The Origin of Chrysanthemum Segetum Plenum. 169 
were removed so early that there was no clanger of these 
15 being fertilized by them. These plants flowered abun- 
dantly from their lateral shoots but exhibited no tendency 
to form a curve with an apex at 21. They were there- 
fore sufficiently pure representatives of the supposed 
race. 
In September I harvested the seeds of the 13-rayed 
plants which I had spared, and sowed half of them in the 
789 10 II 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 
A B 
Fig. 31. A. Chrysanthemum segetum. B. Chrysanthemum 
segetum grandiflorum (after purification). Curves of 
the races after isolation. A, Curve of the 13-rayed race 
in 1894. B, Curve of the 2i-rayed race in 1897. The 
ordinates give the number of individuals with like 
number of ray-florets in the primary inflorescences of 
the individual plants. The numbers of ray-florets them- 
selves are given below the abscissa. 
following spring (1893). I raised 162 flowering indi- 
viduals, and recorded the numbers of rays on their ter- 
minal heads. The curve representing this generation 
was steep, monomorphic and symmetrical (see Fig. 31 A 
for 1894), and agrees satisfactorily with the curves, 
given above, for the plants from the wild locality (p. 167 
and Fig. 32, Vol. I, p. 152). Therefore there can be no 
