INDUCTION OF INSTABILITY AT SELECTED
. LOCI IN MAIZE

BARBARA McCLINTOCK
Department of Genetics, Carnegie Institution of Washington,
Cold Spring Harbor, N. Y.

Received April 14, 1953

JX previous reports (McC.iinTocx 1950, 1951), studies of the origin and

expression of genic instability at a number of known loci in the maize
chromosomes were summarized. It was concluded that changes in genic expres-
sion result from chromosome alterations at the locus of a gene and these are
initiated by. units other than those composing the gene itself. The mutations
are considered, therefore, not as changes in the potentials of genic action, but
rather chromosomal modifications at the locus effecting the kind and the degree
of genic expression. The extragenic chromatin units have specificity in that
differences among them may be recognized. Each exerts a specific type of con-
trol of the action of the gene with which it becomes associated. These units
may be transposed from one location to another within the chromosome com-
plement. When incorporated at a new location, each expresses its mode of
control of the action of the associated gene, and in a manner similar to that
which occurred at the former location. These conclusions have been supported
by extensive examination of the action of one particular system that has modi-
fied genic action at a number of different loci. It is the so-called Dissociation-
Activator (Ds-Ac) two-unit system.

Both Ds and 4c are single chromosomal units. Their locations in particular
chromosomes may be determined by linkage relations to other established
markers. Each, however, may move from one location to another within the
chromosomal complement. Such transpositions of Ds or Ac, or both, may
occur in a few sporogenous cells. Consequently, a few gametes may be formed
with Ds or Ac, or both, located at new positions. Following such transposition,
each remains at the new location until, in a subsequent cell or plant generation,
transposition to another location again occurs. Either may be inserted at vari-
ous locations within the complement. It has been determined that the process
of transposition involves some initial change at either Ds or Ac that can result
in chromosome breaks and fusions. It has also been learned that in order for
transposition of Ds to occur, 4c must be present in the nucleus. Besides trans-
position, there are other known consequences of events initiated by Ds and
again, these appear only when 4c is also present in the nucleus. They are
dicentric chromatid formation, deficiency, duplication of segments, inversions,
ring-chromosomes and reciprocal translocations between chromosomes. In all
of these cases, the chromosome breaks are produced as the consequence of
some initial change involving the Ds unit.

Genetics 38: 579 November 1953.
580 BARBARA McCLINTOCK

When Ds is transposed to the locus of a known gene, it may immediately or
subsequently affect its action. This is expressed either by partial or complete
inhibition, or by a previously unrecognized type of altered genic expression.
As long as Ds remains in this position, genic action is subject to further
change. It has been possible to determine that the subsequent changes are
reflections of alterations occurring at the locus, and these are initiated by Ds.
No Ds-initiated changes will occur, however, unless Ac is also present in the
nucleus. If 4c is absent, no further modification of genic action occurs. The
mutation present at the time of removal of 4c will be stable in expression in
subsequent generations until Ac is again introduced into a zygote. In some
cells of the plant arising from this zygote, alterations at the locus of the gene,
initiated by Ds, again will occur. Some of these may result in further modifi-
cations of genic action—that is, new mutations. Both the initial mutation and
subsequent mutations are expressions, therefore, of interaction between Ds and
Ac. Thus, these two chromosomal units comprise a nuclear system capable of
controlling genic expression; Ds initiates changes in genic expression and tc
controls when they will occur. Since Ds may enter various locations in the
chromosome complement, 4c-controlled mutability can be expected to arise at
a number of different loci of known genic action. In previous reports, the
origin and subsequent behavior of 4c-controlled mutability at the C, Bs and
"x loci in the short arm of chromosome 9 were described. It is the purpose
of this paper to indicate that, in conformity with expectation, 4c-controlled
mutability can appear at other loci of known genic action. It will also be shown
that this type of control of instability may be obtained at selected loci.

To test the prediction that genic instability under the control of 4c may be
obtained at selected loci, two, considered to be particularly favorable for initial
tests, were selected. These are the locus of 4; in chromosome 3 long arm, and
that of 42 in chromosome 5 short arm. The genetically active components at
these two loci affect anthocyanin pigment formation in both the plant tissues
and in the aleurone layer of the kernel. The known recessive alleles are desig-
nated a; and a». When either is homozygous, no anthocyanin pigment is devel-
oped in the plant tissues or in the aleurone laver. Both recessives are stable in
the presence of Ac; tests designed to show whether or not mutations of a, or
a would occur when Ac was present in the nucleus were negative, Conse-
quently, it was possible to perform the following experiment. A number of
plants were grown that had a Ds unit located in the long arm of one chromo-
some 5 and an Ac unit located in another chromosome of the complement. All
plants were homozygous for 4, and Shy and for 4, and Bmy,. (Shz, normal
development of endosperm; she, shrunken endosperm, located a quarter of a
crossover unit distal to 4;. But, colorless secondary cell walls in plant tissues :
bm, brown mid-rib, brown color in secondary cell walls; located 6 crossover
units proximal to 4s and in same arm of chromosome 5.) The silks of some of
these plants received pollen from plants homozygous for ay, Shy and Ay. The
silks of other plants received pollen from plants homozygous for ag, bm and 44.
None of the plants used as pollen parents carried 4c. The ears resulting from
INSTABILITY AT SELECTED LOCI IN MAIZE 581

these crosses were examined to determine if any kernels exhibiting variegation
for aleurone color were present.

If a sufficiently large number of ears were obtained from the described
crosses, one or more variegated kernels should appear. This conclusion is based
on the following reasoning. In plants carrying Ds and Ac, transposition of Ds
may. occur in a few sporogenous cells. Should one such transposition insert Ds
at the locus of A, total or partial inhibition of 4; action could result. The
gametophyte and consequently the gametes derived from this sporogenous cell
could carry an 4, locus with a Ds-initiated altered capacity for action. Should
Ac also he present, further Ds-initiated changes at the locus could occur in
some cells during development of the kernel that arises from functioning of
this gametophyte. In the progeny of such cells, these subsequent changes in
4, action could be expressed if the pollen parent had contributed the recessive,
a,, which is stable in the presence of Ac. The aleurone layer of the mature
kernel would exhibit a variegated pattern of anthocyanin pigmentation, either
with respect to presence and absence of pigmentation, or to intensities of pig-
mentation, or both. This kernel could then be removed from the ear, a plant
grown from it and tests conducted with this plant and its progeny to determine
whether or not the variegation is an expression of instability at the 1, locus
and, if so, whether this instability is 4c-controlled. In a similar manner, alter-
ations at the 4» locus that arise in the Ds and Ac carrying plants could be
detected if the pollen parent contributed the stable recessive, a». In order to
establish that the unstable state in either case originates in the Ds and de
carrying plant, it is necessary for one of the parents to introduce a second,
closely linked genetic marker. In progeny tests, linkage of unstable 4; with
Shy or of unstable 4 with But, would indicate inception of instability in the
Ds and .dc carrying parent plant.

