W. A. Richardson — The Origin of Cretaceous Flint. 541 



In a recent paper J. Stansfield, working with gels and solutions of 

 different concentration, has added considerahly to our knowledge of 

 the rhythm. The following among his conclusions may be noted : — 



(«) Increase in the distance between the bands is due to progressive 

 dilution of the reagent, and the rate of diffusion is an important 

 controlling factor. Under certain conditions the bands may be equally 

 spaced, or spaced at decreasing distances. 



(I) Banding may sometimes break down altogether, and be 

 replaced by a granular zone. This, again, may be followed on further 

 diffusion by a resumption of banded precipitation. 



(«) (b) (c) 



Bichromate Gels. Flint in Chalk. Coal-seams. 



FlG. 2. — Khythm curves obtained by plotting distance separating bands against 

 the position of bands measured from a chosen origin. 



Now measuring up Stansfield's Fig. 2, and plotting the results 

 besides those of Liesegang, the rhythm-curve given in Fig. 2a is 

 obtained. 1 There is the closely banded stage AD as before ; a stage 

 DE of increasing separation not quite linear; a granular zone EF 

 where the precipitate is in spots scattered throughout the zone 

 without linear or other arrangement; and a final region FG where 

 the precipitate is in bands separated by constant intervals, but much 

 greater than in the earlier banded stage DE. 



We may now return to a consideration of the Chalk. In Fig. 3 

 I have constructed vertical sections for three coastal localities, 

 putting in the position of the flint bands. Quantitative data are 

 available for practically the whole of the Kent section, and in the 

 others it is possible to fill satisfactorily the gaps in the measurements 

 by qualitative statements, which are fortunately abundant. Now 

 all of these show a sufficiently striking resemblance to the 



1 J. Stansfield, Amer. Journ. Sci. , ser. IV, vol. xliii, p. 1. 



