1895.] Antidromy in Plants. 977 
mode of folding of the embryos of the two seeds, produced by 
one flower, of Maple (A. platanoides L.). In Horse-chestnut the 
radicles of different seed incurve antidromically (a and c of fig. 
9), and the young leaves of the plumule (situated inside the 
radicle at p of fig. a, enlarged at fig. b) show the leaflets differ- 
ently arranged at the two 
sides, indicating the same 
primitive torsion as in 
other plants. The torsion 
of the plumule of ¢ would 
be antidromic as com- 
pared with that figured. 
That the place of origin z 
of the seeds is ordinarily Fig. 9. 
the determining cause Of Embryo and Plumule of Horse-chestnut. 
this character is proved by 
Corn, Coffee, Bean, Lepidium, and other seeds. In Gymnosperm 
the bilateral origin of the seeds, and the spiral arrangement of 
their numerous cotyledons point to the same inference, which 
is confirmed by the phyllotaxy, and by the primary spirals 
formed by the scales of their cones, as well as by the lateral 
bending of their woody tissues. We may ascribe to this cause 
the habit of splitting of tree-trunks in contrary spirals, and I 
think that the same tendency sometimes shows itself in the 
sculpturing of the cortex, so that from the bark of Chestnut and 
hard-barked Hickory I can infer the direction of the phyllo- 
taxy without seeing the leaves. 
` Direct evidence as to the Convulvulacee is difficult because 
of secondary distortions. But indirect evidence is available. 
Morning-glory has an incumbent curvature of the embryo 
as in many Crucifere, indicating such a diversity between the 
two seeds in a locule as produces in Crucifere antidromic 
phyllotaxy. This may also help to explain the twist of the 
embryo of Lepidium virginicum L. which has puzzled botanists, 
and if our sufmise is good, we may expect to find the embryos 
of two seeds of the same fruit antidromically twisted. 
After writing as above I examined the seeds from the two 
carpels of a flower of L. virginicum, and found them anti- 
67 
