5. T H E L O C O M O T O R Y B E H A V I O U R OF CELLS
181
roundings might take the form of a change in this bias. For example, it
may have an independent rhythm of speed or of frequency of turning,
or it may show a non-random distribution of size of turn (a tendency to
oscillate in place; or conversely to deviate little from a straight line as
Levi and Meyer, 1941, believe neurites d o ) , or of direction of turn
(circling or alternating). Any of these may change in various ways, the
response nevertheless remaining unoriented in relation to the environment.
In an oriented response, the cell is reacting to the spatial pattern of
its environment. Speed or frequency of turning or amount of turning or,
indeed, any of the internal biases becomes different in different
environmentally specified directions. Such responses in animals—the
taxes—are commonly made to a vector property of the environment,
such as direction of lighting. But cells will more usually be responding
to a pattern of scalar properties, such as to a diffusion gradient of concentration which is detected by difference of intensity at different parts
of the cell surface or at successive moments in time, or to a discontinuity
of substrate structure simultaneously detected at different parts of the
cell.
C. U N O R I E N T E D RESPONSES
1. Speed
The speed of movement of an individual cell in vitro is the easiest form
of behaviour to measure roughly, at least when the cell is confined to a
two-dimensional substrate, but there are inevitable uncertainties even
then which limit precision. In the first place, an object so deformable,
externally and internally, has no simply-determined fixed point on
which to place the ruler. Inaccuracy due to this difficulty is diminished
by measuring speed over a distance long in relation to the cell size. In
the second place there are commonly considerable irregularities of
direction; and in practice the higher the rate of change of direction the
more the true speed is underestimated (Abercrombie and Heaysman,
1953). Inaccuracy from the second difficulty can be diminished by
reducing the time intervals at which measurements of distance are
made. These difficulties may at times introduce a bias, when comparing
different cells, which needs to be watched; but provided total distances
are long and time intervals short useful estimates may be obtained,
though rather few have been.
One way in which reliable estimates cannot as a general rule be
obtained has been much exploited. This is to measure, at daily intervals,
the linear distance moved by the edge of an outgrowth that is extending
from an explant. The most important questionable assumption in this
method is that the cells move radially outwards in a straight line.
Occasionally the substrate structure and hence cell movement is so
181
roundings might take the form of a change in this bias. For example, it
may have an independent rhythm of speed or of frequency of turning,
or it may show a non-random distribution of size of turn (a tendency to
oscillate in place; or conversely to deviate little from a straight line as
Levi and Meyer, 1941, believe neurites d o ) , or of direction of turn
(circling or alternating). Any of these may change in various ways, the
response nevertheless remaining unoriented in relation to the environment.
In an oriented response, the cell is reacting to the spatial pattern of
its environment. Speed or frequency of turning or amount of turning or,
indeed, any of the internal biases becomes different in different
environmentally specified directions. Such responses in animals—the
taxes—are commonly made to a vector property of the environment,
such as direction of lighting. But cells will more usually be responding
to a pattern of scalar properties, such as to a diffusion gradient of concentration which is detected by difference of intensity at different parts
of the cell surface or at successive moments in time, or to a discontinuity
of substrate structure simultaneously detected at different parts of the
cell.
C. U N O R I E N T E D RESPONSES
1. Speed
The speed of movement of an individual cell in vitro is the easiest form
of behaviour to measure roughly, at least when the cell is confined to a
two-dimensional substrate, but there are inevitable uncertainties even
then which limit precision. In the first place, an object so deformable,
externally and internally, has no simply-determined fixed point on
which to place the ruler. Inaccuracy due to this difficulty is diminished
by measuring speed over a distance long in relation to the cell size. In
the second place there are commonly considerable irregularities of
direction; and in practice the higher the rate of change of direction the
more the true speed is underestimated (Abercrombie and Heaysman,
1953). Inaccuracy from the second difficulty can be diminished by
reducing the time intervals at which measurements of distance are
made. These difficulties may at times introduce a bias, when comparing
different cells, which needs to be watched; but provided total distances
are long and time intervals short useful estimates may be obtained,
though rather few have been.
One way in which reliable estimates cannot as a general rule be
obtained has been much exploited. This is to measure, at daily intervals,
the linear distance moved by the edge of an outgrowth that is extending
from an explant. The most important questionable assumption in this
method is that the cells move radially outwards in a straight line.
Occasionally the substrate structure and hence cell movement is so