. ^ 



KUNADIYAPACAWITA MOUNTAIN, CEYLON. 



145 



rogafding each species, including- an analysis of tlieir distribution. This 

 catalogue clearlj indicates that the great majority oE the Kunadiyan flora 

 is endemic, while the base of the mountain is fully supplied with forms not 

 entirely confined to Ceylon, 



It is striking tliat while many of the (example?, in tlif* case of Kunadiya- 

 parawita, appear to be confined to a vertically narrow zone, or belt, of 

 distribution^ tlie same spucios in anol.hcr part of tlie island appears to l;e 

 more widely spi'ead. 



Two theories present themselves Eur consideration in the matter, that may 

 be best illustrated bv the following diagram : — 



r:-i^;::-!:;'i-;-i'. -':^:;:::■.■ 







I _^. J. 



' 3i 



^^\^- 



I I d 1 I 1 1 





~L I J I 11 



_i _r .1 



h'-I 



A, C, B, represents the section across a valley as it is at present^ but the 

 dotted line A^ C was possibly the outline at a very remote period till erosion 

 cut back the intervening formation to its present outline. The distance from 

 I to ]i, and from in to iv, is vertically the same, but^ svjicrJiciaUy^ iii to iv is 

 Inrger as regards the area on which plants could obtain a foothold. Equally, 

 owing to the angle from in to iv being lo\ver, the direct action of gravitation 

 on such fruits as are unequipped with special developmentsf or dispersal 

 must be more favourable to lodgment than if these fruits, or seeds, fell on 

 ground of a high :ingle of inclination. 



If this hypothesis bo correct, it seems to me that not only is the area i, ii 

 less, but gravitation would tend to reduce still further the areas occupied by 

 species on steep land, as opposed to land more gently sloping, for the steeper 

 the land the more rocky in general will it be^ and so the less chance of 

 foothold will it afEord. It would follow, therefore, that the proportion 

 of successful fruits would become less and less as the land became steeper 

 and steeper. 



If this theory is examined further, it appears to be reasonable to accept 



N 2 



.^. 



