, 1910] SAXTON—WIDDRINGTONIA 41 
| appearance of which indicates that they probably arose as cleavage 
planes. In fig. 41 a transverse section of the upper four nuclei is 
shown. The presence of kinoplasmic radiations (spindle fibers ?) 
between these four nuclei indicates that probably they had a common 
origin and that after the first division of the oospore the basal nucleus 
remains undivided. It is evident that the early development of the 
proembryo does not show any resemblance whatever (after the first 
division) to that described by LAwson (5) for other Cupressineae 
(Thuja and Libocedrus), and by OTTLEY (9) for Juniperus, where 
eight free nuclei are organized before walls are laid down. 
Fig. 40 shows diagrammatically the structure of a ten-nucleate 
proembryo entirely filling the archegonium. It was not quite clear 
whether the walls extended to the upper three nuclei or not. The 
stages between figs. 39 and 4o are missing, and it is difficult to guess 
how the second form arose from the first. Probably there is some 
variation in the earlier divisions; compare, for instance, fig. 40 with 
the two proembryos figured in my preliminary account (10, fig. 16). 
Embryo development has already been briefly described. 
Fig. 42 is a drawing of a germinating seed before the cotyledons 
_,are withdrawn from the testa. Fig. 43 shows the upper part of the 
: same seedling after removal of the testa. Often the testa is carried 
up on the tip of one of the cotyledons in germination. The seedling 
_ structure (anatomy) of Widdringtonia cupressoides differs somewhat 
from that recently described by Hitt and DE FRAINE for other species 
(3), and an account of it has already been published from this labora- 
_ tory, together with a description of a remarkable twin seedling of 
_ the same species (Morris 8). 
In the plumular development a pair of opposite leaves succeeds 
the cotyledons, and is found in a plane perpendicular to that of the 
cotyledons. These leaves are followed by from about three to about 
_ ten alternating whorls of four leaves (text fig. 2). The structure and 
leaf arrangement of a tricotyledonous embryo have proved rather 
interesting. The three cotyledons are equal in size and have prob- 
ably been equivalent in development and are succeeded by alternating 
whorls of three primordial leaves. This is shown in the photograph 
(text fig. 3). The transition region is longer than in the normal 
seedling, and only at its lower end is there any indication that the 
