1871.] 



321 



[Wilder. 



HODENTIA. 





D. 



L. 







Hystrix cristata 



15 



4 



19 



Owen. 



" alopha 



14 



5 



19 



« 



Lepus timidus 



12 



7 



19 



" 



Castor fiber „ 



15 



4 



19 



« 



Fiber zibothicus 



13 



3 



16 



Flower. 



Capromys 



17 



6 



23 





IiOiicberes 



17 



8 



25 



<« 



Hydromys cbrysogaster 



14 



7 



21 



Owen. 



Myoxus 



13 



7 



20 



Cuvier. 





D. 



L. 







Dasypus peba 



10 



5 



15 



Owen. 



Bradypus 



17 



3 



20 



Flower. 



" 



15 



5 



20 



" 



" tridactylus 



16 



3 



19 



Owen. 



Manis 



13 



5 



18 



Flower. 



" pentadactyla 



13 



4 



17 



Owen. 



Myrmecophaga jubata 



15 



3 



18 



it 



Orycteropus capensis 



13 



8 



21 



" 



Megatherium 



iH 



3 



19 



a 



Cyclothurus 



15 



2 



17 



Flower. 



MARSUPIALIA. 





D. 



L. 







Most genera 



13 



6 



19 



Owen. 



Phascolomys wombat 



15 



•4 



19 



" 



" latifrons 



13 



6 



19 



Flower. 



Phascolarctos 



11 



8 



19 



« 



Petaurus macrurus 



12 



7 



19 



" 



After making due allowances for differences in the interpretation 

 of facts by different observers, the preceding tables are very sug- 

 gestive. 



1. The different groups are seen to be unlike as regards the con- 

 stancy of the vertebral formula; the adherence to 20, among the Car- 

 nivora (with but two exceptions so far as I know) is as startling as is 

 the adherence to 7 with the cervical vertebra; the number 19 is 

 equally characteristic of the Artiodactyla; while in striking contrast 

 to these two groups are the Perissodactyla and the Insectivora, which 

 certainly do not differ widely enough in their habits from the Artio- 

 dactyla and Carnivora, to give a clue to the reason for these discrep- 

 ancies. 



2. Although in most cases, the species of a genus differ only by 

 the greater or less development of the rib-process, so that the total 

 number of thoracico-lumbar vertebrae is the same, yet in some cases, 

 (Equus, Otolicnus, Loris, Sorex), this number varies by a single verte- 



PROCEEDINGS B. S. N. H.— VOL. XIV. 



21 



APRIL, 1872. 



