616 The Phylogeny of the Camelide. [July, 
inwards of the second row on the first, in both the fore and hind 
feet. This rotation has resulted sooner or later in the loss of the 
internal digit (thumb and great toe) from both extremities. In 
the history of this sliding inwards of the second row, the out- 
side element of the row has always preceded in time the inside 
element. The Amblypoda (Fig. 5) show this clearly. The unci- 
form bone has extended inwards so as to support the second bone 
of the first row (lunar) in part as well as the one which properly 
rests on it (cuneiform). But the magnum has not slipped inwards so 
3 as to support the scaphoid of 
the first row. That continues 
to be supported by its proper 
successors below, the trape- 
zoides and the trapezium, the 
latter taking half the burthen. 
This structure (Fig. 5) is ab- 
solutely intermediate between 
that of the Taxeopoda (Fig. 
4), and that of the Diplarthra 
(Fig. 6), and I imagine that 
all ungulates in passing from 
the taxeopodous to the diplar- 
throus stages traversed the 
conceivable path would have 
been through a type in whi 
ee ee at se rng, below. te or 
Wyo a pienis ha of Wind river, the unciform did not pe ae 
bg ad ind ag Report U. S. wards beyond the limits © 
cuneiform. No such type has 
been found. On the other hand, I have shown that the Oreodon- 
tide! have pushed the transposition of the bones of the sec 
carpal row to such an extreme that the magnum has gotten €n- 
tirely under the scaphoid, while the unciform supports the lunar 
completely. Thus the alternating position with its useful mechan- 
ical consequence has been lost to this group, the effect a 
oe being exactly that seen in the Amblypoda. This may have ha 
| Something to do with the extinction of the Oreodontidæ | 
- * Proceedings Amer. Philos. Soc. 1884, pp- 504-9, and 1884, P. 23 (Palæontolog: 
Bull., No. 39). 
amblypodous. The only other 
