5 88 
Holden and Bexon.—On the Seedling 
lateral bundle only, producing a compound strand ( D ) the marginal (. E) 
remaining independent. From this cotyledon, therefore, a median proto- 
xylem strand ( Cpx. 2) enters the hypocotyl with a single collateral (C) on 
one side and two such strands (. D and E) on the other. A few small xylem 
elements become detached from the inner side of the marginal strand (E) 
and move away from the parent group until they occupy a position midway 
between strands D and E. Although at first they lie within the phloem of 
the main bundle they ultimately resemble a protoxylem group closely 
(Fig. 63, Px. 3) and during the development of root structure behave as such. 
This condition is maintained for some distance down the hypocotyl, the only 
change being the division of the metaxylem groups B, C, and E into 
B l , B 2 , C 1 , C 2 , E 1 , E 2 ; the metaxylem group A does not divide until con¬ 
siderably later and then becomes resolved into two very unequal portions- 
During the passage of these half-bundles down the hypocotyl a solitary, 
small protoxylem-like element appears and disappears between A and 
E and a similar one, at a slightly lower level, between B and C. Towards 
the base of the hypocotyl the protoxylem from the cotyledon showing 
abnormal transition (Cpx. 2) dies out and the metaxylem group C 1 divides 
into two. 
From this complex five root poles are organized as follows : 
1. A normal cotyledonary pole is formed from B 1 and the bulk of A. 
2. The remnant of A forms a second pole with E 2 . 
3. Strands E l and D concentrate on Px. 3 to form a third pole. 
4. Strand C 2 divides into two, half forming a fourth pole with B 2 and 
half forming a fifth pole with C 1 : the remaining part of C 1 , though 
persistent, takes no part in root-pole formation. 
It is worthy of note that the poles formed by the remnant of A with E 2 , 
and that of B 2 with C 2 , occupy the positions of the isolated protoxylem-like 
elements which appeared and disappeared higher up in the hypocotyl. 
Discussion. 
In considering the possible theoretical significance of the seedling 
anatomy of the sycamore it will perhaps be well to outline the ontogenetic 
succession which Chauveaud ( 3 ) has shown to be characteristic of a large 
number of angiosperm seedlings. This investigator has established the fact 
that the vascular elements pass through a series of phases which are remark¬ 
ably uniform and which he regards as constituting a phylogenetic sequence. 
He considers that each xylem group has its phylogenetic origin in a radial 
file of vessels developing centripetally, the phloem at this stage being repre¬ 
sented by a series of tangentially distributed elements occupying alternate 
radii to the xylem. This condition is termed the ‘ disposition alterne \ 
The xylem and phloem elements developed subsequently to those exhibiting 
