INFLUENCE OF BURROWING ANIMALS. 297 



The only personal observations I have been able to make in regard to 

 the depth of these excavations were along the line of a canal in the dia- 

 mond regions of Minas Geraes ; there the chambers or enlarged galleries 

 were found from 4 to 8 feet below the surface, the chambers being ar- 

 ranged in tiers or on floors one above the other. Sampaio gives one sec- 

 tion through a nest over 7 feet deep and another 11 J feet deep.* I have 

 frequently been told, however, of these galleries being as much as 12 feet 

 deep in the old sauba colonies, and that the ant-killers on coffee planta- 

 tions often open them out to this depth. 



The Reverend H. C. McCook has described the habits and abodes of a 

 Texas ant (Attafervens) which give some idea of the extent of ant burrows.f 

 This species makes its galleries as much as 12 feet in diameter, 15 feet 

 deep and 120 feet long. Belt tells of the leaf-cutters (GEcodoma') having 

 galleries 4 or 5 feet deep. J 



The Reverend J. G. Wood tells of an instance of saubas having " ruined 

 a gold mine for a time, breaking into it with a tunnel some 80 yards in 

 length and letting in a torrent of water, which broke down the machin- 

 ery and washed away all the supports, so that the mine had to be dug 

 afresh." § 



The chambers in the galleries examined by me were all found to con- 

 tain loose balls made of the fragments of leaves that had been carried in 

 by the ants. || 



The quantities of vegetable matter carried into their burrows is almost 

 beyond belief. I have seen a full-grown orange tree completely stripped 

 of its foliage within a few hours. In the coffee regions the damage done 

 by them is so serious that the Brazilian government at one time offered 

 a large premium for a saccess^ul formicida or ant-exterminator.^ 



I can vouch for the accuracy of the description given by Lund of the 

 work of these ants. The case is quoted here for the purpose of giving 



*Sauva ou Manhu-uara. A. G. de Azevedo Sampaio. Sao Paulo, 1894, pp. 22, 52, 64. 



fProc. Phil. Acad. Nat. Sci., 1879, pp. 33-40. Abstract in Annals and Magazine of Nat. Hist., 5th 

 ser.jVol. iii, pp. 442-449. 



J A Naturalist in Nicaragua, p. 76. 



§ Wanderings in South America. Charles Waterton. London, 1882. Explanatory index. Rev- 

 erend J. G. Wood, p. 47. Formigns sauvas, extracts from an article by Dr G. S. Capanema. Aux- 

 iliador da Industria Nacional, xliv. Rio de Janeiro, 1876, pp. 32-36. 



II On the injury done by ants to crops see the Revista Agricola do Imperial Instituto. Fluminense 

 de Agricultura, ix, 1, March, 1878, p. 21 ; xiv, 4, December, 1883, p. 215. O Auxiiiador da Industria 

 Nacional, xliv, Rio de Janeiro, 1876, p. 31. Tratado descriptive do Brazil em 1587. Obra de Gabriel 

 Scares de Souza. Revista do Inst. Hist, do Brazil, 1851, xiv, p. 271. See also ibidem, p. 163. 



l[Belt thinks the leaves carried into the ground by these ants are used to grow fungi for food 

 (Naturalist in Nicaragua, p. 80). The balls of leaves I have seen were all covered by fungi. Bur- 

 meister thinks the leaves are allowed to decay and are then fed to the larvse (Reise nach Brasilien» 

 p. 372). Barao de Capanema says the young ants are fed upon a white mould wliich is planted and 

 cultivated with care by the ants (Auxiiiador da Industria Nacional, xliv, 1876, p. 34). This has now 

 been established by Dr Moeller (Die Pilz-Gaerten einiger Sudamerikanischer Ameisen, Jena, 1893. 

 See also Boletim do Museu Paraense, i, no. 2, pp. 90, 91. Para, 1895). 



