hands of the most experienced hulero produces an enormous mass 
of woody debris which clings to the cuts, and subsequently are 
taken up with the latex. The extraordinary mass of wood and 
bark in the Centra! American rubbers is entirely due to the tapping 
being performed with the machete. There is certainly no difficulty 
of removing from the latex these particles of wood and bark, but it 
goes without saying that if it should be possible to avoid this or 
any other contamination at the outset, it is much to be preferred to 
any, however effectual, process of subsequently removing it from 
the latex. 
For this reason a narrow plane has been recommended for the 
tapping of the trees, but there are several drawbacks to it. It 
certainly gives a perfectly continuous cut, and one free from the 
above-named debris but it very soon gets clogged. The layer of 
bast to be cut through in order to obtain the full yield of latex is 
very considerable, and necessitates the blade of the plane protruding 
very considerably, a circumstance not calculated to facilitate its 
use. The thickness of the layer of bast to be cut through not only 
varies not inconsiderably from tree to tree, but it varies also in the 
same tree at different heights of the trunk. A plane gives under 
these conditions very little chance of adapting the depth of the cut 
to the depth of the layer of bast, and consequently, according to 
the setting of the plane either the layer of bast is not entirely 
penetrated, or the cuts pass more or less deeply into the wood 
itself. 
I therefore made experiments with a triangular cutting tool, the 
cutting edge of which is formed by one of the (rounded) angles, and 
after several modifications, I arrived at a form which answers the 
purpose admirably. To cut or tap the trees with this instrument 
requires some experience but certainly very much less than the 
?nachete. The cuts are absolutely clean and continuous and then- 
width is naturally regulated by the cutting angle and the depth of 
bast to be cut through. This is exactly as it should be. For trees 
of different age, or of different diameter, tools with different cut- 
ting angles may be employed. 
It has often been proposed to provide the instruments used for 
tapping with guards so as to prevent their penetrating into the 
wood. This is, no doubt, a very praiseworthy notion, but unfortu- 
nately one which it will be found practically impossible to adopt 
for the already stated reason that the thickness of the layer to be 
cut through varies within such wide limits as to render the proposal 
quite impracticable. 
The just described tapping tool does not, of course, prevent the 
penetration of the cuts into the wood, though I may at once state 
that its liability to do so is certainly much less than with any other 
instrument operated by striking such as the adze or machete. But 
in the experiments I concluded at Las Cascadas I gradually came 
to the conclusion, rather heretical in view of the nature of the 
numerous statements on this point, that the tapping to begin with 
has surprisingly little effect upon the well being of the tree, and 
further that cuts penetrating into the wood of Castilloa elastica, 
