318 



THE SAYRE OBSERVATORY. 



The probable error of a single determination computed from tbe daily mean of all, 

 and therefore including that part due to the declinations employed, is found to be as 



follows : 



// 



Evening observations 0.223 

 Morning observations 0.240 



The Constant of Aberration. 



As each pair of stars, with two exceptions, was observed both evening and morning, 

 when the effect of aberration was not far from its maximum value, this series is well 

 adapted to an investigation of that constant, if by any means the effect of the latitude 

 variation can be eliminated. In fact this formed a part of the original plan in arranging 

 the scheme of observation. 



Unfortunately, however, in a series like this, the effect of aberration cannot be sepa- 

 rated from the latitude variation without some assumption with regard to one or the 

 other. In the present case it has been assumed that the latter can be represented by a 

 single periodic function, the period being fourteen months. Each observation will then 

 give an equation of the form 



+ A + Ax H By -|- Op |- T/i. = 



Where 



(?)„ is an assumed latitude, 

 A a constant correction, 

 Ax + By the periodic terms, 



Cp the correction on account of error in Struve's aberration constant, 

 Tp uniformly progressive change in the latitude. 

 For fill! details as to the process of combining and solving these equations, reference 

 may be had to the series of 1892-93. 



The resulting correction to Struve's value of the constant of aberration, which is 

 the only term we are now concerned with, is 



Struve's value 

 Final result 



| ".0027 ± 0140 



20". 4451 



20". 448 +014 



It is no doubt an open question whether this value is entitled to very great confi- 

 dence as a determination of the aberration constant. If the series had been continued 

 for three months longer, so as to embrace an entire period of fourteen months, or if it 

 had happened to be arranged symmetrically with respect to a maximum and minimum 

 value of the latitude, its weight would no doubt have been very materially increased. 



