E. LÖNNBERG, MAMMALOGY OF ECUADOR. 91 



that Caenolestes and its allies should be reckoned as more 

 related to the Polyprotodonts tlian to the Diprotodonts.» 



It appears, however, that this question could be discussed 

 from a somewhat different point of view. Caenolestes is un- 

 doubtedly a very primitive form, and it is therefore less 

 suitable \o compare it with the recent Diprotodonts, with 

 which nobody bas wanted to unite it. The now living 

 Diprotodonts are much more specialised than that bvmg 

 rehet Caenolestes, and they are also generally more specia- 

 lised than the Polyprotodonts. But it is not correct to con- 

 sider all primitive characteristics, which Caenolestes and the 

 Polyprotodonts may have in common, as proofs for a closer 

 relationship between them than between Caenolestes and the 

 ancestors of the Diprotodonts. 



Miss Pauline H. Dedener bas concentrated^ what she 

 thinks speaks for the Polyprotodont affinity of Caenolestes 

 in 10 paragraphs. In the first of these she says that the 

 dental formula of Caenolestes is *like that of the Dasyurid 

 genera». This is true with regard to the number of incisors 

 and the size of the oanines, which are simply primitive cha- 

 racteristics, but quality is of more importance than number. 

 It must be remembered that with regard to the mcisors 

 even those of the upper jaw Caenolestes has been subjected 

 to a distinct specialisation in Diprotodont direction (ef. Ma- 

 cropus), while those of Phascologale retain a more primitive 

 type In the second paragraph the resemblance to Phasco- 

 logale in external form is spöken of. I do not think, that this 

 is of any importance, when there is no specialisation expressed 

 in the external shape. It is also like a big shrew, or a rat 

 without that this involves any affinity, but it bears some 

 resemblance to the diprotodont Dromicia as well. In para- 

 graph 3 is spöken about the »resemblance to Dasyurus skull 

 in general shape», which is of little importance, and with 

 reaard to the slender pterygoid processes and general form 

 of' alisphenoid bullae. About the value of these two latter 

 characteristics I am not prepared to express any definite 

 opinion, but in »general form» the bulla of Caenolestes does 

 not appear to resemble Dasyurus more than it does f. i. 

 Dromicia. Paragraph 4 points out the resemblance ^to 

 AntecUnom ys and Sminthopsis skulls in size, shape and deii- 



1 The American Naturalist, 1909, p. GU— 618. 



