GASTEROSTEUS ACULEATUS, ETC. 313 



This fish was rather small, had the throat of a bright reel 

 colour, and the eyes of a brilliant bluish green. At first, all the 

 others were pale ; but, in the course of a few days, one of them 

 gradually assumed the rich hues ot that just described, and soon 

 afterwards it also became attached to a spot, taking up its abode 

 in one of the corners of the trough. On examining attentively 

 the two selected localities, a nest was found in each, composed of 

 a collection of delicate vegetable fibres, resting on the bottom of 

 the trough, and matted into an irregularly circular mass, some- 

 what depressed, and upwards of an inch in diameter ; the top 

 being covered over with the same materials, and having, in the 

 centre, a large hole. The fishes scarcely ever strayed from their 

 nests, but were constantly on guard, defending or repairing them ; 

 they were perpetually prying into the hole at the top and thrust- 

 ing their heads right into it. On one occasion, one of them 

 entered by this hole, and slowly forced itself right through the 

 side of the nest j as it gradually moved onwards, its body had a 

 peculiar, lateral vibratile motion. They would frequently seize 

 hold of the nest and give it a violent tug, shaking and tearing 

 loose the vegetable matter of which it was composed ; at other 

 times they would carry to it, in their mouths, fine conferva-stems, 

 and press them with considerable force into the walls of the nest, 

 or thrust them into the hole, which by this means, was sometimes 

 partially concealed. Occasionally, each was observed hovering 

 over its nest, with the head close to the orifice, the body being 

 inclined upwards, at an angle of about 45°, fanning it with the 

 pectoral fins, aided by a lateral motion of the tail. This curious 

 manoeuvre was apparently for the purpose, so to speak, of venti- 

 lating the spawn, which could be distinctly seen through the orifice 

 at the top ; at least, by this means, a current of water was made to 

 set in towards the nest, as was rendered perfectly evident by the 

 agitation of particles of matter attached to it. This fanning or 

 ventilating process was repeated, at short intervals, during the 

 day, and every day until the spawn was hatched, to accomplish 

 which took between two and three weeks. 



Only one nest contained spawn ; the other was torn in pieces, 

 and the materials scattered about, in the hope that we might have 



