74 
ZOOLOGICAL SOCIETY BULLETIN 
Professor Carl H. Eigenmann in “The Fresh 
Water Fishes of British Guiana” writes of a 
Rivulus that he found near Kaieteur Falls, on 
the Potaro River,—“One of these fishes taken 
in Shrimp Creek jumped against the vertical 
face of a huge rock and clung by the adhesion 
of its tail. From this point by another flop it 
made and clung to a point much higher up the 
face of the rock.” This movement which Pro¬ 
fessor Eigenmann describes, has happened 
many times in the aquariums at Kartabo, and 
in the pools wherein Rivulus have been ob¬ 
served. 
The ability to move about on land more or 
less independently of aquatic conditions, has 
brought about some interesting complications 
in the distribution of these fish. 
On a flat surface the method of progression 
is quite different. When not alarmed the fish 
may be found resting flat on the ground with 
the pectoral fins outspread and forming props 
to support the body. The posterior part of the 
body is twisted around so that the tail lies flat 
on its side on the ground. By shifting the tail 
from side to side with a thrashing motion and 
wriggling the body, the fish moves in a forward 
direction, and is able to traverse relatively long 
distances in this manner. One individual which 
was left on a table near a bird’s nest formed of 
small sticks, reached the nest, climbed within 
the interstices, and with a little wriggling pene¬ 
trated deep within the nest, from which it was 
with difficulty extracted. 
These fish are quite abundant in the jungle 
and occasionally along the shores of the rivers 
near Kartabo. They are not dependent for 
their existence upon any type of surrounding, 
the only necessary factors in their distribution 
being quiet water and shade. We have found 
them in small streams, in pools in the ground 
within the jungle entirely isolated from other 
water, within small water pockets in great but¬ 
tress tree roots, and about the roots of trees 
extending into the river. 
A typical habitat is to be found on one of 
the small islands in the Cuyuni River. Within 
a space of four hundred feet, entirely within 
the dark, shaded forest, a small stream starts 
as a series of connected shallow pools in the 
higher portion of the island, passing downward 
through a group of flat rocks where the pools 
become narrower and deeper, meandering 
around the roots of trees, before entering the 
river, through broad, shallow, leaf-filled, mud- 
bottomed pools, where the water is from two to 
four inches deep. Rivulus are to be found 
throughout the stream, in greater numbers in 
,the broad pools and fewer in the more swiftly- 
moving narrower parts farther up. 
Most of their time is spent beneath leaves, 
either resting or using the protection of the 
leaves as a convenient place from which to dart 
out and capture insects which alight upon the 
surface of the water. 
They are not difficult to capture, but slow 
'movements of a net near their pools are futile,— 
the fish becoming alarmed and dashing away 
'at the first sign of any moving object. When 
using a small landing net, the best method is to 
dip the net as deeply and as rapidly as possible 
into the pool, drag it along the bottom, gather in 
some of the leaves and pull it straight up and 
out. The fish will be found among the leaves 
in the bottom of the net, lying quietly until 
disturbed. They then jump about, landing on 
the sides of the net, and if not speedily caught, 
continuing their flipping until they escape over 
the top. 
Given two or three inches of water in an 
aquarium they live well, distributing themselves 
(through the water, preferring to be neither near 
the bottom nor at the top. They are somewhat 
pugnacious, and when hungry are likely to chase 
and bite off the tails of their aquarium mates, 
The food of Rivulus consists mostly of in¬ 
sects, with occasionally a small amount of vege¬ 
table matter. Examination of the stomach con¬ 
tents of a number of freshly killed fish reveals 
ants as the most plentiful item of diet, flies and 
beetles following, miscellaneous insects and 
vegetable matter last. In captivity they will 
eat scraped meat and insects of all kinds. 
Grasshoppers and other insects over three-quar¬ 
ters of an inch are attacked, but the fish are 
not enthusiastic about them and prefer smaller 
prey. Mosquito larvae, earthworms, ant 
cocoons and larvae, are all devoured with great 
eagerness. 
Note. —It will be gratifying to members of 
the Zoological Society, as it has been to the Staff 
of the Research Station, to read what the emi¬ 
nent naturalist. Prof. T. D. A. Cockerell, writes 
of the work of the Station in “Science,” No. 
1422, pp. 350-351. His words apply with equal 
force to birds, reptiles and fish as well as in¬ 
sects. “If we are to take full advantage of the 
wealth of biological opportunity afforded by the 
insects, we must turn to the tropics where the 
number and diversity of species is at a maxi¬ 
mum. In the tropics essentially similar climatic 
conditions have persisted for ages, permitting 
the development of biocoenoses which may be 
