UPPER EOCENE (BARTON AND UPPER BAGSHOT FORMATIONS). 621 



they are entirely confined to a special horizon is no doubt due to 

 the absence elsewhere of any similar pockets into which such small 

 shells were drifted and have been preserved. A tiny coral and the 

 fingers of small crabs' claws are mingled with them in equal profu- 

 sion. Many of the species are exceedingly like living shells from 

 Australia and Japan, and seem to indicate a considerable depth of 

 water with light drifting currents. Many rare freshwater shells 

 are met with in this fauna, the larger of them being almost invariably 

 abraded, as if brought from long distances. 



The second fauna is best represented in the Middle Barton, though 

 few of the species are actually confined to it. The shells are of 

 large size, and comprise the bulk of the typical Barton forms figured 

 by Brander. Most of the striking ones are extinct, but others, such 

 as Ficula nexilis, Cassidaria nodosa, and the species of Pleurotoma and 

 Natka^ are so nearly identical with living forms, that representatives 

 of them may be said to exist. 



The third fauna is that of the CAamcf-beds, comprising a number 

 of exquisite and entirely distinct shells of moderate size, whose sud- 

 den appearance is to be attributed less to an interval of time than 

 to a change in the outfall of the river, by which the muddy water 

 and silt of an estuary gave place to clear water and a sandy bottom. 

 An enormous colony of CJiamas and the ubiquitous Turritella took 

 possession of the area ; but not the least remarkable circumstance 

 is that the old representative species of several genera were suddenly 

 replaced by others that, though quite distinct, seem closely allied. 

 Thus Voluta humerosa replaces V. maga, Murex tripteroides super- 

 sedes M. asioei% TypJiis Jistidosus displaces T. pungens, &c., while 

 nearly the entire tribe of Pleurotoma give way to clear-water 

 Cowries, Cones, Mitroi, Murices, &c. The survival of stray and 

 often water-worn specimens of Middle Barton species does much, 

 however, to rob the Chama-heds of the peculiar facies of their fauna 

 when tabulated, and renders the break far less apparent than it 

 actually is in the field. 



The fauna from the Long-Mead-End Sands is again very distinct 

 indeed in its general facies from those which precede it. Its most 

 noticeable feature is the large proportion of Cerithia and of OUva 

 Branderi. It possesses a peculiar Natica and Marginella^ and species 

 of Melania and Melanopsis similar to those of the Headon Beds above, 

 while, owing probably to an influx of brackish water, the whole group 

 of Volutce, Pleurotoma?, and Murices so characteristic of the Lower 

 and Middle Bartons have disappeared. About a dozen of its com- 

 monest species are, in fact, indicative of brackish, if not of fresh 

 water, while an equal number of hardly less abundant, truly marine 

 forms pass up from below. 



The list of Barton fossils comprises 23 Vertebrates, 47 Invertebrates 

 other than MoUusca, 257 Gasteropods, and 150 Bivalves, exclusive of 

 over 120 undetermined species. Twenty-eight of the Mollusca first 

 appear in the London Clay and range for the most part above the 

 Lower Barton, though 7 of them are absent in the Lower and 

 3 in the Upper Bracklesham. A further 37 species first appear 



