OP ARTS AND SCIENCES. 161 



of the Chief Engineer of General Sheridan's staff) which I used for 

 this purpose at Bismarck, D. T., Oct. 9, 1873. It was my first fair 

 trial of the method. The stars were selected from my catalogue of 

 981 stars, published in that year by the War Department, and are all 

 thoroughly well determined, — an advantage which Bessel's method 

 has had over Talcott's. 



This scheme requires four hours observing. If I recollect rightly, it 

 was somewhat interfered with by clouds, and would otherwise have 

 been sooner finished. I was anxious to succeed that evening, as time 

 pressed ; and I therefore did not attempt to accumulate as many ob- 

 servations as possible, but preferred to make sure of a few. The 

 result was sufficiently accurate for geographical purposes, having a 

 probable error of ± 0"31. The instrument was quite indifferent in 

 its optical portion. The Land Office at that time required a proba- 

 ble error of less than 3" ; and does still, so far as I know. There 

 are Land Office determinations extant which are far more than this 

 in error. 



In more precise latitude and longitude work, the instruments used 

 have generally 3 inches aperture, and 30 to 36 inches focal length. 

 Such an instrument should be very solidly built and set up. The one 

 with which I am most familiar is Brigham Young's, in the Temple yard 

 at Salt Lake City : it is by Wiirdemann, and was originally placed 

 there at the time of the Coast Survey determination of longitude at 

 that place. I found it very firm and strong : its level, collimation, and 

 azimuth errors were constant, though not very small, as the Mormon 

 astronomers seem not quite expert in adjusting. But I did not for 

 this reason neglect to make the full number of observations, nor to dis- 

 tribute them precisely as if I were observing with a smaller and worse 

 instrument, partly from habit, and partly because the instrument at 

 Evanston, W. T., with which I was comparing time, was the instru- 

 ment by J. H. Temple, mentioned above ; and the observers were ex- 

 changed in the middle of the series. 



There are two essentially different patterns of large portable tran- 

 sits. The one, the German or Russian, has a prism between the object- 

 glass and the eye-piece ; and the eye-piece itself is at one end of the 

 horizontal axis. I have used such instruments only for trials of per- 

 sonal equation. There is one such at the Harvard College Observatory, 

 and others were made for the American Transit of Venus expedi- 

 tions. These transits are very convenient. The level is always upon 

 the axis. The observer sits in one position between reversals, and has 

 not the troublesome necessity of bending his body into inconvenient 



VOL. XI. (n. S. III.) 11 



