SOME PHYSIOLOGICAL RESULTS OF TEMPERATURE VARIATIONS. 121 
tions. It then rises through March and April, falls in May, rising through 
June to its maximum in July. It falls again through August, September, to 
November, rising through December, January, and so on. 
Now, what can cause this ? 
One is struck by the fact that the number of conceptions is greater in the 
hot than in the cold months. The effect of temperature becomes more evident 
when the “Temperature Chart” (Chart I.) is referred to. 
It is then seen that when the temperature rises the number of conceptions 
increases, and that the temperature reaches a maximum in July, in which month 
the number of conceptions is also greatest. The temperature and conception 
curves then sink together to a minimum. 
On referring carefully to this Chart, it will be seen that, although the 
two curves correspond in their principal features, in detail they do not always 
| tally. 
In the first place, a unit increase of temperature does not always produce a 
corresponding increase in the number of conceptions ; thus, in the temperature 
curve, from August to September there is a fall of 4°, while the conception 
curve shows hardly any depression. Again, in three instances can be remarked 
a distinct deviation from the similarity. The number of conceptions in 
| December and January increases, while the temperature slightly descends ; 
/ and in May it descends while the temperature rises. It is seen then, that if 
'the temperature curve is a factor—and we have not distinctly proved that 
_yet—it is not the only one. 
We shall afterwards find that the temperature and marriage curves are the 
| great causes of the variations in the number of conceptions (with also other 
smaller factors). 
Influence of Temperature Curve. 
Although to the eye the temperature and conception curves so much agree, 
‘it is necessary to prove that the latter actually depends for its formation upon 
‘the former, and is not merely the result of some strange coincidence. 
Having proved that, it will be necessary to formulate, as a law, the tem- 
perature influence, and to state arithmetically the influence of an increase of 
a temperature unit on the number of conceptions. 
| By means of “ Fourter’s Harmonic Analysis” we can analyse a curve into 
its harmonic constituents. The temperature and conception curves may be 
analysed into curves recurring respectively every twelve months, six months, 
‘four months, and so on. By this method the phase is given, and by compar- 
‘ing the two curves thus analysed the dependence can be seen (if any). 
Thus, if it be found that there is one primary temperature curve (with 
‘maximum every twelve months), whose maximum falls exactly in the middle 
