632 The Earl of Rosse [May 31, 



6 American Journal of Science,' and in the ' Memoirs of the National 

 Academy of Sciences.' 



In 1883 Langley commenced his experiments on the relative per- 

 centage of the Sun's heat and the Moon's heat transmitted by glass, 

 obtaining not widely differing numbers for each with the interposed 

 glass near to the face of the bolometer; but subsequently, with the 

 glass removed to a greater distance, his results approximated to mine, 

 75 per cent, being the value found for sunlight, and 12 to 14 percent, 

 for moonlight with the same sample of glass. 



Langley, as far as he was able, observed the eclipse of October 4, 

 1884, and he says the eclipsed Moon rose behind clouds, and 

 there was still haze as the penumbra passed off. " The inference from 

 the observations, so far as any could be drawn, was that about the 

 same amount of heat was received as was to be expected had there 

 been no previous eclipse." 



He remarks of the Birr Castle observations during this eclipse 

 (before the complete accjunt had been published) : " They appear to 

 bear but one interpretation, that all the heat from the Moon disap- 

 pears immediately that it passes into the Earth's shadow, and there is 

 no evidence of any being retained for any sensible time, more than if 

 it were reflected." Tins we have seen appears to have been very 

 nearly, but perhaps not quite the case. 



Langley, in ' Memoirs of the National Academy of Sciences,' vol. iii. 

 infers : " 1st. That there is a certain presumption that the Earth's 

 atmosphere is diathermanous to heat of lower wave length than has 

 been heretofore supposed, and of lower wane length than appears to 

 reach us from the Sun. 



" 2nd. If we may make any inference from comparisons with the 

 Leslie cube, the sunlit surface of the Moon is not far from the 

 freezing temperature, but not so far below it as we might expect to 

 find that of an airless planet. 



" 3rd. So far as our limited observations allow of an inference, 

 there is a " not very materially greater coefficient " [query less] " of 

 transmission for lunar than for solar." 



The second of these inferences he makes from the distribution of 

 heat in the lunar spectrum, a very difficult thing to measure satis- 

 factorily. This he admits, and says that the work has been one of 

 great labour, but justified by its having been given to a question of 

 abstract interest, but one " to which the whole subjects of terrestrial 

 radiation and the conditions of organic life upon our planet are inti- 

 mately related." 



Remarks of a similar character may be made Avith regard to the 

 results of the eclipse observations. What limits the extent of our 

 atmosphere ? Is it boundless ? Does it always follow Boyle's Law, 

 or where does it depart from it ? These may be questions of interest 

 to physicists. The observations before the commencement of the 

 Eclipse of 1888 seem, therefore, so far as they go, to be of extreme 

 importance. It is certainly unsatisfactory, and perhaps it maybe pre- 



