•ON THE LAJBORATORT OF THE MAEINE BIOLOGICAL ASSOCIATION. 549 



appearances, but which, when its hunger has been appeased, noses its food 

 before eating, is an illustration of, perhaps, the minimum use of smell 

 observed by me. Many other habitual sight-feeders make, or can make, 

 lauch more extensive use of the nose. The well-known fact that stale 

 bait appeals to few of our food-fishes points to this, and my experiments 

 bear it out. 



So far as I could determine, fish that are not very hungry habitually 

 smell food before taking it. The pollack seems usually to be ready for a 

 naeal, and on almost all occasions when anything eatable is thrown into 

 the tank in which it is swimming it rushes towards it, and bolts it. It 

 does not hesitate to take stale food or food that has been steeped in 

 strong smelling fluids ; and time after time I have been amused to see 

 its too-late repentance after it had swallowed clams that had been 

 saturated with alcohol, chloroform, turpentine, &c. It is only when it is 

 satiated with frssh food or disgusted with what is nauseous that it takes 

 the precaution to smell before eating. On the other hand, various fish 

 that are equally keen-sighted, and habitually recognise their food by 

 the use of their eyes, are more prudent. The whiting {Gadus mer- 

 langus), for instance, appears to pay much more attention to smell, and, 

 as a rule, turns about and withdraws on approaching within a few inches 

 of high-smelling objects that the pollack would take without hesitation. 

 Even whiting, however, cease to be delicate if they are very hungry, and 

 if other fish are present to compete for the food that is thrown to them. 

 In such circumstances bait that is very distasteful may be taken by even 

 the most cautious of sight-feeders ; and likewise in such circumstances a 

 quite smell-less artificial bait may be successfully employed. Where large 

 shoals of fish are, there are likely to be many that are very hungry, and 

 the consequent keen competition will lead to hasty feeding by sight alone ; 

 and hence it is, probably, that lead-baits are successfully employed in cod- 

 fishing in the Moray Firth and ofi" the Northern Islands, while they are 

 of no avail among the scanty fish further south. 



It may be said that in these cases the fish actually search for their food 

 by sight alone, and merely test the quality of what they have found by 

 smelling it ; and Bateson quite recognised this. But more is possible : 

 habitual sight-feeders can he induced to hunt by smell alone. The pollack, 

 which is such a pronounced sight-feeder that it will take a hook baited 

 with a white feather or a little bit of flannel and trolled along the sur- 

 face, is yet able, when blinded, to get its food with great ease. Several 

 blind specimens in the Plymouth tanks were carefully watched by me ; 

 and I had no difliculty in deciding that it was by smell alone that 

 they found their food. Their conduct was exactly such as was seen in 

 the smell-feeders, to which I shall presently refer. 



Again, the cod (Gadus morrhua), which Bateson puts among the sight - 

 feeders, is generally believed — and with good reason, I think — to feed 

 more by night than by day ; which suggests that it, too, not only tests 

 its food, but actually hunts by smell. 



Lastly, in this connection I would state the results of my experiments. 

 I worked with a number of fish, and always with the same success ; but I 

 shall here only refer to one case — that of the dabs (Pleuronectes limanda). 

 That they were sight-feeders was evidenced by their behaviour when I 

 lowered a closed tube full of water, and with a worm in the middle of it, 

 into their tank : time after time they bumped their noses against the 

 glass at the very spot where the worm was situated. That they could 



