108 



KNOWLEDGE. 



[May 1, 1900. 



1 -04-111011 object o-lass, whose focal length is 5'12 inches. 

 Yellow screen of moderate intensity, on account o£ the 

 deep blue sky. Stop=-T- Exposure = 1 full second. 



Plate, Fig. 2.— Nimbus covering the setting snn, whose 

 rays gild the upper edge* of the cloud bank, duriug rainy 

 weather. Showers are actually seen falling from the 

 cloud on the horizon. Same object glass. No yellow 

 screen, owing to the absolute hu-k of blue. Stop = J-. 

 Exposure =^ second. 



Figure 1 (text). — Peaceful cumuli during warm weather. 

 Same glass. Faint yellow screen. Stop=-'_. Exposure 

 =:| second, in strong sunshine. 



Figure 2. — Sunset behind ragged clouds. Same glass. 

 No screen at all. Stop=,/l. Exposure= J^ second. 



Figure 3. — Rainbow, with a nearly horiz.ontal sun. 



strong, while haste is particularly nugatory. A slow, 

 methodical development always yields good results. At 

 Juvisy, pyrogallic acid is used in preference to other sub- 

 stances. An enfeebled old solution, intensified when 

 necessary, is, moreover, preferable to a new one. Finally, 

 the question of the moment of stopping the development 

 is one which experience only can decide. 



3Ltttcrs. 



♦ 



[The Editors do not bold themselves responsible for the opinions 

 or statements of correspondents.] 



IS THE STELLAR UNIVERSE FINITE? 



TO THE EDITORS OF KNOWLEDGE. 



Sirs, — I cannot enter into this discussion, but may I 

 suggest, as a point which appears to me to have been 



Fig. 2.— Sunset effect, 18<J9, Septeml)cr, 2id. 5h. 2om., loral time. 



The secondary bow is very marked on the negative ; but 

 there are no supernumeraries. Same glass. No yellow 

 screen. Stop = whole aperture. Exposure=oV second. 

 Short exposures are jMrticularlij effective on rainhows. 



The excellent isochromatic plates of Messrs. A. Lumiere 

 et ses Fils, of Lyons, have been invariably used here by 

 M. Mathieu and the writer in their attempts at cloud 

 photography. 



The development of cloud plates is done in the ordinary 

 fashion, though plates sensible to the red should be 

 developed with a very weak red light only, in order to 

 avoid togging. The solution should, of course, not be too 



* These golden fringes, due to the illumination of the cloud from 

 behind, are also visible roiuid cumuli at night during lightning 

 flashes. lu a paper published, however, in 1S'J9, in the " Bulletin de 

 la Societe Astrononiicjue," a M. Touehet, quite misuuderstauding^the 

 nature of the oeeurrence, introduced tlie iinseicntific notion of 

 " lightning ribbons running along the clouds' edges." ■ 



overlooked, that the Ether as a transmitting medium 

 for light IS limited in extent, and belongs speciallv to 

 a cluster of stars of which our solar system is a part, 

 and that it thins out at a finite distance from that 

 cluster (in the same way that our atmosphere thins out 

 at a distance from the earth) until it is incapable of 

 transmitting light. In that case, the Stellar Universe 

 might be infinite, but there would be a " rapid decrease 

 in the luminosity of the stars ' at a great distance from 

 us, owing to the decreased transmitting power of the 

 ether ; and though the stars were just as numerous 

 bovond the ether, there would be nothing to tran.'^mit 

 their light. Geo. Phelps. 



Sirs,- 



TO THE EDITORS OF KNOWLEDGE. 



-Mr, Anderson says that " individually they " 



(the brighter and fainter stars) " do not lose the same 



