704 CLIMBING PLANTS. 



often fuse at the places where they adjoin one another and increase in circumfer- 

 ence, frequently becoming as thick as a man's arm. The illustration on the next 

 page (fig. 168), taken from a photograph at Darjeeling in the Himalayas, shows 

 these stems, which look as if they had been actually tied on to the smooth trunks 

 of tall trees, and wliich bend away somewhat from the support, and then ramify 

 and develop abundant leafy branches. 



Many tropical species of fig, which may serve as representatives of a fifth type, 

 exhibit the following peculiarities : — their climbing roots, nestling to the sub- 

 stratum, flatten and spread out like a doughy plastic mass; the adjacent roots 

 fuse together, and in this way irregular lattice- works, or incrusting mantles, only 

 interrupted here and there by gaps, are formed, which lie on the supporting trunks 

 and are firmly fastened and cemented to them without fusing with it or deriving 

 nourishment from it. Frequently not the trunk only but the branches of a tree 

 serving as support are incrusted with the flattened clamping roots of the climber. 

 Sometimes the climbing Ficus sends columnar aerial roots down to the gi-ound, 

 whilst its leafy branches intersect those of the supporting tree; so complete is the 

 entanglement that at first sight it is hardly possible to distinguish what belongs 

 to the support and what to the climbing plant. Fig. 169 is a faithful reproduc- 

 tion of a sketch by Selleny drawn at Kondul, one of the small Nicobar Islands, 

 .showing one of these remarkable climbers with flattened roots incrusting the 

 support, i.e. Ficus Benjannina on a supporting uiyrtaceous tree, the latter obviously 

 sufiering under the burden of its oppressor, and already in a dying condition. 



These " tree constrictors", as one might call them, although they do not absorb 

 materials from their supports, as was formerly supposed, are certainly not inditi"erent 

 to them, and may injure and even kill them like the constricting, twining stems 

 described and figured on pp. 159 and 160. The entwined tree decays and its wood 

 disintegrates, perhaps termites assist in carrying away the remains of the dead 

 trunk, but the climbing stem with the flattened, climbing roots remains still 

 vigorous. It has meanwhile created a sufficient support for itself by its prop-like 

 aerial roots, and these prevent it from falling. As Hooker says in his Himalayan 

 Journals: — " We found great scandent trees twisting ai-ound the trunks of others 

 and strangling them: the latter gradually decay, leaving the sheath of climbers as one 

 of the most remarkable vegetable phenomena of these mountains". When at length 

 the climber, deprived of its original support, also dies, its roots and stem-structures 

 become bleached, and its curious forms, in which to speak with Martins, "the 

 excited imagination fancies it recognizes fantastic spectres and gigantic voracious 

 monsters", rise up weirdly against the dusky background of the primeval tropical 

 forest. 91 



The manner in which cHmbirg roots become fixed upon their supports is not 

 less varied than their manifold structural modifications. It has already been stated 

 that the climbing roots are light - avoiding, and that their growing points are 

 directed towards the rocky faces and boughless tree - trunks upon which they 

 climb. Should the distance between the stem and the wall be not great, the 



