DEPARTMENT OF COMMERCE AND LABOR BUREAU OF FISHERIES GEORGE M. BOWERS, CommU«lon«» THE MUSSEL FAUNA OF THE MAUMEE RIVER Bureau of Fisheries Document No. 757 WASHINGTON GOVERNMENT PRINTING OFFICE 1912 *» ^^•' DEPARTMENT OF COMMERCE AND LABOR BUREAU OF FISHERIES GEORGE M. BOWERS, Commissioner THE MUSSEL FAUNA OF THE MAUMEE RIVER 15 Bureau of Fisheries Document No. 757 WASHINGTON GOVERNMENT PRINTING OFFICE 1912 A V X^ h^ > THE MUSSEL FAUNA OF THE MAUMEE RIVER By h! Walton Clark and Charles B. Wilson Bureau of Fisheries Document No. 757 CON TENTS. Page. Introduction 3 Geology and geography of Maumee and upper Wabash Basins 4 Biology of Maumee River 8 Mussel survey 10 Headwaters of Maumee and Little Wabash Rivers 10 St. Marys River and Spy Run 11 St. Joseph River and feeder canal 18 Maumee River 25 Discussion of mussel species 39 General distribution of shells in Maumee River 55 Food of Maumee mussels 57 Parasites of Maumee mussels 61 General conclusions 71 2 T) THE MUSSEL FAUNA OF THE MAUMEE RIVER. By Charles B. Wilson and H. Walton Clakk. INTRODUCTION. This report embodies the results of an investigation of the mussel fauna of the Maumee River, carried on under the auspices of the United States Bureau of Fisheries during the summer of 1908, and it represents a part of the series of investigations undertaken by that Bureau looking toward means of continuing the suppl}^ of mussel shells. Upon this supply, which had begun to show signs of serious depletion, depends one of the important industries of the country, the pearl-button industry of the Mississippi Valley. The scope of the inquiry embraced whatever appeared to be closely related to mussel life, such as the history and character of the river and its present mussel supply as regards abundance, quality, and species. It was desirable also to ascertain where good breeding stock might be obtained, the location, character, and condition of mussel beds, and circumstances in which different species appeared to thrive. Such features of general life history as food, habits, time and manner of spawning, enemies, and parasites were also noted whenever opportunity offered. In the choice of the area to be studied, the features to be emphasized during the investigation, and similar details we have had the direc- tion and advice of Dr. Barton W. Evermann, Chief of the Division of Scientific Inquiry of the Bureau of Fisheries. His counsel has been of gi-eat assistance in the preparation of this report. In addition to the Maumee River, the upper part of the Wabash Basin was also investigated, in order that the two basins might be compared where they are close together. This latter part of the work also yielded many interesting results, especially in the spawn- ing habits of mussels and the life history of parasites. The results of the Wabash investigation are frequently referred to briefly where they carry on lines left incomplete by the Maumee studies. 3 4 MUSSEL FAUNA OF MAUMEE RIVER. The authors had as an associate in the work Mr. Donald M. Earll, of Washington, D. C, to whom is due much credit for the results. During the Maumee investigations, not only the main river, but also the two streams which unite to form it, the St. Joseph and St. Marys, were examined for the last few miles of their course, and the other principal affluents, the Tiffin and the Auglaize, were likewise investigated for a short distance above where they join the Maumee. Other bodies of water flowing into the Maumee or closely related to its basin, such as the feeder canal at Fort Wnyne and its reservoir, and Spy Run, a small tributary to the St. JNIarys not far above its mouth, were studied. Below Defiance, Ohio, the IMiami and Erie Canal runs parallel with the river, and was followed to avoid the riffles and difficulties of navigation to be encountered in the main stream. It otfered a favorable subject of research in itself and the Maumee could readily be examined from time to time, as that river was never any great distance away. During the trip to the United States fish hatchery at Put-in Bay, Ohio, o])portunity was taken to examine the mussel fauna of that region. This gave data for inference as to v\diich species had prob- ably ascended the Maumee from Lake Erie and which had entered more recently from the Wabash system. It also furnished a basis for comparison between the river and the lake mussels as regards size, etc. The authors are about equally responsible for the work. To Mr. Clark belongs the credit for the final determination of the mussel species, all of the observations on the food of mussels, and most of those upon the parasites. Dr. Wilson supervised the work and has furnished the geological and geographical distribution and most of the notes upon the various stations. GEOLOGY AND GEOGRAPHY OF THE MAUMEE AND UPPER WABASH BASINS. Although the Maumee River flows into Lake Erie and is therefore a part of the great St. Lawrence system, while the Wabash flows into the Ohio and thence to the Mississippi, the basins of the two rivers have been so intimately connected in their geological history and topographical features, as well as by the artificial connections established during the existence of the Wabash and Erie Canal, that they are practically one. Even at the present day the headwaters of the Little Wabash River approach within 3 miles of the St. Marys River and still nearer to small affluents of that stream, and when the white settlers first came to Fort Wayne they found this point an important portage for the native Indian tribes. There is no per- ceptible divide between the two basins, and a large open country ditch still unites them through an old flat lake plain known locally MUSSEL, FAUNA OF MAUMEE RIVER. 5 as "the' prairie." A careful examination of this ditch revealed no mussels at all, and it is not likely they could live in the mucky bottom, though it is possible that glochidia might be carried through the stream by migrating fishes. During the glacial period the entire region was buried under the ice, and every form of molluscan life was exterminated. At the melting of the ice sheet great bodies of water were formed which could find no outlet to the eastward on account of the still unmelted ice, and therefore all the water was compelled to drain toward the south and west into the Ohio and Mississippi Valleys. The old channels can still be traced and are represented on the geological maps of the region. The geologists tell us that at first there were two of these lakes entirely separated from each other. The eastern- most one. Lake Maumee, covered the present basin of the Maumee Eiver and the upper part of the Wabash Basin and was 100 feet higher than Lake Chicago, which lay to the west. Lake Maumee drained through the Maumee and Wabash Valleys into the Ohio, while Lake Chicago found an outlet along the Des Plaines and Illinois River Valleys. The restocking of the region with mussels must have been effected by migration through these respective outlets. At this period sev- eral of the more hardy species came up through the Wabash Basin into Lake Maumee, as is attested by the finding of their shells among the fossils of the period. At the beginning, therefore, the Maumee Basin was restocked with Ohio Valley species of mussels, which came to it by way of the Wabash River." After the ice had retreated far enough northward to reveal the surface features of the region under discussion, it exposed the valley of an ancient preglacial river, sometimes called the Huronian River, and now known as the Saginaw-Grand Valley. This river valley runs east and west across the southern part of the State of Michigan and became a channel connecting the two lakes just mentioned. The opening of this channel gradually lowered the water in Lake Maumee to the level of that in Lake Chicago. A series of terminal moraines which had formed along the southern margin of Lake Maumee were enabled by this lowering of the water to assert them- selves and close the Wabash outlet. These moraines are known to geologists as the Mississinewa Moraine, to the south and west of which lies the Mississinewa River, the Salamonie Moraine, to the north and east of the Salamonie River, the Wabash Moraine to the north and east of the main branch of the Wabash River, the St. Joseph-St. Marys IMoraine to the east of those two rivers, both of which drained originally into the Wabash, and the Defiance Moraine " For an interesting discussion of tliis subject, see Simpson, American Naturalist, vol. XXX, p. 379-384. 6 MUSSEL FAUNA OF MAUMEE RIVER. to the east of the Tiffin and Aiighiize Rivers. The former drainage of the St. Joseph and St. Marys Rivers through the "Wabash is in- dicated by the peculiar turn they make at Fort Wayne, where they unite to form the Maumee. This sudden change, of course, is so striking a feature that it has excited comment by abnost everyone who has considered the topography of the region. The situation is well expressed by Mr. Simpson who says: The St. Josephs, St. Marys, and Auglaize Rivers, tributaries of the Maumee, flow in the direction of the Wabash ; the two former join at Fort Wayne and flow partly bacliward as the Maumee, the whole looking like a tree with its branches broken down and hanging against its trunk." From the time of the closing up of the Wabash outlet until the St. Lawrence was reopened, the only outlet for Lake Maumee was through the Huronian channel into Lake Chicago, and thence into the Des Plaines and Illinois Rivers. All the mussels, therefore, which took part in restocking Lake INIaumee during this period belonged to the Illinois River fauna and entered the lake by this roundabout channel. The Unionida^ are essentially inhabitants of shallow water and would have spread along the margin of these glacial lakes. As the level of the lakes was gradually lowered and the present system of drainage was established, the mussels naturally followed the retreating waters, and thus finally found their way into the river beds where they now live. This account of the origin of the present mussel fauna in the Maumee and upper Wabash Basins is adapted from several geologi- cal reports of the States of Indiana and Michigan and from an excellent paper on the " Distribution of the Unionidse in Michigan " by Bryant Walker." From the foregoing account of the manner in which the stock- ing of the Maumee Basin took place, it would naturally be inferred that in the lists of species of mussels obtained at the various stations on the Maumee and Wabash Rivers by the party, there should be more than the usual amount of similarity between those from the headwaters of the two rivers, and this similarity should decrease the farther down the respective rivers the comparison is made. Such we find to be actually the case. (See p. 38.) But there is another factor which may at least claim a share of the credit for the correspondence in species. The Wabash and Erie Canal was in active operation between these two rivers for nearly half of the last century. Starting at Toledo, Ohio, it ran along the bank of the Maumee for the entire length of that river to Fort Wayne. There it crossed the narrow strip of country between the headwaters of the two rivers, ran along " American Naturalist, loc. cit. "Autlior's separate of a paper read before (lie Michigan Academy of Science, Mai'cli, 1899. MUSSEL FAUNA OF MAUMEE RIVER. 7 the eda^e of the old lake plain or " prairie," near the Indian portage and at no great distance from the Little Wabash, struck the main Wabash below Huntington, Ind., and then followed the latter river down to the Ohio. This canal was fed from the two rivers and opened into them at various points along its course. There was thus established a channel of communication which, during its half century of existence, must have resulted in the ex- change of many species of mussels, as well as other portions of the fauna of the two rivers. Such a canal furnishes an excellent breeding ground for mussels, as is well shown in the case of the old feeder canal at Fort Wayne (p. 19) and the Miami and Erie Canal still in existence between Defiance and Toledo (p. M). This artificial opportunity for interchange of species, occurring so recently, must have greatly supplemented the original similarity due to the former identity of the drainage basins. It would also go far toward destroying any local peculiarities that might have developed since the tw^o basins were separated. The Maumee Basin is situated in the northeastern part of Indiana and the northwestern part of Ohio, and covers an area of C,500 square miles. There are no large hills in or around it, it being a flat glacial plain, covered with moraines and low ridges. The basin abounds in clay and solid subsoil, so that the river is rather muddy most of the time. The water is also quite hard, furnishing an abun- dance of lime to the mussels for tlieir shells. In the upper part of the river, as far down as Defiance, Ohio, there is considerable sand and gravel bottom, but below Defiance a good deal of the bottom is solid bedrock. As already stated, the Maumee is a very young river in point of geological time, yet, next to the Detroit River, it is the largest tributary of Lake Erie and is one of the largest rivers within the State of Ohio. The Maumee is formed by the junction of the St. Joseph and St. Marys Rivers, just beneath the Columbia Street Bridge in Fort Wayne, and flows northeast to Lake Erie at Toledo, Ohio. From origin to mouth is about 100 miles in a straight line, but as a result of the angular or crescentic form of the glacial moraines which separate the Maumee from the Wabash Basin, the river winds greatly, especially that portion of it above Defiance, and thus adds fully 50 miles to its length. As might be expected, the current is not very swift except during high floods, since the river falls but 164 feet in its entire course, or a trifle over a foot a mile, and much of this is accumulated in a series of riffles at Defiance and Grand Rapids. Elsewhere the river is made up of long, sluggish stretches, alternating with short riffles of varying lengths, and it is in the vicinity of the latter that most of the mussels are found. 8 MUSSEL FAUNA OF MAUMEE RIVER. Through the removal of most of the forests which once covered the area about its headwaters, the current and depth of water are not so constant as they once were, but the river rises and falls more sud- denly, and is usually considerably more turbid or muddy than formerly. The height and activity of the river in times of flood are strikingly manifested by the tangled piles of drifted logs and brush up on the flood plain high above the normal surface of the river. This change in the steadiness of flow of the river, with the accom- paniment of a more shifting bottom, has probably had considerable eft'ect on the mussel life of the stream. f BIOLOGY OF THE MAUMEE RIVER. Through the abandonment of the Wabash and Erie Canal, which took place nearly 30 years ago, and the draining and gradual drying up of the old lake plain or prairie, the Maumee River has become isolated from the Wabash, so that there has been no transference of species for some time. In the course of our work our attention was devoted chiefly to the mussel fauna. The general features of the river and its inhab- itants, both plant and animal, undoubtedly have many important relations to the mussels, but, on account of the limited amount of time at our disposal and the impracticability of carrying on several lines of work at once, only such features were noted as appeared to have some evident bearing on mussel life. In the St. Marys River and about the region of Fort AVayne, where conditions were favor- able, the river was more thoroughly explored than elsewhere, the flora and fauna were noted, and the stomach contents and parasites of the mussels were examined. All along the way stomach contents and parasites were preserved for future study. A valuable addition to our knowledge of the biology of the Maumee Basin has been furnished by a report of the investigations made by Philip H. Kirsch during the summer of 1893 and published in the Bulletin of the United States Fish Commission (vol. xiv, for 1894). This report refers to an abundance of algae in various places, and furnishes a list of 87 species of fishes, 3 species of mussels in the Maumee, 4 species of crawfishes, 7 of batrachians, 6 of snakes, and 6 of turtles. In this report the remark is made that mussels are common." They are, however, only incidentally mentioned, so that " The mussels obtained by Kirsch during these investigations were later turned over to Mr. C. T. Simpson for examination. In an interesting discussion concerning the " Mis- sissippi Valley Unionidae found in tlie St. Lawrence and Atlantic drainage areas," in the American Naturalst, vol. xxx, p. 379-384, Mr. Simpson remarks of these shells : "This changing of characters" la number of changes have just been discussed] "has been well illustrated in a lot of Unionidae recently submitted for examination by Prof. B. W. Evermann, of the U. S. Pish Commission. * * * Unin lutcolus Lam., U. subrostratus Say, U. circulus Lea, l'. phaseolus Hild., U. multiplicatiis Lea, U. multira- diatus Lea, and Anodonta grandis Say, are so dwarfed and stunted and changed in color as to be scarcel.v recognizable, while the same species from tlie Wabash from which these have no doubt all been derived, are as vigorous and finely developed as any in the Mis sisslppi Valley." MUSSEL FAUNA OF MAUMEE EIVER. 