■)0< 



NA TURJi 



[Al'KIL I I. 189: 



LETTERS TO THE EDITOR. 

 The Age of the Earth. 



I AM surprised to observe, in the article which Prof. Sollas 

 has written on this subject in your issue of the 4th inst., p. 53 j, 

 that he speaks with approval of Dr. A. R. Wallace's method of 

 calculatine the earth's age. About two years ago (I hive only 

 this week's number of Natl'RE at hand) I wrote to you on this 

 subject, and was undenthe impres>ion that I hid proved the 

 complete fallacy of Dr. Wallace's method of calculation. 



To put Dr. Wallace's view briefly, he assumes that deposition 

 within a limited areaof, if I remember rightly, 3,000,000 square 

 miles, goes on 19 times as fast as denudation over the whole 

 land area, which is 19 times as sreat, and then argues that the 

 whole maximum thickness of the stratified rocks (and hence the 

 earth's age) could be deposited m 1/19 of the time required to 

 carry away from an equal area of land an equal bulk of material. 



The falKicy consists in assuming that a great rapidity of deposit 

 over a limiied area can in some way allow of the deposit or 

 formation of sedimentary rocks at a greater rate th.in that of 

 denudation. 



It is obvious that, in a given lime, no greater volume of deposits 

 can be formed than the volume of material denuded in the same 

 time. If, therefore, as Prof. Sollas assumes, 1/2400 of a foot of 

 sediment per annum is denuded from the land ari-a, by no ar- 

 rangement can a land area of equal extent, consisting of sedi- 

 mentiry rocks of the same composition and thickness as those 

 which actually constitute the land area, have been formed as a 

 whole more rapidly than I fool thickness over 5 7,000,000 square 

 miles area in 2400 years. Taking the estimate of Prof. Sollas, 

 viz. 164,000 feet, as the maximum thickness of the sedimentary 

 rocks, and taking the existing land area to be accounted for as 

 57,000,000 square miles, the lime required to form an area of 

 57,oco,ooo square miles of rock 164,000 feet thick, at 12400 

 of a foot per annum, is 393,600,000 years, unless the area under- 

 going denudation was greater or less than it is at present (and 

 it could not be four times as great as at present). No con- 

 centration of the deposit over a small area would shorten the 

 lime required by a single moment. Bernard Hobson. 



If, in the compass of a short article, I did not allude to the 

 controversy which followed the attack made by Dr. Hobson 

 (Nature, vol. xlvii. p. 175, 226) on Dr. Wallace's method of 

 estimating the age of the stratified "seiies, it was because I 

 thought, as I do still, that the honours of that controversy rested 

 entirely on the side of Dr. Wallace. 



There is no fallacy in Dr. Wallace's argument, but a strange 

 misconception on the part of Dr. Hobson, which arises from his 

 consistent disregard of the word maximum as prefixed to the 

 estimated total thickness of stratified rocks. It is obvious that 

 stratified systems cannot have a /«ajr««/«;« thickness everywhere 

 over the whole 57 million square miles of the land suiface. As 

 a matter of observation, a system attains its maximum thickness 

 over a very limited area, and over a large part of the 57 millions 

 of square miles of land surface it has no thickness at all, or, in 

 other words, is entirely absent. If " maximum " could be made 

 to mean the same as "average," no doubt Dr. Hobson's con- 

 tention would h lid, but those who have made use of a 

 maximum in estimating the age of the stratified scries have 

 observed a strict distinction in the application of the two terms. 



Rathgar, Apnl 9. W. J. Soi.las. 



Poly em bryony. 

 I.N connection with the note in the last number of Nature 

 on the above, I think it should be known that llie phenomenon 

 was incidentally observed some two years ago in the re<l beet 

 (Hela rubra) by the late Mr. Romanes and myself. We found 

 that a single seed might produce as many as four distinct plants, 

 and as (ar as our observations went, polyemhryony was quite 

 the normal condition. It seems lobe more characteristic ol the 

 Gymno-perms than the Angiospcrini, and has ol course been 

 investigated in the former, and in the latter among the Mono- 

 cotyledons (Treljakow) and Dicotyledons (t.g. Citrui-St.ra.-i- 

 burger). The (act of its occurring in such n common type as 

 £. rubra should, I think, lie taken advantage of by some 

 botanist, as the results could not fail to be both interesting and 

 important. Tretjakow's discovery that the supernumerary 

 embryos in Monocotyledons may be produced by the antipodal 

 cells, certainly suggests his compaiison between such embryos 

 and those pro luced by [parlhenogenelic?] apogamy on the 

 prothallia of the lower plants. Kra.sk I. Cole. 



