MAIZE GENETICS

Barpara McCrintock

CONTINUATION OF THE STUDY OF THE
Invuction oF New Mutants IN
CHROMOSOME 9

The study of the induction of new
mutants in the short arm of chromosome
g, following the chromatid and chromo-
some breakage cycles involving this arm,
is nearing completion. At present, nine
new mutant types have been isolated that
are located in the short arm of chromo-
some 9. The mutants pale-yellow seedling
(pyd), white seedling (wd), and yellow-
green seedling and plant (yg) have been
repeatedly induced by the breakage-fusion
cycles. Only one each of the following
mutants has been obtained: bronze aleu-
rone, virescent seedling, pale-green seed-
ling, small-kernel, narrow-leaf, and light-
green plant. Only in one branch of this
study was an attempt made to determine
the proportion of all newly arising seed-
ling or kernel mutants that are located in
chromosome g. Three hundred and eighty-
three plants that showed the chromosome
type of breakage cycle in their early devel-
opmental stages were self-pollinated and
the kernels and seedling progeny observed,

Thirty-two newly arising stable (nonvarie-
gating, see below) recessive mutants were
observed. Twenty-four of these were lo-
cated in chromosome 9 (5 pyd, 11 wd,
5 yg, 1 small-kernel, 1 narrow-leaf, 1 light-
green); four of the new mutants were not
located in chromosome 9, although one of
them showed an unexpected relation to a
previously known mutant in chromosome
g; tests of the four remaining new mutants
have not been completed. In the majority
of the parent plants, only one of the two
chromosomes g could be thoroughly tested
for the presence of a new mutant, because
of the frequent lack of transmission
through the pollen of one of the altered
chromosomes 9.

Cytological observations of the chromo-
some 9 carrying a newly arising mutant
have been made in the case of 13 pyd
mutants, 11 wd mutants, 8 yg mutants,
and the small-kernel, pale-green, and
bronze mutants. In 12 of the pyd mutants,
all of the wd mutants, the bronze mutant,
and the small-kernel mutant, a chromo-
somal deficiency or an obvious alteration
was evident at the “locus” of the mutant.
DEPARTMENT OF GENETICS

In the remaining cases examined, the par-
ticular chromosomal event that gave rise
to the mutation could not be determined
with certainty.

One of the four new mutants not located
in chromosome 9 showed an interesting
relation to the mutant c, which is located
in the short arm of chromosome 9. The
recessive mutant c, when homozygous,
produces kernels that have no color in
their aleurone layer. The dominant allele,
C, gives rise to colored aleurone. The new
mutant, a simple recessive in breeding
behavior, produces small kernels with de-
fective embryos. Plants that were Ce and
likewise heterozygous for the small-kernel
mutant were self-pollinated. Colored and
genotypically colorless (cc) kernels and
normal and small kernels segregated inde-
pendently. The small kernels in the homo-
zygous cc class, however, were not com-
pletely colorless, as expected. Instead, they
showed a faint but distinct blush of color
in the aleurone layer, closely resembling
that produced by the 6z mutant. Cyto-
logical examination has not been made to
determine whether the origin of the small-
kernel mutant may have involved chromo-
some g in some unusual manner. The
peculiar relations of the spotted-leaf mu-
tant, as described below, suggest this
possibility.

MopiricaTion oF Mutant Expression
FotLtowinc CuroMosoMAL
TRANSLOCATION

The recessive mutant “spotted-leaf”
(spl), in early investigations, gave evi-
dence indicating that it was a recessive
located in the short arm of chromosome 9.
This was because the spotted-leaf character
appeared in the appropriate chromosomal
class in Fi plants following the cross of a
spotted-leaf plant to a female plant hetero-
zygous for a relatively long but trans-

177

missible terminal deficiency of the short
arm of chromosome 9. Further study of
this particular mutant revealed that its
inheritance is not typical of a simple reces-
sive and that its locus cannot be placed in
the short arm of chromosome g in the
usual manner.

