We travelled about twelve miles south by west to latitude 16 degrees 47 minutes; at first over an almost uninterrupted box-flat, full of melon-holes, and with many small holes in the ground, which caused our horses and cattle to stumble at almost every step.  The dry melon-holes were covered with dead Paludinas, with shells of a large crab, and of the fresh water turtle.  At about seven miles, we passed a strip of Blackwood forest, with many Nonda trees; and crossed a small creek.  The latter part of the stage was again over a large box-flat, intersected by shallow grassy depressions, timbered with flooded-gum.  We saw on the rising ground some open scrub, with scattered Bauhinias and Cochlospermums.  Our encampment was at a creek on the south side of a slight rise, with Bauhinia trees, and near good water-holes.  The creek, like all the others we had passed, flowed to the westward.

Near our camp we examined three holes, full six feet deep, and four feet in diameter, communicating with each other at their bottom.  They were about three feet apart, and appeared to have been dug with sharp sticks.  I have not the slightest idea for what purpose they were intended.  They were most certainly not dug to obtain roots; and it seemed unlikely for wells; for the water, even in this unusually dry season, was very abundant.

The white ant-hills, which are built in rows, had, during this stage, a direction from north by west to south by east, and, as I have before mentioned a conjecture that the little builders would expose the narrowest side of their habitation to the weather side, the prevailing winds would be from the north.

July 9.—­We travelled thirteen or fourteen miles south by west to latitude 17 degrees 0 minutes 13 seconds, at first crossing a box-flat, and after that a succession of greater or smaller plains, separated by a very open Grevillea forest.  These plains were well grassed, or partly covered with a species of Euphorbia, which was eaten by our horses and cattle; and also with the long trailings of the native melon; the fruit of which tastes very tolerably, after the bitter skin has been removed; but when too ripe, the fruit is either insipid or nauseous.  The bustard seems to feed almost exclusively on them, for the stomach of one, which Brown shot, was full of them.

The apple-gum, which we had missed for some time, again made its appearance, accompanied by another white gum, with long narrow leaves.  As we approached the creek, at which we afterwards encamped, the vegetation became richer, and the melon-holes enlarged into dry water-holes, which were frequently shaded by the Acacia with articulate pods (Inga moniliformis).  The two species of Terminalia, of the upper Lynd, were numerous; and a small green looking tree, which we found growing densely along the creek, had wood of a brown colour, which smelt like raspberry jam; and, upon burning it, the ashes produced a very strong lye, which I used in dressing the wounds of my companions.  This tree was found in great abundance on all the rivers and creeks round the gulf, within the reach of salt water; and when crossing Arnheim Land, though less frequently.