Another experiment was to determine to what extent he was able to enumerate. To this end I constructed a small square box and made a hole in one side of it. The box was cushioned inside so that the contents would not rattle. In the box were placed three marbles of the same size and color. The hole was just large enough for the monkey to withdraw his hand with one marble at a time. After letting him play with these for a while, putting them into the box and taking them out, I abstracted one of the marbles and left the other two for him to play with. On taking them out of the box, he missed the absent one, felt in the box for it, rose, and looked where he had been sitting. Again he put his hand into the box and looked at me as if to say he had lost something. Failing to find it, he soon became reconciled to the loss and began to play with the remaining two. When he had become quite content with these, I abstracted a second one. Thereupon he instituted search and was quite unwilling to proceed without finding the lost marbles. He put his hand into the box, evidently in the hope of finding them. He would not continue to play with the one. I restored one of the marbles, and when he discovered that I could find the lost marble, he appealed to me in each instance to assist him. Then with his little, dirty, black fingers he insisted upon opening my lips to see if it was concealed in my mouth--the place where monkeys usually conceal stolen goods. I repeated this experiment many times, until quite convinced of his ability to count three. Another marble was then added to the number and he was allowed to play with the four until he became familiar with that number. But when one was taken from the four he did not appear to be greatly impressed with the loss. At times he seemed to be in doubt, but he did not worry much about it, though seeming to realize that something was wrong. It is not to be supposed that monkeys have names for numerals, but they surely have a more or less distinct conception of plurality. The same fact is true of birds. It is said that all birds are able to count the eggs in their nests. This is certainly true of those that lay only three or four eggs. During the time that these experiments were being made with monkeys in this country, the late Professor Romanes was making certain experiments with a chimpanzee in London. He succeeded in teaching her to count seven, so that she would count and deliver to him on demand any number from one up to seven. This she did without prompting, and usually without mistake. Among different specimens of monkeys there seems to be a wide range of tastes. In this respect they vary much the same as human beings do. The same is true of their mental powers in general. With some monkeys the choice of color is much more definite and of dimension much more certain than in others, and most of them appear to assign to different numbers a difference of value.