384 
with seeds of Nelumbium which on being opened, showed the 
embryos rich in chlorophyll, with their leaves in one seed folded 
round in the inverse order of those in another seed; besides this 
the insertion of the stamens in the flower of different individuals of 
Nelumbium form antidromic spirals. A special case among the 
Liliaceae is Convallaria majalis, whose two leaves are at the usual 
angle among the Monocotyledones of 120°, and as one of the 
leaves forms a sheath around the other, it is instructive to observe 
how in a bed of Lily-of-the-Valley, all the plants being regarded 
under the same orientation, the inner leaf in half the plants bends 
over 120° to the right, and in the other half to the left. Arthur 
K. Harrison of Lebanon Springs, N. Y., informs me that before he 
heard of my work he had taken note of the double phyllotaxy of 
Veratrum viride. ; 
With a little care we can make out the general law as applying 
to the Gymnosperms; both the phyllotaxy and the spirals of the 
cones of Coniferae show it; and it may be expected to manifest its 
presence in the relative position of the cotyledon, during germina- 
tion; also in the embryo and its suspensor in Cycas. (See figures 
in Engler & Prantl, 2: 1. 17.) 
I have not yet tried the opposite-leaved orders, further than 
the case of Caffea, and that of Bryophyllum, as insufficiently 
tested by the marginal buds on the leaves; if this last observa- 
tion be verified, it may reinforce the old doctrine that ovules are 
the homologues of such marginal buds. Acer platanoides shows 
it by the antidromic folding of the cotyledons of the seeds of ad- 
joining carpels; also Aesculus by the contrary curvature of its 
radicles and by its plumules. Nor have I tried the twiners 
and climbers, saving that I have partially succeeded with 
Morning-glory (Convolvulaceae). Its embryo resembles that of 
some Cruciferae in having the radicle folded “ incumbently’’ upo” 
the cotyledons ; and again (unlike the Cruciferae) its orientation 
is changed so that the radicle lies next the floral axis and the 
cotyledons lie towards the periphery. Now if we reduce the 
orientation back to that of Cruciferae, we shall find a very close 
correspondence of the embryonic structures in the two orders; 
and the same antidromic difference will be found in Convolvu-. 
laceae that is readily shown by the phyllotaxy of Cruciferae. 
