THE ACRASINA
123
to the concentration of wild cells, (b) became proportionately smaller at
higher concentrations, or (c) was independent of concentration. It was
felt that (b) and (c) required special explanation, because linearity was
the proof that the initiator was a single cell. The reader may therefore
feel that Agg-91—an example of (b)—is an unreasonable choice for
analysis. On the contrary, it seems probable that (b) was the most
general result, and that (a) was obtained when the highest wild-cell
concentration used was below the zone of wild excess, and (c) when the
lowest was within it. It may be noted that in large, pure wild populations, the cells were 'in excess' at all densities above the 'optimum'. In
the mixtures, the role of inhibition of centres when only one of the components was in excess is not yet clear; but it certainly must have been
this that prevented the number of centres rising when both components
were present in excess.
Another difficulty in interpreting the numbers of centres in terms of
pre-existing initiator cells was that though the wild strain when alone
seemed to contain no more than 1 initiator in 2,200 cells, more were
revealed when it was mixed with a mutant, the precise number being
characteristic for each mutant, and the maximum observed being 1 in
76 wild cells. Sussman (1952) first supposed that wild cells possessed
varying capacities to initiate, and that the mutants were differentially
sensitive to the initiating stimulus and, though less sensitive than wild
cells, could reveal more initiators, because they could be added at higher
density. The fact that histidine increased the sensitivity of the test
system without increasing the number of primary centres formed at the
optimal density (Bradley et al., 1956) meant that this explanation
obtained only if the 'responder' cells responded to more than one factor.
The same conclusion followed from the finding of Ennis and Sussman
(1958a) that the increase in the ratio of centres to wild cells in mixtures
with a particular mutant did not correlate with the strength of the
mutant's chemotactic response to a wild aggregation on the opposite
side of a thin agar membrane. Calculations from their data show that
neither of these variables correlated with the ratio of the number of
centres to mutant cells in the mixtures. The independence of these three
variables emphasizes the complexity of the cellular interactions and
the difficulty of drawing a simple conclusion from the experiments on
mixtures.
Ennis and Sussman also discovered that aggregateless mutants could
induce extra centres to appear in wild populations that were separated
from the mutants by an agar membrane. This was further evidence that
the mutants were more than mere responders to the initiating stimulus,
and that the main proof (Sussman, 1952) that centres were started by
single specialized initiator cells was invalid. However, the authors sup-
123
to the concentration of wild cells, (b) became proportionately smaller at
higher concentrations, or (c) was independent of concentration. It was
felt that (b) and (c) required special explanation, because linearity was
the proof that the initiator was a single cell. The reader may therefore
feel that Agg-91—an example of (b)—is an unreasonable choice for
analysis. On the contrary, it seems probable that (b) was the most
general result, and that (a) was obtained when the highest wild-cell
concentration used was below the zone of wild excess, and (c) when the
lowest was within it. It may be noted that in large, pure wild populations, the cells were 'in excess' at all densities above the 'optimum'. In
the mixtures, the role of inhibition of centres when only one of the components was in excess is not yet clear; but it certainly must have been
this that prevented the number of centres rising when both components
were present in excess.
Another difficulty in interpreting the numbers of centres in terms of
pre-existing initiator cells was that though the wild strain when alone
seemed to contain no more than 1 initiator in 2,200 cells, more were
revealed when it was mixed with a mutant, the precise number being
characteristic for each mutant, and the maximum observed being 1 in
76 wild cells. Sussman (1952) first supposed that wild cells possessed
varying capacities to initiate, and that the mutants were differentially
sensitive to the initiating stimulus and, though less sensitive than wild
cells, could reveal more initiators, because they could be added at higher
density. The fact that histidine increased the sensitivity of the test
system without increasing the number of primary centres formed at the
optimal density (Bradley et al., 1956) meant that this explanation
obtained only if the 'responder' cells responded to more than one factor.
The same conclusion followed from the finding of Ennis and Sussman
(1958a) that the increase in the ratio of centres to wild cells in mixtures
with a particular mutant did not correlate with the strength of the
mutant's chemotactic response to a wild aggregation on the opposite
side of a thin agar membrane. Calculations from their data show that
neither of these variables correlated with the ratio of the number of
centres to mutant cells in the mixtures. The independence of these three
variables emphasizes the complexity of the cellular interactions and
the difficulty of drawing a simple conclusion from the experiments on
mixtures.
Ennis and Sussman also discovered that aggregateless mutants could
induce extra centres to appear in wild populations that were separated
from the mutants by an agar membrane. This was further evidence that
the mutants were more than mere responders to the initiating stimulus,
and that the main proof (Sussman, 1952) that centres were started by
single specialized initiator cells was invalid. However, the authors sup-