289 
A Phcenological Study. 
1917. Of the 16 weeks up to April 21st, 1917 (No. 26 in 1919 
curve), only three showed a temperature above the normal, and all 
the 1917 plants were markedly later than the corresponding species 
for 1919, and still more markedly later than 1918. It will be noted 
that the Spring temperatures for 1919 are below normal, and on 
the whole only slightly above those of 1917; so that the lateness of 
the Spring flowers in 1917 in comparison with 1919 seems hardly 
explicable by temperature difference. 
Between the flowering dates of 1917 and 1918 the difference is 
much more striking. This is clearly shown on the curves. That 
for 1917 (x) running through April and May, while the 1918 
( + ) record shows the same species flowering in January, February 
and March, and one in April. Here the differences between the 
flowering dates agree with the differences between the 1917 and 1918 
temperatures. The 1919 curve (') shows a few flowerings in January, 
none in February, a few in March, and a considerable number of 
April readings. In fact a general inspection of the 1919 curve from 
January to the end of April shows clearly that it is intermediate 
between the curve of 1918, in which the flowerings are (on the 
whole) early, and that of 1917. But it must be repeated that the 
difference in temperature between the Spring of 1919 and that of 
1917 is not nearly as obvious as is the difference in flowering dates. 
The May readings of 1917 are obviously late, and not till 
No. 42 (Acer Psetuloplatanus) do we find a case in which the flower¬ 
ing date (May 9) is the same as that for 1919. And not till No. 51 
[A jiiga reptans) is the 1917 date earlier than in 1919. This is only 
a temporary difference, and the 1917 readings are equal to, or 
more commonly later, than 1919 dates up to No. 73 (Vicia sativa), 
May 23. Taking the later 1919 readings in May, from No. 75 to 
No. 95, we find the 1917 readings still markedly later than in 1919. 
One plant, No. 76 (A sperula odorata), flowers earlier in 1917, and two 
or three are synchronous with 1919. But the general result is:— 
the 1919 plants flowering near the end of May, and 1917 scattered 
through June; and this in spite of the temperature of 1917 being 
above the normal from the end of April to nearly the end of June. 
In the same way the 1917 dates for the plants flowering early in 
June, 1919, are on the whole later. 1 
From No. 130 to 149 the cases in which the 1917 flowerings 
are earlier than, or equal to, those of 1919 become more frequent, 
e.g., 135, 139, 140, 141, 142, 143, which are all equal to the 1919 
» Except for Nos. 101, 109, 120, 125, 129, 130. 
