£>3 ■TV* Cornell University Library The original of this book is in the Cornell University Library. There are no known copyright restrictions in the United States on the use of the text. http://archive.org/details/cu31924032171377 Cornell University Library arY356 Report of the observations of the total 3 1924 032 171 377 olin.anx / REPORT OF THE OBSERVATIONS OF THE TOTAL SOLAR ECLIPSE, July 29, 1878, MADE AT FORT WORTH, TEXAS. EDITED BY LEONARD WALDO, ASSISTANT AT THE OBSERVATORY OF HARVARD COLLEGE. CAMBRIDGE: PRESS OF JOHN WILSON AND SON. 1879- Acknowledgment. It would be ungracious to allow this monograph to find its way into the scientific world without recognizing the generosity of those ^gentlemen who have by their material aid assisted us in bringing to a gratifying conclusion our observations of the Total Solar Eclipse of 1878. We feel especially indebted to — The Board of Directors of the Bache Fund. The Officers of the Western Union Telegraph Company. The Officers of the Pennsylvania Railroad and of the Railroads connecting St. Louis with Fort Worth. The Officers of Washington University, St. Louis. The Officers of the University Club, St. Louis. S. W. Lomax, Esq., and those other Citizens of Texas who actively co-operated with us. To the following gentlemen of St. Louis : — Gerard B. Allen. W. H. Gregg. Thomas Allen. Edwin Harrison. Robert B. Brown. E. A. Hitchcock. C. B. Burnham. Henry Hitchcock. George O. Carpenter, Jr. George E. Leighton. Daniel Catlin. Charles Parsons. M. Dwight Collier. J. C. Richardson. John T. Davis. R. M. Scruggs. W. G. Eliot. W. A. Thornburgh. C. S. Greeley. James E. Yeatman. And to — Professor S. S. Greene, of Brown University, Providence. E. Ellery Anderson, Esq., of New York City. T. P. Smith, Esq., of Waltham, Mass. D. T. Smith, Esq., of Salem, Mass. The Firm of Alvan Clark & Sons, Cambridgeport, Mass. We trust that these gentlemen will find in the following pages an indication that their aid has resulted in some slight accession to our knowledge of the Sun. THE EDITOR. Cambridge, January, 1879. Contents. PAGE Acknowledgment 3 Introduction 7 Personnel and Outfit of the Expedition 7 Arrangements for the Observations on July 29th 9 Description of the Method of Making the Sketches of Plate IV. n Rksumk of the Observations of the Times of Particular Phenomena 12 Time Observations at Fort Worth 14 The Observations with the " Brown University " Transit 15 The Adopted Chronometer Errors and Rates 15 The Longitude Determination 16 Programme for the Exchange of Chronometer Signals 16 Fort Worth and St. Louis Chronometer Comparison 17 St. Louis Chronometer Error Determination 18 Final Corrections and the Resulting Longitude 19 Comparison of Fort Worth Chronometers with the Washington Noon Signals 20 The Latitude Observations 22 Investigation of the Sextant employed 22 The Observations and the Resulting Latitude from Star Observations 24 Latitude Observations of the Sun 25 Geographical Positions 27 Position of the Stone Monument in Court House Square, Fort Worth 27 Position of the Observers co-operating with the Fort Worth Eclipse Party 28 Reports 29 Report of Mr. Leonard Waldo 29 The Observation of the Contacts ' 30 The Diffraction Fringes 30 Description of the Photographic Arrangements 31 Discussion of the Photographs taken to show the Appearance of the Totality 32 Discussion of the Photographs taken to record Evidences of Polarization 35 Professor Pickering's Memorandum relating to these Photographs 35 6 CONTENTS. FAGS Report of Mr. R. W. Willson 37 Description of the Instruments used in the Observations 37 The Observation of the Contacts 37 Description of Plate II 39 Physical Peculiarities of the Corona 39 Report of Professor J. K. Rees 4 1 Preliminary Arrangements 4 1 Description of the Instruments used in the Observations 4 1 Method employed in Making and Reducing the Sun Observations for Latitude 44 The Observation of the Contacts 4<> The Results of the Examination of the Spectrum of the Corona 4 6 Spectroscopic and Telescopic Observations during the Progress of the Eclipse ......... 47 Report of Mr. W. H. Pulsifer 4» Description of the Instruments used in the Observations 48 The Observation of the Contacts 5° The Reversion of the Fraunhofer Lines 5° The Determination of the Height of the Reversion Layer above the Sun's Limb 51 Report of Mr. F. E. Seagrave 52 Description of the Instruments used in the Observations 52 The Observation of the Contacts 53 General Appearance during Totality 53 The Meteorological Observations ■ 54 Visibility of the Corona previous to Totality 55 Reports from other Observers co-operating with the Fort Worth Eclipse Party 57 Report of Professor S. H. Lockett, descriptive of Plate III., at Forth Worth 57 Report of Messrs. H. A. Hill, J. M. Hobson, and L. M. Rush at Allen 58 Report of Messrs. James Giles and J. H. Breeze at Dallas . . > 59 Report of Messrs. J. M. Pearson, John S. Moore, and W. T. Finch at McKinney 59 Report of W. H. Chandler, Esq., at McKinney 60 Report of W. Quarles, Esq., at Bremond 60 Plates 61 Introduction. BY LEONARD WALDO. The " Fort Worth Eclipse Party " consisted of five observers who organized the party with the special end in view of recording such phenomena as might aid in establishing the correct theory regarding the corona which is seen surrounding the sun during the moments of a total solar eclipse. The members were : — Mr. Leonard Waldo, of Harvard College Observatory. Mr. R. W. Willson, of Harvard College. Professor J. K. Rees, of Washington University, St. Louis. Mr. W. H. Pulsifer, of St. Louis. Mr. F. E. Seagrave, of Providence, R.I. The general plan of work to be done was proposed by myself, and after discussion of it by the party and modification in some of its details, it was adopted. According to this scheme, it was determined that the corona should be examined with the naked eye, the telescope, the spectroscope, the polariscope, and that an effort should be made to photograph the corona, and also to take such photographs as might indicate polarization, by placing between the two lenses of cameras double refracting prisms. Further, that we should make such other observations regarding the duration of totality, the times of contacts, and such miscellaneous observations as should be possible without interfering with our corona work, and that, in furtherance of this plan, we should encourage the general observance of the eclipse by such persons as might be available in the country about our station. The following instrumental equipment was provided : — i. A 5-inch telescope of 73 inches focus, made by Alvan Clark & Sons, mounted upon a steady stand and provided with a battery of eye-pieces, giving 58, 92, 150, 260 diameters respec- tively. Loaned by T. P. Smith, Esq., of Waltham, Mass. 2. A 5-inch telescope of 81 inches focus, made by Alvan Clark & Sons, mounted upon a portable tripod stand, with a battery of eye-pieces, giving 61, 100, and 134 diameters respectively. Loaned by Alvan Clark & Sons, of Cambridgeport. 3. A 4-inch telescope of 58 inches focus, made by Alvan Clark & Sons, mounted upon a substantial tripod, and provided with a battery of eye-pieces, giving 35, 75, 155, and 310 diame- ters respectively. Property of Mr. Pulsifer. 8 INTRODUCTION. 4. A 3-inch telescope of 45 inches focus, made by Secretan of Paris, mounted upon a sub- stantial stand. Loaned by E. Ellery Anderson, Esq., of New York City. 5. A 3-inch telescope of unknown make. Property of Mr. Seagrave. 6. A draw telescope of 1.35 inches aperture, provided with an erecting eye-piece, giving a magnifying power of 24.5 diameters. Property of Mr. Waldo. 7. A draw telescope of 2 inches aperture. 8. A spectroscope by Browning, having a dispersive power of ten 6o° flint glass prisms. Property of Mr. Seagrave. 9. A direct vision spectroscope by Browning, with a train of ten 6o° prisms. Property of Mr. Pulsifer. 10. A spectroscope by William Grunow, having two large prisms by Courvoisier of Paris, each side of whose base is 1.5 inches, and which each give a minimum deviation of the D ray of about 48 42', and a dispersion between the extreme red and the extreme violet of 0.0833. The telescopes are of 1.52 inches aperture and 14.0 inches focus, with triple object glasses by Browning. The instrument is provided with micrometric apparatus, and an additional pair of telescopes of 7 inches focus. Property of Mr. Waldo. 11. A single prism spectroscope by Alvan Clark & Sons. Property of Mr. Pulsifer. 12. A portable transit instrument by Temple, of 1.7 inches aperture and 28 inches focal length. Called " Temple Transit." Loaned by D. T. Smith, Esq., of Salem: 13. A portable transit instrument by Troughton and Simms, of 1.6 inches aperture and 23 inches focal length. Called " Fort Worth Transit." Loaned by Professor Greene, of Brown University, Providence. 14. Sidereal Box Chronometer, "Victor Kullberg, 11 78." Property of Mr. Seagrave. 15. Mean Time Box Chronometer, "William Bond & Sons, 1058." 16. Mean Time Pocket Chronometer, "Johnson, 1436.", Property of Mr. Waldo. 1 7. Sextant, " Stackpole & Brothers, 1 707." With stand designed by Mr. Willson. 18. Large polariscope for photographing an artificial eclipse. Loaned by the Rev. E. C. Bolles, Salem: 19. Two large double refracting prisms. Loaned by Professor A. Litton, of Washington University, St. Louis. 20. Photographic outfit furnished by our photographer, Mr. A. Freeman, of Dallas, Texas, consisting of three cameras complete with all accessories for taking the photographs. 21. Telegraphic instruments. 22. Four stop watches. Loaned by the Auburndale Rotary Watch Company. 23. Miscellaneous apparatus, consisting of a barometer, thermometers, small spectroscopes, Arago polariscope, glass scales ruled by Professor W. A. Rogers, eye-pieces, sun-shades, &c. The selection of Fort Worth as the observing station was based upon the following con- siderations : — i. The chances of a clear day were as great as for any point along the line of totality, according to the valuable paper on this subject put at our and other observers' disposal, by the United States Signal Service. 