252 FOOT-NOTES TO EVOLUTION. 
in growth. All the pieces of whorls are preserved, but 
often it is possible to have a complete series in one 
specimen. The individuals representing stages of 
growth are kept separate, in small glass tubes attached 
to cards for labels, on which are noted the measure- 
ments of the specimen, stage of growth, and such other 
facts as are wanted for ready reference. ' 
The points to be noted in studying the development 
of chambered cephalopods are: character of the proto- 
conch or embryo shell; position of the siphuncle, 
whether internal, median, or external; character and 
direction of the siphonal collars, as on Plate V, Fig. 9, 
where Zropites phebus has the young siphonal collars 
pointing backward, and the old ones pointing forward, 
also change of the siphuncle from the internal to the 
external margin of the whorl; increasing lobation of 
the septa with advancing growth; changes in the spiral 
of the coil; increasing involution, height of whorls, and 
ornamentation of the outside shell. By observing care- 
fully these changes in character, the successive stages 
may be sharply distinguished from each other from the 
beginning of the larval period to the old age of the 
shell. 
No one species in its life history gives the entire 
history of the race; the earlier forms do not get far 
along in development, while the later ones hasten 
through the earlier generic stages so rapidly that a 
Jurassic or Cretaceous species often begins life where a 
Paleozoic form in the same line of descent left off. We 
are thus often forced to piece out the development his- 
tory with successive species, using both the develop- 
ment of the individual and the successive development 
seen in the rocks. The writer has recently worked out 
the development of Giyphioceras of the Carboniferous, 
and Schloenbachia of the Cretaceous, in the same line of 
