of the termites' body weight consist of water, but all the fungus gardens (in the nests of gardening species) were moist, and the atmosphere of the royal cavity was, as usual, saturated with heated water vapour. Whence did this apparently unlimited supply come from?
The termites could not procure it out of the soil, because the earth was so parched that hardly a film of vapour could be produced by distillation of subsoil. The only other possible sources of supply were dew and the sap of living plants. In time, careful observation excluded both these. It was just a lucky chance that put us on the track of the surprising truth. The late Mr Wessel Wessels, a fine practical naturalist and a keen observer, who, as recorded in The Star, was killed by a mamba some years later, first suggested a possible source of supply. He told us he had found termite shafts in the walls of deep wells on two occasions in Bechuanaland. These shafts had been disclosed by the digging of the wells.
The termites went down these shafts to the level of moist earth and from there brought their water to the ant-hill on the surface. This seemed almost as incredible as the theory of the manufacture of water - and yet it was the true explanation. Very shortly after this suggestion was made, we were informed that a well had been sunk high on a hillside on the farm Rietfontein No. 1368, and that a termite shaft had been disclosed going down to a depth of sixty-five feet from the surface.
We had an opportunity of observing for several months this tremendous feat of engineering and the use to which it was put by the termites, and nothing I had previously observed in the life of the white ant was more absorbingly interesting. The well had originally been sunk at a spot pointed out by a waterdiviner, but at a depth of sixty-five feet the work was stopped because no moist earth had yet been found.
In the northern wall of the well was a two-and-a-half inch wide termite shaft which went down to the bottom of the well and disappeared in the ground. The ant-hill to which this shaft belonged was situated a few paces from the edge of the well. We carefully opened up the gallery for its entire length, and disclosed some of the fungus gardens by means of a small aperture which could be closed when not in use.
The nest belonged to a species in which soldiers and workers were not so widely differentiated as in Eutermes (the common high-veld termite largely dealt with in the previous articles). It soon became evident that the termites had been placed in a tragic difficulty by the opening up of their shaft. Although, above everything, they hated the light, they were compelled to use the shaft since a cessation of the water supply, even for a single day, would mean the death of the community.
And so we saw an endless stream of workers going into the depths on the left-