2. The totality would occur with us some fifteen minutes after it had occurred in Colorado, INTRODUCTION. g which would cause the record in Texas to have additional value in questions relating to any rapid change in the corona. 3. There would be, no doubt, plenty of observers who would go to Colorado on such a mis- sion ; but it was not so certain that any other well-equipped party would go to Texas in the summer season. Mr. Willson accordingly left St. Louis, July 8th, and proceeded to Fort Worth to arrange for the rest of the party who were to join him later. Before leaving St. Louis, however, he and Professor Rees had established the " Temple Transit " upon the transit piece in the observatory lately erected by the Washington University at the corner of St. Charles Street and Eighteenth Street. I arrived in St. Louis the morning of July 1 5th, and at once proceeded, with the aid of Pro- fessor Rees, to get ready for the exchange of longitude signals between Fort Worth and St. Louis. In the evening, however, after having completed the first set of transit observations, and upon repairing to the telegraph office, we received word from Mr. Willson that he had been unable to mount his transit instrument as yet; and he further advised our proceeding to join him at Fort Worth, as he was somewhat under the influence of the weather. Accordingly, after a telegraphic comparison of chronometers, we prepared to leave St. Louis on the night of the i6th, which we did, arriving at Fort Worth the afternoon of the 18th of July. Upon presenting the letters we bore to S. W. Lomax, Esq., he immediately tendered us the use of his grounds, and all the facilities his home could offer. The site was extremely favora- ble, about half a mile from the city limits, commanding an extensive view in all directions. We very gladly therefore accepted Mr. Lomax's kind offer, and chose the sites for our instruments to the west of his house. Counting from the north toward the south, the observers Pulsifer, Rees, Willson, Seagrave, and Waldo were in order, and about 40 feet apart. The house-top, which fortunately was flat, afforded an excellent place for the sketchers and for the person who was to call off the seconds remaining of totality. Upon digging to the depth of about 3 feet, we found limestone rock ; and our transit pier, which consisted of four upright scantlings about 3x4 inches, firmly boxed to- gether, rested upon this bed of rock. A gravel and broken stone mixture was then filled in around the scantlings, and the interior of the pier up to the level of the ground was tamped with the same mixture. A wooden block about fourteen inches square made the top of the pier, and a tight box resting on this top was made to set over the transit, and protect it in bad weather. Two posts set about three feet from the E and W pivots held the lamps for illumina- tion of the field of view, without danger of warming the pivots. Within a few feet of this pier, a post was set into the ground to support an artificial horizon at a convenient height for an observer sitting in a chair to use the sextant. We had caused to be reprinted from the excellent " Instructions for observing the Total Solar Eclipse," prepared by Professor William Harkness at the direction of the Superintendent of the United States Naval Observatory, a pamphlet of extracts, including Section I. entire and that part of Section V. included between the sub-heading " The Contacts " and the end of the section. Copies of this reprint were furnished to each of the observers who wished to co-operate with us. A class was at once organized among those willing to aid us about IO INTRODUCTION. Fort Worth, and was exercised in sketching the outlines of the previous eclipses as shown in the various reports, the time of making a single sketch being limited to two and a half minutes. The State press exerted itself in spreading the suggestions made to inexperienced observ- ers, and the reports given in the sequel show that our efforts to interest others in this work were fruitful. Regular routine was at once established. Messrs. Willson and Rees took the meridian transit observations, with the sextant Professor Rees observed the sun daily for latitude, and I observed a Scorpii and Polaris for the same purpose at night. Mr. Seagrave had the meteoro- logical observations to make, and a daily observation of the sun's disc; in the evening, he assisted me. The direct preparation for the eclipse, however, was the most arduous task. The spectroscopic work must all be rehearsed ; and, since it was extremely warm in the sun, it became a trying thing to work more than a few minutes at a time over instruments which were too warm to handle. Mr. Pulsifer joined the party four days preceding the eclipse ; and we immediately chose our assistants from the helpers who had gathered at Fort Worth for this purpose. A heavy storm beginning the Friday before the eclipse, and lasting over Saturday and Sunday, so seri- ously interfered with our preparations that we were fearful Monday would find us not in readiness. Particularly was I fearful about the photographs. Mr. Alfred Freeman could not come to Fort Worth until the Wednesday previous to the eclipse. Thursday and Friday had been fully occupied in rigging up a dark room and testing the chemicals. The cameras were all in readiness, but we had had no chance to try their rates of motion with the improvised clock-works. Nor did there any chance occur until Monday afternoon, and then I found it was too late to make any thing like a good adjustment of its motion ; and so all of our photographs are " moved pictures," and therefore they only indicate the position of the promi- nences without showing much detail. A carefully prepared and detailed scheme dividing the work during the eclipse among the various observers was arranged. By this scheme, the appointment of the instruments named in the preceding list and the work to be accomplished was as follows : — Mr. R. W. Willson, the general telescopic and polariscopic phenomena. Instruments, — Nos. 3, 15, Arago polariscope, hand spectroscope. Professor J. K. Rees, general telescopic phenomena during the partial phases, and the measurement of coronal lines during totality ; the extension of the corona away from the sun as determined spectroscopically. Instruments, — Nos. 1, 10, and a stop watch. Mr. W. H. Pulsifer, spectroscopic observations of the contacts ; Young's reversion layer ; the position of the " 1474" line, and a determination as to whether it joins the sun's disc, is broader at its base than its apex, &c. ; the examination of the prominences for new lines, and the cusps for rays. Instruments, — Nos. 3, 5, 9, n, and the watch, " Jurgensen, 13548." Mr. F. E. Seagrave, spectroscopic observation of the contacts ; the discovery of any new prominence lines, particularly beyond F. ; the micrometric measurement of the line " 1474," and the meteorological observations. INTRODUCTION. 1 1 Instruments, — Nos. 2, 8, 14, and thermometer " Casella, 29563." As for myself, I expected to be fully occupied with the photographs during totality, and with the general work always necessary during the partial phases. Instruments, — Nos. 6, 16, 20, and a stop watch. Mr. A. M. Britton of the City National Bank occupied a position on the house-top, from which the approach of the eclipse could be seen ; and it was his duty to call out in a strong tone the number of seconds remaining of totality, each 1 5 seconds after its commencement, assum- ing the totality to last 2 m 30 1 . He also was to obtain a determination of the duration of the totality with his stop watch, " Ls. Audemars, 1 1855." The class of sketchers were seated by Mr. Britton, and before each one of them a plummet was suspended by a cord, which should serve as vertical reference lines in their drawings. Eight sketchers furnished drawings of the totality ; and four of the drawings have been selected for publication, and are given in Plate IV. Professor Lockett's drawing, which is the most valuable of the naked-eye sketches, was made by a careful draughtsman, and is a capital representation of the corona as I saw it with the unaided eye. In preparing all of our plates for publication, the original drawings were photographed directly on the block. They seem to me to be extremely faithful copies of the originals. Professor Lockett's drawing better gives the effect of an oil painting he has since made, by viewing it from a distance of about ten feet. Phase. Observer. Fort Worth Mean Time. Remarks. Page of this Report. h. m. s. I Contact. L. Waldo. 3 " i8-7 Small telescope. See Report. 3° R. W. Willson. 11 10.7 40 J. K. Rees. 11 11.3 Estimated. Telescopic contact. 47 W. H. Pulsifer. Lost. 49 F. E. Seagrave. 10 44.4 Disappearance of the C line. 10 prisms. 53 II Contact. L. Waldo. Not observed. 3° R. W. Willson. 4 17 6.1 This is probably recorded io* early, as Mr. Willson 40 J. K. Rees. 17 19.4 had been carrying the chronometer beat for 2 m 47 W. H. Pulsifer. 17 15-3 in his head. — L. W. 5° F. E. Seagrave. 53 Ill Contact. L. Waldo. Not observed. 3° R. W. Willson. 4 19 47- See Report. 40 J. K. Rees. Not observed. 47 W. H. Pulsifer. Not observed. 5° F. E. Seagrave. Not observed. 53 IV Contact. L. Waldo. 5 19 14-5 Small telescope. Observed through clouds. 30 R. W. Willson. 19 18.3 40 J. K. Rees. 19 28.5 Cloudy. 47 W. H. Pulsifer. F. E. Seagrave. > Too cloudy to be seen with the spectroscopes. ■] 50 53 Duration of Totality. A. M. Britton. 2 28.75 11 For convenience of reference, I have gathered together the observations which include the element of time ; and I give them in the accompanying abstracts, the first of which relates to the 12 INTRODUCTION. observations made at our own station. The observations made on the day of the eclipse will be found recorded in the reports following the Introduction. The approximate geographical positions of the following observers are given on page 27, and the details of the observations are given on the pages indicated in the last column : — Locality. Observers. I Contact. II Contact III Contact. IV Contact. Duration of Totality. Local Time used. Page of this Report. Allen, Collin Co., Texas. H. A. Hill, ^| J. M. Hobson, \ L. M. Rush. J h. 3 fit. S. 19 8 h. in. s. 4 23 54 A. m. s. 4 25 39 h. nt. s. 5 20 54 tn. S. 1 45 Houston. Slow of Washing- ton signals \k 21m. 58 Dallas, Dallas James Giles, ) J. H. Breeze. } Co., Texas. 