JAKUABY 2, 1914] 



SCIENCE 



tions in the last two years, respectively. 

 Variables of short period complete their 

 changes in a few days, or hours. Professor 

 Bailey has found five hundred such ob- 

 jects in the globular clusters. In one of 

 these clusters, Messier 3, out of a thousand 

 stars one seventh are variable, all have a 

 period of about half a day, and their 

 periods are known within a fraction of a 

 second. Their light changes so rapidly 

 that in one case it doubles in seven min- 

 utes. It is a strange thought that out of a 

 thousand stars, looking exactly alike, there 

 should be a hundred little chronometers 

 keeping perfect time, and whose rate is 

 known with such accuracy. About a hun- 

 dred and fifty variables belong to the Algol 

 class, in which the light is uniform for a 

 large part of the time, undergoing a sud- 

 den diminution at regular intervals. This 

 is due to the eclipse of two bodies, one 

 darker than the other, revolving around 

 their common center of gravity. An elab- 

 orate theoretical study of this problem has 

 been made at the Princeton Observatory 

 and, largely from the photometric and 

 photographic magnitudes made at Harvard 

 and elsewhere, the dimensions of a large 

 number of these systems have been deter- 

 mined. 



Photography still can scarcely compete 

 with other methods where the greatest ac- 

 curacy is desired, as for instance, the meas- 

 ures with the polarizing photometer by the 

 late Oliver C. Wendell. The masterly use 

 of the selenium photometer by Professor 

 Stebbins gives results for bright stars of 

 still greater accuracy, while the experi- 

 ments in Germany with the photo-electric 

 cell by Rosenberg and Guthnick give re- 

 sults which promise to revolutionize our 

 present methods. The principal source of 

 error appears to be the varying trans- 

 parency of the air. The trial of the instru- 



ment in a location where the air is exceed- 

 ingly clear and steady for long periods is 

 greatly to be desired. 



During the last twenty-five years photo- 

 graphs have been obtained by the Harvard 

 Observatory in order to furnish a history 

 of the stellar universe. Two similar 

 8-inch photographic doublets have been 

 used, one mounted at Cambridge for the 

 northern, and the other at Arequipa for 

 the southern stars. "With each of these 

 instruments about forty thousand photo- 

 graphs have been taken. The total weight 

 of these plates is about forty tons. As 

 each plate covers a region ten degrees; 

 square, every part of the sky has been 

 photographed, on the average, a hundred 

 times. This work is now supplemented by- 

 two small Cooke anastigmat lenses, each 

 having a field thirty degrees square. The 

 number of plates taken with these two in- 

 struments are nine thousand and fourteen 

 thousand, respectively. The exposures 

 with the larger instruments are, in general, 

 ten minutes, showing stars of the thirteenth 

 magnitude. The exposures with the 

 smaller instruments are one hour, showing 

 stars of the eleventh magnitude. A con- 

 tinuous history of the sky is thus furnished 

 from which the magnitude and position of 

 any stellar object of sufficient brightness 

 can be determined for a large number of 

 nights during the last quarter of a century. 

 A striking illustration of the value of this 

 collection occurred when the planet Bros 

 was discovered in 1898. It appeared that 

 this object was nearer the earth in 1894 

 than would occur again for thirty-five 

 years. An examination of the photographs 

 showed its presence on 23 plates, and from 

 their positions the parallax of the sun and 

 mass of the earth were determined with an 

 accuracy equal to that of any of the 

 methods previously used, and on which an 