9 Kirsch's report and the present one may be regarded as supplemen- tary to each other. The following general observations were made concerning the biology of the stream : Among the aquatic plants, Oscillatoria was abundant in the St. Marys Eiver and the upper portion of the Maumee down as far as Kern Reservation, 7 miles below Fort Wayne. In the St. Marys this plant was found floating in large blackish or blue-green bunches on the surface of the river; " farther down much of the floating Oscil- latona seemed recently washed in. Filamentous algae, probably Cladophora^ grew on the rocks in the bottom and occasionally on the shells of the mussels. Wild celeiy {Vallisneria spiralis) was abundant on the bottom of the St. Marys at Fort Wayne, but w^as- not conspicuous farther down. The dense-flowered water willow {Dianthera am-ericaiia) fringed the edges of the river and lower portions of the canal all along the way and was one of the plants most constantly seen. During the latter part of the cruise along the Miami and Erie Canal this plant was covered with large tangled clumps of dodder, so that the prospect either way along the canal was lined with great masses of yellow here and there. Algaj of many sorts gi-ew in great abundance in the Maumee below the dam at Defiance. Portions of the canal from Napoleon, Ohio, to the locks below were covered with a dense scum of delicate little w-ater fern, Azolla caroliniama. Along the lower stretches of the canal Cera- tophyUum was abundant in stretches of quiet water. Typha latifolia, the narrow-leaved cat-tail, formed a dense hedgelike growth along the edge of the canal in places. The spatterdock, Nymphaa advena^ was fairly common along the lower reaches of the canal and the white water lily, Castalia odorata^ occasional. Duckweeds were common in the canal, but not noted in the river. As to animal life, various species of protozoa were very abundant in one place in the St. Marys, as will be described later on. (See p. 13.) They were not noted farther down the river; however, there was little opportunity to examine carefully for them. No plankton was taken, but the material examined from the stomachs of mussels probably gives a fair notion of the nature of the smaller free plank- ton elements. Sponges and flatworms were rather common on rocks and shells the whole length of the river. In addition to the mussels, %arious mollusks such as Pleurocera^ Ancylus, and Sphrenurrh were common. Crawfishes were abundant everywhere in the river, but were not often noted in the canals. xVquatic insects and larvae were not " Further observations show that the presence of Oscillatoria and other minute organ- isms is a periodic phenomenon. They accumulate during periods of low, stagnant water but altogether disappear during periods of high water with rapid current. 10 MUSSEL FAUNA OF MAUMEE EIVEK, especially abundant, except below the clam at Defiance. According to residents of the region along the river, fishing was in general rather poor, although in a few places, as several miles above Defiance and below Defiance Dam, angling was enthusiastically carried on by men wading out in the stream in high boots and casting. In the autinnn of 1907 a good deal of angling was observed in the St. Joseph River just above its mouth, at Fort Wayne. The anglers there were catching a good number of the wall-eyed pike, or pike perch, Stisostedioit v'ltreum., which they called " pickerel.'" We did not see many fishes in the water during most of the cruise, although at Fort Wayne, Avhere we had opportunity to examine carefully, a number of species were observed, and all along the Maumee the small-mouthed black bass seemed fairly common. The darting fishes frequently struck the boat with resounding whacks and occasionally jumped over the boat or even over our heads as we sat in the boat. They frequently jumped against different members of the party and fell back into the water, and G examples, most of them quite large and fine, jumped into the boat and were captured. At one place, a little above Woodburn, Ind., we passed through a large school of basking gars. Among a(iuatic birds and waders, kingfishers {Ceryle alcyon) were fairly conunon, green herons {Ardea vlresceiw) and the great blue heron (Ardea herodias) were quite abundant, the former along shore or on branches of trees and the latter standing statuesquely on rocks in midstream and rising with a low level flight as we approached. We saw several small flocks of wood ducks {Aix sponsa) which seem to be summer residents. Sandpipers were common along shore and flocks of killdeer occasional on bars. On account of the northern lati- tude it is perhaps worthy of note that the song of the Carolina wren came to us frequently from the woodlands along the upper stretches of the river. No aquatic mammals were seen, although occasional shell piles gave evidence of the presence of a few muskrats. MUSSEL SURVEY. HEADWATERS OF MAUMEE AND LITTLE WABASH BASINS. The region in and around Fort Wayne is of peculiar interest, since it is the borderland between the two river basins, and also furnishes much evidence of the influence of the old Wabash and Erie Canal upon the mussel fauna. For this reason, considerable time was spent here before beginning the specific work upon the basins themselves. Most of the localities reported upon were worked over several times MUSSEL FAUNA OF MAUMEE RIVER. 11 in order tliat there might be a reasonable certainty that all the resi- dent species had been found. A careful study was made of the species themselves and data collected as regards food, parasites, etc. The localities examined include the St. Marys and St. Joseph Eivers, Spy Run, a small tributary of St. Marys, and an old feeder canal and its reservoir which was formerly connected with the St. Joseph Eiver. The Aboit and Little Wabash Rivers, which belong to the AVabash system, were also examined. The stations at these localities are designated by lettei's, those on the main river by numbers. The rela- tive abundance of the different species at each station is indicated by giving the niunber of examples obtained, or by remarks. ST. MARYS RIVER AND SPY RUN. In the St. Marys River the shells are scattered irregularly eveiy- where, so that along the bank a few dead shells may be found every now and then, or an occasional live mussel seen in the water. There are a few shell beds, all of small size, however, where conditions appear to have been exceptionally favorable for the development of molluscan life, and at such places the mussels are fairly abundant. Station A. Above the Blujfton Road Bridge, at Fort Wayne. — The highest point at which the St. Marys River was examined was above the Blutfton Road Bridge, about 4 miles from the mouth of the river. At this place there were riffles and the remains of an old dam, and the point lies above the entrance of most of the sewage of the city. The water was shallow and the bottom covered with gravel, sand, or mud, according to the rapidity of the current. At the date examined (June 26) the water was quite turbid and nothing could be seen on the bottom. We had heard reports of a goodly number of shells there, but by wading about considerably we found only a few. The living mussels were hard to find, partly because of the stony bottom and partly because of their actual scarcity. A man fishing at the place was offering boys who were swimming near by 5 cents apiece for mussels, which he wished to use for bait, and it was noteworthy that although the boys were searching diligently the fisherman was not oversupplied. We found a fair number of dead shells on the shore, representing 18 species. The little Truncilla sulcata, which seemed on the whole a rather rare shell, and of which during the entire summer's work we found only 1 live example (although scattered valves were occasionally found along the river bank), seemed unusually common here; we obtained 4 entire shells and 10 valves. None of the larger species of mussels were present in sufficient numbers to be of commercial value. 12 MUSSEL FAUNA OF MAUMEE EIVER. while all of the others were either too small or so discolored as to be unfit for use. The river was followed from this point down to Swinney Park, a distance of about a mile and a half, but practically no shells were seen, and men wading in the river fishing for crawfishes with dip nets knew of none. From Swinney Park down to the riffles, soon to be discussed, the river had been examined fairly well during the previous year, and no shell beds of any importance had been found. Occasionally dead shells had been seen scattered along shore and in one place a few large Lampsilis ligamentinus, which had probably been killed by fishermen for bait. Station B. Rifjles^ St. Marys River helow the Van Buren Street Bridge. — The most important bed found in the St. Marys River was at the riffles, about an eighth of a mile below the Van Buren Street Bridge and across from the city waterworks pumping station. The riffles are conveniently near the residence of Miss Elizabeth J. Bowman, at 719 Michael Avenue, who very kindly gave us the use of her house and grounds as a laboratory headquarters for the work in and around Fort Wayne. Being so favorably situated, this station was repeatedly visited and the mussels obtained were carefully exam- ined for food, parasites forming pearls, etc. As the opportunity for doing laboratory work is rather rare in moving field parties, and as it was supposed that this mussel bed would be fairly typical of others we would encounter but would not have opportunity to study fully, this bed was investigated and the observations noted in considerable detail. The stretch of river above the mussel bed was about 8 feet deep at its greatest depth and 60 or 80 feet wide, with a sandy bottom, and was used by the boys as a swimming-hole. A large quantity of sand for building purposes had been hauled out of the river at this place and screened some little distance from the river bank. Among the screenings was a fair number of mussel shells of different species and shells of Pleurocera and S phwriuTn. The species of mus- sels represented by the shells among the screenings were Qnachvda ruhiginosa, Q. undulata., Pleurohema clava, TJnio gihhostis, Sym- fhynota complanata^ Anodonta grandis, Lampsilis gracilis, L. rectus, L. ligamentinus, and L. luteolus. There were not many live mussels to be found in this stretch of the river, probably on account of dis- turbances caused by the hauling of sand. There were also few found below the riffles, as the stones were too thick in the river bed to furnish any foothold. The shell bed itself was therefore on a slope between the shallower portion of the riffle and the deeper sandy bottom upstream. The bottom on which the MUSSEL FAUNA OF MAUMEE EIVEE. 13 mussels were located was composed of mingled sand and pebbles. The water over the mussel bed was shallow, averaging about 2 feet in depth during the period of observation, and the current was mod- erate, 3 or 3^ miles per hour. The water varies considerably in depth and turbidity at diiferent times; even a small local shower raises the surface perceptibly and makes the Avater very muddy, after wliich it slowly abates in volume and becomes clear. While this mussel lied Avas under observation it was occasionally rendered unworkable for a few days by local showers. On the other hand, the water was sometimes found clear enough to see the bottom with a fair amoimt of distinctness. The life conditions of the rifHes and the swimming hole just above it were studied in considerable detail. The river flora did not appear to be particularly rich. The only conspicuous alga was a species of OsciUatoi'ia, which was very abundant on the surface in large floating blue-black patches from which the filaments radiated in every direc- tion. Of phanerogams, the only aquatic plants growing in the river at this place were eelgrass or wild celery {ValUsneria spiralis) and the dense-flowered water willow {Dianfhera americana). The animal life was rich and rather varied. One of the most abundant animals was a V 07'ticella-\\kQ protozoan, probably Epistylis^ which thickly covered with a uniform hoary coating the blades of ValUsneria and the branches and leaves of willow which dipped into the water. The same or a similar protozoan grew in great abundance on the crawfishes of all sizes, especially thickly on the antennae, which were rendered quite white and plumose. Examples of the protozoan taken from the crawfishes showed the zooids thickly clus- tered at the apex of a long nonretractile dichotomously branching stem. On account of the nonretractility of the stem, this white coat- ing can not contract into a mass wdien irritated as can colonies of V orticella. A small colony of Vorticella w^as found clustered on an Atax taken out of a mussel. A species of fresh-water sponge was very common below the shell bed, frequently coating the inside of dead valves with an expanded thick feltlike covering. The brown hydra [Hydra fmca) was quite abundant, and numerous examples were seen attached to the back of shells of Ancylus. Leeches did not seem to be abundant ; the only one noted was a fish leech attached to a sunfish caught in the river. In addition to the mussels, discussion of which wnll be reserved till the end of the description of the locality, various univalves such as Pleurocera^ Physa, and Ancylus were common. The crawfishes were so abundant at this place as to deserve special mention. So far as examined the species was Camharus virilis 14 MUSSEL FAUNA OF MAUMEE RIVER, Hagen, and the river bottom appeared to be almost overpopulated with them. The young were often seen crowding up close to the shore. As noticed above, many were thickly covered with a growth of protozoa, and some of the young were covered with a green algal growth. In the mussel bed the crawfishes were seen sunning them- selves on the rocks which projected out of the water, lifting them- selves high upon the pincers in a most peculiar fashion. Upon wading out to a large projecting rock situated in the river about in the middle of the mussel bed they were found to be clustered thickly on the sloping surface of the rock just at the water line, their heads projecting out of the water. Many were found living in dead mussel shells just below the bed; indeed, it was hardly possible to pick an old shell or tin can out of the water without obtaining a crawfish. Some of those thus captured were of very large size. The crawfishes were frequently observed devouring the flesh of dead mussels, but it is doubtful whether they were able to kill any, especially the larger ones. They may possibly be an intermediate host of some of the mussel parasites. The crawfishes were caught in considerable num- bers, either by dip nets or minnow seines, for fish bait and were used in the St. Joseph Eiver. On July 2 a man was observed seining crawfishes just below the mussel bed at the riffles. The seine was drawn over the bottom, which was covered with rounded pebbles about the size of hen's eggs or larger, and the ends of the brails poked about among the rocks. The crawfishes, of which it seemed there was one under nearly every stone, backed up into the net and were caught in great numbers. The fisherman wanted only the soft- shells and ''peelers" (crawfishes about to molt) and threw the rest back. A great number were molting at the time. The " peelers " can be easily distinguished from other crawfishes by pressing the sides of the carapace, which arches up free from the body. They also have a somewhat different color. The molted carapaces of crawfishes and gastroliths associated with them were quite common along the shore of the river. No crustaceans except crawfishes Avere noted in the river itself, though small Entomostraca were abundant at the edge of the river, in overflow ]">o()ls. Aquatic insects and larvje did not appear to be particularly abund- ant so far as species were concerned; a few forms, however, were quite common. The red "blood worm" (a Chtronom'us larva) was present in large numbers, and along shore, just under the water surface and attached to submerged objects by a short slender pedicel, was a great number of small gelatinous pear-shaped o])jects full of minute dots, probably Qgg bunches of Ghironomus. Midges were abundant, emerging in great numbers; on the morning of June 21 MUSSEL FAUNA OF MAUMEE RIVER. 15 there were many exiivite on the water surface and on objects along shore. Many of the insects died shortly after emerging. On the first visit to the swimming-hole, the afternoon of June 14, a large number of johnny darters {Boleosoma nigrum) and a few minnows were observed dead along shore. All the fishes were quite small, apparently the young of the year. It is very likely that some sudden pollution of the river about that time Avas the cause of the death of the fishes as well as that of a great number of large craw- fishes on shore and mussels in the water. Some time during the sum-] mer it was reported that dyestuff had been poured from the Imitting mills some distance above into the river and had caused a wholesale destruction of fishes, which died in such numbers as to render por- tions of the city along the river scarcely habitable. In addition to the dead fishes mentioned above, the following species were observed : Catfishes (young, probably Ameiurus natalls), mad toms {Schil- heodes gyrinus), golden shiners {Ahramis crysoleucas)^ carp {Gypri- nus carpio)^ and sunfishes {Eupomotis gibhosus). In the previous year a number of small black suckers {Catostomus nigricans) were seen at this place. Of reptiles, a few painted turtles {Ghrysemls marginata) were found. There Avere no indications of aquatic mammals, such as tracks or piles of shells along the shore. Late in autumn, however, a good many crawfishes were seen along shore with backs bitten open, and this may have been done by a muskrat or a mink. As has been said, although a few mussels could be found above the main bed, the greater number were to be obtained just at the head of the riffles in a fairly compact bed. The mussel bed itself is quite small, probably not more than 2,000 square feet in extent, and commercial operations would clean it out in a very short time. On several occasions the water was clear enough to see the bottom and the distribution of the mussels fairly well. The mussels did not appear to be moving. They were headed mostly downstream, the natural position for feeding nuissels, and were not deeply buried; about half of the shell projected out of the mud. The projecting portion was more or less stained and frequently covered Avith an algal growth. On June 20 this bed was gone over with considerable thorough- ness and a number of mussels collected. These were placed on shore, counted, and sorted over, and, after a number of representative shells such as Avere desired for specimens Avere picked out, the rest were thrown back. All the shells, especially L. rectus and L. liga- mentinus, were large and fine, A number of the living animals were measured and Aveighed, Avith the results shoAvn on the next page. 22367"— 12 2 16 MUSSEL FAUNA OF MAUMEE EIVER. Measurements and AV eights of St. Marys River Mussels. Serial No. Species. Length. Height. Thick- ness. Weight. Quadrula undulata do do do do do do Lampsilis rectus '.