Memorandum to Marcus Rhoades. First Section. Jantlary 1949 Outline of topics discussed. Introduction. Part I. The c-ml and standard Ds loci compared. Page 4 1. 2. 3. 4. 5. 6. 7. 8. 9. The c-ml locus: General considerations. Page 4 Stabilization of the mutated c-ml locus. PRge 13 Evidence for the conclusion that no Ds locus waa present at the standard location in the original c-ml carrying chromosome. Page 15 The standard Da locus: TN0 contrasting types of mutation behavior. Page 18 The relation between the two types of behavior of c-ml and of the standard Ds locus, Page 20 Comparison of rates of mutation of c-ml and of the standard Ds. Page 22 The AC control of mutation during deveiopment. Page 23 The mode of origin of mutable loci. Page 26 Possible mechanism responaible for mutable phenomena. Fage 31 Part II. The c-m2 locus. Page 42 1. Classification and description of types of mutations occurring at the c-m2 locus. Page 42 2. The origin of the c-m2 locus. Page 52 'x .- . Illustrations of the phenotypic changes that are associated with c-m2 mutations. Appendix - Extra Tables. Not explained in text. Memorandum to abrous. Jmmary, 1949 Introduotion The orno locrus (hereafter written o-ml for speed in typing) appears to have arisen as the oonssqusnoa of a ohange in loostion of Df3: r8mOVe1 from its fonner position and Insertion into for close to) a norm::1 0 loous. In this new position, it behavea just RPI it did in its former position. The o to 0 mutations appear to be assouistad with its behavior in this new position, aa I ehall try to point out. In this new looation, more has been learned about IIs behavior than oould have been learned from Da in its "standard" location. The position in crhrozaosome 9 where De was ffrst discovered will be Galled the Hstandardw looation. A position of' reference is needed beclause Ds does change its looation. In Its standard position, it ia known that Da changes its mutation pattern from one general type of behavior to another, i.e., the Qarly" loss type and the Yew-late" loss type ot behavfor (for details of these two general types, see below). Previously, the behavior of the Ds loous in the mfew-late" state had been difficult to analyze and interpret. In its new position at the C locus, various types of behavior of Ds are subject to analysis, as will be indioated. Becrondly, ohromoaome breakage of' a type not previously recognlved may be rseponslble for the mutation phenoaena associated with mutable lo&.. The type or breakage and aubsequent fusion of' broken enda that will ocmur in any one "state" of a mutable 100us my be controlled by the particular organization of' the inserted loous reaponslble for thi 8 breakage phenomenon, at the time of Its Insertion into a new looation. Thfs breakage prooess need not lead to the simple criss-oross type of fusions as previously suspeotad, nor to the formation of U-shaped dioentrio ohromatids already known to arise from mutations occurring In one "state" of the Ds loous. rt Is probably something quite different. This breakage mouhanism will be outlinsd later. This meohanfsm should sometimes alter the organization of the inserted locus so that new types of organizations would be bormed. Thess altered organizations will give rise to nsw "states" of the mutable locus, that is, dlffsrent rates of producrtion of visfble mutations, of stabilization of loci, rend of the typas of' fusions that will ocour following breakage. The state of a nutable loous is something more than a quantitative change at a loous. Starting with any one organization that gives a particular series of breakage-fusions events, the types 0r ohanga to new organizations giving an altered series of evsnts may follow rather definite sequsnoes. Thirdly, stabilization of a mutating loaus may follow as a consequenoe of lossiduring a mitotia ayole,of the Inserted locus responsible for the mutation phenomena or of Its aotlve component. #ith some organization8 of the inserted locus, each mutation of the mutating locus,with some exoeptions,will result in stabilization of the mutating locus. '#ith other organizations, some of the mutstions will not reault in elimination of the acltivs loous. In these oases, continned mutations will ooour. This mutating locus may be a dominant, an intermediate allele or a recessive but mutwtlons will oontlnue. Fourthly, I suapeet that many loei, such as the C loous and the AL and A2 looi are crompound loe,i involving ganic aotion assocla~ted with SUOO~SS~VS or related steps in the oonverslon of substrate to produot. The type of mutations ooourring followin$ insertion of a "Da" loaus into suoh looi will depend upon (1) the posltlon of insertion, (2) the organization of the inserted louus at the time of its insertion and (3) the changes in organization of the Ds locus that will oa~lur subsequently. If the aoaumulating data has been lntarpreted correctly and continues to give further supporting evidenoe, it will be neaessary to ahange the symbol "Dew to something more expressive of its basic aotion. The observed dissooiation phenomenon (dioentric ohromatid formation) is only one of the ooneequenoes of mutationa of Ds that ooour. The various typss of Ds mutation behavior are probably expressions of one basic type 0r event. This. is the important event underlying all expressions of mutability of mutable loci. The newer evidenae has made it possible to consider this event more specifl- oally even though there is muoh that is purely speculative. Neverthe- less, the expanded knowledge allows for better dlreated experimen- tation and observation. Fifthly, there are probably a number of unstable dominant looi in my material. These wfll eventually throw unstable reoesstves. The wx-ml loous arose from an unstable dominant locus, designated Vx-m. This unstable dominant loous has now been isoltlted. It will throw more unstable reaessives. One suoh recessive, a new*-m, is growing in the greenhouse bgrt many more should be detested following suitable crosses now that the unstable dominant that throws unstable recessives has been isolated. The instability of dominents and the unstable reoessives they throw are probably ~11 expressions of the one kind of event, i.e., ohromosome breakage. Mutable expressions go in a aycle from unstable dominant to unstable reoessive to unstable dominant, eta., or to various quantitative levels between full dominant and full rsoessive. (Full dominant is a relative concept. Super dominants can be isolated.) Stabilization may occur at any quantitatively eipressed level, however. Part I. The c-ml and standard Ds loci oompared. The o-ml loous; general aonsiderations For the details, the c-ml 0888 may be oonsidsred first. A brier review of the behavior of this loous will give the essential raots neoessary to an understanding of the discussion. The a-ml locus L,,l 0 37&,l,'"' arose in the ~lros5 of a yg a sh Bz wx ds ae female plant b a mele J I j j J i* 0 .i , _ .,_< ..-. . -2' plant that was Yg C Sh Bz wx Ds / Yg C Sh Bz wx Ds, Aa ac.: Ten crosses of this male parent were made to e sh ds aa famnle plants. The ratios were: 2077 C, non-variegated (ac ac ac) : 1199 C to c variegated, early losses (AC ao ac) : 317 I: to 0 then classified as having the few-late type of Ds action (mainly Ao ac ac) : 1 c to C variegated kernel. This last kernel W~'IS obviously not the regular C to c variegation expected as the oonsequenoe of Ds aotion. It showed the opposite type of variegation, from o to C. A plant was grown from this aberrant kernel in the summer of 1947 (plant 4204). The plant was variegated. It showad streaks of yg. This variegation was not unexpeotad beaause it was known that both chromosomes 9 in the male parent oarried Ds. If Ac were present, just such yg streaks should .appear. Xt was only necessary to determine that AC was also present in this plant; the later tests proved that plant 4204 was AC ao. The aonstitution of this plant was then believed to be Yg 0 sh wx ds / Yg c-ml Sh wx Ds, Aa ao. The data and analysis given below shows that the looation of Ds is not as given. Instead, it is probably at, within or oloee to the C locus. In this plant (4204) no ,evidenoe has been obtained for a Ds in the "standard" Plant 4204 was orossed to: (1) yg o sh wx ds ao (2) C sh ba wx ds ae (3) C 3h wx De / C Sh wx Ds, do ac (4) pyd c Sh tix da / yg c Sh WI de, ao ac. From aross (11, It was clear that the male parent possessed two ahromosomes $,ons with Q o-ml 812 wx and one with yg c ah wx. (For kernel taunts, sac Table I f Appendix.) From cross (2) it was olear that a D s loous was likewise present in the o-ml 9h wx ohromosome. . of the 8h kernels (the Ao oontaining kernels) were C B-C bz variegated. (See Table 2 , Appan- dix.) There were few suuh kernels in the sh class. (This latter observation should have made me suepicious about the looation of De but 3 failed to register a reaction. At the time, I did not know that Ds oould ohange its lactation. Realization of this did not enter my oonsoiousness until late this spring, following the harvest of .1 ,c : r*: I the greenhouse crop. From oross (3) it was likewise olear that a Dr loous was present in the o-ml. Sh wx ohromosome. Some ot the kernels were C to c variegated following Da mutations in the G Da ohromosome contributed by the female.parent; in some of these kernels, the o areas showed o ,I,. * I. q ')'i f I, 'A to C m&atfoti$ due to o to C mutations ooourring at the o-ml locus in the ohromosomes 0 oontributed by the male parent (4204). These latter kernels most always showed homozygous deficient tissues of the type oocurringS when three Ds loci are present, one in each chromosome 9. The oombined evidenoe indioated quite definitely that a Ds locus was present in the o-ml oarrying chromosome of plant 4204. From oross (1) it was obvious that the expresoed mutation rates of o-ml are not alike in all the kernels. The majority of the kernels show relatively few o to C mutations. A small number of Sh kernels, however, had very many o to C mutations. I assumed from this that the kernels showing only a few c to C mutations were either (1) those having a o-ml Sh wx Da constitution and that the Ds mutations were eliminating the segment of the abort arm oarrylnR the a-ml locus from many sector8 of these kernels so that the o to C mutations oould not ooour In these sectors; or (2) that the state of the o-ml locus was low, i.e., few o to C mutations \bWttr ooourria@; even in those kernels with o-ml and no Ds (those derived from cross-over ohroma- tias). Again, it was assumed that those kernels with many o to C mutations probably possessed no Ds beclause Ds oould have been removed from the chromosome as a oonsequenoe of arossing over between Sh and the assumedhDs loous in the parent plant (4204). Saoause both uhromosomea 9 in thin&rent carried wx, there was no immediate genetio determination of such a cross-over. The sh kernels showing o to C variegation obviously oarried a cross-over ohromatid. Barrin double crossovers, which are rare, the e-ml chromosome in these kernels should have no