210 CARNEGIE INSTITUTION OF WASHINGTON. 



curves were obtained showing the transmission of windows made of different 

 materials. The microscope cover-glass used for stellar observations up to 

 March 17 of the present year was found to transmit most of the radiation to 

 5.5 fi with some transmission to 7 fi. The transmission of crystals of various 

 salts was also studied, with a view to finding a substitute for the water-cell. A 

 plate of potassium alum 1.7 mm. thick, which may be obtained in large 

 crystals and shows a satisfactory degree of permanency when polished, is 

 found to transmit throughout the visual region and as far as 1.5 /x without 

 absorption bands. At this point the transmission falls off sharply. This 

 material, therefore, should make a desirable substitute for the water-cell. 



The results of other investigations show: first, that there seem to be no 

 grounds for suspecting a variation in the transmission of the water-cell due to 

 temperature or prolonged use ; second, that the reflecting power of the silver 

 films in the telescope appears to be constant for angles ranging from 45° to 

 perpendicular incidence; third, that the variation in radiation of the 100-watt 

 nitrogen-filled lamp used at the telescope as a check on the sensitiveness of the 

 apparatus is directly proportional to the power used between the limits of 40 

 and 110 watts; fourth, that the source of drift appears to be inherent in the 

 thermo-couples, some being entirely free from it and others showing it to a 

 marked extent. 



The D'Arsonval galvanometer as at present used for stellar observations is 

 placed at a distance of 7 meters from the registering apparatus, this limit 

 being set by the size of the room available. By the aid of multiple reflections 

 between the galvanometer mirror and a fixed mirror the deflection may be 

 increased in arithmetical progression with the number of reflections. As many 

 as 5 reflections were used on one occasion, which gave as the sensitivity of the 

 galvanometer 5X10""" ampere per millimeter. Observations of the deflec- 

 tions given by the thermo-couple with changes in the amount of radiation 

 received show the scale to be uniform over a wide range. 



THE REGISTERING MICROPHOTOMETER. 



Several improvements in this instrument have been made by Pettit. 

 Among these are a motion which quickly returns the plate-carriage to the 

 starting-point, and a Julius suspension for the galvanometer which eliminates 

 small irregularities in the photographic curves due to earth tremors. A study 

 of the resolving power of the instrument shows that after the width of the 

 first slit is reduced to 0.01 mm. no gain can be made by further narrowing, 

 on account of the effect of diffraction. An increase may be obtained by 

 reducing the width of the slit in front of the thermopile, but with the result 

 that some scattered light from the illuminated film falls upon the junctions. 

 Investigations have been made of the variation of the deflections with slit- 

 width and slit-length and with changes in the voltage of the battery which 

 supplies current to the lamp. 



The microphotometer has been used by St. John and Miss Ware in studies 

 of line-structure and wave-lengths in the solar spectrum, by Hubble in 

 investigations on nebulae, and by Pettit in mapping special regions of the 

 solar spectrum, with a view to the determination of the intensities of the 

 spectral lines. 



