NOTES ON THE NEW ZEALAND FROST-FISH. 157 
are more widely found, and they have evidently had time to 
divide into several genera. The marine form from which they 
sprung appears to have become extinct, and to have left no 
marine descendant at all near the line of direct descent. I have 
lately taken a marine Philougria at Coogee Bay, near Sydney, 
New South Wales, but it resembles terrestrial species of PAzlou- 
gria so closely that I am rather inclined to consider it a terres- 
trial form that has returned to the sea—something like the seals 
among the mammals—than one that has always lived in it. The 
Amphipoda appear to be only now developing terrestrial forms, 
and a splendid series could be made out of existing species from 
Nicea, living wholly in the water, through Alorchestes, &c., which 
live in rock-pools, Lut can walk and live on land with great 
agility, Zalorchestia, &c., living just above high-water mark, and 
only occasionally splashed with salt water, to species of Orchestia 
and Zalitrus, suchas O. Sylvicola, which live far away from the 
sea. 
NOTES ON THE NEW ZEALAND FROST-FISH. 
<> 
The habits of this singular fish are so long in being discovered 
that any scraps of information which may be obtained from time 
to time become correspondingly precious. Were the fishermen of 
‘our coasts, however, accustomed to a more careful examination 
of the specimens they procure, a great deal more knowledge 
might soon be within our possession. It was with much interest 
then that the writer read Dr. R. v. Lendenfeld’s remarks in the 
N.Z. JOURNAL OF SCIENCE for May of this year on this fish. 
From his researches this observer deems it probable that the 
Frost-fish during winter, which is also assumed to be its breed- 
ing season, draws into shallow water, as compared with the great 
depths it had previously frequented, for the purpose of spawning. 
That this habit results in such a reduction in the pressure of the 
water as causes an expansion of the air bladder, so great as to 
rupture that organ, thus causing the death and stranding of the 
fish. Another discovery of interest is his report of the full ovaries 
and milts at the same period. 
Dr. v. Lendenfeld’s theory is so far a good and ingenious one, 
and he infers a certain amount of corroboration as due to the 
fish having never been netted, nor any small fish ever seen. But 
the theory is hardly consistent with itself, or it requires further 
proof, and is quite at variance with facts known to other 
observers. | 
First, then, as to the theory. The discomfort to which the 
fish would be subjected at the first portion of its ascent from as- 
sumed great depths towards shallow water, would instinctively 
cause it to return to those depths long before it had lost all con- 
trol over its air bladder. Or, supposing the bladder to be dis- 
tended to the bursting point, then the fish would float belly up- 