The spotted-leaf mutant is not detected
in the young seedling stage. Chlorophyll
development in the seedling usually ap-
pears quite normal. As the plant continues
to grow, the chlorophyll throughout the
plant changes to a bright yellowish green.
Later, further changes in the chlorophyll
occur. In evenly distributed, small spots
throughout the leaves, the chlorophyll ap-
pears to break down still further. Thus,

on a light-green leaf there appear many

small yellowish spots. The mutant has
consequently been called “spotted leaf.”
Following the production of the spots, the
plant may exhibit various degrees of
necrosis of the cells of the leaf, beginning
at the tips of the leaves. This degeneration
is accelerated by excessive heat and light.
Relatively few spotted-leaf plants have sur-
vived to maturity.

The spotted-leaf mutant first appeared
in the progeny of a self-pollinated plant
that was being investigated for purposes
other than the detection of new mutants.
This parent plant had one chromosome 9
with the minute terminal deficiency that,
when homozygous, produces pale-yellow
(pyd) seedlings. This chromosome 9, of
known constitution, had been introduced
into the parent plant by a controlled cross.
Its homologous chromosome g had pre-
viously been altered in constitution follow-
ing a series of breakage cycles. During
these breakage cycles, an additional chro-
mosomal alteration occurred. It involved
the translocation to the end of the short
arm of chromosome 8 of a segment com-
posed of six chromomeres derived from
the end of the short arm of chromosome 9.
178

As this segment of chromosome 9, before
its translocation to the short arm of
chromosome 8, had been subjected to the
breakage-fusion cycles, it was not an un-
modified terminal segment. Its exact
composition is still unknown. Cytological
observations indicate that chromosome 8
could not have lost more than a chromo-
mere from the tip of its short arm during
the translocation process; if as much
chromatin as this has been removed, the
resulting deficiency does not interfere with
pollen transmission to a marked extent,
for this chromosome is readily transmitted
through the pollen. (Translocations in-
volving the broken end of chromosome 9
during its chromatid breakage cycles with
other chromosomes of the complement
are not infrequent. See Year Book No.
41.) The parent plant had, then, a pyd
chromosome 9, the newly modified chro-
mosome 9, a normal chromosome 8, and
the chromosome 8 with the translocated
segment. This plant was normal in ap-
pearance. The new spotted-leaf mutant as
well as the expected pyd mutant appeared
in the progeny of this plant following
self-pollination.

All evidence to date indicates that the
spotted-leaf character will appear when-
ever a plant is homozygous for any one of
the pyd-producing deficiencies of chromo-
some g and also has one normal chromo-
some 8 and one chromosome 8 with the
translocated segment. In other words, the
chromosome 8 with the translocated seg-
ment is responsible for the appearance of
the mutant spotted-leaf, but only when a
short terminal homozygous deficiency is
also present in chromosome 9. A simple
interpretation, subject to verification, of
this relatively complex genotypic expres-
sion is possible. The terminal deficiency
that produces the pyd mutant is large
enough to be cytologically detected. Possi-
bly the segment of the tip of chromosome

CARNEGIE INSTITUTION OF WASHINGTON

g that has been translocated to chromo-
some 8 is itself incomplete and has a very
minute deficiency for a segment of chro-
matin within the limits of the larger pyd-
producing deficiency. Thus, plants having
two deficient pyd chromosomes g and the
duplication carried by chromosome 8
would be homozygous deficient for a
minute segment of chromosome in the
tip region of chromosome g even smaller
than that which produces the pyd char-
acter. If this is so, the spotted-leaf character
would be expected to appear as a simple
recessive mutant, located close to the tip of
the short arm of chromosome 9g, in the
progeny tests of some plants whose
chromosomes g have undergone the break-
age cycles. It would be expected to show
allelism with the pyd and wd mutants.
That it has not been detected so far in
progeny tests of such plants may be due
to the difficulty of detecting this mutant
in the seedling stages, which have been
used almost exclusively for detecting the
newly produced mutants.

The known conditions connected with
the origin and action of the spotted-leaf
phenotype are instructive in a considera-
tion of the problems associated with some
previously studied types of interrelated but
independently inherited mutant factors.
A succession of a few relatively simple
chromosomal aberrations may well ac-
count for some of these observed relations.