Ds locus. These kernels, however, did not show high rates of c to C mutation. That a changed looation of Da had ooourred when o-ml arose oould not be clearly FeveRled from an analysis of the crosses made during the summar of 1947. That something was wrong soon beoame obvioua. The nature of the disorep- enofes ~111 be indicated. -I- The seleotion of kernels for tests In the greenhouse last winter,, oentored on eetablishing the, sontrol of c-ml mutation phenomena by Ao. From this crop, it wc98 determined that Ao OontrOll8d the c-ml mutations; without Ao, no mutations ocourred. It alao showed that the time in development when mutations would oocur depended upon the dosage of AC. In this respect, it was quite similar to Ds in response. In the progeny derived from crosses of those plants that were drown from the highly o to C variegated kernels, this greenhouse orop also showed the exact oorrespondsnoc in pattetrns of mutation to those already known for one typa of Ds. You will remember aeeing, this aimllarlty in comparing kernels of the oonstitutlon C da / C ds / I ?7s, Aa ac ac (and AC Ao ac) with those kernels having o ds / c ds / c-ml, AC ac 80 (and ho AC a~). The two sets of kernels ware indistingulehable, respectively, in ooior pattern of variegation. The mmmer planting (1948) included a selected number of 'sx kernels showing c) to C variegation from oross (4) above. The Vg df co< f+fv plants arising from these kernels were till variegated for pyd. This ? was not antioipated for 1 had expected a few of these plants to have a o-ml chromosome with no Ds beaause of crossing-over between o-ml and wx in the male parent. It did not bother me too much as the sample was small. In order to be sum to eliminate Ds from a c-ml oarrying chrol[Dosome, some of these plants were crosses (1) to o sh wx ds ao, [2) to 0 sh bz WI da ac, and (3) to plantls having the rearranged chromosome 9 with o sh %K and a female-transmissible deficient chromosome 9 oarrylng 3h and wx but deficient for the c locus. There were other reasons for these orosscs but mainly it was to observe the efteots of c-ml mutations when Ds was certainly removed f mm ths ohromosom. No Ds (standard location) should be present in most of the cr-ml Wx ohromatids. Only double croBsovera ~0uld give the 00nstitutPon c-ml lrJx Ds and these are rare. The rsaults from cross (I) were Strikingly different from 0xpsatat~on. E'early all the -$r kernels thet showed o to C vnrie@- tion were also heavily W Lo wx varfsgst.ed; this :;$x to wx variegation resembled ehe usual Ds typs'variegation. In these kernels, the rate of c to C variegation was low, except for Sectors in Sorie of the kernels, A few of the '#x kernels, however, Were heavily VariegEited for c to C. These kernels showed only a faw wx areas and they uauc.lly were quite small. A tabulation of the olasses of kernels resulting from oross (1) mty interest you for they tell the full story much better than words (table 1). Because the male parent was heterozygous for Ao, only half of the kern818 having a o-ml locus could show c to C variegation. Also, the kernels that received the stable o locus (from the perentel c Sh Ix chromosome) will not show c tc G variaga- tion with or without AC. You will note that only 14 of the $90 o tix kernels (no visible c to C mutations) Showed 'iix -wx variega- tion whereas all of the 140 c to C variegated kernels that were `#x likewise showed 3x to wx variegation. The prasenae of deteotible sh sector8 in some of these heavfly Wx-wx variegated kernels made it clear that 1088 of a segment of the short arm of ohrOmO8Om8 9 was responsible, or aften associated -Jiith, the appearance of the wx sectors. Some kind of Da-like action %a8 occurring in these kernels -Mith a c-ml :Vx chromoa0ma. This Da action waS not associated with a standard Ds loous in the c Sh ~Yx chromosome of the male parent. If' R 11s locus had been present in this ohromosome, many of the o 'Sx kernels shouL6 have been #x - WL varfEqted but they were not. Secondly, all the ax kernels showing o to C variegation likewise 43504 x 44343-3 46 90 141 27 3 1 1 4346-18 x w 40 80 146 10 1 cl" 0 4349-2 x 34 03 X48 22 0 0 1 Totals 120 253 4s 67 4* 1 2 c 2 .C c7C c CrbSS c wx c .I:X s 9x c: wx wx 3x 3 wx i~X+wX 4346-16 x 4434h5 27 57 73 u 2 ? Cl 2 ----m-e 4340-31 x rl * 29 3 1 3 1 1 * -26 x iv w 37 99 148 28 0 2* 2 f' -34 x '1 ?' 34 100 134 15 4+ 0 3 Totals 1.37 367 75 1ov 3 11 - - Bhbwed 3~ tb WI: variegfitibn. These represent the kernela receiving a crossover chromatid that brought o-ml and #x onto the same chromosome. Since all of these ohromosomes show this Ds-like action, a Da-like locus must have been present in the o-ml 3h wx chromosome of the male parent. Its location must be in, close to or to the left of the c-ml locus. Thirdly, of the 16 C kernels arising from germinal mutation of a-ml, only one showed any C to c variegation and this was Q peaulhar type. (It will be polinted out later that just such new variegations are to be expeoted. They represent changed looation or changed organizations of' the inserted locus responsible tar mutable phenomena.) Unless all such C kernels heve an ac ac ac constitution, a Ds type variegation should be present in some of' them, on the assumption that a Ds is in some ponition other then the c-ml lacus. Fourthly, extimination of the C areas in both the c to C, ViX - wx kernels and the c to C wx kernels were likewise instructive. In these kernels, with aotne expectted sxceptions, the C areas were nut v8riegated for C to c. In the c to C, Wx - wx kernels, the C areas were Yx, not ';r'x -wx variegated as would be expected if Ds were present to the left of the C locus. Obviously, then, when k mutation to C occurred, the N-like action In this chro?nosome ceased (with some sxoeptions, to be consSdered later). The oombiaad observations made me suspsot that the Ds action IRS aseociated with the presenue of Es in the o-ml locus and that the Da locus was usually lost from the chromosome when a mutation to C occurred. The following conclusions were formulated: that (1) the o-ml loaus tlrom f'mm 8 c Sh wx IIs chromosome fcllowing removal of the Da locus from its standard locaticn and insertion into the normal C locus; (2) when a mutation or o-ml to C occurs, Da disappears; (3) the insertion of Ds into the C locus brought about an inhibition in the expression of the genea at the C locus; consequently the modified C loous resembles o in phenotypic expression. (4 Yhen this 953 locus is removed from the ohromosome, the normal qenio action of the C loous cm reappear. (5) Two contrasting types of Ds action a*,,,., MM be reoognized when Ds is in this new position: (a) one giving the typloal. diosntric ohromatid through breakage snd fueion at this new location and (b) one giving tha c to C mutations that u+ally result in no visible alteration of the chromosome. For an understanding of the events underlying mutability phenomsno, conclusions (4) and (5) above are most important. They bring up the question: Does Ds in its standard location show these same two contrastine types of behavior? A positive answer to this question is indicatert. Two dissfmilar types of behavicr of Ds in its standrrrd location have long been recognized. The extreme of one type gives many dicentric chramatids RS the visible conaecuance of Ds mutations. The other extreme gives the *few-late".pattern of chromo- some segment losses, i. 0., few dicentric-forming mutations occur. The relationship between these two contrasting types of bb mutation behavior of' Ds in its standrsrd location will be indiaated after 8ome of the supporting evidence for the conclusions (1) to (4) are given. Cross (2) -- C sh bz wx ds au female by o-ml Sh Rz wx / c Ph Bz !ivx, AC ac male -- will be mentioned first. On the ears resulting from cross (2), a namber of kernels were vsriegated for Be -bz. In those that had the YX locus, the C b-r Areaa most always were variegated for 3x -wx, i.e., $3~ was lost before 'pix. This would b8 expected if dloentric chromntid formntion rasul.ted from -110 Ds mutations at the C looue. (If Ds wem in its standard position, the bz areas would but wx.) The results from one such uross are a8 follows: C sh bz wx da ac x (Plant 4684-7) o-ml 3h Bz wx / a Sh Bz Wx, AC ao [Plant 4434D-5) c Bz $i'x non-var. C Bz - C bz c Bz" C Bz -C bz #x .Am wx wx not obviously VEtI?. 124 12 89 41 The kernel@ resulting from crosses of type (3) were likewise useful because the classes sre those expected on the basis of the comlusione given abave. 088 such cl-089 gave: ac! acz x c-ml Sh wx / o Sh Wx, AC ac (Plsnt 4642-6) ( Plant 4434D95) S C Sh Wx non-varhegated. 1 C Sh wx non-variepted 11 o to C lightly variegated; Wx-#i;r variegated; homozygous deficient tissue8 (from dioentrio crhrormtids following Da mutations). 19 c to C lightly variegated; wx. Homozygous defiolent tissues present (front dioentrio ohromatid formation resulting from Da mutations). 5 c) to C, heavil variegated; wx. tiid 370 obvious hoaozygoua defioient Yuoh tissuea would not be detected if dicentria ahroarat2d formation oarrurred late.) 63 o sh wx; no homozygous defioient tissue. wwwwwww"w-wwwwwwww-wwwwwwww--wwwwwwww~www"wwwww"--wwww----w-----w--w * Unless early losses of 32 oocur, it is often not possible to detect C Bz-C bz variegation. Bz substame from the surrounding Bz Areas diffuses into the bz areas giving them the Bz phenotype. A few of the kernels in this *may belong in the C Bz-C bz class. -120 43 ct to C, ve&eg~&tjd mostly lightly variegated; Wx; no homozggous de- fioient tisaua 240 c Sh Wx; non-variegated; no homozygous deficient tissue. The plant arising from the original kernel showing c to C varla- gation (plant 4204) ia now &$%.