144
B. M. SHAFFER
paths taken by propagated fronts of activity. Experimental barriers to
their spread deform the pattern as expected.
The next point is how a secretion front is started. The simplest assumption is that the trigger, whether a cell or cell group, is merely slightly
ahead in development—a case of primus inter pares. I t could, for
example, have the lowest threshold in the population and thus respond
by much increased secretion to small amounts of inducer secreted spontaneously by other cells. But on this basis, it would not be easy to account for the origin of successive fronts, both because we have already
had to assume that the transmitting cells 'recover' some time before they
relay the next front, and because such fronts may travel from a sparser
to a denser area of a stippled aggregation. Moreover, in P. violaceum
founders cannot be induced to appear simply by heaping cells together,
yet in low-density cultures they become active when distant from their
nearest neighbours. Furthermore (Shaffer, 1961a), both darkness and
sandwiching the cells somewhat inhibit the differentiation of founders,
and in combination they do so very severely; but neither factor hinders
the growth of streams, once started, and in fact sandwiching slightly
facilitates the operation of the relays, and we have supposed that it may
lower the inducer threshold. The situation in Acytostelium is still more
striking (Shaffer, 1959; 1961b). Sandwiched cells are unable to aggregate spontaneously even when illuminated; but a homologous aggregation—though not one of D. discoideum or P. violaceum—at the edge or
on top of the agar overlay induces them to form streams which are commonly far more extensive than its own, again indicating the enhanced
functioning of the relays in the sandwich layer. In Acytostelium, as in
P. violaceum, an overlying aggregation does not induce founders, so the
streams wander about rather aimlessly till they form rings or a network;
and the relations of these structures to the original aggregation show
that their cells have been dominated by the attractor they have made
themselves. How then does an inducing aggregate act? Perhaps it
supplies a factor that the sandwiched cells remain unable to produce.
But the more appealing possibility is that it simply exposes them to
supraliminal inducer, after which they are able to secrete this. It is true
that some time after the overlying aggregate has been removed or transformed into fruiting bodies, induced streams will disperse; and so too will
an inducing aggregate itself, or even a culminating grex, if covered with
agar; but this does not necessarily support the first explanation. Thus,
from various species, there is evidence that quite different factors are
involved in 'pulling the trigger' and propagating the front.
What is the relation between triggers and centres? Clearly, an active
trigger will be the temporary focus of aggregation, but whether or not it
will become the nucleus of a centre will depend on how far and in what
B. M. SHAFFER
paths taken by propagated fronts of activity. Experimental barriers to
their spread deform the pattern as expected.
The next point is how a secretion front is started. The simplest assumption is that the trigger, whether a cell or cell group, is merely slightly
ahead in development—a case of primus inter pares. I t could, for
example, have the lowest threshold in the population and thus respond
by much increased secretion to small amounts of inducer secreted spontaneously by other cells. But on this basis, it would not be easy to account for the origin of successive fronts, both because we have already
had to assume that the transmitting cells 'recover' some time before they
relay the next front, and because such fronts may travel from a sparser
to a denser area of a stippled aggregation. Moreover, in P. violaceum
founders cannot be induced to appear simply by heaping cells together,
yet in low-density cultures they become active when distant from their
nearest neighbours. Furthermore (Shaffer, 1961a), both darkness and
sandwiching the cells somewhat inhibit the differentiation of founders,
and in combination they do so very severely; but neither factor hinders
the growth of streams, once started, and in fact sandwiching slightly
facilitates the operation of the relays, and we have supposed that it may
lower the inducer threshold. The situation in Acytostelium is still more
striking (Shaffer, 1959; 1961b). Sandwiched cells are unable to aggregate spontaneously even when illuminated; but a homologous aggregation—though not one of D. discoideum or P. violaceum—at the edge or
on top of the agar overlay induces them to form streams which are commonly far more extensive than its own, again indicating the enhanced
functioning of the relays in the sandwich layer. In Acytostelium, as in
P. violaceum, an overlying aggregation does not induce founders, so the
streams wander about rather aimlessly till they form rings or a network;
and the relations of these structures to the original aggregation show
that their cells have been dominated by the attractor they have made
themselves. How then does an inducing aggregate act? Perhaps it
supplies a factor that the sandwiched cells remain unable to produce.
But the more appealing possibility is that it simply exposes them to
supraliminal inducer, after which they are able to secrete this. It is true
that some time after the overlying aggregate has been removed or transformed into fruiting bodies, induced streams will disperse; and so too will
an inducing aggregate itself, or even a culminating grex, if covered with
agar; but this does not necessarily support the first explanation. Thus,
from various species, there is evidence that quite different factors are
involved in 'pulling the trigger' and propagating the front.
What is the relation between triggers and centres? Clearly, an active
trigger will be the temporary focus of aggregation, but whether or not it
will become the nucleus of a centre will depend on how far and in what