[[[do'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'.'. do do do do Lampsilis venlricosus... do do Lampsilis ligamentinus. do do do do do do do do do do do do Lampsilis luteolus. . . do do.. do do do do do. Symphynota costata. do do do do do do mm. 115 104 102 8(5 80 70 72 158 159 134 150 153 141 154 126 110 83 137 135 145 135 131 132 130 126 131 122 115 97 120 122 124 127 134 127 126 118 102 92 123 95 111 113 95 111 101 Ounces. 10.75 8.25 8.50 4.75 3.75 3.50 3.00 13.75 14.50 7.75 13.75 12.75 12. 50 8. 75 13.75 8.25 4.75 16.50 15.50 18.25 15.75 14.75 13.75 11.75 12.00 12. 75 12. 25 10.75 5.25 11.75 10.00 9.25 10.75 11.00 11.75 10.75 9.25 6.50 4.00 6.75 3.25 4.75 5.75 3.00 5.25 3.50 BeloAv the mussel bed and also in it were a large number of dead shells. These were covered with water and appeared to be still as good for commercial purposes as ever, since they retained their original texture and luster. Some of the shells that had been longest dead were stained black, perhaps from iron in the water, but this color was superficial and could have been easily ground off. The mussels appeared at the time of our visit to be dying in some num- bers, for a good many still retained the position they had during life and the flesh was still clinging to the interior of the shell. Some of the larger living shells were diseased and stained brown or green in the neighborhood of the cardinal teeth. The stomachs usually contained considerable mud, in which was scattered a few diatoms, ScenedesTnus, Phacus^ etc.; the amount of organic material was quite small. The results of studies of stomach MUSSEL/ FAUNA OF MAUMEE RIVER. 17 contents are given in detail under the general heading at the close of this report. (See p. 57.) When examined in June these mussels were strikingly barren. The only gi'avid individual noted was an example of Lampsilis ventrivosus with the shell of a marked female contour very much inflated, and this was partly barren, the gills appearing shrunken. Subsequent investigations show^ed that the sterility of this bed was due partly to the season. As w^e went downstream and the season advanced we noticed gills of various species beginning to take on a dusky tinge at the margin and filling up with fertilized eggs and glochidia. Moreover, on a visit to this mussel bed on September 21, of 28 examples of L. ligamentinus procured 8 were gravid. The other species procured in September were : Qnadrula tuherculata^ 1 ; Q. ruhigiriosa, 1 ; Q. uridulata^ 3 ; Symphynota complanata^ 1 ; LampsiUs 7'ectus, 3; L. luteohis, 2; none gravid. The percentage of breeding mussels was hardly as great as one might expect under favorable conditions. No small shells were found, the smallest being of medium size and apparently several years old. The conditions on the whole seem to indicate that this mussel bed is dying out. It would have been possible to secure a number, perhaps a half ton, of good shells here without disturbing the mussels. The riffles were incidentally visited in November, and although the water w^as quite cold the protozoa were still abundant on craw- fishes and dead leaves in the water, and the conditions bore con- siderable resemblance to those observed earlier in the year." None of the mussels was badly parasitized.^ Atax ypsilophorus, Aspidogaster conchicola^ and Cotylaspis Insignis were rather com- mon, but not abundant. The marginal cyst distomid was rather frequent, and it was here Ave obtained material with which we were able to associate this organism with jDcarl formation. It was most common in LampsiUs ligamentinus. From the riffles down to the mouth of the river mussels appeared to be quite scarce. " On Juno 20, 1000, as opportunity permitted, this place was revisitod. It was durin.i; a long period of high water. The river, though turbid, was not filthy and appeared reasonably clean. At the riffles it was racing at a rapid rate and the current could hardly be stemmed when about breast-high, but would sweep one off his feet. No Pro- tozoa were seen, and only a few young crawfishes, which were very active and alert. A few gray-back minnows were swimming about on the surface, and schools of small minnows darted from shore. By wading about a great deal only two live mussels (L. ligamentinus) were found, though there were a number of dead shells. There were no evidences of unhealthy conditions or overcrowding ; on the contrary, the conditions seemed to be those of an ideal mussel stream. This visit served to emphasize the fact that during periods of high water sewage is comparatively innocuous and that while the water of the river has great variation the river population is fairly constant ; In periods of low water the stream is overpopulated and, to use a chemical phrase, the in- habitants are almost precipitated out. * The parasites are more fully discussed and described at the end of the report. 18 MUSSEL FAUNA OF MAUMEE EIVER. Station C. Spy Run. — Spy Rim is the name of a small shallow stream which enters the St. Mary's about a half mile above its mouth. The bottom is of fine sand and the water at the time visited was quite clear. There were many small fishes, chiefly the johnny darter {Boleosoma nigrmiv)^ a few golden shiners {Abramis crysoleucas) and chubs (Semotilus atromaculatus) , and a number of crawfishes, all free from Protozoa. No living mussels were found, but about a dozen dead shells, all of which, except fragments of one LampsiUs iris, were Anodontoides feiv/ssaeianvs snheylindraceus. ST. JOSEPH RIVER AND FEEDER t'AXAL. The St. Joseph River, more commonly known by the shortened local name St. Joe, is a larger and clearer stream than the St. Marys, and since only the very lowest portion of its course enters the city, it is much freer from sewage. Station D. St. Joseph River at Robinson Park., 6 miles north of Fort Wayne. — The highest point at which the St. Joseph River was examined was at Robinson Park, a pleasure resort about 6 miles above the city. This point could easily be reached by trolley. Con- siderable change had taken place in the river here within the last few years. At the time the park was established there was a dam in the river which had kept the feeder canal, just above, full of water. This had raised the water of the river and had expanded it somewhat into a quiet, pondlike basin. The breaking of the dam had reduced the water to its nornud height and emptied the canal. At the time of our visit the river was rather high and turbid from recent rains. We procured a boat and rake at the park and rowed up to the riffles about half a mile above the park, examining the river and shore. There was a fair number of shells scattered along here, all of which were large and of excellent luster. The large shells found at the park itself had been dead some time, and may have been killed by the lowering of the water when the dam broke; the LampsiUs hfteolus were exceptionally large and fine. Farther up the stream large dead shells were collected on the Inink, chiefly Quadrida undn- Jata. They were freshly dead and had probably been killed for bait. On investigating the river it was found full of snags in some places; in others the bottom was of hard clay, too compact for mussels to live in. In the beds of. gravel where mussels were likely to be found the bottom was full of sharp rocks wdiich in the turbid water were difficult to distinguish from mussels, so that collecting the latter was uncertain. We found very few live shells, but were told that they were easy to obtain when the water was low and clear. MUSSEL FAUNA OF MAUMEE RIVER. 19 The manager of the boathouse at Robinson Park told ns of having seen a good bed of mussels here during clear low water. He also reported that where he used to live on the Ohio River they ate the " small sweet mussel," but the larger kinds were too tough for food, except that they could be used for flavoring soup. Station E. Feeder canal from Robinson Park to Fort Wayne. — In the days of the Wabash and Erie Canal, the dam mentioned above was built across the St. Joseph River just below what is now Robin- son Park, and a feeder canal was dug to convey the water from above the dam to the main canal at Fort Wayne. After the main canal had fallen into disuse this feeder was still used as a source of power for mills at the outskirts of the city, and the water was turned back into the St. Joseph River. AMien Robinson Park was established the canal formed a charm- ing water lane leading from the city out to it. About the summer of 1906 or 1907, however, the dam in the river broke and let the water out of the canal. The dry canal bed offered one of the most remarkable opportuni- ties possible for the study of mussel distribution within a small area. It was 6 miles in length and, with the exception of a few places be- low the general level, it dried up so quickly that the mussels are still left in their original positions in the mud on the bottom of the canal. The shells are half to two-thirds buried in the dried mud, the great majority of them with the posterior end directed northward — that is. against the current when the canal was full of water — and with the valves nearly closed, just as the animals died. This region had been visited and preliminary studies made in the autumn of 1907, and it presented the same features that it did early in the summer of 1908 — a long, dry, and cracked mud flat forming a vista of projecting shells, the short tracks the mussels had made during life being still distinct. In the canal bed it was possible to study all the species that had lived there, their abundance, distribu- tion, and the like. Many of the young shells of Lampsilis luteolus exhibited brilliant rays; they appear to have been exceptionally highly colored. Toward the upper end of the canal, in a place where the bottom was 15 feet wide, the mussels were counted for a stretch of 10 feet along the canal bed and the following species noted : Quadrula riib'i- ginosa^ 11; ^- cylindrical 1; Q. vnchdata^ 86; Anodonta grandis., 6; Ptychohranchus jdiaseolus., 1 ; Lampsilis ligamentinus, 5 ; L. luteolus., 6. The width taken was the total width of the bottom of the canal and was considerably wider than the space occupied by the mussels. About a mile farther down the canal a space 10 feet square was measured off in the bottom of the canal and the following species 20 MUSSEL FAUNA OF MAUMEE RIVER. were found: Quadi'ula riibigmom^ 6; Q. undiilata-^ 60, all rather small; Pleuroheina clava^ 1; Alasmidonta tnincata^ 2; Symphynota com'planata, 2; S. costata^ 5; Anodonta grandis, 15; Ohovaria eir- eulus, 4; Lampsilis ligamentinus, 5; L. luteolus, 1; L. ventricosus^ 4. This gave a little over 1 shell per square foot. In 1908, in a square meter of bottom near the Eod and Gun Club, the following sf)ecies were noted: QuadruJa nihiginosa^ 9; Q. undulata, 36; Symphyiwta complanata^ 1; Anodonta grandis^ 17; Ohovarla circulus^ 11; Lamp- silis iris, 2; L. ligamentimbs, 2; L. luteolus, 3, giving a total of 81 per square meter. In addition to these shells there were many small Sphaniums, the ground being paved with them, 34 Campelomas, and 23 Pleuroce7'as. The square meter referred to above represents, as nearly as could be judged, an average number rather than either extreme. The general impression obtained concerning the abundance of vari- ous species is given in the table on page 37. Figure 2, plate i, represents a portion of the dried canal bed near its upper end. It shows admirably the mud cracks formed on drying and the mussel shells still protruding in situ. Realizing the value of an opportunity to study so extensive a tract with the mussels still in their original positions and in plain sight, several visits were made to the canal, and as a result the following conclusions were reached : 1. Each of the species found was well distributed throughout the entire canal; that is, there w^as no marked gathering together or colonizing of any one species to the exclusion of others. So fully was this true that a careful examination of a few rods of the canal bed anywhere was reasonably certain to yield specimens of all the species, except possibly one or two that were very rare. In this connection it must be kept in mind that the bed of the canal was very different from that of a river, in that it was practically the same throughout its entire length. It was all shallow mud, with little sand or gravel; the water was of nearly uniform depth, and the current was of uni- form rapidity. In short, there was almost nothing except the con- sistency of the mud upon which a mussel could base any preference of locality. Slight differences in this respect caused an increase in the numbers of mussels in some places, but affected all the species equally. 2. The number of mussels which can find accommodation under fairly favorable conditions upon a definite area in the canal bottom may be judged by the counts on measured areas given above. 3. A canal makes an excellent breeding ground for mussels, as was particularly attested by the presence of a large number of younger examples among the adults, showing all stages of growth. The cur- rent in the canal appears to have been just fast enough^ food material U. S. B. F.— Doc. 757. Plate 1. Fig. 1. — Bed of Maumee River just below Columbia Street bridge, Fort Wayne, Ind., showing mussels killed by sewage from city gasworks. ■^■il^C^K^^ liT* . S^ -*x' Fig. 2. — The dried bed of the Feeder Canal, showing mud cracks and mussel shells in situ. r^'^^lSy*^ Fig. 3.— Pile of shells, chiefly m u c kets . Lampsilis ligamen - tinus, containing 25 tons, gathered fromthe Maumee River near Fort Wayne, Ind. MUSSEL, FAUNA OF MAUMEE RIVER. 21 constant and ample, the supply of water uniform the year round, and the bottom seems to have been well suited to a great variety of species. This gives the mussels a uniformly large size, while the shells at the same time are of extra quality and luster. 4. These facts would indicate that a canal would furnish one of the best locations that could be obtained for artificial propagation of mussels. Furthermore, the water in a canal can be drawn down from time to time without much trouble, thus allowing examination of the mussels. Examples suitable for breeding or for commercial purposes could then be selected easily and rapidly without in any way disturb- ing the others. Such possibilities are at least worthy of a careful consideration in future attempts at the artificial propagation of mussels. The last visit to the canal was made on July 27, after a series of heavy rains. The canal bed was at this time under water for nearly its entire length, and w^as covered w^ith grass and weeds that had sprung up in a short time and effectually concealed the shells. This visit threw light on a phenomenon which had proved quite puzzling during previous visits. Occasionally, shells of Qvadrula undulata, only recently dead and with bits of flesh still clinging to them, had been found in dry places in the canal l)ed some distance from any jjool. At the time it was hard to understand how they came to be there, as it seemed impossible that they could have " mudded up " and survived the long period following the breaking of the dam. This visit showed that during the high water of winter or spring the whole canal bed had probably been submerged, and that the mussels had migrated from the deeper portions at that time. During our previous visits, also, a number of pools, comprising the deeper portions of the canal bed, were observed, and a few live mus- sels, chiefly Anodonta grandis and Qnadrula undidata, had been noted near the edges, but the pools were not fully investigated. On this last visit the pools were thoroughly investigated and were found to contain an abundance of moUuscan life. Fine examples of the folloAving species of mussels were secured: Qnadrula coccinea, 1; Q. ruhiginosa, many, deeply buried; Q. ci/lhidrica., 1; Q. nnduhita, many; AlasmMonta truncata^ 1; Symphynota costata^ several; Ano- donta grandis^ a few; Stt'ophitus edentalus, a few; Ptychohranchus phaseolus, 1; Obovai^i circulus, 1; Lampsilis iris, 1; L. rectus, 2; L. Ugamentinus, many, large and fine; L. luteolus, several; and L. ventricofius, 1, large. Among other mollusks, Ancylus was abundant and large, usually attached to the shells of living mussels. Algae of various species were common, covering the water surface. In one place Hydrodlctyon was exceedingly abundant. This investigation w^as interesting and important, as it showed that mussels will live for long periods in shallow pools where they once 22 MUSSEL FAUNA OF MAUMEE EIVER. have obtained foothold, even though the conditions under which they once throve and muhiplied have ceased. While it is true that mus- sels die rather quickly when placed in foul water or unfavorable conditions, both our investigations here and our experiments in trans- planting elsewhere indicate that they can live for long periods where they probably would not naturally develop, and that they can easily be stored in jdooIs of limited size. Station F. Reservoir of the feeder canal. — This is an artificial pondlike area at the lower end of the canal and within the outskirts of the city of Fort Wayne. This reservoir, which originally con- nected with the canal, covers about an acre, and the water, which at the time of our visit was only a foot deep at most, probably never gets much deeper, as it can easily drain into the river below. The bottom of the reservoir is of a bare, yellowish clay, and there is a great deal of miscellaneous trash over it, indicating that it has been used to some extent as a sort of dumping ground. There were quite a number of mussels in this pond. Quadrula rubighiosa was com- mon in the warm shallow water near shore, and on the date of our first visit (June 30) was remarkably active, almost every example being at the end of a long, curved track. They were quite high up out of the mud, and several had fallen over and were lying on their sides. Quaditda undidata and Anodonta grandis were also common, and there were a few Lampsilis luteolus. Some dead L. glans and one dead L. venirlcosus, which quite closely resembled L. eapax^ were found along shore. More very young mussels were seen here than anywhere else. They were at the end of long, narrow tracks resem- bling snail tracks, in the shallow water along the north shore of the pond, and were nearly buried in the mud, only the posterior tip of the shell slightly projecting. Their presence indicates that the reser- voir was used quite actively as a breeding ground for the mussels. There were a fcAv fishes in the reservoir, chiefly carp and suckers, and boys were occasionally seen fishing there. Other mollusks, especially large Camfelorrbas.