Tue Unexpecrep APPEARANCE OF A
Numper or Unsraste Mutants

In the cultures that were grown to ob-
serve the mutations produced as a conse-
quence of the breakage cycles of chromo-
some 9, an unusual and unexpected series
of new mutants appeared, characterized by
types of instability known in genetic litera-
ture as mutable genes, variegation, or
mosaicism. Although maize has been ex-
DEPARTMENT OF GENETICS

tensively investigated, the appearance of
“mutable genes” has been relatively rare.
In the present comparatively restricted
study of the progeny of plants having
newly broken chromosomes 9, fourteen dis-
tinctly new expressions of instability of
genic action have been isolated, although
more types have been observed. Six of
these have received at least preliminary in-
vestigation. It is not possible, however, to
follow the genetic or cytological factors
associated with all these new variegated
types, because of the extensive amount of
time and labor involved in the study of
any one. The plant characters that may be
most readily detected in variegated or
mosaic patterns involve some form of color
change in a tissue. Thus, most of the
variegating types that have been recog-
nized involve either anthocyanin or chloro-
phyll pigments. It should be emphasized
that no one of these variegating mutants
has received more than a preliminary
study; however, a brief description of four
of these mutable types will serve to illus-
trate the nature of the phenomena and the
problems involved.

Vartegated white seedling. This mutant
first appeared in the progeny of a self-
pollinated plant in which both chromo-
somes g had been involved in the chromo-
some type of breakage cycle in the very
young embryonic stage. This first progeny
consisted of 108 normal green seedlings and
19 white seedlings. In the white seedlings,
sectors of pale-yellow or of normal-green
tissue were present. Each pale-yellow or
green sector was sharply delimited; all the
plastids in the pale-yellow cells were yellow
and all the plastids in the green cells were
normal green. The size of the pale-yellow
sectors varied from large, indicating an
early mutation from white to pale-yellow,
to very small, indicating a late mutational
change to pale-yellow. Most of the green
sectors were small, indicating that the

179

mutations to green usually occur late in
the life of the tissue. A few early muta-
tions occurred, however, as there were a
few rather wide green sectors in the leaves
of some of these white seedlings. The rate
of mutation to pale-yellow and to green in
any one seedling was relatively constant;
in general, each seedling was distinguished
by one particular rate of mutational
change. In some of the seedlings, distinct
sectors were present in which the rate of
mutation had increased considerably, al-
though within. these well defined sectors
themselves uniform rates of mutation were
discernible. This variegating mutant be-
haved as a simple recessive upon further
testing. The phenotypic expression of the
variegation—that is, white, pale-yellow,
and green—is strikingly similar to the
allelic series composed of wd, pyd, and
normal-green associated with overlapping
deficiencies at the tip of the short arm of
chromosome g. A further similarity to the
wd mutant is apparent: when plants
heterozygous for this variegating white
mutant are selfed, a number of kernels
with defective embryos are produced. If
the proportion of kernels with defective
embryos is high on any one ear, the ex-
pected proportion of variegated white seed-
lings is correspondingly lowered. In each
case, the total number of defective embryos
plus white-variegated seedlings approxi-
mates the one-fourth expected in the
progeny. This behavior is quite similar
to that of the wd mutants that have been
extensively studied. Nevertheless, when
this mutant is combined with the wd
mutants, only normal-green seedlings and
plants are produced. Furthermore, tests
have shown that this mutant is not located
in the short arm of chromosome 9.
Variegated light-green. The recessive
light-green mutant appeared in the progeny
of a self-pollinated plant that had received
a chromosome 9 which had been under-
180

