8 4 
Edith Chick on Torreya Myristica. 
Torreya seedling recalls vividly that of Gingko in appearance. 
In each the cotyledons are hypogeal structures, thick and fleshy, and 
bearing very little resemblance to leaves. (PI. viii., Figs. 2, 3 and 4). 
The petioles, starting from either side of the hypocotyl, bend round, 
each through 90°, to meet in a position half-way between, and become 
closely adpressed even outside the seed. (PI. vii., Figs. 1, 2 and 3, 
PI. viii., Fig.4). There is a considerable variation in the length of 
petiole outside the seed. In the two Orton specimens (PI. vii., Figs. 
2 and 3) this was only a few millimeters, while in the one from Kew 
there was a length of nearly 4 cm. 
Within the seed it is difficult to say where the petiole ends and 
the lamina begins and there is also a great variation in the form of 
the adpressed cotyledons (PI. vii., Fig. 2, and PI. viii., Figs. 3 and 4), 
those of one seedling being sickle-shaped, of another broadening to 
a spathulate apex and of the third being almost tubular. 
The appearance obtained on breaking open the seed and 
dissecting out the cotyledons, leaving them lying in position, is most 
striking: The cotyledons, which are of a vivid green colour, lie in 
the white endosperm, into which penetrates in all directions the hard 
black nucellus. The endosperm is cut up into a body of very 
4 
irregular shape owing to this rumination of the intrusive nucellus, 
which, brown at first, has become black at this stage. It is possible 
that the rumination of the endosperm may account for the variations 
in shape of the cotyledons and for other points about them which 
will be dealt with later. 
Form of the Cotyledons. 
The cotyledons having the simplest form were those from the 
seedling shewn in PI. vii., Fig. 2. The portion within the seed was 
35 mm. long and each cotyledon possessed a petiole-like region and a 
spathulate expanded apex, which may be called for convenience the 
lamina. The two cotyledons were similar in all respects and were 
free from each other throughout. The epidermis, however, of each, 
on the side directed towards the other, was in close contact with 
the endosperm, and probably was performing an absorptive function. 
Each cotyledon had a single vascular bundle which ran the 
whole length of the cotyledon without dividing, merely broadening 
in the laminar region. The anatomical details will be considered 
later. 
In the seedling from Kew, (PI. vii., Fig, 1 and PI. viii. Figs. 1, 
2 and 3), the cotyledons were tubular and fused together by their 
