February 22, 1900] 



NA TURE 



391 



LETTERS TO THE EDITOR. 



[Tie Editor does not hold himself responsible for opinions ex- 

 pressed by his correspondents. Neither can he undertake 

 to return, or to correspond with the writers of, rejected 

 manuscripts intended for this or any other part of NATURE. 

 No nottce is taken of anonymous communications.'] 



Lord Kelvin on the Origin of Granite. 



In 1897 Lord Kelvin delivered an address to the Victoria 

 Institute on the age of the earth, Sir G. G. Stokes lieing in the 

 chair. Incidentally, the address included theories of the origins 

 of granite, basalt and continents, and touched on the question 

 of the inclusion of gases in various rocks. 



At the late meeting of the British Association two sectional 

 Presidents referred to Lord Kelvin's theory, both apparently 

 accepting his lordship's conclusions without noticing his 

 premisses. 



My old master, William Pengelly, used to teach that as every 

 theory depended on many facts, if but one fact were disproved 

 the theory fell to the ground. From this aspect there are 

 several points in Lord Kelvin's theory which seem to require 

 elucidation. 



Lord Kelvin starts from a period in the earth's evolution 

 when a lava ocean forty kilometres deep covered a solid 

 nucleus. The specific gravity of the liquid lava is assumed to 

 be 2 50. It is also assumed, on what seem firm grounds, that 

 solid lava would sink in fluid lava, and on more doubtful 

 grounds, that all minerals crystallising out of lava would also 

 sink. On this assumption the lava ocean silts up, and the sur- 

 face does not " freeze " until the forty kilometres of the crust 

 (excepting the future ocean basins^ are composed of solid 

 crystals set in an interstitial mother liquor. 



According to Lord Kelvin, continents arose from the drifting 

 and unequal distribution of the crystals falling like a "snow 

 shower " through the lava ; the ocean basins arising from the 

 contraction of the mother liquor in cooling. The theory is 

 brilliantly unfolded, but there are many technical difficulties, 

 e.g. Lord Kelvin's granite, besides being an exclusively 

 primeval rock, is composed of drifted crystals of felspar, mica, 

 quartz and hornblende (or some of them) set in a basaltic 

 matrix. The ultimate mother liquor is made to serve the 

 purpose both of basalt and of the matrix of granite. One gas 

 at least, viz. C.Oo, treated as original in basalt, might well arise 

 from the calcite which so often occurs in that rock. But the 

 most serious and far-reaching difficulty is involved in the 

 specific gravity of lava minerals. Assuming the liquid lava to 

 be 2 '50, there are several minerals which would float in such 

 an ocean, and if there were but one, it would suffice to provide 

 a floating crust or slag which would blanket the glowing lava 

 and entirely upset all heat calculations based on the consolida- 

 tion of the earth from within to without. 



What perplexes me in the matter is that so many philosophers 

 who accept Lord Kelvin's conclusions hold themselves at 

 liberty to reject his premisses. 



For instance, the President of Section E, while declaring his 

 adhesion to Lord Kelvin's time-views as against the geologists, 

 entirely ignores both his continental theory and the main 

 premiss as to the specific gravity of the lava ocean. Lord 

 Kelvin assumes the lava to be 2 "So, while Sir John Murray 

 assumes the crust to be 2*50, a most fundamental distinction. 



Petrologists have fought desperately over the question of the 

 origin of granite, but so far as I am aware they are agreed on 

 all the main points. 



I believe myself that every known fact fits most exactly 

 into the grand working hypothesis that granite is a plutonic 

 rock formed by hydrothermal action and pressure out of a 

 previously existing rock, which consisted in the first place of 

 those light aluminous soda and potash silicates which first 

 consolidated on the .surface of the primeval lava ocean. To 

 these silicates we have but to add water, in order to form (so 

 far as constituents go; a typical muscovite granite— absolutely 

 nothing more ; indeed, less, as we may omit the soda. The 

 constituents of muscovite (K20,2H20,3Al203,6Si02) are simply 

 leucite (K„0,Alj03,4Si02) and water. Introduce sodalite, an 

 even more likely constituent of the primeval crust than leucite, 

 and we have all our materials for ordinary granites, except 

 magnesia, with which, however. Lord Kelvin supplies us by 

 means of eruptions of the basaltic mother liquor. 



