back saw Nigger Boy swimming after us. A number of vaqueros attempted
to catch him, but he outswam them and came out with the cattle; the
excitement was too much for him to miss.

Each trip was a repetition of the former, with varying incident. Every
hoof was over in less than two hours. On the last trip, in which there
were about seven hundred head, the horse of one of the Mexican
vaqueros took cramps, it was supposed, at about the middle of the
river, and sank without a moment's warning. A number of us heard the
man's terrified cry, only in time to see horse and rider sink. Every
man within reach turned to the rescue, and a moment later the man rose
to the surface. Fox caught him by the shirt, and, shaking the water
out of him, turned him over to one of the other vaqueros, who towed
him back to their own side. Strange as it may appear, the horse never
came to the surface again, which supported the supposition of cramps.

After a change of clothes for Quarternight and myself, and rather late
dinner for all hands, there yet remained the counting of the herd. The
Mexican corporal and two of his men had come over for the purpose, and
though Lovell and several wealthy rancheros, the sellers of the
cattle, were present, it remained for Flood and the corporal to make
the final count, as between buyer and seller. There was also present a
river guard,--sent out by the United States Custom House, as a matter
of form in the entry papers,--who also insisted on counting. In order
to have a second count on the herd, Lovell ordered The Rebel to count
opposite the government's man. We strung the cattle out, now logy with
water, and after making quite a circle, brought the herd around where
there was quite a bluff bank of the river. The herd handled well, and
for a quarter of an hour we lined them between our four mounted
counters. The only difference in the manner of counting between Flood