176 MESSRS. A. J. JUKES-BROWNE AND W. HILL I [May 1 896, 



very highest hed, so that if two thirds of the French Cenomanian 

 are to be correlated with the thin bed of Greensand near War- 

 minster on the strength of the similarity between the faunas, we 

 must imagine that this thin bed has expanded to a thickness of 

 80 feet or more in Prance, although the beds both above and below 

 have very greatly diminished in thickness. 



Our lists, however, show that, of the fossils which may be 

 collected from these beds at Havre in a few days' time, quite as 

 many occur in our Chalk Marl as in the Warminster Beds. We 

 suspect that the occurrence of Pecten asper has influenced the French 

 systematists more than any other element in the fauna ; but our 

 experience in England has convinced us that P. asper is very 

 sporadic in its mode of occurrence, and required a very special kind 

 of environment. 



We consider it very unsafe to trust to echinoderms or ground- 

 feeding molluscs in correlating at so great a distance apart forma- 

 tions of different lithological character, because the conditions of life 

 may also have been different in the two areas. One condition alone, 

 namely, greater depth of water, might be sufficient to exclude from 

 the one formation species which found a suitable location in the 

 other area. This, we think, is the reason why Pecten asper and 

 many other of the Warminster molluscs and echinoderms do not 

 occur in the Chalk Marl, not because they had everywhere ceased to 

 exist, but simply because the water was too deep for them. 



The case is quite different with the cephalopada, for they could 

 move freely from place to place, and were not directly dependent on 

 depth of water or on the nature of the sea-floor. They must there- 

 fore be much more trustworthy chronological guides ; and if we rely 

 upon their guidance in the present case, we find that they lead us to 

 the very conclusion which we regard as correct, for the cephalopoda 

 of the lower part of the Cenomanian of Havre are identical with 

 those of our Chloritic Marl and Chalk Marl, and they do not occur 

 in the Chert Beds of our Upper Greensand. 



Let us now take another point of departure, and consider what 

 modifications are likely to present themselves in our Chalk Marl 

 when traced into shallower water and nearer to a line of coast. If 

 we examine the Chalk Marl of the Isle of Wight under the micro- 

 scope, we find that it actually is a Craie glauconieuse; it con- 

 sists essentially of minute grains of quartz and glauconite, and of 

 shell-fragments embedded in a fine chalky matrix. The material 

 of the corresponding part of the Cenomanian at Havre has less 

 of the chalky matrix, and a larger proportion of the inorganic 

 materials (quartz and glauconite). 



There is, in fact, just the difference that one would expect to find 

 in a contemporaneous deposit accumulated rather nearer to the land, 

 and we have shown that this change gradually increases till the 

 chalky ingredient disappears entirely, and we reach shallow-water 

 sands and sandstones. Would it not be surprising if there were not 

 a corresponding change in the fauna ? Is it likely that the same 

 assemblage of fossils would inhabit the littoral sandy floors, the 



