230 FOOT-NOTES TO EVOLUTION. 
allel developments from a common stock, or else the 
separation from the oldest stock of crinoids took place 
in the misty pre-Cambrian time. 
The vertebrates, too, show evolution in a general 
way in the appearance of types, for we find fishes from 
the Lower Silurian on; at first only placoderms, bony- 
plated fishes, then upward by degrees to the teleosts 
(scale fish with bony skeleton), which began in the Up- 
per Jura. Amphibians had branched off from the fish 
stock by the end of the Devonian, genuine reptiles from 
the amphibians by the end of Carboniferous time. Mam- 
mals appeared first in the Trias, probably as descendants 
from the amphibian stock. The first birds we find in 
the Upper Jura as transitions from reptiles. 
The method used by most naturalists in the study of 
phylogeny has been a comparison of a series of adults 
from successive geologic horizons, together with a study 
of present and past distribution and migration of ani- 
mals. The results of this are seen in all our classifica- 
tion and in all family trees. Such work as Marsh’s 
origin of the horse, Cope’s phylogeny of the reptiles, 
Baur’s contributions to the origin of the mammals, and 
many others, abundantly justify this method of research. 
But interesting and valuable as are the investigations 
in phylogeny made in this way, such genealogies can 
not, as a matter of course, be more than approximate, 
for we have nowhere a uniform succession of rocks, and 
nowhere an unbroken genetic series. Each region has 
often changed, belonging now to one faunal province, 
now to another, each great change in faunal geography 
showing some physiographic revolution here or else- 
where. Thus the local series is broken and filled out 
from other regions, species usually being classed to- 
gether because of mere resemblance, while their real re- 
lationship is unknown. 
