Sept. 21, 1871] 
NATURE 
407 

arteriosus, and provided with transverse series of valves 
in its interior, he found that such a duldus arteriosus was 
likewise present in ZLefzdosiren, but with a very different 
valvular arrangement. This peculiarity, combined with 
the development of a lung, he considered to be sufficient 
to distinguish Lefzdoszren as the type of a separate sub- 
class, which he named Dzxoz, and placed at the head of 
the entire class. 
Thus, then, Lefzdostven was finally placed among the 
Fishes ; but from the time of its discovery dates the ten- 
dency of zoologists to subdivide the assemblage of cold- 
blooded animals xo¢ only where the development of a lung 
ceases, but also where the development of gills begins, 
Or, in other words, systematists became more and more 
convinced that the old division of Reptiles and Fishes was 
insufficient, and that three classes of living cold-blooded 
Vertebrates should be distinguished, viz., Reptiles, Amphi- 
bians, and Fishes, some regarding the second as even 
more closely allied to the third than to the first. 
When we find a group of animals represented by a very 
small number of forms in the existing Fauna, we look to 
Palzontology to fill up the seeming blanks ; but Lefzdo- 
stven did not appear to have any fossil representatives. 
Prof. Owen stated (in 1839) that its teeth resembled “in 
their paucity, relative size, and mode of fixation to the 
maxillz, those of the Cizm@ra and some of the extinct 
cartilaginous fishes, as Cochliodws and Ceratodus,;” but 
no further inference was made from this fact as regards 
affinity. And Prof. Huxley (in 1861), when drawing atten- 
tion to analogous structures in Lefédosiven and certain 
Devonian fishes, still maintained the entire absence of 
the Dipnoous type in the fossil state. 
The discovery of a “gigantic Amphibian allied to the 
genus Lefidosiren, from rivers in Queensland,” and 
named Ceratodus Forsteri by Mr. Krefft, promised to mark 
another step in the advancement of our knowledge, and to 
lend additional aid in determining the natural affinities of 
these animals. As soon as Mr. Krefft had recognised the 
importance of this discovery, the trustees of the Australian 
Museum of Sydney took steps to secure well-preserved 
examples. They sent a collector into the district where 
the animal was known to occur ; and, with their usual 
liberality, they despatched to the British Museum, for 
examination, the first specimens they could spare, by 
which I was enabled to present a full account of its 
organisation to the Royal Society. It is not my intention 
to enter here into the details of the results of this examina- 
tion ; I must be satisfied with giving a short description 
of it, pointing out some of the bearings which this dis- 
covery ha#€pon the advancement of science. 
The fish (for this it proved to be, and even more so than 
Lepidostren) appears to be not uncommon in some dis- 
tricts of Queensland ; specimens have been obtained from 
the Burnett, Dawson, and Mary rivers, some high up in 
perfectly fresh water, others descending into the lower 
brackish portions. It is said to grow to a length of six 
feet, the largest example sent to the British Museum being 
about three and a half feet long. The flesh is excellent 
eating, and of salmon colour, hence it is called by the 
squatters Burnett or Dawson salmon. Its food consists 
of the decaying leaves of myrtaceous and other plants, 
with which the stomach and intestine are crammed. 
Probably now and then it swallows, perhaps accidentally, 
some aquatic animal ; but it is rather doubtful whether it 
can be caught by using living animals as bait. It is also 
stated that it is in the habit of going on land, or at least 
on mud-flats ; and this assertion appears to be borne out 
by the fact that it is provided with a true lung. On the 
other hand, we must recollect that a similar belief was 
entertained with regard to Lepzdostiren, of which now 
numerous examples have been kept in captivity, but none 
have shown a tendency to leave the water. I think it 
much more probable that this animal rises now and then 
to the surface of the water, in order to fill its lungs with 


air, and then descends again until the air is so much de- 
oxygenised as to render a renewal of it necessary. When 
we recollect that the animal evidently lives in mud or in 
water charged with the gases which are the product of 
decomposing organic matter, the usefulness or necessity 
of such an air-breathing apparatus, additional to the 
gills, becomes at once apparent. Further we shall see 
that the limbs of this unwieldy and heavy animal are 
much too feeble and flexible to be of much use in locomo- 
tion on land ; they may assist it in its crawling, in water, 
over the muddy bottom of a creek ; but the chief organ 
of locomotion is the compressed, broad, and flexible tail, 
denoting by its shape and structure that the fish can exe- 
cute rapid swimming motions. However, it is quite pos- 
sible that it is occasionally compelled to leave the water, 
although I do not believe that it can exist without it in a 
lively condition for any length of time. It is said to make 
a grunting noise, which may be heard at night for some 
distance. This noise may be produced by the passage of 
the air through the cesophagus, when it is expelled for the 
purpose of renewal. 
It deposits a great number of small eggs, which are im- 
pregnated after deposition. Nothing is known of their 
development ; but we may infer that the young are pro- 
vided with external gills, like those of some other Ganoid 
Fishes. 
The Barramunda (we will use the name given to it and 
other similar fishes by the natives) is eel-shaped, but 
considerably shorter and thicker than a common eel, and 
covered with very large scales. The head is flattened and 
broad, the eye lateral and rather small, the mouth in front 
of the broad snout and moderately wide. The gill open- 
ings are arather narrow slit on each side of the head. 
There are no external nostrils. The tail, which is of about 
the same length as the body without the head, is com- 
pressed, and tapers toa point, but it is surrounded by a 
very broad fringe, supported by innumerable fine and 
long fin-rays. There are two fore and two hind paddles, 
similar to each other in shape and size, and very different 
from the fins of ordinary fishes ; their central portion 
being covered witi a scaly skin, and the entire paddle 
surrounded by a rayed fringe. If we were to cut off the 
hind part of the tail of a fish, the piece would bear a 
strong resemblance to one of the paired paddles. The 
vent is situated in the median line of the abdomen 
between the paddles. 
In order to obtain a view of the inside of the mouth, 
it is necessary to slit it open, at least on one side. We 
then notice that there are a pair of nasal openings within 
and on each side of the cavity of the mouth. The palate 
is armed with a pair of large, long, dental plates, with a 
flattish, undulated, and punctated surface, and with five 
or six sharp prongs on the outer side, entirely similar to 
the fossil teeth described under the name of Ceratodus. 
Two similar dental plates of the lower jaw correspond to 
the upper, their undulated surface fitting exactly to that 
of the opposite teeth. Beside these molars the front part 
of the upper jaw (vomer) is armed with two obliquely 
placed incisor-like dental lamellz, which have no corre- 
sponding teeth inthe lower jaw. As we know the kind 
of food taken by the Barramunda, the use of their teeth 
is apparent. The incisors will assist in taking up, or even 
tearing off, leaves, which are then partially crushed be- 
tween the undulated surfaces of the molars. 
The skeleton consists of a cartilaginous basis, in the 
form of along tapering chord for the body and tail, and 
in that of a capsule forthe head. No segmentation into 
separate vertebrz is visible in any part of the notochord 
but it supports a considerable number of apophyses, the 
abdominal of which bear well developed ribs, all being 
solid cartilaginous rods, with a thin sheath of bone. In 
the same manner no part of the brain-capsule is ossified, 
but it is nearly entirely enclosed in thin bony lamellae. 
This is also the structure of the appendages of the skull, 
