ASTROPHYSICS IN THE UNITED STATES BOSLEE. 361 



of the collimator, a condition evidently indispensable for the use of 

 all the available light ; the tube is sufficiently rigid to allow the 

 realization of this. It is well known that one of the principal diffi- 

 culties in the accurate measurement of radial velocities results from 

 the changes of temperatures which change the indices of refraction 

 of the prism during the necessarily long exposures and thus produce 

 a false displacement of the spectrum lines. In order to escape this 

 danger the whole dispersive system is surrounded by a thermostat 

 which maintains it at a constant temperature. It consists of a 

 wooden box, lined with thick felt. By means of an electric fan and 

 a fine German-silver wire passing along the sides of the box and 

 traversed by an electric current of moderate intensity the box may 

 be rapidly warmed when the temperature becomes too low; a ther- 

 mometer, by means of an auxiliary current, stops the warming cur- 

 rent when the proj^er temperature has been regained. 



Naturally the great equatorial of this observatorj^ is particularly 

 suited to measures upon double stars ; we ourselves were able to sepa- 

 rate the two components of the close double 8 Equulei (0.3" apart), 

 and E. G. Aitken, the astronomer in charge of these researches, on 

 nights of good seeing does even better and can separate those as close 

 as 0.14". He has already published 2,000 new doubles hitherto un- 

 catalogued. 



Notwithstanding the superb qualities of this instrument, its angu- 

 lar aperture (about 1 to 19) is too small for use in photographing 

 ver}^ faint objects, such as nebulae and comets, and for this purpose 

 they therefore employ the Crossley reflector, whose ap(;rture is 36^ 

 inches. This instrument has had a somewhat peculiar histoiy and 

 its mounting has been wholly rebuilt during the last few years in the 

 shops of the observatory. As they had especially in view its use in 

 photogi'aphy, rigidity was considered a most essential qualitj^ In 

 order to follow a star across the meridian without having the tele- 

 scope strike the j)ier and thus avoiding the necessity of then revers- 

 ing the instrument, which would have been very undesirable in the 

 middle of an exposure, they adopted the English form of mounting 

 where the two extremities of the polar axis rest upon separate piers. 

 In order to diminish the moment of inertia of the moving parts, 

 which weigh 6 tons, they gave the polar axis the bizarre eccentric 

 form shown in plate 2; the eccentric portion serves partially as a 

 counteipoise and the radius of gyration is decidedl}^ reduced. The 

 driving sector has a radius of 2.50 meters so that any irregularities 

 due to its movements are much reduced. In order to keep the image 

 of the star being pliotogi-aphed constantly at the same point of the 

 plate it has usually been the custom to move the whole telescope by 

 the means of slow-motion mechanisms; with the Crossley reflector 

 only the plate holder is moved. Two micrometer screws give it the 



