapids above and below. My guide showed me the height to which the river reached in the great floods of the preceding December and January. Be- tween the river and the station the land rises a little and in the same floods—as Mr. Hess told me at another time—the Reventazén broke through its banks farther upstream and came down the railroad tracks to a depth of four feet near the station. Owing to landslides above and below which TURRIALBA AND PERALTA 263 carried tracks, telegraph poles and wires and everything into the river, they were seven days with no communication with the outside world, while day and night the air was filled with the continuous roar of the rocks and boulders swept downstream by the tremendous force of the river and constantly striking and loosening others as they were hurled along. A conductor supplied another picture of the Man- ager of the United Fruit Company obliged to stand on a chair in the chair car on his train because the car was flooded. But I saw no dragonflies by the Reventaz6n in the fast falling darkness. Following a suggestion of Mr. Hess, on August 9 I went along the track southward, as far as “mile post 55” from Limon, stopping on the way at a small stream crossed by a short railroad bridge where I took a couple of additional species of dragonflies for Peralta. A short distance beyond post 55 was a siding turning off to the east; just north of it was a corrugated iron cabin and a pretty grass-covered path also leading east to the Reventazén. This soon brought me to the river which at this point had no pro- jecting rocks although above and below there were foaming rapids and boulders. A wire cable was stretched across the river here from platforms of logs lodged in tree-trunks about eight feet above the ground on either bank. On the cable, at each bank, was a tiny car consisting of a couple of grooved wheels from which hung two short pieces of rope each sus- pending a wooden crosspiece. There was no pulling rope across the river and apparently the only way of crossing was to sit upon one crosspiece, put the legs upon the other and then laboriously pull the car along by hand-over-hand work on the wire cable—probably cutting one’s hands to pieces in the attempt. Mr. Hess had said that I could cross the river by this car if I wished—but I did not wish! At the opposite end of the cable a man was apparently about to 264 A YEAR OF COSTA RICAN NATURAL HISTORY cross by the car at that end, but if so he gave up the idea and disappeared into the forest. A black cormorant was sitting on the rocks of the rapids downstream but did not linger, although I did not actually see him fly away. The river banks were lined with fairly large trees whose branches, on the river side, overhung the water for many feet; they were heavily loaded with epi- phytes—ferns, Spanish moss and other bromeliads. Looking down on the rocks at the very water’s edge, I saw a basilisk lizard (Bastliscus) perhaps a foot and a half long, with a long conical projection on the top of its head and a fold of skin along the back. As I stepped a little nearer to get a better view, it disappeared in a twinkling behind a rock and per- haps into the water. While returning to the railroad by the grassy path a second and similar basilisk gave me a better opportunity to examine it. The lizard was bright grass- green in color, somewhat over a foot in length; its head was produced upward in a conical fashion that gave it a most grotesque appearance and there was a dorsal crest to its body seemingly not fully elevated but plaited. When I came very near it ran off into the underbrush. Some ‘‘bobos,” fishes of delicious taste which I had pre- viously eaten at Guapiles, were for sale at the Peralta Com- missary on August 10, a negro having caught them in the Reventaz6n with a net. The negroes present pointed out, as characteristic of the bobo, a horny tubercle on the front end of the middle line of the upper jaw. A few days later Professor Alfaro gave me the technical name of the bobo as Joturus pichardi; it is a member of the mullet family (Mugilide). Mr. Hess suggested that I go north along the tracks to the Chiriqui(?) River, whence, by a trail leading upstream, I could reach another piece of forest. He had once followed this trail on horseback as far as two “crater lakes” of which TURRIALBA AND PERALTA 265 little was known, he said; one, much the larger, was eighty manzanas in area, the other about thirteen manzanas. (A manzana is a Spanish square measure equalling 10,000 square yards.) The crater lakes were too far away, by his account, to be reached on the cloudy morning of August 10, but other- wise I adopted his suggestion. The Chiriqui(?) River was crossed by the railroad at about milepost 5314, if there were one present but I failed to find it. The long-tailed, black and green, day-flying moth Uranidia fulgens, mentioned on page II0 as so much resembling a Papilio butterfly, was very abundant this morning and a little north of post 5314, I stopped, at 7.50 A. M., to make a definite count. There was evidently a migration in progress, for the moths were flying northwest to southeast steadily without returning or diverging. Most of them were high in the air above the tops of the forest trees remaining near the tracks, but enough were flying lower to be plainly identified. I counted 176 passing just above me in two minutes and then stopped because the glare from the sky was too hard on the eyes. There was no wind; the sky was mostly clouded over, the sunshine feeble. The migrating column was of no great width. The banks of the Rio Chiriqui were heaped with rounded boulders. A negro who was gathering old ties and logs told me that the trail I was seeking was on the north side of the river, and crossing the railroad bridge I found its starting point at the very abutment. But before turning into it I walked along the tracks a little farther to look at the Reventazén, which was filled with great rocks that made it a boiling rapids opposite the mouth of the Chiriqui. The trail entered the woods a short distance from the railroad. The principal characteristic of this piece of forest was the great number of lianas which hung from the trees. They were of all sizes from those little thicker than light twine to 266 A YEAR OF COSTA RICAN NATURAL HISTORY huge cables as thick as a man’s body and composed of sev- eral smaller lianas, each as thick as one’s wrist, wound round each other. The trail broadened to ten feet or more in width, was much cut up by horses’ hoofs and in many places extremely soft and muddy. It crossed several small streams which were not affluents of the Chiriqui but flowed in the opposite di- rection. Finally I reached a clearing with much grass and a few trees, beyond which was a grove of large trees with no undergrowth or grass, the ground being rough and muddy. Crossing this clearing I caught a Megaloprepus dragonfly, much to my surprise—the locality was indeed far from water but certainly not “deep woods.” The buttress-roots of some of the trees in this grove enlarged the diameters of their trunks at the ground to twenty or thirty feet. The altitude of the little hill on which this grove was situated was 1400 feet by my aneroid. From here I returned to Peralta as I had come. In the morning of March 25, 1910, I went to the Rio Chiriqui again, but instead of going into the woods border- ing it on the north, as I did before, I went along the river bank itself, which is formed of boulders of various sizes. The river had several channels and in the southernmost of these, where the current was less swift, were not a few small dragonflies with brilliant coppery heads and thoraces and bright sky-blue abdomens (4rgia near cupraurea). When I had gone upstream several hundred yards from the rail- road two deer crossed the river about fifty feet away from _me. I saw them before they discovered me but one soon turned and ‘spied me; they quickened their pace but did not run and soon Aeanpeured into the shrubbery, which on both banks intervened between water and forest. These deer stood not more than four feet high and had no horns. Ac- cording to Elliott, there are four species of deer which prob- TURRIALBA AND PERALTA 267 ably occur in Costa Rica, Odontocelus truei, costaricensis and nemoralis, and Mazama sartori. Miller spells the first of these generic names Odocoileus and gives tema as the specific name of the Mazama. Odocoileus, which includes the Virginia deer of the United States, has branched or un- branched antlers (but on the male only) and usually a long tail and a metatarsal gland. O. truei has (in its summer coat) the upper parts light chestnut or tawny, the tail tawny above, the height at the shoulders 27 to 29 inches. 0. costaricensis is larger than O. truei, is mixed black and buff above, has the tail cinnamon above and a total length of 56 inches. The upper parts of O. nemoralis are yellowish- brown and gray, the tail dusky above, the height at the shoulder 2714 inches (Lydekker). Mazama has simple unbranched horns which are directed backward, a short tail, the hair on the forehead forming a tuft, but no metatarsal gland. M. tema (or sartori) has the upper parts of the body and of the tail brownish-red, the height at the shoulder 2014 inches. It is one of those deer known as brockets. Between four and five that afternoon I saw a few toucans by the Chiriqui, both perching and flying. When flying, a toucan is elongated and bent down in front and so at the opposite extreme from a parrakeet in flight, the latter appear- ing very short and “sawed off” in front. The green-and-golden-eyed, blood-sucking flies, Chrysops tanycerus, were here in numbers to-day. This genus of the horse-fly family is widely distributed over the earth and per- secutes man and other mammals wherever it occurs. This day being Good Friday, not a train or an engine moved in either direction on the railroad and a construction train which arrived at Peralta on Thursday evening waited until Saturday before going on. Such was the custom of the country rather than the wishes of the railroad management. CHAPTER XV SANTA CLARA The level country on the north side of the bases of the Cordillera Central is known as the “Llanuras de Santa Clara” or, by abbreviation, merely as “Santa Clara.” Our first acquaintance with this region was made on Tuesday, June 1, 1909, when P. took a letter of introduction from Mr. E. J. Hitchcock, Manager of the United Fruit Company for Costa Rica, to Mr. E. W. Reed, Superintendent of the Santa Clara Division, located at Guapiles. As far as La Junta the route was the main line from San José to Limén; at La Junta the tracks diverge to the so-called “Old Line.” About two o’clock I was deposited on a tiny platform between two tracks at La Junta and the train went on to Limon. There was no one about but negroes who worked on the adjoining banana farms. A row of their cabins faced the track. My questions brought the answer that the train for Guapiles would stop at that platform about three o’clock —the schedule time for leaving La Junta was 2.40 P. M— and that there was no station here. The temporary bridge in use in June was a little upstream from the site of the old one, which caused some change in the track arrangements. It soon began to rain, so I went to one of the cabins, where a young negro woman was washing clothes, and asked permission to sit on her little porch under shelter. This was readily granted and a chair was also brought out for me. A number of men were lounging on the porch of the next cottage and remained there all the rest of the afternoon, one of them strumming on a guitar; the most frequently SANTA CLARA 269 repeated air had a refrain of which the intelligible words were “Sally Maun day.” I saw many other Jamaican negroes during this week for they do nearly all the work in the hot banana fields. They speak English, nominally, but it was often difficult to understand them and apparently for them to understand us. Opposite the cabin where I sat was a tall tree whose trunk rose straight from the ground without a branch for about fifty feet; then its branches spread rather widely with not very thick foliage and with long depending wisps of Tillandsia. Attached to many of the branches were the hanging nests of the lowland oropéndolas. Each nest looked to be one and a half to two feet long, thickening somewhat from top to bottom and with an opening on the side. I counted fifty-four such nests on this one tree. The birds made frequent visits to the nests; their flight was strong and their call a long and somewhat calliope-like whistle. By and by the rain stopped and I walked a little in the neighborhood of the junction though I dared not go far since the train was already overdue. A banana plantation came down to the tracks on the southern side and a little pool lay near it. As I approached, two green lizards, each a foot long, left the water’s edge and darted into the bananas. Following them, I saw on a tree (forming part of the fence around the bananas) at about two feet from the ground, one of those tree nests of termites or so-called “white ants,” well-known in the lower parts of tropical America. This nest was about the size of a man’s head, brownish and made as usual of the wood which has been chewed by the termites and so worked into building material. Breaking a small hole into it, the termites could be seen running about actively within. The sound of a locomotive broke off my observa- tions and I returned to the platform. It proved a false alarm and by five o’clock (the scheduled time for arrival at Gua- 270 A YEAR OF COSTA RICAN NATURAL HISTORY piles) when no train had yet appeared, I began to wonder whether it had been omitted for the day. However it came a little later and left La Junta for Guapiles at 5.25 P. M. It was a mixed train with more freight cars than passenger, and the only one of the day in that direction. I learned later that it was usual to go on past La Junta to Siquirres, where the Guapiles train started, and wait there, but no one had told me this. Further, Tuesdays, Thursdays and Saturdays were heavier freight days than the others and in consequence the train was always later on those days. Between La Junta and Guapiles Mr. Reed found me on the train and I had the pleasure of his company the rest of the way. He told me that in the twenty miles between the two places there was the possibility of sixty-three stops on the railroad. This was because there were practically no roads for teams in this part of the country and conse- quently everyone depended on the railroad for the carriage of all goods. We did not make the sixty-three stops but there were enough of them and enough freight to be unloaded to make it after nine o’clock when we reached Guapiles and Mr. Reed took me to “Salvador,” as the Superintendent’s house was named. Here I met Mrs. Reed and here I unex- pectedly found Mr. Schaus and Mr. Barnes who had ar- rived the day before to collect Lepidoptera. Salvador was approached from the railroad by a sivaight walk bordered on each side, for part of its length, by a row of majestic royal palms; some gaps which had been unac- countably left had been filled in with younger specimens. The house itself, about three years old, was of frame, as was the case also with all the farmhouses of the region. It was two-storied, square, with a wide veranda running completely around all four sides on both stories, a central hall for the entire length of both, and in the rear a detached but con- nected building containing a kitchen and pantry below and SANTA CLARA 271 a well-equipped bathroom and other small rooms above. In the main house the interior of all the rooms—walls and ceil- ings—was finished in dark stained wood. All the sleeping rooms were on the second floor and at night all windows were left open—a welcome change from conditions enforced by sleeping on the street level at Cartago. From the upper veranda at the front of the house there was, in the clearness of each early morning, a magnificent view of the Cordillera Central lying just south. From left to right one saw Turrialba, Irazi, Barba and Pods. The view of them from Guapiles is more imposing than from Cartago and San José because Guapiles is but 1000 feet above the sea while Cartago is 4750 feet and San José 3800 feet in elevation, so that the mountains gain immensely in height when seen from here. Turrialba, which is more or less hidden by Irazti' when looked at from Cartago or San José, is especially fine. The northern slopes of all these mountains were well wooded to the summits, which was in great contrast to the appearance of Irazii from Cartago— the forests being largely replaced by cattle pastures. Mr. Schaus and Mr. Barnes offered to show me some of the best collecting grounds which they had found on their previous visits to Guapiles, an offer which of course I was very glad to accept. On the morning of the second we crossed to the south of the railroad, went along a row of wooden houses one of which bore a large sign “Blue Nose Villa” for no visible cause, passed a sawmill, crossed two potreros diagonally and so arrived