M. Carbonnier has recorded how the male of the curiously grotesque telescope-fish, a variety of _Carassius auratus_ (Linn.), acts as accoucheur to the female. Three males pursued one female which was heavy with spawn, and rolled her like a ball upon the ground for a distance of several metres, and continued this process without rest or relaxation for two days, until the exhausted female, who had been unable to recover her equilibrium for a moment, had at last evacuated all her ova.[117] That adult fish are capable of feeling affection for one another would seem to be well established: thus Jesse relates how he once captured a female pike (_Esox lucius_) during the breeding season, and that nothing could drive away the male from the spot at which he had perceived his partner slowly disappear, and whom he had followed to the edge of the water. Mr. Arderon[118] gave an account of how he tamed a dace, which would lie close to the glass watching its master; and subsequently how he kept two ruffs (_Acerina cernua_) in an aquarium, where they became very much attached to one another. He gave one away, when the other became so miserable that it would not eat, and this continued for nearly three weeks. Fearing his remaining fish might die, he sent for its former companion, and on the two meeting they became quite happy again. Jesse gives a similar account of two gold carp.[119] Anger is strikingly shown by many fish, and notoriously by sticklebacks when their territory is invaded by a neighbour. These animals display a strange instinct of appropriating to themselves a certain part of the tank in which they may be confined, and furiously attacking any other stickleback which may presume to cross the imaginary frontier. Under such circumstances of provocation I have seen the whole animal change colour, and, darting at the trespasser, show rage and fury in every movement. Of course, here, as elsewhere, it is impossible to be sure how far apparent expression of an emotion is due to the presence of that mental state which we recognise as the emotion in ourselves; but still the best guide we have to follow is that of apparent expression. Following this principle, we are also entitled to attribute to fish the emotions conducive to play; for nothing can well be more expressive of sportive glee than many of their movements. As for jealousy, the fights of many male fish for the possession of females constitutes evidence of emotion which would be called by this name in the higher animals. Schneider, in his recent work already often quoted, says that he has observed a male fish (_Labrus_) show jealousy only towards other individual males of his own species--chasing these away from the neighbourhood of his female, but not objecting to the approach of fish of other species.