254 CARNEGIE INSTITUTION OF WASHINGTON. 



lines, which fall upon the solar lines and interfere with their measure- 

 ment. An attempt will therefore be made to imitate closely the results 

 obtained with the compound quarter-wave plate by constructing a 

 compound iodine- vapor cell, with absorbing chambers 1 or 2 mm. wide, 

 separated by vacuum cells of the same wddth. If this simple device 

 is successful, the iodine Unes superposed upon the solar spectrum will 

 appear on the odd strips, while the solar spectrum will appear alone 

 on the even strips. Wliatever be the outcome of this experiment, 

 it seems certain that some such differential method, requiring only a 

 single illumination of the grating, will be necessary to meet the present 

 rigorous demands for increased accuracy of measurement in solar and 

 laboratory spectroscopy. For the elimination of personal equation, 

 a problem of great importance, the Koch microphotometer seems to be 

 entirely satisfactory, and it is to be hoped that it will come into more 

 general use. 



But spectroscopic measurements are not the only ones in which 

 increased accuracy is needed. In work with the spectroheliograph it 

 is often desirable to determine small displacements of flocculi not 

 measurable by existing methods, because of local distortions due to 

 drift of the image during the exposure, temporary poor seeing, changing 

 refraction during long exposures with low sun, and other causes. 

 The 13-foot spectroheliograph recently built for the 60-foot tower 

 telescope is accordingly provided with two camera sUts, one of which 

 may be set on Ha. (or any desired line), while the other, at a distance of 

 several inches, is set on the continuous spectrum. Thus two images 

 of the same region of the sun may be photographed simultaneously 

 on a single plate, one showing the floccuU surrounding a group of sun- 

 spots, the other the spots themselves. Fine wires stretched across the 

 coUimator slit together with the images of the two camera sUts at the 

 end of the run furnish the necessary fiducial lines, with reference to 

 which the positions of spots and flocculi can be measured. Thus the 

 motions of flocculi can be determined differentially with reference to 

 sun-spots on solar images affected in precisely the same degree by any 

 source of distortion. 



INSTRUMENTS. 



The completion of the remodeled 60-foot tower telescope and its 

 auxiUaries, which include a spectroheliograph of 13 feet focal length, 

 has greatly strengthened the equipment available for solar research. 

 The improved 30-foot spectrograph, which can now be easily trans- 

 formed into an 18-foot spectrograph for use with a quartz invar inter- 

 ferometer, has been provided with vacuum and mercury arcs, a device 

 for exposing simultaneously on arc and solar image, and other new 

 accessories. The 13-foot spectroheUograph has proved to be very 

 satisfactory. The frame of angle iron which supports the optical 

 parts is suspended in the underground chamber from a heavy carriage 



