170 Observation of the Origin of Varieties. 
doubt that the wild form exists in the mixtures grown 
in botanical gardens. But in order to strengthen this 
proof I have cultivated the isolated race for one more 
generation. For this purpose I selected three vigorous 
plants from amongst the 1893 crop whose terminal in- 
florescences had 12 ray-florets, and left them to be fer- 
tilized by themselves and by their like after all plants 
with 13 or more rays had been eradicated. From these 
ff 
three seed-parents I harvested the seed separately and 
raised three families, in 1894, on different beds. The 
rays of the terminal inflorescences were recorded, and the 
experiment brought to an end. 
I shall now give the results of these three counts 
made in 1894 together with that of 1893. It will be 
seen that the series of figures correspond with one an- 
other exactly ; at any rate as nearly as is necessary for 
the object of this experiment. The composition of the 
four cultures in the two generations was as much the 
same as we should expect four samples of an ordinary 
species to be. 
THE 13-RAYED RACE. 
VARIATION IX NUMBER OF RAYS, IN TWO GENERATIONS. 
Ray-florets 8 9 10 11 12 13 14 15 16 17 18] 19 20 21 
1893 2 1 7 13 94 25 7 7 1 2 3 
1894. First family 00 01 10 59 18 2341021 
Second (< 00 14 11 89 11 5002 100 
Third " 01 2 3 10 73 21 1200000 
Total, 1894 1 3 8 31 221 50 8 5 4 3 1 2 1 
The total for 1894 is given in the form of a curve 
in Fig. 31 A. The whole number of individuals dealt 
with in this year was 338. 1 
In order to isolate the 21 -rayed race out of the same 
1 For a detailed comparison of the curves of the two years see 
hh' fiir Entzvlckelungsmcch., II, 1895, toe. cit., p. 62. 
