Seed-Scattering 257 



kept afloat for some time by means of air-bubbles. 

 Large fruits often float longer than small ones, and 

 could hardly be transported by any other means than 

 water. 



Then, however, comes the all-important question, 

 how far the seeds are affected by remaining for some 

 time in water ; and here again they vary considerably, 

 some being far more hardy than others. 



The coco-nut, for instance, being enclosed in a mass 

 of fibre, floats well, and is able to stand immersion in 

 either fresh or salt water for an unusually long time, 

 without losing the power of germinating ; and coco- 

 palms, self-planted, are the first trees to spring up 

 upon any newly exposed coral-reef, the nuts having 

 been floated thither from some more or less distant 

 coast. When making experiments to ascertain how 

 long seeds might remain in salt-water without being 

 killed, Mr. Darwin was delighted to find that some 

 grew after twenty-one days' immersion. Many ocean- 

 currents, as he reckoned, travel at the rate of a mile 

 an hour, so that these seeds might be floated five 

 hundred miles without being any the worse. But, 

 alas ! he had overlooked one thing. The seeds had 

 been under water all this time; and as Dr. Hooker 

 reminded him, ' If they sink, they won't float /' Seeds 

 vary much as to the length of time they are able to 

 remain afloat, and these seeds could not have been 

 transported at all by water, except under different 

 circumstances, such as while they were still enclosed 

 in their seed-vessels, or even attached to the plant or 

 branch on which they grew. 



Some few seeds grew after being kept for 137 days 

 in sea-water j so that, if able to float, they might have 



*7 



