1907.] Wheeler, Fungus-groiving Ay^ts uf North America. 795 



few Myrmicine genera, such as Cryptocerus, Atopomyrmex, Myrmecina, 

 Pheidologeton, Aeromyrma, Carebara, Lophomyrmex and certain species 

 of Tetramorivm. In my opinion no great taxonomic importance is to be 

 attached to this character; nevertheless its constant occurrence in the true 

 Attini must be taken into consideration." 



In a later paper (1895) Emery groups the genera above mentioned as 

 follows : 



Tribe Dacetii: Daceton, Acanthognathus, Orectognathus, Strumigenys, 

 Epitritus, Rhopalothrix, Cerafobasis. 



Tribe Attii: embracing besides the genera and subgenera enumerated 

 in the introduction to this paper, Wasmayinia and possibly also Ochetomyr- 

 mex. 



Tribe Cryptocerii: Procryptocerus and Cryptocerus. 

 Tribe. Cataulacii: Cataulacus. 



Emery is apparently of the opinion that the Attii are related to the 

 Tetrarnorii through such intermediate genera as Wasmannia and Ocheto- 

 myrmex, whereas Forel is inclined to seek their origin among the Dacetonii 

 through such a series of genera as Cyphomyrmex, Rhopalothrix and Strumi- 

 genys. Morphological considerations may be adduced in support of either 

 of these contentions. The question then naturally arises as to whether there 

 are in the Dacetonii or Tetramorii any ethological peculiarities which by 

 further development could lead to the highly specialized fungus-growing 

 habits of the Attii. 



Forel (1902) regards Cyphomyrmex as the most primitive genus of Attii 

 and believes that some of the species do not raise fungi, whereas the others 

 make very imperfect gardens on insect excrement. These ants would thus 

 be transitional in their habits to the Dacetonii, many of which also live in 

 damp places in rotten wood, where fungi grow in abundance and where 

 there is plenty of insect excrement that might gradually come to be em- 

 ployed as a substratum. In an earlier paper (1893) Forel quotes in support 

 of his view an observation of H. Smith on the West Indian Strumigenys 

 smithi Forel, a species which nests in rotten wood. Smith says that "the 

 cavities in which these ants are found are ahoays black inside, as if with some 

 fungoid growth." Forel infers from this that some species of Strumigenys 

 cultivate fungi. It seems to me, however, that his view evaporates into a 

 mere hypothesis when the facts are more closely scrutinized. In tlie first 

 place, there is no known species of Cyphomyrmex, nor in fact any Attiine 

 ant, which does not cuhivate fungi. I have sho\yn in the third part of the 

 present paper that statements to the contrary in regard to C. rimosus are 

 false and due to superficial observations. In the second place, there is not 

 a particle of evidence to prove that the Dacetonii cultivate fungi. The 



