EE 
leaves in whose axils they arise ; while the vascular traces run 
out not into the leaves but into the cladodes. 
There yet remains to be mentioned a matter of some interest 
and no little importance. Owing to the increasing girth of the 
seedling the sheathing base of the cotyledon becomes pushed to 
one side of the lower extremity of the stem; while opposite it, 
and at the same level, arise the one or two kataphylls, now no 
longer enfolded by it (C, Fig. 3). A. officinalis, the only species 
previously described, has only one of these, and Henslow (8) 
and others have thought that it represented a second cotyledon, 
delayed in development and reduced in size. Following out this 
idea still further, it has been suggested that monocotyledons 
have been derived from a dictotylous ancestor by the sup- 
pression of one of the cotyledons. A more detailed account of 
the changes which take place in the plumular meristem during 
.germination will be necessary before discussing this question. 
I may say, however, that while it seems quite likely that the 
ancestor of all present-day angiosperms was a dicotylous 
type, the above attempt to prove this will not stand examina- 
tion. 
The Development of the Plumular Meristem. 
After the examination of a large number of monocotylous 
embryos I have found that, knowing the relative development 
of stem and leaf in the mature plant, one can guess with fair 
accuracy what the structure of the embryonic plumule will be. 
In other words, I believe that the characteristics of the adult 
produced in response to its environment, have tended to appear 
ever earlier in the life-history of the individual, till now they are 
found impressed even on the embryo, shut up though it be 
within the seed. Moreover, I am inclined for this reason to 
doubt the utility of searching for evidence regarding a plant’s 
phylogeny in this region at least of its embryo. Applying what 
has been said to the genus Asparagus, let us consider what might 
be expected in the plumular meristem, and then observe how 
this agrees with its actual structure, as described later on. We 
are dealing with plants whose leaves are reduced to mere scales, 
while very much specialised branches function in their stead ; 
the stem has in fact become dominant over the leaf. It would 
therefore be natural to infer that the formation of leaf-primordia 
might possibly not yet have taken place in the embryo, and that 
in its subsequent growth the greater part of the plumule would 
be devoted to the formation of axis. 
Though in the case of many monocotyledons the plumule 
_ of an embryo taken from a ripe seed will be found to consist of 
one or more leaf rudiments arising from an axial part, in the 
