THE AFRICAN RUBBER TREE 181 



Rubber from 1st tapping . . 1 



,, ,, ^ti\.i ,, . . . -J 



(h) A tree was tapped by the full-herringbone system, 

 the vertical channel being 8 ft. long and the lateral in- 

 cisions 1 ft. apart. The lateral cuts were reopened on 

 alternate days by paring the lower surface in the manner 

 adopted for Para trees : 



Rubber from 1st tapping 



2nd „ 



5j >j 3rd ,, 



4tli „ 



,, ,, 5th ,, 



Total . 



1-5 

 0-5 

 0-25 



0-2 

 2^ 



No definite conclusions as to the best method of tapping 

 can be drawn from these experiments, but the results may 

 be summarised as follows : 



(1) The spiral system gave the best yield of rubber 



and then the method of long, vertical cuts. 



(2) There appeared to be no wound response, as the 



yield of rubber steadily diminishes with succes- 

 sive tappings. 



(3) The yield of rubber obtained from a tree after a 



year's interval was less than the original yield, 

 no matter whether the new tapping was made 

 on the same side of the tree or on the opposite 

 side. 



II. The trees which were tapped in 1909 for the first 

 time were retapped on the same area in 1910, but the 

 results were very poor, the amount of latex obtained 

 bemg only one-tenth of that yielded in 1909. 



III. Durmg 1911 tapping experiments on the Fun- 

 tumia trees in the Mamu Reserve were undertaken on a 

 larger scale : 5,456 trees were tapped, of which 2,039 

 were wild trees and the remainder cultivated trees. 

 The full-herringbone system was employed, the cultivated 

 trees being tapped by means of the Christy knife to a 

 height of 12 ft. and the wild trees by the native knife 