-to have had the constitution gg c ah Bz wx da / Yg o-ml Ds Sh Bz wx ds, AC ac. Before th8 trans- position af Ds to the C locus was re:ilized, the aelection of sh kernels that showed c + C variegation in the aross of this plant to CI ah wx da ao plants, was expected to give kernels having a e-ml loous but no Ds lccus. Suoh cm-l carrying chromosomes represent orosa-over ohromatids and if Dig had been present at its stsndard location in the a-ml Sh wx chromosome of plant 4204, they should heve no Ds loaus. Plants were grown from some of these o + C eh kernels. Croasss of' these plants were made to (1) C sh bz wx da ao plants and to (2) plante having a deficient chromosome 9 (the d8fiCienCy included the C locus) Sh wx and a resrranged chromosome 9 with c sh Wx [es wall as to c sh de ac plants) Following cross (11, C Bz - C ba variegated kernels appeared on the ear. Following Ctross (2), homozy- gous deficient tisrsus appeared in the o --i, C Sh wx kernels. Obvicuely, the Ds aotion had not been removed from the c-ml ohrOmOSOm8 by crosaina-over between o-ml and Sh. Beoause the transposed position of Da in the o-ml ohromosome was not suspeoted this past summer, the most favorable types of orosses to illustrate a new position of Ds were not spst8matioally made. Nevertheless, some additional ClrosseB, made for other purposes, have been helpful beaause they are eonsiatent with the above interpretation, An illustration of one of auoh oade is given In the aross of a c-ml sh `-4X / c-ml sh wx, Au ao male plant to an ac ac female plant that had -x3- a terminal defialent ohromosome 9 with C Sh wx. This croaa gave: 90 I @lx kernels 3 c A% I&ml 34 C wx normal; no homozygous deficient tissue 20 C WI, with large amounts of homozygous deficient tissue. %The horaogygous deficient tissue in the 20 C wx kernels srlaes as the ~-.l~v. 37 .d consequence of dioentsic'formntfoa assocfatec! with !M mutetfnns at the o-ml locus. There are several points that require further oonsfderatlon. (1) Why does a stable C loaus arise when a c to C nutation occurs9 '&hat happens to the Ds locus durin@ such ~tn event? (21 dhet is the relation between the two visibly different types of events thnt cm occur at the c-1111 loaua; (a) diaentric chromRtid Formation snd (b) the c to C nutntione that do not alter the chromosome mornholoqgQ (7) Does Ds in its standard looation ahow these game two mein types of behavior? (4) #hat i t n errelations exist between these two contrasting behaviors of Da? (5) iiiihat is the evidence Car the oon- elusions that no II3a locus is present at t3e standard locatIon in the original. c-ml 9h wx chromosome? _tabili?ation of the C locus fcllowi~ a o to C mutation of o-ml will be uonsidered first. Stabilization of the mutated o-ml loous. On examining the variegated kernels arising from the ffrst crosses (c sh wx ds aa x 4204), it was noted that the mefority of the C areas srising irOle mutations of' c-ml to C were not varieRa+,ed for C to c, as could be expected. This expectation :vas based on the assump- tion thut a Us locus was present in its standnrd loaation in the c-ml carrying c3hrozPoeome of the naale parent. ff this had been true, TO to 80 paroent of the e to C variegated kernels should hove had this Ds locus. Consequently, this absence of C to c varlegotina in the S preae waB unexpected. I wondered at the time whetherBin this case, a c to C mutation was associated with a change in AC activity co that no further AC controll& mutations would occur in the sector. ?he evidenae for suah a conclusion was not substantiated in other crosses. Tt was known that c to C mutations in one chromosome could occur following i3a mutations Ln the hamol~gous chrolatosoaee. This wa:; revealed in kernela resulting Prom the orom of C wx DE? / C wx Ds ac ec female x 4204. It was obvious hem thet dicentric- chrom&tid-forming Da mut;ztLons aould occur in the fermtle chl*omsomeEl: 9 early in dev~lnpmnt and that these could be fo2lomd in Zzlter eel.1 generatfons by c to C muttl t ions clit the o-n1 locus in the chronosorne 9 contributed by the lU~.10. In the croaseh~ or Q sh XX ds ac x 4204, a nuaber of geruinal mutations of c to C appeared on the ears. )rTone of these C kernels showed an.y C to c variegation of the Ds type. If a sttindard Ds locus were prezerit in the C chromosome of these kernels, many of these C kernel8 should have shown the Ds activity by giving C to c variega- tion, for ha31 of thalm 83ould have an AC HC ac constitution. 29 c, non-variegated kernels arising from these crosses, end from some similar irnt>s mado in the greenhouse last winter, were removed from the ears and planta grown from them in the field this past summer to determine (1) if an Ac locus were present, (2) If Ds were present and (3) if the G locrua arising from a o-ml tmtcation was stable in later ge:zcrc?;x kernels (ge,rmSnal mutations) are C ho c variegated with the o areas being wx:typioal DY type. There oan be no question, then, that Ds mutations oari occur in a o-ml carrying chromosome following a c to C mutation if Ds is in its standard locntion. Gith Ds at the C locus (c-ml) a mutation of c to C usually eliminates the I)s locue or its action and the newly srieine C producing locus is stabilized against further mutation phenomena. Mutations OP c-ml to C do not uaucllly result in mor?hologiaally altered chromosomes 9. One obtains a normal ohromosome 9 with a nora? C loaus action.* Nevertheless the Ds loous at c-ml shows two main contrasting types of behavior (1) the production of diaentrio chromatids and (2) the mutations giving rise to a C phenotype without alteration of the chromosome and usually with elimination of further Ds activity. ---------"cI--l-"-----.--------------------------"------------------- * The use of *normal C loouat* must be qualified. "Jlutations at the o-ml loous have given many apparently normal reaoting C loci. There are a number of sub-levels and some super-levels as well as a few resembling the pink mutations (very rare) of; c-m2 arising, as the consequence of mutation8 of c-ml. These represent both quantitatively different reaction oapaaftise a8 well AS qualitatively different reaation oagaoities. Aooordlng to my grasent interpretation, suuh quentitati ve levels of expression of C aotivity are to be expected as well as very much less frequent ahclngee that will incorporate adja- cent looi (the pink reaatlon). It would be more profitable to post- fwciabl poue this disoussion until after & mechonia,,responsible for mutable phenomena haa been discussed. -18. The standard Ds looua: Two contrasting types of behavior As the last senI;t..noo above indicates, it is concluded that 33s activity is responsible no S only for the dioentric chromatid ?orma- tion at the o-ml, locus but also for the c to C mutations of this locus. This conclusion LY based upon the striking resemblance between two contrasting tgas of Ds activit,y when Ds is In its standard J.ooRtion and the two contrasting types of events occurring at the c-ml loous. The nature of the evidence for this general conclusion will be considered. That the 119 locus in Ita standard pcsition gives ~H'G Tain type8 of recognizable behavior,has been known for a Zong time. In sn AC aa a0 constitution, Do looi have been isolated thet give meny dicentric chromatids as the conseguenos of mutations of l?s. Kernels with appropriate genie oonstitutfone reveal this clearly. Sny of these mutations occur ear3y in development. Such Ds loci have been iiesignatsd Vs-early". A contrasting type of Ds behavior results in only a few losses of segments of the short arm of chromosome 0 and they usuilly OCJCUF late. Such Ds looi have been designated '3s-fe+ 1Hf;tt" * liz the kerneiu 4th a. few-late I)s locus, many aberrant types of chromosome breakage ocour; ,y:, b \' the brecnklsg8fior%ginate at positions other than the P>s Iccus. The significance of this will be discusser! 1Eiter. uy, LO show that the fey;-late type of 9s behavior F3 associated with ti particular state (orgenizntion) of 2he D3 loous ani! the*: it is not due to inodff'iers loctated e3ssv:here in ti-,e chromoaone complement, the fallo-9ing ~ros3es were n;uds. Plants were obtained that ha:! the Gcdristituiisn: I ra Ez 1-d Es-early / 1 3h Rz wx Us-fsirv lute, kc ac. These plants 'vvere orossed to female plants with C sh bz wx ds ac. -19- The types of variegated karnels,with respect to &'x and wx constitu- tlons.wer8 tabulated. The results of these oroases are given in ., . table 3. Crossing ov8r between W 8Dd Ds is low, only 1 to 2 or 3%. The 10 I - C bz, wx kernels showing early losses probably 8ro98 from urossing over between wx and Ds. The 34 I - C bz, Wx - wx k@rn8lS, with the few-late loss pattern of Ds,r8prasent oross-overa between %r end Ds together with newly arfsing few-late Ds loci derived from early Ds looi. The few-lfite type of Ds loous ariasa rAther L ~6 he 0breyoed as yet) frequently 819 a mutation of the early Ds COCCUS but only rnrely4doee a Ds-early arise from 8 DS-reW-18t8 locus. The evidenae from table 3 strongly supports the Bonolusion that the early and the few-late types 0r variegation refleut two ditferent Qtates" of the Ds locus. Other observations make this COnd.USiOn even stronger. One pertinent observation concerns the apgearanoe of sectors showing the r8W-hit8 type of Ds behavior in kernels that otherwise show the early type of Ds behavior. If the f8w-late type of Ds behavior arises rather frequently from the early type of Ds behevior, then mob sectors should be seen. The most instructivs 8vid8no8 for the prssenm of two suuh contrasting types of Ds behavior can be obtained from uomparisons of these two types of behavior of Ds at the standard location with the smie two types of behavior of Ds When at the a-ml locus. I shall attempt now to point out the sinailarlties. 4363-B x 4679-2 46 4 4684-10 x * 1 73 0 51. 4362-23 x 4679-3 126 93 15 117 6 114 4363-35 x 4679-5 77 72 1 63 2 87 4 4364-1 x 4679-5 120 92 13 120 ) 2 79 5 Totals 361 -20. The, relation between th8 two types of behavior of o-ml and the two 1 types of behavior of DS at its standard looatlon. In the or088 of a female c1 sh wx da 80 plant by a male e-ml 3h wx o Sh '#I, AC aa plant, the Ix kernels showing a to C variegation were of two distinat types. These were Oonsidered earlier but wfll be further uonsidered here. On8 type was most frequent. It @IV@3 Very f8W Viaibl8 o to C mutations but many mutations leading to dioentric pi&:,. 5:. &(;,l ohromosomes. There were a few #cJr kernels, hOWeVer, wfth V8I7 many c to C rautations. In the88 kernels, few dioentrio forming mutationa oaaurred r0r very few wx spots were preerent. In general the pattern of the88 WI Spots resemble in distribution and si.28 the patterns given by the rteW-ht8 5)s in its Standard looation. In both camet, losaea of Segments of the short arm of ohromoeome 9 occurred infrequently I: i : .;.l.J 9' ': ;' and,late in d8VelOpm8nt. ', Mth respect to typee and frcquenulee 0r loss of segment8 of the short 8x7~ of ohromosome 9, both ban b8 alassi- fied 8S NDs-f8w-loteR States. Again, a Similarity is found between those c-ml blrx kernels Showing low rates of a to C variegation but high rates of dioentric formation with th8 pattern produced by Da-early. Furthermore, in some of these kernelS, distinct Seaters appear with high rates of o to C mutatione. These sectors are Wx and have few if any wx spots In them. To review: The k833818 with high ratasl of 0 to C mutations give low rates of cbegmental losses in the Short arm of chromosome 9. The pattern of these losses r888mble the pattern8 of losses obtained from the f'ew-late DS in its standard location. The patternS of losses in the kernels with high rate?.of 1oSsrr of Segments of the short arm of the o-ml chromosome~resemble the patterns of lospleg obtained from the standard Da-early loous. In the forms? kernels, the seotors with high rates of' o to C mutations resemble in frequenay and ooaurrenoeYtha motors in the latter kernels that show low rates of segmental loss (Ds-few-late sectors). Again, o-ml loci giving high rates of aeignental loss throw a-ml loci giving the low rates of se@nental loss. Mmy of these, in turn, have a high rate of c to C mutation. It is concluded, then, that the state of De-early in the standard looatlon and the stete of o-ml giving high segmental losses ropresen0_ the same mutational state of Ds whereas the Ds-few-late In the standard