NO. 1.^78. VOL. 51] 



IMPROVEMENTS I.V PHOTOMETRY. 



XT EARLY sixty years have passed since it first oc- 1 

 -'-^ cuired to the philosophic mind of Sir John 1 

 Herschel to attempt an arrangement of the relative 

 brilliancy of the stars, upon a method that should be 

 more secure than the eye estimations that had done duty 

 for many centuries. It is not necessary to enter into any 

 description of his method, which may be regarded now 

 as entirely superseded. Doubtless, had he been sur- 

 rounded by skilled workmen, furnished with better tools, 

 the cumbrous method employed would have been sim- 

 plified, but the establishment of an observatory remote 

 from the assistance and contrivances of the workshop is 

 not without drawbacks, as he and others since h.ive dis- 

 covered and regretted. About the same time, Seidel, 

 in Germany, was at work on the same problem, and the 

 fact that two astronomers, independently of each other, 

 undertook the solution of the same problem, is a proof 

 that it was ripe for mature consideration, while the series 

 of astronomers who have laboured in the same path 

 confirms the suspicion that this kind of investigation too 

 long neglected offered a field having a rich prospect of 

 reward. 



But a photometer at once convenient and capable of 

 general application to the stars remained to be invented, 

 and this want was effectually supplied by ZoUner, who 

 proposed a form of construction which has certainly ob- 

 tained the most general use of any of the suggestions 

 that have been from time to time put forward by astrono- 

 mers, who have recognised its deficiencies and tried to 

 remedy thetn. The distinguishing characteristics of the 

 Z<)liner photometer are the introduction of an artificial 

 star formed from a lamp shining through a small aper- 

 ture, and the controlling of the light of that star by means ] 

 of polarisation. This principle is now of such general use I 

 that no lengthened description is necessary. But to ex- 

 plain the reason for the introduction of other forms of 

 photometer, it is necessary to point out what are, or 

 what were, considered to be its defects by those who tirst 

 used the instrument, defects which it is believed care and 

 experience have sinccdone much todiminish.if not entirely 

 to remove. A source of error might be anticipated in the 

 varying brilliancy of the lamp employed to form the 

 artificial star, and in the early days of the instrument 

 this was a fruitful source of annoyance. Next, the light 

 of the lamp had to strike no less than twenty-eight sur- 

 faces, and apart from the difficulty of getting so many 

 surfaces true, and ensuring the parallelism of the Nicol 

 prisms by which the diminution of the artificial star !■; 

 effected, there is also to be considered the inevitable loss 

 of light at so many surfaces. One consequence of this is 

 that the brightest stars of the heavens are a])t to be 

 brighter than the artificial star, and since the observa- 

 tion is inade by reducing this light to match that of the 

 real star, it is necessary to have recourse to some such 

 expedient as reducing the aperture of the telescope. 

 And then a difficulty is encountered which has not yet 

 met with a complete explanation. The light deducted 

 from the star, as seen with a reduced aperture, does not 

 coincide with that which would be predicted from theory. 

 In some of the recent series of observations the dif- 

 ferences between observation and theory are as great as 

 they are perplexing. " There can be no doubt," wrote 

 Mr. C. S. I'eirce, of Harvard, twenty years ago, " that the 

 errors introduced by the use of these diaphragms are 

 by far the most serious of those by which my observa- 

 tions are effected." Dr. Wolff met with similar difficulties, 

 and dotibtless anomalies such as these have encouraged 

 the production of other photometers which should be 

 free from the suspicion of error. Having regard to the 

 photometric work actually accomplished, we may confine 

 attention to two forms of apparatus known as the Picker- 

 ing Meridian Photometer and the Pritchard Wedge 