^ were abundant. The mussels of this place, especially the Anodontas, were more heavily infested with parasites than those obtained anywhere else, with the exception of some Lampsilis alatus taken at Grand Rapids. Ohio, which contained Atax in great numbers. The common mus- sel parasites, At(u\ Cotylaspis, and As pidog aster ^ were present in nearly every individual examined. The reservoir was again visited on July 23, and it w^as found that the Quadrula ruliginosa, which on the previous visit were high up out of the mud and actively moving about, had by this time buried themselves deeply. It was also noticed that at the same date, in the pools of the canal itself, they were likewise deeply buried. It is probable that their great activity during our first visit vv^as due to MUSSEL FAUNA OF MAUMEE RIVER. 23 restlessness caused by the warmth of the shaHow water and that their action of deeply burying themselves was in anticipation of the dry- ing up of the reservoir. A number of the Anodontas obtained at this time were examined as to food and parasites, with the following results : No. 1. — Parnsites: Ala.r, 3, in gi.ll axils; Cotylastpifi infiiffnis, 1, in axil of Inner gills; Aspidoguf t; c^ (N.-H ^"a^ ^ : qj o ® O fa fa \<^^ fa fa fa Ofa fa ■ i "i -^ -^ ^ ps"^ a" a tD 0} a> a> o o fa fa fa fa O O t- ^IN ■'° ^§B^ a i - a>"a ;a" : p; .2 T3 > ^ s"^ ■J. "li |S isa?^& fa 0) 01 fafafa fa fa fa 0) o ® » faofafa ^ OJC^ rt o>-< 00 crarH(Mooo)cort t-H t>- —1 iM •-:> rt Or-H CO M t^ CJ rt -0-. to O pq a a a i s . J4' +-> Ml *^ >> ^ : :S ^ ^ 08 5? C3 X! 1 i ■£-2 £ t p, ■ft |5 -2 3 2 -y 1 5 --ra P.V C3 .aSi.2 «^ 3 3X hP.o.B.p,c aaaat 1 3 C3 3 II a c 3 3^ c3 c3 J 1^ = r E 0. ""1 .2S&,- 03 o i o caC ill ^-a O r D i 1 i 3 ft£c } p.-g -2 > 'Ell F ■I F j ! 1 ^- C i-a ^c? G ^d- s: < W a ^^•" < u 'fr p. c ,_ 3 a ! 1- J c- c3 CI i-:!h. J c 'i i- 88 Mussel fauna op maumee eivee. W 3 MUSSEL FAUNA OF MAUMEE RIVER. S^ DISCUSSION OF MUSSEL SPECIES. 1. Qitadnda fuherculata (Rafinc^^qne). Purple warty-back. This is not a very common species of mussel in the Maumee or its tribu- taries, only a few occasional examples having been taken. It was most abundant in the upper portions of the river, near Fort Wayne, and down as far as Maumee Center Bridge. No living examples were obtained below Defiance, Ohio, and only a very few dead shells were picked up below here, the last one being obtained at Waterville, Ohio. All specimens seen were kept, and only about 52 were obtained in the Maumee system altogether. The range of this species is given by Simpson" as "Mississippi drainage generally; southern Michigan; San Saba County, central Texas." While it is in the main a Missis- sippi form, it is one of those species which occur occasionally within the St. Lawrence Basin, and although not usually a lake-dwelling species a few old valves w^ere found along the shore of Put-in Bay, The Maumee examples show very little variation. Although most of them were rather small, some of the shells seen at the clammer's camp near Fort Wayne w^ere quite large. The few mussels of this species obtained alive usually contained small baroques in the dorsal region. This mussel is of no value for buttons on account of its color. In addition to being purple, the luster is usually dull. The epidermis, which is commonlv brown, is somewhat tinged with green in examples from Maumee Center Bridge. In some places the distomid of Kelly, associated with dorsal baroques, is a common parasite. The examples found in the Maumee River, however, are comparatively free froifi this parasite. 2. Quadrula coccinea (Conrad). This species is not as common in the Maumee Basin as Q. tvhercvlata^ only about 31 specimens having been obtained. It was more abundant in the upper course of the river, but dwindled out and finally disappeared; the greatest number (10) was obtained at Kern's Reservation, not far below Fort Wayne, and the last one was taken at Florida, Ohio, in the ritlles below the bridge. It is a species common to both the Mississippi and the St. Lawrence drainage. The larger number of examples seen were white nacred, though a few were rosy. The larger white-nacred specimens would make excellent button shells, classing with the pig-toes, though considerably thinner. 3. Quadrula ruhiginosa (Lea). Quadrula rubiginosa is one of the most widely distributed of our fresh-water shells and occurs in a great variety of situations; it is found in lakes as well as in rivers. It is fairly common all along the Maumee and its tributaries (except the Tiffin, which was not thoroughly examined and in which few " Proceedings U. S. National Museum, vol. xxii^ p. 795. 40 MUSSEL FAUNA OF MAUMEE RIVEB. slie41s of any species were found). It was very abundant in the feeder canal and in its reservoir. A number of shells of this species were also found on the shore of Put-in Bay Island, Lake Erie, Although during the summer's collection we obtained material exhibiting considerable variation, all examples found within the Maumee Basin were fairly constant in form, though some were short and humjDed. This species very closely resembles Quadrula cocci nea, the only apparent ditierence being in the prominence of the ridge on the posterior umbonal slope; hence some specimens are difficult to classify. Young shells of this species exhibit faint rays which disappear with age. The specimens obtained in the reservoir dilfered from most of the others in having a very satiny epidermis. The greater number of the shells have a white nacre, although in some cases it is yellowish or rosy. None of our examples are as rosy, however, as Q. cocc'inea frequently becomes. The flesh of this species varies from salmon or orange to deep red, the richness of its color probably vary- ing with the nearness to the breeding season and the abundance of ova present; these latter are of a bright carmine color. Some speci- mens were found becoming gravid and one nearly ripe in the feeder canal July 27. On account of its commonness not all of the shells seen were collected; however, we obtained 112 examples from the Maumee Basin. What seemed a peculiarity in the action of this mussel was the habit of burying itself in the early summer. On a visit to the reservoir of the feeder canal on June 30, the mussels were all pro- jecting far out of the mud and actively moving about, while on July 23, at the same place, they were buried quite deep, and hard to find. They were also deeply buried in the feeder canal pools about the same date. A probable explanation of this peculiar action is that the reservoir was quite shallow and the water became warm. The water becomes low during drouths, and the mussels may have gone down into the mud in anticipation of the pool going dry. This species is sometimes more or less affected with Atax^ though not badly. Those found in the reservoir were more affected than others. Although its small size is something of a disadvantage, this species would make a fair button shell. It has good luster, and its compara- tive flatness would permit its being worked up economically. Though there are a number of river shells superior to it, it is one of the best if not the best button species that would thrive well in ponds. A considerable number were weighed and measured, and the largest, obtained at the Defiance Dam, was 88 millimeters high by 306 millimeters long, and the dry shell weighed 101 grams. MUSSEL FAUN^A OF MAUMEE RIVER. 41 4. Quadnda piiHtulo.sa (Lea). Pimple-back: '.varty-back. Qiiad- Tula pustulosa is another of the widely distributed mussels common to both the Mississippi and St. Lawrence drainage systems; but, although thus widely distributed, it is somewhat choice of its habitat. It shows a distinct preference for more or less sandy or gravelly bars with a live current. It does not usually form beds, but is gen- erally more or less mixed in with other species. It was found throughout the length of the river, but most abundantly in the cen- tral portion. A few examples, some of them of good size, were found at Fort Wayne, especially in the feeder canal. The greatest number taken at one place was at Kern's Reservation, where -tC were obtained in a shallow bottom of sandy gravel. All these examples, and, indeed, nearly all found in the Manmee until we reached the Auglaize River, were small, dwarfed shells; those found in the Auglaize River and near the dam at Grand Rapids were of consider- ably larger size. A few of the shells were almost devoid of pustula- tions, and one small smooth shell, much produced forward and rounded, bore considerable resemblance to Ohovana retusa. The two largest examples obtained were 85 millimeters long and weighed (^dry shell) 145 grams. This is one of the species commonly used in button making, and where it reaches large size is fairly well adapted to that purpose. In the Maumee, on account of both their small size and their fewness, they are not an important item. Most of the examples secured were retained and numbered in all 156 shells. 5. Quadrulci lacliryniom (Lea). Stranger; monkey-face. This species was found only in the lower courses of the river; none were encountered above the Auglaize. In that stream three examples, the first seen, were secured. From this point they were occasionally met with all the way down the river, and in the canal. Thirteen speci- mens were taken below the Defiance or Independence Dam, and below the dam at Grand Rapids, Ohio, 24 were obtained. This is the greatest number for any locality. The largest one seen was obtained here; its length was 115 millimeters and its weight (dry shell) 221 grams. All examples seen were retained, the total number being 46. This shell is used in button making, and ranks in this respect along with the three-ridges. The irregular sulcation of the valves, how- ever, puts it to some disadvantages in cutting, so that while it is an interesting and well-marked shell, it does not possess enough favor- able qualities from the commercial point of view to make it desirable for propagation. 6. Quadrula cylindrica (Say). Rabbit's-foot. This has been re- garded as a species of the Mississippi drainage, belonging -principally to the Ohio, Cumberland, and Tennessee River systems. Only a few examples were found and these were in the feeder canal and upper 42 MUSSEL FAUNA OF MAUMEE RIVER. parts of the Maumee. It is therefore probable that they represent the result of migrations by way of the Wabash and Erie Canal. One example was found in the St. Marys River and four in the Maumee in the vicinity of Antwerp, Ohio ; a few dead valves were occasionall}" seen, but not more than two or three at the most. None were found below the point just mentioned. In the feeder canal, where condi- tions were particularly favorable for a study of the fauna, four exam- ples were found, and two fine valves, not mates, were obtained in the mouth of the St. Joseph. The shells found in the feeder canal were of an elegant compressed form with numerous triangular tubercles on the anterior portion, and represent the subspecies striglUatu^ B. H. Wright, which is considerably more attractive than the type form. They are rather dark in color, with green triangles, as is usual, and fine capillary rays. One example contained two Atax. This attractive species is not well suited to button making on account of its form and the fact that the nacre is frequently diseased ; hence it is practically unknown to the trade. 7. Quadi'ula undulata (Barne^:) . Three-ridge, blue-point. With the exception of Lampsilis ligamentinii^, Quadrida undulata is the most abundant species of mussels in the Maumee Basin. All the undulate mussels found were referred to this species except those at Put-in Bay, Lake Erie. Notwithstanding the statement made in nearly all the literature at hand that '"the two species" are perfectly distinct and there is no need of confusing them," many examples were found that were hard to place. Different authors do not agree perfectly as to the differences between these species and, generally speaking, the characteristics given are rather vaguely expressed, a fault not due to the carelessness of describers but to the nature of the case. All de- scriptions seem to agree in stating that the umbones are more inflated in plicaia, but we find this feature as varial)le as any other. In the large collection of shells of the Bureau of Fisheries belonging to these two types we have many intermediate forms and several series exhibiting almost perfect transition from one species to the other. This species, as we have it, is exceedingly variable, so much so that in looking over the entire summer's collection it was possible to give the locality of a number of shells simply from their appearance. (J. imdulata was found to be fairly connnon along the river where there were any shells at all, and on account of its abundance only a small proportion of the shells seen were retained. Fourteen were obtained at the riffles in the St. Marys, where it was fairly common. In the St. Joseph Eiver at Robinson Park and in the feeder canal it was abundant and attained large size; 10 examples were obtained from the St. Joseph and 23 were taken of the large number seen in the feeder canal. It also throve in the reservoir of the feeder canal. " Q. undulata and Q. plicata. MUSSEL FAUNA OF MAUMEE EIVER. 43 Very few were found in the upper Maumee. None were noted in the shell pile collected by the clammers near Fort Wayne and only 10 were obtained in the river above Defiance, Ohio. In the Auglaize River the species is quite abundant and 52 specimens were secured. It was fairly common in the Maumee below Defiance. Ten examples were secured at the Defiance Dam, 6 in the canal at Texas, and 16 near the dam at Grand Rapids. It was the most abundant shell in the canal below Grand Rapids. From Defiance down the river and in the canal these shells were l)Iacker and considerably more inflated than those above; the speci- mens found in the feeder canal at Fort Wayne were usually quite compressed and brownish in color, while those below Defiance were full and black, suggesting Q, plicata. These latter shells were usually overgrown with patches of the bryozoan Pliitnatella polymorpha. This is one of the staple species used in blitton making, the nacre having a good luster and the shell being thick and heavy. The small end is beautifully iridescent and frequently of an attractive blue color. This species, along with its relatives, is in good repute among pearl hunters. We found a few minute pearls and small rough baroques, but no more than in other shells. It is fairly free from parasites; a very few Atax were found. The nacre is frequently stained brown. As the species is hardy and thrives in a variety of conditions, espe- cially in canals, it might be a good sort of propagate where better species would not thrive. The total number of examples obtained in the river was about 180. In Lake Erie, at Put-in Bay, we found a dwarf species of shell, Q. plicata hippopma (Lea), resembling this. 8. Pleurobeina clava (Lamarck). Simpson" regards this as a shell of the Mississippi drainage and gives its range as " Ohio, Cum- berland, and Tennessee River systems, reported from Iowa City, Iowa, St. Peters River, Minn., and from Nebraska." Call reports it from the St. Marys and from another stream in Indiana (St. Joseph) of St. Lawrence drainage. In the Maumee Basin it was nowhere abundant, but fairly well distributed along the upper course of the river. One was found above Blulfton Road Bridge, and in the St. Joseph River two dead shells were found, one quite large. Two shells were found in the Maumee near Fort Wayne, while at Maumee Center Bridge we obtained 12 moderate-sized specimens. At Antwerp two small shells were seen, and dead shells were still fairly common several miles below that place. None were seen as far down as Defiance, nor any below there, nor were any foimd in Lake Erie at Put-in Bay. The whole number of specimens obtained » Synopsis of the Naiades, Proceedings U. S. National Museum, vol xxii^ p. 746. 44 MUSSEL FAUNA OF MAUMEE EIVEE. was 20. The Maumee specimens are brightly polished and can be distinguished by this feature from the rest of the summer's collection. This is an exceedingly handsome shell, but of no commercial im- portance, as both its small size and peculiar form are disadvan- tageous. The nacre, though usually white, is occasionally of a beautiful warm pink. 9. Unio gibbostis {Burnes). Spike. Unio gihhosu^ was f Airly com- mon the whole length of the river. It is one of the species of wide distribution and considerable variability. Our specimens do not exhibit much variation, however. Most were of the solid, short- ened, somewhat humped type. The long examples, which have pro- cured for the species the name " spike," sometimes closely re- semble Lampsilis rectus, from which they can be distinguished by the sculpture of the umbones and by the shape of the lateral teeth, while the short humped form is frequently almost indistinguishable from Ptychobranchus phaseolus; about the only way to tell them apart is by observing the color and texture of the epidermis, that of U. gihhosus usually being darker, and by the shape of the body cavity. One shell was found in the St. ]\Iarys, two in the St. Joseph near its mouth, and three in the Maumee near Fort Wayne. It was most abundant in the central portion of the river from Maumee Center Bridge to the bridge at Florida, Ohio, and was common in the Auglaize River. Lower down the Maumee it gradually disap- peared, the last few shells having been seen at Grand Rapids, Ohio, below the dam. In all, 63 shells were obtained in the Maumee Basin. It appears to be one of the common shells in Lake Erie, 16 having been found on the shore of Put-in Bay Island. Both purple and wdiite nacred forms were present. The greater number, about two- thirds, had white nacre, one had rosy nacre, one had purple nacre, white at the edges, and the others were intermediate between w^hite and purple. Young examples showed delicate rays. This species is of no commercial value, as even the white forms have the nacre quite dull. In some localities the distomid of Kelly, associated with dorsal baroques, is a common parasite, but they are not abundant in the Maumee. 