278 THE BIOLOGY OF THE SEASONS 



and gas-floats. In no other way could coco-nuts be carried 

 to isolated and uninhabited coral-islands, where they 

 sometimes form the beginning of terrestrial vegetation. 

 Writing of Keeling or Cocos Islands, coral formations 

 in the Indian Ocean, about six hundred miles distant 

 from the coast of Sumatra, Darwin called attention to 

 the number of seeds carried over from Sumatra and Java. 

 " It is interesting thus to discover how numerous the seeds 

 are, which, coming from several countries, are drifted over 

 the wide ocean. Professor Henslow tells me, he believes 

 that nearly all the plants which I brought from these 

 islands are common littoral species in the East Indian 

 Archipelago. From the direction, however, of the winds 

 and currents, it seems scarcely possible that they could 

 have come here in a direct line. If, as suggested with 

 much probability by Mr. Keating, they were first carried 

 toward the coast of New Holland, and thence drifted 

 back together with the productions of that country, the 

 seeds, before germinating, must have travelled between 

 1800 and 2400 miles." 



One of the most effective means of securing dispersal 

 is that which seems at first glance to be least propitious 

 that the fruit should be eaten. What seems, for a moment, 

 like a full stop, works well when it works at all, namely, 

 in cases where the seeds are not digested. Succulent 

 fruits are eaten by many birds and by a few mammals ; 

 they are eaten for their own sake, and the hard envelopes 

 of the seeds huthe case of berries, the hard endocarps of the 

 fruit in the case of drupes, save the seed from being digested. 

 It is passed out from the food-canal none the worse, in 

 some cases probably the better, often, naturally enough, 

 far from the place where the fruit was eaten. In this way 

 we can interpret the occurrence of an isolated gooseberry 

 bush far from any human dwelling. 