position and the o-ml with high rate8 of Q to C mutations represent the sme mutationcsl state of Ds. From this, it Is conclutisd (388 figure 1) that the Ds-few-late, in its standard looetion, is undergoing many mutations ah are hiddsn because n+o visible phenotypio Cfects 70110~ from such mutatPons. In arder to see this type of mutation behavior of Ds, the Ds loous must be inserted into or olose to a loous that will give a phenotypicallg deteotible etfeot, following such a muttstion. This oucurs when Ds is In the c-ml locus but does not occur when inserted in Its standard looation. One more resemblance between o-ml and DEJ (standard locationjshould be mentioned. Xhen kernels with high rates of o to C mutations of c-ml are seleoted and plants grown from them, the high rate of mutation of c to C is retained in later generations. Again, In this respect, the high rate of o to C mutation pattern of c-ml resembles the few-late type of Da behavior (stzndard position) because in both cases, mutations back to the dicsntric chromatid forming type of Da behavior occur rarely. The "few-late" type of Da behavior represents a relatively stable stats of a Ds loous. Figure 1 Comparison of Ds activity at the standard looation and at o-ml Standard Looation Ds behavior o-ml Ds behavior Ds -early Same erant 1) low rata of hidden mutations i St )l) lowratesora~c mutation a) high rate of visible mutations (segmental losses) \ ., 2) high retes of segmental losses Ds-few-late Same event 1) high rate of hidden mutations n 1) high rate of o -. C mutationtj: 2) low rate of segmental losess w 2) low rate of semental \ lossas -220 Comparisons of rates of mutations of c-ml and standard Ds. St follows from these observations and conclusions that one is in a position to interpret the mutation rates in the kernels with C de / C ds / 1 Ds-early, AC ac ac (or AC AC ac, AC AC AC) with thoss having a / c / c-ml, Ac ac ac (or At! AC ac, AC AC Aa). Because the color pattern produced by losses of I' that occur as a consequence of dicentric forming Ds-eerly mutations are the same as the color pattern produced by I: to C mutations, direct comparisons of mutation rates and patterns can be node. Yith these particular constitutions, the kernels showing high rates of c to C mutations and thoae having a Da-early (standard location) may be indistinguishable in phenotype. Since c-ml loci giving high rates of a to C mutation show reletivelg few segmental losses and since it has been concluded that thel'loci have a few-late type Ds locus, it is further concluded that the rates of Ds mutation in both the early Da state and the ft:w-late Ds state are really the same. AC controls the time and frequency of some event that occurs at a Da locus whether or not the consequence of thI.s event lea&to a detectable change in phenotype. The mutations resulting in no C&actable change in phenotype may be called "hidden" mutations. The presents of hidden mutations will remain a factor aausinq difficulty in the study of all mutable loci. Their presence Is suspected in all the Ac controlled mutable loci. They are a factor that will interfare with any attempt to evalutlte mutation rates of any mutable locus. ,Means of detectinq and cltiasifyin% the types of hidden mutations Is absolutely necessary if an understandin& of the factors involved in bringing about visible mutF-tions Is to be realized. The Ao control of mutation during development The sirail:-$rity in rr?tes of mutations at the a-ml locus, when the tz -$ C rates are high, to segmental losses rates at the standard Ds loous when these are high, has presented an opportunity to determine something about the nature of AC action in particular cells during development. jYhy do mutations occur in some of the cells during development and not in others? If two &z-controlled mutable 1001 are present in a cell, will Q mutation occur at each locus at the some time in the same cell? I3ow could this be determiaed~ Crosses of plants having e-ml wx ds in chromosome 9 (high rate of c 7 CT mutations) bg plants having c (stable) *!r Ds-early (stenderd loaation) in chromosome 9 should make this relationship ene'pznble. If, in the resulting kernels, the c-ml locus in a c-ml wx ds chrorna- -some mutates to C, a C urea will arise, the size Rnd position de?endinq upon ilrhen and where the u to C mutation occurred, that is, in what cell this occurred during development. If', in this same cell, a mutation ocourred at the Da (standard) locus in the c .7x 3s chromosome, the 8x locus will be deleted and the oells arising following this mutation will be wx. By examining the starch constitution of the cells underlying the C areas in the kernels resulting from suah a cross, it should be possible to determine whether or not particular conditions,- controlled by Ao, arise in oertain ~~11s and whether these aonditions affeut all Ac controlled mutable loci elike - that is, produue a mutation in all Ao-oontrolled mutable looi. If so, the C areas should have underlying wx staroh constitutions and an exact correspondence in sectors should appear; or some relationship should be visible between t&c :::,a typos of mutations that will tell whether such a coincidence :t mutations occurred, (that is, whether one-strand or two atrends are involved in the muttltion process in a particular cell). Fortunately, in some of these crosses, the frequency of coincident mutations of o-ml to C in a c-ml oarrying chromosome and of Ds muta- tions in the c Vx Ds chromosom8 is remarkably high. A large propor- tion of the C areas have underlying wx starch. This is easily seen in the Ao ac 80 oonstitutione but it is better observed in AC AC ac constitutions when a relatively early mutation occurs. In this latter case, the coincidence mutations are very frequent. There are a number of exceptions but suoh exceptions should be anticipated. This is beoause several types of hidden mutations occur end also because one of these changes the state of the locus. These newly arising states will give cthanged rates of visible mutations. Also, the mutation grooess itself frequently gives rise to exoeptionel conse- quences some of which are visible others of whiuh are hidden, as will be shown later. The high rates of coincident mutations, mentioned above, might lead one to conclude that the Au control of mutability is related to some particular &-controlled condition that arises in particular ~11s during development. llihen this condition arises, it will affect a mutation at all AC controlled mutable 1001 in this cell whether or not the mutation occurring at any one pertfoular loous will result in a changed phenotypic expression. The Ac locus Itself may mutate at the same time also. All of this may well be true but the diffiaulties in proving it are great. The crosses described above certainly suggest this conolusion regarding AC control of mutable loci. Nevertheless, other oroa885, originally thought to be similar, do not give such a striking coincident rate of visible mutations of these two Ao controlled mutable loci even though many ooinoident mutations did occur in the -2% development of the kernels on thelre ears. I am inclined to believe, from observing patterns end rates of' visible mutations, that there oan be a limited range of statea pf Ds loci, from ones giving few hidden mutations and many viaible`mutationa to othera giving many hidden mutations and few visible mutationa. Combinations of c-ml ds / o Wx D8 that are not reolprooally balanced In this respect would show a number of non-coincident visible mutations. This would not mean, however, that aoinoident mutations did not oocur. Rather, it was not possible to see them. A possible method is available for examining whether or not a range exists in the rates at which these two classes of mutation will occur. It involves various isolates of the c-ml locue. Here, two of the types of' mutations of a single Ds locus c1an be distinguished in c ds/ c ds / a-ml Wx oonstituticns -- the o to C mutations and the segmental loss mutations. Also, some of the mutations leading to changes in state niap be recognized. Selection of o-ml loci giving intermediate frequencies between high c to C and low se*gmental losses vs. low o to C mutation and high segmental loss can be made. As yet, X do not have available for examination a seleotion of such c-ml looi with the proper constitutions. However, a comparison of the behavior of the c-ml locus derived from single plants In the oroa9es of these planta to both (1) c sh wx ds ac and (2) C sh bz ds ac certainly presents su&eative evidence for the preeence of' such intermediate states of Ds. The o to C rate of mutation following cro8s (1) hae been compared with the C Bz - C ba variegation rate following cwos9 (2). Medium rates of o to C variegation appearing in the kernels followinhz; cross (1) are correlated with medium rates of loss mutations in the kernels arising from cross (2). If this proves to b8 ~8Il8lXilly tl3l8, it will explain why the oroeses combining dSZfsrsnt o-ml wx 4s ohrOmOsOm8e with 6 Wx Da ohromosonms showed differcent colnoident rates of visible mutations or the two 1001. If the observed oolnoidsnt ratee are relatively low, the rates of hidden and visible mutations, in the two loai, may not be properly balanoed to show the rates of oolncident mutations that aotually ooourred. The mode Of origin Or mutable loOi The analysis of the o-ml oaasa has been of oonsiderable interest beOaUS8 It point8 towards the mod8 or Origin of mutable looi. It la only neoe88ary, it would appear, to insert a Ds loous into (or adjaoent to) a normal dominant loous. Inhibition of the dolainant looue, muert be aasooiated with this insertion, if a changed phenotypic expression is to be reoognized following insertion. If the organization of the inserted Ds loous gives the "h3W-lat8f pattern of behavior, which uaually involves 108s of the Ds locus as a oonsequenoe of a mutation, the inhibitor aotion will be removed and re8BtabliSbment of' the normal organization and genie action may occur. Suoh mutatlons will occur, of oourae, only when Ao i.8 pr888nt. If this projected mode of' orl~in of mutable 1001 ia the oorreot one, many newly arising Ao oontrollsd mutable loci should appear in Ao ao and Ao ko planta that also have DY loui somewhere in the ohromosomal complement. That such newly arising Ao-controlled mutable loci are oontinually appearinq in the AC AC and AC ao plants is well known already. The rate at whiah such new mutable loci oould arise would depend upon the frequenoy of tranapoaition of Ds loci from one position in the chromosome 3amplsment to another position and also upon the relative number of loci present in the ooaaplement thaf oan give a ohanged phenotvplo expression mi0wing such Insertiona. The type Of mutable phenomena to be first observed aftar suoh insertions of Ds 1001, would Aepend on the organization of the Ds loaua at the time of its inaertlon. There is suffioient evidence now to indioate that the rate of trans- position of the Des loous is rather high. This adds ooneiderable