going the chromatid type of breakage
cycle. Approximately one-fourth of the
seedlings in this progeny were very light
green. All the light-green seedlings showed
a number of sectors of normal-green tissue,
most of them appearing as fine green
streaks in the light-green leaf. In the cells
that are light green, the plastid size and
number are normal but the chlorophyll
color in all plastids is light green. In the
sectors of normal tissue, the chlorophyll in
all plastids is normal green in color. In
any one seedling, the mutations to normal
green occur at an approximately constant
rate. Very wide differences, however, ap-
pear among the various seedlings; some
have very high rates of mutation, others
have very low rates, with many seedlings
showing gradations between these ex-
tremes. These definite rates of mutational
change from light green to normal green
are maintained throughout the life of an
individual plant, except for distinct sectors,
presumably arising from individual cells,
that show an increased rate or a reduced
rate of mutation, respectively, over that of
the plant as a whole. In a number of cases,
two distinctly delimited but adjacent sec-
tors showed inverse relations in their muta-
tion rates. In one sector the rate of muta-
tion, as expressed by the number of green
streaks in the leaf tissue, was greatly in-
creased, whereas in the sister sector the
number of green streaks was greatly re-
duced. The positions of these twin sectors
in the stalk and leaf suggested that they
arose from two sister cells of the growing
point. Such striking changes in the rate
of mutation of cells derived from a single
somatic cell of a plant have characterized
all the variegating types that have been
under study. The yet-unknown alteration
that occurs in this cell determines the rates
of mutation of the recessive mutant to a
dominant allele in all cells that arise from
it, until another such change suddenly

CARNEGIE INSTITUTION OF WASHINGTON

occurs. The frequent appearance of twin
sectors showing inverse rates of mutation
leads to the suggestion that the controlling
factor for future mutational occurrences
may be altered or segregated at a somatic
mitosis so that the mutations that will oc-
cur in the cells arising from one daughter
cell are increased over the potential rate of
the mother cell; conversely, the mutations
that will occur in the cells arising from the
other daughter cell are reduced.

The light-green character is inherited as
a simple recessive. In individual cultures,
the F2 ratios frequently approximate three
normal-green plants to one light-green
plant. This usually occurs when the muta-
tion rates in the light-green plants are uni-
formly low. However, in some F? cultures
segregating light-green plants with low
mutation rates, the ratios were distinctly
aberrant. When the mutation rates are
high in the light-green plants of an F:
culture, a deficiency in the light-green class
frequently occurs. This suggests that muta-
tions of the light-green locus to normal
green have occurred at a relatively high
rate in the Fi heterozygous plant and result
in an increase in the relative number of
gametes carrying normal green. It is pos-
sible that the light-green locus may occa-
sionally mutate to intermediate alleles, as
several distinctive types of the light-green
mutant have appeared in a few of the
cultures.

The luteus mutant. The recessive luteus
mutant is one of the more interesting and
complex of the variegating mutants that
have appeared in this study. The luteus
character is associated with the presence of
small yellowish plastids. The luteus seed-
lings are, therefore, bright yellow. From
one original plant that had received a
chromosome 9 which had undergone the
chromatid type of breakage cycle, two
kernels were removed and plants were
grown from them. Each plant was self-
DEPARTMENT OF GENETICS

pollinated. In the progeny of one plant,
yellow (luteus) seedlings segregated.
These showed no variegation. In the
progeny of the second plant, yellow seed-
lings segregated but many were highly
variegated for normal-green tissue. The
nonvariegated luteus seedlings from the
first plant were inviable, but the more
highly variegated luteus seedlings from the
second plant were viable, presumably be-
cause of the extensive amounts of normal-
green tissue. Further tests showed that the
nonmutating or stable luteus and the
highly mutating luteus represent two ex-
treme states of one particular locus. The
stable luteus is only relatively stable and,
in the tests so far conducted, has only
rarely become mutable again.

In F2 progeny tests, stable luteus usually
segregated quite normally. From 52 self-
pollinated F, plants there were obtained
2306 normal-green seedlings, 773 non-
variegated luteus seedlings, and only 3
variegated luteus seedlings. These 3 varie-

gated luteus seedlings appeared in three ‘

different cultures. In one single F, culture,
231 normal-green seedlings, 51 nonvarie-
gated luteus seedlings, and 10 variegated
luteus seedlings appeared. In this case, it
is probable that in a cell of the Fi hetero-
zygous plant the stable luteus suddenly
became unstable and formed a sector of
mutating luteus. In contrast with the
stable luteus, the highly variegating luteus
is distinguished by great instability of the
luteus locus. It may mutate to normal
green but also, phenotypically at least, to
several distinctive intermediate alleles.
Plants heterozygous for the highly stable
luteus and a mutating luteus may exhibit
various rates of mutability in a manner
quite similar to that observed among the
variegated white seedlings and the varie-
gated light-green plants. A distinctive and
relatively constant rate of mutation from
luteus to green is apparent in each indi-

