336 THE AMERICAN NATURALIST.  [Vor. XXXIX. 
the sea, where they lay their eggs, probably near the mouths of 
the rivers from which they descended,! and then apparently die, 
as no adult fish are ever seen to return. 
The young fry, when about $ to 1} inches in length ascend 
the rivers by thousands during the wet season (August, Septem- 
ber, and October), moving up stream in a continuous line near 
or under the banks, as do the young of eels (Anguilla). When 
in a stretch of comparatively quiet water they move steadily 
onward; but in rapid water they progress by jerks, resting on 
the bottom for a few seconds, then making a fresh dash onward 
and taking a fresh grip on a pebble or rock with the ventral 
sucker, and, after remaining quiet for a few seconds, dashing on 
again. They even ascend vertical or overhanging surfaces, 
over which a small amount of water is running in this way, 
resting for a while, then moving upward an inch or so, resting 
again, and moving on. Sometimes during one of these ascents 
they are swept off and into the eddy below; but in a few min- 
utes they are ready to try it again. I have seen as many as a 
dozen moving up the face of a rounded rock a foot in diameter, 
over which the flow of water was not enough to cover their 
bodies. After a heavy rain the waters of the St. Vincent rivers 
rise rapidly, and then fall again, leaving many little outlying 
pools along the banks, which, under the influence of the 
scorching tropical sun soon dry up, leaving dusty hollows. 
Many of the fishes become cut off from the main stream at 
such times, and, as the pools dry up may be seen jumping about 
in the hollows, entirely covered with a thick coating of dust. If 
these stranded individuals be placed again in the main stream, 
they soon begin to ascend with the others as if nothing had hap- 
pened. The tenacity of life of the young tri-tri is remarkable. 
They will live for several hours in these dry situations, exposed 
to the full rays of the sun. 
The journey of the fry up the rivers occurs at the time when 
! The fact that certain rivers in these islands, apparently suitable in every way 
for these fish are not inhabited by them seems to point toward the fact that the 
fish from the neighboring streams spawn in the vicinity of their mouths; other- 
wise we should expect to find a few of the young straying into these uninhabited 
streams every year. 
