206 
Transactions of the Royal Society of South Africa. 
of various lengths, of the weed could usually be seen floating in the water. 
The explanation of the unexpected appearance of the young fish was soon 
discovered on closer examination, when, what appeared to be such fragments 
of weed, suddenly seemed to come to life one after the other, till a number 
, of them resolved themselves into a small swarm of young fish, which swam 
about actively, and were very^obvious, even to a casual observer. 
The reversal of the process could be very easily observed, for if any 
sudden movement was made, and more especially, if the hand-net was 
passed over the water, the small fish at once became rigid and floated about 
in any position, apparently in a quite helpless inanimate condition. They 
then so closely resembled fragments of weed, which also floated about, that 
they could with difficulty be distinguished from them, even by an observer 
of the transformation. This similarity was increased by the characteristic 
colour of these small fish, which was a dark green, very similar to that 
of most of the fragments of the weed. The weed, when floating, usually 
retains its original green colour, more or less, and when this becomes lost 
it sinks to the bottom. 
The fish, when they have assumed the rigid condition, are not easily 
roused from it, and can readily be caught by means of a hand-net when in 
this state. 
There can be little doubt but that this is a case of protective resem- 
blance, by means of which the fish escapes the observation of its enemies. 
At this stage it is apparently more a surface-feeder, and, being a slow 
swimmer, is exposed to the attacks of the numerous fish and birds which 
procure their food at or near the surface of the water. Later on it frequents 
the deeper water, though still a surface-feeder, and is able to escape its 
enemies by its increased power of swimming. It loses much of its 
characteristic green colour and probably its habit of feigning death. 
Another example of the way in which young fish "may be protected is of 
quite a different nature, and shows not only a superficial but a structural 
modification not found in the adult stage. This occurs in the " klip-fishes," 
which are so abundantly represented in the Cape seas, and which, in the 
adult stage, show some remarkable adaptation to their surroundings. 
The young klipfish are born alive, and are at this time of a clear, glassy 
transparency. They are at once able to swim about actively in the water, 
and, unlike their parents, do not remain amongst the rocks and weeds at the 
bottom of the sea, but swim up easily and at once towards the surface of 
the water ; there they remain, near, or at some little distance from the 
surface, without any special effort. Adult klipfish are unable to remain 
floating in the water for any considerable time, and when they cease active 
swimming movements they sink rapidly to the bottom, the reason being that 
there is no air-bladder, such as we have in most of the fishes, to act as a 
float. In the young, however, there is a large and well- developed air-bladder, 
