CONTROLLED MUTATION IN MAIZE

Barpara McCiintock

Experimentation conducted during the
past year was aimed at expanding our
knowledge of the kinds of elements car-
ried in the maize chromosomes that con-
trol gene action and give rise to changes
in this action, that is, to mutations. Dif-
ferent controlling elements have been rec-
ognized, each characterized by its own
specific mode of control of gene action and
mutation; and this mode of control is
quite independent of the primary type of
action of the genic materials themselves,
which these elements may serve to modify.
Their presence in the chromosome comple-
ment is made evident because they under-
go transposition from one location to an-
other and do not lose their specificity of
action in the process. Were it not for
such behavior, these elements would re-
main undetected. Insertion of a control-

ling element at a locus of known gene
action results in immediate or subsequent
change in this action, or both. Each ele-
ment expresses its own mode of control
of change in gene action, and this allows
the presence of the element at the locus
to be recognized. The same element may
be inserted at a number of different loca-
tions and thus come to control the action
of genic materials at each of these loca-
tions. Conversely, the action of the same
genic material may be influenced by dif-
ferent controlling elements, as a result of
independent insertions of such elements at
one particular locus.

Evidence of the control of gene action
and mutation at a number of different
known loci in maize by the Ds-Ac two-
element controlling system has been re-
ported in past years. Knowledge gained
246

from concentrated attention to the mode
of operation of this particular system has
provided the basic information that now
serves as a guide in planning experimenta-
tion and in interpreting the modes of
operation of other controlling elements or
of systems of interrelated elements. With
this framework of knowledge, advances in
these studies may be made much more
rapidly and effectively. In the past year,
attention has been given to a system that
differs considerably from the Ds-Ac sys-
tem in its manner of control of gene action
and mutation. A tentative hypothesis to
account for the operation of this system
was outlined in Year Book No. 53. Extén-
sive tests have now been conducted, and
the evidence obtained from them fully sup-
ports the hypothesis previously formu-
lated. Much additional knowledge of the
mode of operation of the system has also
been obtained. A summary of this may
now be given.

THE a:*-Spm System or ContRoL oF
Gene Action anp Mutation

The A: locus in chromosome 3 of maize
is a particularly favorable one for examin-
ing the operation of controlling systems.
The genic materials at this locus are con-
cerned with the development of antho-
cyanin pigmentation both in the plant
tissues and in the aleurone layer of the
kernel. When a change in intensity, qual-
ity, or pattern of distribution of that pig-
ment appears in an individual plant or
kernel, the altered phenotype is readily
noticed. Therefore it is possible to detect
insertions of controlling elements at the
locus shortly after they occur. Different sys-
tems controlling gene action and mutation
vat this one locus have been recognized, and
their modes of operation examined. Each
was detected, initially, because of a distinct
deviation from the standard Ai type of
expression of anthocyanin pigmentation,
appearing either in an individual kernel
on an ear or in an individual plant of a
culture. Four of these systems have been
analyzed in some detail and their modes

CARNEGIE INSTITUTION OF WASHINGTON

of operation defined: the a:-Dt (Dotted)
system, the Ds-Ac system, the ar™"*-Spm
system, and still another system operating
at this locus to control the type and dis-
tribution of anthocyanin pigmentation in
both kernel and plant. The system with
which we shall be concerned in this section
is the a.”""-Spm system.

Knowledge of the behavior of control-
ling systems makes it evident that this one
is composed of two interrelated but inde-
pendently located elements. One of them,
becoming inserted at the locus of Au,
caused a modification of the action of the
genic materials located there. The modi-
fied locus has been designated a," to dis-
tinguish it from other modifications that
have arisen independently at this same
locus. Subsequent changes have occurred
at the locus, each effecting a change in
gene action. These are regarded as arising
from alterations of the controlling element.
With a few possible exceptions, the genic
materials themselves do not appear to be
altered by such modifications. Their mode
of action, however, may be decidedly al-
tered as a result of any one change in the
associated element. Thus the observed
changes in gene action are referable, on the
whole, to changes in the controlling ele-
ment and not to irreversible changes in the
gene substances. Moreover, such changes
in gene action, stemming from alterations
in the controlling element at the Ai locus,
can occur only when a second, independ-
ently located element is also present in the
nucleus. The second element of this sys-
tem is called Suppressor-mutator (sym-
bolized as Spm), for the following reasons.
In plants that either are homozygous for
a." or are ai""/a, in constitution, and
that also have Spm in their nuclei, no an-
thocyanin pigment develops in the aleu-
rone layer of the kernel or in the plant tis-
sues until, in a somatic or a germinal cell,
a modification of the controlling element
located at A. allows pigment to be formed
in those cells where it normally develops
when the standard organization at the 4:
locus is present. These modifications effect
DEPARTMENT OF GENETICS