3 11 4 19 58 4 21 56 5 20 Dallas. 59 McKinney, Col- lin Co., Texas. J. M. Pearson, \ John S. Moore, \ W. J. Finch. J 3 10 30 4 13 3 2 s 14 30 1 27 See Report. 59 j> j> W. H. Chandler. 3 8 4 14 5 J 4 30 1 2 9 y A 60 Bremond, Texas. W. Quarles. 1 56^ 60 Three of the party left Fort Worth on the evening of the 29th of July, Mr. Pulsifer and myself desiring to arrive at St. Louis as soon as possible, in order to exchange longitude signals with Messrs. Willson and Rees, who remained in Fort Worth some days longer for that pur- pose. I should mention, before closing the descriptive part of the Introduction, that Messrs. Lomax and Britton, in connection with a few other gentlemen interested in the geographical position of Fort Worth, subscribed a fund for the construction of a monument in the Court House Square at Fort Worth, whose position should be made to depend upon the position of our transit pier by means of a survey executed by Mr. Terry, and which may be found on page 27. The inscription furnished to Mr. Lomax for this pier was, — N. Latitude, 32° 43' ig" W. Longitude, ih aim 7J.57 FORT WORTH ECLIPSE PARTY, July 29, 1878. W. H. Pulshtsr. J. K. Rues. R. W. Willson. F. E. Seagravb. Leonard Waldo. and it is here given, so that it may be easily identified at any future date. INTRODUCTION.— FORT WORTH CHRONOMETER ERRORS. 13 CHRONOMETER ERRORS AT FORT WORTH. The sidereal chronometer "Victor Kullberg, n 78," was used as the standard time-piece of the party. At ordinary temperatures, it performs admirably with a rate of about 2 f .6 losing. The extremely high temperatures to which all the chronometers were subjected at our observ- ing station proved, however, destructive to regularity in their rates. The observations for time were made by Mr. Willson, assisted by Professor Rees. The "Brown University" transit instrument was used, with the exception of two sextant sets made July 1 6th and 19th by Mr. Willson. This transit is an old one by Troughton and Simms, of the pattern figured in Bowditch's Practical Navigator. A new level, kindly loaned to us by Messrs. Buff and Berger of Boston, was adapted to the instrument; and, after the date of July 26th, the ordinary web reticule of 5 webs was replaced by one ruled on glass by Professor W. A. Rogers, having three groups of lines of 3, 5, and 3 lines respectively. The corrections to the first series of webs to reduce each one to the mean were, for Circle E, I. II. III. IV. V. -49 s .2 7 - 2 3 s .98 -C/.63 + 24 , .72 + 49 J .io The Equatorial distance between two consecutive lines of the glass reticule was i2 s .633 and between the nearest lines of two consecutive groups 25 s .266, without any sensible variation. The value of a division of the Buff and Berger level I found to be 1^.00 = i".47o at 79°.5 F. which is the mean of ten settings of the Harvard College Observatory Meridian Circle. The extreme heat, the annoyance caused by the numerous insects which fluttered into the observers' faces, or crept under their clothing, or got into positions such that a movement of the instru- ment would jam them under the pivots, or under the level in reversing it, coupled with the fact that our whole observatory was a temporary one, all conspired to render our observations both for time and latitude less accordant than we wished. Still, I think all our determinations are fully within the limit of accuracy imposed by a rigorous conception of the chief objects of our expedition. Rejecting those nights which are not fairly accordant, the following are the time observations used : — H INTRODUCTION. — FORT WORTH CHRONOMETER ERRORS. Time Observations at Fort Worth. Mean Date. 1878. Star's Name. Lamp. Observed Time. Kullberg, 117S. Aa. Bb. Cc. Error, Kullberg, 1x78. Mean Error. k. m. s. s. s. *. h. m. s. r Herculis. w. 17 59 41.32 - .24 + 2.14 - i-73 + 1 43 34-77 « Ophiuchi. w. 18 35 30.16 + .28 + -99 — 1. 21 + 1 43 34-ii 8 Herculis. w. 18 40 42.82 - .02 + 1.15 — 0.96 + 1 43 34-45 d. [i Herculis. E. 19 25 17.12 + .19 + 1.06 + i-35 + 1 43 34-54 h. m. s. July 20.37 + 1 43 34-45 7 Draconis. w. 19 37 5.20 + 12.97 - .76 - 1.47 + 1 43 26.20 j" 2 Sagittarii. w. 19 41 56.20 - 25-95 - .23 — 1.06 + 1 43 26.36 (i- Sagittarii. w. 19 51 21.70 — 20.76 — .21 - .98 + 1 43 27.70 r\ Serpentis. w. 19 58 45.66 - 14-27 - .41 - .92 + 1 43 26.78 « Lyrae. E. 20 16 15.02 + 3- 2 4 - .98 + 1. 18 + 1 43 26.87 July 22.43 + 1 43 26.78 x Herculis. W. 17 59 28.70 + 2.35 + -45 - .70 + 1 43 23.70 a Scorpii. W. 18 5 29.98 - 6.57 + .21 - -53 + 1 43 23.50 A Draconis. W. 18 11 29.58 + 11. 10 + .78 - i-34 + 1 43 23-69 x Ophiuchi. W. 18 35 23.02 — 2.69 + .29 - .49 + 1 43 23.25 a 1 Herculis. E. 18 52 33.82 + 2.40 + .19 + .15 -f- 1 43 23.73 « Ophiuchi. E. 19 12 44.80 - 2.52 + .10 + .50 + 1 43 23.12 July 23.37 + 1 43 23.50 t Herculis. W. 17 59 24.17 4- 1.26 + .40 - 1.84 + 1 43 l6 -94 a Scorpii. W. 18 5 19.61 - 3-42 + -17 -i-37 + 1 43 I 5-4i ri Herculis. W. 18 22 3.14 + 0.50 + .04 — 1.62 + 1 43 16.34 a 1 Herculis. E. 18 52 24.86 - i-i5 + .88 + 1.30 + 1 43 *7-5 2 44 Ophiuchi. E. 19 2 18.52 - 3.27 + -3° + 1.36 + 1 43 16.63 a Ophiuchi. E. 19 12 36.20 — 1.26 + -3° + 1.29 + 1 43 16.87 July 25.37 + 1 43 16.62 a 1 Herculis. W. 18 52 18.00 + .40 + 1 43 9- J 5 v Draconis. W. & E. 19 37 0.08 + -79 + 1 43 9-40 [i Sagittarii. E. & W. 19 49 40.60 + .24 + 1 43 8.93 July 27.40 + 1 43 9.16 8 Draconis. W. 20 55 23.08 + 12.83 + .80 + 5.20 + 1 43 7-49 8 Aquilse. W. 21 2 33-97 - 4.30 + -33 + 2.00 + 1 43 7-52 i 1 Cygni. w. 21 9 39.80 + 4-38 + .56 + 3-20 + 1 43 7.07 j 1 Cygni. E. 21 9 46.76 + 4.01 + .42 -3.20 + 1 43 7- 12 v Aquilse. E. 21 23 42.70 - 3-36 + .23 -2.05 + 1 43 6.26 a Aquilse. E. 21 28 05.47 - 3-5 2 + .19 — 2.04 + 1 43 7-53 July 28.48 + 1 43 7.16 « Scorpii. W. 18 5 20.38 — 19.82 .00 + .14 + 1 43 1.17 r[ Herculis. E. 18 21 43.46 + 2.92 .00 - -15 + 1 43 0.67 % Ophiuchi. W. 18 35 5-44 - 8.08 + -13 + .12 + 1 43 0.81 d Herculis. E. 18 40 9.72 + 0.40 + .21 - .14 + 1 43 °-39 July 30-35 + 1 43 0.74 a Scorpii. W. 18 4 39-95 + 13-57 + .05 + 2.09 + 1 42 56.16 v. Ophiuchi. E. 18 34 49.14 + 5-72 + .20 - 1.94 + 1 42 56.32 a 1 Herculis. E. 18 52 2.04 + 4-57 + -13 - 1.98 + 1 42 5 6 -45 July 31.36 + 1 42 56.31 d Sagittarii. E. 20 53 19.08 + i3-7o - .24 — 2.46 + 1 42 55.99 8 Cygni. W. 21 24 11.96 + 4-73 -1.38 + 3-2i + 1 42 55-65 a Aquilse. W. 21 27 41.58 + 6.52 — 1. 12 + 2.32 + 1 42 5 6 -73 July 31.47 + 1 42 56.12 INTRODUCTION. — FORT WORTH CHRONOMETER ERRORS. There were also made the following — Chronometer Comparisons at Fort Worth. '5 Date. Victor Kullberg, Bond & Sons, Johnson, Date. Victor Kullberg, Bond & Sons, Johnson, 1878. 1178, s. t. 1058, m. t. 1436, m. t. Observer. 1878. 1 178, s. t. 1058, in. t. 1436, m. t. Observer. ^d. h. m. s. h. m. s. ^d. h. m. s. h. m. s. 18.25 15 26 56.5 5 56 50-0 R.W.W. 24.4O 19 3i 39-5 9 38 IO.O R.W.W. 19-35 17 56 20.5 8 22 3.0 L.W. 24.94 8 32 44-5 22 37 I2.0 L.W. 21.38 18 47 30.0 9 4 28.0 k. m. s. R.W.W. 25.08 11 55 58-o 1 59 53-5 h. m. s. R.W.W. 21.49 11 46 2.5 11 45 40.0 L.W. 25-38 9 33 o-o 9 3 1 34-o L.W. 21.94 8 17 6.0 22 33 54-o L.W. 25.4O 19 29 26.5 9 32 10.5 L.W. 22.38 18 43 15.0 8 57 26.5 R. W. W. 26.10 12 26 21.0 2 26 24.0 R.W.W. j 22.43 10 18 20.0 10 17 40.0 L.W. 26.39 19 18 47.0 9 i7 45-° / R.W.W. 22.46 20 47 8.0 11 1 0.0 R.W.W. 26.41 10 3 10.0 10 1 30.0 L.W. 22.89 7 2 4 29.5 21 36 41.0 L.W. 27.42 10 18 30.0 10 16 37.0 L.W. 23-33 17 42 35-5 7 52 10.0 R.W.W. 27.98 23 25 0.0 23 23 0.5 L.W. 23-37 9 8 30.0 9 7 35-8 L.W. 28.93 8 22 49.0 22 12 10.0 J. K. R. 23.42 19 44 47.0 9 55 2.0 R.W.W. 28.99 9 58 0.0 23 47 5-° J. K. R. 23.96 8 5 2 2 9-5 23 40.0 L.W. 29.O2 10 39 25.0 28 24.0 L.W. 24.08 11 45 25.0 153 8.5 L.W. 29.O2 36 30.0 35 13.0 L.W. 24-37 9 11 30.0 9 10 20.0 L.W. 29.41 30.42 20 3 36.0 20 11 0.0 9 5i 6-5 9 54 42.5 R.W.W. J. K. R. Whence we have the — Adopted Chronometer Errors and Rates at Fort Worth. Date. Error "Kullberg, 1178." Rate K. 1178. Date, 1878. Error "Bond, 1058." Rate "Bond, 1058." Date, 1878. Error "Johnson, 1436-" Rate J. 1436. d. 1878 July 20.37 22.43 23-37 25-37 27.40 28.46 30.35 3 J -36 3M7 h. m. s. + 1 43 34-45 + 1 43 26.78 + 1 43 23.50 + 1 43 16.62 + 1 43 9.16 + 1 43 7.16 + 1 43 0.74 + 1 42 56.31 + 1 42 56.12 -3-73 -3-49 -3-44 -3.18 - 1.89 -3-40 -4-47 - 1.8 d. 16.OO I9.OO 22.46 23.42 25.40 26.39 28.93 30.42 + 6? 7* + 16. 8* + 36.17 + 41.71 + 53-03 + 58.06 + 73-5 ! + 81.75 * Sextant Ob- servations. s. + 3-°3 + 5.62 + 5-78 + 5-7 2 + 5.09 + 6.08 + 5-55 d. 21.49 22.43 23-37 24-37 25-38 27.98 29.02 + 8?2 - 4.O - 12.8 - 22.9 -33-o -5i-8 - 2.9 12.6 9.4 IO.I 10.0 7-3 Set by 1058. 1 6 INTRODUCTION.— THE LONGITUDE DETERMINATION. THE LONGITUDE DETERMINATION. Through the courtesy of the officers of the Western Union Telegraph Company and of the Texas and Pacific Railroad Telegraph Company, arrangements were made to transmit longi- tude signals between Fort Worth and St. Louis on the nights of July 15th and July 31st. With this end in view, there was sent to Mr. Willson at Fort Worth the " Brown University " transit instrument and the mean-time box chronometer, " Bond, 1058." At St. Louis, I used the " Temple transit," and the sidereal box chronometer, " Victor Kullberg, 1 1 78," on the even- ing of July 15th; and the mean-time box chronometer, " Dent, 2748," on the evening of July 31st. For the reason previously mentioned in the Introduction, however, an exchange on the night of the 1 5th proved impracticable ; and we are therefore compelled to rely on the one night's exchange of signals of July 31st. The following programme for the observer at Fort Worth will give the method by which the exchange was effected : — LONGITUDE PROGRAMME. Observing Station at Fort Worth. 1878. July 15. Begin with Circle West. 31. „ „ „ East. Observing Scheme. 1. Level and reverse level. 