Seventy-one ears were obtained from the cross in which the pollen parent
had contributed a,. One kernel was found that clearly exhibited variegation for
aleurone color. There were no colorless kernels on any of these ears. Among
the 120 ears obtained when az had been introduced by the pollen parent, three
kernels exhibiting variegation for aleurone color appeared, each on a different
ear. Again, there were no colorless kernels on any of these ears. Plants were
grown from all four variegated kernels and tests initiated with each to deter-
mine the nature of the instability being expressed. The progeny, in turn, were
further tested. From these studies, the following was determined. In the varie-
gated kernel that had received a, from the pollen parent, and in the plant de-
rived from it, instability of genic action at the locus of 4, was being expressed.
The alteration responsible for this instability occurred at the locus of 4, in one
of the chromosomes 3 of the Ds-de carrying female parent. Mutations at this
modified 4; locus, designated a,""*, occur only when Ac is present. In the
plants derived from the three variegated kernels that appeared when az had
been the pollen parent, the nature of the alteration at 42 could be determined
in only two of them. This was because in one of them, the chromosome 5 con-
tributed by the Ds-.4c carrying female parent was not transmitted to the next
582 BARBARA McCLINTOCK

generation either through pollen or egg. Thus, no definite conclusions may be
drawn concerning the nature of the instability, presumably at the A», locus,
that was exhibited in the kernel from which this plant arose. In the two re-
maining plants, it could be shown that it was the 4» locus in the chromosome 5
contributed by the female parent that had been modified. In one plant, the
mutations occurring at this locus proved to be Ac-controlled. This mutable
locus is designated a,”*. In the second plant, the mutations were not Ac-con-
trolled. This mutable locus is designated ag”"3,

Some of the methods used to determine whether or not mutation at a par-
ticular known locus is 4c-controlled will be outlined. For illustrative purposes,
the 4; locus has been chosen. Four independent inceptions of instability at 4,
have been detected in the Cold Spring Harbor cultures. Two of them, a,”
and a,”*, are not Ac-controlled. Two of them, however, a,”"3 and a", are
Ac-controlled. Both of the latter had their inception in plants homozygous for
the normal 4; locus and in which both Ds and Ac were present. The desig-
nation a,""4 was given to the case whose origin was outlined above. In this
report, the methods used to analyze the factors responsible for mutation at
a,""* will be considered. These same methods, however, have been used to
analyze all cases of Ac-controlled mutability. It is because of the very close
linkage of She to a;”*, allowing the latter to be readily followed in progeny
tests, that instability at the 4, locus has been selected for illustrative purposes.

In considering 4c-controlled mutability, the following facts should be kept
in mind. Mutations will occur only when Ac is present in the nucleus. In the
absence of Ac, the modified genic action at the locus is stable. Instability, how-
ever, will return if Ac is again introduced into a nucleus having this modified
locus. Ac controls when mutations will occur; the higher the dose of Ac, the
later the time during the development of a tissue when mutations at the
affected locus will occur. Ac is inherited as a single unit. Transpositions or
alterations of Ac, however, may occur in a few sporogenous cells resulting in
loss to one of two sister cells, change in location of Ac in the chromosome
complement, or change in dose action (change in state). Some of the gametes
produced by plants carrying Ac will be derived from cells in which such events
occurred. Evidence for these statements will appear in the descriptions of the
tests conducted with a,"4,

The pattern of variegation in the original kernel having a," resembled that
expected of mutable a;, Within the aleurone layer, many small areas of color
appeared in a colorless background. A plant was grown from this kernel in
the greenhouse during the winter of 1951-52, and given the culture number
6110. This plant was: (1) self-pollinated, (2) crossed to two plants having no
Ac factor that were a; She/a, sho, (3) crossed toa plant that was 4, She/a; she
and carrying a single Ac factor, and (4) crossed to a plant with a genetic
constitution designed to test for the presence of Ac in plant 6110. The nature
of the Ac tester stocks will be described shortly. The number of crosses con-
ducted in the greenhouse was limited. The results of these crosses did indicate.
however, that plant 6110 carried a newly arisen alteration at the A, locus; and
INSTABILITY AT SELECTED LOCI IN MAIZE 583

that this alteration was responsible for the expressed instability. The constitu-
tion of plant 6110 was a,”* Sh2/a, she and it carried one Ac factor. The evi-
dence also suggested that the mutations occurring at a,""* were Ac-controlled.
From these greenhouse crosses, kernels with appropriate constitutions were
selected and plants obtained from them in the summer of 1952 in order that
more adequate tests of the nature of the instability could be conducted.

The self-pollinated ear of plant 6110 produced 228 kernels with the follow-
ing appearances: 7 completely colored, Shy: 129 variegated (exhibiting areas
of color in a colorless background), She: 49 colorless, She: 43 colorless, sho.
In the crosses to plants that were a; Sh2/ay sha and in which no Ac factor was
present, 888 kernels were produced. They could be segregated into the fol-
lowing classes: 13 completely colored, Shg: 225 variegated, Sh, : 438 colorless,
Sha: 212 colorless, sz. In the cross to the plant that was 41 Sfi2/a, shy and
had 1 4c factor, 273 kernels were produced: 145 completely colored, Sha : 1
completely colored, sh. :29 variegated, She : 17 colorless, Shz: 81 colorless,
shy. The several distinctive patterns of colored spots exhibited by the varie-
gated kernels derived from this latter cross suggested Ac-control of mutability
at a,"*, These resembled the patterns produced with single to triple doses of
Ac. Kernels with these different doses of Ac were expected to be present on
this ear because each parent carried one .4e factor. Other evidence also sug-
gested that the mutations were -4c-controlled. This appears in the ratio of
variegated to nonvariegated kernels resulting from the cross of plant 6110 to
the plants that were a, She/a, she and had no Ac. If Ac were not linked to
a,"4, only one-half of the a,"* She carrying gametes produced by plant 6110
would also carry 4c. And, if Ac controls mutation at a,"*, then in only one-
half of the a," She carrying kernels could mutations occur, Since ay"+ and
Sho are very closely linked, the expected ratio of kernel types from the de-
scribed cross should be: 1 variegated, Shz:2 colorless, Sho: 1 colorless, she.
The observed ratios were: 225: 438: 212, which is a close approximation to
this expectancy.