stable mutations, in that the altered type
of gene action so produced continues to be
expressed in subsequent cell and plant
generations, both in the presence and in
the absence of Spm. In plants of the above-
named constitutions that do not have Spm
in their nuclei, on the other hand, re-
stricted gene action occurs, and this results
in the appearance of uniformly distributed
pigment both in the aleurone layer of the
kernel and in the plant tissues. This ex-
pression of the genic substance at A: is
constant and stable through successive
plant generations as long as the Spm ele-
ment is absent from the nuclei, for no
mutations occur. Return of Spm to the
nuclei, however, by appropriate crosses,
again initiates the Suppressor-mutator ef-
fect on the element located at di. Gene
action is again suppressed until, in a so-
matic or germinal cell, some change in
this element allows the genic materials to
function in some particular manner.

Unlike its counterparts, Dz in the a:-Dt
two-element system and Ac in the Ds-Ac
two-element system, Spm does not show
pronounced dosage effects that are re-
flected in altered frequencies or times of
occurrence of mutation at the modified
A, locus. Like other controlling elements,
however, Spm undergoes transposition and
consequently occupies no set position in
the chromosome complement. An indi-
vidual plant may have several Spm cle-
ments, each occupying a different site in
the chromosome complement. Because it
shows no dosage effects, the number of
Spm elements present in the nuclei of a
plant, as well as their locations within the
chromosomes, must be determined by
progeny tests.

In addition to the modifications affect-
ing stable gene action, the element at Ai
also undergoes another type of change, but
far less frequently. These changes, called
“changes in state,” are expressed by strik-
ing differences in the types of mutation
that occur subsequently in the presence of
Spm, and also in their times and frequen-
cies of occurrence during development of

247

each tissue. They also affect the degree of
gene action that occurs in the absence of
Spm. This varies among the states, and in
this respect they form a graded series, from
those that produce low levels of pigment
intensity to those that give high levels. A
few of the latter produce pigment intensi-
ties approximating that given by the genic
materials at the standard 4:1 locus. Never-
theless, with these latter states as with all
states of a.”-* examined, pigment forma-
tion is completely suppressed in the pres-
ence of Spm and will appear only after the
element located at 4: undergoes some
mutation-inducing event, or after Spm is
removed from the nucleus. With respect
to state it has also been found that the
type of gene action that appears in the
absence of Spm is not correlated with the
types and patterns of mutation that appear
in its presence. The states of a:”~* will be
discussed more fully later.

Inheritance patterns of Spm. In Year
Book No. 53, evidence was reported of
linkage of Spm with Y (yellow endo-
sperm), located in chromosome 6, in some
plants of a particular culture (see table 17
on p. 257 of that Year Book). In these
plants, only one Spm element was present.
Their constitutions were a1" She/ai sho;
Y Spm/y +. (The a allele in these plants
belongs to the a-Dt system of control of
gene action. It does not respond to Spm
and therefore behaves as a stable recessive
in plants that have Spm. Shrunken endo-
sperm, shz, is very closely linked to a: and
shows less than one-quarter per cent cross-
ing over with it.) When these plants were
crossed by plants homozygous for ai, she,
and y and having no Spm, the types of
kernels on the resulting ears indicated that
approximately 35 per cent crossing over
had occurred between Y and Spm in the
heterozygous parent plants. In the She
class there was a total of 1470 kernels. Of
these, 723 were uniformly pigmented,
showing a pale color in the aleurone layer
(no Spm present); 269 of them were Y
and 454 were y. In 740 of the Sh kernels,
spots of deep pigmentation appeared in a
248

