The inner surface of the culs-de-sac in the Opossum is smooth, but in the lower part of the single cavity in the Kangaroo and Potoroo it presents a reticulate structure. The lining membrane in the lateral canals in all the genera is disposed in regular longitudinal folds, a dis position which characterizes the true vagina in most of the ordinary quadrupeds. In the Kangaroo, as in the other Marsupialia, the lateral canals communicate with the common or urethro-sexual cavity without making a pro jection; but at the distance of three-fourths of an inch from their termination there is a sudden con traction, with a small valvular projection in each (fig. 138). By those who consider the cu l-de sac and lateral canals as a modification of the corpus uteri, these projections will probably be regarded as severally representing an os tincw; but as they do not exist in the Opossums and Petaurists, in which there is simply a contrac tion of the vaginal canals at the corresponding part, and as in both these and the Kan garoo, the true uteri open in the characteristic valvular manner, as in the Rodentia, without the slightest appearance of a gradual blending with the median cul-de-sac, the valvular struc ture in the lateral canals cannot be regarded as invalidating the view here adopted of the vaginal nature of the median cul-de-sac, which is supported both by the general tex ture and connexions of the part in question, as well as by what is now ascertained to be its limited function. Moreover, in the large single vagina of some of the Rodentia, as the Hare, Rabbit, and Paca, there are two corresponding valvular folds of membrane near its commence ment, a little way above the urethral aperture.
In endeavouring to trace the purposes an swered by the different forms of the female marsupial organs above described, considerable difficulty arises from the want of the necessary evidence which would be afforded by the ex amination of the pregnant uterus in each genus, and by the absence of information as to their respective periods of gestation, and the powers of the new-born foetus. As far, however, as a conclusion can be drawn from the relative pe riods of gestation in the Kangaroo and Opos sum, the proportionate capacities of the vaginae to the uteri would appear to be greater as gesta tion is shorter; the vaginae being thus calculated to present fewer obstacles to the escape of the fcetus in proportion to the duration of its uterine existence ; and, consequently, a less capacious and complete external pouch is requisite for its ultimate perfection. From Rengger's descrip tion of the connexion of the foetal Opossum to the uterus, it might be concluded that the gene ration in that animal approximated to the true viviparous mode more nearly than it does in the Kangaroo; but the determination of this inte resting question will require a more exact inves tigation into the nature of the fcetal vessels and membranes in the genus Didelphys. The im
pregnated uteri of the smaller pouchless Opos sums of South America would be objects of peculiar interest and value in the present state of the inquiry.
With respect to the variations of structure in the marsupial female organs, it may also be remarked, that though they are apparently most complicated in the Kangaroos and Pha hangers, yet in reality they deviate from the type of the normal Mainmalia in a minor de gree in these Marsupialia than in the Didel phides and Petauri. For, the essential differ ence being a division of the vagina into two canals, we find this bipartition to be most com plete in the multiparous genera, while in the Kangaroos the division is only partial, and the complexity arises from augmented capacity and extent. It is to be observed, however, that the bipartition of the vaginal canal in the Kanga roos is not continued from the uterus into the va gina, leaving its distal extremity single, but com mences at the urethro-sexual cavity, and is arrested near the uteri, the orifices of which thus open into a common canal.
The situation of the rudimentary vaginal septum or hymen in the unimpregnated female organs of the placental llfammalia before men tioned, corresponds with this formation in the Kangaroo ; and in a case where this septum was preternaturally developed in the human subject, it was found to obey the same law of formation, and at the same time to have been coincident with a completely divided uterus' It is not unusual to find the vagime of the Kangaroo distended with a gelatino-mucous adhesive secretion containing hard irregularly shaped fibrous masses. One of these bodies, which was found in the mesial cul-de-sac of a Kangaroo, was described and figured by Sir Everard Hornet as the vertebral column and occipital bone of a foetus ; and his first theory of marsupial generation appears to have been much influenced by this belief. Professor Leuekart,t who found similar bodies in the vaginal tubes of a Kangaroo, compares them to a mola, or false conception, but observes that there was nothing in their structure that would permit him to form a conclusion that they were parts of a foetus.
In the Wombat the lining membrane of the vaginal culs-de-sac is greatly increased by innumerable irregular rug and papilla!, the urethro-sexual canal is lined by a thick epi thelium, and its surface is broken into count less oblique rugte and coarse papillae; the surface immediately surrounding the urethral orifice, which in this as in other Marsupials is close to the vaginal orifices, is comparatively smooth.
The clitoris is situated in a preputial recess near the outlet of the uro-genital passage : it is simple in those marsupials that have a simple glans penis, but is bifid in those which have the glans divided : and in the Opossum each division of the glans clitoridis is grooved.