Before presenting the evidence that establishes Ac-control of mutation at
a;""4, the phenotypes produced by mutation will be described. The initial change
at 4, resulted in complete inhibition of anthocyanin pigment formation in the
aleurone layer of the kernel and in the plant tissue. Subsequent mutations
occur. Two main classes which result in pigment formation can be recognized.
In one class, the mutations reestablish the original phenotypic expression of
A,: deep pigmentation in aleurone and plant. In the aleurone layer, the borders
of such mutant areas are not precisely defined because of diffusion of pigment
forming substances into the surrounding nonmutant cells. This produces * dif-
fusion rims” about these mutant areas. Within the second class, the muta-
tions give rise to a graded series, with respect to intensity of anthocyanin
pigmentation ; from light to relatively dark shades. The borders of areas having
these mutations are sharp. Diffusion rims, characteristic of the first class of
mutation either are not present or are very weakly expressed. The phenotypes
exhibited by the kernels carrying germinal mutations also express these same
584 BARBARA McCLINTOCK

two main classes of mutation. In these cases, however, the color intensity in
any one kernel is the same throughout the aleurone layer. The plant tissues
likewise show a graded series of intensity of pigmentation among the mutant
areas. Marked changes in state of a," that result in altered relative frequen-
cies of the various types of mutation, were rarely observed. This is in contrast
to ay") and a,”°, both of which have produced many modified states, each
exhibiting a particular type or types of mutation and/or frequency of type.

In order to determine that Ac controls mutation at a", several types of
tests must be applied and positive results obtained from each. These are: (1)
establishment of the presence of dc in plants showing mutations at a,”"*:
(2) establishing that a,)"4 is nonmutable in the absence of 4c but that return
to mutability will occur when Ac is again introduced into the nucleus; (3)
establishing in this case that it is the same Ac factor known to produce breaks
at Ds and to control mutations at other loci of known genic action, which is
responsible for controlling mutation.

The presence of Ac can be detected because it produces breaks at Ds, wher-
ever it may be located, or mutations at some loci of known genic action (see
table 6). In order to be able to test readily for the presence of 4c in a particu-
lar plant, it has been necessary to develop so-called Ac tester stocks. Many of
them utilize genetic markers in the short arm of chromosome 9 affecting the
phenotype of the endosperm of the kernel, and also Ds at its standard location
in this arm. For illustrative purposes, the test procedure with one such stock
may be described. This stock is homozygous for the chromosome 9 endosperm
markers I, Shy, Bs and Wx, and for Ds at its standard location, proximal to
Wx. CU, inhibitor of aleurone pigment formation; dominant to allele C, re-
quired for aleurone color development. Sh, normal development of endosperm ;
shy, shrunken endosperm. Bz in presence of C produces dark color in aleurone
and plant; with the recessive, bz, color is modified to a bronze shade. Hx pro-
duces amylose starch in endosperm and pollen grain which stains blue with
I-KI solutions ; no amylose starch produced by recessive, wr, and starch stains
red-brown with I-KI solutions. The order of these markers in the short arm
of chromosome 9 is: f Sh, Bz H’x Ds centromere.) No Ac factor is present in
the plants of this stock. Consequently, no breaks at Ds occur in these plants.
If, for example, the Ac constitution of a plant homozygous for C, sh, bs and
wx is to be determined, a cross to or by the Ac tester plant is made. If the
plant being tested has no Ac, all the kernels on the resulting ear will be color-
less, nonshrunken, and H’x. If one Ac is present, half of the gametes produced
by the C, shy, bs, wx plant will carry Ac. The other half will have no Ac. The
kernels on the resulting ear will exhibit this gametic ratio for Ac. Those that
do not have Ac will be totally colorless, nonshrunken, and H’x, Those that
have Ac will be variegated. In these kernels, sectors will be present that show
the collective recessive phenotype: C, si, bs and wa. These sectors are pro-
duced because Ac induces breaks at Ds in the J Shy Bz W’x Ds carrving chro-
mosome in some cells during the development of the endosperm. These breaks
result in the formation of acentric fragments carrying all of the dominant
INSTABILITY AT SELECTED LOCI IN MAIZE 585

markers. The fragment, in each case, is lost to telophase nuclei in subsequent
mitoses. Therefore, all cells arising from those in which the-dominant markers
have been removed will express the collective phenotype C, sity, bs and wr.
If the plant being tested has two .4c factors, located at allelic positions in one
homologous pair of chromosomes, all of the kernels resulting from this cross
should be variegated. (Usually, however, a small percent of the kernels are
nonvariegated either because no dc is present or because of a high dosage
action of de which so delays the time of Ds breaks that none occur during the
development of the kernel. These conditions arise from transposition or change
in state of fc that can occur in a few sporogenous cells of lc carrying plants.)
If two de factors are present and these are nonlinked, a ratio approximating
1 nonvariegated to 3 variegated kernels will appear on the ears resulting from
the test cross. If more than two Ac factors are present, not only the ratios
obtained but also the timing of breaks at Ds may be used to detect the number
of Ac factors that are present. In these cases, however, verification must be
obtained by progeny tests.

The type of test outlined above may be used to detect positions of lc when
it is located in the short arm of chromosome 9. For example, plants having the
constitution 4c C Sh, Bs W’x/ae C sh, be wy or ones that are C Shy Be Ac lx
/C shy bz ac wx have been crossed by Ac tester plants that are homozygous for
I, Shy, Bz, wx, Ds and have no Ac. The phenotypes exhibited in the sectors
of the variegated kernels on the resulting ears, and the frequency of kernels
with sectors of a particular type, indicate in each of these cases, the location
of te within the short arm of chromosome 9. Subsequent tests of the progeny
arising from the various crossover classes of kernels in each of these two cases,
affirm the particular location of ie.

Tests showing that 4c controls mutation at a,""* may now be described.
Kernels having particular phenotypes were selected from the self-pollinated ear
of plant 6110, and from the ears derived from the two crosses described above.
In plant 6110, the chromosome 3 factorial constitution was: ay?" Sho/ay she;
the chromosome 9 factorial constitution was: C Hx/C wx. The necessity for
indicating this latter constitution will become apparent as the tests are de-
scribed. One .4c was present in plant 6110 but it was not linked to the marker
in chromosome 9 and appeared not to be linked to those in chromosome 3.
Four classes of kernels were selected from the self-pollinated ear: (1) color-
less, Sho, Wx, (2) colorless, Shy, wx, (3) variegated, Sh, Wx, and (4) varie-
gated, Sh», wx. Since the locus of Shz is very close to that of a,""*, plants
derived from the first two classes of kernels were expected to be either
ay” Shs /ayt Shy or ay”* She/ay she in constitution and to carry no ele fac-
tor. Those derived from types (3) and (4) above were expected to have either
of these two constitutions with respect to markers in chromosome 3; they
should have, however, either one or two 4c factors, and if two, these should
be located, in the majority of such plants, at allelic positions in one pair of
homologues.