colorless background (Spm present); 451

of these were Y and 289 were y. In addition,
there were 7 completely colorless kernels;
4 were Y and 3 were y. Among the 1489
kernels in the sh» class, only 7 carried
ai"; 3 of these were pale-colored (no
Spm), and 4 had spots of deep color in a
colorless background (Spm present). All
other kernels in the sh2 class had com-
pletely colorless aleurone; a 1:1 segrega-
tion for Y and y appeared among them.
Plants were grown from selected kernels
of all classes on these ears, and each was
subjected to a particular set of tests. It was
obvious that three major types of test were
required: (1) verification of linkage of
Spm with Y in plants derived from the
variegated kernels in the Shz Y class,
(2) verification of the presence of a:" but
the absence of Spm in plants derived from
the uniformly pale-colored kernels, and
(3) determination of whether or not Spm
would be present in approximately 65 per
cent of the plants derived from the a: sho Y
class of kernels and in approximately 35
per cent of the plants derived from the
a shz y class of kernels.

Tests other than these three were also
required. It was believed that somatic
losses of Spm from some nuclei were oc-
curring in some cases; this assumption was
based on the presence in some plants carry-
ing a,""* and Spm of distinct sectors show-
ing the phenotype that appears in the ab-
sence of Spm. Tests of the assumption
could be readily carried out when such a
sector extended into the tassel. Pollen col-
lected from the sectorial and nonsectorial
parts of the same tassel could be used in
particularly designed test crosses (see be-
low), which allowed detection of the pres-
ence or absence within the functional pol-
len grains of the Spm element and also of
an 4:""" locus capable of responding to
Spm in the expected manner. Such tests
were made, and they fully confirmed the
assumption that somatic losses of Spm
from some nuclei occur during develop-
ment.

CARNEGIE INSTITUTION OF WASHINGTON

In plants having one Spm element,
transposition of Spm in some germinal
cells can result not only in loss of Spm
from some nuclei, as described above, but
also in changes in its location, or increases
in its number, in others. Since the rate of
transposition of Spm appears to be rela-
tively low, in view of the rather sharp
linkage relations described, detection of
cases of transposition required tests of rela-
tively large numbers of individuals among
the progeny of plants having one Spm
element whose location was known. Such
tests were conducted, and evidence of
changes in location and increase in num-
bers of Spm elements was found.

Test (1) was extensive. It was accom-
plished by crossing each plant with one
that either was homozygous for a1", Sho,
and y and had no Spm or was homozy-
gous for a, shz, and y and had no Spm.
For many plants both test crosses were
made, and the results obtained in each test
were the same except in a few instances
where loss or change in position of Spm
occurred early in an individual cell of the
plant. Most of these were evident because
of the fact that one of the ears produced
by the plant was obviously sectorial with
regard to Spm constitution. A plant homo-
zygous of a:"* but having no Spm is par-
ticularly useful for determining the pres-
ence or absence of Spm in another plant
that is either ay” * far", ar” /a,, or ar/ar
in constitution, and also for determining
the numbers of Spm elements that may be
present in such a plant. An intercross is
made between the two plants. If Spm is
present in the plant being tested, all the
kernels that received it from gametes of
this plant will show colored spots in a
colorless background, and all those that did
not receive it will be uniformly pale in
color. With few exceptions, the ratio of
variegated to pale kernels will indicate the
numbers of Spm elements that were pres-
ent in the zygote of this-plant. The excep-
tions arise from early-occurring losses and
transpositions of Spm, but the frequency
of these is relatively low. If the tester
DEPARTMENT OF GENETICS

plant, which is homozygous for a:”* and
has no Spm, is also homozygous for some
known recessive markers such as wx, pr,
and y, and if the plant being tested is
heterozygous for such markers, evidence
of linkage of Spm to one or another of
these markers, or evidence of absence of
such linkage, is readily obtained.

