334 Prof. Frankland on the Physical Cause 



It is necessary to mention that the surface of the tin-plate 

 cube in contact with the hot plate was coated with a film of 

 lampblack, and that the surfaces of the granite were only roughly 

 ground and not polished; in fact all the experimental conditions 

 were in favour of the transmission through water, and yet the 

 surface temperature of the latter did not rise with much greater 

 rapidity than that of the former. Now taking into account the 

 specific heats and gravities of water and granite, the amount of 

 heat transferred from the base to the summit of each block will 

 be in the following proportion for equal times :— 

 Granite : Water. 

 1 : 2-36. 



Although these determinations can only give a rough approxi- 

 mation to the respective velocities with which heat passes through 

 strata of granite and water, yet they show that the convection of 

 heat through the latter is by no means so rapid, when compared 

 with its conduction through granite, as is commonly supposed. 

 There can be no doubt that, in the case of the ocean, the rapidity 

 of transfer would be increased by polar and equatorial currents ; 

 nevertheless the assumption that the cooling of the floor of the 

 ocean proceeded as rapidly as if it had been freely exposed to 

 the air is altogether untenable ; and I conceive it not only pos- 

 sible but probable that this secular cooling of the earth through 

 the ocean may have continued down to a comparatively very recent 

 geological period, and may, even at the present day, not be alto- 

 gether interrupted. 



The greater facility with which heat is thus conveyed through 

 water would obviously render the rate of downward increase of 

 warmth for a given surface temperature much less -rapid than in 

 the case of granite. Thus I consider it probable that the internal 

 heat of the earth affected in a very marked degree the surface 

 temperature of the ocean long after it had ceased to influence 

 appreciably the external warmth of the land. This assumption 

 acquires considerable support from a consideration of the condi- 

 tions tending to retard the escape of heat from the oceanic sur- 

 face as compared with that from the land. The facility with 

 which radiant heat escapes from equal surfaces of water and gra- 

 nite at the same temperature through perfectly dry air is nearly 

 equal ; but so soon as aqueous vapour is interposed in the path 

 of these rays, the conditions become wonderfully altered ; the 

 escape of heat from both is diminished, but its radiation from 

 the water is retarded in by far the greatest degree. This extra- 

 ordinary intranscalency of aqueous vapour to rays issuing from 

 water has been conclusively proved by Tyndall in a paper just 

 communicated to the Royal Society*. 



* Proceedings of the Royal Society, vol. xiii. p. 160. 



