OBSERVATIONS ON ANTS 117 



lowed by part recovery, then a second spell of weakness, more severe 

 than the first, violent twitching, then rolling upon the back. A 

 second injection of the same fluid resulted in death within half a 

 minute. 



From the above it is plain why the attacking ants leave the thorax 

 of the leaf-cutters untouched, as it contains material highly poisonous 

 to the smaller insects. 



Continuing the work of experiment, a solution was produced 

 from the entire bodies of the smaller ants. A drop of this fluid 

 administered to a wasp between the second and third pairs of legs 

 produced immediate paralysis of these organs. Another injection 

 given just in front of the forelegs produced the same result in that 

 pair. Now, if the small ants are closely watched during their attacks 

 upon the leaf-cutters, they will be seen to attach themselves first to the 

 unfortunate insect's legs. Here they cling for a few seconds before 

 gradually working their way towards the head of their prey. 



At length they arrive at this point and remain doggedly clinging to 

 the mandibles or mouth parts of the larger ant. From what we have 

 observed the following is a possible explanation of what takes place: 



The third experiment has shown that whatever part is injected with 

 the small ant's poison, becomes subject to paralysis in a very short 

 time. The smaller ant attacks the head of the leaf-cutter and the 

 latter, as we have observed, makes no effort to use its mandibles upon 

 its tiny persecutor. The smaller ant's course is plain. It attacks the 

 leaf-cutter's head; paralysis of the muscles governing the mandibles 

 and biting parts results and the larger insect can do nothing to protect 

 itself against a number of the smaller ones, who soon drag it to their 

 lair! 



Before closing this chapter on various phases of ant behavior, that 

 I have been able to record from time to time, it is fitting that I set 



