64 HARVEST FNG ANTS. 



variety. It differs however in its smell, which, when the 

 body is crushed, resembles that of Pheidole Tnegacephala, 

 and is something like aniseed. Habits of structor and 

 barhara. Nest in earth. On one occasion I opened a large 

 nest at Cannes, where the colony was composed in about equal 

 parts of ants which in colour and appearance might be said 

 to represent the three forms, structor, barbara, and the red- 

 headed variety of the latter. There were also a few ants 

 with pale yellowish brown heads. (Mentone and Cannes.) 



B. 



The following Indian species are described by the late Dr. 

 Jerdon as harvesters, in the Madras Journal Lit. and Sc. 

 1851 :— 



(p. 45). Atta rufa. — "Its favourite food is dead insects 

 and other matter, but it also can-ies off seeds like the (Eco- 

 dorna, chaff," &c. &c. (p. 46). (Ecodoma providens. — " Their 

 common food I suspect to be animal matter, dead insects, 

 &c. &c., which at all events they take readily, but they also 

 carry off large quantities of seeds of various kinds, especially 

 light grass seeds, and more especially garden seeds, as every 

 gardener knows to his cost. They will take off cabbage, 

 celery, radish, carrot, and tomato seeds, and in some gardens, 

 unless the pots in which they are sown be suspended or 

 otherwise protected, the whole of the seeds sown will be re- 

 moved in one night. I have also had many packets of seeds 

 (especially lettuce) in my room completely emptied before I 

 was aware that the ants had discovered them. I do not 

 know, however, if they eat them or feed their larvae on them, 

 though for what other purpose they carry them otf I cannot 

 divine. I have often observed them bring the seeds outside 

 their holes, as recorded by Colonel Sykes, and this I think 

 generally at the close of the rainy season ; but in some cases 

 I had reason to beheve that it was merely the husks, of 

 which I have seen quite heaps, and that the ants did not take 

 them back to their nests. If any of the forementioned seeds 

 be sown out at once in a bed, most likely in the morning the 



