Xerophily in the Coniferce and Microphylly. 103 
they would not be found to-day to possess the characters that they 
actually do. A similar idea is expressed by Coulter & Chamberlain, 1 
who consider that the development of the “ small and rigid needles 
or concrescent scales” from their assumed broad-leaved ancestors 
“ cannot be regarded as the result of a general tendency among 
Gymnosperms quite unrelated to conditions of living. The 
leaf is too variable a structure and too closely related in its work 
to external conditions to permit such an explanation of its changes.” 
It appears that in the case of the Coniferae this assumption 
may fairly be criticised. If the theory, advocated by Professor 
Seward, of the Lycopodiaceous origin of the Conifers be accepted, 
there is involved the conclusion that the small-leaved structure is 
primitive for the Conifers. If an analogy only be admitted between 
Conifers and Lycopods, this would appear to indicate that the 
small-leaved type is remarkably constant throughout large groups. 
Even admitting the megaphyllous origin of the Conifers, we are 
confronted with the fact that the same type of small-leaved Coniferae 
has persisted from later Palaeozoic times to the present day. 
Moreover, there is much to be said in support of the view that, 
within the Conifers, the narrow one-nerved type of leaf is more 
primitive than the broader many-nerved type found in a few 
Araucarieae and in Podocaipus § Nageia. 
The evidence therefore seems to point to the conclusion that 
the small-leaved type is extremely stable : and this also seems to 
follow from what we know of modern Conifers. The variety in 
their habitat, though difficult to estimate, appears to be correlated 
with less variety in their structure, relations of leaves to stem, size 
of leaf, etc., than in the case of Angiosperms. The Larch, with its 
deciduous habit and lesser degree of structural protection against 
transpiration, but with its retention of the microphyllous foliage, 
is a striking example of the persistence of what we have come to 
regard as the typically Coniferous form of leaf. Moreover, the 
deciduous species of Taxodium and Glyptostrobus inhabit swampy 
situations without modification of the type of leaf-form. In Phyllo- 
cladus the plant has even resorted to the device of producing 
flattened shoot-expansions in place of increasing the size of its 
leaves: and the same is the case in Hollick & Jeffrey’s new 
Cretaceous Conifer, Androvettia statenensis, which they refer, on 
anatomical grounds, to the Araucarineae. 2 
1 Morphology of Gymnosperms, p. 413, 1910. 
2 Studies of Cretaceous Coniferous Remains from Kreischerville. 
New York, 1909, p. 22. 