Test (1), outlined above, verified the
linkage of Spm with Y that had been ob-
served in the parent plants, and the ratios
of kernel types were the same as those
shown by the parent plants. As expected,
however, a few cases were encountered of
change in location or increase in number
of Spm elements, which had occurred in
a germinal cell of the heterozygous parent
plant. As an illustration of these tests, data
obtained trom fifty-six plants may be sum-
marized. In forty-seven of them, linkage
of Spm to Y was clearly expressed and to
the same degree in each plant. Among the
7705 kernels in the pale class (no Spm),
2534 were Y and 5171 were y. Among the
7434 kernels in the variegated class (Spm
present), 4862 were Y and 2572 were y.
These data indicate that Spm is located
approximately 35 crossover units from Y.
In nine plants, the ratios of kernel types
did not conform with this. In four of
them, one Spm element was present but
its linkage to Y was not expressed with
certainty on any of the ears produced.
Among a total of 1142 kernels on these
ears, 533 had pale aleurone color (no
Spm); 258 of them were Y and 275 were
y. Among the 609 variegated kernels
(Spm), 321 were Y and 288 were y. In
two plants, two independently located
Spm elements certainly were present. On
each ear produced by these plants, a ratio
of 1 kernel with no Spm to 3 kernels with
Spm was observed. A total of 500 kernels
was produced. In the class with pale-
colored aleurone (no Spm), there were 117
kernels; 46 of them were Y and 71 were y.
Among the 383 kernels in the variegated
class (Spm present), 206 were Y and 177
were y. The data suggest that in both
these plants one Spm element was carried

249

in the Y chromosome and the other was
located elsewhere. In the three remaining
plants, the ratio of kernel types on the
ears deviated in another way from that
which might have been expected. Al-
though the number of kernels on these
ears was low, the deviation from a ratio
of 1 Spm to 1 no-Spm was obvious: 27 to
70, 43 to 111, and 36 to 66. On none of
these ears was there any evidence of link-
age of Spm with Y; in the Y class there
were 117 pale-colored kernels to 50 varie-
gated kernels, and in the y class there were
129 pale-colored kernels to 58 variegated
kernels. Frequent but late-occurring losses
or transpositions of Spm may have been
responsible for the observed deviation
from the expected 1:1 ratio, although other
causes may be considered. Progeny tests
are required before any definite conclu-
sions can be drawn regarding cause.

Test (3) is considered an important one
because in the two classes involved, the
presence or absence of Spm could not be
determined by observation of type and
distribution of anthocyanin pigmentation.
All kernels were homozygous for ai, and
since this recessive allele of 41 does not
respond to Spm, anthocyanin pigment was
absent in all kernels of these two classes.
Tests for the presence or absence of Spm
and for its location, if present, were con-
ducted with fifty-six plants derived from
the Y class of colorless, sz kernels and
with sixty plants derived from the y class
of such kernels. Each plant was crossed
by a plant homozygous for a:”-*» She, and
y and having no Spm—the Spm tester
stock described above. If no Spm was pres-
ent in a plant being tested, all kernels on
an ear resulting from this cross would be
uniformly pale-colored. If one Spm ele-
ment was present, half the kernels on an
ear would be pale-colored (no Spm) and
the other half would be variegated, with
colored spots on a colorless background
(Spm present). If more than one Spm
was present, the ratio of variegated to pale
kernels would be higher. With this mode
of testing for Spm, it was possible to learn
250

that no Spm was present in twenty-four
of the fifty-six plants derived from the Y
class of kernels, and that in the remaining
thirty-two plants one Spm element was
present. Its linkage with Y was clearly
expressed. in thirty of these thirty-two
plants. Among a total of 7792 kernels on
the ears produced by the thirty plants,
3985 were uniformly pale-colored (no
Spm); 1366 of them were Y and 2619
were y. The remaining 3807 kernels had
a colorless background in which spots of
deep color appeared (Spm present); 2472
of them were Y and 1335 were y. On the
basis of these data Spm may be placed in
chromosome 6, approximately 35 crossover
units from Y. It will be noted that this is
the same distance from Y indicated by the
data from test (1), given above. The ra-
tios of kernel types on ears produced by
two of the thirty-two plants having one
Spm element did not give clear evidence
of linkage of Spm with Y. On one ear
there were 70 pale-colored kernels and 88
variegated kernels. In the pale class, 30
were Y and qo were y; in the variegated
class, 47 were Y and 41 were y. On the ear
of the other plant there were 123 pale ker-
nels, 66 of which were Y and 57 y, and
154 variegated kernels, 79 of which were
¥ and 75 y.

