s met with on Lake Nyassa, which was said to be
used exclusively for killing men. It was put on small wooden
arrow-heads, and carefully protected by a piece of maize-leaf tied round
it. It caused numbness of the tongue when the smallest particle was
tasted. The Bushmen of the northern part of the Kalahari were seen
applying the entrails of a small caterpillar which they termed 'Nga to
their arrows. This venom was declared to be so powerful in producing
delirium, that a man in dying returned in imagination to a state of
infancy, and would call for his mother's breast. Lions when shot with it
are said to perish in agonies. The poisonous ingredient in this case may
be derived from the plant on which the caterpillar feeds. It is
difficult to conceive by what sort of experiments the properties of these
poisons, known for generations, were proved. Probably the animal
instincts, which have become so obtuse by civilization, that children in
England eat the berries of the deadly nightshade (_Atropa belladonna_)
without suspicion, were in the early uncivilized state much more keen. In
some points instinct is still retained among savages. It is related that
in the celebrated voyage of the French navigator, Bougainville, a young
lady, who had assumed the male attire, performed all the hard duties
incident to the calling of a common sailor; and, even as servant to the
geologist, carried a bag of stones and specimens over hills and dales
without a complaint, and without having her sex suspected by her
associates; but on landing among the savages of one of the South Sea
Islands, she was instantly recognized as a female. They began to show
their impressions in a way that compelled her to confess her sex, and
throw herself on the protection of the commander, which of course was
granted. In like manner, the earlier portions of the human family may
have had their instincts as to plants more highly developed than any of
their descendants--if indeed much more knowledge than we usually suppose
be not the effect of direct revelation from above.

The Mukuru-Madse has a deep rocky bed. The water is generally about four
feet deep, and fifteen or twenty yards broad. Before reaching it, we
passed five or six gullies; but beyond it the country, for two or three
miles from the river, was comparatively smooth. The long grass was
overrunning all the native paths, and one species (_sanu_), which has a
sharp barbed seed a quarter of an inc