Febeuaey 28, 1896.] 



SCIENCE. 



315 



There are two families of fresh-water 

 fishes confined to the cold and temperate 

 waters of the southern hemisphere and 

 generally distributed in such ; they are the 

 Galaxiids and Aplochitonids ; the former 

 were associated by the old ichthyologist with 

 the pikes, and the latter with the salmonids, 

 but they really have no such relationship, 

 but are closely related to each other and 

 segregated from all others. The Galaxiids 

 are represented by one genus, Galaxias, of 

 which about five species occur in South 

 America, five species in Tasmania, ten 

 species in Australia and five species in New 

 Zealand. (A monotypic genus, Neochanna, 

 is confined to New Zealand.) The Aplo- 

 chitonids number only six species, referable 

 to two genera; of these two are found in South 

 America (Aplochiton,) two in Tasmania 

 {Aplocliiton and Prototrodes) , one in Aus- 

 tralia and one in New Zealand {Proto- 

 troctes). 



It was long supposed that no species of 

 either family of Galaxioidean fishes oc- 

 cured in Africa, but last year Dr. Stein- 

 dachner described a representative of Ga- 

 laxias ( G. capensis) and consequently we 

 now have South Africa to consider with 

 reference to a former Qommunity of popula- 

 tion and continuity of land of all the south- 

 ern hemisphere. 



The conditions of existence and propaga- 

 tion of fresh- water fishes were then dis- 

 cussed and the chances against diffusion of 

 any fresh-water fish across the ocean or by 

 other means than natural water courses 

 were weighed. 



In finally taking into consideration the 

 limited distribution northwards and the 

 close relationship of the species of the sev- 

 eral regions referred to, it was urged that the 

 evidence in favor of a former Antarctic con- 

 tinental area was strong, and, in view of the 

 afBnities of the species of the now distant 

 regions, the conclusion was logical that the 

 time of disruption was not remote in a ge- 



ological sense. It was suggested that such 

 disruption might have been coeval with the 

 final uplift of the Andes. 



^ The amphibians and reptiles furnish no 

 data bearing directly on an Antarctic con- 

 tinent, but do yield some (though very 

 slight) bearing on an earlier and more 

 northern connection of the southern conti- 

 nents. Much more cogent and less ambig- 

 uous is the evidence resulting from the 

 study of the fishes. 



The fishes of tropical Africa may be 

 ranked under two grand categories. One of 

 these comprises species of genera or groups 

 represented more largely in Asia, and the 

 other of forms related to types otherwise 

 confined to tropical America. These Afric- 

 American forms belong to the extensive fam- 

 ilies of Characinids and Cichlids or Chro- 

 mids. Fishes of these families are the 

 most conspicuous and numerous in both 

 continents. The representatives of the two 

 families of the different continents always 

 belong to diflerent genera, and often to 

 different groups of genera or subfamilies. 

 We have, therefore, in the fishes, as in the 

 mammals, conflicting evidence. According 

 to one set of facts, the continents of Africa 

 and Asia are similar, and, in fact, they 

 have been united to form one zoological 

 realm ; according to the other the primi- 

 tive fauna of Africa is more like that of 

 America. Just two decades ago (1875) the 

 speaker explained this apparent contradic- 

 tion by the assumption that the aborigi- 

 nal types had been early derived from a 

 common source, and, for that reason, com- 

 bined Africa with South America and Aus- 

 tralia in a zoological hemisphere which he 

 named Eog^a, and contrasted with an- 

 other called C^NOG^A, embracing Asia, 

 Europe and North America. The numerous 

 species congeneric with Asiatic and Euro- 

 pean types, were considered to be recent em- 

 igrants, geologically considered. The pur- 

 port of all the evidence was that there may 



