132 THE BIOLOGY OF THE SEASONS 



tall, tough vegetation standing in a well-nigh perfect circle 

 around the edge of the clearing. The weeds had crowded up 

 as closely as they dared, and were held back from the for- 

 bidden grounds by the insects, whose energy and skill could 

 easily limit their bounds. Certainly, ants capable of such 

 work could readily have cleared away growing stalks of the 

 Aristida. In fact, after the seed has ripened in the late 

 summer, they are said to clear away the dry stalks in order to 

 make way for a new crop. It is this that j ust ifies the reputa- 

 tion of Barbatus as a farmer. She has not been seen so far 

 as the author knows sowing the seeds, but she permits 

 them to grow upon her formicary bounds, and afterwards 

 utilises the product." l 



In a recent careful study of Messor barbatus, a leaf- 

 cutting and seed-gathering ant of Dalmatia, Professor F. W. 

 Neger of Tharandt noted that most of the seeds (of Legum- 

 inosae in particular) were allowed to begin to germinate 

 before the ants put them out to dry. This seems discrepant 

 with what is often stated, that ants treat the seeds in such a 

 way that they cannot sprout. But Neger suggests that the 

 germination permitted has the advantage of bursting the 

 seed-coats. It is then stopped by exposure, so that it does 

 not go far enough to ferment the starch into maltose and 

 dextrin. When the seeds are thoroughly dry and dead, they 

 are taken back again to the nest and chewed into a dough. 

 This is baked in the sun into minute biscuits, which are 

 stored. Here, in fact, we have an industry that comes very 

 near to cooking. 



One of the most extraordinary habits of the termites or 

 white-ants, so abundant in warm countries, is that about 

 thirty different species feed on moulds which are grown 

 within the termitary on specially constructed maze-like beds 

 of chewed wood. The fungi are believed to afford a supply 



1 H. C. M'Cook, Nature's Craftsmen, 1907. 



