l6o CARNEGIE INSTITUTION OF WASHINGTON. 



of the type shown by these experiments to promise success. With 

 such an instrument, immense new fields of investigation of the 

 highest importance in their bearing on the problem of stellar evolu- 

 tion could be immediately occupied. Here again, however, the un- 

 favorable atmospheric conditions at almost all existing observatories 

 would render the construction of a large telescope almost useless. 

 To be successful, such an instrument must be erected at a site where 

 the night-seeing is nearly perfect, the sky clear and transparent, and 

 the average wind velocity very low. Under such conditions, a 

 properly constructed telescope of large aperture would undoubtedly 

 yield results greatly siirpassing those hitherto obtained. 



These considerations are sufficient to define the general character 

 of a site suitable for a well-equipped solar observatory. There 

 are other points, however, which must be taken into account. A 

 solar observatory provided with an outfit of instruments, and then 

 left to do its work without the possibility of improvement or change, 

 could never attain the best results. On the contrary, it must have 

 the means of producing new types of instruments and modifying 

 old ones, as the development of the work may suggest. In other 

 words, a shop completely equipped with all appliances necessary 

 for the most refined construction of both the mechanical and optical 

 parts of instruments, should form an integral part of a solar 

 observatory. A shop of this kind can not be conducted without 

 great difficulty and expense if far removed from large cities and 

 other sources of supply. This is only one of many reasons which 

 would render it desirable to select an observatory site within easy 

 reach of the facilities afforded by a large city. 



In his recommendation for the establishment of an observatory 

 for the purpose of determining whether the lieat radiation of the sun 

 undergoes change. Dr. Langley pointed out the desirability of mak- 

 ing the observations at a height of some 20,000 feet above sea-level. 

 Apart from the excessive difficulty and expense of conducting an 

 observatory at such an elevation, which are best appreciated by those 

 who have worked at great altitudes, the inaccessibility of high moun- 

 tain peaks would stand in the way of sudh an undertaking. But it 

 nevertheless might have been carried out, at a somewhat lower alti- 

 tude, if the recent development of Dr. Langley 's work at Washington 

 had not indicated that the great mass of observations could undoubt- 

 edly be made to good advantage at a much lower station. The 

 increasing perfection of the observational method has, indeed, per- 

 mitted fairly good results to be obtained under the very unfavorable 



