REPORT OF THE ACTING SECRETARY. 75 



eient. For a good bolometer only 1 or 2 per cent of the resistance needs to be 

 thus substituted. The improvement obtained on JMount Wilson is so great 

 that the " drift " in a single holograph now rarely reaches a centimeter, 

 although a rise of temperature of nearly 2 degrees centigrade per hour may 

 be taking place. The bolometric apparatus may be conveniently balanced- by 

 the aid of a variable shunt of high resistance round one of the balancing 

 coils. This device is eni])loyed now instead of a slide wire for determining 

 the sensitiveness of the platinum thermometer used with the standard pyrheli- 

 ometer. 



Some trial has been made of a new device of entirely different principle, as 

 a rival of the bolometer and its class of instruments. This new instrument is 

 not yet fully perfected, but gives some promise of exceeding in sensitiveness 

 either the bolometer, the thermopile, the radiometer, or the radio-micrometer. 



OBSERVATIONS AT WASlIliXGTON. 



Notwithstanding the fact that it has been deemed expedient for the satis- 

 fiictory proof of the variability or constancy of solar radiation to conduct many 

 of the observations at the favorable station on Mount Wilson, still the observa- 

 tions at Washington remain of hardly less importance than before. For the 

 determination of the amount of solar radiation outside the earth's atmosphere 

 depends so largely on our estimate of the loss by scattering and absorption 

 of the rays in the air that very much of the confidence with which we may 

 hope the work will ultimately be regarded must be produced by the number 

 and weight of the circumstances which tend to establish the accuracy of the 

 estimate of the effect of the earth's atmosphere. Chief in importance among 

 such checks is the simultaneous observation of the sun's radiation at stations 

 remote .'rom one another and different in altitude, like Washington and Mount 

 Wilson. When it is said that the solar radiation received at the earth's sur- 

 face on Mount Wilson generally exceeds that in Washington by a third, but 

 that, nevertheless, simultaneous spectrobolometric observations at high and low 

 sun at the two places yield substantially identical values of the solar radiation 

 outside the atmosphere, the soundness of the work seems strongly verified. 

 In order to secure as many checks of this kind as possible, solar constant deter- 

 minations have been made by Mr. Fowle in Washington at all times when the 

 sky conditions warranted. Apparently a very considerable rise of the solar 

 constant values, occurring in .January after Mr. Abbot's return from Mount 

 Wilson, was observed at Washington alone. As in former years, the number of 

 days suitable for these observations in Washington has been small. 



Another impox'tant independent method of securing evidence of the varia- 

 bility of the sun is found in the bolometric examination of the solar image to 

 measure the absorption taking place in the sun's immediate surroundings. 

 This observation depends but little on the clearness of the earth's atmos- 

 phere, and can therefore be carried on about as well in Washington as any- 

 where. As stated in former reports, the Observatory is iirovided for this work 

 with a horizontal reflecting telescope of 20 inches aperture and 140 feet focus, 

 easily one o" the largest telescopes of the world. Observations with the spec- 

 trobolometer at the focus of this great telescope have lieen carried on regularly 

 by Mr. Fowle throughout the year. Improved means for stirring the column 

 of air within the telescope tube to improve definition have been provided, 

 but, requiring additional electrical facilities, their trial was postponed to 

 await the completion of the elocti-ic station above referred to. 



