516 BOTANICAL GAZETTE [JUNE 
strands of the single functioning cotyledon. In these cases, there- 
fore, one cotyledon is developed whether opposing strands unite 
or not. In monocotyledony, therefore, as shown by Agapanthus, 
the number of vascular strands in the single cotyledon is likely to 
be greater than it would have been if both cotyledons had devel- 
oped, for the cotyledon itself is larger. It is such cases that have 
suggested that monocotyledony has arisen by the suppression of 
one cotyledon. It is not so much the suppression of one cotyledon, 
as the growth of the whole cotyledonary zone to form a single 
cotyledon. In other words, in such a case a cotyledon is no more 
suppressed than are petals in Sympetalae. 
A further examination of the proembryos of Cyrtanthus san- 
guineus, reported upon by Miss FARRELL (loc. cit.), shows that in 
this species four primordia may appear upon the cotyledonary zone, 
which for a time develop equally. Then the whole zone becomes 
involved in the more rapid growth, giving rise to the cotyledonary 
ring or sheath, but with the four growing points still prominent. 
In the next stage the cells of the ring between a pair of growing 
points on each side become more active, and the four original grow- 
ing points begin to “grow together”’ in pairs, so that two cotyle- 
dons, each with two points, begin to appear. During this stage, 
whiclf may be called figuratively a “fusion in pairs,” the cotyle- 
donary sheath still continues to elongate. Later on one of the 
two cotyledons begins to develop faster than the other, resulting 
in two unequal cotyledons, which are connected at base by the 
thick ring. Gradually the cells of the smaller cotyledon cease 
dividing, and, those of the other continuing to divide, the result is 
a seemingly single terminal cotyledon. The developmental stages 
in this case are four cotyledons, two cotyledons, and finally one 
large cotyledon, associated with another one so small as to escape 
ordinary observation. The suggestion here of the possibility of 
polycotyledony in Cyrtanthus is plain, and the explanation of the 
polycotyledonous condition among certain Gymnosperms seems 
evident. 
In the current accounts of the embryogeny of Sagittaria and 
of other forms with filamentous proembryos, the development of 
the proembryo from the filamentous condition to the organization 
