NOCTURNAL FISHES. 17 1 



known to display an equal amount of confidence to- 

 wards those who tend them. The food question being 

 settled, another difficulty presented itself, and this 

 time one that threatened, sooner or later, to accom- 

 plish the extermination of the whole shoal. Imme- 

 diately succeeding their advent, a large number of 

 these little fish were found dead each morning, at the 

 bottom of their tanks, a circumstance which at first 

 seemed inexplicable in association with their quiet 

 behaviour during the day. A night inspection, how- 

 ever, happily revealed the cause of their rapid destruc- 

 tion. It was then seen that the nocturnal movements 

 of the herring, at least in confinement, are altogether 

 distinct from those seen by daylight. In the latter 

 instance these movements are very quiet and uniform, 

 the fish swimming round their tank in one shoal and 

 in one continuous stream. At night, on the contrary, 

 the shoal is entirely broken up, each fish taking an 

 independent path, and darting from one side to the 

 other with an amount of agility scarcely to be antici- 

 pated by a mere daylight acquaintance with the 

 species. It was during these active nocturnal move- 

 ments that the fish struck against the rockwork of 

 their tank and came to an untimely end. This mor- 

 tality, however, was soon arrested by placing a dim 

 light over their tank, which illuminated the outline 

 of the rockwork just sufficiently to enable them to 

 recognise and avoid it. With this dim light the fish 

 still retained their active habits, and it was noticeable 



