168 



mon in the woodland districts, though both are to be found in almost 

 every kind of situation. 



Bates, who travelled in the Amazon, and Belt, who wrote the 

 ^' Naturalist in Nicaragua," are both credited with observations on 

 the habits of Atta cephalotes, the latter author being the first to note that 

 the ants CARRIED VEGETABLE MATTER INTO THEIR NESTS, 

 NOT AS FOOD, BUT AS A MATERIAL UPON WHICH THE 

 FOOD WAS GROWN, in the form of the conidia of a fungus. 



The matter however did not receive direct attention until some 

 years later. Early in 1892 I constructed a few artificial nests, and 

 commenced observations, and a resum^ of these was read at a meeting 

 of the Linnean Society in the latter part of 1892. Alfred Moller, a 

 German Naturalist, appears to have been studying these insects at or 

 about the same time, in the province of Santa Catherina, Brazil, in 

 the neighbourhood* of Blumenavi. It was clearly shown by these 

 observations that the food of the ant was really the fruiting portions 

 or conidia of the fungus in the nest, a fact which Belt only surmised. 



This was proved by observing the nurses take portions of the 

 food material and place it directly in the mouth of a larva, and the 

 material placed there was proved to be fungus by again removing it 

 and subjecting it to microscopical analysis. 



Ants in all the mature stages were also observed to feed directly 

 upon the conidia of the fungus, which is the part used by all as food. 



Moller, besides noting these facts, traced the origin of the fungus 

 a.nd found it to be the mature stage of Bozites gongylojjJiora Moller, 

 which is similar in form, though different in colour, to the common 

 edible mushroom Agaricus campestris. It is just possible however 

 that the fungus of our ants nests may not be the same species as that 

 observed by Moller ; and it is also possible that the species cultivated 

 by the different kinds of ant are singular to the nests of different 

 species ; as some recent observations show that the material used for 

 the growth of the " mushroom gardens," as they are called by Moller 

 differ in the nests of various species. It has also been seen that the 

 form of growth of the mycelium in some nests differs materially from 

 that found in others, which leads to the inference that the fungus 

 may be found to be specifically distinct on closer examination ; but of 

 course tintil the fungus of the nests of the various species is observed 

 in its mature form, it is not possible to determine this point with 

 exactness. 