181

vidual seedling. These rates range from
very low, with only a few small green
streaks on a leaf, to very high, where the
leaf appears to be almost mottled because
of the many mutations that have occurred.
Most of the mutations occur late in the
development of the tissues, but a few occur
earlier and give rise to wide sectors of light-
green or normal-green tissue. In a manner
quite similar to that found in mutable
white and light-green, distinctive sectors
appear in these plants with decidedly
altered rates of mutation, either increased
or decreased. Adjacent sectors (twin sec-
tors) also occur, one showing an increased
rate of mutation, the other a reduced rate
of mutation as compared with that in the
plant as a whole.

A further complexity also appeared. It
was first clearly recognized in the progeny
resulting from a cross of plants that were
heterozygous for a stable luteus locus and
its wild-type allele by the variegated luteus
plants that had both a mutating and a
stable luteus locus. Because the Fi plants
resulting from the combination of either a
stable luteus or a mutating luteus with a
normal wild-type allele have been normal
green in appearance, one could expect to
find normal-green plants, variegated luteus
plants, and nonvariegated luteus plants in
the progeny of the above cross. These
three types of plant appeared, but, in addi-
tion, a distinctly new type of variegated
plant was present. These plants were green
but showed fine streaks of luteus. The
numbers of such streaks varied greatly
among the different plants. When very
many streaks were present, they occurred
in well defined sectors within the leaf. In
a few of these plants, wide sectors of luteus
were present. Within some of these luteus
sectors, mutations to normal green oc-
curred, each sector showing its own par-
ticular rate of mutation from luteus to
green. Because these plants were mainly
182

green, with fine streaks of luteus, they
were designated “streaked.” In their pat-
terns, the streaked plants resembled the
variegated luteus plants, but they were
reverse images—that is, negatives (luteus
streaks in green tissue)—of a positive
(green streaks in luteus tissue) variegation
pattern. This suggests that some of the
mutations of luteus to green are unstable
and revert to luteus by processes similar
to those that give rise to the dominantly
directed mutations. As a working hypothe-
sis, to be considered as tentative only, it
may be conceived that the locus concerned
may be present in a very stable recessive
state, rarely mutating to dominant, or in
a very stable dominant state, rarely mutat-
ing to luteus; or that it may be present in
recessive or dominant states with various
intermediate rates of mutability. Further-
more, changes in the stability of any one
state may occur suddenly, following some
yet-unknown event that probably takes
place during a mitosis.

The chromosome-breakage variegation.
The most unexpected expression of varie-
gation that has appeared in these investiga-
tions is associated with the occurrence, in
many somatic cells, of breakage in chromo-
some g that takes place at a particular
locus in the short arm of this chromosome.
This breakage results in the formation of
an acentric fragment and the subsequent
elimination of this fragment at a somatic
anaphase. This phenomenon first ap-
peared in the progeny of a self-pollinated
plant that had started development with
its two chromosomes 9 undergoing the
chromosome type of breakage cycle, al-
though healing of the broken ends had
occurred in early embryogeny. A few of
the kernels on the ear of this plant ex-
hibited a type of aleurone color variega-
tion that had not previously been observed.
In this plant, one newly broken chromo-
some g carried the factors J (dominant

CARNEGIE INSTITUTION OF WASHINGTON

inhibitor of aleurone color) and Wx
(starch in endosperm stains blue with
iodine). Its homologue carried the reces-
sive alleles ¢ (colored aleurone) and wx
(starch stains red with iodine). The aber-
rant kernels were recognized in the hetero-
zygous class that had received both the :-
and the /-carrying chromosomes. Accord-
ing to their genetic constitution, such
kernels should be colorless. The aberrant
kernels were conspicuous because of the
presence of colored (7) areas. In some
kernels, there were well defined sectors
that exhibited a uniform pattern of small z
spots. The pattern in any one such sector
was distinguished by the number of 2 spots,
their uniform distribution within the sec-
tor, and their relatively similar size. The
patterns in these sectors suggested that the
I factor, carried by one chromosome g, had
been systematically eliminated from some
cells and that, in each sector, this had
occurred at a particular rate and at a par-
ticular stage in the development of the
endosperm tissue. Subsequent testing in-
dicated that the I-carrying chromosome
also possessed the minute terminal de-
ficiency that is responsible, when homo-
zygous, for the appearance of white seed-
lings (the wd mutant). Consequently,
green and white seedlings segregated in
the progeny of this self-pollinated plant,
the white seedlings arising mainly from
the I kernels.