Among the sixty tested plants derived
from the a1 she y class of kernels, seventeen
had a single Spm element and forty-three
had no Spm. On the ears produced by the
seventeen plants having Spm, after the test
cross described above, there was a total of
6746 kernels; 3465 of these were pale-
colored (no Spm), and 3281 showed spots
of. deep color on a colorless background
(Spm present).

These progeny tests again indicated that
Spm was located in the Y-carrying chro-
mosome 6 of the heterozygous parent
plants, and again placed it approximately
35 crossover units from Y. In linkage
studies with transposable elements, an er-
ror is always introduced into the. calcula-
tions of crossover distances, and the degree
of this is related to the frequency of oc-

CARNEGIE INSTITUTION OF WASHINGTON

currence of transposition of the element,
before gamete formation, to new locations
in the chromosome complement. As may
be noted from the several tests outlined
above, this frequency in the case of Spm
is not great enough to have a serious effect
on determinations of linkage relationships.

Test (2), mentioned earlier, was readily
conducted, in several different ways. The
plants being tested were assumed to be
a." She/a: she in constitution, and to
have no Spm. The absence of Spm was
confirmed in all cases by means of the
test cross outlined above. The presence of
a,""", carried in the Sh2 chromosome and
capable of responding to Spm, was readily
determined by crossing these plants to
plants homozygous for a: and shz, some
carrying an Spm element and others lack-
ing this element. On the ears produced
by the latter cross, nearly all the SA2 ker-
nels were pale-colored and, as expected,
nearly all the she kernels were colorless;
no variegated kernels appeared. On the
ears produced by the former cross, how-
ever, the two expected classes of kernels
appeared in the Sh class: those showing
spots of deep color on a colorless back-
ground (in which both a."* and Spm
were present), and those showing a uni-
formly pale color (in which a”* was
present but Spm was absent). Also, as
expected, nearly all the sh2 kernels were
colorless. The location of Spm in the a:
sh2 parent was known in some cases, and
the expected linkage with factors carried
in the chromosome that also had Spm was
made evident on the ears that resulted
from their use in these crosses.

The tests outlined above have been de-
scribed here in some detail in order to
indicate the necessary initial analytical
methods in an investigation of the basic
mode of operation of this two-element sys-
tem. With the general mode of operation
defined, it was possible to conduct a num-
ber of further tests. Some of these were
designed to determine the number of Spm
elements present in individual plants of a
particular progeny, when the presence of
DEPARTMENT OF GENETICS

two or more was suspected in the parent
plant. Others were aimed at determining
various locations in the chromosome com-
plement that may be occupied by Spm. At
present, two positions in chromosome 6,
two in chromosome 5, and two in chromo-
some g have been identified. Spm also
occupies other sites in the chromosome
complement that have not yet been located.
In another series of tests, individuals hav-
ing two Spm elements, located at allelic
positions in a pair of homologues, were
tested in order to determine the frequency
of loss of Spm from the female germ cells.
It was found to be absent in approximately
6 to 10 per cent of the female gametes pro-
duced by these plants. The majority of
such losses of Spm occurred late in the de-
velopment of the germinal tissue.

In addition to the tests just discussed, an
extensive series of tests was also conducted
with each of eight distinctly different states
of a”*. This was done in order to ex-
amine the mode of control of change in
gene action at 4: exhibited by each state
in the presence of Spm, to discover the
type of gene action appearing in its ab-
sence, and to determine the stability of
each state—that is, its constancy—in the
presence of Spm. Also, several of the
states were combined in a single individ-
ual, and the independence of action of
each in the presence of Spm was deter-
mined. Allelic relationships of states were
revealed by segregation ratios in the prog-
eny of these individuals.