strength to the plausibility of this Da-insertion oonaept of the origin of new Ao aontrolled mutable loai. With regsrc? to changed looations of Ds, the c-ml case has been particularly fnatructive. It arose in a C Sh wx Ds / C Sh wx Ds, Ao ab plant. This plunt had besn examined cytologioally and wss known to have two morphologically normal ohromosom8a 9 (exoept in thoae cells that had undergone a Ds mutation). Among the 4000 teated msla gametes of this plant, one pessessed a new mutable loous--o-ml. When this chromosoms, oarrying o-ml, was tested, no Ds aotivity oould ha detaoted at the standard looation--the posftion it was known to occupy in the two chromosomes 0 of the parent plant. Instead, typical De aotivity oould be detected at a new looation: 01088 to or at the newly arising o-ml locus. No obvious ohromoaomal alterations sppear to have accompanied this change in looation of Ds. The ohromosomes 9 in plant 4204, :,weFe naorphOlO~iO,ally normal. Also, the a-ml carrying chrOmO8Om8 $ave no svidenoe of rsduoed transmisafon through the pollen. Gross ohromosomal alterations may aooompaay a change in location of Ds. One suoh 0888 has been analyzed (from plant 4306; for 8 summary of this analysis, see "Notes to acoompany annual report" included with this memorandum). This analysis points towards the mechrnnism undar- lying suoh ohangers in looation of Ds and also supports the conolusion that Ds mutations are associated with Some prooess that often results -20. in tearing out of the Ds laoua from the chromosome. This tearinpout prooess produoes broken ends uapable of fusing with other broken ends, not only fn the torn chromosome but also in the torn-out Ds loous. The torn-out Ds 10~8, with broken enda oapable of fusion with other broken ends, may be inserted into a new location ff a coinoi- dental break occwss elsewhere in the chromosome complement. Fusion of unsaturated broken ends, a well established phenomenon, is all thet is required to complete tke promso of change in location of 3s. A seoond aase, possibly quite similar to this one, is now rece!lvinq! analysis. Other cases involving gross chromosamal alterations that aooompsny transpositions of Ds whioh should also add to the evidence on the nature of Ds mutation and transpasition, will certslnly be found now that I am sensitized to recognize them. Because of the ohanged nature of variegation appearing in individual kernels, numerous bases of changes in location of Ds have recently been rsaognized. Such chen%ss in location of Ds are readily deteclted in crosses of C sh bz WC ds ac x I Sh Rz TX Ds, AC. These transpositions lmluda insertions of Ds to the left of I, between 3 end Bz or between Bz and '8x to give Ds I Bz Xx, T Ds I& TX and I Bz Ds 8x constitutions, respeatfvcly. Some of these kernels have been germinated in the greenhouse this winter to obtain ilnes with these various altered louations of Ds. Unfortunately, germlna- tion of some of them did not occur. This suggests that gross chromo- somal abnormalities affeating germination oapaoltfes may have aoooglpanied the transposition of Da in some of these cases. Before considering a possible mechanism responsible for all of the Ga mutation phmomena, it would be profitable to presentJbrleflyJ some observations of relatively infrequent but nevertheless very -EQ- important types of events associated with Ds behavior. These are most easily deteoted in crosses of C sh bz wx ds ac by I Sh Bz :tTx Ds- early or Ds-few-late, AC aales. A change in location of Ds, a8 mentioned above, is one of these relatively infrequent events. These infrequent events are additional cluea to the fundamental mechfinism underlying the Ds mutation phenomena. In examining the variegated kernels of the constitution C 3h bz wx ds / C sh bz wx ds / I Sh Bz Wx Da-early, Acr au ac, some very clear cases of internal chromosome deletion in the Ds carrying chromosome have been observad. These 'involve (1) deletion3 of the Xx locus. The resulting sectors are I Bz wx Typioal dicentrio Ds behavior may or may not be present in such sectors. (2) Th8 deletion may inolude both Bz and jYx. The resulting seotors are then I bz wx in phenotype. Such sectors may ahow typical Da behavior within them or no Ds behavior may be registered. The relative frequency of these deletions is low but their total frequency is sufficiently high to have been observed numerous times. In kernels of the above constitution but having the Ds-few-j,ate instend of the Ds-early, this aberrant 't;ehavior is much more frequent per visible mutation then occurs when Ds-early is present. I believe that many of the observed visible mutations produced by the Da-few-late state are due to such aberrant consoqucnces of 3s mutations. ';r'ithout question, the presence of a P3 locus, whether in the Ds- ec"rly or Ds-few-late 8tate, introduces chronosomal abnormalities other than those resulting from the zore usual types of Ds action. Any hypoth- esis of Ds behavior must consider the origin of these numerous ce3es of well defined types of chromosomal abnormalities that occur .*hen Da and Au are present but do not oomr with Ds aa, da P.c,or ds 80~ The frequenrry of appearmoe of these various classes of duplioetion, deffoiency etnd transposition is too greet to be neglected. They muat represent some of tha oonsequencea of the underlying meohanism assoaiatad with Ds mutation phenomena. The genetic observations of odd types of ahromosomal ebnormali- tieta accompnnying Ds mutations have received cytological confirmation. Ic plsnta that were Ds Da AC 80, clusters of sporocytes or lndividus.1 sporoeytes have been observed with duplioatfons or interns1 defician- ciea in the short arm of chromosome 9. A few have shown inverrJlons Csome periceatrio). In fmm! oases * only one full chromosome 9 -was present. The other chromosome 9 was raprsarnted by a rln+shaped ohromosoma, usually relativel,v small. fn Ao ac plants having a da cerrying chromosome 9 with 8 heterochromqtic .extenaion eg the end of the short arm, and.a morphologically normal chromosome 9 carrying Ds, it crould be determined that th&sberrant chromosome events involved the I9s oarrylng chromosome and not the (3s carrying chromosome. These observations support the oonclusion that the aberrant types of chromoeomal events are asaooieted with the Da mutation phenomena itself--something must ocuur at the fls locus Itself before the observed typea of ahromosomal rearraneenenta will arise. These cytological observations were made early in the study of Ds behavior. They had caused me much concern, even though they were relatively infrequent. Hera, again, their frequency was auoh too high to be ignored. It is realized now that just euoh conditions are to be expaoted. It should be possible to obtain from the Da Ao plants a number of strains with varI.ous chromosomal abnormalities. Al.raQdy, 1. have obtained plants with duplioations and plants with deficianoiea -3l- in the short arm of ohromosome 9 that arose from Ds .GC plants having morphologically normal ohrOmOSom8s 9. The methods of detecting suoh plants pertain mainly to abnormalities within the short arm or chromosome 9. However, I do have a 0888 of the involvement oi the short arm of chromosome 9 with the long arm at chromosome 8. I have not teen looking for such CWWS. @ore should he found. Possible meohanism responsible for mutable phenomena Enough information has been oollecrted on De Ao behavior (1) at the standard looation, (2) when transposed to the right of T (case 4306) and (3) when at the C locus (c-ml) to allow cons!deration of a possible meahanism that could account for the observed types of behavior. These various types of behavior may be enumersted. In the presenoe of Ao, aad onl.y in the presenae of AC, the Da locus is associated with different kinds of ebnomal ahromosome. or sene behavior. These various kinds of abnormal behavior surely must arise as the consequenoe of one fundamental, primary type of event ocourring at the Ds locus. It would be diffioult to picture a number of different kinds of primary events but it is rs1ativel.y easy to picture a humbar at different oonsequenaes of one primary type of event. This event must aocount for the following behavior: I. Ds at its standard looation gives rise to: 1). Dioentria ohronratids: the fusion of sister chronstlda ooeurring at the Ds locus. ( y,< jyc-; This is the usual event with the Ds-early organization of' the Ds loous. Z).Jh,Ds-early state oan mutate to the Ds-few-late state by a ohange et the loaus that arises in one aall (at a mitosis?). This is shown by the sectors of Ds-few-lute appearing in Ds-early kernels. These sectors are sharply Qelimitad. 3). Ds-Pea-late is more stable in organization than Tss-early. This IS evident because it maintains the Petit-late type ot mutation pattern in later generations and rarely (if ever) mutates dfrectly back to an orgnnizntion @.ving the Dcj-early type of mutation psttern. 4). Xn kernels having Ds-early, n number of interm defZcieacies arise that include segments Rdjacent to the 2:s Lo~us...~, ,4.L- The deleted segments are of various Iztngths. The most f requent:,.hele- I tions are short, althcsugh some longer deletions occur. Pollowing such deletions, Ds activity may be retained in the chromosome or it may be lost from the ohromo.some during the event that gives rise to the deletion. 5). In Ds-rew-late, the visible mutations, that is, losses of' scfpents of the short arm oi chromosome 9, often involve some event that is defi.nitely not Ufoentric chrometid formation with fusion of sister chromatida at the Ds locus. The production of internal deficienoies or the production of Elicentric chromatids coming from fusions of sister ahromatids at positions other than 3, CRT! occur. 0). The Ds aotion may be lost from the chromosome completely without uZtering the chromosome raorphology. (The evide nce for this when Ds is in its stfindard looatlon has not been considered up to now. I have obtained some normal chromosomes 9 without Ds that have been derived rrom Da Db Aa ao plants.) This loss of Ds aatioa is inter- pretad to arise as the oonsequenoe of removal of 3s from the chromosome oomplomeot altogether or removal of the 3s locus rrom its stendrlrd positton ana insertion elsewhere. The removal or Ds sctlon rollowing the c to C mutations of c-ml and the removal of Ds from its standard location and Insertion into a normal C locus to give c-ml 18 substantiating evidanas f'or this interpretation. -aa- 7). The Ds locus may be removed inxtac&.