2. Observe five or six stars, two of which are circumpolar (*. e. 8> 65°), observing for level near the middle of the series. 3. Level and reverse level. 4. Be at Telegraph Office at 9.40 Fort Worth mean time, and adjust relay so the back stroke will be strongest, and use this back stroke throughout the longitude work. Exchange the signals as in the scheme following. The 1st night the signals will begin from St. Louis, the 2d night they begin from Fort Worth. 5. Return to observing station, and reverse the telescope upon a circumpolar star for collimation, reading the level before and after reversal. 6. Observe five or six stars as in 2. 7. Level and reverse level as in 1. Telegraph Office at Fort Worth. 1878. July 15. 9* 4S OT Fort Worth mean time. Person using the Line. Messageor Signal. L. W. " Are you ready? " R. W. W. "Yes. Begin." L. W. Rattle at the next 45 s beat of L. W.'s chronometer. Rattle continues from 45* to 55*, and the chronometer beats o* 2 4 6 8 10 12 14 16 18 20 22 24 3° 3 2 34 3 6 3 8 4o 42 44 46 48 5° o 2 4 6 8 s &c, are sent by tapping with the telegraph key for $ m , the last signal being at INTRODUCTION.— THE LONGITUDE DETERMINATION. *7 o s of the 6th minute. Rattle from 5 1 to. 15* of the 6th minute completes the signal, and L. W. then telegraphs the hour and minute of the o s following the first rattle. R. W. W. " Are you ready ? " L. W. "Yes. Begin." R. W. W. Signals precisely as above. L. W. "Are you ready?" R. W. W. "Yes. Begin again." L. W. Signals precisely as above except they continue for three minutes instead of five, and the last rattle occurs 4'" 5 s to 4 m 15*, hour and minute of first signal telegraphed as before. R. W. W. " Are you ready ? " L. W. "Yes. Begin again." R. W. W. Signals as in the preceding clause. R. W. W. "Good-night." L. W. " Good-night." Two break-circuit keys, precisely similar in design and size, were constructed ; and one was used by Mr. Willson at Fort Worth, the other by myself at St. Louis. Both observers having repaired to the Telegraph Office upon the night of July 31st, the following comparisons were made, in accordance with the above programme : — Time of sending signals from Fort Worth by ' 'Victor Kullberg, 1178," s. t. h. 711. s. 19 56 30.0 57 0.0 57 30.0 58 0.0 58 30.0 59 0.0 59 30.0 20 0.0 30.0 1 0.0 20 1 30.0 Time of receiving signal at St. Louis by '*Dent, 2748," m. t. h. m. s. 10 6 27.9 6 57-8 7 27.8 7 57-8 8 27-5 8 57-4 9 27-3 8 57-3 10 27.2 10 57-i 10 11 27.0 II. Time of receiving signals at Fort Worth by " Victor Kullberg, 1 178." Time of sending signals from St. Louis by " Dent, 2748.' h. m. j. 20 s 3-5 5 33-6 6 3-65 6 33-7 7 3-85 7 33-9 8 4.1 8 34-15 9 4.2 9 34-35 20 10 4-45 h. m. s. 10 15 O.O *5 30.0 16 O.O 16 30.O 17 O.O *7 3O.O 18 O.O 18 3O.O 19 O.O 19 30.O 20 O.O i8 INTRODUCTION.— THE LONGITUDE DETERMINATION. III. Time of sending signals from Fort Worth by " Victor Kullberg, 1178." k. m. s. 20 Time of receiving signals at St. Louis by (< Dent, 2748." 20 13 0.0 30.0 14 0.0 30.0 is 0.0 30.0 16 0.0 IV. Time of receiving signals at Fort Worth by " Victor Kullberg, 1178." h' m. s. 19 5-9 19 36.I 20 6.1 20 36-15 21 6.2 21 36-3 22 6-45 h. m. S. IO 22 55-i 23 20.0 23 55- 1 24 25.0 24 54-9 25 24.8 IO 25 54-6 lals f -om St- Louis b h. M. s. 10 29 O.O 29 3O.O 30 0.0 30 3O.O 3i O.O 3i 30.O IO 32 O.O The time observations made at Fort Worth by Mr. Willson and Professor Rees will be found as a continuation of the regular series made at Fort Worth. The transit used at St. Louis is a very nicely finished one, made by Temple, of Boston. It has an aperture of 1.7 inches, and a focal length of about 28 inches. The value of one division of the level I found to be 1*00 = 8".68 at 74°.8, which is the mean of ten settings with the Meridian Circle of the Observatory of Harvard College. The corrections necessary to reduce each of the five webs in the reticule to the mean web are as follows for Circle East : — I. II. III. IV. V. + 46^.082 + 23^.073 -o r .o5i - 22^890 -46 f -482. The St. Louis observations were very imperfect previous to going to the Telegraph Office since I was hardly able to get the instrument approximately into the meridian, and its level and collimation adjusted, before it was time to repair to the Telegraph Office. The observations are as follows : — Observed Time. Date. 187S. Star's Name. Lamp. Bent, 2748, St. Louis m. t. Aa. Bb. Cc. Error, Dent, 2748. d. h. m. s. s. s. S. in. s. July 31.4 H Herculis * E. 9 6 18.18 — 1. 12 — 0.22 - 0.36 + 2 54.88 i^j 1 Draconis E. 9 8 33-78 + 9.68 - o-57 - 1. 05 + 2 55-7° y Draconis * E. 9 18 18.64 + i-93 - 0.39 - 0.51 + 2 54-99 r Aquilse * W. 11 22 27.36 - 2.84 — 0.42 + 0.32 + 2 54.82 « Cephei W. " 36 54-63 + I5-25 — 1.70 + i-47 + 2 55.69 £ Cygni * E. 12 3 1 43-36 + 4.28 - 0.35 - 0.38 + 2 55.12 ( Pegasi * E. .12 40 20.18 + 8.32 - o-35 - o-35 + 2 55.21 (3 Cephei E. 12 51 40.63 - 36-38 — 1.02 - o-97 + 2 53.63 s Pegasi * E. 13 1 58-I7 + 11.89 - 0.44 - 0.33 + 2 55.22 INTRODUCTION.— THE LONGITUDE DETERMINATION. iq The observations show a disturbance of the instrument after the observation of * Cephei, which at first was so irregular in its nature that several stars observed with lamp W have been rejected. Solving for a new azimuth after * Cephei, and taking the mean of the time stars, which are denoted by a (*), we have, after reducing the errors to the same instant,* — The error of Dent, 2748, July 31^.47, is + 2 m 55*.oi ± 0^03. Now, since the signals were sent by hand, we shall arrive at a more probable comparison of the chronometers, if we take the means as the comparisons stand than if we use the coincidences only. Whence we have, — Victor Kullberg, 1 178. Dent, 2748. h- m- s. h. m. s. From I. 19 59 0.00 = 10 8 57.46 II. 20 7 33.95 = 10 17 30.00 III. 20 14 30.00 = 10 24 24.94 IV. 20 20 36.17 = IO 30 3O.OO and the final mean, which will be free from errors of time of transmission and armature time, — St. -L. m. t. Victor Kullberg, n 7 8(s. t.) Dent, 2748 (m. t.) «• *»• J. h. m. s. 1878. July 3i rf .44. 20 10 25.030 = 10 20 20.600 And we have for the errors of the two chronometers, — For July 31^.44 St. Louis m. t. = July 31^.42 Fort Worth m. t. h. m- s. V. Kullberg 1 1 78 = + 1 42 56.22 (s. t.) Dent 2748* = +02 54.96 (m. t.) Whence we have the final comparison, reduced to their respective meridians, but still affected by the personal equations of the observers, R. W. Willson and myself, — Fort Worth sid. time. St. Louis mean time. ^. '«• J- k. m. s. 1878/ July 31^.4. 18 27 28.81 = 10 17 25.64 Assuming the right ascension of the mean sun for St. Louis for this date to be 8* 36"* 53^.33, and applying the difference of the personal equations of Mr. Willson and myself as determined by transit observations after our return to Cambridge, and found to be o J .2 7 additive to the observation of an equatorial star by Mr. Willson, we have then the difference of longitude between the observatory of Washington University, St. Louis, and the transit pier in the rear of Mr. S. W. Lomax's residence, near Fort Worth, Texas, is, — o h 2% m 3 1'. 3 1 and if we apply the small correction necessary to reduce the position of the transit pier of the Washington University Observatory to the position of the transit pier occupied by Pro- fessor William Eimbeck in the St. Louis — Washington longitude campaign, as described by Professor William Harkness (Washington Astr. Obs. 1870, App. I.), and adopt the longitude of Professor Eimbeck's pier, as described in the paper mentioned, we have, — Fort Worth Transit, west of Washington University Observatory o h 28"* 31^.31 Washington University Observatory, west of Professor Eimbeck's pier o m 0^.13 Professor Eimbeck's pier west of United States Naval Observa-"! h tory at Washington, D. C. J o 52" 36^.90 And finally, the longitude of our observing station at Fort Worth is west 1* 2\ m 8^.34 of the dome of the United States Naval Observatory at Washington, D. C. * Hourly rate deduced from sextant observations July 27th and August 5th is + o*.o6i. 20 INTRODUCTION. — THE LONGITUDE DETERMINATION. We desired to receive the Washington noon-signal on the morning of the 29th of July, in order to compare our chronometers with the Washington clock. Professor Rees, who had this matter in charge, embodies the result of our efforts in this direction in the following words : — " In the signal received, a rattle announced the coming of the taps; then, after a pause of a few seconds, a tap was sounded for every second, till exactly at twelve o'clock Washington mean time a double-beat was tapped. The double-beat was continued to exactly one minute after twelve o'clock Washington time, then was changed into a single tap for each second, which lasted for several seconds, when a rattle was sounded, and the signals ended. We had then the means of two comparisons, one at twelve o'clock Washington mean noon, and another one minute later. "July 29th, 1878, a.m. — Comparison of chronometers to show whether Bond, 1058, suffered any change of rate in the carrying to and from the telegraph office : — Mean time, Bond, 1058. Sidereal time, Kullberg, 1178. k. m. s. h. m. j. 22 12 IO.O = 8 22 49.O " After receiving the signals, we have — Bond, 1058. Kullberg, 1178. h. m. s. h. m. s. 23 50 0.0 = 10 o 55.0 "The mean solar interval between the two comparisons is i* 2>7 m 50 J .o, which gives, reduced to sidereal time, i* 38"* 6^.07. The noted sidereal interval is seen to be 1* 38" 6 j ;o. But 1* 37 m 5o'.o is not the true solar interval of the comparisons, but is too great by the amount that 1058 gains in that interval. Since the rate of 1058 is + 5^.55, it would gain in the interval 0.4 1 ; hence, the true mean solar interval is 1* 37™ 49^.6= 1* 38"* 6^.07 sidereal time. The noted sidereal interval, corrected for the loss of Kullberg in the interval, gives 1* 38** 6^.23. Thus, we seem to have evidence of a small change of rate in 1058. If we consider, as we think we have a right to do, that the change occurred partly in coming down and partly in going back to our station, then the change that had occurred up to the time of the signals will be small. " July 28.94. — Signals from Washington. Washington. Bond, 1058. h. /«. j. h. m. s. o o 0.0 10 40 8. (comparison not good.) 1 0.0 41 9. (comparison better.) " The signals were constantly broken in on by the agents along the railroad line, and I was very much disturbed. "Assuming the error of Bond, 1058, at the time of comparison, to be 4- i m 13^.57, we com- pute the following : — West of Washington. k. tn. s. Longitude from first comparison .... 