86 
Ethel N. Thomas. 
( b.)—Distribution of Types in Relation to Habit of Plant. 
That the differences found in the anatomy of seedlings may 
be directly correlated with their habit very naturally suggests 
itself to one’s mind. There is no doubt that habit has played an 
important part in the evolution of these types, but operating as an 
ancient differentiating force, so that its influence is seen, for the 
most part, over a large area embracing several families, and not in 
isolated genera and species. A secondary effect of habit can 
however be seen in individual members of families, as for instance 
in Calycanthus, which conforms to the general Ranales type, but 
the diarch root is only formed at the base of a long stem-like 
hypocotyl. 
The intricate relation between habit and anatomy cannot be 
discussed here, but allowing that a certain broad correlation exists, 
one may consider whether this supports the case for progression 
or for reduction. 
On several grounds one may perhaps assume that the primitive 
Angiosperm is likely to have been a plant with a fairly large, robust, 
slow-growing seedling, and therefore probably with the tetrarch 
type of anatomy. From this arose the more slender, quick-growing, 
herbaceous plant with its early production of perfect foliage 
leaves, 1 and reduced seedling anatomy. The original type, however, 
probably persisted in some families, possibly more or less correlated 
with the tree habit. Thus the Leguminoseze, Rosaceae, Aceracese, 
Sapotaceae, Euphorbiaceae, etc., are largely composed of shrubby 
and tree-like forms. Looked at in this way, it would be easy to 
understand why some plants, such as Casuarina and Cassytha, 
might have escaped anatomical reduction, on account of the 
insignificance of the early plumular traces. On the other hand, 
it would be difficult to account for them on the habit hypothesis, if 
tetrarchy be taken as a progressive character, for the effect could 
not well precede the cause, although anatomical reduction might 
lag behind modification of habit of hypocotyl, particularly if other 
necessary co-operating factors were absent. 
(c.)— Internal Evidence. 
From a careful consideration of the types as seen in Angiosperms 
and Gymnosperms, it appears easier to imagine the derivation of all 
the modifications found from a tetrarch than from a diarch 
1 It is always the intercotyledonary poles—and therefore those 
opposite the main plumular traces—which are suppressed. 
