292 SIDMAN
When we came to examine those slides from the first experiment, we
were gratified to see with the microscope that there was indeed a pattern
of labeled cells and that the pattern in the embryonic brain was different
at each of the three time points.^11 In the brain sampled at one hour after
injection, most of the radioactive cell nuclei were oval-shaped with their
long axis radial to the brain surfaces, and were located at a distance from
the inner surface of the brain, while in the six-hour specimen, most labeled
nuclei were near the inner surface (that is, the ventricular surface), and all
cells that were actually dividing at the moment of fixation were radioactive.
It took some time to figure out what this meant, but the two keys were
already in the published literature. First, I found a trio of very obscure,
largely forgotten papers from the mid-1930s by a Midwestern embryology
professor named Frederick Sauer, in which he showed that many so-called
“multi-layered” epithelia were actually composed of pseudo-stratified,
elongated cells with their nuclei at different distances from the surface.^12
The nuclei of these cells, he inferred correctly, dynamically moved toward
the inner surface as the cells prepared to divide, and cell division actually
took place at that surface. It seemed that in our specimens, those cell
nuclei which lay at a distance from the surface were the ones to become
radioactive, as seen at one hour after injection of tritiated thymidine,
and that those same nuclei must then move toward the inner surface of
the brain, taking about six hours to get there, and divide at that surface.
The other key publication was a more recent and timely one–a brief,
conceptually vital paper by the Strangeways Laboratory investigator,
Stephen Pelc, which established that cells replicate their DNA prior
to cell division, not during cell division.^13 Pelc was responsible for the
nomenclature everyone has come to use: S for the DNA synthesis phase,
M for the mitosis phase, G1 for the gap phase between mitosis and
synthesis, and G2 for the later gap phase between synthesis and mitosis.^14
In one of those rare flashes of insight that make the labor of scien
tific work unmatchably rewarding,^15 the conclusion seemed to me un
avoidable that in the developing brain, cells in S phase have their
nuclei at a distance from the ventricular surface, and that those nuclei
translocate toward the surface during G2, go through mitosis, M, at the
surface and then withdraw again from the ventricular surface during
G1. Examination of our 24-hour specimen indicated that some of the