The constitutions with respect to siz in the plants derived from the above
TABLE 1
Constitution of plants of culture 6424 (column 1) and types of crosses (columns 3 to 8).
that bave revealed these constitutions for each plant,

Crosses with plants of constitutions given in column headings

 

 

sos Plant Self- a,sh,/a,sh a, Sh,/a, sh. a, Sh,/a, Sh Ac tester a,/a
Constitution of plant no. pollinated “No Ac "No Ac "No Ac plant Ac
a, Sh,/a,™* Sh,; No Ac 1 + + +
(plants 1 to 5) 2 +
3 + + + +
4 + + +
5 + + +
a,” Sh,/a,"~* Shy; 1 Ac 6 + +
(plants 6 to 8) 7 + +
8 +
a," Sh,/a™™ Shy; 2 Ac 9 + + + +
a,"~* Sh,/a, shy; No Ac 10 + +
(plants 10 to 17) 11 + +
13 + + + +
13 + + +
14 + + +
15 + + +
16 + + + +
17 + + + +
a" Sh,/a, shy; 1 Ac 18 +
(plants 18 to 25) 19 + + + +
. 20 + + + + +
21 + + +
22 + + +
23 + +
24 + + + +
25 + :
a," Sb,/a, shy; 2 Ac 26 + + + +
(plants 26 to 29) 27 + + + +
28 + +
29 +

98S

MDOLNIIDW VaVEUVa
INSTABILITY AT SELECTED LOCI IN MAIZE 587

selected kernels were determined by crosses to plants that were either a; she
/ay she or a She/a, she. (Supply of plants of the former. constitution was
limited due to early death from disease of many of them. In some of the tests
it was necessary, therefore, to substitute plants of the latter constitution.) Ac
constitutions were determined for many of them by crosses with Ac tester
stocks. Detection of the presence of a,"* in plants derived from the colorless
kernels required the introduction of Ac. For this purpose, plants that were
homozygous for a, and that carried one or more Ac factors were used in
crosses with the plants derived from these kernels. To show that the factor
controlling mutation at a,”"* is the same as that which controls breaks at Ds,

TABLE 2

Ac constitution of gametes of plants of culture 6424 (table 1) determined by

crosses to plants’ of Ac-tester stock, Constitution of tester stock: 1Ds/IDs;
A,/Ay no Ac.

 

 

 

Plant Parentage Kernel types on resulting ear
Ac constitution no. in of 6424 -
of plant culture plant in Colorless Variegated: colored areas in
6424 cross (no Ac) colorless background (Ac)
Group I, No Ac 3 é 324 0
5 co 128 0
12 3 260 0
14 é 123 0
15 3 11 0
16 3 216 0
Group I, 1 Ac 6 3 251 248
7 3 252 288
19 3 377 397
20 3 135 142
20 3 206 188
21 2 151 135
22 $ 154 128
23 g 42 46
23 é 145 134:
24 So 182 191
Group III, 2 Ac 9 $ 47 410
26 3 15 457
27 3 23 434

 

some plants arising from the colorless, Sho, w+ class of kernels were crossed
by plants with the constitution : a,/a,; Wa Ds/we ds;1 Ac. The Ac factor was
not linked to markers in chromosome 9.

The constitutions of 29 plants in the progeny derived from self-pollination
of plant 6110 are entered in table 1 along with the tests conducted with each
plant that served to indicate its constitution. These plants were grown in cul-
ture number 6424. Nine plants in culture 6424 were a,"* S ho/ay™* Shy. Five
of them had no 4c (plants 1 to 5, table 1), three had one de (plants 6 to 8,
table 1) and one had two Ac factors (plant 9, table 1). Twenty plants in cul-
‘ture 6424 were a," Sho/a; sho and among them, eight had no dc (plants 10
to 17), eight had one Ac (plants 18 to 25), and four had two Ac factors
588 BARBARA McCLINTOCK

TABLE 3

Types of kernels appearing on ears resulting from crosses of plants of culture
6424 (table 1) with tester plants having constitutions entered in column I. Part 1;

Tester plants have no Ac. Part Il: Tester plants have 1 or 2 Ac factors.

 

 

 

 

Parel
Constitution Plant Parentage Kernel types
of tester nl in ef 6424 Colored Colored Varie- Varie- Colore Colore Totals
plants “E404. plant in Sh, ° ih gated gated less less
bahia 2 SO Sb, shy Shy sh,

41 hy/a, sh, 1 2 0 0 0 0 230 0 230
No Ac 3 2 0 0 0 0 133 Oo 133
3 3 0 0 0 0 573 0 $73

4 2 0 0 0 ) 298 0 298

Totals, plants 1 to 4 0 0 0 0 1234 QO 1234
6 g 1 0 132.0 126 0 259

6 3 7 0 507 0 554 0 1068

7 g 1 0 1332, 0 151 0 284

7 é 15 0 275 0 295 0 585

8 9 3 0 168.0 188 0 359

Totals, plants 6 to 8 27 0 1214 0 1314 0 2555
9 é 26 0 584 0 95 0 705

12 2 0 0 0 0 120 125 = 245

16 2 ) 0 0 0 131 120-251

16 o 0 0 0 0 144 «137,281

17 g 0 0 ) 0 65 55-120

Totals, plants 12 to 17 0 0 0 0 460 437 897
18 g ) 0 75 0 73 134-282

19 g 3 0 62. 0 83 167 315

19 é 12 ) 199 ) 216 =90375~—s« 802

20 9 0 0 53 0 58 114 225

20 3 12 0 2340 267 444—<O9S7

21 9 4 0 8 oO 75 168) =—-333

22 2 1 0 or) 0 79 «154 333

23 2 1 0 69 0 70 157 297

23 o 20 0 179 1 185 333s 718

24 é 9 0 174 3 210 363759

25 g 1 0 54 0 72, «123~——(«250

Totals, plants 18 to 25 63 0 1284 4 1388) =2532— $271
26 g 1 0 166 1 0 152 320

26 s 4 0 140 0 6 141-291

27 g 0 0 134 0 35° «178 = 47

28 g ) 0 151 0 1 146 298

Totals, plants 26 to 28 5 0 591 1 42 617 1256
29 2 0 0 72.0 28 «125225

 
INSTABILITY AT SELECTED LOCI IN MAIZE 589

TABLE 3 (continued)
PartI (continued )

Plant Parentage Kernel types

 

 

 

 

 

Constitution no. in of 6424 -
of tester culture plant in Colored Colored Varie- Varie-~ Color- Color- Totals
plants 6424 ae Sh sb gated gated less less
2 2 Sh, Shy Sh, shy