1 21 5.57 Longitude from second comparison ... 1 21 4.57 " On informing Mr. Todd, at Dallas, that I had not received a satisfactory set of signals, he answered that he had received uninterrupted ones, and would send me signals, using his side- real chronometer. INTRODUCTION.— THE LONGITUDE DETERMINATION. 21 " He preceded his signal by a rattle of a few seconds, and then sent a signal every ten seconds by his chronometer, the time of reception being noted on Bond, 1058. The signals closed by Mr. Todd sending me the time of beginning, and his error on Washington sidereal time. " The following is the record : — Bond, 1058* k. m. s. 23 o 21.5 = 3I.O = 41-5 5I.O 6l.O = I IO.5 = 20.6 = " Reducing all these comparisons to the same instant, we get : — h. m. s. 23 o 21.5 = 21.03 = 21.56 = [Mr. Todd's chronometer was 21.08 = g m 49^.3 fast.] 21. 11 = 20.64 = 20.76 = " Taking the mean, applying the Bond and sidereal chronometer corrections, and reducing Washington sidereal time to the corresponding mean time, we have : — Washington mean time. Fort Worth mean time. h. m. s. h. vi. s. 24 20 I5.79 = 22 59 7.4I Hence the difference in longitude is, by the Washington signal : — h. m. s. I 21 8.38." Washington h. 8 Sidereal Chronometer. m. s. 59 0.0 10.0 20.0 30.0 40.0 50.0 9 00.0 we get : — h. 8 7ft. 59 s. O.O 22 INTRODUCTION.— THE OBSERVATIONS FOR LATITUDE. THE OBSERVATIONS FOR LATITUDE. The following observations were made with the sextant " Stackpole and Brother, 1 707," by myself. The sextant is of the pattern originally designed by Professor William Harkness, and described by him in the " United States Naval Observatory Eclipse Report," of 1869; with the exception that it has an arc of seven inches radius instead of six inches, as there mentioned. It was received directly from Mr. Stackpole's hands, and was found to be in excellent adjustment throughout. A method of determining the eccentricity of a sextant, where the conveniences of a large circle are at hand, occurred to me, and I have adopted it in the following investigation ; using the large Meridian Circle of the Harvard College Observatory for this purpose, through the courtesy of Professor W. A. Rogers. I replaced the finding level on the index-arm of the sextant by another, so sensitive that an inclination of 10" would displace the bubble about 0.03 inches. The sextant was then firmly secured to the meridian circle, so that the plane of the sextant should be, as nearly as-practi- cable, parallel to the plane of the circle. It then became a simple matter to set the vernier of the sextant at consecutive io° grad- uations of the sextant-limb ; and the mean readings of the meridian circle, when the sextant- level is brought to the same position by moving the circle in declination, gave the true intervals between the zero of the sextant-limb and the points at which the vernier was suc- cessively set. Putting As = the sextant correction for any reading 22, E = the number of degrees from the diameter of maximum eccentricity to the zero of graduation of the sextant, counted in the direction of the graduation, e = the maximum eccentricity expressed in seconds of arc, we have by the usual formula, with all necessary accuracy, an equation of the form — As = e sin (z + E) for each interval determined. Also, As = e cos E sin z + e sin E cos s, whence we form the normal equations, — 2 sin z As = 2e cos E sin 2 z + 2e sin E sin z cos z ; 2 cos z As = 2e cos E sin z cos z + 2e sin E cos 2 z. Proceeding in the manner indicated, I found for the sextant under consideration, — e= 22". 1, E = i4°.o; whence we derive the following table of corrections to any observed angle 2: — INTRODUCTION.— THE OBSERVATIONS FOR LATITUDE. 23 As z Az - 5.35 10 - 8-99 20 - 12.35 30 - 15-35 40 - 17-87 5o - 19.88 60 - 21.23 70 -21.99 Referring these angles to the sextant-limb, and making the correction o" at the zero of the graduation, we have, denoting the observed sextant reading for an angle measured by «, — u> Ao> 0> Am - 0.0 10 - i-9 20 - 3-6 30 - 5-4 40 - 7-o 50 - 8-7 60 - 1 0.0 70 - n-3 80 - 12.5 90 - 136 100 - 14-5 no - 15-3 120 - 1.5-9 In the observations for latitude, the fourth column indicates the reversal of the artificial horizon cover, I. and II. being the ends respectively nearest the observer. The fifth column indicates the number of double altitudes corresponding to the mean of the times and double altitudes given in columns six and seven. Column eight contains the correction to my pocket chronometer, "Johnson, 1436," in Fort Worth local mean time. Column nine contains the index correction, which always depends upon from six to ten observations made either between the sets of double altitudes taken in reversed positions of the horizon glass cover, or else immediately following the double altitude observations. An eye-piece magnifying about seven diameters was used, and after the date of July 24?3 a stand was employed to support the sextant. Mr. Seagrave recorded the observations at my dictation. The resulting latitudes in the case of « Ursse Minoris were obtained by using the tables published in the supplement to the "American Ephemeris and Nautical Almanac for 1878," for the factors A, B, C, D. Since the hour angle of the star is nearly 6* in the observations made below, we shall not introduce sensible error by assuming the motion of a Ursa? Minoris in altitude to be proportional to the time for the short intervals used. Accordingly, the observations have been taken in sets of five, and their means used in the reductions. In the case of the circum-meridian altitudes of a Scorpii, since the zenith distance is 67 , we shall not require to consider the remaining terms of the formula (following Chauvenet's notation), — k x = h a + Am a in which h the mean of the observed altitudes corrected for index error, eccentricity, and refraction. A = cos <£ cos S cosec £,. £ = the assumed zenith distance. m = the mean of the values of m. 2 sin 2 J t m = sin 1" ' , where t is expressed in sidereal seconds, and indicates the distance from the meridian. 24 INTRODUCTION.— THE OBSERVATIONS FOR LATITUDE. And, accordingly, the resulting latitudes have been computed by the formula, k t = h a + Am Q . The' assumed position of the observing station is \ = i* 2i OT .o W. of Washington, = 32 44'.4 N. ; and the star places adopted are — 1878 July 25 d .4 a Ursse Minoris \ h \a t m 1&.6 + 88° 39' 29".2 a Scorpii 16 21 59.6 — 26 9 49 .4 Sextant Observations of a Scorpii for Latitude. Date. Bar. Ther. Horizon No. Obs. Chronometer "Johnson, 1436.'' Sextant "Stackpole,i707" At, "J. 1436." Index. Resulting Latitude. 1878 July 23.34 29.28 81.7 24-34 2 5-33 29.12 83.0 29.00 82.0 8 8 8 8 8 o 25 2 10 4 20 5 17 6 33 7 5° 8 52 8 11 17 8 12 14 8 15 16 8 17 49 8 18 48 8 20 43 8 22 10 8 24 3 8 10 50 8 12 12 8 14 8 8 IS 7 8 16 s 8 16 58 8 19 33 8 21 20 8 25 30 8 26 45 8 1 32 8 5 So 8 6 55 8 9 26 8 12 38 8 16 20 8 17 57 8 20 30 8 22 10 8 23 13 62 4 15 62 6 35 62 9 10 62 10 o 62 12 o 62 14 o 62 13 50 62 14 20 62 15 20 62 15 50 62 15 50 62 15 IO 62 14 o 62 12 20 62 II 30 62 16 30 62 16 O 62 15 O 62 14 20 62 15 O 62 14 62 12 62 10 62 3 62 I 62 15 O 62 l6 IO 62 16 26 62 16 O 62 15 30 62 11 35 62 10 o 62. 6 10 62 4 o 62 1 45 4- o 12.4 - 1 56 + O 22.6 + o 32.5 •32 44 21.6 -33 - 10 -32 44 25.9 -32 44 45.9 INTRODUCTION. — THE OBSERVATIONS FOR LATITUDE. 25 Sextant Observations of a Urs^e Minoris for Latitude. Date. Bar. Ther. Horizon No. Obs. Chronometer "Johnson, 1436." Sextant "Stackpole,i7o7" At, J. 1436. Index. & Resulting Latitude. d. 1878 July 21.46 in. 1 5 h. m, s. 11 1 57.1 / // 65 24 50 tn. s. - 8.5 / // - 2 37 n — II a 1 II 32 44 35-2 21-47 22.41 29.28 80.7 11 1 5 5 11 16 26.6 9 5 1 47-8 65 35 3 64 38 30 - 8.4 + 4.5 - 2 37 - 2 31 - II - 10 44 i3-9 44 8.3 22.42 11 5 9 59 59-6 64 45 4 + 4.5 - 2 31 - 10 44 42.8 23-37 29.28 8l.0 1 5 8 43 31.6 64 5 + 12.8 - 2 29 - 10 44 4i-7 23-38 11 5 8 52 4.6 64 4 19 + 12.9 - 2 29 - 10 44 21.7 24-37 29-34 83.O 1 5 8 44 8.0 64 1 45 + 22.9 - 2 29 - 10 44 3 - 1 24.38 11 5 9 5 i-2 64 13 20 + 23.0 - 2 29 - 10 43 5i-9 25-37 25-38 26.38 29.OO 29.04 82.O 8l.O 1 11 1 5 5 5 8 47 11.6 8 59 8.0 8 56 22.6 64 5 56 64 13 3° 64 14 15 + 33.0 + 33.1 + 40.3 - 3 43 - 3 3 - 2 47 - 10 - 10 - 10 43 57-9 44 10. 1 44 3 2 -o 26.39 11 5 9 16 10.6 64 26 3 + 40.4 - 2 47 - 10 32 44 1 1-8 The observations of a Ursae Minoris are more numerous than those of a Scorpii ; but since the observations of a Scorpii were made after the stand was in use, and I had become better accustomed to the insects which, attracted by the light, caused considerable annoyance during the observations, I shall give equal weight to each of the two stars ; and we have, from star observations, the latitude of the transit pier of our observing station is, — Whence we have — By 60 Double Altitudes of a Ursae Minoris 32 44' i8".8 By 35 Double Altitudes of a Scorpii 32 44 31 .1 $ = 32° 44' 25". The following sun observations were made and reduced by Professor Rees at my sugges- tion Civil Date. 1878. Corrected. Ear. Therm. Number Qbs. Means of Bond, 105S, A.M. Means of Q 2 alt. Means of Bond, 1058, P.M. Index Correction. Eccen- tricity. Deduced 0. F. k. »Z. s. / h. tn. s. / n II o / it Sat. July 20 „ 20 29.56* 29.56* 96* 96* 6 5 10 34 5-3 tt 132 20 129 O I 46 8.0 - 2 23.I - 2 I4.9 ~ 16.3 - 16.3 | 32 44 io.gof Mon. „ 22 29.36 96 20 10 22 56.71 127 30 I 50 28.76 — 2 23.O - 16.3 32 45 19- | Tues. „ 23 „ „ 23 29-34 29.24 9 1 92.2 10 10 10 24 5.79 127 40 126 30 I 52 38-59 - 2 33.O - 2 41.3 - 16.3 - 16.3 | 32 44 53-8f Wed. „ 24 29.25 100.8 20 10 25 15.2 127 50 I 48 38.74 - 2 34-4 - 16.3 32 45 44- Thurs. „ 25 Total Observations 29.II 98.8 24 10 26 40.47 128 5 I 47 28.08 - 2 34.I5 - 16.3 Mean 32 45 5 2 -6 95 32 45 12.06 * Assumed. t These have been deduced by using the formulae for single altitudes of the sun and taking their mean (Chauvenet, Vol. I. p. 230, University edition). tf This observation is of the sun's, lower limb. 26 INTRODUCTION.