10. Alasmidonta truncata (B. H. Wright). This is not an espe- cially common shell in the Maumee Basin. According to Simpson" it is a Mississippi Valley species which has migrated into the upper parts of the St. Law^rence Basin. It is rather variable in form and is sometimes difficult to distinguish from StTophitus edentulus. It was found in the feeder canal and St. Joseph River at Fort Wayne. Occasional examples were encountered along the course of the river, as 2 at Kern Reservation, 1 at Maumee Center Bridge, 1 at Ant- werp, Ohio. Three large ones were obtained in the Auglaize River " Synopsis of the Naiades, Proceedings U. S. National Museum, vol. xxii, p. 671. MUSSEL, FAUNA OF MAUMEE KIVER. 45 ind three in the Maiimee below the Independence Dam. The hist example was found below the dam at Grand Rapids, Ohio. In all, 15 specimens were taken. This shell is handsome and odd, brightly splashed with blue or green and greatly inflated. I^ is of no commercial importance on account of its thinness. The nacre is usually whitish, though fre- quently with a bluish tinge and occasionally a beautiful pink or rosy. 11. Symphynota complanata (Barnes). Heel-splitter; hackle- back. This species is common about Fort Wayne, being especially abundant in the feeder canal, where it was one of the most con- spicuous members of the mussel fauna. Of the great number seen in the feeder canal 18 fine examples were taken; 2 were secured in the reservoir of the feeder canal and 5 in the St. Joseph River near its mouth. This species is evidently an overflow from the upper Wabash drainage. Only one specimen was obtained in the Maumee River and that was at the very head near Fort Wayne; one which had been found at New Haven, about 7 miles below Fort Wayne, had been kept and was shown as a curiosity. This shell was formerly not much used in button making, as it is usually rather thin for that purpose. Its broad flat surface and fair luster, however, are greatly in its favor, and more buttons can be cut out of one of its valves than from any other species except the large Quadrula heros. It appears to breed abundantly and bring forth large numbers of young. Gills of examples obtained in the Little Wabash River near Fort Wayne, September 26, 1906, were full to bursting with glochidia, which are brownish and give the whole mass a distinct brown color. The species appears to have bred successfully in the feeder canal, as there were a great many you.ng shells there. It would probably not thrive well in ponds, though a healthy example was seen in the reservoir of the feeder canal. It is not found in lakes, but is fre- quent in small streams with rather muddy bottoms, and is somewhat similar to Anodonta in habits. It is frequently the host of Atax^ and is one of the species especially susceptible to infection by the parasite Aspidogaster conchicola^ which inhabits the pericardial cavity, without, however, doing much apparent harm. As this species lives where there is rather little current and thrives on a muddy bottom, it would probably be a valuable subject for culti- vation. Its abundance in the canal suggests that it would respond well to artificial conditions. 12. Symphynota costata (Rafinesque). Fluted-shell ; squaw-foot. This mussel was fairly common in the INIaumee River throughout most of its length. One example was found in the St. Marys River 46 MUSSEL FAUNA OF MAUMEE EIVEE. above the Bliiffton Road Bridge, and 11 were obtained at the riffles. In the feeder canal 2 examples were secured. It was probably not uncommon there, but on account of its general abundance no more were taken. From 1 to 3 were found at most of the stations on the Maumee, so that it would be unnecessary and tedious to give its occurrence in detail. It was common in the Auglaize, where 13 ex- amples were secured, and abundant in the Maumee below the dam at Grand Rapids, Ohio, where we obtained 23. One was found in the canal just below Grand Rapids. Some of .the examples are beautifully rayed. The nacre is usually of a peculiar and pleasing yellowish shade. On account of the usual thinness of the shell, how- ever, the species is commercially unimportant, few or none being used in the button business. Altogether we obtained in the Maumee Basin 125 shells belonging to this species. 13. Symphynota- com.'pressa (Lea). In the Maumee Basin we found only one example of this widely distributed species. This was found in the St. Joseph River near its mouth. The species is rather small and thin, and of no commercial importance. 14. Anodontoides ferussacianiid sicbclyindraceiis (Lea). This subspecies was found in only one locality in the Maumee drainage, and there it was the only species found except one fragment of Lamp- silis iris. This was in a small stream, Spy Run, which enters the St. Marys at Fort Wayne. In this stream seven examples were found, all small dead shells. This species is of no commercial impor- tance on account of the thinness of the shell. 15. Arwdonta grandis (Say). This species delights in a muddy bottom and a slow current. It was rather common in such places as the feeder canal and its reservoir, and a few large examples (dead shells) were found along the St. Joseph River near Robinson Park. It was found only occasionally^ in the rivers; 5 were obtained at the riffles of the St. Marys River; 10 were taken of many found in the feeder canal ; 12 were collected in the reservoir of the feeder canal. We found 8 in the St. Joseph River near its mouth. In the Maumee it occurred rather sparingly, 1 example having been obtained at the Mau.mee Center Bridge, 10 in the Auglaize River, 1 in the Maumee below the dam at Defiance, 5 below the dam at Grand Rapids, and 2 in the canal below Waterville. A number were seen below the dam in the canal at Maumee. Only a few of the examples seen were col- lected ; the total number taken was 51. The specimens from the feeder canal were unusually thick, appear- ing as if a second nacreous layer had been deposited. Of all the species of mussel encountered, Anodonta, was the richest in parasites, harboring the greatest number of species, and (except in the case of some Lampsilis alatus obtained at Grand Rapids which MUSSEL. FAUNA OF MAUMEE EIVER. 47 contained great numbers of Atax) the largest number of individual parasites. Most of the shells obtained were more or less stained red interiorly. This stain, caused by a parasite, the distomid of Osborn, varied greatly in extent, beginning in the umbonal cavity and extending from that region in two diverging lines, one anteriorly and the other posteriorly, ventrad. The V-shaped area thus formed gradually fills in with red until the whole surface is involved. This process has been quite fully described by H. L. Osborn « and by H. M. Kelly.^ The minute distomids, showing pink against the mantle, were present in numbers varying from a few to several hundred. Occasionally mussels of this species were found in which the dis- tomids mentioned above had not yet escaped from the sporocysts. The shells of such mussels frequently had the nacre raised in sharp pearly projections, and in some cases the sporocysts seemed to have been overgrown with the nacre of the shell. In addition to the distomid of Osborn, A. grandis also usually has a number of Aspklogaster conchicola in the pericardial cavity, several individuals of Cotylaspis insignis in the axils of the inner gills, and one or more species of Atax among the gills; even leeches are fre- quently found within the shells. The leeches are no doubt parasitic, as is evidenced by the fact that where a pile of freshly cleaned shells killed by muskrats were found under water on another occasion (at Lake Maxinkuckee) the leeches almost covered the insides of the valves. On account of the thinness of its shell, Anodonta grandis^ and indeed all the species of this genus, are of no value commercially. 16. Anodonta imbeeiUis (Say). This species has not heretofore been reported from the Maumee Basin. It was only infrequently found; one specimen was obtained on the shore of the St. Marys River, at Fort A¥ayne. After this none were seen until at the very end of the investigation, in the canal at Maumee, Ohio, below the dam; at this place 38 examples were taken. Their abundance here indicates that they were probably abundant also throughout the length of the canal, but do not " bite " at the dredge hooks. As none were found in the bottom of the feeder canal, it is probable that they do not occur there. This is a very pretty species, but it is of no com- mercial value. 17. Strophifus edentuhts (Say). Squaw-foot. This species is not common in the Maumee Basin. Examples were occasionally found in the St. Marys River at Fort Wayne, in the feeder canal, and in the St. Joseph and upper Maumee Rivers. One was obtained at Grand "Zoological Bulletin, vol. i, no. 6, 189S. "Bulletin Illinois State Laboratory Natural History, vol. v, p. 400, 1899. 22367°— 12 4 48 MUSSEL. FAUNA OP MAUMEE EIVER. Eapids, below the dam. None were seen in the Auglaize River, nor in the Maumee at Defiance. One shell was found on the shore of Put-in Bay, Lake Erie. In all, only 8 examples were obtained in the Maumee Basin. It is of no commercial importance. 18. Ptychohranchus phaseolus (Hildreth). This species was not abundant anywhere in the Maumee Basin, but was scattered all along the length of the river. In the autumn of 1907 a fair number were obtained in the feeder canal. We found 16 good specimens in the St. Joseph River at Fort Wayne, 2 in the Maumee near its head, and 2 or 3 at occasional stations along the river. It was fairly com- mon in the river below Antwerp, Ohio, where 10 shells were obtained at one station. Nine examples were secured in the Auglaize River, 2 in the Maumee below the dam near Defiance, and 3 at the bridge at Florida, Ohio, and 2, the last seen, were taken below the dam at Grand Rapids, Ohio. The examples obtained in the Auglaize August 8 were beginning to become gravid. In all, 53 specimens were secured from the Maumee Basin. Shells of this species were fairly common on the shore of Put-in Bay, where it was small but beautiful in color, luster, and form. The outer gills of gravid examples are remarkable for their folded and plaited appearance; the gills look like nests from which insect larva^, like " Avigglers," are ready to emerge. The nacre of some of the young shells is of a beautiful faint pink, but that of all the large old shells was white. This shell has not been used to any extent, if at all, probably on account of the scarcity of supply, and it is not known to the trade. Its flatness, nearly uniform thickness, and excellent luster would make it worthy of experimentation in the button business. It would probably rank with the " butterfly," Plagiola secvris, in that respect The presence of a number of young shells in the feeder canal indi- cates that it might be easy to propagate in artificial channels. 19. Ohliquaria re-flexa (Rafinesque). Warty-back. This species was first encountered in the Auglaize River, where 1 example was secured. Below the Defiance Dam 3 were obtained; 1 was found at Florida Bridge, and 3 at Grand Rapids below the dam. In the stretch of canal below Grand Rapids it was apparently abundant. One was picked up on the shore at the vei-y entrance of the canal and 22 in a pile left by a mink or muskrat along the side of the canal. A few were found in the canal below the dam at Maumee. This species, being quite thick and of good luster, is used to some extent commercially, but on account of its small size a shell yields only a few buttons. 20. Plagiola donaciformis (Lea). The first example of this species was obtained in the Auglaize River. No others were seen until we reached Grand Rapids, Ohio, where 1 was found in the Maumee below the dam. In the stretch of canal near Grand Rapids it ap- MUSSEL FAUNA OF MAUMEE EIVER. 49 peared to be fairly common; 10 shells were taken from a pile left by some predaceous animal, and of a large number seen in the canal below the dam at Maiimee 14 were secured, some of them quite young and small. This is one of the most beautiful of our fresh-water shells, and the markings are very attractive. It is too small to be of any im- portance commercially. 21. Plagiola elegans (LiQn). Deer-toe. No examples of this species were seen until the Auglaize River was reached, where we found 10. It was fairly common in the river and canal below this point. Living mussels and shells of some recently killed were very common in the Maumee below the dam near Defiance. Of many examples seen we obtained 17 shells here. Below the dam at Grand Rapids, Ohio, we secured G shells, and at the pile of shells near the canal bank, Avhich has been frequently referred to, we obtained 49 shells of this s])ecies. Some were seen along the shore of the Maumee 4 miles above Waterville, and it was common in the canal below the dam at Miami. This is one of the prettiest of our fresh-water bivalves, but it is too small to be used much for buttons, though the largest examples arc sometimes utilized. By clammers it is called the "deer-toe." It is fairly constant in form, but varies considerably in color and mark- ings, the ground color being sometimes yellowish and sometimes green. The rays are often broken up into V-shaped blotches. The total number of examples collected was 97. 22. Plagiola securis (Lea). Butterfly. Of this species only 1 ex- ample, a large coarse specimen closely resembling some of the short- ened and flattened forms of Lam^psilh U game tit hius, was secured. This is a species belonging to the Mississippi drainage, and it is one of the most valuable shells for buttons. The specimen was obtained at the clammer's camp near Fort Wayne and was from the upper part of the Maumee. 23. Ohovaria circulus (Lea). There appears to be considerable confusion among authors regarding the three so-called species Oho- taiia circuluSy 0. letu^ and O. leihii, and by some authorities the two last named are regarded as subspecies of the first. We obtained the small inflated 0. leibil on the shore of Put-in Bay Island. In the Maumee we obtained flatter, larger shells, Avhich are similar to those in the collection of the Bureau from the Ohio and Tennessee Rivers. This species [circuliis) occurs occasionally throughout the length of the Maumee. Three examples were found in the St. Marys River at Fort Wayne. It was fairly common in the feeder canal, where 16 specimens were secured, and in the St. Joseph River near its mouth, where we obtained 10. Six examples were obtained at Ivern Reservation, five at Maumee Center Bridge, and 14 in the 50 MUSSEL FAUNA OF MAUMEE EIVER. Auglaize River, in addition to occasional shells picked up here and there. The last one seen was at Grand Rapids below the dam. The whole number obtained in the Maumee Basin was 60. This species is frequently infested with the distomid of Kelly, and contains small dorsal baroques. Diseased shells with one or both valves roughened and yellow on the inside were quite common. The shell is usually too small to be of much use commercially, but it has a beautiful luster, Avhich continues in dead shells when exposed to the weather longer than in any other species observed. The nacre is usually white, though in some instances it is pink or rosy. 24. LampsiUs gracilh (Barnes). Lamps His gracilis is fairly com- mon and well distributed in the Maumee Basin, but is nowhere par- ticularly abundant. Seven examples were obtained at the riffles in the St. Marys River and three in the St. Joseph near its mouth. In the upper part of the Maumee River it is rather scarce; only 1 was taken near Fort "Wayne, 1 at Maumee Center Bridge, and 1 below Antwerp, Ohio. In the Auglaize River it was fairly common; 13 were obtained there. In the Maumee below the dam near Defiance it was rather abundant, 10 being found at that place. It was also abundant at Grand Rapids beloAv the dam ; here 10 were secured, and 2 in the canal below Grand Rapids. It was quite common in the canal below the dam at Miami, Ohio, though none were collected there. In all, 63 examples were secured, and 10 additional dwarfed speci- mens on the shore of Put-in Bay Island, where it was quite common. This is one of the most iridescent shells we have, the posterior por- tion shading into a fine rainbow blue and purple. It is of no com- mercial value, however, on account of the thinness of the shell, which usually cracks badly in drying. 25. Lampsih's alatus (Say). Pancake. This species is of occa- sional occurrence in the Maumee Basin, but not abundant. Along the upper parts of the basin they were rather rare. Three were obtained in the St. Joseph River at Fort Wayne, and 1, small, in the INIaumee, not far below its source. In 1907 several examples were obtained in the Maumee near Fort Wayne. A few scattered speci- mens were found below Antwerp, Ohio, and 1 in the Tiffin River a half mile above its mouth. Three were taken in the Auglaize and 11 large ones below the dam near Defiance, where it was beginning to become common. Three were found at Florida Bridge. Below the dam at Grand Rapids, Ohio, it was one of the most common shells and great numbers that had been opened by pearlers were seen on the rocks. We secured 23 examples here. A few were found in the second stretch of canal and in the Maumee above Water- ville, Ohio. It was common in the canal below the dam at ]\Iiami, Ohio. In all, 56 specimens were taken in the Maumee Basin. At MUSSEL FAUNA OP MAUMEE RIVER. 