528 THE AMERICAN NATURALIST. (VoL. XXXIX. 
not in any case attain to the extreme preponderance noted in the 
Cordaite and Conifer, etc., is precisely in harmony with the 
general approximation of the Cycadacez as a whole to their 
filicinean ancestors and the lack of need for that peculiar form 
of mechanical support which is demanded by trees of large size, 
and which is provided in all the arborescent forms of the seed 
plants. With these statements it will be unnecessary to follow 
the Cycadacez in further detail, but rather to consider the 
course of events in the more arborescent Cordaitales and Conif- 
erales where the analogy with the angiosperms is closest. It is 
now a well recognized fact that the protoxylem of the gymno- 
sperms constitutes a transition zone within which peculiar and 
often somewhat remarkable evolutionary changes take place. 
The protoxylem or transition zone is a region peculiarly sensi- 
tive to the controlling influences of environment, and its response 
is so immediate as to give rise to structures of a very diverse 
nature. This has been shown to be true of the Cycadacez 
(Penhallow, :04c) to a notable extent, while it is no less promi- 
nent in the Cordaitz, both of which groups are of a recognized 
primitive character among gymnosperms and stand in somewhat 
close relations to their Cycadofilicinean ancestors. But a very 
important difference exists between these two groups with respect 
to the survival of the protoxylem and the precise composition 
of the secondary wood. In the Cycadaceze the growth of the 
stem in annual increments involves the complete repetition of the 
entire xylem structure and the formation of secondary growth 
in a manner which is to be met with in no other group of gym- 
nosperms. The growths of successive years are not immedi- 
ately joined to one another and therefore conterminous, but they 
are separated radially by very definite zones of fundamental 
tissue of the original cortex. In each zone of growth we recog- 
nize spiral and scalariform elements of a transitional form, 
together with tracheids bearing bordered pits and forming the 
secondary xylem. These tissue regions are disposed precisely 
as inthe growth of the first year; they bear the same relations 
to one another both in position and in development, and in each 
case they perform the same functions as in the initial zone. 
From this it follows that there is a complete regeneration of the 
