THE SPHAEROBOLUS GUN 323 



did not entirely lose its elasticity ; for, when it was pushed upwards 

 from below by pressure exerted by a pin, it everted, although very 

 feebly. 



The alcohol and iodine solutions must have killed every cell 

 in the membranes subjected to them, yet the membranes behaved 

 in these solutions like those which had been immersed in 10 per 

 cent, potassium nitrate, i.e. they contracted but did not entirely 

 lose the elasticity of the semiglobose lower part ; for, on pressing the 

 membranes below in the manner already described, sudden e version 

 took place, albeit feebly. 



The results of the experiments just described seem to show that, 

 while the osmotic pressure in the palisade cells is of primary im- 

 portance in providing the force for the ejection of the Sphaerobolus 

 projectile, the elasticity of the rather thick white walls of the palisade 

 cells, and possibly also that of the walls of the hyphae of the fibrous 

 layer, is an additional factor in the process. 



If the cell-walls of the palisade layer and the fibrous layer were 

 hgnified and stiff, like those of wood cells, instead of being non- 

 lignified and highly elastic like those of collenchymatous cells, the 

 discharge mechanism of the Sphaerobolus gun — despite the high 

 osmotic pressure in the cells of the palisade layer — would not work. 

 We thus see that the successful operation of the discharge mechanism 

 depends not entirely upon osmotic pressure but also upon the 

 mechanical properties of cell- walls. 



We have seen that the inner everting set of membranes of the 

 stellately opened Sphaerobolus gun is attached to the outer non- 

 everting membranes solely by the tips of the 6-9 teeth, the two sets 

 of membranes being separated by an air-space everywhere else 

 including the tooth-sinuses (Fig. 161, A and B, p. 316). The attach- 

 ment of the two sets of membranes to one another at the tips of the 

 teeth is a very firm one, so firm indeed that, normally, it is not 

 broken when the projectile is suddenly discharged (cf. Tigs. 147, 

 p. 298, and 172, p. 361). Some idea of the mechanical strain 

 tending to separate the inner and outer sets of membranes from one 

 another at the moment when the projectile leaves the gun may 

 perhaps be gained when one considers that the inner set of 

 membranes (as will be shown in a later Section) turns inside out 



