June 8, 1900.] 



SCIENCE. 



901 



the Characinids and the Lepidosirenids, 

 which are the most prominent constituents 

 of the African fauna. These families are also 

 equally characteristic of tropical America, 

 but the representatives of the two conti- 

 nents belong to different genera. 



The deduction seems to be inevitable that 

 the main element of the piscine fauna was 

 derived from the same source as that of 

 America. The fact that generic differentia- 

 tion has supervened to such an extent sug- 

 gests, if it does not prove, that the time 

 that has elapsed since the derivation of the 

 respective faunas is great. Equally inevi- 

 table appears to be the fact that the cypri- 

 noid element has been derived from an 

 Asiatic source, and the slight differentia- 

 tions indicate that the introduction of that 

 element has been comparatively recent. 



If we now examine the piscine fauna of 

 Madagascar we find that one of the most 

 characteristic African genera (^Tilapia) is 

 deiveloped in that island and that there is 

 nothing in that fauna to contradict the evi- 

 dence of that genus — that it has been de- 

 rived from Africa or the same main source 

 as the African species. 



These views are identical with those pro- 

 mulgated a quarter century ago (in 1875) 

 in the Annals and Magazine of Natural History, 

 and then Africa and South America were 

 associated together with Australia in a hem- 

 isphere called EoG^A. contrasting with an- 

 ther named C^nog^a, comprising North 

 America, Eurasia and India. The accum- 

 ulating testimony of the succeeding years 

 has added to the cogency of the argu- 

 ment. 



If we now look at the mammalian fauna 

 with the light thus reflected, we may appre- 

 ciate facts of an analogous nature, but more 

 obscured or complicated by recent inter- 

 changes of faunal constituents. 



On the one hand are numerous and con- 

 spicuous mammalian types congeneric or 

 closely related to Eurasian or Indian forms. 



On the other hand are many smaller and 

 less obtrusive mammals peculiar to the 

 continent and without any near rela- 

 tives elsewhere in the present geological 

 epoch . 



It has been wisely said that " the final 

 test of a scheme of zoological distribution 

 must be the paleontological test." But the 

 paleontology of Africa has not j'et yielded 

 the test. The evidence of paleontology, so 

 far as it goes, points to the origin or devel- 

 opment of most of the conspicuous animals 

 of Africa elsewhere than on the continent. 

 It is true that Africa has been declared to 

 be especially the ' center of adaptive radia- 

 tion during the Tertiary period ' of the 

 Proboscideans, as well, as of the Hyracoi- 

 deans. The evidence, for this claim, how- 

 ever, is only negative. At least, so far as 

 the printed record goes, no early remains of 

 Proboscideans or Hyracoideaus have been 

 found in Africa, and their former existence 

 there apparently has been assumed because 

 their remains have not been found in bet- 

 ter explored lands. The assumption may 

 be right, but it must not be forgotten that 

 it is a pure assumption. Madagascar, how- 

 ever, can not be assumed with strict pro- 

 priety to be the ' chief centre of adaptive 

 radiation ' of all Lemuroideans, inasmuch 

 as that order was formerly widespread, and 

 the great island is rather the last strong- 

 hold of the restricted group. 



If, as Professor Osborn well urges, it " is 

 our problem to connect living distribution with 

 distribution in past time and to propose a sys- 

 tem ivhich ivill be in harmony ivith both sets of 

 facts," with the facts of distribution of the 

 fishes and even that of mammals in view, 

 the association of the so-called Arctogsean 

 realms is illogical and falsifies the record. 

 Whatever facts a classification may be in- 

 tended to embody, the African fauna must 

 be isolated. If we wish to express, in our 

 terminologj', a former condition of affairs, 

 Eogsea is a term adapted to do so. 



