1885.] DR. H. GADOW ON THE SPECIES OF RHICA. 313 



portion. These ribs are therefore anterior psendo-thoracic. The 

 last three ribs of the whole number of eight are in all Rheas likewise 

 not connected with the sternum — true lumbar or posterior pseudo- 

 thoracic ribs. The skeletons of the American Rhea in the Cam- 

 bridge and Heidelberg Museums have each four pairs of true sternal 

 ribs, each being furnished with an uncinate process. 



Rh. macrorhyncha varies from the other two species in having 15 

 true cervical vertebrae, the 1 6th of the series bearing the transitional 

 rib, whereupon follow two long pseudo-thoracic, then three true 

 sternal, and lastly tliree lumbar ribs. The 23rd vertebra (iu Rhea 

 darioini and atnericana), hut the 24th in Rh. macrorhyncha, is the 

 first which is firmly ankylosed with the ilium and with the following 

 vertebra. The same happens, however, to the next one in front in 

 most old specimens. 



The acetabular connection of the ilium with the vertebral column 

 is effected by the 26th and, chiefly, the 27th vertebra, but in Rh. 

 macrorhyncha by the 2/'th and 28th. 



The true primitive sacral vertebrae, as determined by Gegenbaur 

 for the avian pelvis, are in all specimens the 33rd and 34th. 



In the male specimen of Rh. clarwini the 34th nerve received no 

 connecting branch from the 33rd, the latter therefore belonging 

 entirely to the ischiadic plexus, just like the 34th in Rh. macro- 

 rhyncha. At any rate the ischiadic plexus is in all the specimens of 

 Rhea examined by myself composed of six postacetabular spinal nerves. 



Considering these facts, we conclude that the 34th vertebra is 

 serially homologous in all the specimens as being the second primitive 

 sacral vertebra, and it can in our comparison be looked upon as the 

 starting point or zero, but that the whole ischiadic and crural 

 plexuses of Rh. macrorhyncha are placed one metamere further tail- 

 wards than the corresponding portions iu Rh. darwini and Rh. 

 americana. Moreover, since the portion from the transitional 

 vertebra to the last rib-bearing vertebra is homodynamous in all the 

 three species (although falling under the category of parhomology 

 or imitatory homology of Fuerbringer), we conclude rightly that 

 Rh. macrorhyncha possesses one true cervical vertebra more than 

 the other two species. This excess in number is compensated by the 

 shortening of the number of presacral pelvic vertebrae from 6 to 5. 



The number of vertebrae constituting the various regions is there- 

 fore : — 



33 33 33 



