to an Astronomical Clock and Telegraph Register. 211 



Locke and myself, on the evening of (he 17th of November, 

 while watching its performance for a term of two hours or more' 

 Having reflected much since, and consulted with my friends both 

 m and out of the coast survey, I am more and more convinced 

 that it is difficult to form an over estimate of its importance in 

 every department of practical astronomy that involves the nice de- 

 termination of absolute dates, or of their relative intervals in time. 

 Throughout the whole range of practical astronomy, epochs of 

 time and their relative intervals have been hitherto measured by 

 astronomers by listening to the beats of a clock or chronometer, 

 and estimating the fraction of a second between them when any 

 event has occurred, such as a phase of an eclipse, an occupation 

 w a star, or a bisection of a star, comet, or planet's centre or 

 limb, by the wires of a transit instrument. This use of the ear 

 has been a matter of necessity, not of choice. Tt is, in every re- 

 spect, in the subdivisions of time and space, a very imperfect or- 

 gan. While the eye readily estimates the proportional parts of a 

 line with the precision of a tenth, the ear seldom distinguishes 

 smaller portions of an interval of time than a fifth. 



In the case of observing transits of a star with a seconds clock, 

 when greater precision than the estimate of a fifth of a second is 

 required, it is only obtained by long practice and experience, and 

 by the use of an ocular estimate of portions of space where the 

 star was at the time of bisection, and where it was seen to be at 

 fhe time when the two beats immediately preceding and follow- 

 ing this bisection were heard. 



In order to increase the precision of the comparison of two 

 sounds together, in point of time, it is necessary gradually to 

 diminish their interval asunder until the minimum audible of 

 about a hundredth of a second is attained. This advantage at- 

 tends the comparison of chronometers by coincidence of beats, 

 when one has a more rapid movement than the other: but it 

 cannot be brought to bear upon the association of the time of a 

 visible bisection of a star with the date of the hearing of the beats 

 ot a clock. The average uncertainty of such an association, as 

 appears from my last report of the result of four thousand exper- 

 iments on this subject, among the officers of the coast survey 

 ls ^enteen hundredths of a second. Of this sum, the greater 

 P°rtion is probably due to the imperfection of the ear as an organ, 

 ln associating its sensations with those of sight. I do not speak 

 DOW from actual experiment, as I hope to do another year. It i 

 We 'l known, however, that the nerves and sensations of sight 

 a «d touch are far more intimately connected together than those 

 of the eye and ear. 



I have repeatedly noticed the advantage of the substitution of 

 the sense of touch for that of hearing in the use of chronographs 

 for making a visible register of the dates of bisection of transit 



; 



