308 
Variation in Anemone nemorosa 
the B distribution of the previous year more nearly than the A distribution, the number of six- 
sepal flowers having risen from 515 to G14 per thousand. C has not been similarly affected 
at all, the distribution for April 8th — 12th, 1899, being very like the distribution for April 
21st— 22nd, 1898. 
TABLE I. 
Frequencies of specimens of Anemone neniOTOsa witJi different numbers of 
sepals gathered in different j)laces in the years 1898 — 1900. 
Year 
1898 
1899 
1900 
Place 
A 
B 
C 
C 
A 
C 
G 
Column 
1 
2 
3 
4 
5 
6 
7 
Date 
April 20-23 
April 21-23 
April 21, 22 
May 7 
April 9 
April 8-12 
April 15 
6 
CO fy 
° 8 
1 9 
a 10 
7 
515 
414 
49 
13 
1 
1 
3 
31 
657 
271 
35 
2 
1 
12 
448 
363 
135 
33 
5 
4 
34 
576 
276 
92 
14 
4 
4 
20 
614 
306 
44 
14 
2 
2 
28 
460 
390 
94 
24 
2 
6 
380 
448 
138 
24 
4 
Total 
Number gathered 
1000 
1000 
1000 
1000 
1000 
1000 
1000 
500 
1000 
500 
1000 
500 
1000 
500 
Mean number of 
sepals 
S.D. of sepals ... 
6-55 
0-68 
6-31 
0-62 
6-76 
0-90 
6-51 
0-87 
6-42 
0-69 
6-63 
0-81 
6-81 
0-80 
On April 15th, 1900, C was visited for the fourth time, the distribution being given in 
Column 7. It will be seen that it has changed its character very considerably, set'e/i-sepal flowers 
being now more numerous than sixes. In the April gatherings of 1898 and 1899 the frequency of 
sixes was roughly 450 per 1000, and the probable error of this on a gathering of 500 blossoms is 
only 14 or 15, so it is very unlikely that the low figure noted in 1900 was a mere random deviation. 
But if o?ie local race can change its character as much as this from year to year, what stress 
can be laid on difierences between local races only noted at one time in one year ? The only 
point in which the race G has constantly difiered from A and B is its greater variability, as 
measured simply by the S.D. 
I regret I found it impossible to visit either A or B in 1900, and in 1901 I had no opportunity 
for observation at all. 
I cannot suggest any definite reason for the change in the G distribution in 1900. The low 
underwood of 1898 had grown to a height of six feet or more in 1900, thus rendering the ground 
more shady, and also screening the wind to some extent. This is an approach towards the 
conditions of place B, but the distribution is not at all like that of B for 1898. The growth of 
underwood in place A may, possibly, account for the change in that distribution between 1898 
and 1899, but there can be no certainty about such a conclusion. 