The states of a.™* and their significance.
From an examination of the various states
of a1"* it has been possible to learn about
the modes of control exerted by this
a.™*Spm system on mutation types, fre-
quencies of occurrence, and times of occur-
rence during the development of a tissue.
Control of all these resides in the element
located at Ai, and this is not influenced by
the number of Spm elements that may be
present, although Spm is required for the
manifestation of these controlled types of
expression. Seven of the eight different
states that have been studied were isolated

251

after a change that occurred in the element
originally introduced at the 4: locus. This
original state of a:”"* gives rise, in the pres-
ence of Spm, to many early-occurring
mutations, both in the plant tissues and in
the aleurone layer of the kernel. The in-
tensity of pigmentation these mutations
produce ranges from faint to deep. A
number of germinal mutations also occur,
and these give rise to alleles that are stable
in the presence of Spm. When plants hav-
ing this state of a:™” are crossed with

plants homozygous for a1, kernels on the

resulting ears will occasionally show a
decidedly modified pattern of mutation.
The kinds of mutation may be altered, or
their frequency of occurrence may be dif-
ferent, or their times of occurrence may be
shifted; or combinations of these several
identifiable alterations of expression may
appear in such kernels. Some of these ker-
nels were removed from ears, and plants
were grown from them. These plants, in
turn, were examined to determine the
behavior of a”? in them. It was found
that in the presence of Spm the pattern of
mutation appearing in the kernel from
which the plant arose reappeared in the
following generation. In other words, the
alteration at ax" responsible for the
changed pattern of mutation was main-
tained in chromosome reduplication and
thus was heritable.

A description of several of the derived
states may be used to illustrate the range
of their expressions. One state produces, in
the presence of Spm, only a relatively few
dots of color in an otherwise colorless ker-
nel, but these dots are intensely pigmented.
The plants also show only a few small
streaks of deep pigmentation in a nonpig-
mented background. In the absence of
Spm, the plants are darkly pigmented but
the kernels are only faintly colored, and
these expressions are constant in successive
generations as long as Spm is absent.
When Spm is returned to the nucleus, the
pattern of expression described above again
appears—a few dots of deep pigmentation
in a colorless background in the kernel and
252

a few streaks of deep pigmentation in the
plant. Only very rarely does this state of
a," give rise to a mutation in a germinal
cell, and no subsequent change of this
state to another state has yet been iden-
tified.

Another state, derived from the original
one, is somewhat similar to that just de-
scribed in its behavior in the presence of
Spm. In the kernels, dots of deep color ap-
pear in a colorless background, but their
number is larger. On an occasional ker-
nel, a large deeply pigmented spot may
also be present. The plants having this
state show a number of fine streaks of
deep pigmentation in a nonpigmented
background. In the absence of Spm, how-
ever, this state is readily distinguished
from the one just described, for now both
the kernels and the plants are intensely
pigmented. Return of Spm to the nucleus
brings back its suppressor action and calls
forth the pattern of mutation characteristic
of this state. Few germinal mutations oc-
cur, and changes of this state to another
state are rare.

Still another state gives rise to dots of
deep color in the presence of Spm, but
these are so numerous that the kernel ap-
pears almost solidly colored when viewed
from a distance. The plant also shows
numerous small streaks containing antho-
cyanin pigment. Only a few germinal mu-
tations occur. In the absence of Spm, the
kernels are lightly but distinctly pig-
mented, and the plants are also pigmented.
Another state gives rise, in the presence of
Spm, to many early-occurring mutations,
and these express the higher levels of pig-
ment intensity. Many germinal mutations
occur. In the absence of Spm, the kernels
having this state are lightly pigmented and
the plants also are pigmented. In this
case, as with all states so far examined, re-
turn of Spm by appropriate crosses in some
succeeding generation will call forth the
pattern of mutation characteristic of the
state.

One state differs from all others with
respect to the types of mutation it pro-

CARNEGIE INSTITUTION OF WASHINGTON

duces. In the presence of Spm, the muta-
tions may occur early in development. In
the kernels, the colored patches, represent-
ing mutant areas, show low levels of pig-
ment intensity. Only very rarely, indeed,
does a colored patch appear that expresses
the standard 4, phenotype. Many germi-
nal mutations occur, and among the ker-
nels having them the same low levels of
pigment intensity are shown. The plants
derived from these kernels are pigmented,
the intensity in each case corresponding to
that shown by the kernel from which the
plant arose. In the absence of Spm, the
kernels having this state show either no
color at all or only a very faint trace at
their base. In the plants, also, no antho-
cyanin pigment is detected on visual ex-
amination,