51 Put-in Bay Island, Lake Erie, 10 small dwarfed examples were found ; it appears to be common there. The JMaumee Basin examples, especially those obtained in the upper course of the river, differ from the usual type of the upper Mississippi Eiver specimens in being considerably thicker, shorter, and more inflated. Aspidogaster conchicola is a fairly common parasite in the pericardial cavity, and the examples seen at Grand Rapids were infested with Atax to a remarkable degree, some in- dividuals harboring nearly a hundred of the mites apiece. There is a great deal of variation in the color of the nacre of this species, from pink to rich coppery red. Some of the shells were beautifully iridescent. On account of its color and the thinness of the shell it is of no use for buttons. In the upper Mississippi, where this species thrives in great num- bers, it forms the bulk of the cull shells left on shore, and on beds devoted to commercial purposes would probably be regarded as a nuisance. Eichly colored baroques, usually of small size, occasionally occur in this species. 26. LampsUis glaths (Lea). This small, inconspicuous species is very likely to be overlooked in collecting, and is probably more com- mon and more generally distributed than reports would indicate. It inhabits both ponds and rivers. In the Maumee Basin only 6 examples were secured — 4 in the feeder canal, 1 in the reservoir, and 1 in the St. Joseph River near its mouth. All were dead shells picked up on shore. 27. Lojnps'dis iris (Lea). This attractive little shell does not appear to be common in the Maumee Basin. Two examples were secured in the St. Marys River above Bluffton Bridge, 4 in the St. Joseph near the mouth, 2 in the feeder canal, and 1 at Maumee Center Bridge, making 7 in all. It is probably more common than collections would indicate, as it is frequentl}^ found in abundance late in the fall after muskrats have begun collecting, where it has been difficult to find in numbers before this. It seems to be one of the favorite articles of diet with the muskrat. 28. Lampsills rectm (Lamarck). While not particularly abun- dant, this is a fairly common and well-distributed shell in the Mau- mee Basin. Its distribution is so general that it is unnecessary to give details. It was quite common in the riffles in the St. Marys, and present in the feeder canal, St. Joseph River, along the Maumee, and in the Auglaize. Nearly every example seen was kept. In all, 63 were taken in the Maumee Basin and 3 in Lake Erie at Put-in Bay. Only a few young were seen, and a few gravid ones were obtained. The young are more or less distinctly rayed ; in the adult the epider- mis is about uniformly black, but the rays are represented by fine 52 MUSSEL FAUNA OP MAUMEE RIVER. plaitings. The female shells are more or less inflated and pro- duced postero-ventrally, so that the ventral outline is somewhat arched. Generally speaking, the color of the nacre is variable in this species; in some localities many of the shells are pink or purple. The great majority of the shells in the Maumee Basin are white, though the cardinal teeth may be pinkish, pinkish-purple, or purple. Coty- laspis is a frequent, and At ax an occasional, parasite. In a number of examples the nacre near the cardinal teeth was black and diseased. Z. rectus reaches quite a large size in this basin ; the largest example was 172 millimeters long and the dry shell weighed 375 grams. On account of its relative scarcity the value of this species as now occurring in the river would not amount to much. It is, however, the most valuable indigenous species so far as price per ton of shell is concerned. In luster it is not greatly inferior to L. anodontoides, the most valuable of the fresh-water mussels. Its elongate form makes it as well suited to the manufacture of knife handles as L. anodon- toides and for pocket knives the purple or pink nacred shells would not be so objectionable as in the manufacture of buttons, so that the advantage the other species has lies chiefly in the attractive color of its skin. At Grand Rapids we were told that a price of $40 per ton was offered for these shells. As this species lives under widely dif- ferent conditions, such as rivers, canals, and even lakes and ponds (though it is usually dwarfed in lakes), it would be one of the most, if not the most, promising species with which to attempt artificial propagation in the Maumee. 29. Lampsilis ligament inns (Lamarck). Mucket. This is by all means the most common shell in the ]\Iaumee Basin, occurring so frequently in the river that it would be both needless and tedious to go into details. At Fort Wayne it was common in all three rivers — the St. Marys, St. Joseph, and the Maumee — and in the bed of the feeder canal. It formed the great mass of the 25 tons of shells col- lected by the clammers at the head of the Maumee. At Fort Wayne it reached its maximum development, and though not actually im- mense, it was quite large and heavy, as heavy indeed as could be worked up profitably. The shells decrease in size as we go down the river, so that those in the middle portion are of only medium size. Those in the Auglaize River were fairly large. Although L. liga- mentimis was found at most of the points examined for the whole length of the river, it became scarcer and showed a tendency to dis- appear from the lower stretches. It was not common in the second stretch of canal, though a few were seen in the canal bed below the dam at the very end of the trip. Only a small portion of the great number seen were taken ; however, we saved 282 examples. A number of the shells were measured and weighed, of which the largest was MUSSEL FAUNA OF MAUMEE RIVER. 53 145 millimeters long and Aveighed 4G7 grams. The nacre of all the shells was white. L. ligamentinus^ the mucket, while not the very best of button shells, is not greatly inferior to the best, and on account of its com- monness and favorable shape and size is the most common species in commercial use. In the upper Mississippi it occasionally has to be discarded on account of its pink nacre, but this feature is wholly absent from Maumee shells. Perhaps 90 j3er cent of the shells of the Maumee Basin that possess conmaercial value are L. ligameiitinus. This species is rather free from parasites. A few Atax frequently occur, but Cotylas'pis^ Aspidogaster, and the other parasites found so frequently in the thin-shell mussels, are usually rare or absent here. Encysted distomids are fairly common, though hardly so common as in L. ventrlcosus. They are usually found about the edge of the mantle, though they may also be distributed throughout the body. The common form, designated in this report as the marginal-cyst distomid, is of especial interest in this species, as it is occasionally the cause of the formation of pearls. This species was not usually found gravid until late in the season. At the beginning of the Maumee investigation, in June, only a few examples were gravid, but the greater number w^ere barren. As we proceeded down the river, it was noticed that in the Auglaize River, August 8, the gills had begun to fill with glochidia and had a faint blackish edge. On a visit to the riffles of the St. Marys on September 27 a large number of the mussels were found to be gravid. LamjmJ'ts ligamentinus is one of the most easily propagated of fresh- water mussels, and where it produces a heavy shell and white nacre, as it does in the Maumee, is one of the most valuable species for propagation. Although well-marked specimens of L. Ugamentinus are easily distinguished from the following species, L. luteolus, one frequently finds shells which are difficult to identify, apparently standing some- where between the two. 30. Lampsilis luteolus (Lamarck). Fat mucket. Lampsilu lute- olus was found in about the same situations as L. ligamentinus but not in such abundance. Most of the examples seen were collected and the number of shells in our Maumee collection is 134. None of our specimens are of the variety rosaceous^ which is a form belonging to the St. Lawrence drainage. Unlike Z. ligamentinus^ this species is usually abundant in lakes, as well as being an inhabitant of rivers. It is indeed one of our characteristic lake shells. We obtained a number from Put-in Bay Island, Lake Erie, but they were rather small, as is usual with lake- dwelling individuals. The examples found in the St. Joseph River and the feeder canal were remarkably large and fine. This species 54 MUSSEL FAUNA OF MAUMEE EIVER. appeared, like many of the others, to become smaller as we proceeded downstream. L. luteoUis is subject to infection by the same parasites as those which attack L. ligamentinus, viz, Atax and occasionally Cotylaspis. One example obtained in the mouth of the St. Joseph Kiver con- tained a number of Bucephalus polyTrwrphus. This shell has nacre of about the same quality as that of L. lig- amentinus^ but the shell is usually smaller, and has more lateral curvature, so that it can not be worked up so advantageously in making buttons. It is well adapted to the manufacture of fancy buckles, however. 31. Lampsilis TnuUiradictus (Lea). This very attractive little mussel was not common in the Maumee Basin, and nearly all those found were dead shells, so nothing was learned concerning its habits. Only 9 examples were found, and these were in the upper two-thirds of the river basin, in the St. Marys, St. Joseph, and Maumee Rivers, and in the feeder canal. In 1907 a fair number were found in the feeder canal, and it had apparently thriven there to some extent. It is of no commercial importance. The species is a resident of lakes as well as rivers. 32. Lampsllis ventncosus (Barnes). Pocketbook. This shell, though not especially common, occurs rather frequently in the Mau- mee Basin. At the riffles of the St. Marys 7 large examples were found. Very fine examples were also obtained in the feeder canal. In the reservoir of the feeder canal a fine but small specimen was found, pretty closely resembling L. capax. In the mouth of the St. Joseph River 18 were secured. Occasional examples were obtained all the way down the Maumee; in the Auglaize 14 fine specimens were taken. The largest ones found were in the Maumee below the dam at Defiance, and it was also common below the dam at Grand Rapids, and 2 more were secured in the Maumee above Water ville, Ohio. Altogether 83 examples were obtained. This shell also lives in lakes and 12 were obtained in Lake Erie at Put-in Bay, all dwarfed. Although well-marked specimens of this species are easily recog- nized, it has many deviations from the typical fonn. Some closely approach L. capax^ and we have seen shells from Lake Maxinkuckee which seem to lie between this species and L. Ivteolus. In other localities it appears to run into a species which in its typical form is known as L. ovatvs. However, in the Maumee Basin it is well marked and fairly uniform. Z. ventricosus is used to some extent in the button business, but it is quite frequently too thin for this purpose. In spawning season this is a very striking shell, the breeding mussel projecting long vividly colored flaps from the mantle and mov- MUSSEL FAUNA OF MAUMEE RIVER. 55 ing" these with a spasmodic motion, while the animal lies on its back. The spawning season does not appear to be strictly confined to any particular season of the year, but spawning individuals can occasion- ally be found throughout the summer. One was observed in the St. Joseph Eiver at Fort Wayne on July 20. The principal parasites of L. ventricosus are Atax^ usually few in number, the distomid of Kelly which is frequently present but never numerous, and the marginal-cyst distomid, which is occasionally present in great numbers in the margins of the mantle and through- out the body. 33. Micromya fahalis (Lea). This is the smallest and one of the most attractive of fresh-water mussels. It is rather rare in collec- tions, on account of its small size being easily overlooked. It was exceedingly abundant in the Feeder Canal; in 1908 we obtained a half dozen examples, and in 1909, by going over the ground thor- oughly in search of this particular species, several hundred were secured. 34. Truncilla sulcata (Lea). Scattered valves of this species were seen along the shore of the St. Marys and Maumee, and now and then whole shells. Shells and valves were found on the shore of the St. Marys above Bluffton Street Bridge, along the St. Joseph near its mouth, and in the Auglaize River. Only 1 living example was found, and this was in the Auglaize. In all only 15 shells were obtained. Truncilla triqvstra was not found in the Maumee Basin, but we picked up several vales on the shore of Put-in Bay Island. GENERAL DISTRIBUTION OF SHELLS IN MAUMEE RIVER. In general, as to distribution of mussels in the Maumee River, cer- tain species were found to diminish in number or size, or both, in a downistream course. Qiiadrula hibercidata diminishes in both number and size until in the lower part of the river only a few dead and rather small shells were obtained. Dwarfed specimens were also found at Put-in Bay. Q. coccinea gradually diminishes and finally disappears at Grand Rapids, Ohio ; it was found at Put-in Bay. Q. ruhiginosa greatly diminishes in numbers, but a few remain until the very last. Q. cylindrical sparingly present around Fort AVayne, w^as not found far below Antwerp, Ohio ; no living specimens were obtained in the Maumee. PleurobeTMi clava diminishes in numbers and disappears at De- fiance, Ohio. Symphynota complanata was common and large in the feeder canal and is occasional in the St. Joseph at Fort Wayne. One speci- 56 MUSSEL FAUNA OF MAUMEE ElVER. men was seen at New Haven, Incl. We ourselves found none in the Maumee, except at its very head. Ptychohrmichus phaseolus decreases in numbers and disappears at the very last near Miami, Ohio. Dwarf examples were found at Put-in Bay. Ohovaria c-ircidtos diminishes rapidly in numbers, and the last one was found in the Auglaize River at Defiance, Ohio. Lampsilis 7n idtiradiatus^ fairly well distributed in the upper part of the river, wholly disappeared at Defiance. L. ligaTnentinus decreased in number more than in size, but diminished considerably even in size. Trimcilla sulcata^ never very common, was not found below De- fiance, Ohio. The following species increase in number or size, or both, with the descent of the river: Quadinda pustidosa increases both in number and size until it is fairly common below Grand Rapids, Ohio. Q. lachrymosa first appears in the Auglaize a little above Defiance and becomes quite common in the lower part of the Maumee and in the canal. Unio gihhosus increases both in size and number and becomes one of the most common forms in the lower part of the river. Anodonta imhecdlis appeared in considerable numbers at the very last, in the canal below the dam at IMiami. Ohio. Ohliquaria refexa first appeared in the Auglaize River, near De- fiance, and became fairly common below there; the same may be said of Plaglola donaeiformis and PJagiola eJegans. Lmnpsilis gracilis shows considerable increase both in numbers and size. The same may be said of Lampsilis alatus^ of which very few were found at Fort Wayne, while it was abundant in the lower stretches of the river. The following species are of about equal size and distribution throughout the entire length of the river: Quadrula undulata was fairly common everywhere. It prefers a solid bottom overlaid with a thin stratum of rather soft mud, and wherever these conditions were encountered it Avas pretty sure to be found. Symphynota eostata was also fairly common everywhere; it in- creased somewhat in size as we descended the river, the examples found in the Auglaize being the largest obtained. A7iodonta grandis was nowhere abundant, but was found at nearly every station. It prefers a soft muddy bottom. Lampsilis iris was rare everywhere, but was found all along the river. MUSSEL FAUNA OF MAUMEE EIVER, 57 L. rectus was nowhere found in abundance; the examples collected near Defiance were especially large and of fine quality. L. ventncosus was nowhere abundant, but was found occasionally all along the river. Of exceptionally rare shells, Plagiola securis was reprasented by only 1 specimen, obtained at the clammer's camp near Fort Wayne. In searching over the pile of shells found in the poultry yard referred to in the early part of this paper, it is believed that an example of Unio crassidens was seen, but as the shell does not appear in our collections its presence in the river can not be predicated witli certainty. In his paper on " The hydrographic basins of Indiana and their molluscan fauna," *» Dr. Call gives a list of 27 species of Unionicla; from the Maumee River; of these, 5 species, viz, Qnadrvla hci'o.s, Ahtsmidonta ealceola^ Ohovaria retiisa^ Ohovaria ellipsis, and Lanvp- silis subrostratus, were not seen by us. FOOD OF THE MAUMEE MUSSELS. The character of the food of mussels may be better appreciated after a short account of the method of feeding. If one looks at a mussel in its natural position in the water one of the first things to attract the attention is two openings, one large and usually conspicuously fringed with elongate papilla?, the other smaller and fringed or not, according to the species. These openings project from the posterior or sharp end of the mussel which is directed upward from the bottom. The larger fringed opening is toward the ventral portion of the mussel and is an incomplete tube formed by portions of the mantle placed together. This is the in- halent opening, and here water is taken into the gill chambers by means of waving cilia Avhich beset the gills, mantle, and other parts of the body bordering on the mantle chamber. The water thus brought in contains oxygen and food particles. The small cilia move the particles up to the anterior end of the body between the large labial palps or lips, which form a sort of funnel to the mouth. Here the food passes through the short gullet into the stomach, and thence into a long convoluted alimentary canal, which finally ends at the smaller of the two openings noted above, the exhalent aperture. There are minute pores in the gills, and through these the water is forced by cilia into the exhalent aperture, from which it returns to the lake or stream in which the mussel lives. Large robust mus- sels are able to produce quite a strong current, but this is usually not visible in the rippling water where they naturally live. It can be best observed in large river mussels removed from their native stream « Proceedings Indiana Academy of Sciences, 1896, pp- 248-257. 