NOVEMBEE 22, 1895.] 



SCIENCE. 



679 



fishes. This great group includes nearly 

 two-thirds of all fresh-water fishes and 

 comprises the Characinids of America and 

 Africa, the Gymuotids of America, the 

 Cyprinids of the northern hemisphere and 

 the various families of Nematognaths. 

 These groups, in most systematic works, 

 have been widely separated and severally 

 associated with forms with which they have 

 no intimate relationships. As long as such 

 views prevailed, the appreciation of the 

 great importance of the geographical dis- 

 tribution of the groups was concealed from 

 view. But with the recognition of the 

 unity of organization, and, consequently, 

 unity of origin of the whole, a fresh con- 

 ception of the relations of that whole to 

 the faunas of the present and past breaks in 

 upon us. We are now justified, from the 

 morphological data at hand, in claiming 

 that all the groups enumerated as Ostaeio- 

 PHYSI and belonging to the orders Plectos- 

 PONDYLi and Xematognathi are naturally 

 segregated and not closely related to any 

 existing aboriginal marine types. The 

 marine forms of the family Plotosidce and 

 the siluroid sub-family Tachisurince must be 

 regarded as divergents from fresh-water 

 forms. With this assumption it becomes 

 necessary to postulate that all the numerous 

 families of the Plectospondylous and Nema- 

 tognathous orders are derivatives from 

 primitive fresh-water types. The extent of 

 this divergence may be inferred from the 

 numerous morphological modifications. 

 The antiquity of the origin of the super- 

 order must be commensurate with the ex- 

 tent of divergence. Far from originating 

 in the advanced tertiary, it is not unreason- 

 able to infer that the parent stock had be- 

 come acclimatized in the fresh water as far 

 back as the early mesozoic; instead of the 

 parent land being the Himalaya region 

 or highlands of Asia, as claimed by Dr. 

 Giinther, it is much more likely to have 

 been in the southern hemisphere— possibly 



an antarctic continent. At any rate, the 

 present geographical distribution of the 

 representatives of the respective orders 

 seems to render such an origin most prob- 

 able. The reasons were given in detail. 



The distribution of the group thus out- 

 lined is to some extent collateral with that 

 of certain mollusks and crustaceans, and 

 the facts respecting the range of the union- 

 aceous bivalves and the ostracod crustaceans 

 were especially discussed. 



It is quite true that there is no paleonto- 

 logical evidence for the inferences and as- 

 su.mptions thus made, but this is simply 

 because the geological record is wofully im- 

 perfect and many of the changes took place 

 in continental areas now submerged or little 

 explored. !N"o remains of Ceratodontids, 

 which must of course have lived to continue 

 the line from the Jurassic species to the 

 present, have been found. The same con- 

 ditions that have afiected the one must 

 have prevailed for the others. 



DRY DREDGING IN THE MISSISSIPPIAN SEA. 

 The U. S. National Museum has recently 

 secured large collections of Devonian fos- 

 sils, chiefly corals, from New York, Ontario 

 and Michigan. The first casts were made 

 in the Corniferous limestone in the vicinity 

 of LeEoy, New York, where the cherty 

 limestone underlying the quarry layers is 

 charged with an abundance of corals, the 

 net sometimes having masses of Diphy- 

 phyllum of more than a hundred pounds 

 weight. At Williamsville, near Buffalo, 

 corals are also plentiful, but here the fauna 

 is smaller and the species are not so com- 

 mon as at LeEoy. However, a short dis- 

 tance west of Buflklo, to the north and west 

 of Port Colborne, in Ontario, well-preserved 

 Corniferous corals are present in great 

 variety and abundance. Also at Hagers- 

 ville, large masses of various compound 

 species are numerous, many hundred tons 

 of which, two years ago, were broken up 



