SPERMATOPHYTES 
227 
The cotyledons are not steadily two in number, as in the preceding 
groups of gymnosperms, but vary from two (mostly so in Cupressineae 
and Araucarineae) to as many as fifteen (in the Abietineae). It is be- 
cause the pines and their allies are the most familiar gymnosperms that 
gymnosperms are commonly described as polycotyledonous ; but it 
should be remembered that the occurrence of more 
than two cotyledons is a feature of only two tribes 
of Coniferales (Abietineae and Taxodineae), and 
that two cotyledons occur in some of the members 
even of these tribes. 
Conclusions. It seems most reasonable to 
conclude that the Coniferales have been derived 
from the paleozoic Cordaitales, which also gave 
rise to the Ginkgoales. The Coniferales, however, 
have retained fewer primitive characters than the 
' Ginkgoales, and are especially noteworthy in 
having lost the swimming sperms. 
In comparing the six tribes of Coniferales, the 
testimony as to their relationships is very con- 
fusing. The testimony obtained from the geo- 
logical record is necessarily incomplete, but so far 
as it is available the relative ages of the tribes are 
as follows: The Abietineae have been traced to 
the Paleozoic, and in all probability are the oldest 
of the Coniferales. The Araucarineae have been 
traced through the Mesozoic, in which period they 
were very abundant ; and in all probability they 
are but little younger than the Abietineae. The 
Taxodineae and Cupressineae are recognizable in ( above ) two cells of the 
...._- ._ , rosette, two cells of the 
the Lower Mesozoic; while the Taxaceae (Podo- suspensor> and an early 
carpineae and Taxineae) are not known below the stage of the terminal 
middle Mesozoic. It may be that this sequence embryo. After COUL- 
. ,. , , c . ., , . . TER and CHAMBERLAIN. 
indicates the actual sequence of the tribes, but it is 
hard to reconcile it with the morphological characters detailed above. 
In any event, the fact that Coniferales as a whole have developed from 
the Cordaitales seems to be sufficiently clear, and is all that need con- 
cern the elementary student of the group. It should be remembered that 
origin from Cordaitales means also a connection through them with the 
ferns, and therefore that all gymnosperms have descended from ferns. 
FIG. 510. Embryo 
of Pinus, showing 
