1898 - 1902. No. 19.] STRAY CONTR1BUT. TO THE BOTANY OF N. DEVON. 13 



from the mainland by a strait in which the strong current keeps the water 

 open most of the year, or at least before the beginning of the breeding- 

 season. The distance from the mainland of North Devon to Castle 

 Island is somewhat less than two miles, to Devil's Island less than one 

 mile. Such short distances can, of course, be easily surmounted by 

 plant-migration, if, for the transporting of plant species, we should 

 attribute any greater importance to those means of conveyance which 

 are generally reckoned with for the stocking of islands, viz., wind, birds 

 and currents (including floating ice). 



We shall now see what inferences may be drawn from a comparison 

 of the plant life of these islands with that of some neighbouring points 

 of the larger islands. Firstly it is to be observed that the number of 

 flowering plants is nearly the same in both islands, viz., 11 on Castle 

 Island and 10 on Devil's Isle. Two are found in the former alone 

 (Draba hirta, Cerastium alpinum), one in the latter alone (Glyceria 

 angustata), nine are common. If now the vegetation of the nearest 

 points in North Devon is drawn into comparison, we will see that all 

 these species are found there except Glyceria angustata, which may 

 easily have been overlooked. The flora of Mount Belcher, as far as 

 known, reckons 17 species, that of Cape Vera 24, and that of the dry 

 limestone ness to the south-west 21. The nearest points in Ellesmere- 

 land Gull Cove and Falcon Cliff which have the same geological 

 nature, affording similar conditions of life as being breeding-places for 

 gulls and other birds, show at least 30 and 40 to 50 species respectively . 

 The list of higher plants found in North Kent reaches 33. 



Thus it appears that the two islands which are isolated by open 

 water, show a considerably poorer flora than that of the mainland 

 localities under similar conditions, notwithstanding the short distance 

 that separates them from the nearest land (the distance over to Gull 

 Cove also is small enough, about 11 miles). The relation of the flora 

 of either of these islands to that of Falcon Cliff will be about 1:4 or 

 5, to that of Gull Cove 1:3, to that of Cape Vera 2 : 5 or, more 

 probably, about 1 : 3, as without doubt several species have been over- 

 looked during the short visit to Cape Vera; whereas the flora of the islands 

 must be looked upon as thoroughly known, at least as far as flowering 

 plants are concerned. If we take the mosses for comparison we will 

 get similar figures. Of course it may have happened that I did not get 

 some species or other in my collections, notwithstanding that I carefully 

 tried to make my moss collections as complete as possible; but I think 

 we may look upon the numbers 31 for Castle Island and 35 for Devil's 



