122
B. M. SHAFFER
it would be biologically surprising if there did exist initiators that could
be produced only during a stage when they were not needed and which
was not succeeded after any definite interval by one in which they were
used.
Mixtures of wild strain with aggregateless mutants. The proof that initiators were single cells (Sussman, 1952) was based on experiments in
which, so it was believed, the number of initiators and of cells responding to them could be varied independently by mixing wild cells with
mutants that were unable to aggregate alone. For example, the number
of centres that developed in populations containing a given density of
wild cells and covering a standard area increased with the density of
the mutant Agg-91 (Ennis and Sussman, 1958a) till it reached a constant value. The mutant was said then to be present in excess. The
number of centres was then linearly proportional to the number of wild
cells, namely 1 per 118 if the density of these cells was not too high,
though at greater densities the ratio decreased progressively. The authors
concluded that each centre was started by a single specialized wild cell,
and that on average 1 wild cell in 118 was an initiator.
Comment. The key assumption in these experiments was that only
the wild cells could be initiators and that mutants did no more than
respond to them. But an aggregateless mutant's developmental mechanism might have broken down at any one of a number of points, and
thus there was no basis for predicting its performance when closely associated with wild cells. The assumption was immediately disproved
(Sussman, 1952) by the fact that wild cells that were too dispersed to
aggregate by themselves could be made to do so by adding aggregateless cells; and indeed Sussman (1954) has since described many other
examples of synergistic development in mutant mixtures. From the
original data obtained in the experiment with Agg-91, it can be shown
that the number of centres that developed in populations containing a
given density of mutant cells increased with the density of wild cells
till it reached a constant value. The wild cells may be said then to have
been present in excess. The number of centres was then linearly proportional to the number of mutant cells, namely 1 per 25,000, if the
density of these cells were not too high, though at greater densities the
ratio decreased progressively. Therefore it is just as valid to conclude
that each centre was started by a single specialized mutant cell, and that
on average 1 mutant cell in 25,000 was an initiator, as that 1 wild cell
in 118 was one.
Those aggregateless mutants with which the number of centres
formed when mixed with wild cells reached a constant value in the
presence of sufficient mutant cells were classified by Ennis and Sussman
(1958a) according to whether this value (a) was always linearly related
B. M. SHAFFER
it would be biologically surprising if there did exist initiators that could
be produced only during a stage when they were not needed and which
was not succeeded after any definite interval by one in which they were
used.
Mixtures of wild strain with aggregateless mutants. The proof that initiators were single cells (Sussman, 1952) was based on experiments in
which, so it was believed, the number of initiators and of cells responding to them could be varied independently by mixing wild cells with
mutants that were unable to aggregate alone. For example, the number
of centres that developed in populations containing a given density of
wild cells and covering a standard area increased with the density of
the mutant Agg-91 (Ennis and Sussman, 1958a) till it reached a constant value. The mutant was said then to be present in excess. The
number of centres was then linearly proportional to the number of wild
cells, namely 1 per 118 if the density of these cells was not too high,
though at greater densities the ratio decreased progressively. The authors
concluded that each centre was started by a single specialized wild cell,
and that on average 1 wild cell in 118 was an initiator.
Comment. The key assumption in these experiments was that only
the wild cells could be initiators and that mutants did no more than
respond to them. But an aggregateless mutant's developmental mechanism might have broken down at any one of a number of points, and
thus there was no basis for predicting its performance when closely associated with wild cells. The assumption was immediately disproved
(Sussman, 1952) by the fact that wild cells that were too dispersed to
aggregate by themselves could be made to do so by adding aggregateless cells; and indeed Sussman (1954) has since described many other
examples of synergistic development in mutant mixtures. From the
original data obtained in the experiment with Agg-91, it can be shown
that the number of centres that developed in populations containing a
given density of mutant cells increased with the density of wild cells
till it reached a constant value. The wild cells may be said then to have
been present in excess. The number of centres was then linearly proportional to the number of mutant cells, namely 1 per 25,000, if the
density of these cells were not too high, though at greater densities the
ratio decreased progressively. Therefore it is just as valid to conclude
that each centre was started by a single specialized mutant cell, and that
on average 1 mutant cell in 25,000 was an initiator, as that 1 wild cell
in 118 was one.
Those aggregateless mutants with which the number of centres
formed when mixed with wild cells reached a constant value in the
presence of sufficient mutant cells were classified by Ennis and Sussman
(1958a) according to whether this value (a) was always linearly related