August 19, 1922] 



NA TURE 



the lithosphere at this depth, even at 10 miles, we have 

 as vet no positive knowledge, but the distribution of 

 igneous rocks at the surface of the earth, and a com- 

 parison of oceanic and continental regions, give us 

 some important facts to guide our reasoning on this 

 matter. We are probably not far from the truth if 

 we assume that the granitic portion of the lithosphere 

 is largely restricted to the continental regions of the 

 earth, and its thickness may not exceed an average 

 of about 5 miles. If so, assuming this granite layer in 

 continental regions to contain on an average 70 per 

 cent, of silica, and assuming that it is underlain to a 

 depth of 10 miles from the surface by basalt containing 

 on an average 48 per cent, of silica, this would give us 

 a silica percentage of about 59 for the average igneous 

 rock of the lithosphere in continental regions down to 

 a depth of 10 miles, which is in agreement with the 

 average of the igneous rock of the " crust " as estimated 

 bv Clarke and Washington. 



It should be noted that this takes no account of the 

 " crust " of the oceanic regions, which is probably in 

 large part basaltic. We may for the purpose of this 

 argument assume that the granite shell of continental 

 regions covers half the earth. This is an extravagant 

 assumption, but as it doubtless errs substantially in 

 exaggerating the acidity of the " crust," the error is 

 on the right side so far as the present argument is 

 concerned. If we further assume the sub-oceanic 

 " crust " down to a depth of 10 miles to be basaltic, 

 and to contain on an average 48 per cent, of silica, 

 this would give us an average igneous rock containing 

 about 53J per cent, of silica for the outer 10 miles of 

 the lithosphere all round the earth. 



Extending our considerations to a depth of 20 miles, 

 there can be little doubt that we should regard the 



deeper 10 miles as on the whole more basic than the 

 basaltic material of the outer 10 miles, and it is reason- 

 able to assume that this deeper layer of basalt does 

 not contain on the average more than 46 per cent, of 

 silica. If we make this assumption, then the average 

 rock of the earth's " crust " as a whole down to a depth 

 of 20 miles would contain not more than about 50 per 

 cent, of silica. 



Comparing these with the figures given above by 

 Clarke and Washington, the inference we draw is that 

 they have probably much understated the basicity of 

 the earth's " crust." Their average down to a depth 

 of 10 miles is, as we have seen, only acceptable for 

 continental regions, and cannot be admitted for the 

 earth as a whole. Still less can their average for the 

 lithosphere down to a depth of 20 miles be admitted, 

 for, as we have seen, there is good reason for believing 

 that the average rock down to this depth probably 

 corresponds to a gabbro, containing about 50 per cent, 

 of silica, rather than, as they infer, to a granodiorite 

 containing 59 per cent, of silica. 



This question of the average composition of the 

 earth's " crust " has important bearings on many 

 scientific and economic problems. It is quite commonly 

 assumed that the average igneous rock is intermediate 

 in composition, and that granitic and basaltic eruptives 

 are products of differentiation derived from inter- 

 mediate magmas. It seems highly probable, however, 

 that the average igneous rock of the earth's crust is 

 basic ; and although differentiation does undoubtedly 

 play an important part in the formation of igneous 

 rocks, the claim that granites and basalts are in general 

 differentiates from magmas of intermediate composition 

 has no adequate foundation in the facts known to us 

 concerning the petrology of the earth. 



Centenary of the Death 



ON August 25, 1822 — a hundred years ago — 

 William Herschel died at Slough, aged eighty- 

 three years and nine months. His scientific activity 

 had continued almost to the end of his long life. His 

 last published paper was read before the Royal Astro- 

 nomical Society (of which he was the first President) 

 in June 1821. It is the onlv one of his seventy memoirs 

 which was not published in the Philosophical Trans- 

 actions, of the yearly volumes of which for the years 

 1780 to 1818 inclusive only those for 1813 and 1816 

 contain nothing by him, while not a few volumes include 

 several papers from his hand. Even in the last year 

 of his life, when his son, under his continual guidance, 

 made and figured the i8f-inch mirror, which was after- 

 wards used by Sir John Herschel at Slough and at the 

 Cape, it is recorded that " the interest he took in this 

 work and the clearness and precision of his directions 

 showed a mind unbroken by age and still capable of 

 turning all the resources of former experience to the 

 best account." 



When Herschel, on March 1, 1774, began to keep a 

 record of what he saw in the heavens with telescopes 

 made by himself, it was natural that he should for some 

 years show no decided preference for any particular 

 branch of astronomy. At first he paid some attention 

 to the planets, and determined the rotation-periods 

 of Jupiter and Mars. But it did not escape his clear 



NO. 2755, VOL. I 10] 



of William Herschel. 



perception very long that what was urgently required 

 at that time was a systematic study of the vast number 

 of celestial bodies outside the solar system. If Herschel 

 had not early grasped this fact, and persevered all the 

 rest of his life in his devotion to sidereal astronomy, 

 he would never have become a great astronomer, but 

 would merely, like his contemporary, Schroter, have 

 been known as an indefatigable observer who occasion- 

 ally did some good work. But on his way from the 

 solar system out into space beyond it Herschel found 

 a new planet (Uranus), about twice as far from the sun 

 as what had up to then been considered the outermost 

 planet. This was not a lucky accident, but a discovery 

 which was bound to be made sooner or later by an 

 observer who searched the heavens as systematically 

 as he did. It was the first time since the prehistoric 

 ages that a new planet was discovered. Herschel 

 afterwards found two satellites of Uranus and two of 

 Saturn, but his principal work was always on subjects 

 connected with sidereal astronomy. 



" A knowledge of the construction of the heavens 

 has always been the ultimate object of my observa- 

 tions." This was the opening sentence of his paper of 

 181 1, and as he had said much the same in the con- 

 cluding words of his first paper (of 1784) on that 

 subject, we see how faithful he remained to the plan 

 of work he had adopted early in his scientific career. 



