74 THE EXPLORATION OF THE NIGER Beyond Sansanding the course of the Niger changes considerably. Thus far the river flows between pretty straight uniform banks, but now 'the hills are lower, and behind them the country is perfectly flat, without so much as an undulation, so that they are completely flooded, often for an immense distance, when the water is high. Here and there villages rise from slight eminences, the clumps of hibiscus surrounding them rendering them visible from afar. Now too appears the sweet grass which the natives call burgu, a special characteristic of the riverside vegetation as far as Say. It is a kind of aquatic couch-grass nearly level with the soil when on the subsiding of the floods the ground becomes dry again. Directly the soil is once more inundated, however, the burgu sends out shoots with extraordinary rapidity, and they grow so fast that they soon reach the top of the water. The natives make a sweet beverage of the leaves of this grass, of too sickly a taste to be fancied by Europeans, but negroes are very fond of it.

For our hydrographical surveys the burgu was a most invaluable help, growing as it does, as I have already remarked, wherever the solid ground reappears after the floods. If, therefore, we should be overtaken by a tornado on the open river, we can always take refuge from the waves by anchoring in the middle of the submerged tracts.

On January i we reached Gurao on Lake Debo, where I had recently resided for two years in charge of the Niger flotilla, consisting of the two gunboats Niger and Mage, and a few barges made of the wood of the country. Two of these barges, it will be remembered, were now part of our exploring expedition.

We paid a visit between whiles to the tomb of Sidi Hamet Beckay, in the village of Saredina. I shall often have occasion later to refer to this worthy, so I will content myself with adding but a few words about him here. It