1893.] Zoology. 393 
but even here the distinction is not absolute, for the Tinamous approach 
in many respects more nearly to the Ratite than to the rest of the 
Carinatæ. Still it seems probable that the various genera of Ratite 
must have diverged from the main line of descent at a comparatively 
early period, though perhaps not earlier than some of the existing 
orders of Carinatz. The Penguins, for instance, are far more reptilian 
in their vertebral column and less typical in the structure of their 
wings than the Ratite. The Ostrich, however, shows the unique and 
very reptilian character of two claws on the wing, and the very gen- 
eral presence of wing-claws in the group is a distinctly primitive 
character. 
“ Leaving the skull, in which the whole group shows primitive char- 
acters, and the wing and related parts, in which the resemblances 
between the genera are largely adaptive, we find the range of variation 
in the Ratitæ to be very great indeed. Two genera (Apteryx and 
Dinornis) have a normal four-toed foot; in three others (Cassowary, 
Emu and Rhea) the hind toe or hallux has disappeared ; while in 
another (Ostrich) only two toes are left. The pelvis of the kiwi and 
moa is of the simplest avian type, both pubis and ischia being free; 
in the cassowary and emu the ischium unites with the ilium ; in the 
rhea the ischiæ unite with one another above the intestines—a unique 
arrangement ; in the ostrich the pubes unite to form a symphysis as 
in most of the higher vertebrates. The feathers have an aftershaft in 
the emu, cassowary, and moa, none in the ostrich, rhea, or kiwi. In 
no order of carinate birds do we find such a wide range of variation 
as this, and when we add to the characters enumerated the extraordi- 
narily aberrant skull and the structure of the egg shell of the Apteryx, 
the total atrophy of the wings in Dinornis, and even of the shoulder- 
girdle in some species of the genus, and the striking differences between 
the sterna, the shoulder-girdle and the wings of the various genera, we 
are forced to the conclusion that the existing or lately extinct cursorial 
birds now known to us are divisible into five well-marked orders, each 
the equivalent of an entire order of Carinate. Of these one order 
contains the ostrich alone, another the rheas, a third the emu and the 
cassowaries, a fourth the moas, and a fifth the kiwis. ; 
“« As to the relation of the kiwi to the other genera it has been shown 
to be most nearly allied, as far as its skeleton is concerned, to the moa, 
differing from it, however, in many important respects. It must cer- 
tainly have been isolated at a very distant period, and as far as we can 
sée, some of its more striking peculiarities are distinctly correlated to 
its method of feeding. Most nocturnal animals have large eyes suited 
