466 Syncotylous Races. 
In the spring of 1894 I planted out into separate beds 
some syncotyls and amphi-syncotyls from my tricotylous 
intermediate race of Mercurialis annna, which produces a 
large number of syncotyls every year. There were three 
female plants and some male ones. Saved separately, their 
seeds produced 4, 19 and 24% syncotyls, amongst which 
hemi-syncotyls and amphicotyls appeared in considerable 
numbers. There is little doubt that I would have suc- 
ceeded in isolating the syncotylous intermediate races 
from Mercurialis (with 24%) as well as from Ccntran- 
thns (with 37%), and also perhaps from Picris (8%) 
and Valcriana (6%) ; but the difficulties, due mainly to 
the retarded growth of the plumule, have led me to limit 
my experiments in this direction to a single species, 
Helianthus animus. 
10. HELIANTHUS ANNUUS SYNCOTYLEUS.l 
As early as 1887 I found a syncotylous race of the 
ordinary sun-flower and isolated it immediately. Since 
then I have grown this race for ten generations. It has 
always been extremely variable, both in the extent of 
the fusion of the cotyledons and in the percentage num- 
ber of syncotylous seedlings. On the other hand it has 
proved perfectly constant and immutable. 
For instance, every effort to bring it to a uniform 
condition, free from atavists, or, on the other hand, to 
reduce it to the corresponding half race, has failed, in 
spite of the most stringent selection. Therefore we have 
here as good an instance of a constant type, which, in 
its visible properties merges into other races, i. e., is 
transgressively variable, as was furnished by the best 
1 Over de erfelykhcid van Synfisen. Kruidkundig Jaarboek, 1895. 
pp. 136-142 and Plate IV. See also the same journal, 1894. 
