362 
Tricotylows Races. 
a seedling of Acer Psciido-Platanits with four seed-leaves 
which I found in the spring of 1887 in a forest and trans- 
planted to my garden. Here it developed its stem. As 
soon as this had definitely split, and just before the cotyl- 
edons were about to fall away 
I dried this specimen in order 
to keep and photograph it. 
It is obvious that such twins 
do not belong to the tricotyl- 
ous race, that is to say, that 
their anomaly is due to some 
other elementary character. 
Therefore they should not be 
counted when recording the 
seedlings, nor be used as seed- 
parents. But as their nature 
can only be determined for cer- 
tain in some cases, it is not al- 
ways possible to take this pre- 
caution ; and the fact that the 
hereditary values obtained from 
tetracotylous individuals are 
sometimes worse than those 
from the corresponding trico- 
tyls may in part be due to this 
circumstance. 
For the rest tetracotyls do 
not in any essential respect be- 
have differently in inheritance 
from the tricotyls. From their 
seeds are produced, besides the atavists, mainly tricotyls, 
with hemi-tricotyls and tetracotyls in the usual dimin- 
ishing proportions. My tricotylous half race of Scrophu- 
Fig. 72. Amarantus specio- 
sus. Forking of a stem 
of a tetracotylous plant 
with two leaves in the 
fork which have grown 
together dorsally. The 
figure shows also their 
axillary twigs. 
