OF ARTS AND SCIENCES : SEPTEMBER 20, 1864. 357 



of the American Nautical Almanac, and a statement that the elements 

 there used are erroneous. 



It is not my intention to enter upon any extended examination of 

 this paper, on which Mr. Saffoi'd seems to have bestowed much labor, 

 nor, at present, to express any opinion upon the justness of his criticism 

 or the correctness of his conclusions ; but simply to call attention to the 

 magnitude of the error which he claims to have detected, and to show 

 to what extent the usefulness of the Nautical Almanac has been im- 

 paired by the alleged error, if Mr. SafFord is right. 



I am led to do this by the apprehension that the impression may 

 have been given that some serious injury has been done to the public 

 service by the use of an erroneous value of the place of this star on the 

 part of the Coast Survey, the National Observatory, and the Nautical 

 Almanac, — or that there is some culpable and important uncertainty 

 about a fundamental element in astronomy. This apprehension is 

 justified by the fact that it has since been published in pamphlet form, 

 with a preface by Professor Bond, in which Professor Bond states that 

 this matter is of moment to American astronomers, because the alleged 

 erroneous value of the mean R. A. of the Pole Star has been adopted 

 by the public institutions above mentioned. 



Mr. Safford treats of the use of the Almanac place as leading to errors 

 o^ prediction, and presents the nearer fulfilment of his predictions as con- 

 firmation of the superiority of his determination of the element in ques- 

 tion. The phenomenon predicted is the appearance of the Pole Star 

 on the meridian. Now to form a distinct idea of the consequence of 

 the use of false data we must examine its effect on this appearance. 



The motion of the Pole Star is so slow, in consequence of its prox- 

 imity to the pole, that an error of time in a prediction which would be 

 quite considerable in some questions in astronomy would here cor- 

 respond to a very small change of place of the star, or a very slight 

 difference in the appearance to the observer. Thus a second of time 

 would only correspond to a change of position amounting to about one 

 third of a second of space. We will now consider how small a quantity 

 a second of space is. The following illustration will place it before the 

 mind quite distinctly. If a human hair supposed to measure g-^jy of an 

 inch in diameter were placed at the distance of 86 feet from the eye, 

 its thickness would subtend an angle at the eye of 1", and would shut 

 out just that amount of celestial space. 



If now we compare the prediction of a transit of Polaris made from 



VOL. VI. 38 



