622 MESSRS. GARDNER, KEEPING, AND MONCKTON ON THE 



in the Lower Bracklesliam, only 11 of which do not range ahove 

 the Lower Barton, These are reinforced by no less than 108 

 additional species in the Upper Bracklesham, 35 of which die 

 ont with the Lower and 24 with the Middle Barton. Thus of the 

 407 species, 175 range below the Barton, against 56 that pass up 

 into the Headon; but of the latter 30 are also Bracklesham and 

 London-Clay species. The upper limits of the Barton formation 

 are thus much more sharply defined, palaeontologically, than the 

 lower ; but we must remember that in the former case the passage 

 into fluviatile beds is abrupt, and the marine beds next above are 

 separated by a considerable thickness of freshwater deposits, while 

 in the latter the transition lies everywhere in marine deposits. The 

 reason for drawing the line between Oligocene and Eocene in our 

 area, here if anywhere, is quite obvious if our statistics are at all 

 reliable. 



The close connexion between the Upper Bracklesham and Lower 

 Barton is rendered very striking by these tables, no less than 35 

 species being quite peculiar to the two horizons when combined. 

 This contrasts with the 12 which are peculiar to the Lower and 

 Middle Barton combined, and the less than half a dozen peculiar to 

 the combined Middle and Upper Barton. The upper limits of the 

 Bracklesham should obviously, on palasontological data, have been 

 drawn much lower down. Only 16 are peculiar to the Upper Barton 

 and Headon combined, and these are mostly freshwater or brackish- 

 water stragglers. 



Of the tabulated Mollusca, 124 are absolutely peculiar to the Barton 

 formation in this country, though we must not lay undue stress upon 

 them, as we have seen that many of those most rigidly limited in 

 range in our area have a more extended or a different range in 

 the Paris area. Of these, 15 range through the three divisions, 

 12 through the Lower and Middle, 5 are confined to the Lower and 

 Upper, 3 to the Middle and Upper. This somewhat capricious dis- 

 tribution may be partly due to the extra turbidity of the water in the 

 Middle period. There are 51 species absolutely confined to the 

 Lower, 10 to the Middle, 28 to the Upper divisions. 



The distribution by genera is equally instructive : but in order to 

 have made an analysis, we must have introduced subgeneric names, 

 which would have lessened the value of the list for general comparison. 

 We have for the same reason retained many familiar generic names 

 which, on the ground of priority, must disappear. The general 

 resemblance between the facies of much of the Barton fauna and 

 that of the London Clay is not apparent in the table, perhaps because 

 species which did not hold their ground during the Bracklesham 

 period, but emigrated, were so considerably modified during the 

 interval that they can be distinguished as new species on their reap- 

 pearance ; while the modifications undergone by those that remained 

 were so slight that in the presence of connecting links they are 

 specifically inseparable. 



If we confine our attention to the species whose range is marked 



