298 



SCIENCE. 



[N. S. Vol, XV. No. 373. 



land's map were for a short time very in- 

 tense, while others, such as the aluminium 

 line of intensity 20 at X 3,961.674, disap- 

 peared entirely. Two sharp bright lines 

 appeared at i 3,884.67 and /I 3,896.21 

 These were strongest in the spot, and did 

 not extend to the limits of the disturbed 

 area. Full details will be published in the 

 Astrophysical Journal. 



The Bruce Spectrograph of the Yerkes Ob- 

 servatory: Edwin B. Frost. 

 The equipment of the Yei'lces Oljserva- 

 tory has been recently enlarged by the 

 'Completion of a spectrograph designed for 

 ■the special purpose of the determination of 

 the motion of the stars in the line of sight. 

 The addition of this instrument to the 

 accessories of the forty-inch telescope was 

 made possible by the liberal gift of $2,300 

 from Miss Catharine W. Bruce and $500 

 from the Rumford Fund of the American 

 Academy of Arts and Sciences. The spec- 

 trograph is very rigidly constructed, chiefly 

 of iron and steel, and the prisms are main- 

 tained in a fixed, invariable position. The 

 whole instrument is inclosed in a large 

 aluminium ease with double walls, for pro- 

 tection against changes of temperature. 

 Coils of wire inside this case can be heated 

 by the 110-volt current of the observatory 

 mains, and it has been found not difficiTlt 

 to keep the temperature of the air in the 

 prism-box within 0°.l C. during exposures 

 of an hour or more. A correcting lens 

 placed one meter in front of the slit makes 

 the visual forty-inch object-glass efficient 

 for the violet light (■'■ 4,500) which passes 

 through the prism-train at minimum devia- 

 tion. The collimator is of 2 in. aperture 

 and 38 in. focus ; and two cameras are pro- 

 vided, one of 3 in. aperture and 24 in. focus 

 and another, a Zeiss anastigmat, of about 

 2.8 in. aperture and 18 in. focus. The first 

 three lenses are of triple construction, de- 

 :signed by Professor C. S. Hastings and 



made by Brashear. It has proved to be a 

 matter of great difficulty to obtain prisms 

 of the large size necessary to transmit a 

 two-inch beam which are sufficiently homo- 

 geneous. (The face of the largest prism 

 is 133 mm. long and 57 mm. high.) Aftei' 

 an unsuccessful experience with a set of 

 prisms of what appeared to be excellent 

 glass from Mantois, a quantity of glass was 

 ordered from Scliott & Co., of Jena, which 

 should be finely annealed and of the quality 

 of telescope objectives, a requirement 

 which, strangely enough, does not appear 

 to be customary in respect to glass for 

 prisms. This new set of prisms shows con- 

 siderable improvement over the first ones, 

 but the definition is still not the same over 

 the whole surface of the faces. This is 

 now assumed to be due to the moulding 

 of the prisms in trangular shape at Jena 

 instead of melting disks, as we had desired, 

 from which the prisms would be cut by 

 Brashear. Although the speetograph 

 dees not fully realize the resolution which 

 the length of the faces of the prisms would 

 imply, the instrument has nevertheless been 

 shown to be capable of furnishing results 

 of a very high degree of accuracy, as illus- 

 trated in the paper by an example of a 

 plate of a Arietis. The comparison spectra 

 so far employed have been the spark of 

 titanium and of iron and the helium tube. 

 During the exposure on a star the observer 

 guides the telescope by light reflected from 

 the speculum slit jaws, which are not in the 

 same plane, but symmetrically inclined 

 away from the line of collimation, each 

 making with it an angle of 92" 55'. It is 

 also possible, by merely turning a mirror, 

 to observe in the same guiding telescope 

 the light that has gone through the slit and 

 has been reflected at the first surface of the 

 first prism. The method of measuring and 

 reducing the plates briefiy described by the 

 writer at the meeting of the Society in 1899 

 is still in regular use. Plates are measured 



