SHUFELDT. ] OSTEOLOGY OF THE EREMOPHILA. 641 
The sacral vertebre are invariably confluent throughout the chain in 
the pelvis of the Horned Lark; indeed, it is only by a process of staining 
this compound bone, and the aid of a strong light, that they can with 
any satisfaction be counted. There are eleven of them; exceptionally, 
twelve. 
The neural canal, circular at the outstart, shows the usual pelvic szell, 
chiefly anterior to the acetabula, conformable with the ventricular dila- 
tation of the myelon in that locality. The exit of this tube distally is 
likewise nearly circular. The foramina along the bodies of the centra, 
in the vicinity of the dilatation referred to, are double and placed one 
above another, for the separate egress of the roots of the pelvic plexus. 
The anterior aspect of the first sacral vertebra presents every element 
and process requisite for articulation with the ultimate dorsal segment. 
It is largely overshadowed by the ossa innominata. Opposite the iliac 
contraction, in the neighborhood of the fourth and fifth sacrals, these 
vertebre throw out their par- and di-apophysial processes far enough to 
meet and brace the iliac bones. Wedo not meet with such braces again 
until arriving opposite the acetabula and beyond, where the parapo- 
physes project upwards and unite with the outer margins of the trans- 
verse processes, thé ilia articulating with the free and united borders. 
Foraminal deficiencies not unusually occur among these processes, 
more particularly between the last two sacrals, where they seem to be 
constant, though of varying size and shape in different individuals. 
The last sacral vertebra is compressed from above downwards, retain- 
ing, however, all the elements required in articulation with the first and 
much-modified coceygeal vertebra. 
Viewing the confluent sacral vertebra, or the “ sacrum”, from above, 
we find the united neural spines, as a vertical lamina, dividing the ante- 
rior interiliac space into twocapacious ilio-neural grooves at that moiety 
of the bone. 
This common neural spine and the ilio-neural grooves proceed back- 
wards until the gluteal ridge of the ilium curves outward to the antitro- 
chanter on either side. At this point the spine disappears with the 
grooves, the sacrum becomes nearly flat and spreads out, to gradually 
contract again before its ultimate dilatation in the diapophyses of the 
last vertebra. 
It is a well-known fact to the ornithotomist that the pelves of birds 
differ with respect to the ilio-neural grooves in one of two ways. As 
seen in this bird they are grooves, as the ilia do not meet the united 
spines of the vertebra; they are very frequently, however, converted 
into canals in other orders, by meeting of the interested bones above. 
The condition as defined though, in the previous paragraph, as relating 
to Hremophila, seems to be characteristic of American Oscines. ‘The sa- 
crum is slightly convex from before backwards on its upper surface, 
moderately concave along the confluent centra below. 
The pelvis of this bird is uncommonly wide and short, and the ischi- 
adic and pubic posterior extremities remarkably flared outwards. The 
anterior and inner angle of each ilium, apparently assisted by the dia- 
pophysis of the first sacral vertebra from beneath, is pointed ; the ante- 
rior border slopes backwards gradually, for a distance of 3 or 4 milli- 
metres, to the rounded anterior external angle of the ilium. Between 
this point and the acetabulum the iliac border is strongly concave in- 
wards, as ‘is the surface of the bone above it, the preacetabular being 
included between this border and a well-defined gluteal ridge. The 
superior postacetabular iliac surface is nearly square in outline, convex, 
and equal to a little more than one-third of the bone. It is thin and 
41 H 