$rom Its standard location and be Inserted alaewherc. A spontaneoua breakage elsewhere in the oomplement is probably responsible for the position of insertion; as the anslysis of oaae 4306 shows (see *Notes to aocoatpany Annual Report'). In these cases, the Ds locus aust be freed from the chromosome completely because It oan enter between the two broken ends at the position of the spontaneous break; fusions of broken ends then follow find the Ds locus is now In Its new location. The freed Ds locus must have unsaturated broken ends in order that it saturate other broken ends by fusion. Therefore, breakage of some kind must be involved et the Ds locus to give a Ds locus with unsaturated broken ends. Th8 freed Da loous need not carry with it a lerie segment of ohromosome. In Its new location, it need not interfere with crossing over. This is shown in the c-ml ease and the 4306 transposition- trhnalooation case. In both oases, crossing over at the region of insertion is not altered. Gross chromosome1 abnormalities, arising from vttrious fusions of the several broken ends when spontaneous chromosome rrithes than chrowetid breakage occurs at a locus other than Ds, my accompany a ohnnge in looatlon of Ds. +?he transposition of 93, in these cases, need not involve a segment of chromatin that oerriss Ds with it; for it has been shotlrn that Ds can move as a sub-miaroscoplc, independent unit in these cases. The analysis of ohangee in location of Ds indicates that the Ds locus may be removed from the chromosome as a sub-microscopic, indepen- del?t unit and that, as mch, it my be inserted elsewhere in the chromosome complement. It indicates, also, that the freed Ds loaus has unsaturated broken ends beaause fusions of unsaturated broken ends must have taken plaae in the 4306 transfer of Ds. Such fusions would be expeoted only if the Ds cootstining, sub-mimosoopio fragment of chromtin had unsaturated broken ends. The "early* state of 3s (Es-early), giving many dloentric chromtids as the consequence of Do mutationa, strongly supports a breakage-fusion mohenism as the causa- tive factor underlying Ds -AC muttrition phenomena. On the interpreta- tion that only one kind of avant underlies all AC controlled mutations, re.Tr:rdless of the visible consequences, some fom of chrolnoaone.1 breskags must be susgected as the primary event res:;onsible for nuts- tion phenomena. IX. Ds, transposed to the normal Z locus, hf+s given ripe t,o the c-vi1 mutable locus. 1. In this position, Da behaves just 83 It does at the standard lomtion or xhe:I trwnsposet- 3 to the right of I (CRSB 4706). a). Da-early type mutations et c-&L give dicmntric rthroma- tids with fusfon of sister ehrcmatids occurrfn:? at r close to the c-ml locus 8s the min type of consequenoe of nrutation. b). This i3s-emly stete r:t c-ml. say autate to the Ds-few- lht&? strta. &hen this occurs, dicentric chromatid forztatioh ceases as the most frequent visual consequccce of 3s rwt?tions. Instead, the rote of' c to C aututions r,l,ay rise: cbruptly. The frequ~ncp of these c to C Hiutations my be the sazw AS the frequenclr of' dicentric ohromatid forming mutations of standard 11s when this 03 is in the 'Tearly** stete. 2. c to C matations oi DES RX c-tel result in ~~orpholopicnlly nomml chromosomes 9. a). The C Loous is usually stable efter such nn event. b). The Da activity ceases follotiin~ such a c to C mutation. i?st sppeers to hem been lost from thr+ locus altogether tis the conse- -as- quence of the c to C mutation. 9 s. . In the heavily veriagated c to C kernels (the c-ml D8 locus aquiva*lent to the standard Du-few-late locus) where many mutations may be observed, a nLrnber of abnormal events have been observed. These ohservstions (progeny test m&e In 1 case only) suggest that occasionally: a). The n~-~lW loous (whloh has the genes of the nor-1 C locus) my be removed frolrz the ohromosome along with Da. This resultrr in a stabilized c locus, no longer capable of' matating to 9. b). The 3s locue may be removed from the c-rr.1 locus and be inserted elsewhere, often in the short arm of chromoao?ne 9. ~.i, rl phenotype rztuy result but continued 11~ activity occurs at the new location. It $,ioes: (1) Rer)er;ted loages of C in the C Peotors rarisinq from such an event if Ds is inserted to the right of C. (2) Losses of C ia the C sectors if inserted to the left of the C locus. In this cane, ho'ticver, there are brenkflge-fusion- bridge cycles that result in twin Hectors of c 8nC deep G in c / a /a-ml Ro nc 80 constitutions. These various events suggest thftt the c to C mutetiona of o-ml result from sozne form of chromosome breekage that usually eventuates in loss of the 3s locus from the chromosome. Sometimecj, however, the removed I)a locus may be inserted sbsewhera if, fit the 8ame time that R P)B mutation ocours, e coinctdantsl break occurs elsewhere. The P8 locus may then be Inserted between these two newly broken ends. Mutations of c-ml to C most often restore the full crtpressfon of the C locus at one single step, as r'ar as one can state from obssrva- Sian3 of the color intensitiea.produued. There are a number of Eutations, huvmvm?, that give rise to much reduced color intensities 02 occevfoually to cieeper intsnsitiee than a normal C loous. These quantitative changes are part of the evidence that roust he oonsidared in my hypothesis x-slating to the primary event associated with Ds action at the c-n1 iocue (or Flsewhere). The separate hypothesis of quantftatfve units at a loclue offers no difficulties, as see11 80 rar, to a s'kraight i'orwurd interpretation of events underlying 3s mutations. If the different quantitative expressions of the various mtstions of a single loous are related to some Inhibitory factor that can expred:; itr;elf'quantitatively, then a new a$ yet unstudied wr5ahle rnctor will have 50 be included in the over-all hypothesis. It relates to the nature of genicr organization resulting from the primary evmt but not necessarily to the primary event itself. Yhst kind of a mechcaissl will give rise to these various events with the expected relative Frequencies? I have thought of one kind of secbanis~ that does not see.n too inconccivabfe. It involves the fcllo3ving a.ssu=lptions and ictwpretations. 1. The main event; responsible for all Ao controlled behavior is related to the reduplication of the 3~ locus at the Lime of gene re~3uplicntion. 2. Cruelly, following chromosome reduplication, each new gene mlscule lies adjacent to the mother molecule but is not joined to the nather molecule by any chemical bonding. '1 c . Hedu;jlfcatfcn of gcrne molecules is orderly in that a11 the dau;hter mol8cLiles lie Eidfc:cent to the mother molecule. Also, all dc:ughtez' aoleculeo lia in a single plane--that i23 the new dau;zhter chroxi~~some ia not twisted about t;hrj mother chromosome. r'he behavior of ring-chromosomea during mitosis ShOWS this. -as- . _. 4. During prop)mse, a repulsion occurs between mother snd daughter chromosomes. This repulsion occurs simultaneously or penrlp so along the whole chromosome. The mother and daughter chromtlds become separated by a rather specific distence, as a consequence of this repulsion force. This is seen by an exmination of %h.e two chromatids at somatic prophasc. In order to give this srecfse spatial relationship between two chrormtids, some form of repulsion force, following chromosome reduplication, is required. 5. '#hen a Ds locus is present in an ac ac constitution, the Ds locus behaves rts other loci do during reproduction nnd thereafter. 6. tihen both 8 Ds locus and an Ac locus are gressnt, the mother end daughter Ds molecules remin chemically bonded together following reduplication of the Ds molecules in oert%in aitoses [contrclled by what Ac is doing at the time). 7. 'khen the repulsion force that separates the mother and daughter chromatids sets in, the mother fjnd dsughter Ds molecules are still tied together. In the adjacent loci, however, the rnother and daughter moleaules are free from one another. Tllrection of repulsion force n n t-4 J-T- $+- Direction of repulsion fC?TYitt " " " " Ds -33- a. Because the bonds uniting mother and daughter Ds molecules are still present, a break must occur in this region. The CR locus mc~y be yanked out of both chromotids: 1 2 -v X .a Direction i of repulsion.~> force ~'I \i Y -4 X 3 4 x P broken ends capable oi fusion with other broken ends. 9. IIecause of the repulsion forces, broken ends ! and 2 and broken ends 3 and 4 will come to lie neer to one another, respectively. Fus&on of these broken ends will occur. The Ds locus will be lost to both sister chromatids. A morphologically normal chromosome will result but it will have no Ds locus. In the case of the Pew-late type of Ds action, this mechanism seems satisfactory as an explanation. It is oecessery to essu~~e that the original insertion of Ds into the C locus, in the case of c-ml, resulted in an inhibition of the action of the C 106~s. Xhen 3s is removed, by the above mechanism, the original orgnnizatiDn of the C locus may again be restored and a normal. C action may nqsin occur. This will explain why the majority of c to C mutations at the c-ml l.ocus give rise to morphologically normal chromosomes 9 with stable ': loci showing no further Ds activity. Tension produced by the united Ds molecules m;sy extend along the chromosome during the repulsion of sister chromatids. Breaks mey occur not as neatly as diagrammed but mriy sometimes result in %Cjscent segments ~~being pulled out of one or both chromatids. -39- This could give slmall internaf aef$ai?naies in the re,-iLox? of the 3s o???? o This a~'~y be responsible far tha very l.tirgc number of new h<- rece~sfve mtants that have ny;rpesred in thmm,atooks. AgAn, Ds my be pulled out of one GhI'ozatid but not out of- the other. Cr, the Ds locus may be gull&? out of bzth chrornatids but in the process be itself broken. Tivo Da loci could then enter one of the sister ohromatid+$ if the varlouo broken ends wepe close to one another. This could give u ne:s, compound 3r; locus. 3~ch compounding cmld oontinue to build up a rather complex, corngound 9:~ locf. ,I`_ I believe . ihS8 must ooour. LI Or, if a coincidental break oclcurrad elsewhere, the free broken ecds of the extruded Ws locus might unite with these other broken ends. Transposition of Ds could then occur. If broken ends 3 or 4 were likewise olose to the new brokm ends, compound, gross chromosoael renrrnngements could arise. Various abnornrlities could be expected to arise and be visible genetically, when appropriate genie markers UC prenentp to maks the events realized. h;.thou@, this CIctgmrtmed scheme aeemz sstisfsctory, in q~:qeral, to account for the few-late 33 behavior it will. not tell -*hy on6 sets the 3s-curly behavior, that is, dicentric forztisn because of fusion of sister chromatfds at the 2s 2.ocus. As stateC ebove, t,he few-late tyoe ,I?' Da behavior is rather stable but the early .J f 9 type frequently thrmva the few-late type. The early type Ds behavior probably involves P. zore complex organization of tho 3,s locus, ptjSSSbl:r a cowp~~i;d organizotioa with dui;licate Dr3 loci having, thcref~m, mny bcrnda following gene rcdu;~lication. This may be: Or, other gene3 may bEt pressEat between some of the reduplioatsd Ds loci: The extensive binding action of these reduplioated Ds loci may result in rather complex tensions and breakage8 during the repulsion period that mny result in plaoing broken ends of sister ohromatids alose to one another so that fusions readily ooour. Also, a reversion to a simpler form of Da organization through deletion of some of the ..