2891.] D. Prain — The Vegetation of the Coco Ch'oup. 395 



The list is so short that ^ analysis of it is unnecessary ; it is suffi- 

 cient to note that the possibility of introduction from Malaya or from 

 Iiido-China is, so far as its evidence goes, evenly balanced. 



While the two lists probably include all the species usually intro- 

 duced by being attached externally to birds they do not exhaust all the 

 possibilities of the case. For, if the mud of a marsh may fix the seeds 

 or fruits of paludine species to the feet or head of wading-birds, other 

 substances may fix the seeds of forest species to the bodies of forest-birds. 

 There is almost no limit to the number of species that might be suggested 

 as introduced in this way, provided their seeds be sufficiently small ; this 

 very circumstance, combined with the necessarily hypothetical nature of 

 the subject, makes it impossible to attempt the suggestion of this mode 

 of dispersal in connection with any particular species.* 



The next kind of " bii"d-introduced " species to be considered — those 

 introduced in consequence of having been eaten — may also be con- 

 veniently divided into two sub-groups ; one consisting of species where 

 dissemination by birds is an every-day process, the other consisting of 

 species that can only be occasionally disseminated in this fashion since 

 the process implies the destruction of the bird itself. 



The first sub-group corresponds fairly closely with those species 



* The following facts will shew that, though necessarily hypothetical, the sub- 

 ject is not far-fetched but is, on the contrary, highly deserving of attention. When 

 in Narcoiidam the writer was particularly anxious to obtain the seeds of a species of 

 Bombax present there, for sowing at Calcutta ; for some days the search was hopeless 

 because the capsules as they ripen are broken open and the seeds are eaten by a 

 species of Horn-bill that is common in the island, while any seeds that escape the 

 birds and fall to the ground are devoured by the rats that swarm in the place. At 

 length under one tree, where there happened to be on the under-growth one or two 

 large spider's webs, four seeds were found sticking in these webs ; these were the 

 only seeds he was fortunate enough to obtain ; they were brought to Calcutta, 

 germinated there, and the four young trees are now alive in the Botanic Garden. 

 This will shew that seeds easily may, and at times do, stick in spider's webs. 



In spring 1890 a Barbet was found lying on the ground in the Botanic Garden 

 unable to fly ; on being picked up and examined it was found that its left wing and 

 left leg were fixed together by means of a spider's web; on freeing these it was found 

 that the toes of its left foot were further bound up in a ball and the flight-feathers 

 were firmly tied together. When finally completely freed from its entanglement the 

 bird flew away, frightened, but physically uninjured. This will shew that birds do 

 sometimes come in contact with spider's webs and that these are capable not merely 

 of fixing objects to a bird's feathers but of fixing these feathers so that the bird itself 

 cannot move them. 



All that is therefore required in order to establish the truth of the hypothesis 

 is direct observation of a bird having come in contact with a spider's web which 

 happened to have seeds lodged in it at the time, and of its carrying away seeds and 

 web together. 

 51 



