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
197
1934). The ability of populations to orient the substrate around them
is both important for the explanation of behaviour in vitro in a plasma
clot, and, as Weiss has pointed out, can be applied to explain many of
the organized movements of cells and nerve fibres in vivo.
Weiss (1945) suggested that cells can no longer be thought to behave
as individuals; they behave as a population unit because any group of
them exudes a common "ground-mat" on the solid substrate. This
material, he supposed, becomes oriented by its own flow, and the cells
then react to it, showing contact guidance. The evidence that such a
ground-mat appears from tissues other than the degenerating peripheral
nerve that Weiss was observing is not strong. Since the directional
movement of cells from an explant can now be accounted for in terms
of a known reaction—contact inhibition—the hypothesis is at present
unnecessary.
There may be other ways, as yet not properly analysed, by which
contact-mediated responses produce general trends of movement. The
concerted movement of epithelial cells is one of the cases not yet
adequately explained. O n a plane surface an epithelium moves en masse
as a sheet of cells, with ruffled membranes only at the free edge. In a
clot, it moves often in the form of strands, again with only terminal
pseudopodial activity (Holmes, 1913). Individual cells may, however,
break free from the sheets or strands, and move by themselves (Holmes,
1913; Lewis, 1923; Levi, 1925). Within a sheet there may be a certain
freedom of cells to alter their relative positions (Herrick, 1932; Wilbur
and Chambers, 1942). The sheet is often under great tension (Lewis,
1923), and retracts instantly if cut; but this is not always so, and sometimes a large area may be firmly attached to the substrate (Wilbur and
Chambers, 1942). It seems likely from this evidence that at times only
the peripheral cells are doing the work, dragging the other cells behind
them, while at other times all the cells are contributing to the movement. It has been suggested that when adherent cells move in tandem,
the leading cell may guide the locomotion of the following one by the
tension it puts upon it (Abercrombie and Ambrose, 1958; Abercrombie,
1961a). A similar suggestion has been made for slime moulds, and
appropriately termed "contact following" (Shaffer, 1962). Something
like this may happen in epithelia. If so, the cessation of movement that
occurs throughout the confluent sheet when one epithelial sheet runs
into another would then be explicable even when all the cells were
moving actively; the mutual contact inhibition of the colliding peripheral cells would release the tension on the cells behind. It remains to
collect some evidence bearing on the proposed mechanism.
197
1934). The ability of populations to orient the substrate around them
is both important for the explanation of behaviour in vitro in a plasma
clot, and, as Weiss has pointed out, can be applied to explain many of
the organized movements of cells and nerve fibres in vivo.
Weiss (1945) suggested that cells can no longer be thought to behave
as individuals; they behave as a population unit because any group of
them exudes a common "ground-mat" on the solid substrate. This
material, he supposed, becomes oriented by its own flow, and the cells
then react to it, showing contact guidance. The evidence that such a
ground-mat appears from tissues other than the degenerating peripheral
nerve that Weiss was observing is not strong. Since the directional
movement of cells from an explant can now be accounted for in terms
of a known reaction—contact inhibition—the hypothesis is at present
unnecessary.
There may be other ways, as yet not properly analysed, by which
contact-mediated responses produce general trends of movement. The
concerted movement of epithelial cells is one of the cases not yet
adequately explained. O n a plane surface an epithelium moves en masse
as a sheet of cells, with ruffled membranes only at the free edge. In a
clot, it moves often in the form of strands, again with only terminal
pseudopodial activity (Holmes, 1913). Individual cells may, however,
break free from the sheets or strands, and move by themselves (Holmes,
1913; Lewis, 1923; Levi, 1925). Within a sheet there may be a certain
freedom of cells to alter their relative positions (Herrick, 1932; Wilbur
and Chambers, 1942). The sheet is often under great tension (Lewis,
1923), and retracts instantly if cut; but this is not always so, and sometimes a large area may be firmly attached to the substrate (Wilbur and
Chambers, 1942). It seems likely from this evidence that at times only
the peripheral cells are doing the work, dragging the other cells behind
them, while at other times all the cells are contributing to the movement. It has been suggested that when adherent cells move in tandem,
the leading cell may guide the locomotion of the following one by the
tension it puts upon it (Abercrombie and Ambrose, 1958; Abercrombie,
1961a). A similar suggestion has been made for slime moulds, and
appropriately termed "contact following" (Shaffer, 1962). Something
like this may happen in epithelia. If so, the cessation of movement that
occurs throughout the confluent sheet when one epithelial sheet runs
into another would then be explicable even when all the cells were
moving actively; the mutual contact inhibition of the colliding peripheral cells would release the tension on the cells behind. It remains to
collect some evidence bearing on the proposed mechanism.