@, Shz/ a, shy 1 3 0 ) 0 oO 573 0 573
No Ac 3 3 0 0 o oO 934 0 934
4 3 0 0 0 0 846 0 846

5 3 0 0 0 oO 894 0 894

Totals, plants 1 to 5 0 0 0 0 3247 0 3247
9 3 22 0 392, 0 49 0 463

10 é 0 0 0 oO 364 106 470

11 3 0 0 0 0 731 -236—Ss«9G7

12 é 0 0 0 0 705 «210295

13 s 0 0 0 oO 715 234 «= 949

14 é 0 0 0 0 644 200 844

15 3 0 0 0 0 431 127-358

17 ¢ 0 0 0 0 192 86 6-278

Totals, plants 10 to 17 0 0 0 0 3782 1199 4981
19 $ 4 0 1820 385 198 769

20 3 8 0 108 0 185 88 389

21 3 12 0 205 0 438 224 879

22 3 2 0 206 0 525 249 982

24 é 6 0 130 1 237. «134 = 508

Totals, plants 19 to 24 32 0 831 1 1770 893 3527
26 3 8 0 532 0 294 280 1114

28 é 3 0 457 0 243 194 897

Totals, plants 26 and 28 11 0 989 0 537 474 =«-.2011
@,Shy/a,Sh, 16 s ft) 0 o 860 535 0 535
No Ac 20 3 8 0 132 0 432 0 572
24 3 3 0 111 0 347 0 461

Totals, plants 20 and 24 11 7) 243 0 779 0 1033

Part II

@,Shy/a, shy 4 3 0 0 244 0 243 0 487
1 Ac 11 é 0 0 78 0 75 145 298
12 3 ) 0 76 fr) 60 123 259

13 Ss 0 0 125 ) 111 268 504

15 3 0 0 34 0 35 60 129

16 3 0 0 68 0 62 135 265

17 3 0 0 54 0 61 149 =. 264

Totals, plants 11 to 17 0 0 435 0 404 880 1719

 
590 BARBARA McCLINTOCK

TABLE 3 (continued)

Part II (continued)

 

 

 

 

 

c oe Plant Parentage Kernel types
onstitution no. in of 6424 . .
of tester culture plant in Colored Colored Varie- Varie- Color- Color- Totals
plants 6424 cross Sh sh gated gated less less
2 3 Sh, sh, Sh, shy
aSh7/ ashy 1 3 0 0 2510 210 0 461
1 Ac 2 2 0 0 161 0 158 0 319
3 3 0 0 387 0 384 o 771
4 és 0 0 157 0 157 0 314
5 g 0 0 119 0 115 0 234
Totals, plants 1 to $ 0 0 10750 1024 0 2099
10 g 0 0 117, oo 203113 433
12 9 0 0 98 ) 184 83-365
14 8 0 0 103 202. «100 = 405
15 & 0 0 122, 0 207.0 «122——Ss«453
Totals, plants 10 to 15 0 0 440 0 796 418 1654
4 Shy/ a; Sh,
lAc 10 é 0 0 54 0 143 0 197
a, Sho/ a, sh,
2 Ac, non-
allelic and
non-linked 12 3 0 0 248 860 228 150 626

 

+88 of these kernels had relatively few, late occurring mutations (4 Ac); 160 had many
more mutations that occurred earlier in the development of the kernel (2 Ac).

(plants 26 to 29). The plants having no Ac factor were derived from the
kernels that were colorless. Those carrying Ac were derived from the varie-
gated kernels, There were no exceptions,

Direct tests for the presence of Ac were conducted with 19 of the 29 plants
in culture 6424. The Ac tester stock used in tests of 17 of them was homo-
zygous for A, and carried J and Ds (standard location) in each chromosome
9; no Ac was present. The types of kernels appearing on the ears when crosses
were made to plants in the tester stock are entered in table 2. The nature of
this test has been described above but may be summarized here. If all of the
kernels are colorless, then the plant being tested has no Ac. If half of the ker-
nels are colorless and the other half are variegated (colored areas in a color-
less background), one Ac factor is present in the plant being tested. If nearly
all of the kernels show colored areas in a colorless background, then the plant
being tested carries two dc factors. Two of the 19 plants directly tested for the
presence of de (plants 13 and 17) carried wx in each chromosome 9. The
silks of these two plants received pollen from plants homozygous for a; and
carrying C, Wx and Ds at its standard location in one chromosome 9 and C,
we and ds in the homologue; no Ac was present in these plants. All of the
kernels on the resulting ears were colorless; also, in those kernels that were

Wx, no wx sectors were present, indicating the absence of Ac in these two
plants.
INSTABILITY AT SELECTED LOCI IN MAIZE 591

The types of kernels produced on ears resulting from crosses of these 29
plants of culture 6424 to plants homozygous for a, are entered in table 3. The
a, stocks were either homozygous for she, heterozygous for she, or homozygous
She. Table 3 has two parts. In part I are entered the types and frequency of
type of kernels appearing on ears derived from crosses of 28 of the 29 plants
with plants homozygous for a, and having no Ae factor. In part II are entered
the types and frequency of type of kernels appearing when plants derived from
the colorless, She classes of kernels in culture 6424 (plants 1 to 5 and 10 to 17)
were crossed with plants homozygous for a; but carrying an Ac factor.

An examination of the data entered in part I of table 3 will reveal that all
of the plants having one or two Ac factors, as determined by direct tests
(Groups II and III, table 2), produced variegated kernels in crosses with
plants having no Ac factor. Also, germinal mutations were evident. Those
plants in culture 6424 with no Ac (Group I, table 2), on the other hand, pro-
duced only colorless kernels from the same type of cross. When, however,

TABLE 4

Relation of mutation at a“ in chromosome 3 to breaks at Ds in chromosome 9,
Constitution of parents: Sa,™*/a,"*; wx/wx, no Ac X 3 a,/ay Wx Ds/wx ds; 1 Ac.
P 1 1 1/4

Kernel types

 

 

 

Variegated Colorless Variegated Colorless
Parent plants aleurone aleurone aleurone aleurone Totals
wx areas wx areas
Wx in Wx Wx in Wx wx wx

2 3 background background
6424-2 6427B-9 5 61 77 0 102 103 348
6424-3 6427B-8 7 89 101 0 86 105 388
6424-4 6427B-7 6 50 60 0 63 55 234

Totals 18 200 238 0 251 263 970

 

these same plants (3 and 5; 12 to 17) were crossed with plants homozygous
for a, but carrying an Ac factor, variegated kernels appeared (part II, table 3}.
In both part I and part II of this table it may be seen that the variegated
kernels were present in the expected classes, and in the expected frequency
within each class, on the assumption that Ac is the factor responsible for the
occurrence of mutation at a,"*.

