368 SEMI-CENTENNIAL OF TORREY BOTANICAL CLUB 
TABLES 5, 6, and 7 present data for the same wild white- 
flowered plant (A) for three successive years (1913, 1914, and 
1915). Тһе age of this plant when transplanted in 1911 from 
Wisconsin to the experimental garden was unknown. There has 
been no very marked difference in its general vigor and habit of 
growth in the five years that it has been under observation, except 
that in 1914 part of the roots which were inadvertently somewhat 
exposed were killed by winter freezing and there were fewer main 
branches produced from the cluster of roots. 
The performance of this plant (A) in each of the three succes- 
sive years shows a seasonal decrease that is quite characteristic of 
the species. There is also rather close agreement in the ranges of 
partial variability, these being 21-13, 22-15, 22-12. The average 
number per head and the standard deviation are also quite uniform 
as follows: 1913, 17.21.79; 1914, 17.81.41; 1915, 17.31.72. 
There is also rather close agreement in the values for the first date 
of blooming (a), but there was a considerable increase in the length 
of the flowering period, (2) in 1915 over that of the previous years. 
The rate of decrease (b) was lowest in 191 5. The significance of 
these facts and the means of proper comparison of such data will 
be discussed presently. 
The principal interannual partial variability is seen in the 
length of the blooming period. When this is considerably shorter, 
as in the year 1914, and the total amont of decrease remains much 
the same, the rate of decrease (b) is necessarily more marked. 
The wild white-flowered plant considered above was grown 
from roots obtained in the field and its growth and vigor were 
much more uniform in successive years than is the growth of 
plants grown from seed. The latter, as a rule, exhibit in the second 
year of growth a marked increase in general vigor as measured by 
the number and size of the main stems,w hich gives a corresponding 
extension of the flowering period with the production of more 
flowers. 
These aspects of interannual partial variability may be illus- 
trated by TABLES 8, 9, and то, which present the data collected 
from a plant in the first, second, and third year of growth. 
The records of this plant in the first, second, and third 
years of growth agree in giving lower numbers per head as the 