at a bit of forest into which a trail wide enough for a rider entered and soon forked right and left. The second of the above-mentioned fields still contained a number of trees and shrubs, remains of the forest that once covered all this country solidly, as indeed it still did in large part. One of the tall trees, with all the characteristics of the tree I noticed at La Junta, had like- 272 A YEAR OF COSTA RICAN NATURAL HISTORY wise a population of hanging “yellow tails’”’ nests and the call of the birds could be heard for a considerable distance. The number of nests was much smaller than on the tree at La Junta. Not long after entering the forest, I had my first view of a living example of the largest species of dragonfly now known to exist—Megaloprepus cerulatus—found in the lower parts of tropical America from Mexico to Bolivia. Its very slender body attains, in the largest individuals, a length of five inches and. it has a wing-spread of more than seven inches. It owes its specific name, cerulatus, to the fact that on its large otherwise clear wings is a broad, dark, purple- blue band which makes the insect conspicuous when flying through the forest. Some writers (who have not seen this insect in life) have suggested that the dark bands on the wings, by resembling flickering shadows cast by leaves, etc., are thereby protective, but I could not see that the supposed resemblance exists. This day I saw two or three individ- uals flying from dense vegetation into the trail. When fly- ing all four wings were spread far apart from each other, the body was horizontal. The flight was sufficiently slow to allow the movement of each wing to be seen; but in spite of this slowness the insect can dodge. Mr. Barnes com- pared the motions of the wings to those of a windmill, but I should say that the effect produced by the moving wings was more like that of a jumping-jack with movable arms and legs pulled by one string, rather slowly but at regular intervals. When caught and held by the wings, the legs were folded against the thorax and remained immobile, even when they were touched or rubbed—the insect seemed to “play ’possum.” When held by the mid-abdomen, move- ments in legs and wings soon appeared. When I struck at individuals in the forest with my net and missed them they usually rose in the air, at the same time making into SANTA CLARA 273 the vegetation. A male was kept alive in an envelope and in the evening was let loose in my bedroom. The walls and ceiling were of dark-stained wood, while there was yellow straw matting on the floor; the height of the ceiling was be- tween eight and nine feet, and from the center hung an in- candescent electric light which I placed at about mid-height. Above the bulb was a flaring fluted shade, the effect of which was to render the space above darker than that below. With the illumination afforded by this one light, the Megalopre- pus was repeatedly released and repeatedly flew up to the ceiling. Once indeed it got under the fluted shade and bumped up against the latter several times until I gently moved the insect out from under the shade, when it flew up- ward to the ceiling again. Nothing whatever is known of the nature or habits of the larva of this species of dragonfly. Another interesting dragonfly first found this day, of smaller size than the above and with a beautiful red abdo- men, was Orthemis cultriformis, originally described from Paraguay and which had not hitherto been known north of Panama. Many beautiful butterflies were in this wood, some not greatly unlike those at home, others of characteristically tropical groups. On June 4 I visited this bit of forest again; there was a young bull in the trail but he gave no trouble. I went up first the left and then the right fork of the trail already men- tioned as far as a point on each where it had become so choked with bushes that further progress was difficult. An early prize this morning was a great green, strong-bodied dragonfly, Staurophlebia reticulata obscura, originally de- scribed from Surinam and hitherto recorded only from Hon- duras, Nicaragua and Chiriqui in Central America. We had seen this species here on our first visit but had failed to capture one. Another prize was a species of Uracis hitherto 274. A YEAR OF COSTA RICAN NATURAL HISTORY entirely unknown to Mexico and Central America. The point on the right fork of the trail at which I stopped, and where I ate my lunch and in all spent a couple of hours, was a small stream which I suppose, from Pittier’s map, to be an affluent of the Rio Toro Amarillo, or simply Amarillo. There was an exquisite little dragonfly here with bright red eyes, brilliant coppery thorax and blue and black abdomen (Argia, near cupraurea). The much duller female of this species was laying eggs in the floating, more or less sub- merged vegetation, some of the insects themselves being submerged during the operation, as is not uncommon in the whole group to which they belong. . The stream-bed, as usual in all this country, was filled with stones and boulders so that it was possible to go some distance downstream from the trail by walking on these, although the banks were impenetrable owing to the vege- tation. I was so engaged when a great metallic-blue Morpho, perhaps the handsomest butterfly of all Central America, came by and fluttered around me so close that I could not use my net at once. I was astonished at its boldness for, notwithstanding the strikes which I made toward it with my net in the endeavor to capture it, it came back twice to where I stood on the rocks, each time after an interval of five minutes or so, circling around me in real mockery of my impotence. This piece of forest differed from that described below, through which the Florida road passed, in the much smaller number of palms to be found in it, although the difference in elevation is not more than one to two hundred feet. It may have been partially cleared once, but as many of the tall exogenous trees were untouched and surrounded by thick growth, it seems unlikely that these would have been left and the palms cut out if they ever had been here, since comparatively few species of these palms are of use to man. SANTA CLARA 275 On the afternoon of June 2, Mr. Barnes, taking his gun to try for a wild turkey, led me north from Salvador past the original and much plainer farmhouse and the machine shop to a pasture where the calves were kept. Crossing this we reached and crossed the Rio Gu4piles by a rough bridge. From this point the river was bordered by a narrow strip of woods and undergrowth separated by grassy pastures from banana fields lying farther east. Mr. Barnes left me here to examine the possibilities, but neither of us did much for it soon became very dark and fearing a general downpour we hastened back to the house. Saturday, June 5, was devoted to this strip of woods along the Rio Guapiles, as on the previous hurried visit I had no- ticed some Palemnema, a genus of dragonflies peculiar to Mexico, Central America and northern South America, of which I had never seen many specimens. I was not dis- appointed for this day too they were abundant in the low plants a foot or two high growing in the fairly dark shadows of the trees. They had the peculiarity of remaining closely in one place and not flying swiftly or vigorously, so that in most cases I could, with a little caution, pick them up with thumb and forefinger more readily than secure them with a net. They also had the peculiarity of keeping the body motionless in the same position and location in the air while the wings vibrated rapidly, thus suggesting a habit of the hover-flies or Syrphidz. Without much exertion and with- out extermination, I gathered fifty-four specimens of two species of the genus (paulina and nathalia) whereas for the Biologia Centrali-Americana the various museums had been able to furnish only seventeen specimens of five species and of these seven were imperfect while my catch of to-day was perfect. Another matter of interest is that these two spe- cies, which completely intermingled and did not, so far as I could see, possess different habits, were not the two most 276 A YEAR OF COSTA RICAN NATURAL HISTORY closely related species of the genus but (if I may so express it) species at least once removed in relationship. The ob- servation has some bearing on recent discusssions of the origin of species. Those writers who believe that geo- graphic, or topographic, isolation plays a great part in differ- entiating species point out that it is not the most closely allied species that occur together; on the contrary, the most closely allied are always separated from each other by some kind of barrier which prevents their intermingling or inter- breeding and consequent loss of distinctness. The larve of Palemnema were entirely unknown at this time (although we subsequently obtained them at Juan Vinas) and the abundance of the winged insects, a few of which had evidently only recently transformed, gave me hopes of finding the early stages. But although I looked long in the waters of the nearby river, under stones, in the floating and submerged vegetation, and in the sheathing bases of bromeliads and other plants, I was totally unable to find the objects of my search. There was a number of brilliantly-colored day-flying moths belonging to the family Syntomide, in this strip of woods. One which I secured, Cosmosoma teuthras, had the thorax bright red, the abdomen black with rows of metallic blue spots, the wings clear almost lacking scales except. around the edges, where they were reddish along the front and hind edges and dark brown along the outer edge. When the insect flew about, the red and blue markings were very conspicuous. Another species of the same family was of such interest to Mr. Schaus that I gave him the specimen. It likewise had transparent wings and its general appearance was wasp-like, due in large part to the way in which its abdomen was marked. Although this region of the body was broad at its front end as is usual in moths, the central part of the abdomen was white, the sides black so that look- SANTA CLARA 277 ing down upon the insect when at rest the black sides were unnoticed while the white central portion was highly conspic- uous. This naturally gave the appearance of the slender- waisted abdomen of a wasp, an illusion which was increased by the narrow transparent wings. I ate my lunch on a great rock in the midst of a smaller arm of the Rio Guapiles which flows past a little island. From this rock I could look into a quiet pool seemingly about four or five feet deep in which were fishes six to seven inches long. They reminded me of that fine fresh-water fish, the bobo, which was twice served for dinner at Salvador while I was there. One of the two, seven pounds in weight, was said by Mrs. Reed to be the largest she had met with in her housekeeping. Mr. Reed said that fishes in these rivers will not bite at hooks but are obtained by exploding a charge of dynamite in the river which stuns, and kills many fishes which are then picked up by men who wade out to gather them. It is a most wasteful method and should be prohibited, or at least restricted, by the Government, but nothing had yet been done. No one at table suggested a comparison of bobo with one of our fishes, but all agreed that it is superfine and there were no small bones to detract from the enjoyment of eating it. This fish has also been mentioned on page 264. On the third of June, arrangements having been made the day before, Mr. Schaus, Mr. Barnes and I went to the rail- road where a hand car was waiting, with a negro and a Costa Rican to work it. This took us westward along the tracks to San Jacinto, a farmhouse, near which a negro was waiting to guide us to the present starting point of the “Florida Road,” a trail leading into untouched forest. This was a favorite collecting ground of the two butterfly hunters but since their last visit here, two years ago, much of the forest had been cut down and the entrance to the trail consequently 278 A YEAR OF COSTA RICAN NATURAL HISTORY changed. The negro led us in a stiff walk for an hour, down a pasture by a little river, through a banana field, then some standing timber partly cleared, out into open sun again by a very winding path where the clearing had proceeded farther until finally the welcome shade of the forest itself was reached. The trail led westward here but was said to be part of a route to Nicaragua—in any event it traverses many a league of forest. We followed it for perhaps a mile and then returned as we had come. Its width varied and was often not greater than that of one’s body although the trail was often traveled on horse or mule-back. On each side of the trail the vegetation was so thick and dense that one could not make way for more than a few feet without the use of a machete such as all the natives and negroes carry. In this forest were mingled exogenous, deciduous trees of many kinds and a variety of palms. The former were rather tall on the whole but there were many exceptions. All the trees lacked branches until they were many feet above the ground, but this lower space was occupied by the epiphytes and parasitic plants, bromeliads, ferns, orchids, lianas and other vines which shot upward from the trunks or hung from the far-away branches and twisted and twined in both profusion and confusion. On the ground itself were shrubs and smaller plants reaching up to mingle with those descend- ing. There was much difference in the tree-trunks as to bareness or as to being covered with lichens. The large exogenous trees rose out of the soil with numerous ridge- like buttresses, while many of the palms had aerial roots. Most of the birds were far away overhead in the foliage of the trees proper, but sometimes an opening above, admitting sunshine, gave a glimpse of them or there was a swarm of butterflies around a full-flowered tree. There were so many new things to look at, both those of which I had read and those totally unknown, and so many SANTA CLARA 279 insects to be stalked, that I was soon the third to go along the trail, Mr. Barnes being in the lead and Mr. Schaus be- tween us. Once, a couple of peccaries rushed across the trail just in front of the latter; I was too far behind to see them but the scent was still very marked when I reached the spot where they crossed—it was like that of strong onions. Mr. Barnes saw some red monkeys feeding quietly in a tree-top, learning of their presence first by the debris they dropped. He told me of them when we met later and I went on to look for them but was unsuccessful. I did not go again to the Florida Road during my stay at Guapiles, but Messrs. Schaus and Barnes repeated their visit and brought me a few dragonflies, while in return I gave Mr. Schaus any of the Lepidoptera I picked up that were of interest to him. One of the dragonflies caught by Mr. Barnes on the Florida trail was of especial interest, as it belonged to the Cordulines, a subfamily of dragonflies never before found in Central America or in Mexico except once near Monterey in northern Mexico. It has in conse- quence been supposed that the absence of this group was one of the faunal characteristics of this region, but Mr. Barnes’ capture proved this idea to be erroneous. His specimen has been described as Neocordulia longipollex; its congeners are known at present only from Ecuador and Brazil. The unexpected taking of this Corduline in the Florida Road indicates how much collecting and study are still necessary for a thorough knowledge of the neotropical region, and how rash it is to generalize from collections made during one visit only or at one season only. Mr. Schaus had collected Lepidoptera in various parts of tropical America for over twenty years, of which three were spent in Costa Rica. As a result of his long experi- ence he preferred to collect in the forest, during the day- time at least, since it was his opinion that specimens to be 280 A YEAR OF COSTA RICAN NATURAL HISTORY found in open country such as banana fields, pastures, grassy fields, etc., were of species having a wide distribution over all or a large part of tropical America, while those living in the forests were more often peculiar to a limited region. Beside collecting with a net in the usual way in daytime, he did much around electric lights at night. At Salvador Mr. Reed had an arc lamp arranged for him under the veranda of a whitewashed building and here he spent several hours each night. One evening two enormous beetles, a male and a female of the same species (probably Megasoma elephas, but the specimens are not accessible) came to the light and were caught. Each must have measured five to six inches long, two to three inches wide and correspondingly thick so that they were quite bulky and heavy. The male is some- what larger and has a slender projection forked at its tip, on its head; this projection alone is nearly an inch long. Somewhat to his surprise, I think, Mr. Schaus did not find the night collecting at the electric light as prolific as at Juan Vinas (at 3500 feet) and was disposed to think it due to the rather unusual dryness of Guapiles at this particular time. In fact his collecting experiences inclined him to the opinion that in Costa Rica there was no seasonal distribu- tion of insects on the Atlantic slope, but that the presence or absence of adult insects depends in any locality on the wet- ness or dryness; this is a condition which may be quite differ- ent in the same place in the same month of two consecu- tive years. On the night of June 3 we saw clearly a total eclipse of the nearly full moon; it began about 6 P. M. on the lower right hand edge of the moon and gradually extended to the upper edge so that for a time the entire disk was covered.’ The Florida Road was visited a second time on Novem- ber 18, 1909, when A. and P. started thither on horseback from La Emilia. For two miles we rode westward on the SANTA CLARA 281 railroad, sometimes between the tracks, sometimes along a crooked narrow trail one side of the rails. At San Jacinto we turned into a field containing a sawmill and as there was an extremely heavy cloud hanging over the field and the forest, we waited on the porch of this house until this par- ticular shower should have passed, which it soon did. Then we rode on around great numbers of logs, through deep slippery mud full of stones and stumps, and after fording a small milky stream traveled along its bank until we neared a large banana plantation. Our way led through this, but seeing a well-traveled road running directly into the woods we turned aside to the left and followed it. From its general direction we thought it must finally join the Florida Road, and probably it did at one time, but after we had ridden about a quarter of a mile—over a rough, “‘rooty,”’ muddy, slippery, loggy path, including a gate to be opened—we came to a great heap of brush which effectually blocked the road. This part of the forest was pretty enough with its huge trees, many epiphytes, and numerous palms, and we saw a Megalo- prepus, but we wanted to visit the real Florida Road, so there was nothing to do but ride back to the fork and go through the bananas. And here was just where our troubles began, for there were many paths through the bananas with branches in every direction and all looked alike. We had been offered a guide, but as P. had been there before he thought he could surely follow the same path. He was not prepared for the way the trails grow up and change their locations and as luck would have it we followed every trail but the right one! Fortunately the day was cloudy with some showers so that it was not so hot as usual in the ba- nanas and we took a few insects, although there was nothing remarkable. About 11.30 we started. back to the forest we had first entered, having convinced ourselves that we could not find the main Florida Road, and retraced our way as far 282 A YEAR OF COSTA RICAN NATURAL HISTORY as the gate in the woods, where we left the horses. Here we ate our lunch, stopping to catch a Megaloprepus that obligingly alighted near us. Then we rambled about, fol- lowing little paths where we saw tapir tracks, enjoying the many new plants but seeing very few dragonflies. One striking plant was Uragoga tomentosa, a Rubiaceous species peculiar to Costa Rica. This was a bush with opposite, reddish leaves, the heads of small white tubular flowers subtended by large fleshy bracts of a flaming red. In fruit it was still more conspicuous, for the bracts persisted and the berries were three-quarters of an inch long and bright purple in color. Another was an epiphytic vine (Columnea sp.), which sent out stiff short branches from each side of its main stem, the latter closely following the host trunk. The leaves of this plant were three to four inches long; from above they looked bright green with a small white patch not more than one-quarter inch across, from below they showed symmetrically-placed blotches of clear crimson. The tubular flowers were borne on slender peduncles two to three inches long, depending from the stiff little branches. The flower itself was a beautiful red, with a five-parted calyx of delicate, tender green set with soft whitish prickles, and in the forest of dark greens the tree bearing this vine was an exquisite sight. All this time the clouds were growing more threatening and by one o’clock rain began; as it increased steadily and rap- idly we hurried to the horses and started back. The road was even worse than it had been on the way out and P. was soon wet through; my poncho protected me quite well from the water descending as rain—but not from the water ascending as splashes, for in addition to the general wetness of the road we forded four rivers, two of which were deep enough to come over my boot tops unless I held my feet up out of the stirrups. Our horses were excellent, however, sure- SANTA CLARA 283 footed as goats, and we had no trouble in getting over or through these bad places. La Emilia, the starting and re- turning point of this second expedition to the Florida Road, was a cattle farm one mile east of Guapiles belonging to the United Fruit Company, where we were the guests of Mr. R. E. Woodbridge, the manager. We had left Cartago on November 15, on the “passenger”’ train for La Junta, where we waited for the Old Line train. The fine new bridge over the Reventazon at La Junta was completed but there was still no station there and we spent the time (nearly an hour) sitting on a bench on the little porch of a negro commissary, ‘“‘restaurante y cantina” that stood on one side of the track near the switch. The heat was not at all noticeable as the day was cloudy and it was fairly comfortable there. We watched the Reventazén tearing along, a big swift stream, much nearer to the shop and cabins than would be agree- able to us if we had to live there, and the cocoanut palms and breadfruit trees made the place somewhat picturesque, otherwise it was simply a collection of negro huts, railway tracks, bananas and plantains. We left La Junta about three o’clock, having planned the trip so that we should not go up the Old Line on a freight day. We therefore did not stop at every possible station and the delays were not long, but the train was exceedingly slow and jerky and it was six o’clock when we reached La Emilia. The “‘stations” along this line were simply sheds built over the path leading up from the track to the company’s farmhouse of that particular district, and clustered around the “station” were the cabins of the negroes who worked on the place. There were no roads except the railroad and the footpath along it and no villages except these clusters of cabins. This region was all forest until the Fruit Com- pany cleared it for bananas or pasture land and it had been 284. A YEAR OF COSTA RICAN NATURAL HISTORY unoccupied land for many years, probably centuries. That it was once occupied, however, perhaps thickly settled, is shown by the great number of Indian graves which con- tain quaint gold ornaments of forgotten workmanship and pottery of shapes and glazing far beyond the skill of the present native potter. The pottery was not firebaked and when first taken from the graves was as soft as mud, but if carefully removed and exposed to the sun it became as hard as rock although brittle. The gold ornaments were usually in the form of animals or birds, often quite accurately worked, and were found only in certain kinds of graves, supposed from the better workmanship and finer quality of pottery and the golden ornaments to be those of chiefs or important men. The house at La Emilia was a two-story wooden building with a long ell and balconies nearly all around on both floors. In front of the house, between it and the tracks, was a formal garden with beds in geometrical patterns and the path edged with beautiful pink and white ‘‘mariposas”’ or peri- winkles. The garden was rather new as yet but gave prom- ise of being very pretty. Halfway down the walk wasa pair of large “gavilans” (Pentaclethra filamentosa), leguminous trees with much divided leaves, but having the leaflets so closely arranged that they look quite solid and undivided. They were stiff and glossy and did not present at all the fern-like appearance of the guanacaste, for example, which has its leaves divided to the same degree. The pods were large, twelve to fifteen inches long, and borne stiffly at the outer surfaces of the crown of the tree. Mrs. Woodbridge was extremely fond of flowers. All around the balconies were numerous potted plants and hang- ing baskets of orchids, ferns, caladiums and begonias. Many of the orchids were now blooming beautifully and all the plants were luxurious and thriving. One of the doorways SANTA CLARA 285 (all the rooms had doors and windows on the balconies) was draped in green, for an asparagus (sprengeri or near that species) growing in a small tub, had been trained up one side of the doorway, across the top and down the other side. On one side of the porch there was a wire network over which was trained a vine foreign to Costa Rica and highly esteemed here—white and yellow honeysuckle. On an- other side was a screen of Clerodendron with red and white flowers. The orchid tubers were favorite places for the nests of a little black bee (Trigona cupira) which has the habit of going always into the hair or beard to sting, never attacking the bare face or hands. Fortunately its sting is not severe, but it was necessary to look out for these bees when going about among the baskets on the balcony. After coffee on the sixteenth we walked back (south) of the house into a potrero bordering a good-sized stream, the Rio Santa Clara, Mr. Woodbridge riding with us a little way to start us in the right direction. The potrero was pretty, though not so beautiful as the one we visited at Turrialba, and had many grand forest trees standing in it, left purposely for shade. Some of these trees bore an in- credible weight of epiphytes, others having a smoother bark or the habit of peeling off the bark (like our northern sycamores) were perfectly clean. We spent parts of sev- eral days in this potrero, at times down among the rocks of the river, at times in the pasture itself where there were some small streams and pools but without finding anything very valuable. There were few dragonflies flying and those seen were not particularly rare. We noticed a large fallen tree, with glossy compound leaves, which bore curious fruits—small woody pods split in two, of which one side fell away leaving the other attached to the stem. The contents, one seed in a mass of pure white pulp, fell out and hung by some threads to the lower end of the pod and as the pod it- 286 A YEAR OF COSTA RICAN NATURAL HISTORY self was colored a bright red inside and the fruits hung in large clusters, it was very conspicuous. The white pulp was sweet, with a curious flavor suggesting both turpentine and peppermint, and the leaf-cutting ants were busily cut- ting it up and carrying it away to their nests. Nearby was a fair-sized tree, which Sr. Tonduz later identified, from some pods we gathered, as a “‘Lorito” (Pithecolobium sp.). The fruit was a twisted legume four to five inches long with thick, fleshy, bright scarlet walls and containing a few fully uncovered shiny black seeds each half an inch long. The red and black of the fruit made a marked contrast with the shining green leaves. On the edge of the potrero and also in the banana fields we found a large purplish lubber grass- hopper (Teniopoda centurio) with bright red under-wings. We often hunted for dragonfly larve between the bases of bromeliad leaves. There was a great variety of creatures living there and we found earthworms (Dichogaster hilaris*, Andiodrilus biolleyi), pill-bugs (Isopods) daddy-longlegs (Phalangids—Cynorta biguttata* and longispina), pseudo- scorpions (Chelanops uniformis*), millipedes (Aphelidesmus calverti), earwigs (Sparatta calverti), roaches (Anaplecta mexi- cana), several species of beetle larve, as well as an adult beetle (Phenonotum sp.), caterpillars, larve of three species of mosquitoes and of one other dipterous insect. Those whose names are marked with an asterisk we found also in bromeliads at Juan Vifas at an altitude at least 1500 feet higher. But the most astonishing creatures, and the most exciting, were two snakes, coiled up into a very small space, each in his own leaf. The bromeliad containing them was a single stock growing on a small tree at a height of about five feet from the ground. The first snake we dislodged with a stick, disabled it and carrying it as far as the house measured its length as twenty-five inches. The Costa Rican gar- SANTA CLARA 287 dener called it a “tamaga” and told us it was very deadly, as poisonous as the coral snake. If this were so, it seemed to us at the time, that this snake must chew and not strike. The head and fore part of the body were brought back to the United States and identified as Leptodira albofusca, widely distributed throughout tropical America. It has a pair of enlarged grooved teeth behind the other maxillary teeth, and the groove, on the front of each tooth, “‘con- ducts the secretion of the enlarged upper labial glands.” Leptodira is a member of the subfamily Dipsadomorphine, one of the divisions of the Opisthoglyph Colubride. Of the Opisthoglypha, Dr. Gadow says (in the Cambridge Natural History): “Apparently all these snakes are more or less poisonous, paralyzing their prey before or during the act of deglutition. So far as man is concerned they are rather harmless, since the poison is not very strong, not available in large quantities, and above all because the small poison-teeth stand so far back that the snake cannot easily inflict wounds with them. The Opisthoglypha are of considerable morphological interest, since they connect the Colubridz with the Viperide, the characteristic poison- ous apparatus of which seems to have been derived from that of the Opisthoglypha by the reduction or shortening of the anterior portion of the maxillaries and the harmless teeth, so that the posterior or poison-fangs come to the front.” The second of our bromeliad snakes escaped unhurt. Both were very sluggish with eyes filmed over and, we sus- pected, were about to moult. On November 18 we investigated some bromeliads on two trees that had just been cut down and were lying on the ground, practically untouched, in the potrero west of the stable of La Emilia. A peon cut off eight or ten of these epiphytes for us with his machete, when I examined them, 288 A YEAR OF COSTA RICAN NATURAL HISTORY but although they still contained water and various in- sects, including the beetles Alegoria salle1 and Metamasius ochreofasciatus, there were no dragonfly larve. In all I must have spent two hours on a dozen plants which had been fifteen to twenty feet above the ground. The next day we pulled down a fair-sized bromeliad, perched ten feet up on a tree, by means of a convenient, nearby liana; still no Odonate larve. Thus although fifteen bromeliads at least were examined during our stay at La Emilia, all containing insects, some or all of which were associated in the same plant with dragonfly larve at Juan Vinas, not a single one of the last named was found. However, none of these bromeliads were in actual forest, but were on trees forming a mere fringe along the Rio Santa Clara, or were in open potreros, and the situations may account for the ab- sence of these particular insects. We had very few adult dragonflies to show for our stay at La Emilia, owing to the little sunshine and much cloudiness and rain. Our object was, indeed, to observe these insects under wet season conditions, but usually the mornings are bright even in this period, whereas we encountered a “tem- poral” or steady rain of several days’ duration. The slightness of our results was doubtless due to the absence of sunshine. When P. left Guapiles on June 6, he took the train for Guacimo, about eight miles to the east on the way back to La Junta, Mr. Reed having telephoned the night before to Mr. Stahle, the Mandador or manager, that I would come. Mr. Stahle (originally from Stuttgart) met me near the station and took me to the house, much less pretentious and comfortable than Salvador and tenanted by men only with the single exception of a Costa Rican(?) woman cook. I was given a room at the east end of the second floor having two windows and three doors, and stayed here two days. SANTA CLARA 289 As usual the farmhouse had verandas on both floors run- ning completely around the house. From the veranda on which my room opened, one looked down almost directly on the Guacimo River, the constant roar of its waters usu- ally making me think it was raining. Following the river upstream (south) the eye passed over the forest to the im- posing crest of Turrialba; the other three volcanoes could not be as well seen here as from Salvador. Mr. Stahle and Mr. Blair (Assistant Manager, a Harvard graduate of 1908) were about to go into the bananas on horseback and offered to show me a trail into the forest. They offered me a horse also but I declined it, knowing that in all probability I would frequently want to go