INSTINCTS AND EMOTIONS IN FISH. 33 



ternal ears likewise are wanting. Still we perceive one very distinct 

 expression in this vertebrate class which is absent, or but slightly- 

 developed, in many of the higher animals, namely, change of colour*. 

 Most of us are aware that when a fish sickens its brilliant tints 

 become less and less, or even entirely fade away, while the same 

 result may follow being vanquished by a foe. But when in good 

 health and residing in suitable localities, especially during the 

 breeding-season, their colours become vivid, and even a temporary 

 accession of anger may cause a similar result. 



The first subject for investigation is, Are the finny tribes desti- 

 tute of affections ? Here 1 purpose inquiring whether fish are 

 monogamous or polygamous, whether they show signs of affec- 

 tion to their companions, if they construct nests, guard their 

 nests or eggs, protect their offspring, and, lastly, if they are 

 ever known to exhibit traits of affection for human beings. In 

 some parts of the world, more especially in fresh waters, we have 

 monogamous as well as polygamous fishes — the former, as a rule, 

 not depositing so many eggs as the latter, probably for two 

 reasons, (1) that they breed more frequently, and (2) that they 

 generally protect their offspring. The Gouramy (Osphromenus 

 olfax) at the Mauritius commences breeding at six months of 

 age, whil& their fecundity is astonishing. During the breeding- 

 season they frequent the sides of tanks, where shelter is afforded 

 them by the grasses and weeds growing in the water, For several 

 days they are very active, passing in and out of their grassy cover, 

 and in some places thickening it by entangling all trailing shoots, 

 and forming what is generally considered the spot under which 

 the ova are depositedf . They continue to watch this j)lace with 



* It is not here held that change of colour is always due to emotional 

 sensations. Thus Professor Agassiz observes that in young Vleuronectoids the 

 embryos were very sensitive to light, both sides changing colour rapidly at will. 

 He considers that as soon as the two eyes become situated exclusively on one 

 side of the head, the nerve controlling the colour-cells, of what has now become 

 the eyeless side, becomes gradually unable to act, consequently the under or 

 blind side becomes colourless. It seems to be an almost invariable fact that 

 the under surface in fish is less vividly coloured than the upper surface, and 

 that such is occasioned by the influence of light ; but whether such is entirely 

 due to the action of the optic nerve is open to grave doubt, for were it so, all 

 blind fish would be colourless ; and here, again, we must distinguish between 

 those which are sightless owing to living in dark caves, and others which are 

 so consequent upon accidental loss of vision in their embryonic stage. 



t On nest-building fishes, see Martens, Zool. Grart. 1872, pp. 107-111. 