Results of the test, recorded in table 4, leave no doubt concerning Ac-con-
trol of mutation at a,"*. It is the same Ac factor that is responsible for pro-
ducing breaks at Ds, wherever it may be located, and for controlling mutability
at some other loci of known genic action (see table 6). This test was conducted
with plants 2, 3 and 4, each of which was homozygous for a,"* and She in
chromosome 3 and for wa in chromosome 9. Pollen from plants homozygous
for a), and carrying C, Wx and Ds (standard location) in one chromosome 9
and C, wx and ds in the homologue and also a single Ac factor, not linked with
the markers in chromosome 9, was placed on the silks of these three plants. If
Ac is the factor controlling mutation at ai” *, the types of kernels on the
resulting ear, and the frequency of each type, may be predicted. In the wr
592 BARBARA McCLINTOCK

class, two types of kernels should appear. Half of them should be colorless
(no Ac) and half should show variegation for color in the aleurone layer due
to mutations at a,"* (Ac present). In the Wx class of kernels, three types
should appear. Half of the Wx kernels should be colorless, and no we sectors
should appear within these kernels (no Ac). The other half of the Wx kernels
should be variegated for aleurone color (Ac present) and the majority of
them should also have sectors showing the wr phenotype. This is because Ds is
located close to Wx. Relatively little crossing over occurs between the loci of
these two factors. Thus, the majority of the Wx class of kernels will have Ds
in the W’x-carrying chromosome; in only a small percentage of them will Ds
be absent. When Ac is present, breaks occur at Ds in some of the cells during
the development of the kernel. These result in elimination of Wx. As a conse-
quence, sectors having the wa phenotype will appear in the mature kernels.

Among a total of 970 kernels entered in table 4, 501 were completely color-
less and 469 were variegated for aleurone color. Within the colorless class, 263
were wa and 238 were Wx. In these Wx kernels, no sectors showing the wr
phenotype appeared. Within the variegated class of kernels, 251 were com-
pletely wa and 218 were Wx. Among these Wx kernels, on the other hand,
sectors exhibiting the wx phenotype were present in 200 of them. In only 18
of them were wr sectors absent. It is obvious, therefore, that the factor con-
trolling breaks at Ds—that is, 4c—is also responsible for mutation at a4.
Without Ac, neither breaks at Ds nor mutations at a,” will occur. When Ac
is present, both will occur. With this information in mind, it is possible to
interpret readily the data entered in tables 2 and 3 that have not received
detailed comment. It is only necessary to indicate the essential correlations that
these data reveal.

In direct tests of 4c constitutions, table 2, eight plants proved to have one
Ac factor (plants 6 and 7, and 19 to 24). Plants 6 and 7 were homozygous for
She. Since She is very closely linked to a,"4, these plants could be expected
to be homozygous for a,"*. When they are used in crosses to plants homozy-
gous for a, and carrying no Ac, a ratio of 1 variegated to 1 nonvariegated
kernel should appear on the resulting ear. This ratio was obtained, as shown
in part I of table 3. Plants 19 to 24 were Sho/sho. Therefore, they are expected
to be: a,”* Sho/a, shy. When crossed to plants homozygous for a; and having
no Ac, a ratio of 1 variegated to 3 nonvariegated kernels should appear on the
resulting ear. If the tester plant is also homozygous for she, the variegated
kernels would be present almost exclusively in the Sho class, and the ratio of
kernel type in this class would be 1 variegated to 1 nonvariegated. The ratios
obtained from tests of these six plants (part I, table 3) fit this expectancy.

Direct tests of Ac constitutions showed that plants 9, 26 and 27 had two Ac
factors (table 2). Plant 9 was homozygous for Sh. and therefore could be
expected to be homozygous for a,”-4. The ratio of variegated to nonvariegated
kernels on the ears resulting from crosses to plants homozygous for a, and
having no Ac factor, should closely reflect the gametic ratio of presence and
absence of Ac. The ratios obtained (part I, table 3), show that this expectation
INSTABILITY AT SELECTED LOCI IN MAIZE 593

is fulfilled. (Because some of the gametes produced by these plants will not
carry Ac, a small percent of the kernels should be colorless, nonvariegated.
Also, a few of them may have Ac but carry an altered state of a,"* that no
longer gives visible inutations or gives so few of them that, in individual ker-
nels, none may have occurred. In order to distinguish between these two possi-
bilities, it is necessary to conduct progeny tests with the plants arising from
the nonvariegated kernels.) Plants 26 and 27 were Shz/sh2 and therefore could
be expected to be heterozygous for a,"*. Approximately 1 variegated to 1 non-
variegated kernel should appear in the backcross test to plants homozygous for
a, but having no de. If the tester plant is also homozygous for she, the varie-
gated kernels should be confined, with few exceptions, to the Shz class. As the
data in table 3 show, this expectation is fulfilled.

Once the fact is established that 4c controls mutation at a,"* and that this
mutable locus responds in the expected manner to increased doses of ec, it is
no longer necessary to apply to each plant all of the tests that have been out-
lined above. Constitutions may be determined by ratios of kernel types in back-
cross tests. Plants carrying altered states of a;"* or changes of dc affecting
its state, location, or number, may be identified by such tests. In these latter
cases, however, progeny tests often must be conducted to establish with cer-
tainty a particular conclusion drawn from the backcross test.

Mutation or change in state of an Ac-controlled mutable locus rarely occurs
early in the development of a plant, although a few such occurrences have been
noted in all examined cases. Most of the changes at the locus take place in the
later developmental stages of the sporophytic tissues. It is such late occurring
mutations that are responsible for the completely colored kernels recorded in
part I of table 3. Such kernels did not appear in the tests recorded in part II
of this table. In this part of the table, the a," carrying plants did not have Ac.
Thus, no germinal mutations were expected and none were found. The types
of germinal mutation occurring in plants having dc resemble in kind those
appearing in the variegated kernels. Among them are darkly pigmented kernels
and kernels showing various lighter shades of pigmentation.

The results recorded in table 3 suggest that very little crossing over occurs
between a;”* and Shs. Only 6 variegated kernels appeared in the sh class,
and 4 of these 6 appeared when pollen from one plant was used, plant 24.
Three of them appeared in crosses that were made with a single collection of
pollen from this plant. This suggests that mutation rather than crossing over
may be responsible for some of the variegated kernels in the she class.