To investigate the nature of the phe-
nomenon associated with the appearance
of the i spots, the green plants arising from
some of the J kernels were grown in the
summer of 1945. In many of these plants,
fine streaks of white tissue, rather uni-
formly distributed over the leaf, were ob-
served. In some plants, there were only
a few such non-chlorophyll-bearing white
streaks; in others, there were more; in one
plant, the number of white streaks was
relatively large. In most of the plants, dis-
DEPARTMENT OF GENETICS

tinct sectors were present in which the
number of uniformly distributed white
streaks was greatly increased or decreased
over that of the plant as a.whole. In these
aspects, the patterns of variegation re-
sembled those observed in the three pre-
viously described variegating types.

In agreement with the presence of the
i spots in some of the J kernels, the pres-
ence of white streaks in the leaves of many
of the plants again suggested that a seg-
ment of chromatin of one chromosome 9
was being systematically eliminated in so-
matic cells. Elimination from one chromo-
some g of the chromatin segment that
covers the white-producing deficiency in
the other chromosome 9 would allow the
non-chlorophyll-producing effect of this
deficiency to be expressed. According to
this interpretation, it would be necessary
for the streaked plants to be heterozygous
for the wd mutant. Tests confirmed the
presence of the wd mutant in all the
streaked plants.

The white-streaked plants used for fur-
ther crosses carried wd, I, Bz, Wx in one
chromosome g and Wd (i., chromatin
segment covering the white-producing de-
ficiency), i, Bz, wx in the homologous
chromosome g. Pollen of these streaked
plants was placed on silks of plants carry-
ing various recessive mutants located in
the short arm of chromosome g. One such
cross, to 2 bz wx, was particularly instruc-
tive. The locus of bz (bronze, very light
aleurone color) lies between those of J and
Wx. The W« locus is closest to the centro-
mere, but the loci of Wd, I, Bz, and Wx
are all included in the distal two-thirds
of the short arm. On examination of the
types of kernel resulting from this cross,
it was again apparent that some kind of
chromatin loss, involving a segment of the
short arm of chromosome g that included
at least the J, Bz, and Wx loci, was respon-
sible for the variegation patterns observed

183

in some of the resulting kernels. In these
crosses, the behavior of chromosome g in
endosperm development could be critically
evaluated, from a genetic point of view,
only in those kernels that had received an
Iarrying chromosome from the white-
streaked parent. The aleurone variegation
pattern, involving the alleles 7 and 7, Bz
and bz, or Wx and wx in the kernel, and
the white-streaked pattern, involving Wd
and wd in the plant, clearly differed from
the previously known chromatin-loss pat-
terns associated with ring chromosomes or
with the chromosome or chromatid bridge
cycles. In all appropriate crosses, an addi-
tional fact was noted. Somatic elimination
of chromatin was observed mainly in those
I kernels that had received an J wx chromo-
some—that is, a chromosome g resulting
from a crossover between the loci of J and
Wx in the white-streaked parent plant.
Relatively few 1 Wx chromosomes were
undergoing chromatin elimination. This
would suggest that if a locus in chromo-
some g is responsible for the somatic elimi-
nation of segments of chromatin, its posi-
tion in the chromosome is proximal but
close to the wx locus. This would place
the chromatin-elimination locus in the
Wd i wx chromosome of the streaked
parent plant. To illustrate the present
knowledge of the nature of this chromatin-
elimination phenomenon, the subsequent
behavior of one I wx chromosome will be
briefly described.

