The Isolation of Tricotylous Intermediate Races. 417 
each (1899). This year 77 plants gave a sufficient har- 
vest, the figures for the three groups (each derived from 
a single grandparent) did not exhibit any differences 
worth mentioning. The minimum was 2%, the mean 
12, and the highest value 25%. Therefore a consider- 
able advance on the preceding generation had taken place. 
In 1900 I had only 31 plants, bearing seed, which were 
cultivated in the same way as in the previous year. They 
constituted three groups, each from a single grandparent, 
but without exhibiting any differences worth considera- 
tion. The hereditary values varied between 5 and 17% 
and their mean was 10 to 13%. The culture of the last 
year (1901) embraced 40 plants, the values of which 
varied between and 21 and had a mean of 10%. 
The result, like that of the two foregoing species, 
may now be summarized in tabular form. 
VALUES DETERMINED IN THE FOLLOWING 
GENERA- YEAR OF SPRING 
TION FLOWERING 
MEAN HIGHEST SEED-PARENTS 
1. 1896 6 
II. 1897 4 12 12 
III. 1898 4-8 16 15-16 
IV. 1899 12 25 23-25 
V. 1900 10-13 16-17 14 
VI. 1901 10 21 
The prospect of raising an intermediate race seems 
therefore in this experiment to be as small as in Ainaran- 
tus and Scrophnlaria. 
6. THE ISOLATION OF TRICOTYLOUS INTERMEDIATE 
RACES. 
It is just as easy to isolate an intermediate race from 
seed which has been bought or obtained from any other 
source if it happens to be already present in it, as it is 
