Primitive Angio sperms . 143 
of the low place in the botanical scale held by Vascular Cryptogams. For 
many years fossil botanists were accustomed to place all the stems with 
normal secondary thickening among Phanerogams without further question. 
While this view prevailed, the absence of cambium in the stems of Mono- 
cotyledons was naturally considered as a primitive character. Now, however, 
a cambium is known to be the rule among the extinct Vascular Cryptogams 
whose tissues have been sufficiently well preserved for microscopic exa- 
mination. The marked difference between living and extinct forms in 
this respect is no doubt due to the fact that living Vascular Cryptogams 
are herbaceous, while the forms preserved as fossils are mainly trees. In 
such forms a cambium would be found if it existed within the group 
at all. 
The evidence then shows that a cambium is much more ancient than 
the Primitive Angiosperms, and that they probably inherited one, from 
whatever race of Gymnosperms or Vascular Cryptogams they descended. 
Comparison of seedling with mature Dicotyledons shows that the ring 
of leaf-traces is visible in the plumule so soon as vascular tissue is differen- 
tiated within it. Cambium is present from the first within each trace, and 
appears between them very shortly. The single ring of open bundles is 
formed even in the seedlings of such forms as Podophyllum , in which the 
later bundles are closed and scattered within the stem. 
Thus the primary stem-structure of Dicotyledons is the same in the 
seedling as in the mature plant. If the Monocotyledonous type were 
primitive, we might expect some of its features to appear in the young 
Dicotyledon. Amphivasal closed bundles should be formed first, and pass 
gradually into the open collateral form. But this does not happen. 
The earliest bundles formed are collateral, and have a well-marked cam- 
bium. 
Among Monocotyledons, on the other hand, the characteristic stem- 
structure is reached only when the stem approaches maturity. The bundles 
of the seedling internodes are collateral, and frequently contain traces of 
a cambium. Their arrangement within the stem can be traced satisfactorily 
only in the rather exceptional forms in which the early internodes are 
well developed. In such forms Professor Jeffrey ( 46 , pp. 314, 315) has 
pointed out that the bundles of the first internodes are commonly arranged 
in a single circle, and orientated as in Dicotyledons. It is true that, while 
there are still very few leaves, their traces would naturally fall into a single 
circle. But their regular orientation in that circle — xylem inwards, phloem 
outwards — suggests that they were formerly linked together by a cambium. 
Besides, the occasional presence of a short-lived cambium within the traces 
of the seedling, and their strictly collateral structure, are very strong evidence 
for supposing that to be the primitive structure of the leaf-trace instead 