58 MUSSEL FAUNA OP MAUMEE EIVER. and placed in the calm shallow waters of a lake. It is probable that these mussels breathe extra heavily in the lass highly oxygenated lake water. The current from the exhalent siphons boils up vio- lently and is ejected intermittently at more or less regular intervals. In many mussels there is no direct communication between the gill chamber into which the water, bearing solid particles, enters, and the suprabranchial chamber from which it is ejected through the exhalent aperture. Some mussels have the gills free above pos- teriorly, so that the two chambers have direct communication, and in these it is possible for the mussels to shunt solid particles directly from the gill chamber to the chamber above without passing through the alimentary canal. We have no evidence, however, that this is ever done. The mussel probably exercises little choice as to the nature of its food. The papilla? which fringe the edge of the in- lialent orifice are supposed to be water-testing organs, and the mussel can close itself against foul water or against positively objectionable material. In the feeding mussel, as observed in the stream or lake, there is no appearance of nicety or fastidiousness as to the particles swept in ; the wide open, immobile, and expressionless lips admit the general current and its diverse minute population with a catholic impartiality. The alimentary canal of the mussel is usually filled with a sort of impalpable or sometimes gritty mud, and mixed in Avith this are found scattered organisms of various sorts which serve as food. The impalpable mud usually corresponds closely in color and general appearance with the bottom in which the mussels are found ; in rivers with yellow clay bottoms it is yellowish; in dark, muddy bottoms it is dark. It consists of fine particles which have been held in suspension in the water. In turbid streams it consists of the material which gives the water its turbidity, in clear ponds with bottoms of fine mucky silt it is black, and in those still waters full of diffused minute alga? and flagellates, which are common in ponds and pools, it is predominantly organic and of a greenish cast. In the latter case it usually consists almost entirely of colonies of plankton-scum (water bloom) alga^, which are frequently inclosed in clear vacnolelike spaces. The moderately long alimentary canal suggests that the ingested material contains a low amount of nutriment; observations also seem to indicate that the digestive powers of the mussel are slow and feeble. In mussels which had been kept for several days ScenedesTrms was found as fresh and green as ever, and in one mussel obtained in the reservoir of the feeder canal green flagellates were still alive and ac- tive after having been at least 36 hours in the mussel. While in these instances it is possible that digestion was delayed by the capture of the mussel and placing in changed surroundings, there are other indi- MUSSEL, FAUNA OF MAUMEE EIVER. 59 cations that digestion is normally slow and imperfect. In mussels taken from the beds, washed, and immediately examined, diatoms found at the posterior end of the alimentary canal are frequently alive and exceedingly active ; they appear to be even more active than when taken from open water, and the form and color of the chromato- phores are unchanged. It is possible that in these cases the mussel obtains some nutriment from the outer coating of the diatom and that the removal of this makes the latter especially active. The organisms found in the alimentary canal of mussels vary con- siderably with the places in which the mussels live, but, so far as observed, do not differ much, if at all, among the various species. They consist of such planktonts as are small enough to be taken in, and not active enough to resist the force of the inward current pro- duced by the mussel. The largest and most active planktont taken to any extent was the entomostracan Bosminia. This was found with considerable frequency in lake-dwelling mussels, but as all seen were dead, and many were simply empty shells, it was not possible to ascertain wdiether the mussel had taken in the living animals or not. Actively moving flagellates are quite frequently ingested. In the gill chamber of some Quadrula tiihercnlata^" a species with an exceptionally large inhalent opening, we found such objects as Chiro- nornv,^ larvae and one small mature dipterous insect. There are several organisms which are likely to be found in mus- sels wherever taken. Species of Scenedesmus are almost invariably present, as is also Navicula along with other diatoms and species of Pediastrum^ such as P. pertuswn with great frequency, P. horymvum rather commonly, and P. ehrenhergii^ which appears to be a rare species everywhere, only occasionally. The rotifer Anurma cochlearis is also found frequently and under all sorts of conditions, in lakes, ponds, and rivers. The part usually found is the empty lorica. Another common object found in the alimentary tract of mussels is a club-shaped, several-celled, brown object, probably the spore of a fungus or lichen. With the exception of those just mentioned the organisms found within mussels are very diverse, and the creatures seem to be quite omnivorous within the limits of what they can capture. Just as in any stream one finds various types of planldon aggregates from the extreme type of " potamoplankton " or characteristic river plankton of the mid-current, recognizable by the scarcity of water-bloom alga?, to well-developed types of " limnoplankton " or lake plankton in the quiet sloughs and bayous, so in these different portions of the stream one finds different sets of organisms in the mussels. In lakes full of diffused water-bloom algse, the material which forms the mudlike matrix of the food in river mussels is to some extent replaced by the « Not in the Mauniee River, however, where few were examined. 60 MUSSEL FAUNA OF MAUMEE RIVER. water-bloom algse. These may serve also as food. The stomach con- tents of such mussels, when placed in vials of preservative, differen- tiate into a lower stratum of black mud and an upper flocculent stratum of fairly pure algse. The mussel may be regarded as a sort of living filter, feeding upon the filtrate it separates from the water. It would be worth investi- gating to ascertain whether they arrest and destroy pathogenic bacteria and thus become efficient purifiers of water. They are valu- able aids in plankton investigations, capturing many minute organ- isms which escape methods of collection ordinarily in use. Though at the outset it hardly looks as if they could cope with the problem, their capacity for ingesting water-bloom algse {Clathrocystis, Lynghya^ etc.) suggests that it might be worth while to investigate their efficiency as reducers of plankton scum. The Avater bloom, as is well known, frequently becomes a nuisance. It collects in. noisome masses of scum along shores of otherwise at- tractive lakes and reservoirs. Its taste and odor renders the water of reservoirs undrinkable, and for a remedy of this condition the copper sulphate treatment has been devised. Its presence in large quantities along the shores of our most beautiful lakes renders them unattractive during the late summer season, and in many places brings the bathing season to an early close. Aside from the filthy appearance of the w^ater, many persons claim to be actually poisoned by the water bloom, and there are instances on record of live stock being fatally poisoned by drinking water covered with plankton scum. In addition to this, the great amount of decaying mate- rial in the water is said to take up oxygen, making the lower strata iminhabitable for fish. It is possible that by planting large numbers of mussels, supplemented by planting ^^g masses of Ghiro- nomns, which appear to hatch easily and the larvae of which eat the coarser alga?, that the plankton-scum nuisance can be gTeatly abated. The stomach contents of mussels taken from the main current of the St. Marys, St. Joseph, and Maumee Rivers were rather note- worthy for their paucity of organic material. Through the large mass of muddy matrix filling the stomach were usually scattered a few Scenedesmus, various diatoms, and an occasional Pediastrum or Cosmanum. At the riffles small brown cystlike objects, which may have been a species of Trachelcmonas, were quite common ; with the exception of this the mussel contained very little. Among the organ- isms noted were Seenedesmus cmidatus^ Cmlastrum microsforum^ Pleurosigma, several forms of Nmncula^ PJiacus longicaudus^ Pe- diastrum haryanum^ Gomjjhonema^ a sponge spicule, and an active Euglena-Wko, organism. The stomach contents of mussels talvcn else- where along the river and in its tributaries, as in the Auglaize at Defiance and the Maumee at Grand Rapids, were not essentially dif- MUSSEL FAUNA OF MAUMEE KIVER. 61 ferent. One of the examples of L. ligarrhentinus taken in the Auglaize, however, contained many very minute cylindrical objects with rounded ends, which were probably bacteria. The stomach contents of the mussels found in the reservoir of the feeder canal, a shallow pondlike body of water, bore considerable con- trast to that of the river. This reservoir was well populated with small gi-een monads and similar organisms, and during the last days of our observation became covered over Avith a green scum. It was " puddle plankton " rather than that which is characteristic of either lakes or rivers. The stomach contents of the nnissels were gTeenish in mass, and were found to consist of various organisms, mostly green in color and actively moving. The main mass consisted of small globular and thick celled, green or brownish flagellates, probably T raehelernonas lagenella. Among other organisms were species of Phacus^ several specias of Scene- desmus^ Pediastrwm lyertuHum^ GonvphoTiema^ several species of Navicula^ a little Botryococcus brwini^ Anunea cochZearls^ Cos- marium, small fragments of a Conferva-Vike alga two or three cells long, fragments of the test of Ceratium hirundinella, and the brown objects resembling fungus spores. There were numerous narrow curved objects which were probably loricas of T raehelernonas. PARASITES OF THE MAUMEE MUSSELS. During the progress of the Maumee lliver investigations the para- sites encountered were noted and examples saved for further study. Generally speaking, Anodontas and other thin-shelled mussels are more heavily parasitized than the species having thick shells, and parasites are more abundant in shallow, warm, and quiet pools than in rivers. The mussels of the feeder canal reservoir were more heavily para- sitized than those of the main streams. The following parasites were observed : 1. Cotylasjns indgms Leidy, a small organism, which to the naked eye resembles a pale leech. The body is trumpet shaped, and the ventral surface has a large ovate disk cut up by partitions. By this disk the animal adheres to its host, and it usually inhabits the axils of the inner gills. It was most common in Anodonta gmndis and occurred occasionally in most other species of mussels examined. An examjole of Lampsilw rectus obtained at the riffles contained the sur- prisingly large number of 67 of these parasites. So far as could be ascertained, this parasite produces no marked physiological effect upon the host and appears to do no particular harm. 2. Aspidogaster concMcola Von Baer bears a close general resem- blance to the above, but is considerably larger, and the adhering 62 MUSSEL FAUNA OF MAUMEE EIVER. disk, which is hirger and more complicated, has been aptly compared to the sucking disk of Remora. This parasite affects various species of mussels, but its favorite hosts are species of Anodonta, Synvphynota complanata,, and Lamp- s'dis alahfs, in which considerable numbers were found. It usually inhabits the pericardial cavity of the host, though when considerable numbers are present they may be found in other parts of the body. The finding of a small elongate " hinge pearl " in a mussel affected by these parasites suggested that, under peculiar circumstances, the bodj^ of one of these creatures may form the nucleus of such a pearl, since the shape of the two is quite similar. This supposition, however, needs further investigation. Aspidogaster reproduces within the jDericardial cavity of the host without any marked metamorphosis, and they are found of all sizes, the minute ones having the same form as the adults." 3. The onarginfd-cyst disfomid. — A distomid forming spherical cysts was fairly common, especially in Lampsilis Ugamentimis and L. ventricostis, along the whole length of the river. A few of these parasites were also found in one Quadrula vndvlata and one Sym- pliynota costata. It is probably the species discovered and briefly described by H. M. Kelly,* who noticed it in four examples of L. ligamentimis. The cA'sts are usually found along the edge of the mantle, generally in the muscular portion below the pallial line. They are also fre- quently embedded in other parts of the body, such as the adductor muscles, and esjDecially in the keel of the foot, where they are occa- sionally found in great numbers. There are indications that they prefer muscular tissue. Cysts of various ages were frequently present side by side in the same mussels. The youngest cysts are translucent and refractive, faintly yellowish in color, and resemble minute pearls embedded in the edge of the mantle. As they grow older they form conspicuous black dots. One of the smaller cysts measured was 0.3 millimeter in diameter, and a large blackish opaque one was 0.9 millimeter. Examined with a microscope the younger cysts appear as brick-red spheres, crossed and recrossed by irregular cracks like those of a dried mud flat. Upon carefully breaking open this crust a minute distomid is released, which slowly crawls about on the slide. One of the dis- tomids thus released was colorless, 0.365 millimeter long and 0.21 millimeter wide at its widest portion. Viewed from above, it was pear shaped in outline, the anterior end being acute, the body gradu- ally broadening behind and the posterior end being rounded. It had " The development of this species is given by Huxley, Anatomy of Invertebrate Animals, p. 178. * Bulletin Illinois State Laboratory of Natural History, vol. v, p. 40G. MUSSEL FAUNA OF MAUMEE EIVER. 63 tv.o suckers, an anterior one at the small anterior end and a ventral one, about the same size, placed far back near the swollen posterior end. The posterior part of the distomid contained a mass of black material, probably food. As the C3^st increases in age it becomes black and forms a conspic- uous dot about the size of a pinhead in the tissue. It is found on examination that the cyst itself, which is surrounded by the black material, has not increased in size, and the wall retains its brick-red color. The appearances indicate that the black material, which is probably a product of excretion of the parasite, has been passed through the cyst wall. In our alcoholic material the cysts, which appear as small black spots to the naked eye, show under the simple lens as minute orange globes, surrounded by a semitranslucent halo, the alcohol having apparently dissolved out the black material around the cyst wall. The marginal-cyst distomid reminds one in its general appearance of Diplostomum^ which is frequentl}'' found encysted in the skin of fishes, especiall}^ minnows, forming small black spots. In the fishes, however, the black color surrounding the cyst appears to be due to a deposit of pigment by the fish. The effect of this parasite (the marginal-cyst distomid) upon the mussel appears to be quite various, according to the location of the cyst. In many cases, probably where the cysts are near the inner surface of the mantle, no effect whatever appears to be pro- duced upon the shell. Under other cysts the shell is stained brown, and where the cysts are near the outer.surface of the mantle the whole edge of the shell is blistered irregularly and stained a steel-blue color. Moreover, as the cysts increase in size they form firm nodules in the mantle and cause the shell to be built out around them, so that they leave pits in the nacre similar to those left by pearls. In studying the collection of shells of L. Ugamentinvs at the Bureau of Fisheries, which contains perhaps 2,000 specimens of this species and is represented by shells from widely separate parts of the country, peculiar malformations of the shell were observed, which were probably caused by the parasite under discussion. Among these markings are pits near the margin of the shell, such as we have frequently observed in the living mussel underneath the cyst, and sometimes these contain small attached pearls. One of the most marked features is a tendency of the antero- ventral portion of the shell to cease growing outwardly and either become thickened or gape away from the middle line so that the mussel can not tightly close its shell in front. It appears that the presence of the cysts keeps the affected portion of the mantle in a constant state of con- traction, thus preventing it from building the shell outward in the 22367°— 12 5 64 MUSSEL FAUNA OF MAUMEE EIVEE. normal fashion. In some of our alcoholic material this contraction and thickening- of the mantle is quite markedly shown. The effect of this peculiar form of growth on the shell as a whole is to give it a somewhat more inflated and elongate appearance than is character- istic of the species. Shells of L. Ugamentinus thus affected approach the contour of L. luteohi8\ and while close approximation of these two species is not rare in normal shells, some of the puzzling cases are probably L. ligameyitiniis which has been modified by this dis- tomid. In L. ventiicosKS^ which is ver}^ susceptible to the attacks of this parasite, we frequently have gaping shells, and others which are markedl}^ thickened; Some of these thickened shells closely resemble those of L. ligmnentinus. The marginal-cyst distomid is of especial interest on account of its connection with pearl formation. Suspicion that this distomid had some connection with the forma- tion of pearls was aroused when the objects were first encountered, as their form and position was especiallj^ suggestive. This suspicion was intestified by frequently finding small round pearls in the mantle of mussels, closely associated with the cysts. Finally, on decalcifying and clearing in paraffin and x^dol, preparatory to embedding, a por- tion of the mantle containing a small round free pearl surrounded by the cysts, a c^yst could be clearly distinguished in the center of the pearl. Our material, which was preserved by simply dropping the mussels into strong alcohol, was not in good condition for histological jDurposes, so that we were unable to demonstrate the nature of the tissues surrounding the pearls. Our studies indicate that the marginal-cyst distomid is widely dis- tributed. "We have found the cysts in alcoholic material collected in the Mississippi River, in L. Ugamentinus collected by Freeland and Williams in the Illinois River in 1907, in L. ventricosus collected in the Yellow River, Ind., by ourselves in 1008, and in L. ligam£ntinus collected in the Ohio by Boepple in 1009. From the appearance of shells examined, we judge it to be common in the Minnesota River. 