where a horse could not or where it would be in the way. At all the banana farms there run off from the main railroad tracks into the plantations, spurs which form right angles to the main tracks and are parallel to each other. To-day we followed the main track a short distance west from Guacimo farmhouse and then north along Spur 1. Workmen (negroes chiefly) were busy—Sunday seemed to be no different from other days here—cutting bunches of bananas, loading them on mules, bringing the loaded animals to certain points on the spur where the fruit was piled up on banana leaves laid on the ground and finally covered with the same when the pile was complete. The bananas are cut while still green at stages known as ‘‘three-quarters” or “ three-quarters full”? and much depends on the judgment of the cutter in selecting fruit just at these stages. If cut ata younger stage the fruit will not ripen when it reaches the United States or Europe, if cut a little riper it will spoil before it reaches its destination. Here and there along the spur were bunches with some yellow and consequently ripe bananas. These were rejected and left to rot. I ate some of these ripe bananas and could not find that they differed 290 A YEAR OF COSTA RICAN NATURAL HISTORY much in taste from those to which I was accustomed at home. Spur 1 was, I believe, about three miles long, and the cutting done to-day was chiefly in the farther half and yielded some 2000 bunches or “‘stems” as they were called here. Stems bearing nine or more “hands” or tiers were reckoned as “firsts,” those with a smaller number as “‘seconds.” In some places ‘‘thirds” were also distinguished as stems with six or fewer hands. Each hand is composed of twelve or more “fingers” that is, individual bananas. It was no uncommon thing for a stem to comprise 150 bananas so that 2000 stems would mean 300,000 bananas for this week’s cutting. In about a week later, depending somewhat on moisture and heat, the same ground can be cut over again—that is, other shoots of banana plant will have come into the cutting stage and this process is continued throughout the entire year for seven or eight years or even longer in some spots where the surface soil is deeper. This constant productivity of the bananas is brought about by the care taken to plant continually new young shoots to replace those which have borne, for each shoot yields but one stem, after which it is cut down and, so far as my observation went, allowed to decay on the ground. According to Mr. Blair (who supplied much of the preceding information) the productive soil at Guacimo farm is only two to three feet deep and is underlaid by red clay and loose rounded stones. When this surface soil is exhausted the bananas no longer thrive. In such cases it was the policy of the United Fruit Company to let the land lie fallow since it was cheaper to clear off forest and make new plantings. The abandoned banana fields soon become filled up with a dense growth including saplings. Mr. Blair’s experience was not long enough for him to say certainly whether the forest eventually reclaimed its own, although he inclined to think that the reclamation did come albeit slowly. SANTA CLARA ° 291 The tracks of Spur 1 were supported on iron ties ballasted with small rounded stones and for much of the way through the bananas ran on a low embankment, intersected fre- quently by “culverts” or short wooden trestles crossing streams of swift-flowing water. Horsemen traveled along the embankment until they came to a culvert, where a side path led down from the tracks to the stream and ford and up to the tracks again. Progress on horseback was only at a walk and not much faster than that of the pedestrian who could follow the tracks uninterruptedly across the culverts. As I walked along Spur 1 every now and then the brilliant green-and-black-barred day-flying moth (Uranidia fulgens), rose from the ground as I approached and fluttered on ahead or into the bananas. A great heavy lumbering grasshopper (Teniopoda centurio) of Indian red, mottled with darker colors and with short bright red wings edged with black, sometimes crawled off into the grass; its weight seemed too great for it to execute more than very short flights. At one culvert, where the water was easy to reach and rippled over small stones, were some dragonflies whose flight up and down short reaches of the stream was so swift that I could not recognize them. I stationed myself to watch and to catch them. When several were together they flew so fast that recognition and capture alike were impossible even at a foot’s distance. When one was alone, however, it flew more slowly. I found that by holding the net a few inches above the water, waiting until the insect was underneath and then suddenly submerging it, the current carried the dragonfly into the net. I thus secured in all six specimens of the insect (Brechmorhoga precox) but it took about half an hour in the full heat of the sun. There was localization in distribution too in the banana field dragonflies. Near the end of the spur, four or five yards from the tracks, was a narrow ditch containing a little 292 A YEAR OF COSTA RICAN NATURAL HISTORY slowly-flowing water. The six or eight species found here were all different from those on the larger swift-flowing streams. The road which had been cut into the forest for a distance of 2300 meters in direct continuation of Spur 1 had not been entirely cleared but here and there was filled with grass and shrubs, at other places was open with rain-water puddles on its clay surface. The forest immediately bordering this road had few palms but these increased in number to the eastward. Crossing a pasture with cattle, horses and mules, directly opposite the Guacimo farmhouse, brought me into another part of the bananas. Following a very narrow gauge tram track and then a log-hauling trail, I reached a piece of forest where palms were largely in the ascendant although exoge- nous trees were not lacking. The trail was in places com- pletely overarched and was then dark and muddy. In one of these dark spots, amidst a tangle of lianas where the net was not usable, I caught in my fingers a great dragonfly (Gynacantha membranalis) never before known north of Chiriqui. It was three and one-quarter inches long, with a wing-spread of four and a half inches; it had brilliant metallic green eyes and a reddish-brown patch on the bases of the hind wings. At another lighter place there was on the ground sis at first looked like a long, slender, wriggling, whitish worm, but a second, nearer glance showed at once that it was a procession of termites (“‘white ants” or “wood ants,” which of course are not ants at all). The procession issued from a hole at the surface of the ground on one side of the trail, the objective point being a tunnel running up a small tree to a nest on the opposite side. The trail was about two feet wide but there was water and wet mud in the middle so that to cross on dry land the route lay down one side of the trail SANTA CLARA 293 to a dry spot, then across and up the opposite side to the lower end of a tunnel almost directly opposite to the hole from which the insects issued. This meant a journey of three or four yards to accomplish two feet of air-line. The termites walked singly, or by twos or threes, in a continuous line, those behind practically touching those in front. Every little deviation from a straight line made by the insects in front was exactly followed by those behind, these deviations being often not explicable by wetness or roughness of the soil. The procession consisted of two kinds of individuals, the larger workers and the smaller, dark-headed ‘“‘nasutes,”’ so called because their heads are prolonged forward into a rather sharp process suggesting a snout. The nasutes were fewer than the workers. The tunnel to which the column proceeded was one of the usual kind—a flattened arched way running up the tree-trunk to the nest and made of wood which had been chewed by the termites. After watching the procession for a few minutes, I took a little clump of wet mud and put it on the insects’ path. Those approaching this obstacle were instantly confused on reaching it, for although winged termites have eyes these workers and nasutes are eyeless; they have antennz, how- ever. Some turned back and went for several feet along the way they had come, followed by an increasing number, even after a way around the obstacle had been found. I think that those which turned back were exclusively workers. In the meantime some workers and the nasutes were explor- ing to each side of the lump of wet mud, which was not more than an inch in length, some going to one side, some to the other. The mud thus produced a break in the procession, and some nasutes that had already passed the place where I put the mud came back as if to learn what was wrong. Finally it was found that it was possible to pass around one end of the mud and, this done, to find the original trail 294. A YEAR OF COSTA RICAN NATURAL HISTORY and in a short time the procession was again continuous, every insect in the line making a slight detour around the wet mud. Then taking a little stick I drew it across the line of march so as to form a tiny trench about an inch long, an eighth of an inch wide and an eighth to a quarter inch deep. Again were the termites troubled when they reached its edge, and some time elapsed before a detour was made around it and the march resumed. A few also made their way down into and up out of the trench in the same line as they had origi- nally marched. In both interruptions it seemed to be the nasutes which led in finding the way, although a worker was usually not far behind. I broke a little of the arched tunnel running up the tree- trunk; within were ascending termites. When I came back to this spot, in less than half an hour, the procession was over, only a few straggling individuals being visible. The last five days of walking over wet, rough, uneven ground had so chafed my heels that they were now very sore. But I could not bear the thought of losing another oppor- tunity of visiting the forest as the weather was good and I needed to return to Cartago on the morrow. So on the sev- enth, declining another offer of a horse for the same reason as before, and taking lunch with me, I set out for the forest road at the end of Spur 1 which I had already visited. Walk- ing being slow and painful, it took me the greater part of two hours to cover the three miles, including, however, a half hour’s stop in a broiling hot sun at the ditch near the spur’s end. A train of a few freight cars had been run up the spur this morning and was stopping at the piles of bananas, which were loaded into the cars by negroes. Banana cutting was also proceeding on the nearer half of Spur 1 and 1600 stems were expected for the day. From the farms the bananas SANTA CLARA 295 were carried by rail to Lim6én where they were transferred to steamers of the United Fruit Company or Hamburg- American Line, fitted up with cooling appliances which keep the fruit at about 60° Fahrenheit during the voyage. Several species of dragonflies which I had not previously taken at Guacimo or Guapiles rewarded me on this walk and made me feel justified in the day’s program. Another in- teresting insect was a long-horned beetle (Ptychodes lecontet) 27 mm. long, whose black and yellow striped body and legs spread out in various directions caused me to think it at first glance, a spider, like the large black and yellow Argiope we have at home. I left the forest about 1.30 P. M. for the long hot walk back, although this day I had brought an umbrella as protec- tion from the sun, which blazed fiercely down on the track. A number of bright and interesting butterflies enlivened the way, however. There were swarms consisting of twenty to forty individuals of three species, with bright orange (Callidryas argante), yellow (Callidryas trite), or white (Pieris ilaire) wings, all of nearly the same size, the wing- spread being two and three-eighths inches. These settled near together between the tracks. As I approached they rose gradually and for a few steps I would be in the very center of the flying insects, realizing exactly Andrew Lang’s simile of “‘living blossoms of flying flowers.” ‘Then they flew off in advance, in what from behind looked like irregular confusion, but which I think was really a single file whose members made a series of undulations but nevertheless strictly ‘‘followed their leader.” Then by and by the line reassembled in a close mass on the ground until my approach again dispersed them. The spots at which they settled did not seem to be determined by presence of moisture as is often the case. Where some rotting bananas lay on the ground there was 296 A YEAR OF COSTA RICAN NATURAL HISTORY frequently to be seen a moderate-sized butterfly (Victorina stelenes) whose wings, spreading three inches, were pale green marked with brown; the brown marks were so arranged as to cause the paler color to form two rows of spots on the front wings, a band and a row of spots on the hind wings, the inner row or band wider than the outer row in both cases. I came to think of it as preéminently a banana butterfly, but its swiftness and wariness prevented my taking it, at least in my physical condition at that time. Subsequently, however, I caught it at Turritcares and Alajuela where there were few bananas. Once when resting near a pile of bananas to see if the butterfly would return, I heard a loud clicking noise that I at once supposed to proceed from a grasshopper or katydid and on looking around to find the musician was astonished to see two butterflies in the spot whence the sound came. Once again at another place I heard the same sound from a pair of the same kind. They belonged to the genus Ageronia and the wings were spotted pale blue and white on the upper side. I saw many individuals during the two days I was at Guacimo but never heard the sound except when two were together, so it may be a sex call. So few Lepidoptera make sounds that the case is noteworthy. One observer’s idea, quoted in the Cambridge Natural History, was that the sound attracts a certain bird; I did not know of this when I heard the butterflies, but I saw no bird pursuing the in- sects. Several times in the banana fields I saw handsome black and reddish-brown butterflies of powerful and swift flight, which were, I think, Aganisthos orion. I had come to consider it as very wary so that I was astonished this afternoon to have one, on two distinct occasions, fly round and round me close to my person and at length, on the second occasion, alight on my leg as I was standing up. The only explanation SANTA CLARA 297 I could think of was that the insect was attracted by the perspiration, as has been noticed in some other species. My insect net these two days was often much hampered by brushing against seeds (which we think of as autumnal) which then adhered to it and tangled it up. These were beggars’ ticks very like our own Desmodium, small burrs a quarter of an inch long and seeds resembling “‘Spanish dag- gers” but narrower and having three prongs at the end instead of two. By way of summary of the Guapiles-Guacimo trip, I found on returning to Cartago that in the six days of June I caught or definitely identified specimens of forty-four species of dragonflies, not one of which had I met around Cartago; that of the forty-four, three had not been recorded from Central America or Mexico and may possibly be en- tirely new, one of the three representing a subfamily never before met with in these countries; fourteen had not been previously recorded for Costa Rica, and thirteen, although known from the Pacific Slope of this country had not pre- viously been recorded from the Atlantic side, so that thirty of the forty-four were additions to knowledge of geographi- cal distribution. CHAPTER XVI THE BANANA RIVER COUNTRY About six miles south of Limon, the Banana River empties into the Caribbean Sea. The United Fruit Company has a number of banana farms in its watershed and a railroad runs to the river, ascends its left bank, crosses and goes on five miles beyond to the Rio Bananito, which also discharges into the Caribbean. On the morning of November 4, 1909, P. left Limén by the Banana River train, which ran south along the beach above high tide, among cocoanut palms and here and there patches of swamp palms. At Westfalia Commissary Sta- tion, about six miles from Limén, we left the beach and soon after passed a plantation of cacao, with trees only ten to fifteen feet high; they were planted in rows and bore fruit on their trunks which was now green tinted with pink or rose. For the rest of the way we were among bananas. At Bearesem East Farm the track was very close to the Banana River. Sometimes going ahead, sometimes backing, we came to, stopped at and went on from Beverly Farm, and Beverly Commissary, which were on the level, and Beare- sem West Farm to Philadelphia South Farm. The farm- houses on these last two farms were each set on a hill some hundred feet above the tracks. At Philadelphia