978 Mil. It. I. rococK ox the [June li, 



seemingly primitive nature of this glitnd in Pudu, I infer that 

 this was tlie view he hehl. If this be so, the shiillow or short 

 pocket seen in such genera, for example, as Alee, Redunca, and 

 rsendois preceded, in an evolutionary sense, the well-develo])ed 

 gland p()s.sessetl respectively by such genera as Capreolus, Felea, 

 and Ovin. Further, it will follow as a jJO.ssibility, perhaps as a 

 probability, that the pedal glands have been developed inde- 

 pendently more than once in various groups of liovidaj and 

 Cervida' ; and that the genera within these families characteri.sed 

 by pa.sterns without any iiiterdigital space have an earlier form of 

 foot than those in which such a cleft is found : that tlie hind foot 

 of Cert'KS, for instance, to take a concrete case, is more primitive 

 than that of JJama. 



I believe this view to be erroneous ; for everything we know of the 

 evolution of the feet in lluminantia points to the conclusion that 

 their structural peculiarities are due to descent from four-toed feet 

 with the phalanges separated to a greater or less extent. Gradually 

 the lateral digits became shortened, raised from the ground, and 

 practically f unctionless so far as aiibrding suppoi't to the body was 

 concerned. The weight was borne by the two middle toes, which 

 became gradually strengthened for the purpose by the extension 

 of the integument between them. Now when Mammalisiof other 

 groups develop an interdigital integument or " web," this web 

 always, I believe, extends ^dong the po.sterior or inferior aspect of 

 the toes, leaving more or less of a cleft between them in front or 

 above. This is well shown in such forms as Luira and the Canidte 

 and in various amphil)ious genera of Rodents and of other ordei's. 

 For these i-easons 1 believe that the earliest type of Ruminant 

 foot is that in which the interdigital web consists of a fold of 

 integument extending along the back of the pastern only, leaving 

 a wide and deep depression between the bones above the hoofs in 

 fi'ont. ]iut when two surfaces of skin are opposed to form a 

 cavity or crease, conditions favouring the activity of the skin- 

 glands, or at all events conditions jjreventing the evaporation of 

 integiunental secretions, are set up. Hence, in a foot with a 

 posterior web holding the digits together behind and a cleft 

 between them in front, we have conditions conducive to the 

 development of a scent-gland ; and 1 conceive that it was from a 

 foot of this structure, containing a potential peihxl gland, that the 

 various types of feet of the Ruminantia, whether containing 

 glands or not, have been derived, and that the glandular pockets, 

 whether large or small, are the renuiants, specialised for the 

 secretion of scented material, of a primitive anterior interdigital 

 cleft. 



This conception seems to me to suj)j)ly the most satisfactory 

 explanation of the prevalence and diversity of the i)edal glands of 

 the lluminantia, all the glands being thus traceable to a common 

 origin. 



Of existing forms of this grou]) tlii' most primitive type of foot 

 is found in the Tiagulina, itself a [)rimilive group combining 



