I. Ostcology.—It can scarcely be said that there is any peculiarity of structure in the ske letons of the whole of the insectivora, in which they differ essentially from other groups ; but on the other hand, there are many in which they differ from each other, according to the very striking and obvious diversity of their ha bits. In the family Talpide, the genus 2'alpa (fig. 440) presents itself as the type. In these animals the cranium is greatly elongated, and of a tapering or conical form, a character which it partakes with the Soricide, but to a still more remarkable degree. The cra nium of the Chrlisochloris, or Cape-mole, the South African representative of this family, presents this character in the most regular form. It is, in fact, a perfect cone, short, pointed at the muzzle, broad behind, where the base of the cone is distinctly circumscribed by a crest, which passes from the root of the zygomatie arch of one side, over the vertex, to the cor responding point on the other; on the sides of the head the zygomatie arches themselves com plete the cone, passing obliquely and in a straight line from the maxillary to the temporal bone, and beneath it is completed by the in clination inwards of the symphysis, the rami, and the coracoid processes of the lower jaw. The portion posterior to the base of the cone is rounded.
In the genera 2'alpa (fig.441) and Condytura the head is equally but less regularly conical, and the snout is more elongated; the zygomatic arch is extremely slender, and rises obliquely, joining the cranium considerably above the audi tory meatus. A similar conformation is seen in the genus Scalops, which I have already mentioned as leading from the Talpide to the Soriciihr. This form is admirably suited for their subter ranean progression, as they push their way through the soil by their long moveable snout, which acts in some measure as a wedge. Amongst the Saricide, the Scalops approaches in its structure most nearly to the falpida., having the almost perfectly conical form of the head which belongs to that family ; and all the Soricide partake of it to a greater or less degree. But the cranium of the Erinaccade approaches more nearly to that of the Camivom ; and viewed from above, the sides are nearly parallel, the zygomatic arches projecting further than the posterior part of the cranium. The muzzle is shorter, more obtuse, and somewhat narrower than the cranium, which is coin pressed forwards. In Centeno the head is much more elongated and conical. The genus 7'upaia has the head nearly oval, the muzzle straight, prominent, much smaller than the cra nium, the zygomatic arches but slightly pro minent, and the circle of the orbit closed pos teriorly, a circumstance which is not found in any other of the order Insectivora. In this genus the orbit is therefore circumscribed by being closed posteriorly by the union of the post-orbitar processes of the frontal and jugal bones ; in all the other genera the orbit and the temporal fossa are confounded in one cavity, without the trace of any attempt at forming a division between them. This peculiarity in the
genus Tupaia spews a marked tendency to wards the insectivorous Quadrumana.
The bones of the face are very early united in the Talpidee and the Soricidee. In the genus Centencs there is no jugal bone. There is therefore no zygomatic arch, notwithstanding the masseter muscle is of great size. This is attached by a single tendon to a sort of tu bercle, which represents the zygomatic process of the maxiliary bone. In Tupaia the inter maxillary bones are very large, and their suture descends vertically, nearly half way be tween the nares and the orbits. The Shrews, like Ccntenes, have no 'malar bone, and the zygomatic process of the maxillary is even less conspicuous than in the former genus. Most of the subterranean forms have at the extremity of the muzzle, that is to say, around the open ing of the nares, a small rim or border, which is especially conspicuous in the Chrysochloris, for the attachment of the large and moveable cartilages of the nose, so important to these animals in turning aside the earth as they make their way through the ground, as well as in seizing the worms and insects on which they feed.
The proportions of the spine vary greatly in the different families of Insectivora, as may be anticipated from the great difference in their habits. On this subject it may not be useless to subjoin the following table of the numbers of the separate classes of vertebra, which is taken from the last edition of the Lecons d'Ana tomie Comparde of Cuvier observe a tail as long as that of a squirrel, and obviously for the same object, that of balanc ing the animal when taking leaps from one branch to another ; and in the Water-shrew the tail is lengthened to assist in swimming, for which purpose it is also fringed on each side with stiff hairs. Of the habits of the Coady tura* we know little, and of that little nothing which accounts for the considerable proportion of the tail.
Of the form of the different vertebra in the various groups of the Insectivora, little need be said, as there are few circumstances con nected with these bones which bear materially upon any physiological point. It may be ob served, however, that in the Talpithe and the Soricide the cervical vertebra form large rings, have strong transverse processes, and, except ing the second, do not possess any spinous processes. In the Erinaceatke, and particu larly in the Tenrec (Centencs) the transverse processes are particularly large, and the rings smaller than in the former families. The spi nous processes are also wanting in the dorsal vertebra of the mole.