457 
Phloem in the Stems of Dicotyledons . II. 
Although all types of inflorescence and all kinds of vegetative habit 
are found in the Monocotyledons, yet no one has ventured to suggest 
that the scattered disposition of the bundles in stem and leaf is due to more 
than a single cause, that cause being recognized as lying in the geophytic 
habit (modern or aboriginal) of the plants corhposing the class. In the 
same way, it is more than likely that there must be a single cause for the 
same type of vascular system of the closely-allied class of Dicotyledons. 
Just as we regard the two great classes as having had a common 
origin in the past, in the same way we must regard the vascular system 
of each as being derived from a common ancestral type of vascular 
system. 
The question therefore arises, What has this ancestral type of vascular 
system been ? Has the structure of Compositae (a type of most other 
Dicotyledonous orders) been derived from that which consisted of a scattered 
disposition of the bundles in stem and leaf, or from that which consisted of 
a ring of bundles (as seen in transverse section) in the stem and either a ring, 
or arc, of bundles in the leaf? Another form of the question is : Were the 
ancestors geophytes, or were they non-geophytic herbaceous or arboreal 
forms ? 
A comprehensive survey of all the facts clearly suggests that the 
ancestors were either geophytes or semi-geophytes. The Compositae and 
the other Dicotyledons are now emerging from this ancestral condition ; 
a small minority of Compositae have, indeed, actually become arboreal in 
habit ; but the herbaceous habit is certainly primitive in this order, as most 
systematists who are conversant with it will probably agree. 
Jeffrey gives an account, in his chapter entitled ‘The Herbaceous 
Dicotyledons ’, of the structure of the vascular ring of the stem of Compo- 
sitae at the point where the leaf-trace is detached. He compares the 
structure of the more woody and the more herbaceous types respectively. 
He concludes, in effect, that the facts constitute clear evidence that the 
herbaceous type of structure has been derived from the woody type. The 
present writer desires, however, to state most emphatically his complete 
inability to discover any evidence whatsoever in the facts set forth by 
Jeffrey with regard to the structure of Compositae, or any other order, 
which would tend to indicate that the herbaceous type of structure has been 
derived from the woody type. The entire chapter is devoid of scientific 
argument, consisting of scarcely more than a descriptive statement of the 
structural facts. 
If the view be held that the Compositae are primitively herbaceous, 
then the sporadic occurrence of medullary bundles, their presence in some 
individuals, species, or genera, and not in others, their predominantly rudi- 
mentary, imperfect development and structure, as also their late ontogenetic 
appearance as compared with that of the main vascular system, all become 
