20 NATURAL HISTORY READER. 



much divided, and end in delicate branchlets, as thin as 

 bristles, and very flexible, each bearing a minute double 

 hook at its tip. These are formed of a hard, woody sub- 

 stance, and are as sharp as needles ; a single tendril may 

 bear between ninety and a hundred of these beautiful little 

 grapplmg-hooks. The flexibility of the tendrils is of ser- 

 vice in allowing them to be blown about by a breath of 

 wind, and they can thus be made to seize hold of objects 

 which are out of roach of the ordinary revolving movements. 

 Many tendrils can only seize a stick by curling round it, 

 and this, even in the most sensitive tendril, must take a min- 

 ute or two ; but with G'ohma the sharp hooks catch hold of 

 little irregularities on the bark the moment the tendril 

 comes into contact with it, and afterward the tendril can 

 curl round and make the attachment permanent. 



12. The movement of the little hook-bearing branches 

 is very remarkable in this species. If a tendril catches an 

 object with one or two hooks, it is not contented, but tries 

 to attach the rest of them in the same way. Now, many 

 of the branches will chance to be so placed that their hooks 

 do not naturally catch, either because they come laterally, 

 or with their blunt backs against the wood, but after a 

 short time, by a process of twisting and adjusting, each 

 little hook becomes turned, so that its sharp point can get 

 a hold on the wood. 



13. The sharp hook on the tendrils of Coicea is only a 

 very perfect form of the bluntly curved tip which many 

 tendrils possess, and which serves the same purpose of tem- 

 porarily holding the object caught until the tendril can 

 curve over and make it secure. There is a curious proof 

 of the usefulness of even this blunt hook in the fact that 

 the tendril is only sensitive to a touch on the inside of the 

 hook. The tendril, when it comes against a twig, always 

 slips up it till the hook catches on it, so that it would be 

 of no use to be sensitive on the convex side. Some ten- 



