BIOPSYCHOLOGICAL 603 



hens learned to try only every second grain. Then the loose ones 

 were separated by two fixed ones, and the hens soon saw through 

 this. But when the loose grains were separated by three glued grains, 

 the hens gave it up. Yet they learned to take two loose ones in 

 succession, separated by a single fixed one. This did not mean that 

 the hens could count up to two, but not up to three ; it rather meant 

 that they could, so to speak, register a group of two as a unity, 

 but not a group of three! What seems to have been proved was 

 the hen's quickness in learning to suppress useless peckings; but 

 the experiments should be repeated. One would like to know from 

 those who have driven cars about the country for some years 

 whether there is any marked suppression of useless movements on 

 the part of the farmyard hens that straggle on to the roads. 



Some of Revesz's hens had very strong views on the greater pro- 

 fitableness of a ten-grain heap as compared with a six-grain heap, 

 and did not hesitate in preferring 3 to 2, 4 to 3, 5 to 4, and 6 to 5. 

 But this was probably due to volumetric, not numerical, impres- 

 sionism. And so when a brooding bird is uneasy over the theft of 

 three eggs out of six, the probability is that its perception is not much 

 more than a vague awareness that the picture has been disturbed. 



We shall not go back to the "thinking horses" of Elberfeld, 

 which could extract cube-roots, stamping out the answer with their 

 feet ; and we do not hold with the enthusiasts who believe that ants 

 and bees notice the fluctuations in their numbers as such, altering 

 their behaviour accordingly. In fact, we think there is little evidence 

 at present that animals can count more than a very little. For 

 counting requires counters, either words or symbols or tallies, and 

 to find more than the beginnings of these among animals requires a 

 generosity to which we cannot rise. 



Utilisation of the Fortuitous. — Before a chimpanzee hit upon 

 the device of piling one box on the top of another to reach the 

 high-hung fruit, it tried other methods, such as climbing, swinging 

 violently on the hanging rope, and standing on the shoulders of 

 another ape. As these all failed, it was driven to further experimenta- 

 tion. But all that was done was on the line of purposeful and 

 deliberate experiment towards an end in view. Yet there are other 

 cases where the intelligent factor operates by taking advantage of 

 a fortuitous occurrence, and we need not depreciate this too much, 

 since many of man's inventions have been made in a more or less 

 similar way. It is said that the Greek eagle lifts the Greek tortoise 

 in its talons to a great height, and then lets it fall on the rocks below, 

 with the result that the extremely strong carapace is broken and the 

 muscles exposed. This is an ingenious way of utilising an almost 

 invulnerable animal, but it may have been discovered quite for- 

 tuitously. Eagles often lift booty in their talons without stopping in 

 their wild rush, but they often let their captive fall. This would be 