In the cross of a white-streaked plant
with the constitution wd I Wx/Wd it wx
to a normal plant homozygous for Wd i
wx, an I wx kernel, highly speckled with
i, was removed and germinated. A normal-
appearing, nonstreaked plant arose from
this kernel. The plant was self-pollinated.
On the resulting ear, there again appeared
I kernels that were highly spotted with 2.
This time, however, in contrast with the
above crosses, a large number of these
184

kernels were present. The seedlings grown
from the kernels of this ear were classified
into three distinct types: green seedlings
with few or no white streaks, green seed-
lings highly streaked with white, and
totally white seedlings. The proportions of
these seedling types arising from the non-
spotted or only very moderately spotted J
kernels, the heavily spotted I kernels, and
the i kernels are given in the accompany-
ing table.

Types OF SEEDLINGS ARISING FROM THREE CLASSES
OF KERNELS FOLLOWING SELF-POLLINATION
oF A wo I/Wp 1 PLANT

 

 

No. OF SEEDLINGS

 

 

CLASS OF KERNELS Whit
Ate- .
Green streaked White
Loic eee ee 63 5 33°
Highly spotted.... 24 4 1
Desa cp vtec sees 70 1 0

 

 

 

 

All the 10 highly streaked seedlings, and
one obviously peculiar nonstreaked seed-
ling arising from an : kernel, were trans-
planted to the field in the summer of 1946.
The white streaks continued to appear in
the leaves as the heavily streaked seedlings
developed into mature plants. The pat-
terns were similar to those of the original
white-streaked plants, but the total number
of such fine white streaks was enormously
greater. In contrast, the nonstreaked seed-
ling remained nonstreaked throughout its
development.

To observe whether chromatin losses
could be seen in somatic mitoses of tissues
that were relatively late in their develop-
mental period, whole mounts of the young
membranous glumes in the florets of the
tassels of the streaked plants and the non-
streaked plant were stained and examined.
In all examined glumes, many cells
showed, besides their main nucleus, a very

CARNEGIE INSTITUTION OF WASHINGTON

small micronucleus or a very small deep-
staining pycnotic chromatin body. The
majority of the anaphase figures appeared
normal, but in some late anaphases one
fragment or two identical-sized fragments
were observed. Time did not allow a de-
tailed study of the sequence of events or
of the frequencies or types of fragment
formation. An examination of the sporo-
cytes of the anthers of all 11 plants, how-
ever, has thrown some light on the nature
of the fragmentation phenomenon.
The chromosome-g constitution at pachy-
tene in the microsporocytes of all 11 plants
was examined. In all 10 white-streaked
plants, many of the sporocytes showed the
presence of a morphologically normal
chromosome g and a chromosome 9 carry-
ing the minute wd-producing deficiency.
In the nonstreaked plant, two morpho-
logically normal chromosomes 9 were
present in many of the sporocytes; the
wd-producing deficiency was not present
in this plant. In the individual anthers of
all 11 plants, however, the sporocytes ex-
hibited mosaicism for a deficiency of a long
segment of the short arm of chromosome 9.
In many sporocytes, loss of the segment
from one chromosome g had occurred in
a recent premeiotic mitosis, although in
some cases losses had occurred earlier to
give rise to a relatively large cluster of
related sporocytes each having a long de-
ficiency in one chromosome g. In a num-
ber of sporocytes, a small deep-staining
pycnotic chromatin body was present in
the cytoplasm. In many of these cells, in
turn, the constitution of the chromosome-9
bivalent could be analyzed. In the clearest
cases, it could be determined with certainty
that one chromosome 9 was deficient for a
large segment of the short arm. Thus, a
correlation could be obtained between the
presence of the pycnotic chromatin body
in the cytoplasm of a cell and the absence
of a long segment from the short arm of
DEPARTMENT OF GENETICS

