February i6, 1922] 



NATURE 



207 



difference does not appear to be due to error in the 

 density determinations for the mass found in different 

 experiments, for a constant volume of the same 

 specimen of mercury is constant to one part in a 

 million, and with special care it is constant to a few 

 parts in ten millions. 



These experiments indicated, too, that any process 

 of distillation would give some separation of the 

 isotopes of mercury, and the question naturally arose : 

 Upon what evidence has the density of mercury been 

 regarded as constant? The matter had been inves- 

 tigated at the International Bureau of Weights and 

 Measures by M. Marek in 1883, and he writes of the 

 results which he obtained : " It is noticed in com- 

 paring these figures that the density of mercury varies 

 slightly from one sample to another according to the 

 method of purification. This result has already been 

 obtained by Dr. H. Wild in a study of this subject 

 specially undertaken." The results which Dr. Wild 

 published in 1874 are not available to us. M. Marek, 

 however, quotes results communicated to him by Dr. 

 Wild, which, although ambiguously stated, make it 

 appear as not improbable that Dr. Wild more than 

 fortv years ago separated mercury into specimens 

 of different density. T. H. Laby. 



^L W. Mepham. 



^K Natural Philosophy Department, University 



^v of Melbourne, December -^o. 



^^B Where did Terrestrial Life Begin? 



H^PThe question raised by Dr. Macfie in his letter in 

 ^^Natlre of January 26 concerning the place of origin of 

 life on the earth is not one which directly concerns the 

 meteorologist, but Prof. J. W. Gregory's comments 

 upon it seem to call for discussion from the meteoro- 

 logical point of view. Dr. Macfie suggests that in the 

 gradual cooling of the earth mountain-tops would 

 first reach a temperature to make them habitable for 

 human life, while the sea would tor further centuries 

 remain above the critical temperature. Prof. Gregory 

 feels hesitation in accepting the conclusion reached 

 that life would first be found on the mountains, con- 

 sidering that while " the mountain summits would 

 have stood like islands above a sea of hot mist . . . 

 any wind would have at times submerged the moun- 

 tain summits beneath the lower atmosphere, and they 

 would have been subject to violent fluctuations in 

 temperature and moisture which would have been 

 unfavourable to primitive life." 



Now with an atmosphere of homogeneous composi- 

 tion it is impossible to warm a mountain summit by 

 immersing it in vi'arm air drawn from the lower 

 layers ; if the conditions are initially stable, adia- 

 hatic cooling sees to it that the warm bath of 

 ir becomes a cold one before the summit is 

 ached. We must therefore assume that in these 

 Illy days the earth's atmosphere was not homo- 

 geneous, but that hot layers of dense gas occupied 

 llie lower levels, while lighter constituents of low 

 tiniperature floated above. In these circumstances 

 stirring up of the lower layers might raise the 

 nperature at higher levels temporarily, but is there 

 IV evidence that such a condition existed? No 

 ice of separation and stratification of the different 

 prises under gravity is found in the troposphere at 

 I the present time, atmospheric turbulence being suffi- 

 I cient to maintain a similar constitution at all heights. 

 Tf the gases were stratified in the manner suggested 

 it would afford proof that vertical turbulence did not 

 occur, and thus the very existence of stratification 

 . would show that the layers below never rose to the 

 mountain-tops 



NO. 2729, VOL. 109] 



Meteorological evidence does not seem to support 

 Prof. Gregory's conclusion that the mountain-tops 

 would be subject to such violent fluctuations of tem- 

 perature as would render life impossible. 



J. S. Dines. 



66 Sydney Street, S.W.3, February 6. 



Dr. Macfie's letter (Nature, January 26, p. 107) 

 accepts the common idea that the surface of the earth 

 was formerly very hot — an assumption which is 

 probably not well founded. If the earth was formed 

 by accumulation of meteoric matter, it began its exist- 

 ence as a cold body the interior of which afterwards 

 became heated by condensation, aided by atomic dis- 

 integration, while its surface was kept at a moderate 

 temperature by radiation. It is difficult to believe 

 that a globe so small, comparatively, as the earth 

 could produce enough heat to raise its surface tem- 

 perature anywhere near to the melting point ; all 

 igneous rocks are probably formed at some distance 

 beneath the surface. 



I imagine the first beginnings of life to have 

 occurred at a very early epoch in the earth's evolu- 

 tion, namely, as soon as (i) the surface became warm 

 enough and (2) elements capable of forming labile 

 energy-storing compounds were present. It is not 

 certain that solar radiation was necessary at first ; 

 the kinetic energy (heat and electricity) may have 

 been derived from the earth itself. 



Life at this stage would be of the humblest kind ; 

 we should scarcely recognise it as life nowadays. 

 There would be no definite organisms, only diffuse 

 substances trading in energy. Between this stage and 

 the development of cellular organisms an immense 

 period may have elapsed, and that period may have 

 witnessed many intermediate stages. The achieve- 

 ment of the cell-form in living organisms must have 

 marked a most important epoch in the historv of 

 life. 



Chlorophyll may have been evolved at quite a late 

 stage, as the culmination of a series of attempts at 

 the formation of energy-fixing pigmentary bodies, 

 most of which probably had iron as an essential 

 ingredient. 



The high stage of development shown in the earliest 

 known fossils suggests that the geological period 

 occupied by their evolution was vastly greater than 

 the period since. The dawn of life may have occurred 

 before there were either mountains or seas ; all evi- 

 dence of such early life fias been obliterated by the 

 metamorphosis and fusion of the deeper rocks. 



Further discussion on this subject mav be found in 

 a paper by the present writer in the Proceedings of 

 the Birmingham Natural History and Philosophical 

 Society, vol. 11, pt. i. ' F. J. Allen. 



8 Halifax Road, Cambridge, January 28. 



Rainfall and Drainage in 1921. 



I HAVE read with interest the letter of Mr. W. D. 

 Christmas in Nature of January 26 concerning the 

 rainfall and drainage at Rothamsted during the very 

 dry year 192 1. 



A few years ago three rain-gauges were installed 

 at Craibstone, the experimental farm of the North of 

 Scotland College of Agriculture. Like the Rotham- 

 sted gauges, each of these is one-thousandth of an 

 acre in area, and contains a block of soil which 

 has been enclosed in its natural condition without 

 disturbance. The soil at Craibstone differs greatly 

 from the heavy loam of Rothamsted, and is composed 

 of sharp granitic drift which is easily pervious to 