This is but a working hypothesis, but it will work ; whereas 



Lord Kelvin's novel theory throw^s the whole problem of granite 

 into inextricable confusion, even starting from the hopeless, 

 position of disagreement as to what the term granite denotes, 

 geologically, mineralogically and petrologically. 



The whole question seems to turn on one single point, viz. 

 whether the first primeval crust when cold floated on the sub- 

 jacent lava. According to the elements of mineralogy it would 

 do so ; but many physicists have assumed that upon consolida- 

 tion it would sink. Will not some of the distinguished specialists 

 in mineralogy and petrology pronounce judgment on this ques- 

 tion, which is really troubling unlearned and ignorant men who 

 are genuinely seeking information ? The conflict of authority is 

 quite overwhelming. A. R. Hunt. 



Torquay, February 1 3. 



Effects of Lightning upon Electric Lamps. 



I HAVE often seen luminous trails, similarinappearancetothose 

 shown in Mr. Webb's photographs (p. 343), in photographs takerv 

 at night. That there are any effects in these or Mr. Webb's 

 pictures that cannot be explained by a moving camera, I am 

 unable to convince myself of. The identical form of the dis- 

 charges from diflPerent lamps has been explained by the distance 

 of the discharge causing them. Granting that it is possible to 

 have a discharge, so intricate in character, exactly duplicated 

 at a second lamp (which is scarcely conceivable), their magni- 

 tudes in the pictures should be inversely proportional to their 

 distances. But we find that, in the pictures, the scrawls are 

 all of the same size. A lamp close to the camera, and a dis- 

 tant lamp, show the trails on the same scale. 



The beading of the trails can be easily explained by the 

 alternations of the current, the carbons fluctuating in brilliancy. 

 I am informed that in Dover these periodic fluctuations are 

 very conspicuous. Where a trail turns suddenly, the beads are 

 closer together, due to the motion of the camera being slower 

 when the direction of motion is about to change. 



If I remember right, there are one or two cases where we 

 have a very large and brilliant pattern, and several similar ones 

 on a smaller scale. This could be explained by reflection fron> 

 the inner surfaces of the lens. R. W. Wool). 



London, February 20. 



The Fitting of the Cycle to its Rider. 



I HAVE read Mr. Hutchins's communication (p. 368) with con- 

 siderable interest. Mr. Hutchins is at the head of the Forest 

 Department which has been recently established by the Cape 

 Government, and the improved method of riding that he has 

 adopted, in accordance with the views expressed in my recent 

 paper, have evidently been of service to him in traversing the 

 very rough country to enable him to carry out his duties. Mr. 

 Hutchins's experiences so closely agree with my own that I cai> 

 say little in criticism of his letter. I think, however, that he 

 will find that the gain from the lengthened crank advocated by 

 me cannot be explained by the very simple formula that he 

 gives. My son and I went very carefully into this matter at the 

 time I prepared my paper, and I think if Mr. Hutchins refers 

 to it he will find, if we consider the mechanical advantage apart 

 from the gain in nerve waste, the mathematical analysis of 

 ankling given by lengthened crank both give the only ex- 

 planation which would be satisfactory to a mathematician. In 

 our explanation the weight of the leg plays a very important 

 part, and it follows therefore that a heavy legged man gets 

 most from our system of riding. R. E. Crompton. 



Crompton Laboratory, Kensington Court, London, W. 



The point raised by Mr. Hutchins in his letter (p. 368) is 

 worth considering, but Mr. Crompton, who by his wonderful 

 riding has done so much to v>opularise the use of the long cranks 

 first suggested by Mons. Boulay, is not heavily built. A man 

 who is over fifty cannot move his legs so quickly as when he was 

 younger ; and so middle-aged persons, stout or slim, profit 

 greatly by using long cranks and high gears. Most people when 

 in a hurry run upstairs two steps at a time, and bicyclists, 

 whether in a hurry or not, find it an advantage to raise gears and 

 lengthen cranks. 



The question is, how far may we go without unduly increasing 

 the weight of our bicycles ? 



My age is fifty-one, weight 15 stone, height under 6 ft., and. 



NO. 1582, VOL. 61] 



