38
M. J. WELLS
water and terrestrial pulmonates. Thus Humphrey (1930) studied
habituation in Helix-the animal ceases to respond to mechanical
shock after two or three trials. Among marine forms Hirsch (1916)
used Natica and found that the withdrawal response to a shadow failed
after three trials ; half an hour later the animal responded again. Arey
and Crozier (1921) had similar results with the littoral pulmonate
Onchidium. I n more complex experiments, very variable results have
been obtaineq. Fischel(1931), Garth and Mitchell (1926) and Thompson
(1917) all tried to train pulmonates in T- and Y-mazes. Thompson’s
Physa did not learn at all. Fischel was more successful with Ampullaria,
which appeared to master a Y-maze in about ten trials. The animal
forgot, however, by next day, and could not be trained in a T-maze
under the same conditions. Garth and Mitchell’s experiment is perhaps
more convincing, though even here there remains the possibility that
their single Ruminia developed the habit of turning to the correct side
of the T by chance rather than as a result of training. It should be
noted that nearly all the animals that have been tested are terrestrial or
freshwater herbivores, and that attemps to regulate their behaviour
have been based on punishments for incorrect responses. In view of
the results with annelids (see Section 111, B, p. 27) it seems likely that
better performances would have occurred had the animals been handled
less and some system of rewards included in the training programme.
It is surprising to find that nothing has been done with the apparently far more promising marine prosobranchs. The neogastropods in
particular are active predators and scavengers, seeking out their food
in a purposeful manner which implies that it should be no great trouble
to arrange for reward-motivated learning in choice experiments.
B. Homing by limpets
The lack of direct evidence of learning in marine forms is to some
extent offset by a considerable body of information about the homing of
limpets. There has been considerable argument as to whether limpets
do in fact return to individual “ homes ’’ on the rocks. This has arisen
because: (1) limpets a t least in some habitats are comparatively
immobile anyway, so that observation of individuals in the same place
day after day is not good evidence of homing (Villee and Groody, 1940);
(2) the same species may behave quite differently on different substrates ; Patella vulgata L., for example, tends to be most active and to
change its “ home ” most often on smooth damp rocks (Orton, 1929 ;
Jones, 1948) ; and (3) there has been a tendency to compare findings
from different species, which do not all behave in the same way even
within the same habitat. Some species regularly return to a home,
M. J. WELLS
water and terrestrial pulmonates. Thus Humphrey (1930) studied
habituation in Helix-the animal ceases to respond to mechanical
shock after two or three trials. Among marine forms Hirsch (1916)
used Natica and found that the withdrawal response to a shadow failed
after three trials ; half an hour later the animal responded again. Arey
and Crozier (1921) had similar results with the littoral pulmonate
Onchidium. I n more complex experiments, very variable results have
been obtaineq. Fischel(1931), Garth and Mitchell (1926) and Thompson
(1917) all tried to train pulmonates in T- and Y-mazes. Thompson’s
Physa did not learn at all. Fischel was more successful with Ampullaria,
which appeared to master a Y-maze in about ten trials. The animal
forgot, however, by next day, and could not be trained in a T-maze
under the same conditions. Garth and Mitchell’s experiment is perhaps
more convincing, though even here there remains the possibility that
their single Ruminia developed the habit of turning to the correct side
of the T by chance rather than as a result of training. It should be
noted that nearly all the animals that have been tested are terrestrial or
freshwater herbivores, and that attemps to regulate their behaviour
have been based on punishments for incorrect responses. In view of
the results with annelids (see Section 111, B, p. 27) it seems likely that
better performances would have occurred had the animals been handled
less and some system of rewards included in the training programme.
It is surprising to find that nothing has been done with the apparently far more promising marine prosobranchs. The neogastropods in
particular are active predators and scavengers, seeking out their food
in a purposeful manner which implies that it should be no great trouble
to arrange for reward-motivated learning in choice experiments.
B. Homing by limpets
The lack of direct evidence of learning in marine forms is to some
extent offset by a considerable body of information about the homing of
limpets. There has been considerable argument as to whether limpets
do in fact return to individual “ homes ’’ on the rocks. This has arisen
because: (1) limpets a t least in some habitats are comparatively
immobile anyway, so that observation of individuals in the same place
day after day is not good evidence of homing (Villee and Groody, 1940);
(2) the same species may behave quite differently on different substrates ; Patella vulgata L., for example, tends to be most active and to
change its “ home ” most often on smooth damp rocks (Orton, 1929 ;
Jones, 1948) ; and (3) there has been a tendency to compare findings
from different species, which do not all behave in the same way even
within the same habitat. Some species regularly return to a home,