698 TRANSACTIONS OF SECTION K. 
is comparatively simple, yet it has been attacked from different sides, and the 
attempt to give a concise account of the results attained is beset with difficulties. 
For the present, however, I propose to consider only their bearing on the stelar 
hypothesis. 
Indeed, seedling anatomy becomes extremely important when the vascular 
system of the root is compared with that of the stem. For in the seedling we 
have a complete and simple vascular skeleton, which at one end belongs to the 
primary root of the plant, and at the other to its primary stem. There must be 
an intermediate region in which stem-structure passes into root-structure, and 
the method of transition should at least suggest, if it does not precisely deter- 
mine, the relation in which they stand to each other. For this reason great 
value has been attached by anatomists to the transitional region of the main 
axis. It was not completely investigated, however, until the microtome was 
introduced into botanical practice, for the change of structure is often very 
abrupt, and cannot be studied in detail unless all possible sections are present in 
their proper order. 
In this, as in other branches of modern anatomy, Professor Van Tieghem was 
first in the field. In his memoir of 1872, ‘Sur les Canaux Secréteurs des Plantes,’ 
he described the course of the bundles in the hypocotyl of T'agetes patula, an 
example of the second type of transition given in his text-book (1886). The 
three types were, indeed, already identified in 1872, for the first and third are 
defined in a footnote appended to the description of Tagetes. 
Tagetes patula was, of course, examined in 1872 with the aid of hand-sections 
only. Two traces enter the hypocotyl from either cotyledon, and form in the 
end a diarch root. The plane passing through its xylem poles is the median 
plane of the cotyledons. In the upper part of the hypocotyl this plane bisects 
the space which separates the two bundles entering each cotyledon. So far the 
description of J’agetes given in 1872 is identical with the generalised account of 
type 2 in the text-book (1886). But a detail of some importance is mentioned in 
the description of Tagetes which does not reappear in the definition of type 2. 
In each of the spaces just mentioned—called, for convenience, xylem spaces, 
because they lie above the xylem poles of the root—lies an isolated xylem 
element, the direct continuation of the most external element in one of the root 
poles, and this element comes to an abrupt end higher up. 
Thus Professor Van Tieghem has tacitly assumed that 7'agetes is exceptional 
in this respect, and this view was also adopted by Professor Gérard in his 
laborious and accurate paper of 1881. He describes the transitional phenomena 
of a number of Dicotyledons, among them Z'agetes erecta. Not only is the 
transition in this species exactly the same as that in 7’. patula, but the author 
records a similar isolation of primitive xylem elements in Raphanus niger, 
Ipomea versicolor, and Datura Stramonium, still treating the arrangement as 
exceptional. 
These details are important, because if certain protoxylem elements belonging 
to the root are not continued upwards in regular succession into the cotyledonary 
or plumular bundles, but end abruptly in hypocotyl or base of cotyledon, there 
is not that complete correspondence between stem- and root-structure which is 
assumed in Van Tieghem’s three types. In all of them the xylem and phloem 
bundles of the root are continued into the cotyledons or plumule. On their way 
through the hypocotyl they may divide or be displaced, and the xylem bundles 
‘rotate ’—that is, they turn on their own axes until the protoxylem is internal. 
But all the elements present in the root are continued upwards in regular 
succession, and are simply rearranged in the upper part of the seedling. This 
is one of the main arguments advanced by Professor Van Tieghem to support his 
view that the steles of root and stem are identical. 
According to most later observers, however, such temporary prolongation of 
the root-poles upwards as that described by Professors Van Tieghem and Gérard 
in a few instances, and considered by them as exceptional, is really of general 
occurrence. The protoxylem elements, indeed, are not commonly isolated from 
the main xylem of the cotyledonary traces as in 7’agetes, but are in more or less 
complete contact with them on either side. Such contact is approached in 
Raphanus niger, where it is very clearly suggested in Professor Gérard’s figures, 
