1871.] DR. R- O. CUNNINGHAM ON RHEA. 107 



zontal plate of the ethmoid not covered by the frontals and nasals is 

 of a different form, being an ellipse with pointed ends instead of a 

 somewhat lozenge-shaped space as in the latter bird. In the former 

 species there is also a much wider unossified space in the interorbital 

 septum between the basi- and presphenoids than there is in the 

 latter ; but this is most probably a difference connected with age. 

 On the other hand the lacuna in the ethmo-alisphenoid plate, im- 

 mediately beneath the horizontal plate of the ethmoid, is nearly 

 twice as large in Rhea americana as it is in Rhea darwinii. The 

 pterygoid processes of the basisphenoid are also curved more forwards 

 in the former than in the latter. By far the most remarkable 

 distinction in the bones of the skull of the two birds, however, is fur- 

 nished by the lachrymal. In Rhea americana (PI. VI. fig 1) the 

 strong process directed backwards (anterior orbital process of authors) 

 is much more elongated than in R. darwinii, and the form of the 

 descending process is also very different. In the former it curves 

 downwards and backwards so as to produce a deeply excavated space 

 on its posterior border, while in the latter (fig. 2) this process is 

 met by another posterior bar of bone so as to connect the space into a 

 large foramen. This will be more readily understood by a reference 

 to the accompanying sketches, in which fig. 1 represents the bone in 

 R. americana, and fig. 2 in R. darwinii, the letter a in both bones 

 indicating the surface of articulation with the cranium. Other 

 minor differences in the bones of the head are probably due to age. 



VertebrcB. — Except in point of size, the vertebrae of the two 

 species differ but little from one another. I find twenty-one free 

 vertebrae present in R. americana between the head and the lumbo- 

 sacral portion of the vertebral column ; but the axis is wanting in 

 this specimen ; so that the total will be twenty-two should none of the 

 other vetebrae be absent, which does not appear to be the case. As 

 in the specimen of R. darivinii either four or five of the cervical 

 vertebrae have been lost, I cannot speak with absolute certainty 

 as to the number of free vertebrae present ; but in all probability it 

 is the same as in R. americana. The styloid processes (pleurapo- 

 physes) of all the cervical vertebrae examined in R. darwinii exist as 

 separate bones, as might, indeed, be expected from the immature 

 condition of the individual. They are fully anchylosed in the third, 

 fourth, and fifth vertebrae (putting the atlas, in which they do not 

 exist, and the axis, which is missing, out of the question) of R. 

 americana, partially so in the sixth, seventh, and eighth, and sepa- 

 rate thence to the first rib-bearing vertebra. 



In both species there are eight rib-bearing vertebrae, the eighth 

 being partially anchylosed to the lumbo-sacral portion of the column. 

 No very obvious differences, save those of size, are to be observed 

 between the vertebral ribs of the two species. In both but three (the 

 third, fourth, and fifth) are connected with the sternum by means of 

 costal ribs. There is likewise no material difference in the form of 

 the sterna of the two birds, allowing for the influence of age. In 

 R. americana the two lateral halves (pleurostea) are united rather 

 firmly together, but would admit of separation without much diffi- 

 culty ; while in R. darwinii they are separate, partly in consequence 