Similar tests, although less extensive than those outlined above for the prog-
eny derived from self-pollination of plant 6110, were conducted with plants
arising from selected kernels resulting from crosses of this plant to plants that
were a; She/a, shy, and had no Ac, and to a plant that was 4; Sh2/ay sho and
carried one Ac. From this latter cross, variegated kernels considered to have
received an Ac factor from each parent were selected. These exhibited only a
few specks of color scattered over the aleurone layer. Tests of 6 plants arising
from them showed the presence of two Ac factors, but these were not linked
594 BARBARA McCLINTOCK

to each other and not linked to the marked factors in chromosome 3. Direct
tests for dc were conducted with several of these plants. A gametic ratio of
3 with Ac to 1 with no Ac was obtained. The ratio of variegated to nonvarie-
gated kernels obtained when these plants were crossed to ones homozygous
for a, and having no Ac gave the expected ratios: 3 variegated to 1 nonvarie-
gated kernel among those that received a,” (table 5).

The same types of tests that were used in the analysis of a," were also
conducted with plants carrying a2”3 and a)”-4. It was soon evident that muta-
tions occurring at a9”8 are not Ac-controlled. At a2” *, however, mutation is

TABLE 5

Types and frequency of type of kernels on ears resulting from cross of
+ a@,/ay no Ac X 3 a," Sh,/a, Sh: 2 non-linked Ac factors.

 

 

 

 

Constitution of Kernel types
parent with 3 Parent Colored Variegated Colorless Totals
reference to

Sh, and sh, Sh, sh, Sh, sh, Sb, sh,
sh,/sh, 6426A-1 1 0 48 0 13 65 127
85 6 0 214 0 88 318 626
”? B-2 1 0 97 1 22 107 228
Totals 8 0 3591 123. 490 981
Sh/ sh, 6426A-1 8 0 1120 142,87 349
22 29 0 443 0 479 315 1266
Bel 9 0 145 Q 173 107 434
"22 8 0 213 0 252 155 628
3 29 0 380 0 412 234 1055
Totals 83 0 1293 0 1458 898 3732
Sh,/Sh, 6426A-1 12 0 153.0 302 0 467
2 26 0 267 0 576 0 869
Totals 38 0 420 1) 878 0 1336

 

Ac-controlled ; and a3:"* responds to Ac in a manner comparable to that ob-
served for all other known Ac-controlled mutable loci (see table 6). A detailed
account of this analysis will not be included here.

DISCUSSION

From examination of instability of genic action at a number of known loci
in maize, it is concluded that mutations need not express changes in genes, but
may be the result of changes affecting the control of genic action. It can be
shown that this control is effected by nongenic agents carried in the chromo-
somes. Different agents are present, and may be distinguished by specificities
they exhibit in their mode of control of the action of genes. It has been learned
that the same agent may operate at different loci of known genic action and
that different agents may operate at any one locus. These agents, therefore,
reflect the presence of “extragenic” systems carried in the chromosomes,
which control genic expression. One of them is the Ds-Ac two-unit system,
INSTABILITY AT SELECTED LOCI IN MAIZE 595

TABLE 6

Known loci (column 1) at which mutability, under the control of Ac, has arisen
(column 2) and mutability at the same loci controlled by other systems (column 3).

(The figures following the symbols represent the sequence of appearance in the
Cold Spring Harbor cultures.)

 

 

Symbol of normal, Instability aed
dominant factor controlled system od
at locus by Ac ,
y than Ac
c_ Cc om ems
om?
ems
1
Sh, sb™
Bz bz? bam?
bz™"?
bem
Wx : went wm
wx wx
wx ® wxT"4
A an a,™""
1 ames am?
1 1
m4 met
A; a, a
a7"?
me=3
a,

 

‘For origins of sh™ from Sh, see MCCLINTOCK 1952.
whose action at the locus of 4, has been considered in this report. Detailed
summaries of its operation are given elsewhere (MCCLINTOCK 1951). There-
fore, considerations of the behavior of this system need not be repeated here.

Instability of genic action, under the control of the Ds-Ac system, has been
examined at six known loci (table 6). At four of them (C, Bz, Ai and As)
genic action is associated with development of anthocyanin pigmentation. At
one of them, Ix, it is associated with the chemical constitution of starch in
endosperm and pollen grain. In the sixth case, Shy, it is associated with a
morphological structure of the endosperm. At five of these six loci, instability
under the control of other systems has appeared in the Cold Spring Harbor
cultures, as indicated in the table. There appears to be no relation between the
primary type of action of the gene and the type of system that can serve to
control it.

Inception of instability at a locus of known genic action is interpreted as the
result of the insertion of a specific controlling unit at or adjacent to the locus.
Such insertion is an expression of the phenomenon of transposition, which is
characteristic of controlling units. Transposition in the case of Ds or Ac may
be detected readily, and its mechanism has been analyzed (McCLIintock
1951). It is initiated by chromosome breaks and fusions that occur at the sites
of the controlling units; and it results in their removal and insertion at other
positions in the chromosome complement.
596. BARBARA McCLINTOCK

In this two-unit system, it is the Ds component that is responsible, when
inserted at the locus of a gene, for modification of the action of the gene;
whereas the function of Ac is to control the occurrence of changes at Ds. Some
of the initiating events at Ds result in detectable chromosome alterations.
Others produce changes that are not microscopically visible and these probably
are responsible for most of the observed cases of altered genic expression.
Removal of Ds from the locus of the gene is one of them. Others may result in
modifications of the chromatin material at the locus, either structurally, quanti-
tatively, or by insertion of chromatin from elsewhere. In this case, a particular
type of change in genic expression might reflect a particular type or types of
change in chromatin components at the locus. Evidence in support of this
assumption has been obtained (McCrintock unpublished).

The Ac component also undergoes transposition, and its insertion at new
positions has been noted; but no cases of direct Ac control of genic change
have been observed. Ac controls the occurrence of changes at Ds, wherever Ds
may be located. Because methods of detecting the presence of Ac are both
simple and sure, it is possible to determine, in any case of mutability, whether
or not it is controlled by the Ds-Ac system. Also, because Ds may be trans-
posed to various positions in the chromosome complement, inception of Ac-
controlled mutability at different known loci can be anticipated. It is only
necessary to provide adequate means for their detection, such as those outlined
in this report.

