306 Scientific Proceedings, Royal Dublin Society. 



made by birds in search of ants' "eggs." Such holes are seen plastered up 

 with clay by the ants. 



The ants are, perhaps, not absolutely indispensable to the plants. 

 B. rafflesiana has been cultivated in England, and was in a flourishing con- 

 dition two years after its importation, at which time no insects were present 

 in the piteliers. Some imperfect experiments of my own, however, tend to 

 sho iv that, under some conditions, the ants may be necessary to the plants. 

 On two occasions I transplanted plants of D. rafflesiana from the Jungle to 

 trees in my garden, within three miles of the place where the plants were 

 growing, and at about the same altitude. On the first occasion I brought 

 one plant, on the second two : in each case within five or six days all the 

 ants disappeared ; most of them, I believe, died, as their bodies were seen 

 lying- in the neighbourliood of the plants ; nevertheless, all the plants soon 

 started into growth ; about a month later I noticed that growth had stopped 

 and the pitchers were drying up, and within six months two of tlie plants 

 were completely dead, the third, wliich started growth by throwing qut a 

 twining shoot, was still alive, though half the pitchers had withered and 

 dropped off. On examining the remaining pitchers I found tliat numbers of 

 another ant, Cremasioffaster Rogenhoferi, Mayr, subsp. lierrii, Forel, had talien 

 up their abode in them. I do not wish, though, to lay too much stress on 

 these experiments, as other conditions may have been unfavourable to the 

 plants. 



We have seen that ants are constantly present under the leaves of D. num- 

 inutaria ; they have also been observed nesting under the concave leaves of 

 D. Col/.i/ris ; the ontogeny of J), rafflesiana leaves little reason to doubt that 

 it was evolved from a flat-leaved form like 1). niiiitmularia through a form 

 like I). Colli/ ris witli concave leaves ; therefore, I think we may conclude that 

 ants have been closely associated with D. rafflesiana through all stages of its 

 evolution, and have had a considerable share in the evolution of the pitcher. 



Observations on some other species of Dischidia. 



Dischidia singular is, Craib, is epiphytic on trees in evergreen jungle, cliiefly 

 on Qiiercus Junghuhnii, Miq. It is heterophyllous, having flat elliptical 

 leaves and narrow linear leaves with a small projection on either side of the 

 middle of the blade. The plant sends out long, slender shoots which usually 

 hang free, more rarely climb. These shoots may bear either kind of leaf, but 

 chiefly the narrow form; the elliptical leaf is more often found on young plants. 

 When I first saw this plant, I concluded it was parasitic, as it appeared to 



