THE ACRASINA
145
directions the secretion front is propagated, and on whether secretion
is maintained or not, and whether the peripheral secretors can be guided
by contact following or whether they need a succession of triggered
fronts and thus may converge on any new triggers that become active.
One possible reason for the high frequency of aggregations without
obvious centres when certain species are under water or a thin layer of
agar may be that repetitive activity of individual triggers is depressed,
like trigger activity in general. Likewise, one would expect aggregation
towards a point to be discouraged if circulating pulses managed to
become established. Trigger inhibition by the environment could also
explain the observations that led Harper (1926) to suppose that negative hydrotaxis played some part in aggregation. But there are other
obviously relevant phenomena about which it is pointless to speculate
until they have been further investigated. For example, in some species,
and most commonly D. purpureum under agar, independent, parallel
streams may be found leading to the feeding front.
Although a centre cannot dominate orientation throughout a large
aggregation area by its own massive secretion of attractor, but must rely
on the peripheral relays, this is not to say that the concentration of
attractor may not be higher at the centre than elsewhere. The fact that
cells in some cases duplicated centres, but not streams, on the other
side of an agar membrane (Sussman and Lee, 1955; Ennis and Sussman,
1958a) indeed seems to be good evidence for a quantitative or qualitative difference in their acrasin outputs; but possibly this was merely a
result of the geometry of diffusion. My own measurements on the large
Dictyostelium species have shown that continuous streams, centres, and
grex, all commonly secrete attractor at about the same concentration,
and that such differences as do occur are not primarily related to mass.
This contradicts the assumption, made by everyone from Harper (1929)
onwards who has considered the point, that attractiveness is basically additive and therefore directly proportional to the number of cells
in an aggregate, though Arndt had to suppose that they sometimes did
not act at full strength. On the contrary, output per cell must be inversely related to the size of the aggregate. And we must assume, as a
further complication of the control of secretion, that output may be
limited by external acrasin. Restriction of secretion is in any case biologically desirable to prevent the guidance mechanism of responding
cells from being overloaded. Such restriction seems to be another way in
which a group of cells, temporarily or permanently more active, could
dominate its neighbours.
In some species and conditions, even in the absence of competition
from other aggregations, centres may not acquire any streams at all, and
growth stops when they either develop into grex or cease to be attrac-
145
directions the secretion front is propagated, and on whether secretion
is maintained or not, and whether the peripheral secretors can be guided
by contact following or whether they need a succession of triggered
fronts and thus may converge on any new triggers that become active.
One possible reason for the high frequency of aggregations without
obvious centres when certain species are under water or a thin layer of
agar may be that repetitive activity of individual triggers is depressed,
like trigger activity in general. Likewise, one would expect aggregation
towards a point to be discouraged if circulating pulses managed to
become established. Trigger inhibition by the environment could also
explain the observations that led Harper (1926) to suppose that negative hydrotaxis played some part in aggregation. But there are other
obviously relevant phenomena about which it is pointless to speculate
until they have been further investigated. For example, in some species,
and most commonly D. purpureum under agar, independent, parallel
streams may be found leading to the feeding front.
Although a centre cannot dominate orientation throughout a large
aggregation area by its own massive secretion of attractor, but must rely
on the peripheral relays, this is not to say that the concentration of
attractor may not be higher at the centre than elsewhere. The fact that
cells in some cases duplicated centres, but not streams, on the other
side of an agar membrane (Sussman and Lee, 1955; Ennis and Sussman,
1958a) indeed seems to be good evidence for a quantitative or qualitative difference in their acrasin outputs; but possibly this was merely a
result of the geometry of diffusion. My own measurements on the large
Dictyostelium species have shown that continuous streams, centres, and
grex, all commonly secrete attractor at about the same concentration,
and that such differences as do occur are not primarily related to mass.
This contradicts the assumption, made by everyone from Harper (1929)
onwards who has considered the point, that attractiveness is basically additive and therefore directly proportional to the number of cells
in an aggregate, though Arndt had to suppose that they sometimes did
not act at full strength. On the contrary, output per cell must be inversely related to the size of the aggregate. And we must assume, as a
further complication of the control of secretion, that output may be
limited by external acrasin. Restriction of secretion is in any case biologically desirable to prevent the guidance mechanism of responding
cells from being overloaded. Such restriction seems to be another way in
which a group of cells, temporarily or permanently more active, could
dominate its neighbours.
In some species and conditions, even in the absence of competition
from other aggregations, centres may not acquire any streams at all, and
growth stops when they either develop into grex or cease to be attrac-