South I alighted, having a letter from Mr. Hitchcock to Mr. C. J. Veitch, Superintendent of the Banana River Division. Mr. Schaus had suggested Bearesem West Farm as being prob- ably the best for my purposes and Mr. Hitchcock had named this place in the letter of introduction. Mr. Veitch received THE BANANA RIVER COUNTRY 299 “PORTETE ‘P LIMON +t++ Railroads © KM. I 2 3 4 s Oo MILES 2 3 The Banana River country. (From a map of the United Fruit Co.) me cordially and said he would send me to Bearesem West if I wished but thought I would find more comfortable quar- ters with him at Philadelphia South. So I accepted his offer for a day or two at least and having been given a pleas- ant airy room changed to my collecting outfit and went into the bananas at the foot of the hill, collecting along a little brook until breakfast. The elevation was perhaps fifty feet; a refreshing breeze was blowing, but the sun shone brightly and when not shaded by the bananas it was de- cidedly hot. 300 A YEAR OF COSTA RICAN NATURAL HISTORY The location of the Superintendent’s house on the hill made it much more pleasant as regards temperature, as it usually received more wind. The two-storied house, with verandas on all sides, was connected with an office over which were two bedrooms and one of these I occupied. The house presented a very attractive appearance, due to the hanging baskets of ferns and other plants with which Mr. and Mrs. Veitch had adorned it. In front were a few euca- lyptus, cocoanut and orange trees. A garden had been started and on one side were breadfruit trees. A fallen leaf from one of these last was twenty-nine and a half inches long. It was deeply divided into five pointed lobes on each side of the midrib and glossy green. Mr. and Mrs. Veitch were English and had lived in South Africa before coming to Costa Rica. Before he became Superintendent here, they were at Chirripo Farm when the floods of December, 1908, came down the Chirripo River at midnight carrying away seven houses just above theirs and the iron railroad bridge and flowing a foot deep across the office in their first floor. Expecting their house to go next, Mrs. Veitch, her daughter and two negro women servants were put into a dugout at 2 A. M. in pitchy dark- ness amid the roaring of the flood and the boat was then dragged by a rope, while Mr. Veitch waded behind guiding it, to the foreman’s house which was a little higher up. The next day, the floods having subsided a little, they returned to their own house, but other freshets came and before the floods were over they left their house seven times on account of threatening danger. The house here at Philadelphia South was situated on a hill of reddish gravelly clay (ap- parently without pebbles), and seemed entirely free from the possibility of danger from floods. Back of the house were many pineapple plants for the family’s use, the stable yards for the horses, mules and cows, THE BANANA RIVER COUNTRY 301 a pasture, and then the forest, barely a quarter-mile from the house. There was a trail into it whose beginning was marked by a cluster of fine trees called, I believe, ‘‘pavon.” At some distance east along this trail was a clearing with a few scattered bananas and pineapple plants. Among the dry withered banana leaves lying on the ground I had a glimpse of a great brown cockroach, seemingly move than three inches long, but although I searched for him, turning over the leaves, he managed to escape. I did secure, how- ever, a magnificent metallic green wasp, Ampulex neotropica, an inch long, its wings clouded with brown. Under a lens the surface of the entire body is finely pitted, while on head and thorax are larger pits among the small ones and many short whitish or grayish hairs. The head is prolonged for- wards into a triangle and the antennz situated on the pro- longation in front of the level of the large eyes. The large black mandibles are attached to the head, one below each eye, and curved outward and forward in such a way as to leave an open space between ed4ch mandible and the re- mainder of the head, instead of the jaw when closed lying in contact with the head, as is the usual condition in insects. This arrangement of the mandibles suggests the zygomatic arch of the skull of higher mammals, while it must give ‘Ampulex a very wide gape. The maxille and labium are attached to the head at a level in front of the attachment of the mandibles. As the Ampulicide are said to feed on cockroaches the finding of this individual immediately after seeing the large roach may not have been a mere coinci- dence. Then I passed through more forest and finally into a creek, walking down whose stony bed brought me into the bananas of the farm again. Just before reaching the ba- nanas the creek ran through many Heliconias, some of which were the largest I saw, each leaf and leafstalk being about 302 A YEAR OF COSTA RICAN NATURAL HISTORY twenty feet long while the pendent red flower cluster was fully ten feet long. The Heliconias have no trunk, all the leaves springing from the surface of the ground. Else- where pasture instead of bananas bordered the same piece of forest, and in a higher part of the pasture a bank of the red clay was exposed. Many large dragonflies were flying about here, and looking more closely I found that they frequently alighted on the grass stems at the edge of this clay bank, assuming a vertical position and often closing the wings. Even so I caught them with difficulty and found them to be Pantala hymenea, mentioned in Chapter V. On November 4 Mr. Veitch gave me a “‘trolley” to take me to the end of a line of railroad track known as “ Philadel- phia B Line.” A “trolley” is a two-seated hand car manned by two men on the front seat, the passenger sitting on the rear seat. In such luxury I reached the Cieniguita River, which on arriving near Limén becomes the Limon River. I crossed it on a narrow footbridge at a place where, with the dry conditions that had prevailed here for weeks past, the river was not more than ten to fifteen feet wide. The oppo- site (left) bank in Santa Rosa Farm was planted in bananas. The banks of the river were in many places almost perpen- dicular and ten to fifteen feet high, but here and there it was possible to get down to the water’s surface. All the banks were fairly well covered with bushes or high grass and there were many logs and fallen branches. Once while stalking some insects I heard a loud splash, as if some large body had fallen into the water, a few rods farther upstream. Looking in that direction I saw a large iguana crawl out on a slanting log and rest a few seconds at full length upon it so that I had a distinct profile view showing its large neckfold and its serrated dorsal edge. The reptile was fully five feet long. Just then something else attracted my attention for an in- stant and when I looked again the iguana was gone. THE BANANA RIVER COUNTRY 303 A pair of the handsome dragonfly Heterina titia were en- gaged in laying eggs, the male holding the prothorax of the female by his caudal appendages and the two flying along together. They alighted on a bare sloping branch which projected some eight inches above the water’s surface. On the downstream side of this they backed toward the water, the female becoming completely submerged, but the male on reaching the water let go his hold of the female and remained in the air, making short flights from time to time but re- turning to the top of the stick. The female meanwhile backed farther and farther down and was much of the time out of sight, owing to the muddiness of the water. Occa- sionally I could get a glint of light reflected from the white envelope of air around her body and wings. She was in- serting her eggs into the stick and remained completely submerged from 2.35 to 2.47 P. M., then climbed to the water’s surface—to be met by my net and transferred to my collecting box. The male of Heterina titia has much or all of the hind wings dark brown, the front wings bright red at the base, the red bordered externally with dark brown but leaving the outer half of the wing clear except at the extreme tip. The females have the wings yellowish or, when older, dark brown, but without any red at any time. The body of this species is one and seven-eighths inches long and the spread of wings two and one-quarter inches. Heterina titia and a companion species on the Cienguita this day, Neoneura amelia, had not been recorded as occurring in Costa Rica. Neoneura amelia is a much smaller insect, less than one and one-quarter inches in body-length, with clear wings which spread one and two-fifths inches; the body of the male is black and orange-red, of the female black and greenish- yellow; both sexes are very slender. As usual in the hotter parts of Costa Rica there were 304, A YEAR OF COSTA RICAN NATURAL HISTORY many termite nests and tunnels among and on the trees. Many of the inhabited nests were but three feet or so above the ground. I broke a piece of a tunnel and of a nest; in both cases the damage was at once inspected by nasutes, ten or twelve appearing in the latter case, but repair did not begin in either case as long as I cared to spend in watching. Under a very tall tree by the river bank were numerous hard fruits three to four inches long, two to three wide and one to one and one-half thick, flattened or concave on one side, convex on the other, with low ridges radiating from a point near the blunter end and with minute pits be- tween the ridges. The following day Mr. Sullivan told me he thought the tree producing them was called “cacheibo.” The fruits were subsequently identified by Professor Pit- tier as Prioria copaifera, the Costa Rican “camibar.” He adds that from it is extracted the resin known in commerce under the name of copaiba balsam, which has various ap- plications in medicine and in pharmacy. By trolley again on the morning of the fifth I went to the “Philadelphia E Line” of track, lying in Holanda Farm (the Spanish name for Holland, referring to the low swampy character of much of the land within it). Here by arrange- ment of Mr. Veitch, Mr. Sullivan, the mandador, met me to show me the path into the forest lying a short distance north of the “E Line.” Mr. Sullivan was formerly employed in the Botanical Gardens in Jamaica and had some acquaint- ance with plants in consequence. We went through the bananas and then into the forest along Agua Buena Creek, as it is marked on the Fruit Company’s maps of these farms; but the water although moderately clear did not look to be particularly good as there were many shreds of apparently organic matter in it. Its bottom is a bluish clay. The place looked promising for dragonflies but the sky was dull—had the sun shone brightly I think I should have secured some THE BANANA RIVER COUNTRY 305 specimens. In the hope that the light would improve we waited an hour at one place, but in vain. While there, humming-birds came to look at us and to bathe. Three different species were represented at one time or another. One, the largest, had a markedly forked tail and a long curved beak. A second, the smallest, was almost entirely a pale reddish-brown. The handsomest was deep metallic blue with a patch of paler greenish-blue at the throat; this was the “Costa Rican Wood Nymph” (Thalurania colom- bica venusta). As many travelers have noted, they were quite fearless, approaching to within a few feet of us. The blue one bathed repeatedly before us, retiring to a low branch after each dip to shake its feathers. There were many tracks leading down to this creek which Mr. Sullivan pointed out to me as those of the ‘mountain cow” or tapir. A large log which had fallen across the creek about eight feet above the water was a resting place for bats of moderate size, half a dozen being disturbed by the waving of my net as I tried to flush some insects that might be lurk- ing in the neighboring vegetation. As the morning of November 7 was bright, sunny and hot I decided to try Agua Buena Creek again. The trolley was not available and a horse not very useful so I set out on foot across the banana fields. Parts of Philadelphia North and Holanda Farms through which I passed had been abandoned owing to the poor quality of the fruit. These parts were in consequence much overgrown and progress over them was rather tiresome. Every morning during my week in this region I frequently met small flocks of moderate-sized dragonflies along the railroad tracks, these flocks separated from each other by intervals of a hundred yards or even less. The insects usually flew at such a height above the ground as to be beyond the reach of my net, say fifteen to twenty feet. But this morn- 306 A YEAR OF COSTA RICAN NATURAL HISTORY ing I fell in with a flock of about twenty-five individuals flying lower than usual and by standing still in the hot sun and waiting until one was very close to me I succeeded in capturing enough specimens to show that the flock was com- prised of four species—the very widespread Pantala flavescens and P. hymenea and the less widely found Tramea cophysa and Macrothemis hemichlora. These differently-colored and -marked species flew around and among each other, some- times seeming to drift in one direction, sometimes in another, chiefly employed in catching small insects in the air. They rarely alighted and apparently sustained themselves with little muscular effort except when the wind blew. In Chap- ter V we have already given brief descriptions of the species of Pantala. The genus Tramea is closely allied and likewise has the hind wings very broad at the base. Those interested in identifying them may recognize Pantala by the third and fourth segments of the abdomen (counting always from its anterior end) having each three cross-ridges which almost entirely surround the body, while the same segments in Tramea have only two cross-ridges each. Moreover, in Pantala in the apical third of the wings, are two parallel and closely adjacent veins (Mp and Rs) which are strongly waved with two convexities directed forward; these same two veins in Tramea are almost straight and only as they approach the hind margin of the wing are they curved toward it. Tramea cophysa has a brown band, about an eighth of an inch wide, on the base of each hind wing reaching from the fifth vein (counting that on the front margin as the first vein) to the hind margin; this insect is of very nearly the same size as the two Pantalas. The peculiarity of Macrothemis is that the tooth, which is present on the under side of each of the two claws with which each leg of a dragon- fly is tipped, is here so large as to equal or exceed in length and thickness the point of the claw itself. The claw conse- THE BANANA RIVER COUNTRY 307 quently appears divided into two nearly equal branches. Macrothemis hemichlora has a wing-spread of two and one- quarter to two and one-half inches, the wings being usually tinged more or less with yellowish; the females often, but not always, have the apical two-fifths of the front wings browner than the remainder of the wings. Going eastwardly along E Line I came to Agua Buena Creek where it was unshaded except for bananas, and under the little railroad bridge I saw a curious sight. A log a few inches in diameter lay partly in, partly out, of water but spanning the creek from bank to bank. From the soil at one bank ran a small covered tunnel, made of little fragments of wood, on to the upper, dry surface of the log. Farther on this tunnel was entirely open and still farther there was no tunnel at all. The uncovered portion and on to the bank where no tunnel existed was arched over with a mass of worker ants forming living roof and walls for the tunnel. Their antennze were waving to and fro, giving the mass a most dizzy appearance. With the workers were also some soldiers. Within these living walls could be seen two files of smaller workers running with great speed in opposite directions, one file empty-“‘handed” (empty-jawed would be more accurate), the other file bearing pupz and larve. I broke a portion of the tunnel where its roof and walls were formed entirely of wood fragments. Almost immediately a number of the larger workers appeared and stationed themselves along the edges of the broken portion, with their jaws wide open and their abdomens bent outward so as to present both weapons—jaws and stings—to a possible foe and so protect the files of workers which were running so swiftly through the tunnel. The appearance of these guards along the broken edges, with jaws and stings turned out- ward, was quite laughable, but a blade of grass touching them was savagely attacked. On both sides of the creek 308 A YEAR OF COSTA RICAN NATURAL HISTORY the ants entered the soil and without much time and digging it was impossible to ascertain whether a raid by one ant community on another was in progress or not. These proved to be Eciton predator, a well-known “army” or “foraging” ant. As I left the track and went on through the bananas to- wards the woodland, a number of Syntomid moths (Dinia eagrus) with clear wings and dark brown body fringed with bright red each side of the abdomen, were flying through the underbrush in a manner suggestive of wasps. A number of nuts and seeds were floating in the woodland stretch of Agua Buena Creek. In these the dragonfly Neo- neura amelia was laying eggs, the male and female flying together, the former holding the latter’s prothorax by the appendages at the tip of his abdomen. On alighting on the nut, the rather dull-colored female bent her abdomen to insert the eggs into the nut’s substance and owing to her colors did not at once attract the observer, but the male, standing rigidly upright with no support other than that given by his grasp of the female, was conspicuous from his red head, thorax and first few abdominal segments. At one time six to eight pairs of this species were ovipositing side by side. A chrysopid or hemerobiid larva came from somewhere on tomy net. He bore a silken case attached to which were remains of various insects whose juices he had sucked to his own bodily advantage. The case serves as a protective covering, presumably, but one might imagine that the at- tached insect remains might be a disadvantage to the chry- sopid, for, as the latter moved about, an insectivorous enemy might mistake the movements of the case as due to the attached insects and, snapping at them, kill the chrysopid within. There were more insects visible this day than on my pre- THE BANANA RIVER COUNTRY 309 vious visit, at least as long as the sun shone, and had the day remained bright later than 1 P. M. I should have found still others. Lizards and members of the salamander and frog groups were extremely abundant almost everywhere in the Banana River Region at this time. I made no attempt to gather any lizards—which were very speedy for one thing!