Makch 23, 1906.] 



SCIENCE. 



455 



' t = true coordinates of zenith and star pro 



Ay 



jected on plate; 



}= projections on plate of total displace- 

 ment of star due to refraction; 



/tt = refraction constant. 



The tables are so constructed that the 

 terms of Ax and Ay, which represent the 

 displacement common to the star and the 

 center of the plate, have been omitted. 

 Those terms of Ax which are proportional 

 to X and the terms of Aj/ which are pro- 

 portional to y have also been omitted, these 

 corrections being atttomatically applied in 

 the determination of the plate constants. 



A Contribution on Astronomical Refrac- 

 tion: EussELL Tracy Crawford. 

 An investigation of the constant of re- 

 fraction was made in the year 1899 which 

 resulted in the discovery that this so-called 

 constant is a function of the zenith dis- 

 tance. The observations were made with 

 the Repsold meridian circle of the Lick 

 Observatory and reduced by a method 

 which is the converse of Talcott's method 

 of determining latitude. 



This method has both its advantages and 

 disadvantages. Among the former, the 

 most important are: first, total elimination 

 of the latitude, and hence also of its varia- 

 tion ; second, elimination of the nadir read- 

 ing; third, there is no wait of twelve hours 

 or six months in order to observe a star at 

 both culminations; fourth, only one half 

 of any error of observation or in the 

 declinations used enters into the reduc- 

 tions; and, finally, the simplicity of the 

 observations. 



The greatest disadvantage lies in the fact 

 that the declinations have to be considered 

 known. But by taking fundamental stars, 

 such as those of Professor Newcomb's new 

 'Fundamental Catalogue,' and by taking a 

 large number of these stars, this difficulty 

 will be nearly completely eliminated. 



Two independent series of observations 

 were made for this investigation. One was 

 made during the summer months, giving 

 A log M = 0.000101 {oO^.e — »). 



The other made during the winter months 

 gives 



A log /ii = 0.000117 (59°.3— s). 



Combining these into one solution, the two 

 constants of the expression become 



+ 0.000108 ± 0.000010 and 58° zt 8°, 



so that the Pulkowa tables should be cor- 

 rected by 



A log ,11 = 0.000108 (5S°— s) 



where s is the zenith distance in degrees. 

 The efficiency of tables constructed by 

 applying this correction to the Pulkowa 

 tables may be shown by comparing a series 

 of observations reduced by the two. In 

 his preliminary reductions of his observa- 

 tions of the Piazzi southern stars. Tucker 

 has used the Pulkowa refractions. In his 

 final reductions, he has corrected his 

 declinations for errors in refraction which 

 are given as a function of the zenith dis- 

 tance. Auwers has published further cor- 

 rections to reduce Tucker's results to his 

 fundamental system {Abhdlg. z. d. A. N., 

 No. 7). The differences between the cor- 

 rections Auwers + Tucker and those re- 

 sulting from this investigation are less than 

 the probable errors of observations found 

 by Tucker. 



A Solar Planisphere : Sarah F. Whiting. 

 In connection with the students' labora- 

 tory work at Whitin Observatory, Welles- 

 ley College, a device for working problems 

 in relation to the hours of day, night and 

 twilight, at different latitudes at different 

 times of year, has lately been worked out 

 chiefly by E. Rebecca Ellis, assistant. We 

 do not find this simple contrivance, which 

 we call a solar planisphere, elsewhere de- 

 scribed. 



