MOEPHOLOGT OF THE GALLINACE^. 227 



extinct Ceratosanms and Allosaurus (Marsh, Amer. Jouvn. Sci. vol. xxvii. 1884, 

 plate 11), and in the recent Crocodile (Huxley, Proc. Geol. Soc. 1S69, p. 27, fig. 3). 



At present in this, my first, stage, the ilium has developed two broad lobes ; it is no 

 longer a wide fan-shaped plate, but grows forwards and backwards almost equally, and 

 embraces at least a dozen vertebrre out of the fifteen which are covered by it perma- 

 nently. The whole hip-plate at this time is a leafy growth of solid hyaline cartilage, 

 apparently segmented into ilium, ischium, and pubis, as in its ossified condition after- 

 wards, before the sutures arc filled in (see Plate XXV. fig. 12, il., isc.,pb.). Under a low 

 power, the segmentation seems to be perfect ; under a high power, it is seen that on the 

 inner face a thin part about one cell deep is continuous ; but externally there are deep 

 chinks, prefiguring the sutures of its bony condition ; these chinks are filled with 

 nascent fibrous tissue. Thus it would appear that the primordial hip-girdle of the 

 Sauropsida was composed of three distinct diverging cartilages, the proximal elements of 

 the ichthyopterygium ; and that the modern ossification (Plate XXV. figs. 4 and 12) ia 

 both Sauropsida and Mammalia is historical or atavistic. The dilated top of the ilium, 

 at first fan-shaped, has soon grown, fore and aft, into two subequal lobes, the pre-ilium 

 and the post-ilium ; the fore lobe is concave externally, and the hind lobe is convex. 

 The upper and broader bar that runs, now, parallel with the post-ilium is the ischium ; it 

 is still separated by a long notch from the plate above, as in the Apteryx and also in 

 the Tinamou (T. Z. S. vol. v. plate 39, il., isc). At this stage the pelvic plate of the Chick 

 is quite Tinamine. The narrow lower bar, the pubis, is much longer than the ischium; 

 it is quite free from it, and the obturator nerve emerges in its fore end. Then these normal 

 parts of the pelvic plate all meet at the edges of the acetabular fenestra (Plate XXII. 

 and Plate XXV. fig. 3, ac). The ischium is forked ; the lower fork binds under on the 

 dilated and hooked fore end of the pubis, which hooked lobe fits in between the ischium 

 and ilium ; the ascending fork of the ischium ties itself to the lower edge of the fore part of 

 the post-ilium, and thus separates the acetabulum from the sacro-ischiatic space [s.i.f.) — 

 idtiraately a " fenestra." At the lower, convex edge of the pubis there is a smallish 

 triangidar projection, which looks forwards and downwards, and has most of its root in 

 the ilium ; it is somewhat later in growth than the rest. This secondary process is as 

 large as in the adult Tinamou {op. cit. plate 41. fig. 3) ; but less than in the Apteryx 

 (Mivart, Trans. Zool. Soc. vol. x. 1879, p. 38, fig. 41, l.p.). 



As to the nature of this process, which I boldly call " secondary," there is still a strong 

 controversy. This has arisen from two causes : first, it is very large in the Dinosaurs, 

 and is in them ossified by the rotated pubis (see Marsh, Amer. Jovirn. Sci. vol. xviii. Dec. 

 1879, pp. 501-503, plate 3) ; and the second difficulty arises from the fact that its small 

 counterpart in the Fowl, and in svich Carinatse as possess it, is ossified by the ilium. 

 Happily there is an intermediate condition, in which it is ossified by both these bones : 

 this is evident from Prof. Mivart's figure of a young Ostrich's pelvis (Trans. Zool. Soc. 

 vol. viii. p. 437, fig. 74, il.p.). This figure is imperfect, because the unossified cartilage 

 is not drawn ; when present, it projected equally from the ilium and the pubis. The 

 importance of this fact, in an attempt to reconcile the discrepancy between the Dinosaur 

 and the Bird, was first pointed out to me by Mr. Hulke, in a letter written to me on 



