MESSES. A. AND E. NEWTON ON THE OSTEOLOGY OE THE SOLITAIEE. 337 
pubis in Didus by Professor Owen’s artist (op. cit. pi. xv.) is probably altogether erro- 
neous ; but if it be so the error is, under the circumstances, most excusable, for no one 
could have predicated the extraordinary forms assumed by these bones in Pezophaps. 
§ 5. Sternum. 
Of this bone the collection contains fragments belonging to at least six individuals, 
but only one that is at all complete, though some portions of the others furnish infor- 
mation as to certain not unimportant details. The most perfect specimen (Plate XVIII. 
figs. 71-74) also has received much damage on its sides and posterior extremity, so that 
their contour can only be inferred — a matter perhaps of the less consequence since the 
characters afforded by the posterior extremity of the sternum are certainly not so constant 
in birds generally as those to be deduced from the anterior end'*. This same specimen 
is also one of the smallest (female V), and on that account it is possible that some of the 
most remarkable characters of Pezophaps may not be very strongly developed in it. 
* In support of this opinion the following cases may be cited among many others that could be mentioned. 
In the Limicolce (Nitzsch) (= Charadriidce and Scolopacidce, Atjctt.), an eminently natural group, a very great 
diversity of conformation of the posterior margin of the sternum exists even among forms 'which are, both in 
general habits and outward structure, very closely allied. Totanus ochropus differs from all other Totanince 
known to us by possessing only one posterior emargination on each side the keel, where they (and even T. glareola, 
so long confounded with T. ochropus) have two. Machetes has a single emargination. Charadrius morinellus 
is unlike any other Limicoline bird we know, inasmuch as both “ ectolateral” and “ entolateral” processes extend 
backward nearly as far as the keel does, and the former are somewhat everted as in Phalaropus. But (Edicnemus 
and Vanellus have the extremity of the inner emarginations bridged across so as to become fenestrae. C.pluvialis 
and C. hiaticula, with many of the smaller Plovers, have two emarginations like most Totanince. Scolopax 
gallinula resembles these last and the majority of Tringince, while 8. gallinago, 8. wilsoni , and 8. bernieri have 
only one large, and 8. major, like 8. americana and 8. rusticola, one small emargination on each side. As great 
differences are to he found in other Limicolce ; hut here we have only mentioned cases of very closely- allied 
species ; for in aberrant forms like (Edicnemus and Machetes some diversity is not much to he wondered at. 
Examples might also he taken from the Procellariiclce ; hut, to come nearer our present subject, among the 
Columbce we have Diduncxdus with a single pair of broad lateral processes, and the space between the emargi- 
nations without any fenestration (Plate XXIII. fig. 170). In the more normal Columbce the structure is much 
the same (Plate XXIII. fig. 169), except that the lateral processes are much narrower, but a monstrous example 
of C. livia (Plate XXIII. fig. 172) shows two pairs. In a specimen of Patagicenas caribbcea in our possession, 
where the middle of the hinder end of the sternum is very broad, there is a small circular fenestra on the left 
side, but none on the right (Plate XXIII. fig. 173). In Goura and Leucosarcia (Plate XXIII. figs. 176, 171) 
there are two pairs of lateral processes, and in the former (Goura) the outer pair is wide and its front edge 
distinctly turned forwards at its upper extremity. In the latter ( Leucosarcia ) the anterior pair of processes is 
tapering. In Geopelia (Plate XXIII. fig. 175) and for a long time in Chamcepelia there are two pairs of lateral 
processes, but in the latter (Plate XXIII. fig. 174) the inner pair eventually become connected at their extre- 
mities, and thus fenestrae are formed. 
A singular modification of the posterior margin of the sternum seems, in some birds, to be connected with 
diving habits. In this respect we have Cinches differing from nearly all the Passer es, and Panclion from all the 
Falconidce ; while an extraordinary development of the representative of the ensiform cartilage is observable in 
the Ducks of the group Clangula and the Mergince when compared with other Anatidce, and these birds, as is 
well known, are among the most excellent of divers. 
MDCCCLXIX. 2 Z 
