596 Transactions of the Boyal Society of South Africa. 
Various stages in the development of the rudimentary embryos were 
observed (Figs. 35-39). While the seeds are immature, the intermediate 
layers of the ovarian wall, the cells of the placenta, and the suspensors are 
packed with starch. This disappears, and the supensors begin to disorganise 
before the seeds are shed. Suspensors are often seen reaching almost to the 
apex of the integument, but they do not project from the micropyle and 
penetrate the placental tissue as has been described in many Orchid 
species. 
Polyembryony is of very common occurrence in Pterygodimn Newdigatae, 
The two embryos are often of approximately equal size, though one may be 
markedly smaller than the other (Figs 40-42). In order to get some idea 
of the frequency of polyembryony in this species, ripe seeds were removed 
from various parts of mature ovaries and placed in alcohol to allow them to 
separate. The average of a number of counts showed that as many as 12 per 
cent, of the seeds of an ovary may exhibit polyembryony. Three embryos may 
occur within one seed coat, but this condition is not frequent (Fig. 43). It 
is possible that cases of two mother-cells in a single ovule, as figured in 
44, may account for some of the additional embryos. Again, instances 
were noted of an apparent segmentation of the embryonal mass, but the 
detailed explanation of the origin of polyembryony in this Orchid must 
await the examination of more suitable material. 
The dehiscence of the clavate fruit is unlike any of those mentioned by 
Pfitzer in his discussion of this order in Die naturlichen PJlanzenfamilien, 
Part II, sect. vi. The capsule splits by three longitudinal fissures into three 
broad and one narrow valve, which remain connected at base and apex. The 
narrow valve consists of the midrib of the carpel below the odd sepal ; the 
two broad valves lying on either side are each composed of two united half 
carpels together with a carpellary midrib (Figs. 45, 46). 
An attempt is being made to grow this Orchid from seed, in the hope 
that some light may be thrown upon the relationship existing between the 
open and closed flowered forms. Seeds dusted over the outer surface of 
unglazed porcelain cylinders kept constantly moist, were found to be appreci- 
ably swollen less than three weeks later. The seed coats were split, and 
delicate unicellular hairs had emerged from certain of the superficial cells of 
the embryos (Fig. 47). In older seedlings these hairs had increased con- 
siderably in length, some being three times as long as the testa. Where two 
embryos occurred within a single integument, both, as a rule, showed signs 
of germination (Fig. 48). The seeds were collected in November, 1914, but 
seed which had been kept dry for considerably over a year failed to germinate. 
Several distinct species of fungi occur in profusion among the germinating 
seeds (Fig. 49). Whether the seedlings will complete their development, 
thus indicating the presence of the necessary mycorrhizal fungus, remains 
to be seen. 