one chromosome 9g of the same cell. The
cytological studies must be greatly ex-
tended in order to clarify and complete the
needed information, but it is quite clear
that the positions of “breakage” in
chromosome g do not occur at random, In
a large number of figures, it could be
determined that the segment deleted from
the nucleus at an anaphase included the
terminal two-thirds of the short arm; in
other words, the break, in each case, oc-
curred at a position approximately one-
third the distance from the centromere.
The acentric segment that resulted from
such a break was subsequently lost to the
nuclei during a mitosis. Not all were im-
mediately lost, however, for in some sporo-
cytes a bivalent acentric terminal segment
of chromosome g was present along with
the normal and deficient chromosomes 9.
Various types of synaptic association were
observed between the acentric segments
and the short arm of the unbroken
chromosome g. In a number of sporo-
cytes, what appeared, on rapid examina-
tion, to be a normally synapsed bivalent
chromosome g proved to be otherwise.
In the short arm of one of the chromo-
somes of the bivalent, a break was un-
questionably present at a position ap-
proximately one-third the distance from
the centromere. Very occasionally, cells
were observed with other abnormalities of
chromosome 9. It is possible that these
resulted from secondary effects of the pri-
mary breakage process, but any conclu-
sions must await a more extensive study.
The nature of the process that is respon-
sible for this most unusual type of chromo-
some “breakage” is not understood. Two
lines of evidence indicate that the locus of
breakage is not a “weak” spot in the
chromosome that is subject to breakage
following unusual tension on the chromo-
some. If a chromosome with this locus is
subjected to the chromatin bridge cycles,

185

breaks occur at various positions in the
short arm regardless‘of this locus. Further-
more, the presence of distinct sectors, each
one with its own uniform rate of breakage
of the chromosome at this locus, cannot
readily be interpreted on the basis of a
structurally weak spot in the chromosome.
It may be suspected that the “breakage”
occurs during the chromosome reduplica-
tion process as a consequence of some yet-
undetermined abnormality that is present
at this one particular locus in the chromo-
some. Such a modified locus should be
subject to genetic analyses. Preliminary
genetic evidence derived from plants that
were heterozygous for various mutants in
the short arm of chromosome 9, as well as
heterozygous for this altered locus, have
placed this locus in a position which con-
forms to the position that had been de-
termined cytologically.

It should be mentioned that the single
non-white-streaked plant was unusual be-
cause of the very high rate of breakage that
occurred in it. Both chromosomes g under-
went the breakage phenomenon, but only
occasionally did breakage involve both
chromosomes g in a single nucleus. When
this occurred in a meiotic nucleus, it
produced unquestionable evidence for the
precise localization of the breaks in
chromosome 9. A genotypically complete
chromosome-g bivalent was present at
pachytene, but in the form of two entirely
detached segments. The homologously
associated distal two-thirds of the short
arms of both chromosomes g formed one
segment; the other segment was composed
of the homologously associated long arms,
the centromeres, and the proximal third of
the short arms. It could readily be observed
that the break had occurred at the very
same locus in each chromosome 9. This
was the same locus of breakage that had
been observed when only one of the two
chromosomes 9 had been broken. It is
186

possible that in this plant the excessive rate
of loss from somatic nuclei of the large
segment of chromosome 9 was responsible
for its selection as an aberrant seedling.
The leaf tissue of the seedling probably
was heavily mosaic for different genomic
complements of chromosome 9. This may
have been responsible for its atypical
appearance.

The cytological observations of breakage
of chromosomes 9, predominantly at a par-
ticular locus, and of the subsequent elimi-
nation of the acentric segment that results,
are consistent with the genetic observa-
tions. This segment carries the loci of Wd,
1, Bz, and Wx, in this order. If, in hetero-
zygous plants, these dominant alleles are
carried by the chromosome that is under-
going breakage in various somatic cells,
simultaneous losses of these dominant
alleles should occur following deletion of

CARNEGIE INSTITUTION OF WASHINGTON

the acentric fragment from a nucleus. In
the kernels, such simultaneous losses of
the Z, Bz, and Wx loci have been observed.
It is apparent that the loss of the Wd locus
accounts for the white (wd) streaks in the
heterozygous plants.

Although the factors responsible for this
breakage phenomenon are not understood,
nevertheless the factors that control the
frequency and the timing of such occur-
rences are similar to those that control
the frequency and timing of dominantly
directed mutations in the variegation types
previously described. In the case of the
chromosome-breakage variegation just de-
scribed, however, the breakage itself corre-
sponds to the “gene” mutations observed
in the other variegations. Possibly the
resemblance is more than coincidental, in
that the underlying phenomena are basi-
cally similar.