—and I neglected the salamanders, but some of the frogs attracted my attention, as they must everyone’s who sees them, on account of their bright colors. The most abundant (Den- drobates typographus) was bright, flaming red with some irregular black marks on the under surface, and three-quar- ters to an inch in length. Less abundant but not rare was a pale greenish-blue frog (D. tinctorius) with black stripes and spots over the whole body including the legs; the body was one and three-quarters to two inches long. Few ani- mals exceed this frog in color variations. Dr. Gadow (Cam- bridge Natural History, vol. viii, pp. 272-273 and fig. 53) gives examples of their very great range and mentions that the animal’s specific name is due to the secretion from its skin being used to dye the green Amazon-parrots, the sub- sequently appearing young feathers being yellow instead of green. The secretion is also said to be used as an arrow poison by Colombian Indians. Individuals of D. tinctorius may very closely resemble in color those of another Costa Rican frog, Atelopus varius. Externally the former may be distinguished by the tips of the fingers and toes having bifid pads, the toes not webbed and the head in dorsal or ventral view less pointed at the muzzle. 4. varius has the same tips undivided and the toes webbed, but also shows many color variations. Gunther has given some colored figures of these in the Biologia Centrali-Americana and says: “The ground-colour is black, the ornamental colours pink, yel- low and green. The latter are distributed in various pat- 310 A YEAR OF COSTA RICAN NATURAL HISTORY terns, in the form of spots, or longitudinal bands, irregu- larly and more rarely symmetrically. The upper or lower parts may be black, almost uniform, or pink with scarcely a trace of black. In some of the most aberrant colour-va- riations the ground-colour is grayish, and the yellow bands are edged with black; in others the colours assume a retic- ulated pattern; and, finally, specimens occur of a brownish- olive, mottled and dotted with darker, the abdomen in these being yellow with some irregular black linear figures.” A third species, of about the size of the red one, was black with a narrow yellow stripe along each edge of the back; it may have been Hylodes rhodopis. All three species oc- curred both in forest and in banana land. After handling them I could not detect any disagreeable odor on my fingers. I killed a couple of the red frogs and one of the blue and black kind in the cyanide bottle and put them into separate bottles of alcohol, but the red ones lost their red after two days; the blue and black species retained the colors much better. On November 8 the trolley took me to Bearesem East Farmhouse, on the left bank of the Banana River. Mr. Veitch gave me the etymology of the scriptural-sounding name “‘Bearesem”’ as follows: before the farms were laid out there existed hereabouts a Banana River Saw Mill, usu- ally referred to as B. R. S. M. When the time came to name two of these farms some one devised their present names by spelling out these initials Be-ar-es-em, now pro- nounced “‘Béresem.” In advance of my arrival at Beare- sem East Farmhouse, Mr. Veitch had arranged by tele- phone for my use of a rowboat belonging to Mr. Brandon, mandador of the farm. He in his turn procured two ne- groes to do the rowing and also asked permission to accom- pany me for part of the way. The Banana River from Bearesem East Farmhouse to its THE BANANA RIVER COUNTRY 311 mouth is two to three hundred feet wide, with banks fifteen to twenty feet high; these were covered in many places with wild cane, elsewhere bordered by banana-plantations and on the right bank near the mouth by forest to a slight ex- tent. The current was not swift here for the fall was but slight, and within the banks mentioned and at a lower level were frequent stony or sandy beaches, enclosing so-called lagunas or almost landlocked arms of the river. We went downstream slowly, as near as possible to the vegetation of a bank so that I might see and gather any dragonflies to be found there. Once a ripple was pointed out as a place where the negroes had seen an “alligator” sink. We landed on several of the little beaches and these, or rather their lagunas, furnished better collecting-grounds than the river itself. As we neared the mouth of the river a tall tree on the right bank came into view; it was leafless but at its crown were many of the hanging nests of the oro- pendolas. I counted seventy and had then to stop count- ing owing to the changing position of the boat, but I think I had counted only half. Herons, white, gray or black, were disturbed by our approach and flew down ahead of us or turned and, passing over us, went upstream to be again aroused on our return. ‘These herons were not numerous, however. Sandpipers, looking like those at home, ran along beach-edges or skimmed the water’s surface. The river emptied into the Caribbean by a rather narrow passage. I landed on the sandy strip which separated the sea from the river before the latter narrowed. The surf was fine and heavy. Many black vultures were standing on the sand, waiting for river or sea to bring carrion. Four pelicans were swimming inside the breakers, looking like great swans, or flying over the water for short distances. A flock of fifty or more dragonflies (Pantala flavescens, Tramea cophysa, Miathyria marcella) was drifting over the sand 312 A YEAR OF COSTA RICAN NATURAL HISTORY strip near the river, fairly low but rising in the air when I tried to catch them in my net. (Miathyria is another rela- tive of Tramea and has two cross-ridges on the third abdom- inal segment but only one cross-ridge on the fourth segment; Miathyria marcella is one and one-half inches long and its wing-spread two and one-half inches; the hind wings have a brown band at the base which usually does not reach to the hind margin. Like Tramea cophysa it is a widespread species in the American tropics.) Close to the mouth of the river an arm—the Westfalia Laguna—branched to the left (north). Up this I now directed the boat, Mr. Brandon having been obliged to leave us long before; the banks were lined with water-hya- cinths (Eichhornia) which soon became so numerous that the water’s surface was completely covered. We forced our way through them as far as the railroad bridge, beyond which we could not go as a staging for repairs had been con- structed beneath the bridge. We went back to the river, stopping a few moments at the landing-stage of West- falia Farmhouse which latter lies on the beach between the sea and the laguna a few hundred feet north of the mouth of the river. Just above the opening of the Westfalia Laguna into the river, and on the same bank, was another, opening at both ends into the river so as to form a side channel. We went through this also. At some spots in it were millions of small, almost totally transparent fish. At times, a whole swarm of these fish leaped simultaneously out of the water into the air, giving the appearance of a wave. There were many transparent prawns, about an inch in length, here also. Among the water-hyacinths of the Westfalia Laguna, and to a less extent along this second laguna, we took a number of Enallagma cultellatum, a yellow, black and blue dragonfly of small size, new to Costa Rica. Between this second laguna THE BANANA RIVER COUNTRY 313 and Bearesem East Farmhouse, on our return, we picked from the stems of plants along the water’s edge, fifteen or twenty shed skins of a Gomphine dragonfly (Gomphoides) of a species which I never saw in the winged condition any- where in the Banana River region. An almost parallel ex- perience we had along Crum Creek, in Pennsylvania, in July and August, 1908, with another Gomphine and similar observations have been made in other parts of the world. On November 9 Mr. Veitch gave me a horse and tele- phoned to Senor Vivo, mandador of Bearesem West Farm, to meet me at Beverly Commissary. The only road was the railroad and the horses here were accustomed to walk the ties or, where there was a smooth path at the side of the roadbed, to travel on that. In either case the pace can rarely be faster than a walk, so at a walk I rode to Beverly Commissary where Bearesem West Main Line and Phila- delphia A Line met. Beside Beverly Commissary stood a schoolhouse, chiefly, perhaps wholly, for the negro chil- dren, and another edifice bearing the sign ‘‘ Philadelphia Church.” These buildings had red tile roofs which formed a strong contrast to the surrounding green of the bananas as seen from the verandas of the Superintendent’s house at Philadelphia South. At Beverly Commissary Sefior Vivo was waiting for me on horseback. He had agreed to take me to a creek crossed by “Spur 3,” to certain points on the upper Banana River, and to a potrero in (I think) Section 38 of Bearesem West Farm. So we rode first through the bananas from Beare- sem West Main Line to Spur 3 and along the latter track to a creek formerly crossed by a bridge which was carried away in the great floods of the preceding December. The creek was broken up into pools here and there and dragon- flies were fairly abundant but so alert and wary that I could not catch one. 314. A YEAR OF COSTA RICAN NATURAL HISTORY Then we returned to Bearesem West Main Line and rode along it past the farmhouse to the left bank of the Banana River, up that bank, under the railroad bridge, through an independent banana farm belonging to one Schutt and so to a reservoir collecting water for Limon. Under the bridge and along the river bank our way was on a narrow and muddy path over or alongside the two large iron pipes carrying the water from the reservoir to Limén. At the reservoir Sefior Vivo said it was impossible to ride further as the path going upstream was still narrower than before; after giving me directions for proceeding on foot he returned to his house for breakfast, while I, leaving my horse in charge of a man at the reservoir, walked up the river bank intending to rejoin Sefior Vivo at his house later so that we could go to the potrero together. From the reservoir, which was quite near the winding Banana River, two pipes ran up the bank to a small creek supplying water to a higher reservoir. The path followed those pipes. The forest came down to the upper reservoir and I turned into it as there was little to be found at the reservoir or the creek. However, when I first arrived at the upper reservoir, there were on the cement walls a great many of what I supposed to be flies, but on catching and examining a number they proved to be small crickets, a species of Rhipipteryx allied to pulicaria. The body, in the dried specimens, is not quite 4 millimeters long, of a dull pale brown or a dull olive above, with blackish marks on the head (other than the eyes) and on the prothorax. The under surface and the legs are largely blue-black, especially the relatively huge femora of the third pair. As in other species of Rhipipteryx, described on page 202, the wings when at rest project a short distance beyond the tip of the abdomen. In the forest I found a number of dragonflies which I THE BANANA RIVER COUNTRY 315 had not seen since my visit to Guapiles and Guacimo in June. Some of these took a long time to capture, such as species of Gynacantha found only in the dark forest of the warmer parts of the country so far as my experience goes. I noticed an individual in a certain round, slightly depressed area containing no growing plants but with leaves and sticks on the still damp ground—all of which made me suspect that in wet weather this was a little pool. It was surrounded by a circle of small palms two to four feet high and the whole place was deeply shaded by tall trees. When I first found this spot and saw the insect, the dragonfly disappeared and I went elsewhere, after mentally noting the place for a visit on my return. A few hours later I came again and saw two species of Gynacantha there. They were large dragon- flies of strong flight but one of their peculiarities was that they usually alighted close to the ground and almost invari- ably selected the undersides of the midribs of the arched palm leaves. In such positions it was often impossible to use a net and then I tried to slip up to them, so slowly and quietly that they would not be disturbed, and catch them in my fingers. I succeeded with Gynacantha trifida but not with the larger G. gracilis, although once as the latter flew off it rubbed against my dangerous fingertips. It was gracilis I especially wanted and it was essential not to frighten it, for I knew that if that happened it would fly far away and I would never see it again. So I followed it as it flew from one side to the other of the circle of little palms as I dis- turbed but did not alarm it. At times when my fingers approached very near, its wings began to flutter slightly ; if I drew off, the fluttering ceased, if I came nearer the fluttering became flight and I must stalk the prey again. It took fully an hour of these tactics before I finally got into such a position that I felt justified in making a sweep of the net that would mean capture or complete loss—but capture 316 A YEAR OF COSTA RICAN NATURAL HISTORY it was. During this hour of stalking one gracilis, however, I caught two trifida in the same spot. Gynacantha gracilis is similiar to G. membranalis mentioned in the preceding chapter but a little larger (three and one- half inches long, wing-spread four and three-quarters inches). The reddish-brown spot at the base of the hind wings is much smaller, being adjacent to the front margin only in- stead of extending halfway across the wing, as it does in membranalis. Its eyes, in life, were blue above ranging through green and yellow to reddish-brown below, instead of the metallic green of its congener. Previously, gracilis had not been reported north of Surinam. Gynacantha trifida is a smaller species (two and three-eighths inches long, wing-spread three and one-quarter inches) with uncolored wings, its eyes brilliant green, darker above. The Gynacanthas have the absolutely largest eyes of all dragonflies, possibly of all insects. They fly in such dark places as these virgin forests or, if in the open, at dusk and they have occasionally been known to fly to artificial lights at night. At Poyntelle, Pennsylvania, in June, 1904, I found two species of Cordulegaster along the same brook; one species with smaller eyes (C. diastatops) was found only where the brook flowed through open pasture; the other (C. maculatus) with larger eyes, occurred only in the forest courses of the brook where there was much less light. All these observations suggest an inverse correla- tion of size of eyes and quantity of light in which they are used, at least in the case of swift-flying species. But there are other species occurring in equally dark places in these Banana River forests with eyes no larger than those of their light-dwelling brethren. Such are Protoneura re- missa and Philogenia carrillica, which I caught this day in the forest with the Gynacanthas. These two species, however, seem to travel only short distances—a few THE BANANA RIVER COUNTRY 317 yards—and to float through the air instead of moving rapidly. The Protoneura and the Philogenia are much smaller insects, being one and one-half and two and one-quarter inches long respectively, of slighter