— THE OBSERVATIONS FOR LATITUDE. Although the observations agree well inter se, yet I am apprehensive there is a systematic error arising from some or all of the following causes : — i° The intense heat would affect the horizon, the sextant, and the observer. We have no means of determining the effect of heating the sextant so hot as it became in the course of these observations, and I am of the opinion that the results would not be uniform for consecu- tive days. 2° The sun's hour angle was so great that small errors of observation would be far more perceptible than when the sun was near the meridian. 3° The same thing which obliged Professor Rees to observe the sun at a larger hour angle than desirable also obliged him to use the extreme limits of the sextant graduation. 4° The chronometer rate might be irregular under the influence of the heat, and the hour angles be in error for that reason. < It was desirable, however, that solar observations should be made ; and, until the stand came into use for the star observations, it was highly probable that the solar observations would be better than the stellar observations. I am now of the opinion, however, that we shall have a more accurate determination of if we use the value deduced "only from the observations of a Scorpii and a Ursae Minoris. INTRODUCTION. — GEOGRAPHICAL POSITIONS. 27 GEOGRAPHICAL POSITIONS. Position of Fort Worth, and the Approximate Positions of the Towns mentioned in the Reports of other Observers. In order to connect our observing station with some permanent monument in the City of Fort Worth, the City Engineer, Mr. J. C. Terry, made the following survey, in which "station i " designates the wooden transit pier in the rear of Mr. Lomax's residence, and " station 8 " designates the stone monument in the Court House Square: — No. of Station. Angl Course. Distance. Feet. I o° O' North. 200 2 46° 30' Right. N. 46° 30' E. I700 3 22° O' 99 N. 68° 30' E. SOO 4 21° 30' JJ East. 2IO 5 90° O' Left. North. 133° 6 29° 45' 99 N. 29° 45' W. 2800 7 9O 0' Right. N. 6o° 45' E. 400 8 56° 3i' jj S. 63° 15' E. 130 From these data, Mr. Terry finds the monument to be, — North 5454.3 feet, and east 982.1 feet of the transit pier. If now we assume that in latitude 32°45' north, 60" measured on the parallel is 5152.2 feet and 60" measured on the meridian is 6063.6 feet, we have the position of the monument with reference to the transit pier to be, — In latitude + 53".7 In longitude - n".5 = - 0^.77, whence the position of the monument in Court House Square is, — Latitude north 32° 45' 19" Longitude west of the centre of the dome of the United States Naval Observatory at Washington, D.C., h. m. s. I 21 7.57. Mr. D. P. Todd, having made an examination of the existing maps of Texas, kindly informed me that the map issued by the Houston and Texas Central Railroad Company is as nearly correct as any he had seen. I have therefore, as the best available means of locating the observers co-operating with us, measured upon this map the following positions : — 28 INTR OD UCTION. — GEOGRAPHICAL POSITIONS. Place. Longitude, West of Washington. Latitude North. o ' h. m. s. / Hearne r 9 32.O =1 l8 O 30 56.6 McKinney 19 33.I = I 18 12 33 14.1 Allen *9 36.3 = I 18 25 33 6.1 Bremond 19 38.6 = I 18 34 3i 10.6 Dallas 19 43.4 = I 18 54 3 2 48.6 Fort Worth 20 15-5 = I 21 2 3 2 46.0 If we assume there is a systematic error in these positions equal to the error in the dis- placement of Fort Worth, we have a systematic correction of — + 5.6 J in Longitude, — 41" in Latitude, to be applied to each of the positions given above. Reports. REPORT OF MR. LEONARD WALDO. In the division of labor among ourselves for the observations on the day of the eclipse, it seemed expedient that I should become responsible for the photographic records made during totality. This, with that general oversight over all the observations which it was necessary for some one to assume, in order that our work should be complete and symmetrical as a whole, was thought to be quite as much as could be undertaken by one observer. I shall preface the discussion of our photographs, however, with the mention of a few miscellaneous observations I found it convenient to make. Telescopic and other Observations. A few moments previous to the first contact, I placed a blanket on the ground, and arranged over it a camera tripod in such a manner that, lying flat on my back, I could without exertion have an easy gaze at the sun. I then requested Professor Alexander Hogg, of the Texas Agricultural and Mechanical College, to be ready at a given signal, which we had previously rehearsed, to stop the stop-watch I had set with my pocket chronometer, " Johnson, 1436," and compared as follows : — Johnson, 1436. Stop-watch. h. m. s. m. s. 3 7 O.O = O O.O At 3* S m , I lay down upon the blanket, and placed an umbrella so that I was for the most part protected from the sun's intense rays, but still had full opportunity to point my hand-tele- scope of 1.35 inches' aperture upon the sun. I had adjusted to this telescope a terrestrial eye- piece magnifying 24.5 diameters, and a neutral tint shade-glass sufficiently dark to occasion no uneasiness of the eye after several minutes' gazing at the sun. I now rested the object-glass end against the camera tripod, and examined the sun's limb at the expected position angle of first contact. The definition was superb. The edge was hard, sharp, and the curve unbroken. The mottled appearance was well marked. There was not a 3° REPORTS. — REPORT OF MR. LEONARD WALDO. spot visible. As the computed time of contact drew near, I directed my attention more closely to the computed position angle of contact on the sun's limb, and presently recognized a sharp deviation from the convex curve which had previously bound the sun's limb. This phase was recorded by the stop-watch as 4 OT 25%', which would be by Johnson, 1436, 3* 1 i m 25^.5. My note at the time was that this " contact may be 2 1 late." But after seeing how the fourth contact appeared in my telescope, and comparing the last instant I was able to detect a varia- tion in the sun's outline at the fourth contact with the instant at which this outline was as much broken as it was when I observed the first contact, I am confident that 1 2 s would be none too large an estimate of lateness. I would give, therefore, the time of first contact to be 3* n OT 13*. 5 ; and I think this is liable to an uncertainty of 2 s . Immediately after this observa- tion, I made the following comparison : — Johnson, 1436. Kullberg, 1178. h. m. s. h. nt, s. 3 15 S4-o = 13 28 4i-o The error of Johnson, 1436, was — 5^.2 on Fort Worth mean time, and we have the Fort Worth mean time of first contact = 3* 1 i m 18^.7 ± 2 s . The second and third contacts I could not observe without conflicting with other duties. The fourth contact was observed in the same manner as the first, with the exception that I used the beat of my pocket chronometer Johnson, 1436, instead of the stop-watch, which was inadver- tently stopped by Professor Hogg before the contact took place. For some unexplained reason, Johnson, 1436, had stopped previous to the second contact. I reset it immediately, and compared it after the fourth contact as follows : — Johnson, 1436. Bond & Sons, 1058. h. m. s. //. m. s. 5 2 5 °-° = S 2 S 59-4 and I observed the last contact at 5* 19" 30^.0 by Johnson, 1436. Whence we have the Fort Worth mean time of fourth contact = 5* 19" i4*.5, making due allowance for the error of Bond, 1058. Immediately previous to the commencement of totality, I stood with my back to the sun, looking at the face of a stop-watch, and waiting for the final disappearance of the crescent of light. In close proximity to this phase, but whether 2 s before or not I cannot say, I glanced toward the ground, my attention drawn by a fluttering motion of something, I did not know what. Every eye but mine was toward the eclipsed sun. I quickly removed the cap of the camera-box to make an exposure, and again looked at the ground away from the eclipsed sun. What I had suspected before I was now certain of. There were flitting shadows moving swiftly toward the east and south-east. They seemed to me regular in their occurrence, both as to time and distance apart ; but how far apart they were or how many occurred in a second I can- not form an estimate, other than that they seemed like dark crests to waves, and might be a hundred feet apart, occurring three in a second. Of the existence of this phenomenon I am posi- tive ; but the estimates of magnitude and frequency of these shadows should be received with the greatest caution. I suppose this to be an observation of the shadows ordinarily seen at the second contact of total solar eclipses. REPORTS.— REPORT OF MR. LEONARD WALDO. 