10 A/y /sr.A^Ds. 



supposcfl iiii,i;lit oooasionally \h'. iriuisportcd to un on 

 l)it« of tloatiiig trees or matted turf, torn by Hoods 

 from tli()s(3 prehistoric TiUsitaiiiaii or African forests. 

 No sucli hick was ours. Not a sin<,de terrestrial verte- 

 brate of any sort appeared upon our shores Ijefore the 

 advent of man with liis domestic animals, who played 

 havoc at once with my interesting experiment. 



It was quite otiierwise with the unobtrusive amall 

 deer of life — the snails, and beetles, and flies, and earth- 

 worms — and especially with the winged things : birds, 

 bats, and butterflies. In the very earliest days of my 

 islands' existence, indeed, a few stray feothered fowls 

 of the air were driven ashore here by violent storms, at 

 a time wlien vegetation had not yet begun to clothe the 

 naked pumice and volcanic rock ; but these, of course, 

 perished for want of food, as did also a few later 

 arrivals, wlio came under stress of weather at the period 

 when only ferns, lichens, and mosses had as yet obtained 

 a foothold on the young archipelago. Sea-birds, of 

 course, soon found out our rocks ; but as they live off fish 

 only, they contributed little more than rich beds of 

 guano to the permanent colonising of the islands. As 

 well as I can remember, the land-snails were the earliest 

 truly terrestrial casuals that managed to pick up a stray 

 livelihood in these first colonial days of the archipelago. 

 They came oftenest in the egg, sometimes clinging to 

 water-logged leaves cast up by storms, sometimes hiddeu 

 in the bark of floating driftwood, and sometimes 

 swimming free on the open ocean. In one case, as I 

 recall to myself well, a swallow, driven off from the Portu- 

 guese coast, a little before the Glacial period had begun to 



