Decembes 14, 1906.]' 



SCIENCE. 



787 



and their probable presence in earlier forma- 

 tions in this country, is therefore of interest 

 in paleogeography, as it further weakens the 

 evidence for the former connection of South 

 Africa with the other southern land masses, 

 by subtracting this family from the commoh 

 faunal elements peculiar to the two southern 

 continents. 



The specimen which enables us to positively 

 identify Chrysochlorid moles from this coun- 

 try was found by Mr. Albert Thomson, of the 

 American Museum Expedition of 1906, in the 

 Arickaree formation (Rosebud beds) south of 

 White River, South Dakota. It consists of a 

 humerus, complete and well preserved but 

 without any other parts of the skeleton. The 

 humerus of Chrysochloris is, however, so pe- 

 culiar and characteristic in form, as described 

 by Dobson (Monograph of the Insectivora) 

 and shown in the figures and specimens with 

 which comparison has been made, that there 

 can be no doubt that the fossil specimen be- 

 longs to the family, although somewhat less 

 specialized than the modern genus. Dobson's 

 detailed description (p. 116) of the humerus 

 of the modern Chrysochloris applies word for 

 word to the fossil; but his figure and those in 

 de Blainville's ' Osteographie,' as well as the 

 actual skeleton, show a less degree of special- 

 ization in several parts. The associated fos- 

 sils make its age equally certain. Only a 

 small part of the collection has been examined 

 in the museum as yet, but this is amply suffi- 

 cient to fix the fauna as intermediate between 

 the John Day (Upper Oligocene) and the 

 Deep River (Middle Miocene), and of nearly 

 the same age as the magnificent fossil fauna 

 recently obtained by the Carnegie Museum at 

 the Agate Springs Quarry in Nebraska. 



I have for some time suspected that the 

 skull described by Mr. Douglass in 1906 as 

 Xenotherium from the Lower Oligocene of 

 Montana, belonged in or near the Chryso- 

 chloridse, which it resembles in a much more 

 significant manner than it does the Mono- 

 tremes to which it was provisionally referred 

 by the describer. The proof that Chryso- 

 chloridse did inhabit North America in the 

 Middle Tertiary makes it reasonable to refer 

 Xenotherium definitely to the same family. 



Apternodus Matthew, from the same forma- 

 tion and region as Xenotherium, is probably 

 the lower jaw of that genus or some closely 

 related form. It is possible also that one or 

 more of the Insectivora described by Marsh 

 in 1872 from the Bridger formation (Middle 

 Eocene) may prove to be ancestral types of 

 Chrysochlorida). 



The distribution of this rare and interesting 

 family of Insectivora as now known is : 



Modern — South Africa. 



Upper Miocene — South America (Pata- 

 gonia). 



Lower Miocene — North America (South 

 Dakota). 



Lower Oligocene — North America (Mon- 

 tana). 



(?) Middle Eocene — North America (Wy- 

 oming). 



Insectivora are exceedingly rare as fossils, 

 and this is no doubt but a small fraction of 

 the real distribution of the family during the 

 Tertiary. We can not regard the South 

 American representative in the Upper Miocene 

 as descended from the North American spe- 

 cies of the Middle Tertiary, for South Amer- 

 ica, during the Middle Tertiary at least, was 

 an insular continent, and its mammal faunse 

 from the early Eocene until the beginning of 

 the Pliocene, contain no elements of northern 

 origin, but develop on entirely independent 

 lines of evolution. It would appear rather 

 that the North and South American chryso- 

 chlorids are descended from a common pre- 

 Tertiary ancestor. The modern South African 

 form, on the other hand, may be more nearly 

 related to the North American genera if we 

 suppose that the middle or early Tertiary 

 range of the family extended to Europe and 

 Asia, whence it might readily have reached its 

 present home. All authorities are agreed that 

 Asia and North America were united during 

 most of the Tertiary, and Africa was united 

 to the northern land in the Oligocene and 

 subsequently. Hence there are no geographic 

 difficulties in the way of this supposed wider 

 distribution — nor adequate evidence to take it 

 out of the region of conjecture. In fact, until 

 the mutual relationships of the Chrysochloridse 



