194
M. A B E R C R O M B I E
arrangement of epithelial cells is presumably produced by a reaction
similar to the contact inhibition of fibroblasts, but here dispersion is
limited by mutual adhesion.
B. A S S O C I A T I V E B E H A V I O U R
Associative behaviour builds up a local concentration of cells, and
counterbalances the natural "diffusive" tendency to spread. It may
take place in various ways. A local increase of cell concentration within
a larger population is theoretically possible through the action of
kineses, the opposite of that mentioned under dispersive behaviour, but
is hardly likely to be common. Taxes oriented to a source of stimulant
independent of the reacting cells have also to be considered. And
mutually oriented responses of cells to each other may also be involved.
Considering first the chemotactic responses, we may note that a largescale positive chemotaxis would tend to concentrate cells at its source,
as in the well-known grouping of white cells round a bacterial focus; a
negative chemotaxis would tend to concentrate cells where the gradient
ceases to be steep enough to produce an oriented reaction, and such a
region is often clearly detectable at either side of the "no-man's-land"
which results from placing explants of buffy coat near together (Carrel
and Ebeling, 1922; Oldfield, 1963). When cells show a positive chemotaxis towards each other, as in their aggregative phase the cellular
slime moulds do, association will obviously be highly efficient. No
similar case amongst animal cells has, however, been reported.
The existence of contact reactions that tend to produce associative
behaviour has already been mentioned (p. 190). The nature of the
reactions is not clearly enough known for one to be able to do more than
list some of the phenomena in which they are probably involved, and
suggest some possible mechanisms.
The difference between a loose mesenchyme and an epithelium, as
Lewis (1922) Loeb (1922b) and Levi (1925) early pointed out, is a
matter of degree of mutual adhesion. An epithelial cell tends to adhere
edge-to-edge to other similar cells, often in such a way that it can move,
but only provided it does not move away (Weiss, 1958). This may well
be a phenomenon related to that of the formation of bundles of neurites
(fasciculation) as a result of each neurite extending parallel to and in
contact with an already established one, a mechanism postulated by
Harrison (1910) and Weiss (1941) and observed in vitro by Stefanelli
(I960). Parallel orientation somewhat similar to the fasciculation of
neurites is observed amongst spindle cells from spinal ganglia (Weiss,
1945) and in certain fibroblasts growing on glass (Macpherson and
Stoker, 1962).
Related again, perhaps, are cases where a cell seems to move by
M. A B E R C R O M B I E
arrangement of epithelial cells is presumably produced by a reaction
similar to the contact inhibition of fibroblasts, but here dispersion is
limited by mutual adhesion.
B. A S S O C I A T I V E B E H A V I O U R
Associative behaviour builds up a local concentration of cells, and
counterbalances the natural "diffusive" tendency to spread. It may
take place in various ways. A local increase of cell concentration within
a larger population is theoretically possible through the action of
kineses, the opposite of that mentioned under dispersive behaviour, but
is hardly likely to be common. Taxes oriented to a source of stimulant
independent of the reacting cells have also to be considered. And
mutually oriented responses of cells to each other may also be involved.
Considering first the chemotactic responses, we may note that a largescale positive chemotaxis would tend to concentrate cells at its source,
as in the well-known grouping of white cells round a bacterial focus; a
negative chemotaxis would tend to concentrate cells where the gradient
ceases to be steep enough to produce an oriented reaction, and such a
region is often clearly detectable at either side of the "no-man's-land"
which results from placing explants of buffy coat near together (Carrel
and Ebeling, 1922; Oldfield, 1963). When cells show a positive chemotaxis towards each other, as in their aggregative phase the cellular
slime moulds do, association will obviously be highly efficient. No
similar case amongst animal cells has, however, been reported.
The existence of contact reactions that tend to produce associative
behaviour has already been mentioned (p. 190). The nature of the
reactions is not clearly enough known for one to be able to do more than
list some of the phenomena in which they are probably involved, and
suggest some possible mechanisms.
The difference between a loose mesenchyme and an epithelium, as
Lewis (1922) Loeb (1922b) and Levi (1925) early pointed out, is a
matter of degree of mutual adhesion. An epithelial cell tends to adhere
edge-to-edge to other similar cells, often in such a way that it can move,
but only provided it does not move away (Weiss, 1958). This may well
be a phenomenon related to that of the formation of bundles of neurites
(fasciculation) as a result of each neurite extending parallel to and in
contact with an already established one, a mechanism postulated by
Harrison (1910) and Weiss (1941) and observed in vitro by Stefanelli
(I960). Parallel orientation somewhat similar to the fasciculation of
neurites is observed amongst spindle cells from spinal ganglia (Weiss,
1945) and in certain fibroblasts growing on glass (Macpherson and
Stoker, 1962).
Related again, perhaps, are cases where a cell seems to move by