sm is always changing, with the result that the creature moves in
what approximates to a straight line, being however actually a spiral
about the general line of progress. This method of motion is strikingly
like the circumnutation of a plant, the apex of which also describes a
spiral about the general line of growth. A rooted plant obviously cannot
rotate on its axis, but the regular series of curvatures of which its
growth consists correspond to the aberrations of Paramoecium distributed
regularly about its course by means of rotation. (In my address to the
Biological Section of the British Association at Cardiff (1891) I
have attempted to show the connection between circumnutation and
RECTIPETALITY, i.e. the innate capacity of growing in a straight line.)
Just as a plant changes its direction of growth by an exaggeration of
one of the curvature-elements of which circumnutation consists, so
does a Paramoecium change its course by the accentuation of one of
the deviations of which its path is built. Jennings has shown that the
infusoria, etc., react to stimuli by what is known as the "method of
trial." If an organism swims into a region where the temperature is too
high or where an injurious substance is present, it changes its course.
It then moves forward again, and if it is fortunate enough to escape the
influence, it continues to swim in the given direction. If however its
change of direction leads it further into the heated or poisonous region
it repeats the movement until it emerges from its difficulties. Jennings
finds in the movements of the lower organisms an analogue with what
is known as pain in conscious organisms. There is certainly this much
resemblance that a number of quite different sub-injurious agencies
produce in the lower organisms a form of reaction by the help of which
they, in a partly fortuitous way, escape from the threatening element
in their environment. The higher animals are stimulated in a parallel
manner to vague and originally purposeless movements, one of which
removes the discomfort under which they suffer, and the organism finally
learns to perform the appropriate movement without going through the
tentative series of actions.

I am tempted to recognise in circumnutation a similar groundwork of
tentative movements out of which the adaptive ones were originally
selected by a process rudely representative of learning by experience.

It is, however, simpler to confine ourselves to the a