SIXTH ORDINARY MEETING. 49 



The heat generated by the sun in assuming its present density and 

 conformation can not be supposed to be greater than that produced 

 by contraction from the limits of the solar system to a homogeneous 

 sphere of one-half its present radius. 



This would make 44,000,000 years, the limiting age which can be 

 assigned to the Cambrian formations. 



We shall conclude by applying the principle that the absorptive 

 power of a vapour is determined by that of the liquid from which it 

 is derived, to explain the empirical law (enunciated by Mr. McGree) 

 that any increase in annual or diurnal range is accompanied by a 

 diminution of mean temperature. The aqueous vapour of the 

 atmosphere, being derived from water, which has a comparatively 

 high absorptive power, must also possess considerable power of 

 absorption, and Tyndall has conclusively shown experimentally that 

 such is the case. Also, the power of radiation is strictly propor- 

 tional to the power of absorption, as is known both from theory and 

 experiment, so that the aqueous vapour of the atmosphere is both a 

 ' good absorbent and radiant. 



Now, when the temperature is raised, not only will the aqueous 

 vapour of the atmosphere be heated, but a larger amount of it will be 

 formed, and as gases expand when heated, this vapour will tend to 

 rise to the higher regions of the atmosphere, and radiate its heat 

 into space. On the other hand, in a comparatively cold season the 

 relatively cooled vapour tends to descend, the heated vapour from the 

 surface of the earth ascends, and imparts its heat to cold space. 



Also the amount of heat received from the sun may for our 

 present purpose be considered as invariable from year to year, so that 

 the two actions above mentioned show that the radiant absorbent 

 and expansive powers of aqueous vapour combine to lessen the rela- 

 tive amount of heat retained by the earth, during both exceptionally 

 high and exceptionally low temperatures, i. e., during a period 

 of large thermometric range, and consequently to diminish the mean 

 temperature. 



There may be and probably are other con-causes of this effect, but 

 the one we have assigned is certainly a real and efficient factor 

 in producing the apparently anomalous result in question. 



In the discussion which followed the reading of Mr. Clark's 

 paper, Mr. Geo. E. Shaw, Mr. J. G. Mowat, Dr. Jos. Workman, 

 and Mr. J. M. Buchan took part 

 4 



