The Migration of Fishes. 57 
(5.) It explains better than the other theories the appear- 
ance of the fish at the time of their arrival in the spring. The 
menhaden appear to be bottom-feeders. If they migrated 
coastwise to the south they would find these feeding-grounds. 
If they sank to the bottom they would find food, if they had 
sufficient vitality to resurrect themselves in the spring. If 
they passed the winter in the mid-ocean strata they could ob- 
tain no food, and would naturally become emaciated, the ac- 
cumulated fat of the preceding summer being absorbed. 
IV. A NEW CLASSIFICATION, OF THE HABITS< Ge 
THE MARINE FISHES. 
Rimbaud’s classification, which is a modification of one 
recognized in the markets of South France, is very suggestive, 
but it does not appear to me to be entirely applicable to the 
fishes of our coast, at least not in the way in which it has 
been usually adopted. 
Rimbaud makes four divisions, viz.: 
I.—Wandering fishes (Porsson nomade). 
Il.—White fishes (Potsson blanc). 
IIl.—Bottom fishes (Potsson de roche, or Potsson de fond). 
IV.—Alien, or outside fishes (Potsson forain). 
The distinction between classes I. and IV. does not appear 
to be very clearly marked. In the Western Atlantic some of 
the fishes making up class IV. belong to each of the other 
classes. 
A more natural classification would be its three divisions, 
which might readily be co-related with the three kinds of mi- 
gration mentioned in the preceding paragraph. 
The first group would include the wandering fishes, the 
Poisson nomade of Rimbaud, whose migrations are entirely 
