PERMEABILITY AND CLASSIFICATION 107 



ings, my observations clearly indicated that the seeds acquire 

 their impervious character in the closed pod. 



Any injury of the green pod of Guilandina bonducella is 

 likely to produce a defect in the impermeability of the resting 

 seed by interrupting its shrinking process in the pre-resting 

 state. Very premature dehiscence results in excessive shrivel- 

 ling of the seed ; whilst dehiscence at a later stage, but before 

 the seed has completed its normal shrinking, arrests the process 

 and prevents the development of impermeability, though not 

 depriving it of its power of germination. 



Mould or mildew, by attacking the soft coats of the unripe Mould and 



leguminous seed, spoils its prospect of becoming impermeable abSity. 



by partially destroying the cuticle. At times it extends into 



the kernel, and the result is a rotting embryo ; but at other 



times the outcome of its attacks is a permeable resting seed, 



speckled in places where the cuticle is lacking ; and unless 



conditions favouring germination soon arise, the seed shrinks 



more and more and finally loses its vitality. These minute 



fungi take their place amongst the greatest foes of a seed's 



impermeability in the tropics. As illustrated by Entada scandens, 



this subject is discussed at length in Note 7 of the Appendix, 



and it has been already referred to at the close of Chapter IV. 



I may here remark that Dr Gola in Italy made a pro- DrGola's 

 1 , . 1 • r 1 11 f • \i experiments, 



longed experiment on the action or these small tungi on the 



impermeability of the seeds of four plants, of which one was 

 Acacia Farnesiana. After exposing them to moist conditions 

 for four months, after which they were found to be covered 

 with mould, he ascertained that the percentage of permeable 

 seeds was not much increased. Thus, in the case of Acacia 

 Farnesiana the percentage of permeable seeds was only increased 

 from 3 to 5 per cent. These indications are not opposed to 

 my remarks in the preceding paragraph, since Dr Gola ex- 

 perimented on the hard matured resting seed, which would 

 be very likely to resist the attacks of minute fungi. It is in 

 the soft coats of the pre-resting or unripe seed that they find 

 their opportunity. 



