earthen vessel. At the end of three hours
they were alive and well.

No. 5, A bandage was tightly stretched
around the head of one of these fishes, so
as entirely to prevent it from opening its
gills. It was then placed in a globe of
water, and at the end of twenty-four hours
was as lively as possible. In this case there
was direct evidence that the fish must have
sustained life by aërial respiration.

No. 6. A water breather similarly
bandaged died in thirty-four hours; but in
this case, owing to the external form of the
fish, the gill cover could not be entirely
closed.

No. 7. A walking fish was placed in a
dry cloth at 9.55 A.M. and left without any
moisture, the temperature being seventy-
five degrees. It lived until 1.20 P.M., occasionally
opening its mouth and taking in air.
At 12.15 it moved across the table
and fell on the ground; and it had
proceeded several feet across the room before
it was picked up. The fall probably
hastened its death. Another of these fishes
eighteen inches long, lived for sixteen hours,
wrapped up in a dry cloth, and placed in a
closed cupboard.

No. 8. A number of these fishes were
placed in a tub, with a small amount of
water and plenty of common grass. No
other food was allowed them; but at the
end of three weeks they were perfectly well
and lively.

There is considerable discrepancy among
naturalists as to the anatomical peculiarities
which allow these and some other genera
of fishes to exist for a comparatively long
period out of water. Professor Owen
observes that, "Accessory respiratory organs,
acting chiefly as a reservoir or filter of
water, are developed from the upper part
of the pharynx or gullet in the climbing
perch (anabas scandens) and allied fishes
of amphibious habits; they are complex
folds of slightly vascular membrane,
supported on sinuous plates; whence this family
of fishes is called labyrinthibranchii;" and
he copies curious figures of the labyrinthic
reservoir of anabas. Günther states that
''the ophiocephalidae (or walking fishes)
have a cavity accessory to the gill cavity
for the purposes of retaining water."

Following these authorities, Dr. Day
started with the belief that this cavity was
for the purpose of retaining water to be
gradually doled out to the gills when the
fish was out of water, with the object of
keeping those organs moist, and thus able
to obtain oxygen from the air. Personal
observation led him, however, to arrive at
a different conclusion. He found that the
cavity or reservoir does not contain water,
but has a moist secreting surface, and that
it contains air, which is retained there for
respiratory purposes; he believes that this
air, after having been thus employed, is
ejected by the mouth. If the fish be
kept under water, this cavity, which is
surrounded by bony tissue, becomes filled
with water, which cannot be discharged;
and as the cavity cannot be emptied, the
water becomes carbonised, and unfit for
oxygenating the blood. The whole respiratory
process thus becomes thrown upon
the gills; and this will account for the fact
noticed by Dr. Day, that when the fish is
in a state of quiescence it will live much
longer in exclusion from atmospheric air
than when excited and moving about in
the water.

A strange-looking, finless, snake-like eel,
the symbranchus cuchia, found in holes in
the Indian marshes, affords a good example
of an air-breathing fish. The peculiarities
of its breathing apparatus are described
by Professor Owen in vol. i., p. 487 of his
Anatomy of Vertebrates. It is suflicient
for the general reader to know that the gills
are in a mere rudimentary state, and that
the respiratory process is transferred to a
receptacle on each side of the head, above
the branchial arches. The cavities are
connected by an opening with the mouth,
and are lined with a highly vascular
membrane, to which impure venous blood is
conveyed. These cavities thus act as lungs,
and the blood permeating their vessels, is
changed from the venous to the arterial
state. Although the anatomical arrangement
of the blood vessels is such that about
half of the volume of the blood transmitted
from the heart is conveyed to the aorta without
being exposed to the action of air, the
fish (notwithstanding its reptilian form of
circulation) is not "of a sluggish and torpid
nature," as Professor Owen asserts, but is
very active in its movements, and almost
invariably gives rise to an exciting chase
over the grass before it can be captured.

Most of the great tenacity of life for
which many of the Indian fresh-water fishes
are famed is, no doubt, as Dr. Day observes,
"due to their capability of respiring
atmospheric air." In India the majority of
inland acanthopterygians* are compound
breathers, as, for example, the whole of

'*' The acanthopteri or acanthopterygians are an
extensive order of fishes so called from the prickly
and inflexible character of the rays in the fins.