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
195
preference entirely on the surface of another cell ("contact promotion",
Curtis, 1960). It may be illustrated by "peripolesis", the restriction of
a moving lymphocyte to the surface of a tumour cell or macrophage
(Lewis and Webster, 1921; Humble, Jayne and Pulvertaft, 1956;
Sharp and Burwell, 1960). The association of Schwann cells and
neurites (Weiss, 1934; Nakai, 1956, 1960) may be similar, at least in
the initial stages before myelination starts. The aggregation of cells into
clumps, especially after enzymic disaggregation, is perhaps another
instance in this general class. Curtis (1960, 1962) has suggested that as
a result of the disaggregation procedure contact inhibition is temporarily
abrogated, and cells move over each other in preference to moving on
the surrounding substrate. It is possible that the ordinary mechanism
of cell locomotion is not involved here; it may be that the diminished
adhesion to substrate relative to adhesion to each other and absence of
intercellular substances, both produced by disaggregation, make it
possible for the cells slowly to shift their contacts until their areas of
mutual adhesion are as large as possible. Degree of adhesion to substrate certainly seems to be important: on a plasma substrate, to which
adhesion of all cells seems to be strong, disaggregated cells may fail to
reaggregate (Moscona and Moscona, 1952; Stefanelli, 1960).
In considering mechanisms of association one should not forget that
cells are contractile, and when their processes have made contact with
some suitable object, such as another similar cell, they may move
themselves by an active shortening of these links, producing closer
association: Twitty (1945) described this contractile movement in
differentiated melanoblasts in vitro; Nakai (1960) has suggested that
fasciculation of neurites may occur by contraction of the cross-adhesions
between lateral filopodia; and wound contraction and clot retraction
may be examples of the phenomenon in vivo (Abercrombie, 1957a).
Tensions are likely to be produced by the drawing more closely
together of the cells that are involved in associative behaviour, however
it is produced. Such tensions have been observed in clumps of reaggregated cells (Loeb, 1922a; Faure-Fremiet, 1932), and are also
strongly marked in cultures (e.g. Mayer, 1933) even when clumping
fails to occur.
Incipient cell degeneration also leads to clumping of cells in cultures,
reminiscent of the clumping produced by the standard methods of
disaggregation, and perhaps depending on the same processes.
C. T R E N D S OF D I R E C T I O N
The assumption of a common direction of movement by numbers of
contiguous cells is perhaps the population behaviour of supreme
195
preference entirely on the surface of another cell ("contact promotion",
Curtis, 1960). It may be illustrated by "peripolesis", the restriction of
a moving lymphocyte to the surface of a tumour cell or macrophage
(Lewis and Webster, 1921; Humble, Jayne and Pulvertaft, 1956;
Sharp and Burwell, 1960). The association of Schwann cells and
neurites (Weiss, 1934; Nakai, 1956, 1960) may be similar, at least in
the initial stages before myelination starts. The aggregation of cells into
clumps, especially after enzymic disaggregation, is perhaps another
instance in this general class. Curtis (1960, 1962) has suggested that as
a result of the disaggregation procedure contact inhibition is temporarily
abrogated, and cells move over each other in preference to moving on
the surrounding substrate. It is possible that the ordinary mechanism
of cell locomotion is not involved here; it may be that the diminished
adhesion to substrate relative to adhesion to each other and absence of
intercellular substances, both produced by disaggregation, make it
possible for the cells slowly to shift their contacts until their areas of
mutual adhesion are as large as possible. Degree of adhesion to substrate certainly seems to be important: on a plasma substrate, to which
adhesion of all cells seems to be strong, disaggregated cells may fail to
reaggregate (Moscona and Moscona, 1952; Stefanelli, 1960).
In considering mechanisms of association one should not forget that
cells are contractile, and when their processes have made contact with
some suitable object, such as another similar cell, they may move
themselves by an active shortening of these links, producing closer
association: Twitty (1945) described this contractile movement in
differentiated melanoblasts in vitro; Nakai (1960) has suggested that
fasciculation of neurites may occur by contraction of the cross-adhesions
between lateral filopodia; and wound contraction and clot retraction
may be examples of the phenomenon in vivo (Abercrombie, 1957a).
Tensions are likely to be produced by the drawing more closely
together of the cells that are involved in associative behaviour, however
it is produced. Such tensions have been observed in clumps of reaggregated cells (Loeb, 1922a; Faure-Fremiet, 1932), and are also
strongly marked in cultures (e.g. Mayer, 1933) even when clumping
fails to occur.
Incipient cell degeneration also leads to clumping of cells in cultures,
reminiscent of the clumping produced by the standard methods of
disaggregation, and perhaps depending on the same processes.
C. T R E N D S OF D I R E C T I O N
The assumption of a common direction of movement by numbers of
contiguous cells is perhaps the population behaviour of supreme