12. G E N E T I C C Y T O L O G Y
427
cells, which had a generation time of 11-12 h. From the high frequency
of chromatid breaks and exchanges, it may be postulated that incorporation of a fraudulent pyrimidine deoxyriboside induces instability in
the new subunit, which reduplicates itself in the next cell generation
with the break remaining open. However, the results of Somers and
Hsu (1961), using B U D R treatment of only 6-8 h to induce breakages
and exchanges, cannot be explained with the half-chromatid theory
unless B U D R can replace thymidine in the old half or unless an instability in the new D N A strand would induce an isolocus break in the
old. In their studies on chromosome breakages induced by 5-fluorodeoxyuridine ( F U D R ) , Taylor, Haut and Tung (1962) postulated that
isolocus breaks may be caused by the torsion in prophase when the
chromosomes spiralize.
Taylor noted an interesting phenomenon in the chromosomes of
the second metaphase, namely, a chromatid frequently would be
labeled for only a part of its length, and its sister would be labeled from
the point where the labeling stopped to the end or to another point
of exchange. If one adds the labeled segments and nonlabeled segments,
he can, in the majority of cases, account for one completely labeled and
one completely unlabeled chromatid for each chromosome.
The phenomenon of exchanges between chromatids provided a tool
to probe the nature ofthe sister (half) chromatids. If the D N A molecule
is composed of two complementary Watson-Crick strands, would the two
half-chromatids represent the complementary D N A strands? In other
words, are the two half-chromatids also molecularly complementary? If
they are, each must synthesize its complementary chromatid. If we term
the two original half-chromatids A and B, A must synthesize a B halfchromatid, and B an A half-chromatid. Furthermore, there can be no
exchanges between the two original half-chromatids A and B because
they are complementary. Thus exchange must occur when four strands
are present, i.e. during or after the D N A synthesis. The original A strand
can exchange with the A strand newly synthesized by the original B
strand; and the reverse applies to the B half. T o test this hypothesis,
one must assume that the frequency of exchanges is constant during
every cell generation. When an exchange occurs during the first generation, it must involve one newly synthesized half-chromatid which is
labeled. The two chromatids, therefore, would bear opposite labeled
segments. Ordinarily these two chromatids separate into daughter cells.
If, however, colchicine is present to prevent anaphasic disjunction,
the cell would become tetraploid and the two original chromatids
would be present in the next metaphase as two chromosomes. The
exchanged segments on these two chromosomes would be complementary as far as label is concerned. These are termed twin exchanges.
427
cells, which had a generation time of 11-12 h. From the high frequency
of chromatid breaks and exchanges, it may be postulated that incorporation of a fraudulent pyrimidine deoxyriboside induces instability in
the new subunit, which reduplicates itself in the next cell generation
with the break remaining open. However, the results of Somers and
Hsu (1961), using B U D R treatment of only 6-8 h to induce breakages
and exchanges, cannot be explained with the half-chromatid theory
unless B U D R can replace thymidine in the old half or unless an instability in the new D N A strand would induce an isolocus break in the
old. In their studies on chromosome breakages induced by 5-fluorodeoxyuridine ( F U D R ) , Taylor, Haut and Tung (1962) postulated that
isolocus breaks may be caused by the torsion in prophase when the
chromosomes spiralize.
Taylor noted an interesting phenomenon in the chromosomes of
the second metaphase, namely, a chromatid frequently would be
labeled for only a part of its length, and its sister would be labeled from
the point where the labeling stopped to the end or to another point
of exchange. If one adds the labeled segments and nonlabeled segments,
he can, in the majority of cases, account for one completely labeled and
one completely unlabeled chromatid for each chromosome.
The phenomenon of exchanges between chromatids provided a tool
to probe the nature ofthe sister (half) chromatids. If the D N A molecule
is composed of two complementary Watson-Crick strands, would the two
half-chromatids represent the complementary D N A strands? In other
words, are the two half-chromatids also molecularly complementary? If
they are, each must synthesize its complementary chromatid. If we term
the two original half-chromatids A and B, A must synthesize a B halfchromatid, and B an A half-chromatid. Furthermore, there can be no
exchanges between the two original half-chromatids A and B because
they are complementary. Thus exchange must occur when four strands
are present, i.e. during or after the D N A synthesis. The original A strand
can exchange with the A strand newly synthesized by the original B
strand; and the reverse applies to the B half. T o test this hypothesis,
one must assume that the frequency of exchanges is constant during
every cell generation. When an exchange occurs during the first generation, it must involve one newly synthesized half-chromatid which is
labeled. The two chromatids, therefore, would bear opposite labeled
segments. Ordinarily these two chromatids separate into daughter cells.
If, however, colchicine is present to prevent anaphasic disjunction,
the cell would become tetraploid and the two original chromatids
would be present in the next metaphase as two chromosomes. The
exchanged segments on these two chromosomes would be complementary as far as label is concerned. These are termed twin exchanges.