Anatomy of Teratological Seedlings . IV. 465 
respectively. The fourth gap heralds the formation of a new protoxylem 
pole which arises laterally and quite independently of the others. It is evident 
that the tetrarch condition of the main root which characterizes both these 
seedlings is evolved in a manner which differs fundamentally from that of the 
normal type. The question of the origin of the root-poles in the abnormal 
seedlings as a whole will be considered in detail at a later stage, but the 
chief point for immediate consideration is that provided by the behaviour of 
the united median strands, which lose all trace of their double origin and act 
in a manner precisely similar to a normal midrib, forming between them 
a single root-pole. Dr. E. N. Miles Thomas (31) has recorded a somewhat 
similar case in an abnormal specimen of Anemone pidsatilla , the two fused 
midribs of which are continuous with one root-pole of a diarch root, the 
other being in the plane of the first plumular leaf. The same author, jointly 
with Miss Davey (32), has investigated the seedling anatomy of a number 
of Pseudo-Monocotyledons, though their re- 
sults have been up to the present .only 
published in abstract. Material of Anemone 
apennina and Ranuncidns Ficaria , in which 
the single member is more or less bifid, and 
of Co7iopodium denudatum , in which it is 
undivided, was among that examined. In 
all the root is diarch, the xylem plane 
passing through the centre of the first 
plumular leaf, a feature which they share 
with the seedling of Anemone pidsatilla de- 
scribed above, and in which they differ from 
their normal dicotyledonous relatives where 
the diarch plate is at right angles to this 
plane. In Ranunculus Ficaria root structure is initiated below the cotyle- 
donary node, the vascular supply of the first epicotyledonary leaf apparently 
contributing largely to one root-pole. In Conopodium denudatum and 
Anemone apennina , on the other hand, root structure is present from the 
lower half of the cotyledonary ‘petiole ’ downwards, and inferentially, only the 
cotyledonary vascular strands connect directly with the root-poles. It will 
be recognized that the two latter forms bear a fairly close resemblance to 
the ‘ Type 2 ’ seedlings of Impatiens Roylei described here, whilst Ranuncidus 
Ficaria is nearer in type to certain Monocotyledons such as Allium and 
Triglochin . 
Other syncotylous specimens of I. Roylei showing a linking of the ad- 
jacent half-phloems of the ‘..double 5 bundle and occasionally forms with 
a temporary linking of the xylem also were met with, but these showed the 
usual triarchy in the hypocotyl with its accompanying three lateral roots. 
It seems reasonable from the evidence yielded by Seedlings A and B 
