2 A JOURNEY IN BRAZIL. 
sive engagement going on while we sail peacefully along. 
What it means, or how the .hattle ends, if battle it be, we 
shall not know for two months perhaps.* Mr. Agassiz is 
busy to-day in taking notes, at regular intervals, of the 
temperature of the water, as we approach the Gulf Stream. 
To-night we cut it at right angles, and he will remain on 
deck to continue his observations. 
April 3d. The Professor sat up last night as he in- 
tended, and found his watch, which was shared by one or 
two of his young assistants, very interesting. We crossed 
the Gulf Stream opposite Cape Hatteras, at a latitude where 
it is comparatively narrow, some sixty miles only in breadth. 
Entering it at about six o'clock, we passed out of it a 
little after midnight. The western boundary of the warm 
waters stretching along the coast had a temperature of 
about 57. Immediately after entering it, the temperature 
began to rise gradually, the maximum being about 74, 
falling occasionally, however, when we passed through a 
cold streak, to 68. These cold streaks in the Gulf Stream, 
which reach to a considerable depth, the warm and cold 
waters descending together in immediate contact for at 
least a hundred fathoms, are attributed by Dr. Bache to 
the fact that the Gulf Stream is not stationary. It sways 
as a whole sometimes a little toward the shore, sometimes 
a little away from it, and, in consequence of this, the 
colder water from the coast creeps in, forming these verti- 
cal layers in its midst. The eastern boundary is warmer 
* On the 17th of May, nearly a month after our arrival in Rio, this cloud 
was interpreted to us. It was, indeed, charged with the issues of life and death, 
for it was on this day and the following that the final assaults on Petersburg 
were made, and the cloud which marred an otherwise stainless sky, as we were 
passing along the shores of Virginia, was, no doubt, the mass of smoke gath- 
ered above the opposing lines of the two armies. 
