MAY 24, I9I7] NATURE ------245 but I Prof. Stefan lYle:yer may be I .have pointed out the unsuitability of makmg some exammatwn of the radiations of the thonum mmerals for age determination or correlation material, and the results he obtains will therefore be and this is particularly so i.n the case of minerals of very great value in deciding this point. the Palreozoic igneous rocks of Langesundfjord, Nor• FREDERICK SoDDY. way. Mr. Lawson and myself based our former con• Aberdeen, May 14. that lead could not be the end product of thonum largely on analyses of these minerals. How• . PRO.F. SoDDY. having given me the privilege of read• e:ver, I have recalculated the ratiQs on the assump• mg h1s letter m advance, I should like to take the twn that thonum has one-seventh the lead-producing opportunity of directing attention to the geological age power of uranium, and it is satisfactory to find that :>f the thorium minerals of Ceylon, and to a few further when thorium is less than five times a·s abundant statistics bearing on the suggestion that only 35 per uranium, the ratios agree as closely on th'ts calculation cent. of thorium produces a stable isotope of lead. as do the simple lead-ratios. When thorium is more than five times as abundant as uranium neither set I a!ll to my friend, Mr. E. J. Wayland, late of ratios gives any approach to agreement although of Ceylon, for the follow• mg prov1s1onal classdicat10n (in order of age) of the the minerals from anv one locality agree am'ong them• older rocks of the island :- selves. (For the analyses referred to, see Phil. Mag. (6) Newer pegmatites and quartz reefs. vol. xxv.iii., p. 832, 1914; and Proc. Geol. Assoc.: (s) \Velipatanwila series of sediments. vol. p. 302, I<}I5)· Thus, having found from (4) Pyroxenites. the pe_rnicious and irregular behaviour (3) Hornblen?e, zircon, and other pegmatites of the of thonun: mmerals to be very misleading, I Balangoda senes (source of thorite and thorianite). must admrt that therr evidence is worthless in the Galle group, and crystalline limestones ( ?) absence of atomic weight determinations. (2) Charnockite series (pyroxene granulites). An atomic weight determination bv Richards and (r) Older. rocks of. the basal complex, including I;-embert on lead from Ceylon thorianite may be used gne1sses, w1th monaz1te and zircon pegmatitic secre- (m the same way as Prof. Soddy has used his own tions, and dioritic intrusions. ' and Hi:inigschmid's measurements) to test the question of end product. The thorianite referred to contained . classificat!on up much o.f the mystery m wh1ch the thonum mmerals of Ceylon have hitherto 6o per cent. Th, and 20 per cent. U, and therefore if the v:rhole. of the thorium disir.tegrated into lead, the been shrouded.. was at first that they be• longed .to two d1stmct penods (Nos. (r) and (1) in the atomiC wetght should be 207·32; whereas if only 35 per above hst), and as the figures in the table given below cent. of the thorium formed lead, then the atomic clearly indicate, the belief was curiously supported by weight be 206}3·. that the pre• t?e Pb/U ratios. However, it is now fairly estab• sence of ongmal lead rs 1mphed by the high lead-ratios hshed that thorite and thorianite do not occur in the of the a?ove . table, ::end that such exogenous lead o!der rocks of .the basal complex, and therefore the would rarse shghtly the latter figure, one finds with htgher lead-ratios are misleading for age measure• pleasure that the atomic weight actually found was ment. The only lead-ratios of any value for this pur• 206·82. Soddy's suggestion thus affords a happy pose are those of uraninite and zircon the former as to the end products of thorium; there being of the same age as the thorium minerals, while IS already accumulative evidence in its favour and the latter belongs to the pre-Charnockite zirconiferous as yet there is none against it. In particular it is' satis• rocks, and is therefore very much older. to observe if the suggestion should receive In the following table I have recalculated the lead• demonstratiOn, then the estimates of geological time already based on lead-ratios are not appreciablv on the assumption that 35 per cent. of thorium g1ves le