W. A. Richardson — The Origin of Cretaceous Flint. 537 



favour of the replacement of the chalk by silica at the time of 

 uplift. Consideration of the remaining questions will be found to 

 strengthen rather than oppose this conclusion. 



3. The Source of the Siltca. 1 



Some suggestions as to possible sources of the silica may be 

 dismissed : such, for instance, that it was deposited by percolating 

 sea- water. Magmatic sources are excluded at any rate from the 

 English Chalk. Moore's 2 suggestion that the silicates of overlying 

 strata decomposed by carbonated water furnished the supply has no 

 geological evidence to support it. 



Sollas, 3 among recent writers, treating the subject quantitatively, 

 has maintained that the silica was derived from the tests and other 

 remains of siliceous organisms scattered through the Chalk. Jukes- 

 Browne always opposed this theory of disseminated silica, as he 

 called it. 4 Tarr more recently favoured the direct chemical precipita- 

 tion of silica from sea- water, and was impressed apparently by the 

 immense amount of silica present as chert. 



Jukes-Browne, as the result of analyses, stated definitely that "there 

 is no inverse relation between the abundance of flint and the presence 

 of disseminated silica". Since no later analyses seem to be available 

 I have abstracted in Table No. 1 those given by Jukes-Browne 

 himself. However, analyses of certain very siliceous beds in 

 Wiltshire have been omitted. Three of these give respectively 39, 

 19, and 17 per cent of soluble silica, and inclusion of them would 

 considerably raise the mean percentage of available soluble silica. 

 The analyses in the table are plotted in one of the curves of Fig. 1 — 

 the percentage silica horizontally and the height of the sample above 

 the base of the Chalk vertically. It will be noticed that there is 

 a marked increase in soluble silica reported in all analyses below the 

 Melbourn Rock. 



Turning now for a moment to consider the amount of silica 

 represented by the flint, I have made an estimate based on the Kent 

 section, since details are available which give almost complete 

 measurements for the whole of the Kent Chalk. The details of this 

 estimate will be found in Table No. II. In this table some columns 

 are devoted to a statement of the assumptions made when details are 

 not available. Assumptions have sometimes to be made as to the 

 number of bands present, their mean thickness, and the amount of 

 flint in each band. In making these estimates I have been guided by 

 actual statements available for the same zones in neighbouring 

 districts, and by my own observations, chiefly in the London area. 

 The mean results for each zone are plotted on the second curve in 

 Fig. 1 to the same scales as the soluble silica. 



1 In what follows numerical data relating to the Chalk, unless otherwise 

 stated, have been obtained from A. J. Jukes-Browne, Cretaceous Rocks of 

 Britain, Mem. Geol. Surv. ; and from the papers on the White Chalk of the 

 English Coast in the Proc. Geol. Assoc. (1899-1903), by A. E. Bowe and 

 CD. Sherborn. 



2 B. Moore, Trans. Geol. Physics Soc, 1917, p. 1. 



3 W. J. Sollas, The Age of the Earth, London, 1905, pp. 132-65. 

 * A. J. Jukes-Browne, Geol. Mag., 1893, p. 541. 



