STORMS AT SAN CARLOS 485 
thunder was exceedingly close and loud, keeping 
up a continuous roll, and the glare of the lightning 
almost ceaseless. Not much rain fell here and by 
9 o'clock all had cleared away. 
It is difficult to ascertain the frequency of the 
flashes when you are in the midst of the explosions, 
but in coming from Marabitanas last month, when 
we did not reach our sleeping-place until long after 
dark, a heavy thunder-shower passed near without 
giving us a drop, and I could see beautifully the 
flashes issuing from the cloud. I counted the 
pulsations of my wrist between consecutive flashes. 
They varied from two to eight, giving an average 
of five pulsations, and fifteen flashes to a minute. 
In the middle of the rainy season, though the 
amount of rain that falls is greater, these very 
violent thunderstorms are of rare occurrence. . . . 
When I was at the Jauarite caxoeira on the 
Uaupes in October 1852, the lightning struck a 
house at the mouth of the Paapurfs and prostrated 
the inmates, but injured no one save a young 
man in a hammock, who was deprived of the 
use of a leg (whether permanently I did not learn). 
This was late in the afternoon, and on the follow- 
ing morning every person I met had his face and 
arms streaked with red carajurii, intended as a 
protection against the paje whose incantations had 
brought down the thunderbolt on the wounded 
man affecting them in the same manner. 
Near electrical discharges are always followed 
by augmented fall of rain. This is peculiarly 
evident where there is a lull or a slight abatement 
in the shower and a vivid flash of lightning comes, 
restoring the fall of rain in all its force. Generally 
