86 BULLETIN OF THE 



II. The combination of sundry shoaler-water collections, made by 

 Pourtales and Agassiz on the Coast Survey steamers " Bibb " and 

 " Hassler," with the deep-sea dredgings, has proved of the highest im- 

 portance, by completing the evidence in several cases where the absence 

 of material from shoal water would have rendered a suspension of judg- 

 ment necessary. 



III. In several cases where the presence of dead sliells in the deep- 

 water material was the only evidence of the presence of a shoal-water 

 species there, its living presence has not been taken as proved unless 

 the multiplication of instances and graduation of depths confirmed the 

 supposition. If a too great conservatism has been exercised in this way, 

 it is on the side of safety in the generalizations. The names provision- 

 ally adopted in the tables are of a conservative character as regards 

 their limits ; since, in this way, a more just comparison with the lists 

 of authors like D'Orbigny and C. B. Adams is rendered possible ; and 

 this course is also less likely to result in errors of determination due 

 to insufficient study. • 



IV. The absence of any tolerably complete catalogue of West Indian 

 moUusks in accessible shape has interfered with carrying the comparisons 

 as far as might have been desired. The best that could be done was tq 

 compare the lists of C. B. Adams's Jamaican shells and those described 

 in D'Orbigny and Sagra's Mollusca of Cuba, to eliminate identical species, 

 and to assume that the resulting list bore about such a proportion to 

 the w^hole literal molluscan fauna of the West Indies as the " Blake " 

 dredgings do toward the whole abyssal fauna. Upon this assumption, 

 however, though so convenient for a brief comparison, no very impor- 

 tant conclusions are based. As the shells quoted by the above-men- 

 tioned authors were all (or nearly all) obtained in the limits of the 

 shore fauna, they afforded a better means of comparing that faunal re- 

 gion with the abyssal region than more modern and complete lists like 

 that of the shells of Guadaloupe (Crosse and Fischer), w^hich contains 

 many true deep-water species brought up on fishing-lines or by coral- 

 hunters. 



The following are the most interesting and important deductions 

 which seem to result from the facts before me. 



I. The fiicts, already known, that certain species of mollusks have a 

 very limited vertical range, forming respectively a literal and an abyssal 

 fauna, are supplemented by the additional hitherto unrecognized fact 

 that a fair proportion (say 20 per cent in the present case) have a verti- 

 cal range which extends from the true literal region (less than 50 



