the Ovule of Myrica Gale . 359 
determined ; but it is interesting to remember that it occurs in a group 
of recent plants which is regarded by many botanists as primitive. 
Turning now to the other character of the ovule of Myrica Gale 
which has been specially noted — the continuation of the vascular supply 
into the integument— -it does not seem very easy to explain the presence 
of this integumentary vascular system by any special requirements of 
the integument, as it is neither of great thickness, nor does it become 
succulent when ripe. Probably it is correlated with the complete separa- 
tion of the nucellus from the integument, which renders it impossible 
for the integument to receive its nutriment through the nucellus, as 
is possible in most seeds where only the apical portion of the nucellus 
is free. 
Comparing the ovule of Myrica Gale with the seed of Trigonocarpus 
in respect to the vascular system, we find again a certain resemblance 
(Text-fig. II). In Trigonocarpus the main supply bundle to the ovule 
gave off at a short distance below the chalaza a number of branch bun- 
dles, usually six to nine, which, bending outwards, traversed the sarco- 
testa of the integument. In Myrica we find a similar arrangement of 
integumentary bundles given off from the main supply bundle. 
The existence of such an integumentary vascular supply in fossil 
seeds such as Trigonocarpus , where we have an integument free from the 
nucellus, is interesting in comparison, and supports the view of a corre- 
lation of these two characters. 
In Trigonocarpus the main supply bundle, after giving off the integu- 
mental bundles, continued unbranched through the stalk of the seed until 
it reached the nucellus. There it expanded to form an apparently con- 
tinuous mantle of tracheids investing the macrospore and terminating as 
anastomosing strands just below the pollen-chamber. 
In Myrica , of course, there is no necessity for such a system of 
nucellar tracheids as was required by an ovule with a large pollen- 
chamber, filled at the time of fertilization with water ; but there is, as 
described above, the distinct strand of elongated cells running up the 
centre of the nucellus which evidently serves the purpose of nutrition of 
the embryo-sac (Fig. 2, c.si). 
This central strand of elongated parenchymatous cells in the nucellus 
of Myrica Gale may possibly represent the remains of an ancient nucellar 
vascular system, or it may be a new structure developed to facilitate the 
conduction of nutritive material to the embryo-sac. When one remem- 
bers the tracheids found by Miss Benson 1 in the nucellus of Castanea , 
and by Treub 2 in that of Casuarina , regarded by Miss Benson as 1 a 
vestige of some long lost structure ’, one is inclined to regard this 
1 Benson : loc. cit. 
2 Treub : loc. cit. 
