The Or igin of the Seed-Leaf in Monocotyledons, i 13 
F. imperialis has two cotyledonary traces also, and the transition 
follows the usual course until a very massive diarch root-stele has 
been formed. But the primary root in this species is stout and 
unusually long-lived. The diarch stele breaks up immediately after 
its formation into a polyarch stele with six or seven phloem and 
xylem groups. 
It is interesting to note in this connexion that the seedlings of 
Albuca Nelsoni show a tendency to root-reduction. In one of the 
three seedlings I examined, the transition takes place precisely as 
in Anemarrhena. The two cotyledonary traces form a tetrach root- 
stele after the same fashion. But in the other two specimens the 
root becomes triarch by union of two phloem groups and suppression 
of the intermediate xylem ray. The root-reduction now sterotyped 
in Nigella and Eranthis may have begun in a similar way. 
In conclusion, the hypothesis that the vascular system in the 
seedling of Anemarrhena represents the bicotyledonary system of a 
remote ancestor, gains very much in probability when we find that 
a similar vascular reduction has taken place in such a form as 
Eranthis hie mails, where the partial union of two cotyledons is 
undisputed. Moreover, if the hypothesis be admitted, the disputed 
case of Ranunculus Ficaria becomes clear No adequate discussion 
of its structure can be given here, but after careful examination I 
agree with M. Sterckx in considering the cotyledonary member found 
in this species as undoubtedly formed by the union of two cotyledons. 
To conceive of steps by which two separate cotyledons should 
gradually unite is easier than to imagine a single cotyledon splitting 
into two similar members, as suggested by Mr. H. L. Lyon. 1 
Nor is there any evidence of weight for the superior antiquity 
of Monocotyledons. I have regarded them for some time as 
specialized forms of an ancestor with two seed-leaves. The com¬ 
plete union of the cotyledons may very possibly be due to their common 
specialization as a sucking organ. It is true that all cotyledons 
begin life by absorbing nourishment from a food-body within the 
seed, but in dicotyledonous seedlings they commonly lay aside that 
function at an early period, even though they may never become 
assimilating organs. Among Monocotyledons on the cont ary the 
apex of the cotyledon often remains within the endosperm 
throughout the existence of both, a period which covers years in 
Palms and some other plants. Such a habit as this would naturally 
lead in course of time to the fusion of the cotyledons within 
the seed. Quarry Hill, Reigate, April 26th, 1902. 
1 H. L. Lyon, Minnesota Botanical Studies, 1901. pp. 643-655. 