build and with narrower, feebler wings. Their right and left eyes are separated from each other on the upper surface of the head by a distance equal to, or greater than, the width of either eye. The right and left eyes of the Gynacanthas meet, for a relatively long distance on the top of the head, so that they cover a larger part of the surface of the head than is the case in the two smaller species. The heights (in millimeters) of the eyes of these four dragonflies are: Protoneura remissa 1.11, Philogenta carrillica 2.88, Gynacantha trifida 5.76, G. gracilis 7.44. It is well known that the surfaces of the eyes of in- sects are made up of a number of facets, or lenses, usually hexagonal in shape, fitting against each other. The diam- eters of these facets are usually greater on the upper surface of a dragonfly’s eye than on the lower surface of the same eye, and the size is different in different species, Protoneura having the lenses smaller than in any of the other three. The distinctness of vision of an insect’s eye is proportional to the number of these surface lenses whose inner nerve fibers are stimulated by light. Other things being equal, therefore, the greater the number of facets the better the eye for distinctness. The smaller the diameter of the lenses, of course, the greater is the number in a given area. In spite of the smaller diameter of Protoneura’s lenses, a rough calculation shows that its eyes contain only one-half, or less than one-half, of the number of facets of the eyes of Gyna- cantha trifida, and between three-sevenths to one-third as many facets as the eyes of G. gracilis. On the other hand, the intensity of the light stimulation which the inner nerve fibers (corresponding to each facet) receive increases with 318 A YEAR OF COSTA RICAN NATURAL HISTORY the diameter of the facet; the larger facets of Gynacantha gracilis must give it an advantage in this respect over its three associates. While these eyes are not portions of spheres but of much more complicated shapes, we may, for ease of comparison, say that the eyes of Protoneura and of Philogenia correspond to much larger segments of a sphere than is the case in Gynacantha where the eyes are much flatter on the upper surface especially. Taking the greater absolute size of the last into account, it results that the radius of curvature of the eye surface of Gynacantha is greater, and this in turn means that the divergence of the axes passing through the centers of the facets is less than in the two smaller insects. The greater this divergence the less distinct the vision. These considerations tend to show that the eyes of the Gynacanthas are more efficient organs than those of Protoneura and of Philogenia. But since other dragonflies (4nax, 4shna) with eyes almost as large, both absolutely and relatively, as those of Gynacantha, are found in well-lighted places, and as Protoneura and Philogenia live chiefly in the shade, we cannot suppose that the large eyes of Gynacantha are a result of shadow-haunting habits. Rather must we look on their large eyes as giving them an advantage over insects with less efficient visual organs when they took to living in darkness. These reflections on the eyes of shade-dwelling dragon- flies suggest a question as to the condition of the eyes of the shade-dwelling butterflies which have been mentioned in various preceding chapters. A comparison of dried speci- mens furnishes the following results. Callitera menander of Turrialba and Peralta, observed only in dark woods, has eyes not as large as those of Callidryas argante and Pieris ilaire, creatures of about the same total size, bask- ing in the brightest sunshine of Guacimo. The eyes of the Ithomiine butterflies, which we took in the shade in va- THE BANANA RIVER COUNTRY 319 rious places, are not absolutely or relatively large. The largest eyes of all our Costa Rican Lepidoptera are those of some Sphinges and of the great Noctuid Thysania agrip- pina, but even they are not nearly as large as those of Gyna- cantha gracilis, in spite of the heavier bodies and greater wing-spread of these moths. Searching for these various forest species kept me until three o’clock, when I retraced my steps to the lower reser- voir, mounted and rode back to Bearesem West Farmhouse and there told Sefior Vivo that I thought it too late to go to the potrero; then I returned along the tracks to Mr. Veitch’s. There were many hawks of several species in this region at the time of my visit and to-day, as on other days, they often sat upon the tops of the telephone wire poles which were usually rails set up in the ground. I never saw these hawks pay any attention to the flocks of dragonflies which were so common along the tracks, although these insects have often been found in hawks’ stomachs in Europe and in the United States. In the evening Oscar, Mr. Veitch’s colored house boy, . killed a “zorro” that was troubling the chickens and as I did not know what a zorro was, it was brought for me to look at. It proved to be an opossum and at Mr. Veitch’s suggestion Oscar cleaned the skull forme. It agrees with the description of that of Richmond’s opossum (Didelphis rich- mond) originally described from Greytown, Nicaragua, and from San José, Costa Rica. On another evening (Novem- ber 7) a scorpion (Centrurus bicolor) three inches long, ap- peared in the office and was killed. On November 10 two “boys” took me on a trolley past Bearesem West and Beverly Farms to the Bananito River. The last part of the track lay in Bananito Farm, belonging not to the Fruit Company but to Mr. Minor Keith. The trolley boys had orders to wait for me at the Bananito until 320 A YEAR OF COSTA RICAN NATURAL HISTORY I was ready to return. On arriving there I walked upstream along the same (left) bank, first through bananas and cacao trees and then along the stony beaches which lay a few inches to five feet above the water. The general character of the river was much like that of the Banana River but the water was shallower and clearer, this day at least. On the beaches were many caiman tracks, some foot- prints being five to six inches across. The only caiman I saw was about four feet in length; it was lying in mud at the water’s edge, only its head exposed to the air. It slid into the water as I approached but only went a short dis- tance, for a few minutes later I was on a higher bank and saw it distinctly in the shallow water. I threw a stick at it, whereupon it turned around very violently, raising a great cloud of mud, and disappeared. Subsequently I saw one of about the same size in the shallow water at the opposite side of the river, but it may not have been the same one. There were many fish of all sizes in the Bananito, as there had been in the Banana River when I was there, also many shrimps (Palemon jamaicensis). The small fishes from the Bananito proved to be Gambusia (Priapichthys) annec- tens. Here and there along the river was growing an acacia iden- tified by Professor Pittier as Acacia campeachiana !=cochle- acantha) five feet or so in height. It resembled the bull’s horn thorn described in Chapters XVII and XVIII, but had a less woody stem or trunk and paired thorns not curved nor inclined toward each but nearly at right angles to the stem bearing them. There were no little fruit-like bodies at the tips of the young leaflets, but along the mid-rib of each leaf was a row of urn-shaped glands, one at the base of each of the twenty-seven or more pairs of pinne. Even in the dried 1This is the campeachiana of Miller, of the eighteenth century Gardener’s Diction- ary, not the campecheana of Schenck, 1914. THE BANANA RIVER COUNTRY 321 specimens these urns are one-thirty-second of an inch high. In not a single plant of this species did I find any ants, but it looked so much like the bull’s horn thorn as to suggest that it might be the starting point for the development of the latter. In Bananito Farm as in other banana plantations, there were rubber trees scattered here and there, left from the original forest when the ground was cleared. The trunk of every such tree was scarred with cuts made through the bark to secure the rubber. In some of these little scars bits of the rubber could be seen, of a black color and quite elastic. Mr. Veitch told me that the rubber trees in his division had just recently been bled, that the sap is taken once a year, to the extent of about fifteen hundred pounds. It is white when it runs out of the tree, for which reason it is called “milk of rubber,” but blackens on exposure to the air. In cutting through the bark to bleed the tree it is important not to girdle the tree completely, as such treatment kills it. The proper cuts, according to Mr. Veitch, are two half- circles inclined obliquely downward toward each other, but one cut about four inches below the other. If cut in this way, a tree recovers from the wounds during the year be- tween cuts. Near the end of the track was an immense tree (not a rubber tree, however) whose buttress-like roots ran as wooden ridges twenty feet from the trunk and reached a height of four feet, decreasing to a few inches as the distance from the trunk increased. Thence I went down the left bank of the Bananito until the bananas gave way to a dense growth of wild cane. Retracing my steps I crossed to the right bank by a fallen tree which served as footbridge, crossed a re- cently-burned patch of cane where the sun beat fiercely, to some thinned-out forest, but finding nothing returned to my trolley and so back to Philadelphia South. 322 A YEAR OF COSTA RICAN NATURAL HISTORY Although the plants and animals of Cartago and the Ba- nana River district are mostly different, correlated with the differences in altitude and in temperature, there are many species which are found in both places, such as the escobilla (Sida rhombifolia) among plants, Anartia fatima, the most common of Costa Rican butterflies, Pantala flavescens among dragonflies, and among birds the tijo-tijo or garrapatero and the black-headed (zopilote) and red-headed (zonchiche) vultures. The water-hyacinth, so abundant in the West- falia laguna, grew in a ditch in Cartago. CHAPTER XVII ALAJUELA AND THE VOLCANO POAS West of San José, the country descends rapidly toward the Pacific coast, where the fringe of low level land is far less extensive than on the Atlantic side. Much coffee is grown on the upper Pacific slope and the railroad runs through highly cultivated and enclosed plantations. Few good col- lecting grounds were to be found in such a region, neverthe- less the river valleys were often deep and richly wooded and sheltered an interesting flora and fauna. We spent one day in February in the midst of the dry season, in Heredia, the first important town on the railroad west of San José and six miles from: the capital. We went by train in half an hour, and after walking about the plaza and looking into the large and imposing church, started north along a good though hot and bare road toward the village of Barba, two kilometers away. This gave us beautiful views of Poas, Barba and to the east a smaller hill which a passing traveler told us was EI Gallito. We walked almost to the village of Barba— classic ground at that time, as “Don Cleto,” otherwise El Senor Presidente Don Cleto Gonzalez-Viquez lived there— but not quite into the village itself. We paused where the road crossed a little stream, Quebrada Seca, by a wide bridge which was supported on a single narrow arch of masonry, making a sort of ravine. The brook was quite shady and we went down to it first because the shade was so inviting, but the high straight walls of the arch proved a fruitful source of exuvie of transformed dragonflies, so we stayed there. We were fortunate enough to find two spe- 324, A YEAR OF COSTA RICAN NATURAL HISTORY cies (Heterina capitalis and Archilestes grandis) in the act of transforming which we had not known before in the larval stage, and it is much easier if one can find the dragon- fly sitting on its shed skin than to raise the insect from the larva. We spent several hours around this little stream, gathering exuvie and a few living larve. The larve of Archilestes grandis were in a quiet pool where we could see them moving about. They were very minnow-like in be- havior, floating in mid-water, making a short dart in one direction or another as a stick was brought near and, when raised from the water in the net, leaping sidewise into the air as a minnow does in a similar situation. Their semi- transparent bodies increased the minnow-like appearance. The railroad from Limon had its terminus at Alajuela, twelve miles west of San José. Our friends Mr. and Mrs. J. B. Clark lived at a large beneficio, “‘E] Brazil,” of which Mr. Clark was manager, a mile east of Alajuela, and to their kindness and hospitality we owe many opportunities of ex- ploring the neighborhood. We visited the beneficio in Sep- tember, December, January, February and March, that is, both in winter and in summer. Alajuela is a neat and attractive town of 6000 inhabitants, prettily situated among the hills, of which fine views are obtained from the plazas. There is the usual Parque Cen- tral and bandstand and facing the Parque is the chief church, in this case a very large white structure with thick adobe walls and a conspicuous red dome of corrugated iron. In addition there were several large open grassy plazas on the outskirts of the town, on which Mrs. Clark and A. used to exercise their horses at a gallop when the roads were so deep in mud that rapid riding on them was out of the ques- tion. The town contained quite a large hospital, several public schools, and a very interesting school of weaving, “Escuela de Tejidos.” The art of weaving, if it ever existed ALAJUELA AND THE VOLCANO POAS 325 in Costa Rica to any extent, had completely died out, so that when it was proposed to revive it there were no looms. These were imported from Salvador, where weaving is a most important art, and teachers were brought from Salva- dor and Guatemala. The school was designed to teach useful arts to the young people of Costa Rica. They had ten or eleven large hand looms on which they wove a great variety of cotton cloths, white ducks and highly colored denims as well as white and red-and-white cotton towelling. The yarns and threads were imported from England and we were shown several dozens of patterns, stripes, checks, plaids and plain, in a number of weaves. They were all such as were worn here by the poorer or plainer people and found a ready sale. In addition to the weaving of cloth, the school taught cane-seating, basketry, hat-weaving, stock- ing-weaving and leather-work (particularly the weaving or braiding of bridles and whips). The school was becoming deservedly popular and promised to be very valuable, as there were few native arts here compared with those of other Central American countries or with Mexico. The beneficio (a place where coffee is bought, cleaned, dried and prepared for market) of “El Brazil” was on the Ciruelas River, which is there joined by the Brazil River. The house was a pleasant two-story frame building, with both lower and upper porches or “corridors” as they are called in Costa Rica, the upper floor commanding wide views towards the volcanoes of Pods and Barba to the north, the range of Candelaria to the south and a great sweep of rich rolling country between. In front of the house in a tiny garden stood lemon and orange trees and an eucalyptus, the latter brought from California by a former resident. There are many places in Costa Rica where the eucalyptus was introduced, when it was supposed to be efficacious against malaria. The tree thrives and flourishes although 326 A YEAR OF COSTA RICAN NATURAL HISTORY the theory has become obsolete. At the back of the house were several higuerones or wild fig-trees of considerable size. The dwelling was separated by a paved road from the offices, in front of which stood a very fine guanacaste tree (Enterolobium cyclocarpum) and some smaller trees. Our first visit to El Brazil was in September, when we spent a fortnight. ‘This was in the winter or rainy season and all these trees were in full leaf. The roads in every direction were unbelievably deep in mud, the roadside banks were draped with ferns, mosses and selaginellas and studded with the exquisite blue flowers, two inches across, of the little gesneriaceous plant Achimenes longiflora. The pord and madera negra (Gliricida maculata) trees in the coffee plantations—coffee being a crop that must always have some shade—were also in full leaf and the coffee trees were laden with bright green berries. The pastures and fields in every direction were beautifully green, the rivers were rush- ing torrents and the straight-sided ditches that ran through every pasture to water the stock were full of water. Beyond the offices at El Brazil stood the beneficio, a huge structure of corrugated iron, housing not only the coffee machinery but also the sawmill and the dynamos for the electric lights. Behind the house and the beneficio were the great cement platforms or patios on which the caffee was spread to dry. In all, there were several acres of these patios. They were divided into smaller platforms by cement troughs designed to wash the coffee thoroughly and to carry water and coffee to any desired portion of the patio. The machinery of the beneficio was run by power from the river. The water was conducted by a cement-lined channel or in- take and operated a turbine at the bottom of a twenty-foot well. During our first visit, in September, there was of course no coffee handled, but the sawmill was quite busy. The ALAJUELA AND THE VOLCANO POAS 327 logs were cut in the forest, almost universally to a standard length of four varas (11 feet), squared a little and sharply pointed at one end and each log was dragged to the mill, often for great distances, by a pair of oxen. After several days of excessive rain, the mud became so deep that it was impossible to drag the logs at all, so that it was necessary to wait until the mud dried a little. This was the time- honored method but disastrous to the roads over—or through —which the logs were dragged. Since our return from Costa Rica such hauling has been prohibited and all logs must be carried in carts. The number of species and varieties of timber trees in Costa Rica is very great. Mr. Clark had a list of one hundred woods which had been brought to El Brazil or the sawmill at “La Eva” in Sarchi de Grecia. All of these were distin- guished by native names and possessed qualities rendering them better suited to one or another purpose. Many Costa Rican timbers besides the well-known mahogany (Swie- tenia) have most beautiful grains and are capable of the finest finish. Conspicuous among these are the “‘cedros” (Ced- rela sp.) of which ‘“‘cedro amargo” (C. glaziovii) furnishes cigar-box wood and much of the “mahogany” of modern cabinet-makers. Other important lumbers are the numerous “‘quizarras,” all belonging to the Lauracez; the “‘guayacan” or “corteza amarilla” (Tecoma chrysantha), one of the nu- merous woods called Lignum vite and much used for cart wheels; the “‘laurél”’ (Cordia gerascanthus) and the many valuable leguminous trees such as the guanacaste, the guapinol (Hymenea courbaril), the gavilan, the candelillos or vainillas (Cassia spp.) and the guachipelin (Diphysa robin- toides). The last is valuable for posts, foundations, or other underground work, because when buried it remains sound almost indefinitely and is not subject to the attacks of insects. 