- Y 5 . .a ', Ds loci could be anticipated, that is, a few-late Ds organization , ..` @rising from the early-Do organization. The various isolates of Ds standard end of o-ml that show grades of Ds behavior between the extreme few-late and the extreme Ds-early, may be fitted into this scheme. The more and the xider separated the 12s loci, the more the chanae for dioentric-forming fusions of broken ends and the less the chance for broken ends of the same ohromatid to fuse. Alao, sequences of change in state of the mutable loci should be ayiticigated on this meahanism. The high dioentric produoers should be the most unstable in maintcining their ststee -#here&s the loWest dicentric producers should be more stable in their states, raraly* if eter, throwing a really high dioc:ntric prociuoing state through a single mutation. The states producing IatermsAiRte r-Rtes of Eicentric , formatton should go in both directions: from intermediate to hi?h or to low. Also, the new unstable mutable loci that arise ns the conse- q,uence of transpositions of Ds may ref3eat, to some extent, the state of the locus before the transposition. If a compund locus is transposed, in tact, then the mutation pheno;rlreny observed ~3.11 be suite different from that observed if a single 9s locus enters the new position. The hy?othesls outlined Is not oomplex. It does rot call on 5eny assumptions of a purely speculative nature. It is rattier one that could well have been Hnticipated in sdvence of sny sugg,estfve evidence. There Is no raason to assume that something coul.d Dot go wrong rvith the reduplication process that would leeid to just this sort of association of daughter molecule with mother molecule. The aberrant mitotic behavior of ring-shaped ohromosomes auggeets that somethlnq of this mture may 80 on as an occasional error during the reproductive strlge, even fn nor:ul loci. It might likewise be responsible for some of the epontaueous breakage phenomena known to occur rntber frequently in the Maine chromosomes. It is the Ds - AC combinatic?n that brinq thiR to frequen t expression. -42- Part II. The c-m2 loous Classification and description of the types of mutationa occurring at the o-m2 loous Mutations ocourring at the o-m2 loous differ from those occurring at the c-ml locus in several easily distinguishable respects. The c-m2 locus is AC controlled; mutations occur only when Bc is present. Also, the time and frequency of mutations of c-m2 respond to AC dosages as do u-ml and t!:e standard Ds. The mutations of c-m2 may be classified into several cat>gorios: I. The Pink mutations. These give various grades of pink to red color in the aleurone with pr pr constitutions or various grades of purple color with Pr. The term "pink" is laboratory language. It ia not a well chosen descriptive term. Until T know more about the various classes of c-m2 mutations, I shell not try to coin specific designations. I shall use the term rppinkV! to cover a seriaa of mutations that appear to be similar or related. Later, it may be necessary to distinguish between various sub- classes of thls group. a). There is a wide range in the intensities of the color produced lollowing mutations of' c-m2 to pink. Some mutations to pink are too f'aint for the outlines of the seotors to be uertainly defined. Some of' these "hidden" mutations may be detected when special conditions are present. These will be described. Other pink mutations are quite dark. A quantitative series of pink alleles are produced by mutations of o-m2. These quantitative alleles of pink may be obtained and maintained from isolates of germinal mutations. Only a limited series of' suah alleles have -439 been isolateU and studied in later generations but many more are aPailabf8 fOX' 8UCb JRll?pOSCaS. The germinal pink isolates that have been studied so far have been relatively stable in their expression. I suspect, however, that some of these germinal mutations may respond to Ac by giving variegation in the depth of color. I am not too sure of this as yet because the pink phenotype in these isolates does not show uniform distribution of color over the aleurone even in ao ac ac constitutions. It is somewhat mottled although the grades of color contraits in e single kernel are not extreme. The AC se] ac3 constitutions seem to be more mottled and in patterns suggesting changes at the locus. I: don't know now whether this effect is due to ohanges at the pink locus or to changes at other, as yet non-deteoted Ao-controlled loci. b). The pink mut&nts are often assooiated with the produc- tion of soxe diffusible colorless substanoe. This subseance (substanc. 1) can be used by a normal C locus to increase the intensity of the aleurone color in those cells having a normal c locus. Apparently, relatively large amounts of this diffusible substance may be utilized by single cells having a single norms1 c locus. &hen this utilization occurs, the color of the aleurone in these oelle may become very deep, giving a super-C color intensity. The dosage responses of a normal C locus may be reflections of the limited r,uantlties of this substance that one normal C locus iJroduces. The methods of showing the presenoe of this diffusible colorless substance produced by cells having mutations to pink, involves an analysis of the color patterns and distributions In sectors of kernels resulting from the following uombinatfons of loci. These combinations will also show that the colorless diffusible substance is not produced by an unmutated c-m2 locus. (1). o-n12 female x C-normal Ds male; Ao ac ac or Ao AC ac constitution (2). r, * x chromosome $3 with a broken end. This broken ohromosome aarries a normal C locus. A0 ac 8c or AC AC ac constitutions. ( 3 ) . " " x o-ml male; AC ac ac or AC AC Ao. (4). pink * x C-normal Ds male. AC ac RC or AC AC ac. (5). rt n x chromosome 9 with broken end. This broken chromo- some oarries a normal C loous. ac ac se oonsti- tutions as good as those with As. ( 6 ) . *' " x c-ml male. AC ac ac or AC AC ac. Ths color patterns produced following each of thsse crosses will bs desuribed later. 0). The isolated germinal mutations to pink have given dosage responses. This has not been extensively worked out but the grades of intensities of pink color in selfed ears of plants with pink / c compared with Intensities given by crosses of the same plant to c f8male plants, certainly suggest3 dosage effects. More instructive for the dosage studies of pink are the kernels with CI / o / and a chromosome 9 carrying pink that has a broken end. The resulting breakage-fusion-bridge cyolea In the pink crarrying chromosome produce definite patterns of changed color intensities, obviously associated with different dosages of the pink locus. Too few kernels are available for an extensive analysis but those examined clearly show dosage effects. They indicate that the more pink loci present, the deeper the color. The quality of the color and the intensities in the sectors with various doses of pink differ from those produced by similar unit doses of a normal c lOOUS. -4s d). The pink mutants (some of them, at least) are deficient for some substsnce that a normal C locus produces. Two such substanaes muy be deficient but at leaat one is a colorless, diffusible sub- stance (substance 2) that is associated sith the activity of a normal C 10cus. The cells with the pink mutation can use this substance to increase the intensity of the pink color. This has been shown by an analysis of color patterns and distributions In the sectors of the kernels having constitutions given above. In addition, the combination of c-m2 female x C-normal ds male, AC ac ac or AC Ao ac, has been useful. II. Hututions of c-m2 giving conditions that resemble the full C genie activity although varying in quantitative expression of this activity. In crosses of o 80 females x c-m2 AC AC males, pink sectors appear in the kernels of the resulting ear. Xithin some of these sectora, mutations to a type of full C expression often occur. Unless there has been a duplication of the o-m2 locus Rt the same time that a mutation to pink occurred to give two c-m2 loci, each _a capable of independent mutation, it may be concluded that e muta- tion of a single c-m2 locus to pink can be followed by a second mutation at this locus to give ~3. full C type expression. Yone of the germinai pink mutations that have been csrried to a second generation have continued to give these pink to full C mutpitions In Ac constitutions. The selection for continued study of germinal pink mutants with spots of full C activity in them, has come only from the clrosses of e-m2 females x o males. In %hese cases, the female contributed two loui to the endosperm (two chromosomes 9). -460 A pink mutation may have been present in only one of them. The other locus might have been a non-mutated c-m2 locus and capable of mutating to full C. This could give C spots in a pink background. The evidenoe for this is suggestive. In 4 of the 7 tested oases (4451A, 4456-2, 4456-4, 4458A-2) the chromosome entering the egg nucleus likewise oarried pink. In the other ? cases, the chromosome entering the egg nucleus carried a non- mutated c-m2 locus (4456 -1, 4456-3, 44568-1). IQ the mutation to pink oucurred during the divisions of the female gametophyte, just such apparently conflicting w results could be anticipated. Such4female gametophyte could have nuclei with the pink mutation and nuclei with the unmutated c-m2 locus. Because mutations of all AC controlled mutable loci ususlly occur lste in the development of the sporophytic tissues or often not uqtil the gametophyte or endosperm stage is reached, the period of' origin of germinal mutations could be as late us that just described. (It might be mentioned at this time that occasionally an jIc controlled mutable locus may mutate early ir, the development of the spotophytic tissues. It is a very rare event, however,) Tests of the stabilization of pink mutations nlust come from the reciprocal cro3s: c x c-In.2. The germinal pink mutations 3howIng full C spot3 must be selected and tested. They occur but thfs obviously oritical test has not been made. I was not sharrlp in focus on the c-m2 mutations when the selection of germinal wtstions wus made! A singl.e (?) mutation of the c-m2 locus to give the two diffus- ible substances or at least a second diffusible substance (substance 2) that is associated wfth gene activity of a normal C locus can occur. iiel.1 defined seotors assumed to have substance 1 and known ,^ to have substance 2, appear on the kernels coming from the following /_ crosses: (1) o-m2 / e-m2, AC ac; aelf-pollinated. (2) o-m2 female x c male. (AC ac RG or AC :?c ac constitutions) (3) c female x c-m2 male. ( w *' " ,t 1 (4) pink female x c-m2 male and reciprocal. (AC RC ,uc or AC AC? ac constitutions). It is concluded that if not a single mutation then a sinale event at the c-m2 locus may result in a modification of genie organization that will produce both of these substences