ANTS AS FOOD.

In the Amazon region some of the ants are even used by the Indians for food.

The head and thorax are the parts eaten, the abdomen being nipped off (at San Carlos I constantly see them eaten entire), and it is eaten uncooked. The taste to me is strong, fiery, and disagreeable, but those who have eaten the bachaco fried in turtle oil tell me it is quite palatable.1

2

Orton says the saúbas "are eaten by the Rio Negro Indians, and esteemed a luxury, while the Tapajos Tribes use them to season their mandioca sauce."

In the more thickly settled parts of Brazil the custom of eating these ants is either not practiced nowadays, or, if it is, it is not generally known. In the early history of the country, however, when the native Indians were much more abundant than they are now, the custom appears to have been common.

STRUCTURES ABOVE GROUND.

Origin of the structures.-The word "nests" frequently applied to the superficial structures of ants should not be understood to mean nests in the ordinary signification of the word. These structures sometimes contain the queens, eggs, and larvæ, but at other times these are kept in excavations below the surface.

The mounds made by the true ants all begin as small funnel-shaped ridges around the excavations started by individual females. The large mounds are the results of the work of many generations and of a vast number of individuals.

Without going into any detailed description of the habits of the ants, it is worth while to give, for those unfamiliar with their habits, a general idea of the methods followed by these ants in establishing new colonies and in increasing them. When the swarming or mating season of the saúba ant comes, the young females leave their homes and fly away. They seem to fly about very much at random-at least, I have rarely seen them going in any particular direction-and when they have been seen going together it was apparently due to the direction of the wind or the position of the sun at the time, rather than to any definite purpose on their part.

When the female alights after a flight of only a few minutes, she breaks off her wings and at once falls to work at excavating a burrow. All kinds of places are selected for these burrows. It does not appear that the selection is deliberate, but it seems to be determined by the accident of alighting from an aimless flight. Judging from the large number of individual females I have frequently seen in the air and on

1 Richard Spruce: Notes of a botanist on the Amazon and Andes, vol. 1, p. 484. London, 1908.

* James Orton: The Andes and the Amazon, 3d ed., p. 301. New York, 1876.

the ground at one time, the great majority of these young colonies must fail to survive. Often I have seen the young females so abundant that there must have been an individual to every square meter of land surface over areas of many hundreds of acres.

In some places where the new arrivals alight the mounds are already so thick that there is little or no room for new colonies, and it is probable that some of these young females must either be adopted into the old colonies or they are killed or die.1

It is evident from the nature of the case that where such a large number of new colonies is started most of them must perish from mere overcrowding, if for no other reason.

The excavation first made by a young female is small and simple, and the earth taken from it is heaped about the opening without any apparent order. Dr. Huber, in the paper just cited, states that at Para, in a colony started by a single female, the first workers appear at the end of 40 days. Shortly thereafter the queen, or founder of the colony, ceases to be an active worker, and all subsequent excavating is done by the constantly increasing number of workers. As the colonies increase in numbers more underground room is required, and the amount of earth excavated and carried to the surface increases proportionately. This earth is brought to the surface in the jaws of the workers in the form of small pellets which are thrown down apparently without any other object than to be rid of them. Sometimes they are heaped up in funnel-shaped pits; sometimes they are thrown out on the downhill side of the opening. At first these bits of earth form heaps of loose, incoherent material, but in time, and with rain and sunshine, it packs down until it is often as hard as an unbaked brick. As long as the colony is active and growing, additions are constantly being made to these accumulations, and these additions may be at any point over the sides or at the top. Passageways are either kept open through these heaps of earth or they are reexcavated. This is demonstrated by digging into the mounds, but it is evident without opening them, from the fact that the fresh material is brought out and spread over any and all parts of the surface.

Size of the mounds.—It might be inferred that there would be practically no limit to the size of the mounds built in this fashion, and I am not sure that there are any limits save those which may be imposed by certain physical conditions, such as the amount and distribution of the rains, the character of the soil, the area over which the necessary plants or food can be obtained, etc. Of course, the mounds are of different sizes according to their ages; but consid

1 Just how new colonies of saubas can be established by a single female is described by Dr. J. Huber in Biologisches Centralblatt, vol. 25, pp. 606-618, 624-635, and in the Boletim do Museu Goeldi, vol. 5, pp. 223-241. Para, 1907-8. Also in the Annual Report of the Smithsonian Institution for 1906, pp. 355–367.

ering only the largest and oldest ones made by a single species, and found in various different localities, it is noteworthy that there is a great difference in the sizes of the largest of them. Just what determines this variation I can not say positively, but the influences referred to above-that is, rainfall, character of soil, and vegetationnaturally suggest themselves as possible influences.

Nowhere do I remember to have seen more or larger ant hills than along Rio Utinga, in the diamond regions of the interior of the State of Bahia. From the town Riachao, down the river to the village of Pegas, the examples are big and abundant. In a few places they are so close together that, big and little, they appear to cover half of the ground. My notes, written on the spot, say "more than half of the ground." Such places, however, are exceptional. The distribution is always more or less irregular-bunched apparently on account of characteristics of soil or drainage, or for some other reason that does not appear. In some areas of from 10 to 20 acres the ant hills occupy from a fifth to a third of the ground, while over larger tracts they take up from one-eighth to a seventh of the ground. In height the mounds are often as much as 5 meters high, with bases 15 or 16 meters in diameter. In the forests these mounds are generally overgrown with young trees. On many of the big mounds I have seen trees more than 30 centimeters in diameter. At the village of Antonio Jose the people have planted pineapples upon the mounds.

At fazenda Bello Horizonte, about 18 kilometers north of the village of Pegas, the ant hills are so large and stand so thickly upon the ground that they form one of the most striking illustrations I have ever seen of the work of these insects. An area of some 30 acres or more is there covered with mounds resembling haycocks. They probably average 2 meters in height and a diameter of 4 or 5 meters at the base.

At a place called Ponte Nova, on Rio Utinga, 8 kilometers north of the village of Pegas, the ant hills are a remarkable feature of the landscape. To the east and northeast of the Protestant college the mounds cover the old fields. One of the accompanying photographs (fig. 2, pl. 1) and text figure 1 were made in this region.

Six kilometers north of the station one was found by measurement to be 1.8 meters high and 4.5+ meters wide at the base. This mound was not regarded by the people of the locality as anything unusual.

To the east of Serrinha several mounds were observed with a height of 3 meters and a diameter at the base of 10 meters. These mounds, therefore, contain each 78.5 cubic meters of earth.

Along the western half of the Bahia and Minas Railway, that starts from the coast near Caravellas, in the southern part of the State of

the vicinity of Urucu Station (kilometer 226) the mounds are so thick and so close together that the country looks like a field of gigantic potato hills.

Bahia, and runs west 376 kilometers into the State of Minas Geraes, ant hills are big and abundant. The newer ones are steeply conical, but with age they become more or less rounded and flattened. In

FIG. 1.-Ant hills on Rio Utinga, near the village of Pegas, State of Bahia. [From a photograph by R. Crandall, 1907.]

1. NEWLY CLEARED FIELD COVERED WITH MOUNDS OF ANTS NEAR RIO UTINGA, STATE OF BAHIA, BRAZIL.

The largest of these mounds have bases of six or seven meters. Photograph by R. Crandall, 1907.

BRAZIL.

White spots in the background are the mounds. Photograph by Dr. Gonzaga de Campos, 1909.