a 
4 Dr. G. D. H. Carpenter’s Experiments on 
that one could put down insects for the monkey (hereafter 
alluded to as M.) to see as they ran about. 
After the first few occasions the method which yielded the 
best results was soon found. If one wanted to test the 
degree of inedibility of an insect previously shown to be 
more or less distasteful, it was offered either before the 
monkey had had any food at all, or after some vegetable 
food but before any other insects. Refusal under such 
circumstances implies a high degree of inedibility. 
Having started the experiment with such insects, one 
next proceeded to “take the edge off” the monkey’s 
appetite before offering other insects whose edibility was to 
be ascertained. This I look wpon as most important, and I 
never undertook any experiment, after the first, without 
having at hand a large number of some insect which I 
knew M. would always eat greedily. I found Acridiids 
most useful for this, as numbers of one or other common 
species can usually be obtained. These will be alluded to 
as “* Staple food” or “‘ Staple Acridians.” 
It was also equally important to have some available 
at the end of the experiment, so that one could prove that 
M.’s refusal of some insect was not due to repletion. It will 
be seen later what a surprismg amount M. could eat. 
The preliminary feeding accomplished, species were then 
offered the edibility of which one desired to test, and the 
manner in which M. dealt with each was at once carefully 
noted in an abbreviated form, before the next species was 
offered. This I look upon as essential, for if one has a long 
series to test, one’s memory soon becomes confused. These 
short notes were copied out and elaborated as soon as 
possible afterwards, while the memory was still fresh. 
As regards the typically Procryplic insects, it does not 
much matter in what part of the experiment they are 
offered, since they are always eagerly eaten. An Aposematic 
species, or one whose edibility is doubtful, is best offered 
for the first time in the middle of an experiment, and for 
the second time nearer the beginning or at the very 
commencement. 
Finally, staple food is again given and the inedibility 
of insects previously refused is judged to be the greater 
the more staple food is eaten at the end ! 
Precautions should be taken that M. does not suffer 
from thirst, otherwise errors will be made, for he may then 
refuse insects which, when not thirsty, he might have 
te aaa 