By intercrosses, it is possible to combine
two different states in an individual plant
or kernel. When Spm is present, the mu-
tation pattern produced by each of the
states is evident, indicating the autonomy
of each with respect to its mode of action.
This is well illustrated in kernels having
the state just described and also a state that
gives only late-occurring mutations that
are expressed in the kernel as deep-colored
dots. Both patterns of mutation appear in
these kernels: the pale-colored areas, many
of which are large, produced by the for-
mer state, and the deep-colored dots pro-
duced by the latter. There appears to be
no interaction between the states that af-
fects their individual modes of expression.
The autonomy of each state is also made
evident by the ratio of types of kernels that
appear on ears produced when plants hav-
ing two different states of a:”"" are crossed
to plants homozygous for a. The two
states segregate from each other at meiosis,
and a 1:1 ratio appears among the kernels.

Conclustons regarding the operation of
the a:""-Spm system. The general mode
of control of gene action and of mutation
by this a.”"*-Spm two-element system is
now evident. The element of this system
that is inserted at the A: locus plays a
major part in controlling gene action and
DEPARTMENT OF GENETICS

in effecting changes in such action, both
in the presence and in the absence of the
second element, Spm. The Spm element
exerts a direct influence on the element lo-
cated at 41, in two distinctly different
ways. First, in the absence of Spm some
gene action occurs at A:, but in its pres-
ence this is totally suppressed. Secondly,
Spm induces modifications of the element
residing at A: that do not occur in its
absence. Two kinds of modification arise.
One effects a stable mutation, and the mu-
tants so formed give rise to a series of
alleles that differ from one another both
quantitatively and qualitatively. The sec-
ond type of modification, of rarer occur-
rence, leads to a change in the controlling
element at 4:—a change in state—that is
subsequently discerned. In the presence of
Spm, these modifications are expressed by
changes in the kinds of mutations that oc-
cur, their frequencies of occurrence, and
their times of occurrence during the devel-
opment of the tissues. These modifications
of state also affect the degree of action of
the genic materials at the A locus in the
absence of Spm.

Spm may be transposed from one loca-
tion to another in the chromosome comple-
ment, both in somatic and in germinal
cells, without losing its specificity of action
in the process. Thus, loss of Spm from
some nuclei and increase in others may
occur within an individual plant. An in-
creased number of Spm elements in a
nucleus is not made evident by changed
patterns of mutation at ai". This con-
trasts greatly with the case of the a:-Dz
system, where increase in number of Dt
elements is made evident by increased fre-
quencies of occurrence of mutation at the
a: locus. It also contrasts with the behavior
of Ac in the Ds-Ac system. With regard
to a" and ai", both of which express
control of mutation at A: by the Ds-Ac
system, successive increases in number of
Ac elements retard in a stepwise manner
the time of occurrence of mutation at the
modified A: locus in each case.

Knowledge gained from an examination

253

of the mode of operation of this system of
control of gene action and mutation, and
a comparison with other systems operating
at the very same locus, has greatly en-
larged our appreciation of the probable
significance of such systems in regulating
gene action during development. Somat-
ically occurring changes in gene action,
both gross and subtle, can result from their
operation, and these are well regulated
with regard to both time of occurrence and
type of change.

ContTINUED Stupies oF THE Mops oF OPER-

ATION OF THE CONTROLLING ELEMENTS
Ds anv Ac

Several other projects were carried out
during the year. Two of them were con-
cerned with the elements Ds and Ac. The
general mode of behavior of these two
elements has been described in many previ-
ous reports. Although the tests conducted
this year were many and the data obtained
were extensive, only the most significant
evidence and conclusions will be given
here.

The first of these projects was an analy-
sis of the direct control by Ac of change in
gene action at the bronze (bz) locus in
chromosome 9 when it is inserted there. It
could be demonstrated that these changes
are associated with events affecting the de
element itself. They give rise to several
different phenotypic expressions of the
genic materials at the bronze locus: a
stable recessive (fz) expression, a full
dominant (Bz) expression, and an expres-
sion that is intermediate between these
two extremes. Stability of the mutants de-
pends on whether or not Ac is removed
from the immediate vicinity of the bronze
locus by the event that affects it and results
in the change in genic expression. Tf it is
removed, the mutant is stable in subse-
quent generations. If it remains, the mu-
tant is unstable, in that subsequent altera-
tions of Ac may lead to further change in
action of the genic materials at the locus.
The time of occurrence, during the devel-
254

