504 AMERICAN JOURNAL OF BOTANY [Vol. 8, 
lie Opposite, or even a little in front of, the edges of the first leaf (fig. 23). 
With germination the basal sheathing portion of the cotyledon containing 
the vascular loops elongates greatly, so that the foliage leaves issue from a 
long tubular cotyledonary sheath (figs. 22," 24, 25). 
Discussion 
The first question which arises in the consideration of plants, for which 
there is reported so different a history of a conservative structure, is the 
degree of relationship of the forms. Cyrtanthus sanguineus (Lindl.) Hook, 
was the first described (6). It belongs to the subgenus Gastronema, which 
by some has been considered as of generic rank. However, the points that 
distinguish this subgenus are but slight and quantitative rather than quali- 
tative, the flowers being fewer, larger, and more erect than in the other 
subgenera. Cyrtanthus parviflorus was described by Baker in 1891 (i). 
In the Flora Capensis of Thiselton-Dyer he places it in the subgenus Monella 
which differs from the subgenus Cyrtanthus chiefly in having linear rather 
than lorate leaves (9). There is little from the systematic standpoint, 
therefore, to indicate that the forms are not closely related. 
The first paper touching upon the embryo of Cyrtanthus sanguineus was 
that of Miss Farrell in May, 1914 (4). In June of the same year Coulter 
and Land gave some additional details (2). The condition in Cyrtanthus 
sanguineus as interpreted by these writers is, briefly, as follows: At the top 
of the ovoid proembryo arise four swellings or growing points. These are 
soon followed by a ring of growth upon which they are raised up as upon a 
tube. Then they merge in twos by an active growth of the tissue between 
the pairs. Subsequently one duplex structure greatly outgrows the other, 
forming the prominent cotyledon, while the other, which does not shoot up, 
with the tubular portion constitutes the sheath. For her paper. Miss 
Farrell had available but two stages of embryo material, one showing the 
tubular stage with the swellings on the rim, the other a nearly mature 
embryo. This would seem a very small amount on which to base the im- 
portant conclusions at which she arrives. Coulter and Land appear to 
have had additional material, though they do not figure it. As a result of 
their observations on embryos, Coulter and Land conclude in part as 
follows : 
In the embryogeny of both monocotyledons and dicotyledons, a peripheral cotyledonary 
zone gives rise to two or more growing points, or primordia; this is followed by zonal 
development, resulting in a cotyledonary ring or sheath of varying length. If both growing 
points continue to develop equally, the dicotyledonous condition is attained; if one of the 
growing points ceases to develop, the continued growth of the whole cotyledonary zone is 
associated with that of the other growing point, and the monocotyledonous condition is 
attained. In like manner, polycotyledony is simply the appearance and continued develop- 
ment of more than two growing points on the cotyledonary ring. It follows that cotyledons 
are always lateral structures, arising from the peripheral zone developed at the top of the 
more or less massive proembryo (3). 