associe.ted with action of a normal C locus. The amounts of the substances produced need not be at the same levels as those produced by a normal C locus. The levels may be considerably lower if the color intensities shown are any indication of such levels, As stated above, it is possible that successive mutations occur, first to pink,which often produces an excess of substance 1, and then to u mutsnt giving a darker color associated with the production of some substance 2. There is no certain evidence as yet that mutations giving substance 1 are produced without some accompanying pink color formation, however faint this pink color may be. There is some evidence suggesting that muttitions do occur that give substance 2 but not substance 1. The sectors suggesting th!s are pink with deeper colored rims in the pink sector on p. restricted part of the border between a pink and a colorless sector; or, within a pink sector, darker areas with diffuse borders often wppeisr. I am not at a12 certain that this interpretation is the only one that will explain some of the color patterns of these sectors. Xore obse.vatians and thought are required. The mutatioils to full C activity are quantitfitively expressed in that various grades of color intensizies are present. The sectors with these phenotypes may be quite light in color or fairly dark. No germinal mutations to full C color activity have been detected so far. This is not evidence that they do not OCGUI'. I have not looked exhaustively for them. Unless they @vs deep grede3 Of full !: color, I might confuse them with pink. I know now that-:often>the two differ in eppearnnce, reaerdlsss of color Fntensit& sxsressions. It should be possible to discriminate between the t*o, in some cases et least. I have not taken time to find thorn but will do so. ?II. The unmutated c-m2 ~GGUS produces no detectibls quanti- tiea of either suhstunce 1 or substance 2. This will be shown when the kernels resulting from the crosses enumerated above Ere described. IV. Ridden mutations occur at the c-m2 locus. The frequency of hidden mutations seems to be quite hlch. This f~ i?hown by: (I) The types of sectors produced in the cross of G ac females X O-82 i'iG AG IIlQlt'33. ;;hen the variegation pattern in these kernels is oompared with the variegation pnttern produosd by c-ml (from the heavy visible-mutation producers) or aith those produced by DB-enrly, it is evident that tLle visible sectors in the G-PC? kernels rr-!:>resent some form of sub-sector arising in 8 descendtint cell of one t3nt had undergone some primary event. This orininal mutation event in the .a nnGestor cell resulted in a visibly ohanged phenotype only in a sub-sector or sub-sectors. Xithsr some kind of (a) segregation to daughter aells occurs following the initial event or (b) su~cessiv3 mutatfons followr after the original mutation or (c) both types Of eV0lltS OGCUI'. -49- (2). Mutations of c-m2 to stable o loci occur rather frequently, ? ???????? o Some stable c loci have been isolated from crosses involving c-m2 / c-m2, AC AC or AC tic plants. I can not say that all such nutations are completely stable. 3eversl have been carried for three generations without showing nlutatLon in the presenue of AC. Several ears have apperlred in the crosses of c / o-m2,Ac ac females x c,ac meles tht+t show large sectors on the ear with only colorless kernels. This may bo due to loso of the i?c locus (evidence f'or loss of AC will be presented later; it is not uncommon;) or to stabilization of the c-m2 locus. Evidence that will diff'erentiate between these two alternatives will be easy to obtsln. It will be necessary, in any case, to determine whether mutations of c-m2 to stable c are associuted with some segregation of chromatids, i.e., stable c in one chromat,id and an altered organization of the c-m2 locus in the sister chromatid. The sub-sectors, mentioned above, suggest segregations may sometimes occur. (3) Soma "hidden" mutaticns fire actually pseudo-hidden in that they are Fick nutet-lons with such faint colors that detection may be very difficult or not possible, in many cases. (4) I suspect thnt c-m2, as well as c-ml, has a class composed of several kinds of hidden mututions that are? not easy to detect. These include changes in organization of the c-n2 locus that will result in changes in the frequency and type of future mutation; losses of parts of the c-m? locus; duplication of parts; relocation of parts, 8tC. Although c-m2 may illuminate some of the activities of a normal C locus, a study of its mutable behavior 13 discouragingly complex when one has this loous as well as the other mutable loci to carry along. -- I -5O- v. Mutations at the c-m2 locus have not resulted in many obvious losses of segments of the short arm of chromosome 9. The genetic tests for this are inadequate, however. The crosses that could. show this clearly inrrolve o-m2 A'x to: (1) c sh wx da ac and to (2) C sh bz wx ds ac. The ;tix-m locus was carried by the ohromosome with c-m2. The wx areas in the kernels coming from cross (1) Sire usually due to mutations of lVx-m. In or058 (21, only a few kernels ;yith C Ez - C bz tireas were observed. These cress may arise from events other than those associated with the o-m% nutnticrns. It may be stated, at least, that the isolates of the c-m2 locus have not given, as yet, eny 4 m%P- states that regularly result in looa of the short arm of chromosome 9. 4 I believe they will appear, sooner or later. To summarize:- The o-m2 locus may be compound in that the mutations occurring at c-m2 are not asscciated 5sitb the quantitative expression of one type of reaction. Two distinctly different types of visible mutation occur at c-m2. They involve the production of two different diffusible, colorleso substa.nces, both neoessary for full C: expres- sion. The production of substance 1 is often associated with the appearnnce of some pink color. llutations giving both substtince 1 and substsnce 2 may occur. Although both substances may be present in a seotor arising from a cell having such a mutetioa, full C color need not appear in this sector. juantitative expressions of full C activity are observed. The mutations giving pink likewise show quantitative expressions. Some mutations to pink give rise to stable loci that no longer mutate in the presence of Ac. Others rniiy be mutable but thiB is not oertain as yet. Mutations to full C -510 type activity, regardless of the quantitative level, result in production of some diffusible substance (substance 2) that the pink locus can use to increase its color Intensity. This 3mc? substance is likewise produced by a normal C locus. That the weaksned color intensity of pink mutations appenrs to be associatad with some specific deficienoy of a needed substiince is suggested by the dosage responses of the pink mute+,ions. This substance can not be substance 1 for substance 1 is often produced in excess even in the mutcticns aivinq very foist pink. Tt my be substance 2. If so, then all mutations giving pink must, likewise produce some substance 2. Neither substance 1 nor subs%ancF: 2 is produced.by the ucmutated c-m2 Incus, however. If the intensity of the pink color is an exprsssion of the love1 of some limiting substance this substance must be one of tho products of mutations of c-I&2. It is not controlled by other loci, as the doswge responses of pink indicate. This may be oubstence 2, u3 mentioned above, or it may be a third substance associnted %ith ~enic action of a normal C locus. Althou@ this sub&awe-producer interpretation of .mutations of c-m2 fits the observations so far made, 1 believe that one should be canny about accepting it as it has been presented. I have that uncomfortabla feeling of having mentally over-looked the missing link that could simplify the whole interpretation. This is by way of confession and not of retraction. Bsfors illustrnting the color patterns produced in the kernels coming from the enVxuerated crosses, some statements should be given regarding the origin of c-W. -52- The origin of the c-m2 locus The o-m2 locus first appear8d on en ear of a self-pollinated plant--plant BOOOB-2. Thic plant had the constitutions: I , ,&. / *pyd I: m-2 m 3h 'Xx ) AC AC. It is only in this one plant of the culture that I hrjve been able to find nny evidence for a mutable c locus. Alao, I hnve not been able to determine, as yet, whether or not R mutable C locus was responsible for the first appeoracce of this o-m2 locus, or whether it came froa a normal G locus, ~3 did c-ml. All that I ten state is that there is no certain evidonoe for its origin from a C-m locus. Culture 4000 ~8s segregating a nutable pale-green locus. Each plant in the culture *#as self-pollinated. to determine whether or not the pale- green mutable locus ~8s present. The origin of plant 4000B-2, that bed the c-m:! locus, will be given starting as far back 19s the &&j &A r,;it: t +4. cross that Five rise to a kern81 W&S& two broken chromosomes S in the ZygOt8. (1) 2467A-5 X 2476-6 I Ax; G "x/ ,- ,a' i I / I ;Jix @---t-- I \Sh wx 4 G ' :d?r One I-S; *h'x-x kernel MS selected from this cross because both chromosomes S had newly arising broker! ends terminriting the short arm. This kernel gave plant: S~-p~.-,~-~ _. i,;`j$i t- -`., -530 I ;6' a'. f (2) 42~B. The main ear and the tiller ear of plant B-42 were selfed. The tiller ear segregated the new mutable pale-green locus. The progeny from nslfing the main ear did not give this mutable locus. (3) The progeny of the tiller ear gave culture 3592. This culture soaregeted pyd ee well as the new mutable pale-green locus. Cytological examfnnt,ion of a number of pl%nts in this culture were made and genetic trsts confirmed the constitution of the tiller of plant 42-B to hsve % chromosome 9 with 8 duplication of the short %rm and carrying I 8x and a marphologically normal chromosome 9 with pyd C Sh wx. Ao was also present in culture 3592--whether homo- zygous or segregating, I do not know, (4) Plant 3592A-9 was selfed. It had a duplication chromosome 9 with I gx and a no-1 chromosome 9 with pyd C Sh wx. The progeny of this self gave: (5) Culture 4000. 40008 came from the I Wx y kernels. ijelffng 400QB-2 gave the c-m2 locus carried by the aorpholo~i- cs11.y normal chromosome 9 of this plant. Mofie of the other plants in culture 4000 showed the presence of a c-m2 locus. Examintiti~n of the solfsd ears of other 3592 plants hes given no indication of a c-m2 locus, These plants, however, did show other mutable looi. Those recognized were: a). y to Y b). Starch consistency change from wx-like to ipots CarOSG c 3h C-SC c-c, Sh Gh c A c ah 4074-a x 4204 1 66 2 81 152 4075-9 x * 3 55 4 g3 103 4074-18 x ') 5 85 1 117 233 414sc-1 x .! c 33 1 45 69 4078-9 x "* 1 65 1 96 169 4081-28 x fl (ir 3 54 1 92 154 -- 4c74-5 x fl 5 70 1 82 195 Mm-25 x ' 5 51 1 79 102 4oa-3c x 7 1 30 3 40 13% ._ Totals 23 509 15 725 1453 'Total = 2725 3,257 ::;h : 1468 ah 4421 x c 3s a(3 7 CIJOSS CSh C c Sh c sh 911 c+c: S-3 o-jc 3h no C sp0ts 4323-7 1 x 4294"-1 2 1 167 153 11 8 1 2 57