without doubt were near. All occurred in a level area of
shifting sand, sixty yards by twenty, situated among some
high sand-hillocks, and at the distance of about half a mile
from a chain of hills four or five hundred feet in height. The
most remarkable circumstance, as it appears to me, in this
case as well as in that of Drigg, and in one described by
M. Ribbentrop in Germany, is the number of tubes found
within such limited spaces. At Drigg, within an area of
fifteen yards, three were observed, and the same number
occurred in Germany. In the case which I have described,
certainly more than four existed within the space of the
sixty by twenty yards. As it does not appear probable that
the tubes are produced by successive distinct shocks, we must
believe that the lightning, shortly before entering the ground,
divides itself into separate branches.

The neighbourhood of the Rio Plata seems peculiarly subject
to electric phenomena. In the year 1793, [12] one of the
most destructive thunderstorms perhaps on record happened
at Buenos Ayres: thirty-seven places within the city were
struck by lightning, and nineteen people killed. From facts
stated in several books of travels, I am inclined to suspect
that thunderstorms are very common near the mouths of
great rivers. Is it not possible that the mixture of large
bodies of fresh and salt water may disturb the electrical
equilibrium? Even during our occasional visits to this part
of South America, we heard of a ship, two churches, and a
house having been struck. Both the church and the house
I saw shortly afterwards: the house belonged to Mr. Hood,
the consul-general at Monte Video. Some of the effects were