328 A YEAR OF COSTA RICAN NATURAL HISTORY In front of the offices and beneficio was a large paved courtyard, where the logs were piled in sawing time and later the ox-carts full of coffee waited their turn to discharge their loads of red cherry-like berries. Many beneficios have a finca or farm belonging to them, on which is raised at least a part of the coffee beneficiated, although all buy coffee from other growers in addition. El Brazil, however, had no finca and the entire quantity of coffee (nearly 5000 fanegas, equalling 575,000 pounds or 4000 sacks) was purchased. The number of small landowners and coffee growers in Costa Rica is very great and everyone who had any coffee to sell must be dealt with and bargained with separately—at El Brazil there were over six hundred growers to deal with. The competition between the different beneficios ran very high and to get the most coffee without paying for it a price that prevented any profit made the purchasing of this coffee fruit an art in itself. Dr. Karl Sapper, after dwelling in Guatemala for some years and traveling throughout Central America, pointed out in 1902 the great advantage which Costa Rica possesses in the predominance of small landowners, in contrast with the conditions prevailing in the larger countries in that part of the world. Where many own small pieces of ground, a much greater and more general interest in the prosperous development of commerce and of peaceful progress exists than where there are only a few great landowners and the great mass of the population, having nothing to lose, are more indifferent and more easily drawn in the wake of po- litical adventurers. When we went to El Brazil in December the dry season had begun. The coffee was ripe on many farms and the picking, which is done by hand, mostly by women and girls, was in full swing. The coffee trees looked very pretty, with their glossy dark green leaves and showy bright red Volcano Barba and Coffee Patios of “EI Brazil,” Alajuela. Covering Piles of Coffee for the Night, “EI Brazil.” To face p. 328 Guayabo—Psidium guava. Entire Tree and Fruit pinned to Trunk. To face p. 329 ALAJUELA AND THE VOLCANO POAS 329 berries. Owing to eccentricities in the rainfall at the time of flowering the coffee was not ripening evenly and several pickings were necessary on many farms. The ox-carts laden with coffee were rolling steadily into El Brazil all day (ex- cept during the hot noon hours) and part of the night. The berries were measured from the ox-cart into a big double- hectoliter measure, which held about three hundred pounds of fruit and made a load for two men, and were then turned into a receiving vat full of water. There were two large cement-lined vats in the courtyard, with such an arrange- ment of water inflow and outflow that they could be filled or emptied at will, the outflowing water carrying the coffee with it. As quickly as possible after arrival the berries were passed from this first vat to be “‘pulped” by machinery and then carried by water to other cement vats for fermenting from 24 to 60 hours, according to the temperature. Fermenting was conducted for the purpose of removing the mucilaginous substance surrounding the grains, and was continued until they no longer felt slippery when rubbed. After fermenting they were conveyed by means of a centrifugal pump and piping to the troughs surrounding the patios, where they were thoroughly washed. The water ran by gravity (10 per cent fall) through these ditches and there were numerous ingenious arrangements of gates and movable dams so that the coffee was repeatedly washed and the trash and such hulls as had slipped in were carried off. The smaller, inferior grains were also separated from the larger and heavier ones. Then the coffee was spread out on the patios to dry. During the day it was spread out in a comparatively thin layer which was constantly turned over and moved about by bare- footed men armed with a peculiar wooden hoe. Towards evening, or if a shower threatened, it was raked into piles or long ridges and carefully covered with tarpaulin as a pro- tection against rain or dew. When the coffee was one day 330 A YEAR OF COSTA RICAN NATURAL HISTORY short of dry it was finished in a steam drier, sent through hullers, cleaned by machines, and finally sorted by a most ingenious hollow cylinder with holes of varying sizes so that from one machine dropped out three grades or sizes of grains flat on one side, broken grains and Caracol or pea- berry, which last is a grain round on both sides and consid- ered the choicest. After all that, the coffee was hand-picked by women and girls, who removed any grains that were defective or twisted in shape, even if the size was right, for all such things make the coffee inferior. The differences on which this sorting was based were very slight—much slighter than those made use of by botanists and zodlogists to distinguish different species of plants and animals—but the resulting “‘grades” fetched very different prices in the receiving ports, London, San Francisco, New York, or Hamburg. When we commented on these fine distinctions, Mr. Clark told us of two instances known to him of experts who were able to distinguish different sorts of coffee and of tea by taste, this sense furnishing the final test in such matters. When at a coffee broker’s in London he selected samples of sixteen lots of coffee, all grown in Costa Rica but in different parts of the country. A cup of coffee was prepared from each one of the sixteen samples and each cup given a number whose meaning was known to Mr. Clark alone. The broker then tasted each cup and said “‘this is Tres Rios coffee,” ‘‘this is Cochenour’s from Juan Vifias” and so on, and fourteen out of his sixteen iden- tifications were correct. A tea expert in San Francisco, on a wager, identified two hundred and fifty different kinds of tea by tasting the beverage. This is surely carrying the development of the sense of taste to a high degree of dis- crimination. When the sorting was completed and the coffee bagged —each sack shipped from Costa Rica is required by law ALAJUELA AND THE VOLCANO POAS 331 to be marked ‘“‘Café de Costa Rica”—the shipping began. When we were at El Brazil, and during all the preceding years of the existence of the beneficio, its coffee was carried by ox-carts over the Carretera Nacional from the courtyard of the beneficio all the way to the wharf at Puntarenas. It took four to six days to make the journey. The carts traveled all night and rested through the day, saving both drivers and oxen from the midday heat. In the fall of 1910 the gap in the government railroad from San José to Pun- tarenas was completed, and since then the coffee has been shipped by rail. It now reaches Puntarenas in five hours with only one mile of carting to the railroad station. The country about Alajuela was so highly cultivated that the stream ravines with their fringes of trees offered almost the only collecting places of interest, and even these were never productive of many dragonflies. They were very pretty, however, and quite different in their vegetation from the little valleys about Cartago. Along the Ciruelas River were many uruca trees (Trichilia havanensis), mingled with guayabos (Psidium guayava). The latter here grew twenty to twenty-five feet high, and in their growth and habit of branching closely resembled an old neglected apple tree. The bark of course was utterly different, for the guayabo has the habit of shedding its outer layers in large pieces which curl up, usually lengthwise of the stem, into tight rolls before they drop off. As a result of this habit the tree is kept free of epiphytic plants. The fruit was not ripe in September, but we tasted it later. It is very good raw, but rarely reaches perfection because so many small animals are fond of it and consequently it was usually gathered and cooked in some way before maturity. The ciruela or jocote (Spondias purpurea) was very common about Alajuela, both along the river banks and in the “‘living fences” where it replaced the pord to some extent. It is rather a pretty tree and the 332 A YEAR OF COSTA RICAN NATURAL HISTORY purple plum-like fruit is excellent when made into various “‘dulces.”” There were fewer epiphytic bromeliads on the trees here than at Cartago, but the latter were often over- grown with “‘cana brava,” a climbing cane with whorls of short branches and long internodes, or with “matapalo.” The latter resembles our mistletoe in appearance and chokes and kills its host unless removed by man. Often in the fences were tree yuccas or “‘itabos” (Y. elephantipes), ten or more feet high, or the fence or hedge might be formed of narrow-leaved terrestrial bromeliads known as “‘pinuelas,” with spines along the margins of the yard-long leaves, curved toward the leaf tip. Such a hedge, formidable in itself, was sometimes supplemented by posts and barbed wire. Mr. Clark told us that before barbed wire came so extensively into use for fencing, fields were often separated from each other by deep ditches and we saw some of these, widened no doubt by rains and running water, which were three to six feet deep and as wide at the top— a much more laborious and expensive method of marking boundaries. South of the house, on the left bank of the Ciruelas River, was a particularly beautiful pasture, ‘Flores’ potrero,” which sloped steeply down to the rushing, foaming, rocky river. It contained many fine trees, among them the guay- abo we have figured and several magnificent isolated specimens of the vainilla or candelillo. This is a large tree, fifty to sixty feet high, with a spray recalling our New Eng- land elms but with a flatter head, and in September it was covered with pyramidal clusters of brilliant yellow flowers borne on the tips of the branches. One of these great trees covered with its yellow blooms was a dazzling spectacle. The river seen from this potrero was very picturesque with little rapids, overhanging trees and projecting stones. The latter were often occupied by lizards and we sometimes ALAJUELA AND THE VOLCANO POAS 333 amused ourselves by throwing stones toward them until they were induced to move into shelter, skimming over the surface of the water and using both legs and tail as locomo- tive organs. Many lizards can travel a surprising distance in this way. They move so quickly that it is difficult to see just how they do it, but they do not appear to be at all sub- merged. Still farther south, after crossing several potreros, on the banks of a little stream, were numerous bull’s horn thorns growing mingled with urucas and jocotes. Although they did not seem to be growing actively we found some young shoots with the yellow so-called food-bodies on the tips of the tender green leaflets and the three conical glands in a row on the petiole still exuding their sweet secretion. This bull’s horn thorn was Acacia costaricensis and derives its pop- ular name from the strong, sharp, paired thorns, which quite closely resemble the horns and forehead of a bull. In many species these thorns are commonly hollowed out and tenanted by small stinging and biting ants (Pseudomyrma sp.) which feed upon the nectar and “‘food-bodies” and have been supposed to be of great protective value to the plant. The ants on the thorns in this potrero were both nu- merous and aggressive. The largest bush had also a big wasps’ nest hanging from its branches. We wanted to pho- tograph a small spray of this plant, and our efforts to do so, although finally successful, were attended with various amusing difficulties. The plant is extremely sensitive to shaking and no matter how gently the spray was removed and laid on the background—the camera on its tilting top being focused and ready—the leaflets would close a little. The ants occupying the thorns also objected to the shaking, and it proved a hard task to handle the spray “gently” when a stream of savagely-biting little furies poured down the stem ready to attack the intruder with jaws and stings! 334. A YEAR OF COSTA RICAN NATURAL HISTORY And at the most critical moment of one attempt we had to drive away the steers living in the potrero, for their over- mastering curiosity almost caused them to upset the camera. This little ravine contained many groups of a huge un- identified aroid, called by the Costa Ricans “‘hoja de pato” (goose leaf). It is not peculiar to Alajuela, being found in many places over the Atlantic slope as well as the Pacific, The triangular leaves reach a breadth of nearly two feet and a length of two to three feet, while the leafstalk may be nearly four feet long. One that we measured was over six feet long in a straight line from base of stalk to tip of leaf, but in its natural position the leaf and stalk are at right angles. The rhizome, creeping over the ground or sub- merged in the mud, is six to eight inches in diameter. On September eighth we found in lanes between jocote hedges, a number of clicking butterflies mentioned on page 296. There were two species, Ageronia ferentina with the hind wings whitish below and Peridromia fornax with the same surface brownish-yellow. ‘They seemed to prefer to alight on bare or nearly bare tree-trunks with the wings outspread instead of folded together as in most butterflies. The flight also is characteristic—a number of quick flaps of the wings, then they are held still while the insect sails for a short distance, after which the rapid flapping is resumed. We caught a few specimens and took them alive to our screened room, where we released them in the hope of learn- ing how the clicking was produced; but during the three days the insects were under observation we did not hear them click. Hampson? thinks that in Ageronia arethusa the sound is produced by a pair of hooks attached to the base of the fore wing playing against a pair of hooks on the side of the thorax. I have been unable to find these_hooks 1 Proc. Zool. Soc., London, 1892, p. 191, fig. 3. ALAJUELA AND THE VOLCANO POAS 335 in Peridromia fornax, however, even with Hampson’s figure before me. Going upstream, the Ciruelas River led chiefly through a series of small, narrow, grassy potreros, each running to the river bank and each one representing a different ownership. About three-quarters of a mile upstream was a log bridge (frequently swept away by the trees and other débris brought down by the floods of the rainy season) which led into a large potrero where the land rose into a considerable hill. This formed the divide between the Ciruelas and Brazil Rivers, the latter being much the smaller. At a point roughly opposite the log bridge, the Brazil flowed through a little ravine so thickly set with trees that it was exceedingly dark.