TRANSACTIONS OF SECHON C. 737 



not been in operation for a sufficient length of time to eliminate the felspar. A 

 pure quartzose sand is probably never formed by the direct denudation of a granitic 

 or gneissose area. The coarser sediments thus produced contain in most, if not in 

 all, cases a considerable amount of felspar. But felspar is more liable to decompo- 

 sition by percolating waters when it occurs as a constituent of grit than when 

 present in the parent rock. Silica may thus be liberated in a soluble form and 

 subsequently deposited on the grains of quartz so as to give rise to secondary 

 crystalline faces, and kaolin may be produced as beautiful six-sided tablets in the 

 interstices of the grit. When the gi-it is in its turn denuded the felspar is still 

 further reduced in amount, and a purer quartz-sand is formed. As the coarser 

 detrital material is used over and over again, thus measuring different periods of 

 time like the sand in an hour-glass, the felspar and other decomposable minerals 

 are gradually eliminated. The occurrence of a large amount of fresh felspar in the 

 Torridon sandstone might, I say, at first sight be thought to be due to the great 

 age of the rock. Any tendency to accept a view of this kind is, however at once 

 checked when attention is paid to the pebbles in the coarser conglomeratic beds of 

 the same deposit. These consist largely of quartzite — a rock formed by the con- 

 solidation of as pure a quartz-sand as any known to exist in the later formations. 

 We are therefore led to_ the conclusion that the special features of the Torridon 

 sandstone are not a function of time, but of the local conditions under which the 

 rock was produced. 



A similar conclusion may be reached by considering other types of sediment. 

 ^\lien the stratified rocks of the different geological periods represented in any 

 limited area are compared with each other certain marked differences may be 

 observed, but the difl!"erent types formed in any one area at different times can 

 often be parallelled with the diflTerent types formed in different areas at the same 

 time, and also with those now forming beneath the waters of rivers, lakes and seas. 

 Deep sea, shallow water, littoral and terrestrial deposits can he recognised in the 

 formations belonging to many geological periods, from the most ancient to the most 

 recent ; and there is no evidence that any of our sedimentary rocks carry us back 

 to a time when the physical conditions of the planet were materially difierent from 

 those which now exist. After reviewing all the evidence at my disposal, I must 

 however, admit that the coarser as well as the finer deposits of the earlier periods 

 appear to be more complex in composition than those of the later. The grits of the 

 Palseozoic formations, taken as a whole, contain more felspar than the sandstones of 

 the Mesozoic and Tertiary formations, and the slates and shales of the former contain 

 more alkalies than the clays of the latter. This statement will hold o-ood for the 

 British Isles, even when allowance is made for the enormous amount "of volcanic 

 material amongst the older rocks— a phenomenon which I hold to be of purely 

 local significance— but I strongly suspect that it will not be found to apply univer- 

 sally. In any case, it is not of much importance from our present point of view. 

 AH geologists will admit that denudation and deposition were taking place in 

 pre-Cambrian times, under chemical and physical conditions very similar to, if not 

 identical with, those of the present day. 



There is, however, one general consideration of more serious import. Additions 

 to the total amount of detrital material are now being made by the decomposition 

 of igneous rocks, and there is no doubt that this has been going on during the 

 whole period of time represented by our stratified deposits. It follows, therefore, 

 as a necessary consequence that strict uniformitarianism is untenable, unless we 

 suppose that igneous magmas are formed by the melting of sediments. 

 _ So far we have been dealing with the characters of sedimentary rocks as seen 

 in hand-specimens rather than with those which depend on their distribution over 

 large areas. Thanks to Delesse ^ and the officers of the ' Challenger ' Expedition,^ 

 an attempt has now been made to construct maps on which the distribution of the 

 sediments in course of formation at the present time is laid down. It is impossible 

 to exaggerate the importance of such maps from a geological point of view, for on 

 the facts which they express rests the correct interpretation of our stratigraphical 



' Litltologie du Fond des Mers. Paris, 1871. 

 ^ R/^poH on Beep-sea Deposits, 1891. 

 1893. ^ 3 3 