4. The distomid of Oshorn. — Quite frequently during our mussel investigations we have found Anodonta affected with small white sporocysts, which covered the outside surface of the mantle next the nacre. The nacre of these mussels was often raised into a number of sharp pearly blisters, and in some cases small white areas, conform- ing with the size and shape of the sporocysts, indicate that the sporo- cysts had been actually covered with nacre. Mussels thus affected were found in the Little River ditch near Aboite Station, Ind., and in White Bear Lake and Lake Minnetonka, near St. Paul, Minn., in 1008. Notes on alcoholic examples of Anodonta collected in AVild Cat Creek, Carroll County, Ind., in 1809 refer to '' white distomids," which were probably this species. MUSSEL FAUNA OF MAUMEE lUVER. 65 111 the reservoir of the feeder canal at Fort Wayne we found an Anodonta plentifully infected with these sporocysts, and were able to make out in detail some of the steps in the life history of the animal. The sporocysts were small white objects, elliptical in outline, about 1.14 millimeters in leng-th and 0.05 millimeter w4de. The skin of the sporocysts was transparent enough to reveal, in the interior, the cer- caria>, which were three in number in most cases examined, closely doubled up and snugly packed together, and performing an uneasy wiggling motion. The walls of the sporocyst easily rupture; in an example studied at St. Paul the previous year the covering was torn apart by a dissecting needle and the cercarian set free. In the ex- ample now under discussion the sporocysts were rupturing of their own accord, allowing the cercaria? to escape. The following is a de- scription of the various stages observed : Cercaria just escaped from the sporocj'st: Body elongate cylindrical with bluntly rounded ends and divided by a well-defined constriction into an anterior two-thirds, the body proper, and a posterior one-third, the ^ tail.'' Body portion ellip- tical in outline with the two ends similar. Anterior sucker terminal, about one-fourth the greatest body diameter, circular, with concen- tric ridges faintly marked. Posterior sucker one-third body diameter, also circular, but with radiating ridges, situated almost in the exact center longitudinally as well as transversely; constriction only one-seventh the bodv diam- eter with a well-defined groove all the way around. Posterior por- tion (tail) almost perfectly elliptical, twice as long as wide, both ends alike and more bluntly rounded than the body proper. Tail covered with a thick epidermis, raised into longitudinal ridges close together, very narrow and zigzag in arrangement. The internal structure is indistinct, but far enough developed to distinguish the pharynx and digestive canals. The former is rela- tively smaller than in the adult and is much nearer the posterior sucker than the anterior one. This means that in after development the posterior sucker travels backward, while the pharynx travels forward. At this stage the whole interior of the body is filled with large cells having numerous and interlacing intercellular spaces. After remaining under the cover glass for a little time the tails break off at the constriction and the body moves about alone. The following is a description after the tail is shed : Body elongate, ovate, without any traces of segmentation or sepa- ration into parts; anterior end evenly rounded, posterior end drawn out into a point, bluntly rounded at the tip. Anterior sucker terminal, circular, three-sevenths the diameter of the body, with circular ridges. Posterior sucker on the mid line 66 MUSSEL FAUNA OF MAUMEE EIVER. of the ventral surface, four-sevenths of the entire length from the anterior end, circular, with radiating ridges and four-sevenths the diameter of the body. Posterior end of body with traces of the zig- zag longitudinal ridges seen on the tail of the young. Mouth in the center of the anterior sucker, opening into a spherical or transversely elliptical pharynx close behind the sucker and one-third its diameter. From the pharynx two digestive tubes lead back to the posterior end of the body, one on each side, each a little narrower than the phar- ynx, of nearly uniform diameter throughout, and curved outward parallel to the lateral body margin. They reach nearly to the poste- rior end of the body, where each ends in a network of intercellular spaces and tubes, the excretory organ. The newly escaped cercarium is about 1.05 millimeters in total length and 0.2 millimeter wide. The length and width are con- stantly changing, according as the organism is stretched out or contracted. The tail is no longer an organ of locomotion; it is very large and unwieldy and easily detached ; most of the cercaria shed it soon after emergence from the sporocyst, so tliat it is evidently a vestigial organ. After shedding the tail the organism, which now has the typical form of a Distomicm, moves up to the umbonal region of the mussel and acquires a pinkish color. From the umbonea it spreads in di- verging lines postero-ventrally and antero-ventrally and finally fills in the space between. "VVliere it is located it profoundly affects the nacre, staining it a salmon color, and frequently giving it a diseased appearance. The presence of this distomid in Anodonta and its re- lation to the salmon-colored nacre was first pointed out by H. L. Osborn" and further discussed by H. M. Kelly.'' In its general appearance and in its life history this organism agrees very closely with Dlstomum dupllcatwn as described by von Baer, who gives a detailed description of the sporocyst and a history of the escajDe of the cercaria and its subsequent behavior.*' It differis from that species in the fact that the tail is not clavate, is longer than the body, straight, and not pediculate. It differs from the cercarian genus Rhophalocerca Diesing, formed to include D. dupli- eatu?n; only in the fact that the tail is not clavate. Von Baer de- scribes what appeared to him a form of reproduction in this distomid, but from what is now known of related forms it appears probable that he misinterpreted appearances. Dhtomum duplicatum. (known also as Rhophalocerca tardigrada Diesing) is of special interest as the organism first shown by De Filippi to be the cause of pearls in « Zoological Bulletin, vol. \, no. 6, 1898. » Bulletin Illinois State Laboratory Natural History, vol. v^ art. viii. <■ Nova Acta Physico-Medica Academia? Ctesares Leopoldino-Caroliuce, Natural Curio- eorum, vol. xiii, pt. 2, 1827. MUSSEL FAUNA OF MAUMEE RIVEK. 67 Anodonta cygnea of Europe; hence a few remarks concerning it may not be out of place. So far D. (luplicatum lias not been reported from any mussels ex- cept Anodonta. Its discoverer found it usually in the kidneys of Anodonta ventrtvo.lia and more rarely in the same organs in A. ana- tliui. One badly infected example of this latter species had the sporocysts in the kidneys, gills, pericardial cavity, and edge of the mantle, and in an example of .1. veiitncosa they were even present in the general circulation. Filippi found the sporocysts and free dis- tomids in the mantle of A. cygnea^ some of them surrounded by nacre ; and Pagenstccher found them in the tissue of the genital glands, liver, and kidneys of the same species. In this country Leidy° has reported ^'■Rhophalocerca tardigroda attached to the mantle of Ano- donta fwoiatilis; specimens also obtained by Mr. Lea from A. lacKstr'isy The mature form of Dhtomnni dv/plicatum is said to be D. tereti- colle Rudolphi, a parasite of fresh-water fishes. To return to the distomid of Osborn: In addition to its being found in Anodonta grandis, this parasite (in the distomid form) has also been reported from Strophitus edentidus by Osborn. We had very few living mussels of this species to examine; however, we found dead shells with the peculiar salmon-colored nacre, similar to that produced by this parasite in Anodonta. In vicAv of the fact that a Distomum quite similar in appearance lo this is found in other species of mussels, while sporocysts are rare (we found them only in the liver tissue of one example of Qumlnda tiiherculata in Tippecanoe River), it becomes an interesting ques- tion whether this distomid migi-ates from the Anodonta to other mussels after issuing from the sporocyst. In the first place, such a migration is probably not an event closely connected with the life history of the species; for the tail, which functions as an organ of locomotion in those species which do migrate, has here been allowed to become functionless, or even a dead weight. Moreover, if the host of the mature form is a fish, as one might expect it to be from the life history of its near relative D. dupUcatmn,., it would be to the advantage of the distomid to remain in the thin-shelled Anodonta, which is much more likely to fall a prey to fishes, or indeed any other predaceous animal, than the thicker-shelled species of mussels. It is very likely, however, that, just as in greatly infested mussels the cercarise frequently overflow the organs which appear to be their favorite habitat, so a considerable number of distomids may be crowded out of the host mussel and be forced to seek protection and food elsewhere. According to von Baer ^ the sporocj^sts of Distomum « Proceedings of the Academy Natural Sciences, Philadelpliia, 1858, p. 110. '' Loc. cit. 68 MUSSEL FAUNA OF MAUMEE RIVER. dwplicatum' and their contained ceicariro live for a considerable time in the water when removed from the host. Migration wonld there- fore be possible. It has so happened that the Anodontas in which we found sporo- cj'sts were all taken in early and middle sunnner, in Jul}- or August; we found free distomids, hoAvever, as early as work in the field began, in June. It would require the study of infected regions throughout the year to ascertain whether the various phases in the life cycle of this distomid bear any relation to the seasons. Von Baer found his Anodontas affected by the sporocysts of D. dupUcatum in late autumn. 5. The distomid of Kelly. — In several of the thicker-shelled species of mussels, such as Lampsilis ventHcosus, L. Uffamentinus, Ohovaria chxulus, and especially in Qu-adrula tuhercidata and U7uo ffibhosus, one frequently encounters pink distomids bearing a close similarity to the distomid of Osborn mentioned above. Kelley," who first de- scribed and discussed these distomids, was of the opinion that they are identical with those described by Osborn as affecting Anodonta and Strophitus and producing a discoloration of the nacre. Indications pointing to the identity of these two distomids are, first, the fact that dorsal baroques are formed in Anodonta by the distomid of Osborn, and second, that the distomid of Kelly, in addi- tion to forming baroques in Ohovaiia circtdus, frequently causes the nacre to become rough and assume a yellowish color, approximating the salmon color of affected Anodontas. In their typical form as they appear in the thick-shelled species of mussels, these distomids exhibit several minor points of difference from the distomid of Osborn. Practically they present so different a set of manifestations that in our field notes it was found necessary to separate them. After considerable study and comparison we are inclined to think them the same, but in this discussion, since the point of view is ecological rather than taxonomic, it will conduce to clearness to refer to this manifestation of the species under a different name. Simi- larly, it is convenient and customary to speak of the cercaria of a distomid, the nanplius of a crustacean or the leptocephalus of an eel. The distomid of Kelly is usually found in much fewer numbers than that of Osborn, a fact which suggests that it may be erratic or a stray migrant, and so far we have not found it associated with sporocysts or cercaripe except in the single instance of Quadrula tuhercKlata referred to above. It is usually confined to the outer surface of the mantle near the cardinal or lateral teeth of the mussels affected, and it is frequently associated, as Kelly has pointed out, with irregular dorsal baroques or pearls. It does not produce a salmon-colored stain in the nacre of the mussel affected, and usually « Bulletin Illinois State Laboratory Natural History, vol. v, art. 8. MUSSEL FAUNA OF MAUMEE EIVEE. 69 produces no stain at all. The discolorations frequently found ni the umbones of mussel shells can not all be attribut(^d to it, since in one pond where such stains are very frequent, no distoniids could be found. In general, it is a baroque former rather than a stain pro- ducer. In some localities it especially attacks Unio g/hhosus and Qvadrula ttiherculata^ and the host mussels are peculiarly prolific in baroques. Most of the dorsal baroques which the clammers save and sell to dealers in jewelry are probably formed through the ac- tivities of this parasite. The baroques are known among the clam- mers as " chicken feed " and usually bring about $2 per ounce on the ground. In many places this material is fairly common and is a source of revenue among the clammers. It is probable that the distomids which stray into large heavy- shelled mollusks fall out of the normal course of their life history unless they again migrate to thinner-shelled species, for the thick- shelled mussels, after they have grown, are proof against fishes, musk- rats, or any other predaceous animals, and live to die a natural death, or in these recent times to fall a prey to the clammer. This gives the parasitic distomid no opportunity to enter into another host in the usual manner. Crawfishes eat mussels that have died ; but so far as known the host of the mature distomid is always a vertebrate. Moreover, on the supposition that the distomids found in the heavy-shelled mollusks are migrants from Aiiodonta, SfrophUxs, or other thin-shelled species, these mussels, which are regarded as valueless by the clammer and destroyed by the wholesale, are of im- portance as an intermediary in the formation of baroques and perhaps all distomid-formed pearls. Similarly, though in a somewhat dif- ferent manner, the cockle, Cardii/m, acting as an intermediate host between a species of mussel-eating duck and the salt-water mussel Mytilus, brings about pearl formation in the latter species, as shown in the investigations by Jameson « of pearl formation in Mytilus. 6. Bucejjlialus polymorphiis (von Baer). — During the Maumee in- vestigations this species was encountered in only one instance, when an example of Lampsilis luteolus. obtained near the mouth of the St. Joseph River, was found to be a fleeted. In investigations outside the Maumee Basin it was not infrequently met. According to Kelly, who has had considerable experience with mussel parasites, it is fairly common, though not so common as Aspidogastei^^ Atax, or Conchopthirus, and affects a large number of species of mussels. Bucephalus pohjmorphus is of considerable economic significance, since it has been shown by von Baer and later b}^ Kelly tha*: this species (and, as Kelly has pointed out, some other distomids as well) frequently affects the generative tissue of the host mussel to such an "Proceedings Zoological Society London, vol. \, 1902. pt. 2, p. 140-166. 70 MUSSEL FAUNA OF MAUMEE RIVEK. extent as to wholly obliterate it and render the mussel incapable of reproduction. In those mussels {Lampsilis gracilis, L. luteolus, etc.) in which there is a marked difl'erence in the form of the shell of the two sexes, Bucephalus, by arresting sexual development at various jDeriods of the mussel's life, stops also the development of the peculiar form of shell characteristic of the sex. Shells are thus left at vari- ous intermediate stages of sexual development, though they may attain full size, and this has doubtless led to considerable perplexity in identifying them, or even to the descriptions of new species. It may also be partly responsible for the large amount of synonymy found in the Lampsilis rentricosus group. It can be readily inferred that the introduction of BucepJmlu^- infected mussels into places where breeding or breeding experiments are being carried on would be likely to produce disastrous results. In collecting mussels for propagating purposes, therefore, shells of in- different contour as regards sexual peculiarities should be viewed with suspicion. In addition, examples should be dissected to detect signs of the presence or al^sence of the trematode. The mature form of Bucephalus polymorphus is Gasterostonium fnhhriatunv von Siebold, which is a parasite of the pike {Esox) and perch {Perca), and an intermediate host (eaten by the final host) is some species of minnow. {Leuciscus is one of the knoAvn hosts.) In ponds devoted to the breeding of mussels the ravages of Bucephalus might be held in check by avoiding the peculiar combination of mus- sel, minnow, and pike or perch, provided of course these were not the particular species of fishes required to carry the glochidia. 7. Af