420 MR. R. I. POCOCK ON THE EXTERNAL 



elongated area of naked skin, from the lower half of which the 

 clitoris pi-ojects. The clitoris is a closed tube perforated by the 

 urethra. When cut open in an unmated female the orifice of 

 the vagina may be seen to lie a little below the anus, and a little 

 lower down is the urinary orifice. (Text-fig. 28, E.) 



The following table shows the vai'iation in position of the 

 apertures of the anal glands, and other points connected with 

 them, in the genera in which I have detected them : — 



a. Apertures or aperture of anal glands concealed witliin the 

 sphincter of the anus. 

 b. A pair of moderately widely separated apertures above the anal 



orifice DoIicJiotis. 



h'. Apertures not above the anus. 



c. Aperture on each side of the anus Dasijproeta, Coslogenys. 



c'. Aperture or apertures beneath the anus. 



d. Two apertures, one on each side of middle line ... Systrix, Lagosfomus. 

 d' . A single median aperture. 



e. Anal gland small, incapable of evagination Octodon, Capromys, 



e'. Anal gland large, capable of evag-ination and provided with 



four pairs of papillag Ilyocastor. 



a' . Apertures of anal glands forming a pair of elongated slits outside 

 the sphincter of the anus and extending down the perineal area. 



Cavia, Kerodon, HydrocJioerus. 



Tlie intei^esting point to note in connection with this table 

 is the diflference between Dolichotis and Cavia -\- Hydrochoerus 

 with respect to the glands. 



According to Tullberg's descriptions, Thrynomys falls by this 

 table with Cavia and Hydrochoerus under a and Ctenodactylus 

 with Basyprocta and Coelogenys under c. 



The Tail. 



The tail supplies useful systematic characters. If, as is 

 probable, a long cylindrical tapering tail covered with scales and 

 short hairs is the primitive type, that type is prevalent in the 

 OctodontidsB, occurring in many of the mouse-like genera. 

 Within the fainily, however, variations in the length of the tail, 

 in adaptation to habit, and in growth of the hairs ai'e numerous. 

 In Octodon, for example, although the organ is longish, the hairs 

 in the distal portion are developed to form an elongated tuft. 

 In Ctenomys it is shorter, but thicker and somewhat compressed 

 with a very distinct crest of short hairs extending along the 

 posterior two-thirds of the upper edge. 



In Thrynomys the tail is of the piimitive type, but compara- 

 tively short, and the same applies to Capromys, but in this 

 scansorial genus the hairs on the underside are stiflfened to aid 

 in climbing, a variation foreshadowing that which is seen in a 

 much more marked degree in Coendu and Chcetomys described 

 below. In one species (C. prehensilis) the tip of the organ 

 is said to be prehensile, but I am not aware whether the tip 

 is curled downwards over a branch or upwai'ds as in the so- 

 called arboreal Porcupines. In Myocastor the tail is moderately 

 long, cylindrical, and rather stout, but shows no modifications 



