416 BIOLOGIC RELATIONS BETWEEN PLANTS AND ANTS. 



supposed that moisture penetrated to the interior of the grain. This 

 hypothesis is, however, very doubtful. When the seed became suffi- 

 ciently softened and ready to germinate, it was thought that the ants 

 raised the micropylar seal and germination ensued. Since a more 

 advanced growth, says Moggridge, would alter the nutritive qualities 

 of their store, they hasten to gnaw off the tip of the radicle. Having 

 effected this mutilation, they dry the seeds in the sun, then store them 

 up anew. If by chance the seeds are moistened by rain, they are dried 

 in the same way. 



When the entire grain becomes soft and swollen, the ants devour its 

 soft parts, which are charged with saccharine substances, of which they 

 are very fond, and which serve to nourish their larvae. The envelopes, 

 in the form of bran, are rejected and carried outside to the rubbish 

 heap. 



The ants then know how to malt the seed of graminaceous plants, 

 obtaining real malt as our brewers do. We need not discuss here 

 whether they pursue this industry by instinct alone or whether they 

 have acquired their skill by experience. 



By using artificial light Moggridge was able to see how the ants gnaw 

 the seed. One of them grasps firmly a portion of the fariuaceous albu- 

 men while two or three others attack the seed with their mandibles, 

 feed upon it, and finally yield their places to others. Thus we see that, 

 unlike other ants that are nourished only by soft or liquid substances, 

 the harvester ants attack solid substances, rejecting, however, those 

 seeds which have never been softened, aud attacking only those which 

 have been dried after having begun to germinate. The hard envelope of 

 hemp and similar seeds generally resists the attack of the ants, so that 

 these insects wait until these envelopes are softened and burst by ger- 

 mination before they devour the oily contents. Although the buccal 

 apparatus of ants is not suitable for the mastication of hard bodies, it 

 answers perfectly well, when assisted by the hard, toothed mandible, for 

 scraping or scratching small particles of farinaceous matter. 



These ants have also carnivorous habits, and frequently pillage neigh- 

 boring nests, but a consideration of this would take us too far from our 

 topic. Harvester ants exist also in the Tropics. As early as the first 

 of this century Sykes, and then Gordon, noted in India a harvester ant 

 (Pheidole providens). 



Certain ants are not only harvesters, they are also farmers. These 

 American species of the genus Pogonomyrtnex were studied throughout 

 ten consecutive years by Dr. Lincecum and his daughter. The observa- 

 tions of those skillful observers have been published by Darwin. These 

 large, brown ants bore a hole in the ground, about which they heap up 

 earth to a height of from 3 to G inches, forming a low, circular mound, 

 rising by a gentle declivity from its center to its outer border. If, how- 

 ever, the ant is building in low, flat, wet land, subject to inundation, it 

 elevates the mound like a pointed cone to a height of 15 or 20 inches. 



