8io Compton . — A n Anatomical Study of 
in Hemitricotyl A, forming two triads which enter the split cotyledon. 
There were three leaves in the first whorl. 
Hemitricotyl G. A curious example, only observed once, in which the 
vascular system derived from one root-pole began to fork in the way typical 
of schizocotyly (cf. Seedling A), although there was no sign of a split in 
the cotyledon it supplied. It thus appears that the tendency to fork may 
be manifested in the vascular structure, though not in the external form. 
The other cotyledon was forked in the usual way. 
We may summarize the main features of these anomalous seedlings of 
Cannabis sativa as follows : 
(i) The dicotyl is diarch with an intermediate level of transition. 
(ii) The fission of a cotyledon leads to a fission of the corresponding 
root-pole or its connexions, at a higher or lower level in the axis roughly 
according to the degree of splitting. This applies alike to the splitting of 
one or both cotyledons. 
(iii) There is a continuous series of stages from scarcely perceptible 
fission to deep lobing and complete tricotyly . 1 
(iv) The vascular structure seems to lag behind the degree of splitting 
of the cotyledons to some extent (except in Hemitricotyl G), for several 
examples were seen of complete tricotyls with diarch roots ; others, how- 
ever, had triarch roots and the whole structure was symmetrical about 
three diameters. 
(v) An instance of a £ subsidiary ’ cotyledon was observed in Seedling 
D ; the validity of this category is doubtful, for we can hardly assume any 
phenomenon beyond schizocotyly in this case, whatever we may do in the 
Conifers. 
(vi) The plumular symmetry lags behind the cotyledonary : we do not 
find trimery in the epicotyl unless at least three distinct cotyledons are 
present, and occasionally not even then. 
Ulex EUROPAEUS (Leguminosae). 
The dicotylous anatomy has been described recently . 2 Mr. Bolland ? 
of the Cambridge Agricultural Laboratory, gave me a single hemitricotyl 
whose lobed cotyledon was divided to within i mm. of the base. The 
structure of this seedling corresponded to some extent with that of Cannabis 
sativa and may be taken next in order. 
The root was diarch, and the two polar protoxylems remained un- 
divided to three-quarters up the hypocotyl : the usual processes found in 
the dicotylous transition taking place, so that two polar bands of xylem are 
formed each with two phloems. At three-quarters up the hypocotyl one of 
these phloems gives off a branch towards the cotyledonary plane, and at 
1 And probably tetracotyly, which, however, I have not observed. 
2 Compton (’12), p. 28 . 
