MOEPHOLOGY OF OPISTHOCOMUS CEISTATUS. 59 



The occipital condyles are not so bilobate as in the Cracidae. The occipital plane is 

 rounded and complete, not fenestrate as in many water-birds ; in this it agrees with 

 the Fowl tribe. A comparison of Prof. Huxley's figures with those given by me of an 

 old Fowl (Phil. Trans. 18G9, pi. Ixxvii. figs. 4-6) will show how much this bird comes 

 short of the typical Fowl. This is seen also in the postorbital region, where the post- 

 orbital process of the frontal, the zygomatic process of the squamosal, and the sphenotic 

 process of the alisphenoid are all aborted in Opisthocomus, but, together, form a bridge 

 over the temporal muscle in the Common Fowl. The Fowl, like the Goose, has the 

 posterior angular process of the mandible very long ; in Ojnsthocomus it is almost 

 suppressed ; it is best seen in the embryo (Pis. VII. & VIII. fig. 2, p.ag.). The 

 figure of the hyoid arch in the 3rd stage (PI. VIII. fig. 3) shows how near this part 

 comes to that of a typical Fowl. 



IV. The Vertebral Chain of Opisthocomus cristatus. 



In the 1st stage the vertebrae are all formed but not ossified, and the number of the 

 caudal series can be determined accurately ; this part has generally six in the adult, 

 and the last of these is composed of four (PI. VII., cd.v.) very rudimentary segments ; 

 therefore nine may be given as the number in this region. I have met with no species 

 of bird in which the individual variability in the number of the vertebrae is so great as 

 in this, as though the tendency to produce races and species had gone no further than 

 to give rise to unuseful individual modifications. I suspect that this bird bears the 

 same relation to the early Tertiary types as the Tapir does to the Mammals of that 

 period — the Fal(eotherium and its congeners. Where the number of vertebrae is 

 greatest, for instance in the Swans, I have found the number of presacrals vary in 

 different species, but not in diff'erent individuals; and, as a rule, the Carinatae vary 

 more in asymmetry of the vertebrae that are developed than in the number in the 

 chain. A very large proportion of the Passerine Order have 14 cervicals, or exactly 

 twice as many as Mammals. The larger the type the greater the number of vertebrae ; 

 and the largest of the sifting birds — the Swans — have as many as the largest Ratitfe, 

 notwithstanding their extremely high position among the aquatic birds. Opisthocomus, 

 like the Gallinaceae generally, takes a middle place in this respect, halfway between 

 the Swan and the Humming-bird. In giving a vertebral formula I shall, in this case, 

 break up the avian sacrum into four secondary regions; namely, the dorso-sacral, 

 lumbo-sacral, sacral proper, and uro-sacral. The dorsals proper are those that have 

 perfect rib-cinctures, are in front of the pre-ilia, and as a rule are distinct from the 

 first that is overlapped by those plates. The following table will show the variability 

 of this chain of bones in this bird. 



VOL. XIII. — PART II. No. 3. — April, 1891. k 



