MATTHEW, SUPPLEMENTARY NOTE 15 



•or the Hippopotamus (if late), but hardly both. It would be irrecon- 

 cilable with the presence and diversity of the Lemurs and Centetids, the 

 absence of monkeys and of the rest of the later Tertiary mammal fauna 

 ■of Africa. 



10) A land connection throughout the Tertiary would account for the 

 presence of the few mammal types that have reached the island, but not 

 for the complete absence of the very numerous groups of several succes- 

 sive faunas which are not present. The amount and character of the 

 adaptive radiation among the half dozen stocks that are present indicates 

 the absence of monkeys, ungulates, cats etc. in the past as well as in the 

 present, as also that the half dozen different elements of the mammalian 

 fauna were due to invasion by single types at several different epochs. 

 The indicated epoch corresponds in each case to the time at which the 

 geologic record indicates that their ancestral stocks were prevalent on 

 the adjacent mainland. 



11) Madagascar, it should be remembered, is an island of almost con- 

 tinental size and great geologic antiquity. It is not a transitory islet, 

 such as Barbour cites in support of his contention that mammals behave 

 .strangely on oceanic islands. If a continental fauna once gained access 

 to it there Avould be no reason for its extinction, nor opportunity for the 

 expansive radiation of a few stocks. If Doctor Allen's explanation were 

 ■correct, the relations of the fauna to the Ethiopian should be similar to 

 those of Borneo or Sumatra to the Oriental mainland fauna, save for a 

 wider amount of differentiation from the late Tertiary and Pleistocene 

 invaders of the Ethiopian region. The briefest summary of its character, 

 such as is given above, is enough to show how widely it differs from that 

 type. 