Examination of many different cases of instability of genic expression has
shown that the basic mechanism producing it is alike in all cases. The mutation
results from a change occurring at the locus of a gene during a mitotic cycle.
The locus in each sister chromatid may be altered, but the modification need
not be alike in both. Consequently, genic expression may differ in the two
resulting cells and in their progeny. The factor controlling the time of occur-
rence of changes at the locus of a gene in future cell and plant generations—
that is, the de factor in the case described here—may likewise be altered dur-
ing mitotic cycles. As a consequence, sister chromatids may differ with respect
to this controlling unit. Segregation of sister chromatids at anaphase will give
rise to two cells that may differ with regard to this component. In the progeny
of these two cells, this difference will be expressed at the locus of the gene
whose mutation it controls. A twin sector or a twin spot may appear in the
mature tissue. The two components of the sector or spot may show marked
differences in mutation time and/or frequency. (A twin sector may be recog-
nized if the mutation affects a phenotypic change that can be expressed in any
cell of the tissue, such as mutation from w to Wx in the endosperm. A twin
spot will appear if the mutation can be expressed only in a restricted part of
the tissue, such as anthocyanin pigmentation in the endosperm tissue which is
confined to the outermost layer, the aleurone layer. In this latter case, however,
the presence of twin spots reflects the presence of twin sectors.) The two
sectors of a twin may express differences in various ways: by absence of muta-
tion in one and presence in the other, by changes in mutation time in the sector
INSTABILITY AT SELECTED LOCI IN MAIZE 597

showing mutation, or by changes in mutation time in each sector. Ii, during
a single mitosis, a change occurs at the locus of the gerie and also at the locus
of the controlling unit, the result may be twin sectors that exhibit differences
in type of mutation as well as in time of occurrence; that is, they may differ
in presence or absence of mutation, in mutational type in each sector, and in
time of mutation. In the Ds-Ac system, it is known that such coincidences
occur with high frequency.

Recently, BRINK and Niran (1952) and Brink (1953) have presented
evidence, from a study of variegated pericarp in maize, that supports the
generalizations stated here and elsewhere (McCiintockx 1950, 1951) con-
cerning the existence and behavior of controlling units. Recognizing the sig-
nificance of twin spots, they analyzed the progeny of several of them. The
examples selected were characterized by mutation to dominant P in one spot
and change in pattern of variegation in the twin. The time and frequency of
occurrence of mutation was much altered in the latter. Examination of proge-
nies derived from spots of the latter type revealed the presence of a mendel-
izing unit, which BRINK and NILAN call Modulator because its presence is
responsible for the altered pattern of mutation. This unit, in the progeny of
some spots, was inherited independently of P. In the progeny of others, it
showed linkage to P. A striking similarity to Ac is evident in this case. With
respect to 4c it is already known that plants having one Ac factor may produce
sectors in which two Ac factors are present. In some of them, the two 4c units
occupy different positions in the chromosome complement: one at the former
location and one at a new location, or both at new locations. They may also
show independent inheritance in some cases and linked inheritance in others.

Evidence obtained by Peterson (1953) in a study of mutability at a locus
in maize concerned with chlorophyll development (pale green to green) sug-
gests the presence in this case also of a controlling unit, which likewise appears
to undergo transposition.

In a discussion of the possible significance of the evidence obtained from
studies of genic instability (McCiintock 1951), it was suggested that con-
trolling units are present in all nuclei, and that they serve to regulate genic
action during the development of an organism. Their presence, in the cases of
observed genic instability, has been revealed by their displacement. Such dis-
placement, involving units of any one system, might arise in any strain of
maize. Detection of a displaced unit would depend upon available methods. In
the case of Ac, tester stocks have been developed that can reveal its presence.
Dt (Dotted), a controlling unit somewhat similar to Ac in its mode of action,
can be detected through the behavior of the recessive a; when Dt is present,
a, mutates to higher alleles of 4;. NuFFer (1953) examined a number of
strains of maize from various geographic regions in order to determine if Dt
was present in any of them. It was found in one strain from Brazil and in
another strain from Peru. The original Dt, discovered by RHOADES, appeared
in a strain originating in Mexico; and its location was determined as in or
adjacent to the terminal knob of the short arm of chromosome 9. The locus of
598 BARBARA McCLINTOCK

Dt in the strain coming from Brazil is not in the chromosome 9 short arm:
linkage studies suggest that it is located in chromosome 6. The locus of D¢ in
the Peruvian strain has not yet been determined. All three of these Dt factors
behave alike in their mode of producing mutation at a,. Since transposition,
often without change in action, characterizes controlling units, it is not sur-
prising that Dt units are present in unrelated strains of maize, and that their
locations in the chromosome complement differ. As further methods are de-
vised to detect controlling units, it is anticipated that more of them will be
discovered.

SUMMARY

Previous studies of the origin and mode of expression of genic instability
at a number of known loci in maize led to the following conclusions. Extra-
genic units, carried in the chromosomes, are responsible for altering genic
expression. When one such unit is incorporated at the locus of a gene, it may
affect genic action. The altered action is detected as a mutation. Subsequent
changes at the locus, initiated by the extragenic unit, again can result in change
in genic action; consequently, a new mutation may be recognized. The extra-
genic units undergo transposition from one location to another in the chromo-
some complement. It is this mechanism that is responsible for the origin of
instability at the locus of a known gene; insertion of an extragenic unit
adjacent to it initiates the instability. The extragenic units represent systems
in the nucleus that are responsible for controlling the action of genes. They
have specificity in that the mode of control of genic action in any one case is
a reflection of the particular system in operation at the locus of the gene.

One extragenic system controlling genic expression is composed of two
interacting units. It is the so-called Dissociation-Activator (Ds-Ac) system.
Both Ds and Ac undergo transposition. The Ds component, when inserted at
the locus of a gene, is responsible for modification of genic expression. Subse-
quent changes at the locus, initiated by Ds, result in further modification of
genic expression. The Ac component in this two-unit system controls when
the changes at Ds will occur. From the conclusions stated above, it was antici-
pated that the Ds-Ac system could operate at any locus of known genic action.
This is because the Ds unit may be transposed to various locations in the
chromosome complement. To obtain this type of instability at any one locus
of known genic action, it is only necessary to provide adequate means for its
detection. The methods used to obtain and detect this type of instability at the
A, locus in chromosome 3 and at the As locus in chromosome 5 are described.
A detailed analysis of one such case is presented in this report.

LITERATURE CITED
Brink, R. A., 1953 Variegated pericarp studies. Maize Genetics Codperation News
Letter 27: 73-75. ,

Brink, R. A. and R. A. Nitan, 1952 The relation between light variegated and medium
variegated pericarp in maize. Genetics 37: 519-544.
INSTABILITY AT SELECTED LOCI IN MAIZE 599

McCuintock, B., 1950 The origin and behavior of mutable loci in maize. Proc. Nat.
Acad. Sci. 36: 344-355. .
1951 Chromosome organization and genic expression. Cold Spring Harbor Symp.
Quant. Biol. 16: 13-47.
1952 Mutable loci in maize. Carnegie Inst. Wash. Year Book No. 51: 212-219.

Nurrer, M. G., 1953 Additional Df loci. Maize Genetics Codperation News Letter
27: 67-68.

Peterson, P. A., 1953 The mutable pale green locus. Maize Genetics Codperation News
Letter 27: 58-60.