Seedling Structure of Gy mno sperms. I. 693 
ledon-trace immediately effect a junction with the corresponding tissue of 
the plumular strands, and concurrently the metaxylem of the same traces 
move towards, and fuse with, the wood of the epicotyledonary bundles. 
Thus the protoxylems of the seed-leaves are left exposed ; there is no 
definite rotation of this tissue ; its exarch position is attained chiefly by 
the movements of the metaxylem ; but, at a lower level, there may be made 
out a rather indefinite rearrangement of the protoxylem elements coupled 
with a slight centrifugal movement. The resulting root-structure is exactly 
the same as in Cephcilotaxus. 
Taxus cuspidata , Sieb. & Zucc. So far as our inadequate material 
indicates, closely resembles T. baccata. 
Torreya. It is unfortunately impossible to offer any original observa- 
tions on this genus, since none of the many seeds planted germinated. 
This, however, is not of much consequence, as the seedling-anatomy has 
been fully described by Miss Chick , 1 from whose paper the following 
remarks are culled : — 
The seedling is much like Ginkgo in its appearance, with two thick and 
fleshy hypogeal cotyledons which are closely adpressed and may fuse 
together by their morphologically upper surfaces. The seed-leaves show 
much variation in size and in form ; one may be about half the length of 
the other, and, as regards shape, they may be sickle-like, tubular, or with 
a well-marked spathulate apex. Another feature of interest is found in the 
fact that the cotyledons show a marked tendency to lobing, which has been 
described also for Cycads and Ginkgo by Strasburger . 2 Mrs. Tansley 
(nee Miss Chick) looks upon this lobing as being probably due to the space 
relations within the seeds caused by the ruminated endosperm, and she 
cites other cases of seeds with a similar endosperm, viz. Palms and Myristica , 
which also have lobed cotyledons ; but at the same time it is pointed out 
that the cotyledons of Anona show no tendency to lobing, although the 
endosperm is ruminated. We shall offer an alternative explanation later on. 
As regards the anatomy of the cotyledons, the main features are as 
follows : Each has a single vascular bundle, which exhibits centripetal 
wood and transfusion elements. When the seed-leaf is lobed a -branch 
of the bundle enters the lobe, and in some instances the vascular strand 
showed a tendency to branch. 
The transition is, on the whole, fairly rapid. ‘ The root protoxylem 
dies out below the cotyledonary node ; and the protoxylem which accom- 
panies the cotyledon-trace outwards would seem to have a double origin, 
one portion . . . which is directly inserted upon the root metaxylem, and 
the other seeming to belong more intimately to the cotyledon-trace. . . . 
A possible explanation of the “ dying-out” of the root protoxylem is that 
1 E. Chick : The Seedling of Torreya Myristica (New Phytologist, vol. ii, 1903). 
2 Strasburger, loc. cit. 
