In the Armadillo the astragalus, in consequence of the greater production of its anterior part, presents more of an angular than a quadrate figure; and the scaphoid articular surface, being proportionally carried forwards, is altogether separated from the anterior calcaneal surface. The posterior and inner calcaneal surface resembles that in the Scelidothere, but is less inclined upwards; and is continuous with the posterior part of the tibial articular surface. Thus the astragalus in the structure of its two most important articulations, viz. that which receives the superincumbent weight from the leg, and that which transmits it to the heel, presents a closer correspondence in the Scelidothere with that of the Dasypus, than with that of the Megathere or Mylodon. The ungueal phalanx of the Scelidothere before alluded to, is represented of the natural size in Pl. XXVII. The side view, fig. 3. shows the position of the articular surface on the proximal end, sloping obliquely towards the under surface, and overtopped by an obtuse protuberance, calculated to impede any upward retraction of the claw: the present joint, in fact, illustrates in every particular the argument by which Cuvier established the true affinities of the allied extinct genus Megalonyx.[58] The present phalanx is, however, less compressed, and less incurved than those of the Megalonyx, which have been hitherto described; but it more resembles in these proportions one of the smaller, and presumed hinder, ungueal phalanges of the Megatherium. The upper and lateral parts of the bone are rounded, and it gradually tapers to the apex, which is broken off. The osseous sheath for the claw is developed only at the under part of the bone: it presents the form of a thick flat plate of bone, with the margin very regularly and obliquely bevelled off, and having a vertical process of bone attached lengthwise to the middle of its under surface. This process must have served for the insertion of a very powerful flexor tendon. The figures of this bone preclude the necessity of any further verbal description. M. Lund lays most stress upon the argument founded on the inward inflection of the sole of the foot in the Megalonyx, and appeals with greatest confidence to this structure in support of his hypothesis of the scansorial habits of that extinct Edental.[59] It is quite true that the Quadrumana derive advantage from this position of the foot in climbing trees, and that it is carried to excess in the Sloths, which can only apply the outer edge of the foot to the ground. But we may ask, was the inversion of the sole of the foot actually carried to such an extent in the _Megalonyx_? And, admitting its existence in an inferior degree, is it then conclusive as to the scansorial habits of that species?