straight for 6 or 7 feet, sometimes they curve abruptly, sometimes they are nearly horizontal, but near the mouth may be almost vertical in direction. These Termites are very local, but the specimens are numerous when found. Mr. Haviland dug for these Insects at two places on the Tugela river, one of them being at Colenso. It is much to be regretted that he was unable to reach the nest. We figure a soldier selected from specimens sent by Mr. Haviland to the Cambridge University Museum. This Insect is apparently much smaller than Smeathman's _T. viarum_. Other species of Termitidae have been described[301] as forming underground tunnels in Africa, but none of the species have yet been satisfactorily identified. It was stated by Smeathman that some species of Termites had chambers in their habitations in which grew a kind of fungus used by the Insects for food; Mr. Haviland is able to confirm Smeathman in this particular; he having found fungus-chambers in the nests of more than one species both in Singapore and South Africa (Fig. 240). HABITATIONS.—In nothing do Termites differ more than in the habitations they form. Sometimes, as we have mentioned in the case of _Calotermes_, there is no real structure formed; only a few barriers being erected in burrows or natural hollows in wood. In other cases very extensive structures are formed, so that the work of the Termites becomes a conspicuous feature in the landscape. This is of course only the case in regions that are not much interfered with by man; the great dwellings spoken of by Smeathman and others soon disappear from the neighbourhood of settlements, but in parts of Africa and in Australia large dwellings are still formed by these creatures. In the latter part of the world there exists a very remarkable one, formed by an {386}undetermined species called by the officers and crew of her Majesty's ship _Penguin_ the "compass ant." The outline of one of the structures formed by this Termite we represent in Fig. 239. Mr. J. J. Walker, to whom we are indebted for the sketch from which this figure is taken, has also favoured us with the following extract from his diary, of date 4th August 1890: "The most interesting feature in the scenery (about forty miles inland from Port Darwin) was the constant succession of huge mounds raised by the Termites, of which I had seen some comparatively small examples in my rambles near Port Darwin; but these exceeded in dimensions all I had ever seen. The most frequent as well as the largest kind was usually of a reddish or ferruginous colour outside, and generally almost cylindrical in shape with obtusely-pointed top, but nearly always more or less weather-worn, with great irregular buttresses and deep ruts down the sides; many of them look like ruined towers in miniature. Their usual height was from 8 to 10 feet, but many were much higher, and some attained an (estimated) elevation of at least 20 feet.