65 



perfect maturity, since I did not find any indication of 

 it in an unripe fruit. And as te» tlie conditions we must 

 admit tliat tlie tension between the numerous vascular 

 bundles in the periphery of tlie torus and the central 

 parenchymatous tissue becomes so strong that the slightest 

 cause sufflces for the tearing up of the weaker portions. 



There are two more points worthy of notice. 



In the first place the relative scarcity of bifurcation 

 in the summer of 1908 as compared with 1906 and 1907. 



The head-gardener, Mr. Verschoor, after being ques- 

 tioned on this point, attributed this striking différence to 

 the great drought of the raspberry-season, and I think 

 his opinion is the right one. For the fruits, of which I 

 examined several in détail, showed an extraordinarily 

 great number of undeveloped drupels and withered styles, 

 almost hidden between ripe drupels; it is obvions that 

 a defective supply of water prevents a normal and vigo- 

 rous growth and accordingly a powerful tension of the 

 tissues. The other point refers to the number of sepals 

 and its influence upon the phenomenon in question. 



In my former paper in which I could only appeal to 

 eight spécimens. I pointed out that in five of them the 

 number of sepals had been augmented. This fact, which 

 I thought to be in some way connected with the split- 

 ting up of the torus, induced me to look, in 1908, espe- 

 cially for flowers with supernumerary sepals, which it 

 was very easy to find. A close examination of the extra 

 sepals has convinced me that they had arisen not so 

 much from augmentation as from the sepalodic character 

 of some of the pelais. In this way 1, 2, 3, 4, and even 

 5 petals may be affected. The last case, viz. of the whole 

 corolla being sepalodic, may hâve induced Ch. Ferrand to 

 describe a double calyx and to compare it to the caly- 

 culus of the Potentilleae. 



