The cause of 



determined* 

 by variations 

 in the be- 

 haviourof 



pod. Iymfir 



Association 



without 



acters har " 



ioo STUDIES IN SEEDS AND FRUITS 



arrested at an earlier stage than in the case of the smaller, 

 dark-coloured seeds, as is indicated by their size and by their 

 higher percentage of water. But the behaviour of the smaller 

 impermeable seeds themselves when they begin to swell for 

 germination is equally suggestive, since, as soon as they begin 

 to absorb water, they assume the hue, size, and general 

 appearance of the larger impermeable seeds. 



The whole question is bound up with the history of the 

 shrinking process of seeds which is discussed in Chapter II, 

 and behind that lies the story of the maturation and drying 

 of the fruit. The immediate cause of this check to the 

 shrinking process of seeds resulting in the production of two 



types of normal resting seeds must therefore be looked for 

 /*~ 



m the conditions surrounding the final stage or the shrinking 

 process within the drying legume. In the case of Entada 

 polystachya the large pods, which are usually 15 or 16 inches 

 long and 3 inches wide, generally dry on the plant, and as 

 the outer skin scales off during this process they begin to 

 break up transversely into separate narrow joints, each con- 

 taining a seed. It is probable that the check to the shrinking 

 process of the seeds is determined by variations in the 

 behaviour of the drying pod with respect to the shedding 

 of its epidermis and the breaking up into joints. Un- 

 fortunately, this explanation of the matter only presented 

 itself to me after the opportunity of further investigation 

 had passed away. 



It is generally necessary, in the case of impermeable 

 leguminous seeds, that the last stage of the shrinking process 

 snou ^ ^ e completed in the pod. If one gathers a number 

 of full-sized so-called unripe seeds in the swollen, soft condi- 

 tion and allows them to go through the drying and shrinking 

 P rocess detached from the pod, they will usually fail to become 

 impermeable seeds. It is also generally necessary that the 

 seed should remain in the pod whilst it is dehiscing on the 

 plant, since, if we remove the seeds from the unopened 

 fruit, even though it is dry and the seeds are in appearance 



