The measurements available indicate that in a pedigree 36-inch Sheltie of the “forest” type, the front cannon-bones will probably measure 137 mm. by 26·2 mm., that the circumference below the knee will be 5·5 inches, and the distance from the elbow to the ground 21·5 inches—in Eric (36·5 inches) the length from elbow to the ground was 22·25 inches. If these figures are approximately correct, it follows that in a 36-inch Sheltie of the cart-horse type the limbs may be at least an inch shorter than in a dwarfed 36-inch pony of the riding type, and that the dwarfing may be unaccompanied by any loss of “bone.” This conclusion is supported by the measurements of Odin, a 38-inch pony, 6 inches below the knee; of Vulcan, a 32-inch pony, 5 inches below the knee; and of other ponies of the Londonderry type belonging to the Ladies Hope, and also by those of Everlasting, Frederick, and other thick-set Auchlochan ponies. For example, in Everlasting, a 38-inch pony, the distance from the elbow to the ground is 22·75 inches, the circumference below the knee 5·5 inches, the bone being “round,” and the shank ·5 inches broader than in the flat-boned riding pony Eric. In heavy horses, but especially in Shire colts, one (or more) of the limbs has occasionally an extra digit ending in a well-formed hoof. In a three-weeks’ horse embryo there are no rudiments of limbs; at four weeks the limbs are represented by paddle-like structures; at five weeks the paddles contain rudiments of three toes—miniatures of the three toes of _Hypohippus_ (fig. 20). In ordinary circumstances the development of the outer and inner toes (ii. and iv.) is soon arrested, but occasionally one of these rudiments develops into a toe as large and complete as in three-toed Miocene horses of the “forest” type. When this happens, when in addition to the third toe there is a toe corresponding to the human forefinger, we have a marvellous instance of reversion. If in the Shetland breed there is a tendency to reversion, one would expect to find now and then extra digits in ponies of the heavy or “forest” type. I have, however, never heard of a Shetland pony with extra digits. Dwarfing of the face and the reduction of the nasal chambers has apparently been carried further in some of the miniature cart-horses than in Eric and other flat-boned Shelties. In Jupiter,[F5] _e.g._, the head, though wider across the orbits, is shorter than in Eric, and decidedly more dished. Ancient horses adapted for a forest life had face pits in front of the orbits, which probably, like the corresponding pits in deer, lodged scent glands. Further, in ancient “forest” horses the upper lip was probably decidedly longer and more prehensible than in modern breeds. In broad-browed, big-boned Shetland ponies there is no indication of a pit for a scent gland, but there is sometimes an unusually long and decidedly mobile upper lip, which may or may not be due to reversion.