FLOWERING season AND CLIMATE. 339 



As it is very difficult, even after a longer examination of this table, 

 to trace the different relations between the climatic factors and the 

 flowering period I add, in order to give a clearer idea, a graphic repre- 

 sentation of the table. (Plate I.) 



If we compare in the first place the meteorological curves with the 

 curve representing the flowering times of the woody plants, there is 

 one prominent feature which strikes us most, viz., the coincidence of 

 the maximum of rainfall in July with the minimum of flowering times 

 during the same month. The clouds show their maximum at the same 

 time, and humidity is just a little below its maximum in July, whilst 

 temperature, after having reached its maximum in May, is going down 

 as rapidly as it had risen since March. The flowering times reach 

 their maximum in March, and, at the same time, humidity, clouds, 

 and rainfall are almost at their minimum. The meteorological curves, 

 except that of temperature, are continually rising in April, May, June, 

 and July, during which period the curve of the flowering times is des- 

 cending. In August the number of the flowering times is increased, 

 whilst the clouds are diminished slightly, and the rainfall considerably. 

 In September, again, the flowering times are less, and, though rainfall 

 as well as clouds are descending, we find humidity at its maximum. 

 After the month of September, the curve of the flowering times is rising 

 decidedly, whilst all the meteorological curves are descending. From 

 our curves we are not able to decide whether any greater influence 

 upon the flowering time is to be attributed to temperature. For the 

 explanation of the fact that the maximum of the flowering times does 

 not coincide exactly with the minimum total of humidity, clouds, and 

 temperature (as this evidently is not in March but in February, 

 though there is only a very slight difference between the two months), 

 we might adduce the reason that the temperature in February is not 

 high enough for a full development in the sexual sphere. Perhaps the 

 statistics of other regions might contribute towards the solution of the 

 question as to the influence of temperature upon the flowering season 

 in the tropics. This much, for the present, may be taken for certain, 

 that to the maximum of rainfall, clouds, and humidity, there corre- 

 sponds the minimum of flowering times of the woody plants, and to the 

 minimum of the hydrometers the maximum of flowering periods. 



A comparison of the flowering times of the herbaceous plants with 

 the meteorological curves furnishes the following details : The minimum 

 11 



