RODENTIA-SACCOMYINAE-PEROGNATHUS. 
417 
appearing between the condyloid process and the lower angle, makes its appearance quite 
prominently outside of the bottom ’of the notch separating the condyloid and coronoid 
processes. 
The molars are all rooted throughout life. Each one, as far as can be judged, has four 
roots ; at least this is the case with the second and third. The molars are slightly convergent 
anteriorly, their inner lines being straight. In the unworn teeth the crowns are uniformly 
studded with distinct regular tubercles. In the upper jaw the first molar has four such 
tubercles, one anterior, one posterior, and one on each side. The second and third molars have 
two series of tubercles, exactly transverse to the crown, each series consisting of three tubercles 
of nearly equal size, separated by a transverse valley. A similar arrangement may be seen in 
the last molar, but in consequence of its diminished size the tubercles become indistinct. The 
same conditions prevail in the lower jaw, except that the four tubercles of the anterior molar 
are arranged two anterior and two posterior, and the outer tubercles of the rest are rather low ; 
indeed, not more than four can distinctly be discerned in the posterior molar. The last molars 
of both jaws are less than the second and third, which are nearly equal. The first molar is 
larger than the fourth ; in the upper jaw it is nearly equal to the second ; in the lower it is 
decidedly less. 
The worn teeth, however, present entirely a different appearance, resembling more closely 
those of Dipodomys. The tubercles are ground off, leaving two parallel narrow islands of 
dentine enclosed by enamel, the centre of the tooth being thus traversed transversely by 
enamel, which is obscurely divided into two lines by a faint channel, the bottom of the trans¬ 
verse valley between the tubercles already mentioned. Still further wearing would remove this 
partition, throwing the whole dentine into a single island ; of this condition, however, I have 
not seen specimens. The deciduous premolars do not differ materially. 
In the following table I have endeavored to show the points in which Perognathus and Dipo¬ 
domys differ from each other, as well as from most other genera of Kodents. 
1. Height of skull... 
2. Width of skull. 
3. Parietal bone... 
4. Interparietal... 
5. Occipital bone___ 
6. Occiput..... 
7. Pit between last lower molar and 
coronoid process... 
8. Molars_ 
9. Tail. 
10. Under surface of soles to the toes. 
Dipodomys. 
One-half the width_ 
Two-thirds the length_ 
Subtriangular. . J.....— 
Small; elongated longitudinally. 
Not entering into the plane of the occiput. 
Deeply emarginated behind_ 
Present_ 
Rootless; transverse ridges continuous_ 
Very long---- 
Densely hairy__ 
Perognathus . 
Two-thirds the width. 
One-half the length. 
Subpentangular. 
Larger; elongated transversely. 
Forming part of the plane of occiput. 
Nearly plane. 
Absent. 
Rooted; transverse ridges tuberculated. 
Moderate. 
Naked or sparsely hairy. 
There appear to be two quite distinct groups in this genus, one having the original 
P.fasciatus for its type, the other probably including the Cricetodipus parvus of Peale. They 
may be characterized as follows : 
Perognathus .—Ears rather large; a distinct, sub-orbicular lobe to the antitragus. Soles 
naked from the heel. Tail quite densely hairy, sometimes tufted at the tip. 
53 L 
