102 MR. R. Н. COMPTON: AN INVESTIGATION OF THE 
(c) Traces from the first epicotyledonary node represented in the root 
by metaxylem. (Sophora spp.) 
(d) Traces from the first epicotyledonary node represented in the root 
by proto- and metaxylem. (Pithecolobium Unguis-cati, Cosalpinia 
sepiaria.) 
(e) Traces from the first epicotyledonary node represented in the root 
by proto- and metaxylem, from the second by metaxylem. (Many 
Viciez.) 
(f) Traces from the first two epicotyledonary nodes represented in the 
root by proto- and metaxylem. (Vicia Faba.) 
Let us consider the first type of connection more closely. It is in this 
type that the natural distinction between the vascular systems of seedling 
and epicotyl is most pronounced ; and consequently it is here, if any where, 
that we should expect to find special devices with the object of bringing the 
two systems more into physiological connection with one another. Now in 
the case of a tetrarch seedling the two lateral protoxylems lie immediately 
beneath the main vascular bundles of the first two leaves, which are almost 
invariably produced in the intercotyledonary plane ; and consequently it is 
these xylems which are most closely en rapport with the young hypocotyl. 
Hence it is in the intercotyledonary xylems that we should look for indications 
of the effect of the plumule in modifying the structure of the hypocotyl. 
As a matter of fact, in the slender epigeal seedlings so abundant in the 
Trifolieæ, Lotew, Galegeæ, and Hedysareæ we find the polar xylems con- 
stantly present ; while the lateral xylems exhibit a great amount of variation, 
even within the same hypocotyl, diarchy, triarchy, and tetrarchy occurring 
within very limited circles of affinity. Moreover, we find in these same 
groups that the form of the cotyledon trace differs very little whatever be 
the symmetry of the hypocotyl. The lateral root-poles almost always 
diminish in importance as they ascend the hypocotyl ; and ina few cases they 
die out completely towards its summit instead of entering the cotyledons. 
The conclusion is irresistible that the variations exhibited by the inter- 
cotyledonary xylems stand in an intimate relation to the first foliage leaves : 
and we proceed to enquire what is the nature of this relation. 
Two alternatives present themselves, viz.:—(i.) The plumular traces may 
tend to replace the intercotyledonary root-poles, or vice versá, their relation 
to one another being complementary and one of supplantation.  (ii.) The 
intercotyledonary xylem may combine with the leaf trace to form a joint 
conducting channel, their relation to one another being supplementary. 
The first theory was originally formulated by Miss Sargant (1903, p. 61), 
and has been adopted by Miss Thomas (1907, p. 84). On this view the diarch 
type of structure is regarded as being derived from the tetrarch, the change 
