484 Holden and Daniels. — Observations on the 
Dicotyledons, and Compton ( 6 ), whilst apparently favouring the derivation 
of the A nemarrhena type through amphisyncotyly, considers that the exal- 
buminous Helobieae may have arisen by unilateral syncotyly. It is note- 
worthy that in Impatiens Roylei , which possesses an exalbuminous seed with 
a perfectly straight embryo, both amphisyncotyly and unilateral syncotyly 
occur, the latter predominating. The evidence as to the origin of the ‘Type 2 ’ 
seedlings of Impatiens Roylei brought forward in the present paper points 
strongly to their being derived from a dicotyledonous type by a close 
unilateral syncotyly, yet a study of a typical member of this group shows no 
sign whatever of the dual origin of the cotyledonary structure. If this 
evidence is accepted, then we are bound to admit that a midrib phyletically 
the product of two ‘ double 5 bundles behaves throughout in a manner 
identical with that of one of the ancestral midribs. The evidence put 
forward by Thomas and Davey (32) in the cases of Anemone apennina , 
Ranunctdus Ficaria , and Conopodium denudatum , as far as can be judged 
from the published abstract, also seems to point in the same direction. 
A further point in favour of unilateral syncotyly, and one again the 
significance of which Davey and Thomas seem to recognize, lies in the 
position of the first epicotyledonary leaf relative to the median plane of 
the cotyledons. It is extremely rare to find the first epicotyledonary leaf, 
in Dicotyledons, lying in any other plane than that at right angles to the 
median cotyledonary plane. The position of the first leaf in Monocotyledons, 
granted unilateral syncotyly, is strictly conformable with that typical with 
the vast majority of Dicotyledons. On the other hand, unless it is assumed 
that in the Monocotyledons there has been a shifting through 90 ° of the 
position of this leaf, it must remain a stumbling-block to all theories in 
which differentiation of cotylar function or partial or total suppression of 
the second cotyledon is considered as the source of Monocotyledony. The 
support of the theory of the differentiation of cotyledonary function is 
largely due to A. W. Hill (14), and has resulted from his investigation of 
the seedling anatomy of a number of geophilous species of Peperomia which 
yield a series of forms commencing with a normal dicotylous type and 
ending in one in which one cotyledon is purely suctorial in function whilst 
the other is purely assimilatory. Hill considers that monocotyledony has 
arisen as a result of specialization of this kind, the assimilatory cotyledon 
having delayed its development until carried outside the seed, when it 
appears as the first leaf. Hill regards the monocotyledonous seedling of the 
Araceae as conforming more nearly to the primitive condition, and would 
group Sargant’s A nemarrhena- Zygadenus series in the reverse order to that 
adopted by her. De Fraine ( 10 ) supports this view, considering that the 
continued separation of the components of the ‘ double ’ bundle into two 
independent portions would, on physiological grounds, be advantageous in 
an absorbing cotyledon and would thus lead to the condition charac- 
