THE ACRASINA
133
the same high-power microscope field each yield 4 daughters almost
simultaneously. If this synchronization was not accidental, the factor
involved may be responsible for some of the divergences in the accounts
of cell behaviour. I have also found discoideum cells with as many as
10 nuclei, and multinucleate cells tearing themselves apart without
nuclear division. Multinucleate giants are produced in culture by many
amoeboid cells, often in response to particular environments (Barta,
1925; Lewis, 1927); but it is still quite uncertain whether in the Acrasina
they arise by nuclear division, or whether the nonplasmodial slime
moulds do, after all, form plasmodia—as possibly they also do in macrocysts (Blaskovics and Raper, 1957). At least in certain of my discoideum
cultures in which many cells in some parts of the feeding front were
multinucleate, engulfment was very rare. I t may be that, as Ross (1960)
has suggested, sexuality is as variable as much of the rest of slime-mould
behaviour, but the present state of our knowledge is most unsatisfactory.
Further investigation is urgently required because of the morphogenetic
importance of the surface changes that decide whether cells that come
in contact engulf one another, or fuse, or form a stable association
without trying to do either, or separate again without hindrance.
Finally, we may note that even the question of whether cell division
of any kind occurs during morphogenesis has been much disputed. The
traditional view was that there was none, but Bonner and Frascella
(1952) reported that some cells did divide, and Wilson (1952) held that
all of them did. R. R. Sussman and Sussman (1960) sought to determine
whether there was any obligatory relationship. Having chosen a diploid
strain of D. discoideum, they claimed to have observed a considerable
increase in cell numbers during aggregation and fruiting if the cells had
still been full of bacteria when subjected to preparative centrifuging,
but no increase if the cells had stopped feeding before then. However,
there measurements did not exclude the strong probability that the cells
with included bacteria attained exactly the same physiological state as
the others—and therefore completed their divisions—before aggregating. The sole evidence that division did occur during the morphogenetic phase was that, with the cells containing bacteria, the average
percentage increase in very small populations that happened not to
aggregate was given as 17-1, as against 32-8 in those that did aggregate,
the second figure being measured 12 hours later, during which fruiting
bodies were built. Unfortunately, the absolute increases were expressed
as percentages of the sums of the initial and final values in the first
group, but of the means of these values in the second group. When the
second figure is calculated on the same basis as the first one, it becomes
16-4—in such good agreement with 17-1 that it proves the opposite
conclusion, namely that these cells did not divide during fruiting.
133
the same high-power microscope field each yield 4 daughters almost
simultaneously. If this synchronization was not accidental, the factor
involved may be responsible for some of the divergences in the accounts
of cell behaviour. I have also found discoideum cells with as many as
10 nuclei, and multinucleate cells tearing themselves apart without
nuclear division. Multinucleate giants are produced in culture by many
amoeboid cells, often in response to particular environments (Barta,
1925; Lewis, 1927); but it is still quite uncertain whether in the Acrasina
they arise by nuclear division, or whether the nonplasmodial slime
moulds do, after all, form plasmodia—as possibly they also do in macrocysts (Blaskovics and Raper, 1957). At least in certain of my discoideum
cultures in which many cells in some parts of the feeding front were
multinucleate, engulfment was very rare. I t may be that, as Ross (1960)
has suggested, sexuality is as variable as much of the rest of slime-mould
behaviour, but the present state of our knowledge is most unsatisfactory.
Further investigation is urgently required because of the morphogenetic
importance of the surface changes that decide whether cells that come
in contact engulf one another, or fuse, or form a stable association
without trying to do either, or separate again without hindrance.
Finally, we may note that even the question of whether cell division
of any kind occurs during morphogenesis has been much disputed. The
traditional view was that there was none, but Bonner and Frascella
(1952) reported that some cells did divide, and Wilson (1952) held that
all of them did. R. R. Sussman and Sussman (1960) sought to determine
whether there was any obligatory relationship. Having chosen a diploid
strain of D. discoideum, they claimed to have observed a considerable
increase in cell numbers during aggregation and fruiting if the cells had
still been full of bacteria when subjected to preparative centrifuging,
but no increase if the cells had stopped feeding before then. However,
there measurements did not exclude the strong probability that the cells
with included bacteria attained exactly the same physiological state as
the others—and therefore completed their divisions—before aggregating. The sole evidence that division did occur during the morphogenetic phase was that, with the cells containing bacteria, the average
percentage increase in very small populations that happened not to
aggregate was given as 17-1, as against 32-8 in those that did aggregate,
the second figure being measured 12 hours later, during which fruiting
bodies were built. Unfortunately, the absolute increases were expressed
as percentages of the sums of the initial and final values in the first
group, but of the means of these values in the second group. When the
second figure is calculated on the same basis as the first one, it becomes
16-4—in such good agreement with 17-1 that it proves the opposite
conclusion, namely that these cells did not divide during fruiting.