SEEDLING STRUCTURE IN THE LEGUMINOS, 109 
subterranea, Castanospermum australe, Erythrina arborescens, Е. indica, and 
many specimens of Vicia Faba the cotyledons receive complex vascular 
supplies derived from two to four root-poles. Vicia Faba and Voandzeia 
subterranea show certain relations to a tetrareh type ; and the Erythrina spp. 
belong to a tribe which except for them and Voandzeia is entirely tetrarch. 
It thus appears possible that these peculiar types of structure are also derived 
from the tetrarch type; the increase in the number of root-poles being 
almost certainly correlated with the great bulk of the cotyledons in these 
species, which except for Erythrina indica are all hypogeal. 
Tug RELATIONSHIPS OF THE TYPES OF SYMMETRY. 
The result of the foregoing discussion has been to indieate that the stable 
tetrarch type, as seen in large epigeal seedlings, is probably fundamental 
and primitive *. The accompanying scheme (p. 110) illustrates the relation- 
ships of the different types to one another, and summarises the conclusions 
reached above. 
THe SIZE OF THE SEEDLING. 
We have seen in the preceding sections how important a factor is the size 
of the seedling in determining both the level of transition and the type of 
symmetry. We can hardly leave the subject here, but must enquire into the 
causes which determine the size of the seedling. 
Given uniform conditions in the seed with respect to the presence or 
absence of endosperm and perisperm, it appears that the size of the seedling 
depends directly on the size of the seed. In the Leguminosze, endosperm is 
for the most part absent, though in a few cases (e. у. Cassie) it is present in 
considerable amount, and in others a small quantity occurs. If we make the 
assumption that weight o£ the seed is proportional to its size (as is fairly 
legitimate within a single family with moderately uniform food-reserves), 
it is possible to get an indirect estimate of the size of the seedling by 
weighing the seed. 
Now in the Leguminose there seems to be a correlation between the tree- 
habit and the seedling structure. Of the 20 epigeal species of trees whose 
* The primitive nature of tetrarchy has also been supported by Miss Sargant, as a result 
of her studies of the Liliacem, &c. (1903). The Anemarrhena type, on which the question 
turns, appears to be an example of unstable tetrarchy (cf. Eranthis and Podophyllum, where 
diarchy is attained lower in the hypocotyl) with extreme reduction of the cotyledonary 
bundles: it cannot be regarded as primitive among the Dicotyledons, if we accept the view 
here advocated. 
A reduction hypothesis has also heen put forward by Mr. Tansley and Miss Thomas. 
(1906). 
