Anatomy of Teratological Seedlings . I, 525 
bifurcated cotyledon or at the apex of the hypocotyl. This would be 
succeeded by tricotyly with the assumption of diarchy at the apex of the 
hypocotyl, the subsequent stages being provided by a progressively longer 
delay in the fusion of the vascular elements concerned until ultimately 
persistent triarchy is attained. Granted a similar shifting to an earlier stage 
of the apical quantitative division in phylogeny, the transition from dicotyly 
to polycotyly would be relatively simple. It will be obvious, moreover, 
that from such an initial stage a large range of possible developments is 
conceivable, since the process is capable of infinite repetition and may be 
augmented by a qualitative fission such as is illustrated by our type a 
cotyledons. 
A consideration of previous investigations will demonstrate that this 
type of cotyledonary bifurcation is of fairly frequent occurrence. Thus 
Lee ( 15 ) records an instance in Diniorphotheca pluvialis and in Bidenspilosa, 
and Guillaumin (8) in Schinus terebinthifolius. The seedlings of Acer plata - 
noides described by Leger ( 14 ) all seem to fall within this category, though 
the elaboration of the lateral bundle systems produces further complications, 
whilst of Compton’s material ( 4 ), hemitricotyls A and B, hemitricotyls C and 
F, and all the tricotyls of Cannabis sativa , tricotyls D and E of Lepidium 
sativum , and the single hemitricotyl of Ulex europaeus conform exactly. 
With regard to Hill and de Fraine’s Gymnosperm material it will be 
perceived that the four half cotyledons in series A, and the two half cotyle¬ 
dons in series B, Pinus australis , the four half cotyledons (C and D, G and 
H) of Pinus insignis , series B, the two half cotyledons of P. contorta 
var. Murrayana , series I, the two half cotyledons of P . montana var. 
gallica , series C, and the two half cotyledons (f and g) of P. sylvestris , 
series H, are also essentially similar. It is of extreme interest to note the 
behaviour of the median resin duct in the only two members of the above 
series figured by Hill and de Fraine, namely P. australis , series A, and 
P. sylvestris , series H, and contrast it with what obtains in those to which 
we have previously referred, as illustrating qualitative as distinct from 
quantitative cotylar division. Here it will be seen that each half cotyledon 
possesses a median resin duct flanked by the halves of a normal f double 
bundle ’. The resin ducts fuse at the apex of the hypocotyl, whilst the 
behaviour of the xylem and phloem resembles that described by us for 
similar cases in Cheiranthus. 
One further instance from this important series of papers calls for 
comment, namely, that of Araucaria Ctmninghamii , series A and B. It 
will be noted that Hill and de Fraine demonstrated that the appearance of 
tetracotyly is deceptive and that there are really only two deeply bifid cotyle¬ 
dons. The half cotyledons are each traversed by two collateral bundles 
which partially fuse near the apex of the hypocotyl. The four bundles thus 
produced pursue a completely independent course throughout the major 
Mm2 