opment of a tissue, of these mutation-in-
ducing alterations of Ac depends on the
total dose of Ac present in the nucleus:
the higher the dose, the later they will
occur. Such responses to de dose may be
effected either by increasing the number
of chromosomes 9 carrying Ac at the
bronze locus—from 1 to 3 in the kernel
and from 1 to 2 in the plant—or by adding
Ac elements that are located elsewhere in
the complement when only one chromo-
some 9 carries 4c at the bronze locus.
Some of these Ac-altering events at the
bronze locus give rise to a dicentric chro-
matid and the corresponding acentric frag-
ment. Changes in the frequency of occur-
rence of this type of event characterize
some of the changes in state of Ac. One
other modification was detected, and it is
of general significance in considering in-
terrelations that may arise between con-
trolling elements. One kernel was found
that showed a marked increase in fre-
quency of occurrence of mutation to Bz.
The tissues of the plant grown from it
exhibited the same increased frequency.
Tests of this plant revealed that an altera-
tion had occurred at the bronze locus in a
germinal cell of the parent plant, resulting
in a modification of the mode of control
of subsequent mutation. Ac was still pres-
ent and was required for the occurrence of
these mutations, but it was no longer lo-
cated at the bronze locus in the short arm
of chromosome 9. The mode of control
now followed that which characterizes the
Ds-Ac two-element system, in which the
Ds element directly controls the change in
gene action at the locus where it resides,
and the Ac element governs the occurrence
of these mutation-inducing events at Ds.
It must be concluded, therefore, that this
modification arose from substitution of a
Ds element for the Ac element at the
bronze locus; or that the Ac element orig-
inally inserted there is compound and may
be separated into two components, a Ds
and an Ac element; or, possibly, that a
Ds clement may originate from some
modification of an Ac element. No evi-
dence is now available to suggest which

CARNEGIE INSTITUTION OF WASHINGTON

of these alternatives is most probable.
Nevertheless, the observed change from a
one-element to a two-element system of
control of gene action and mutation is sig-
nificant in considering the relations that
exist between controlling elements and
systems of such elements.

An additional project that received much
study was concerned with the behavior of
Ds after its insertion just to the left of
Sfx in chromosome g. In this position, it
induces changes in action of genic sub-
stances located on either side of it, and
these effects may include a segment of the
chromosome six or more crossover units
long in the standard chromosome g. Dur-
ing the past year, several cases of extended
modification of gene action, in which the
genic components farthest removed from
Ds exhibited reversion to standard expres-
sion, were examined. In all cases, it could
be determined that the Ds element was a
component of the segment of chromosome
showing altered genic action, and that the
reversions observed were accomplished by
some change involving the Ds element
itself. The patterns of reversion were those
associated with the operation of the two
elements, Ds and Ac: Ac was required for
their occurrence, and the times of occur-
rence reflected the dose of Ac that was
Present in the. nucleus. In the cases ex-
amined, the reversions to normal action of
a genic component within the modified
segment were not accompanied by loss of
Ds or by change in its location. This is
unlike the behavior of Ds at some other
known loci. In these other cases, it has
been determined that reversion of gene
action is often accompanied by removal of
Ds from the immediate vicinity of the
locus concerned.

Many different tests have been made of
the behavior of Ds when inserted just to
the left of Shi, and all of them indicate
that Ds is effectively fixed in location after
its insertion there. It can continue, then,
to exert its influence on the action of genic
substances located to either side of it. Thus
a sequential series of changes in action of
DEPARTMENT OF GENETICS

these genic materials can occur as long as
Ac is also present in the nucleus, and a
number of such sequences have been fol-
lowed through three or more steps. The
kinds of modification in gene action in-
duced by Ds at this location resemble, in
some respects, those appearing spontane-
ously at some other well-known gene loci
in maize, such as the R locus, whose nu-
merous alleles are known, as well as spon-

255

taneous rates of change to other alleles. It
is possible that there is a controlling ele-
ment or elements present at this locus, and
also at other loci in the standard chromo-
some complement of maize, and that the
modifications in gene action they induce
are responsible for the appearance of the
mutants and for the particular sequences
of change from one type of allele to an-
other that have been observed.