CATKINS 87 



glad to visit its honey-bearing flowers in early spring, 

 when flowers are so few. 



How can we explain the different lengths of the 

 cones or spikes of female flowers? The Willow pro- 

 duces very many female flowers, as also does the 

 Birch ; Alder notably fewer ; Hazel often about eight 

 only, of which less than half usually ripen. The 

 difference becomes still more conspicuous if we com- 

 pare the number of the seeds produced. The Willow 

 may produce a million (I have not counted them) on 

 a single spike ; Birch fewer, but still very many ; 

 Alder perhaps a tenth of the number of the Birch ; 

 Hazel only two, three or four. The size of the seeds 

 is naturally in an inverse proportion to the number, 

 and the quantity of food stored up in the seeds will 

 vary almost directly with the size. What is the 

 meaning of these striking differences ? Willow, 

 Poplar and Birch seeds are 

 dispersed by the wind ; there- 

 fore they must be light, and 

 since many will be lost, it is 

 desirable that they should be 

 very numerous. I think that 

 the seeds of Alder are dis- 

 persed by running water, for 

 they spring up along the FlG . 3T .__ Ripe cone of I]irch 



KanlrQ rf ri\-f>rc anrl KrrrVc broken across to show the 



riVerS and DrOOkS. winged fruits. Magnified. 



Seeds so dispersed need not 



be very light, and it is probable that a smaller propor- 

 tion is wasted than in wind-dispersed seeds. The 

 Hazel-nut is eatable, and is carried off by Squirrels 

 or other animals. Here and there one is dropped, or 



