30 The Irish Naturalist. March* 



without much difficulty bearing the retained fruits with it. 

 If this be what actually occurs then this plant has evidently 

 ev'olved two strings to its bow in the matter of dispersal, 

 a main one — wind, and an accessory one — animals. 



Although outwardly the two kinds of fruit are dissimilar 

 there is not really any very significant difference in their 

 microscopic structure, as will be seen by comparison 

 of figs. 3 and 4 on the accompanying Plate. Each 

 contains a single seed and although the ray fruit is 

 considerably larger than the disc fruit, the seeds themselves 

 are more or less of the same size. But the cavity in the 

 former is larger than that of the latter and the seed does 

 not so completely fill it. 



The principal difference between the two kinds of fruit 

 lies in the structure of their walls. In both cases the 

 main inner portion of the fruit wall is made up of a few 

 layers of corripact fibrous cells with highly refractive 

 thickened walls. In the brown wrinkled disc fruit this is 

 succeeded by a single layer of small thin-walled cells, and 

 this by the epidermis, a single la^^er of cells the brown walls 

 of which give the colour to the fruit. 



In the " peeled banana " ray fruit, however, the fibrous 

 cells are followed by a broader band of thin walled cells, 

 the width of which varies, being greatest at the sides and 

 back of the fruit. Outside of this comes the epidermis, 

 the cell walls of which are not coloured. The hairs arise 

 from the epidermis and are more plentiful on the concave 

 than the convex side. These points will be clear from 

 an examination of the photographs of transverse sections 

 of the two kinds of fruit reproduced on the accompanying 

 plate. 



Becker has shown that in some plants which are 

 heterocarpous, or which bear more than one type of seed, 

 the behaviour of the different kinds of fruits or seeds on 

 germination is dissimilar. To ascertain whether this was 

 the case in Picris echioides the few ray fruits and a 

 corresponding number of disc fruits from each of twenty 

 heads were placed under conditions favourable to germina- 

 tion. Fifty-nine fruits of each kind were thus treated. 

 Thirty-one of the disc fruits and twenty-nine of the ray 