31 During the progress of the eclipse, I examined the sun's disc with Professor Rees's tele- scope, with a diagonal reflector of plane glass and various eye-pieces. My conclusions were as follows : — i° The faculse are seen clearly up to within say 2" of the moon's limb. I think they are sharp to the very edge. The moon's limb has something of the " boiling " appearance ; and I am not sure that the sun's photosphere is sharply denned in the jagged edges of the moon's advancing limb. 2 There seems to be a minute point of light separated from, but in the line of, a prolonga- tion of one of the sun's cusps. (This was after second contact.) I do not know whether this is a phenomenon similar to that which produces " Bailey's Beads " or not. The Photographic Work. i° Preliminary Preparations and Description of the Cameras. Guided by own experience, and strengthened in my opinion by the kindly advice of Mr. D. C. Chapman, who is well known as the assistant who has aided Mr. Rutherfurd in the produc- tion of his exquisite astronomical photographs, I thought that we could find some local photog- rapher who would do the necessary work under proper supervision. We were not able to secure the use of a telescope properly corrected for the chemical rays, and so we depended upon the use of large cameras instead. It occurred to me that we might use an improvised mounting of the following description for these cameras to supply the place of a clock-work motion. The bases of the cameras rested upon a board, one end of which was connected by a hinge to a square wooden plunger which moved vertically in a smooth fitting wooden casing which was firmly secured to a post. To the eastward of this post another post was placed, so that the board, hinged to the plunger at its west end, would rest its east end upon the second post, and be free to slide on this post as its west end followed the plunger in its vertical motion in the cylinder. On the top of their respective boards, the cameras were so arranged that motion in azimuth could be communicated to them by means of a screw controlled by a cord held in the hand. The mode of manipulating the apparatus was as follows : the vertical box being filled with dry sartd, an adjustable opening in its base was sufficiently opened to allow the descending plunger to follow the upper surface of sand on which it rested, with the angular velocity (measured with a radius equal to the distance from the second and eastern post to the lower end of the plunger) of the sun in altitude. Then it was hoped there would be no diffi- culty in giving the necessary slow motion by hand to the large screw-head to counteract the sun's motion in azimuth. As the day of the eclipse drew near, however, it became apparent that there was so much to be done that our time of preparation was all too short. The storm of July 27th and 28th completely prevented any trial of our improvised mountings, and only after the commencement of the eclipse on the 29th was the sun in such a position that it was possible to experiment upon the rate of the two motions. As totality drew near, I found that it would be impracticable to make more than a rough adjustment of either motion in the time remaining ; and so I decided to give up the azimuth motion altogether, and allow the plunger to follow the outflowing sand, with the idea that the motion of the sun in a straight line on the 32 REPORTS. — REPORT OF MR. LEONARD WALDO. plate would present fewer difficulties in determining the position angles of the prominences, than if there was an undetermined displacement in both altitude and azimuth. As for the camera lenses themselves, they were as follows : — i° A Dallmeyer lens described as 6.D. The back combination only of this lens was used, which is of 48 inches focus. 2 A Ross Cabinet portrait lens, No. 3. 3° A Tench portrait lens of 6 inches' focus and 2.75 inches' aperture. Between the lenses of 2 there was inserted a compound prism for polarizing light, made by Duboscq. This prism is composed of two bi-refracting prisms, so turned as to combine the refraction of each : they are achromatized with prisms of crown glass. The axial ray is the extraordinary one. The longer diagonal of the rhomb was nearly vertical, consequently the two images would be nearly horizontal. Between the lenses of 3 there was inserted an ordinary double refracting prism with the shorter diagonal of the rhomb nearly vertical, consequently the two images would be nearly vertical. Both of these prisms were loaned by Dr. A. Lytton, of Washington University. These last two cameras were thus provided at the suggestion made by Professor E. C. Pick- ering in his report to the superintendent of the United States Coast Survey upon the observa- tions of the total solar eclipse of Dec. 22, 1870. At four o'clock, Mr. Freeman, our photographer, retired to the dark room, and prepared three plates, one for each camera ; and at my signal, about two minutes before totality, he placed them in their respective cameras, and again retired to the dark room. Two baths were used in coating the plates, and Mr. Freeman at once proceeded to prepare two more plates. During totality, the exposures were made without mishap, and the results are tabulated below : — Number of the Negative. Number of the Camera. Time of Commence- ment of Exposure, counted from the Beginning of To- tality. Duration of Exposure* Remarks. I 2 3 4 5 1° 1° 1° 2° 3° I s 62 1 130* 2 s 2 s 6o* 65' If i45 J i4S' Shows 3 striae. Coronal extension, / = 120 is 23'; for/ = 300 is 25'. Shows the 3 striae of No. 1 and 3 ad- ditional ones. Coronal extension, p = 120° is 25'; for/ = 300 is 23'. Shown in Plate I., and is the best of the three negatives. Shows the 6 striae of No. 2, and additional ones. Images badly defined and over-exposed. Images fairly sharp. Owing to the short exposure of negative No. 3, the details are less obliterated than in the case of Nos. 1 and 2. Mr. S. B. Wells, of Boston, very kindly made for us a somewhat enlarged copy of this negative, which was afterwards still further enlarged by Mr. Edwards of the Heliotype Company, Boston, and is shown in Plate I. The negative was taken with a Tolles 4-inch objective and extremely oblique light : as a consequence, there are visible scratches on REPORTS. — REPORT OF MR. LEONARD WALDO. 33 the glass plate, which with direct illumination would not have been visible. The unfortu- nate motion which is shown in Plate I., in negatives i and 2, is much more conspicuous, owing to their longer exposure. It will be noticed that the images of the sun's edge at the commence- ment and the end of the exposure are fairly outlined ; and it will suffice for our necessarily approximate investigation of the details shown to consider these outlines as circular, and having a radical axis which is better determined from negatives 1 and 2 than from 3. This axis is inclined to the horizon in negative No. 1, 33°.i, and in negative No. 2, 32°.2. Adopting the mean, and assuming the exposure of No. 3 to have taken place under similar conditions to the exposure of Nos. 1 and 2, it would seem that the radical axis of No. 3 is inclined to a vertical circle 57°.3, counting toward the north, at the time and place of its exposure. The position angle measured by a vertical circle passing through the centre of the sun at the middle of totality at Fort Worth would be 61 ".7; hence the position angle of the radical axis is \°.\. To locate as far as possible the striae which are plainly evident in the area of the photograph comprised in the lunes of the intersecting circles, I measured upon the original negative the rectangular co-ordinates of the terminal points on the circumference of each circle. These measures were made under a magnifying power of fifteen diameters, obtained with a Tolles 4-inch objective and a micrometer eye-piece by Powell and Leland fitted to my microscope micrometer. (For a description see Proc. Am. Acad. Arts and Sci., 1877, p. 352.) Dis- cussing the two images separately, each will afford us a determination of the angular intervals separating the terminal points of these striae. The position of each point may be expressed by an equation of the form — x* +f - r 2 - Ax - By + C= o, where x andjj/ are the rectangular co-ordinates of the centre of the first image, and r its radius. A, B, are twice the measured co-ordinates of the terminal points, and — C= % (Az + B*). In the same manner for the second image, we have — Tabulating the measured values of }4 A, }4 B, and }4 A', }4 B', we have the following table in which the terminal point given in the second image is not necessarily the same point mentioned in the first column, except in the case of the intersections of the two outlines : — FIRST IMAGE. SECOND IMAGE. Designation of the point measured. 'A A. ^B. Remarks. 'A A'. ^B'. Remarks. Intersection Terminal point !) » JJ J? J? 3? » J) 10.3 19.2 20.1 21. 1 2I.4 21.5 21-5 O.7 4.4 6.0 7.8 8.7 I0.8 12.2 Well defined. Faint. Faint. Beginning of the black Hazy. [edge. 10.2 21.4 22.5 23-3 23-4 23-7 23-9 O.O 4-i 54 7-5 8-3 11. 2 11.0 Well defined. Faint. Faint. Beginning of the black edge. 34 REPORTS.- REPORT OF MR. LEONARD WALDO. FIRST IMAGE. SECOND IMAGE. Designation of the point measured. #A. y 2 B. Remarks. #A\ ^B'. Remarks. Terminal point .... 21.3 I4.I Faint. 23-9 II.9 j> >i 20.8 15.8 Broad and hazy. 23-3 14.O >, }) " 20.0 I 7.1 23.0 15-4 » }> 19.2 18.3 Well defined. 21. 1 18.0 Intersection . . I3- 1 21.7 13.2 22.1 Terminal point 5-3 20.9 7-3 20.2 Hazy at edges. 99 99 3- 2 19.1 5-2 18.9 Faint. Point in the circumference . 2.6 l8.2 4.0 17.0 99 99 • 0.4 14-5 2.8 14.8 » 99 0.0 9.8 2.0 11.0 99 99 1.1 6.2 3-3 6.1 99 99 3-9 2.7 5-6 3-° These data give us nineteen equations of condition for each image, involving x, y and their second powers. Forming the equations, adding them together, and taking their means, we have for the two cases — X 1 + y* — r 2 — 25.90 x — 24.08 y + 425.91 = o. x* + y 2 - r* — 29.80 x' — 23.16 y + 470.56 = o. Subtracting each equation in turn from our mean equations, we shall then be rid of the second powers, and may proceed at once to form our normals with all necessary exactness, by rendering the coefficients of x, y, x\ y, successively positive, and adding the resulting series of equations. We shall thus have — 307.10 # + 18.56^-3509.07 = and — 8.30 x + 206.72 y — 2244.49 = o 322.00 x'- 35.96/ -3773-74 = 0. - 41.60 x" + 191.56/ - 1561.06 = o. Solving these equations for x, y, x' and y, and substituting in the mean equations to deter- mine r and r, we have — #=10.796 ^=12.943. y= 11.289 f - r °-954' r- 10.883 *" = 10.893. With these values of the co-ordinates of the centres of the two images, we may now easily refer each terminal point measured to the centre of its image. We have — X Y e X' Y' fl' X Y e X' Y' e' *- 0.5 — 11.0 267.4 *— 2.6 — 10.7 O 2 S6-3 + 10.5 + 2.8 14.9 + II.O + 0.9 4-7 + 8.4 - 6.9 320.6 + 8.5 - 6.9 320.9 + 10.0 + 4-5 24.2 + 10.4 + 3-° 16.1 + 9-3 - 5-3 33°-3 + 9.6 - 5-6 329-7 + 9.2 + 5-8 32.2 + IO.I + 4-4 2 3-5 + 10.3 ~ 3-5 341-2 + 10.4 ~ 3-5 341-4 + 8.4 + 7.0 39-8 + 8.2 + 7.0 40-5 + 10.6 - 2.6 346.2 + 10.5 - 2.7 345-6 *+ 2.3 + 10.6 77.8 *+ 0.2 + 10.9 88.9 + 10.7 - o-5 357-3 + 10.8 — 0.2 358-9 - 5-5 + 9.6 1 19.8 - 5-6 + 9.2 121.3 + ,„, + 0.9 4.8 + II.O — 0.0 360.0 - 7-6 + 7.8 133-3 - 7-7 + 7.9 134-3 * Radical axis. REPORTS. — REPORT OF MR. LEONARD WALDO. 