342 
lamoichthys. The ribs in Polypterus are well known to represent an 
ontogenetic stage in Urodeles !). 
7) Relations with Stegocephali. Several of theallied forms 
of Devonian fish (PANDER) had teeth which were fluted like Laby- 
rinthodont teeth. I hope soon to describe the squamosal region tho- 
roughly in Polypterus and to show that there is a fairly close homo- 
logy with the Stegocephali. 
5) Bone formation. Enough stress has not been laid on certain 
mechanical conditions under which bone is formed. Examining for 
example the ethmopalatine articulation in a young Polypterus peri- 
chondrial bone may be seen to be invading the ectethmoid cartilage 
as the roots of a tree pierce the ground. The ectethmoid becomes 
later an independent bone. The converse holds with the autopalatine. 
Thus the palatine bar may be said to “pick out” a bone from the 
cranial cartilage. In this way arise many bones in the skull throughout 
the animal kingdom. The pterotic of Teleostei may be quoted as 
another good example. So also arise the vertebrae in the vertebral 
column with their numerous variations — not as protecting elements 
around the nerves nor analogously to the folds in the sleeve of a man’s 
coat. Two special instances recur below. 
9) Cheiropterygium. The hand of the Urodele must be de- 
rived from the Crossopterygian fin. The ulna is represented by the 
propterygium ?) the radius by the metapterygium. The mesopterygium 
represents the intermedium and centralia and the foramina in it might be 
called intercarpal foramina. The pro- and meta-pterygium (ulna and 
radius) are ossified and articulate with the shoulder girdle. One has 
every condition necessary for the “picking out” of a humerus from 
the shoulder girdle. When the Crossopterygian Ganoid became a 
marsh animal and a double levered arm became of advantage, this 
portion of the shoulder girdle through Natural Selection would become 
a perfect humerus. Such must be the origin of the humerus, 
Later the development of the humerus is pushed back to the cartilage 
stage in the ontogeny of the higher Vertebrates. 
The view here proposed appears to be entirely in accord with the 
facts of development °). 
1) Hatscuex, Die Rippen der Wirbeltiere. Verhand. d. Anat. Gesell- 
schaft in Berlin, Jena 1889. 
2) See also Emery, Beziehungen des Cheiropterygiums zum Ichthyo- 
pterygium. Zool. Anz. X, 87. 
3) WIEDERSHEIM, Entwicklung des Schulter- u, Beckengiirtels, Anat, 
Anz. 1889 und 1890, 
