1895.] on the Radiant Heat from the Moon during an Eclipse. 625 



the former coincides far more nearly than does the total heat-curve 

 with the latter. 



Slide 5 gives the " Extinction-curve " for heat, compared with 

 Seidel's for light. The curve gives the logarithm that must be 

 added to that of the reading at any zenith distance to obtain the 

 logarithm of the reading as it would have been had the Moon been 

 This curve gives a percentage to be added to the 



in the zenith.* 

 readings of 



1 J at 29° zenith distance 



7 „ 45° 

 19 „ 60° 

 28 „ 65° 

 50 „ 79° 



i— /■' \ 



M : \ 



MO I \ 



*** ''/" V 



"* '/ \ ' 



K // P-.cC..,. V. 



p.... ''.».>' -i 



i - i w n -n , -» t i -1 n a — ! I -■ ■ ■' -j a •» -> ■* g m ■*■. 



Fig. 4. 



It then occurred to us that, though this real absorption of heat 

 takes place so quickly that no appreciable time occurs between its 

 reception and emission when observed through the comparatively 

 slow changes of its incidence in the course of the lunar month, surely 

 with the far more rapid changes of illumination during an eclipse — 

 some three or four hundred times as rapid — the delay in the emission 

 might be easily perceived. 



Accordingly we have been on the watch for every lunar eclipse. 



Such events are not very frequent, and of such as occur some are 

 unsuitable for observation owing to the Moon's low altitude (or to her 

 being below the horizon during the whole or part of their duration), 

 while of the remainder most are more or less useless for our purpose 

 owing to the uncertainties of our climate. 



Before the date of my former lecture one eclipse had been utilised 



* See also Proc. Roy. Soc, 1869, No. 112, p. 436, aui 1870, No. 123, p. 9 ; 

 also Phil. Trans. 