35 The origin has been taken at right angles to the radical axis. The correction to the angle of the radical axis is — Sin Ap = ± J ( x-x'y+ {y-y y 4/- 2 which gives — A p = ± 5°72. By an examination of the photograph and its position with reference to the camera, it has been concluded that the point of the radical axis first mentioned is that point having a position angle of 1 84°4. Applying the correction A p and making this assumption, we have the follow- ing results : — FIRST IMAGE. SECOND IMAGE. Corrected Position angle. Remarks. Corrected Position angle. Remarks. o 184.4 243-3 253-0 263.9 268.9 280.0 287.5 297.6 306.9 314-9 322.5 6.2 42.5 56.0 Diameter parallel to radical axis. Too faint to measure. Diameter parallel to radical axis. 184.4 243-3 252.1 263.8 268.0 281.3 282.4 287.1 298.5 305-9 322.9 5-6 43-7 56-7 Diameter parallel to radical axis. Too faint to measure. Diameter parallel to radical axis. The two columns give the position angles of the prominences as recorded on the nega- tive No. 3 and reproduced on an enlarged scale in Plate I. These results are uncorrected for refraction, and their apparent discrepancy will be readily understood by those observers who have had occasion to measure photographic images of the sun. They are also liable to a small systematic correction, owing to the imperfect means of determining the zero of position. I have no means of knowing whether these striae represent prominences, strictly speaking, or whether they denote a greater intensity of coronal light only at the points whose position angles are given above. In Plate I. the details are fairly given, the extension of the corona particularly. The original negative has a lunar outline of about o ,v,, 43 by direct measurement, and from this an idea of the amplification in the Plate may be obtained. Negatives numbered 4 and 5 were submitted to Professor Pickering, who has kindly embodied the result of his examination of them in the following memorandum : — " Each plate contains two photographs of the corona, formed by placing a double image prism between the lenses of the camera. One of these photographs is well defined (in negative No. 4), while the other is elongated in the direction of the line connecting them. Probably D 36 REPORTS. — REPORT OF MR. LEONARD WALDO. one image of the prisms is nearly achromatic, the dispersion of the other being uncorrected. The elongation is so marked in negative No. 4 that none of the differences can with certainty be ascribed to polarization. " In negative No. 5, the images are nearer together, better defined, and the effect of color is much less. The points marked a, b, are equally dark in the upper photograph, while the corresponding points a', b\ of the lower photograph differ greatly in intensity, b' being much darker than a. This result is confirmed by c, d', which are much darker than /, while e is but little lighter than c or d. " These effects are explained by a radial polarization of the corona, if the plane of polarization of the achromatic image of the prism is perpendicular to the line connecting the two images. Otherwise, they suggest tangential polari- zation. " A line h g is also visible in the upper photograph which does not appear in the other. If due to polarization, this confirms rather than contradicts the conclusions noted above. It is not improbable, however, that this line may be due to a protuberance or bright spot in the corona, which is diffused in the ill-defined image. "In repeating this observation, a Rochon prism of quartz should be used in which both images are equally corrected for color. The angular separation need not exceed one degree, as it is only essential that the photographs should not overlap." After Professor Pickering had completed his examination (in entire ignorance of the direc- tion of the axes of the rhomb), I received from Dr. Lytton the prism with which the photo- graph had been taken. Professor Pickering and myself then confirmed Dr. Lytton's statement that the plane of polarization of the achromatic image is nearly parallel to the line joining the two images, — a conclusion which points to the tangential polarization of the light of the corona. Dr. C. S. Hastings has since courteously complied with my request, and made a critical examination of the photographic negative ; and, though he confirms the description of the appearance of the negative, he dissents from the conclusion that there is positive evidence offered regarding the polarization of the corona. It would be beyond our province to do any thing more in this report than simply to record the phenomena observed. It is not possible to reproduce the photograph in question, and I can only leave the discussion to the physicists, with the remark that the negative, as described by Professor Pickering and its details corrobo- rated by other competent witnesses, should be put in evidence as regards the polarization of the coronal light, unless the method can be shown to be open to serious objection. Professor Pickering, Dr. Lytton, Dr. Hastings, Mr. Wells, Mr. Tolles, and Mr. Edwards have all rendered important service in the preparation of the above remarks relat- ing to the photographs, which I desire to gratefully acknowledge. LEONARD WALDO. REPORTS. — REPORT OF MR. R. W. WILLSON. 37 REPORT OF MR. R. W. WILLSON. In accordance with the plan of the expedition, I proceeded to Fort Worth in advance of the other members of the party, arriving on July 9th. The first week after my arrival was spent in examining various points in the neighborhood of the town, in order to select a place for our observations. After the arrival of the " Brown Transit " and the chronometer, Bond, 1058, on the 15th, I began a series of observations for time, which was continued until the 31st. In the division of labor upon the day of the eclipse, the duties that fell to me were of a general nature. I was to observe the times of the several contacts, and to examine the corona with the polariscope. It seemed best, also, that some one should watch the progress of totality, and obtain a telescopic view of the corona, examine its general structure, and take note of any new or remarkable phenomena which might present themselves. This duty naturally devolved upon me, not as being best fitted by reason of previous close familiarity with the appearance of the sun under all conditions, but as the only observer who had not already assumed some special object for his attention during the eclipse. I was therefore provided with the three-inch telescope, by Secretan, with universal mounting, and a list of questions prepared by Mr. L. Trouvelot, of Cambridge, the answering of as many as possible of which was my principal aim. For observation of the first contact, I made use of a terrestrial eye-piece, magnifying 42 diameters ; and for the last contact, an inverting eye-piece, magnifying 98 diameters. Both powers as well as a power of 1 50 were used at different times for watching the progress of the partial phases. For second contact, and during totality, an inverting eye-piece magnifying 45 diameters was used, the diameter of the field of view being about sufficient to give a clear view of all parts of the corona lying within one diameter of the sun's centre. In this eye- piece was a glass plate ruled with concentric circles, about 6' apart, whose circumferences were divided into arcs of 45 each, by lines radiating from their common centre, giving a simple means of locating the outlines of the corona and the positions of noteworthy portions. I was also provided with an Arago polariscope with selenite plate, and with a small direct vision spec- troscope, by Browning, to be used if occasion should offer. First contact was noted at 3* I2 M 25', and was a very good observation, the sun's limb being very clearly defined and steady, and the irregularities of the moon's limb very marked. Indeed, throughout the whole time of the eclipse, till a few minutes before the last contact, the atmos- pheric disturbances were remarkably small, and the definition such as left nothing to be desired. As the moon advanced, a careful examination of the sun's surface showed no distortion of its details which could be attributed to a lunar atmosphere ; the faculae retaining their shape quite unaltered as they were covered by the advancing limb. The cusps were throughout 38 REPORTS. — REPORT OF MR. R. W. WILLSON. sharply defined, no brushes of light or other similar phenomena were noticed ; but, at times, the points appeared to be slightly curved outwards, an illusion which was less marked with a higher power. At 4* i6 OT the extremity of the lower cusp was separated from the remainder of the visible portion of the sun. As totality was now fast approaching, I replaced my dark shade-glass by a lighter reddish one, in preference to a very light green shade, for which the light was still too strong. To avoid the necessity of another change, I determined to observe the second contact with this, perhaps, objectionable color ; and therefore caught up the beat of my chronom- eter, and waited for totality, watching with my left eye, my right having been for some time protected by a bandage. The progress of the moon now seemed very rapid ; small portions of the cusps, now reduced to thin lines, being broken off by the projections of the moon's limb, and formed into little blotches of light evidently much increased in width by irradiation, which gradually lost their brightness without much change of size or shape, fading till they disappeared. This continued till only a thin strip of light remained, and this at last broke up into several fragments, which in the same manner faded gradually to extinction some four seconds later. The time of losing the last ray of light was, however, very well marked, and was noted as 4* i8 w 2