402 PROFESSOR PETERS ON THE GENUS PECTINATOR. 
those of Hchinomys. The zygomatic process of the maxillary is much more like that 
of Lagotis than that of Dipus or Scirtetes, its upper root being very narrow and not 
enlarged, and the lower having its base horizontal and not vertical. The malar bone, 
although very low, is provided with a very short inferior angle, corresponding with the 
characteristic lower zygomatic process of Echinomys; and its maxillary process is, not- 
withstanding its posterior hamulus, much more like that of Zagotis than that of Dipus'. 
The posteriorly rounded frontals and pointed externo-anterior angles of the parietals are 
shaped exactly as in Habrocoma. ‘The sharp supraorbital margin reminds one of Pedetes, 
but also of Echinomys. The underside of the snout and the large foramina incisiva are, 
carefully compared, more like those of Habrocoma and the Chinchillina; the palate and 
the basis cranii, on the contrary, show a greater affinity to Echinomys; and the meatus 
auditorius externus, differing from that of the Chinchillina and the Dipodina, is in direc- 
tion and form more similar to that of Habrocoma and Echinomys. The cranial cavity is 
very like that of Habrocoma, and differs, therefore, very much from that of the Dipodina. 
The great extent of the interparietal bone is a peculiarity of the Ctenodactyli, when 
compared with the skulls of the related genera. 
The lower jaw differs entirely from that of the Dipodina, and agrees best with that of 
the Chinchillina, except that no distinct coronoid process is to be seen. As in Chinchilla, 
the pointed angular portion of the lower jaw is, on account of the shortness of the roots 
of the incisors, less distinctly placed on its outer side than in Hchinomys and the Octo- 
dontes. Petromys, on the contrary, has the coronoid process and the position of the 
angle of the same form as Habrocoma, and ought, therefore, probably not to be united 
with the Ctenodactyli, but with the Octodontes. 
The spinal column is (in two skeletons) composed of 7 cervical, 12 thoracic, 7 lumbar, 
4 sacral, and 19 caudal vertebre. 
The cervical part is similarly developed to that of Echinomys, Habrocoma, and Chin- 
chilla. It is therefore very different from the broad, very short, or even coalescent?, 
corresponding portion of Pedetes and Dipus, and is alone sufficient to prove that the 
Ctenodactyli cannot be naturally placed with the Dipodina, as has been proposed. The 
atlas and axis, and also the other cervical vertebrae, are more like those of Habrocoma 
than of Chinchilla or Echinomys, although the atlas has, as in Chinchilla, its outer sides 
less concave, and a short ventral spine. The second, third, and fourth vertebrae are 
provided with a ventral backwards-projecting keel. 
The thoracic vertebre are also totally different from those of the Dipodina, and in their 
gradually increasing spines more similar to those of Habrocoma, although in number 
more approaching those of Echinomys. They are, of course, in form also similar to 
those of Echinomys (eayennensis) and Chinchilla; but these genera have the spines much 
" Only Dipus (cegyptius), and not Scirtetes (spiculum), is provided with an orbital process of the malar some- 
what resembling that of the Ctenodactyli. 
° Dipus (egyptius) has four cervical vertebr (the second to the fifth inclusive) coalesced. 
