THE UBRMH OF THE UNIVERSITY OF ILLINOIS UNIVERSITY OF ILLINOIS Agricultural Experiment Station BULLETIN Xo. 320 INHERITANCE OF KERNEL ARRANGE- MENT IN SWEET CORN BY W. A. HUELSEN AND M. C. GILLIS URBANA, ILLINOIS, FEBRUARY, 1929 CONTENTS PACT MORPHOU )( i V < )F KERNEL ARRANGEMENT 301 MATERIALS AND METHODS. .. 302 Review of Previous Work 303 EXPERIMENTAL RESULTS 303 Progeny Segregations and Their Classification 303 Fi Progenies 306 F 2 Progenies 307 Modifications in the Genetic Expression of Rowing 308 ANALYSIS OF THE INHERITANCE OF KERNEL ARRANGEMENT.. 309 Segregations in the F- Progenies 309 Segregations in the Back-Cross Progenies 313 Segregations in the F 3 Progenies 313 Rowing in Relation to Plant Characters 328 PRACTICAL ASPECTS OF THE INHERITANCE OF ROWING 335 SUMMARY AND CONCLUSIONS 336 LITERATURE CITED. . .336 INHERITANCE OF KERNEL ARRANGE- MENT IN SWEET CORN BY W. A. HUELSEN, Assistant Chief in Olericulture, and M. C. GILLIS, Associate Ears of the Country Gentleman variety of sweet corn differ in appearance from those of other varieties by having an irregular or '"zigzag" arrangement of kernels. This condition becomes fixed to a considerable degree under careful and long-continued selection. However, there is a constant recurrence of individuals which are more or less rowed. The lack of uniformity among the segregating ears, as well as the variability in the percentage of segregates, indicated at first that this might be a form of polymorphism, and suggested a careful study of the character. MORPHOLOGY OF KERNEL ARRANGEMENT Stewart 6 and Weatherwax 8 explain the peculiar arrangement in Country Gentleman Sweet Corn as being due to the crowded condition of the kernels, which in turn is the result of the development of both the upper and lower flowers of the pistillate spikelet. In rowed vari- eties the lower flower remains primordial and only the upper flower functions, and the familiar rowed appearance results. The phyloge- netic significance of the functioning of the lower flower is still some- what in doubt, according to Stratton 7 and others. It is probable that reduction in the number of pistillate flowers is the more highly special- ized form. It appears, therefore, that the distinguishing difference between Country Gentleman and rowed varieties is in the functioning of the low r er flower. Kempton 4 mentions a sweet corn in which the irregular kernels are due to the indiscriminate arrangement of spikelets. Weatherwax, 8 however, claims that he has found no variety of corn in which the spikelets are not arranged in rows on the cob, irrespective of whether one flower functions, or both. In w r orking with Country Gentleman sweet corn, the authors have frequently noted the occurrence of rowed individuals in open-polli- nated strains selected by the ear-row method. Occasionally such ears were distinctly rowed like other varieties of sweet corn, but more often the rowed condition was confined either to the butt or to the tip, as mentioned by Stratton. 7 Specimens in which the rowing was intermediate, or indistinct, were of frequent occurrence. Further in- vestigation proved that rowing occurred in strains of Country Gentle- man obtained from a number of widely different sources. The theory of Weatherwax 9 would seem to account for this phenomenon. He 301 302 BULLETIN No. 320 [February, states, "At times, however, a set of conditions, presumably environ- mental, may limit the size of the grain or increase the length of the cob sufficiently that the rows are almost straight, altho each spikelet is still producing two grains." On this basis one would expect that most of the large-sized Country Gentleman ears should have the kernels in rows over all or part of the ear and, at the same time, con- tain two kernels in each spikelet. The authors have observed that all the distinctly rowed ears or parts of ears in Country Gentleman cultures have spikelets with only a single functional flower, which accounts for the regularity in the arrangement of the kernels. In the intermediate type, where the rowing is present but more or less indistinct, paired kernels borne on a single pedicel are interspersed with single kernels in which the lower flower has remained primor- dial. This intermediate type of rowing differs, however, from another type which appears the same but is merely due to incomplete polli- nation in an otherwise distinctly rowed ear. MATERIALS AND METHODS Narrow Grain Evergreen is a 16- to 20-rowed sweet corn having an obscure origin. Certain commercial strains of this variety were selected from crosses between Country Gentleman and Sto well's Ever- green. The strains used by the authors, however, were the result of long-continued selection from Stowell's Evergreen. This may account for the fact that all but one of the rowed parents proved to be homozy- gous for kernel arrangement. Country Gentleman, when true to type, is characterized by "shoe- peg" kernels and by an irregular or zigzag kernel arrangement ex- tending over the entire ear. Crosses were made in 1924 between parents which had been pre- viously inbred for two generations. The F 2 and F 3 generations were grown from ears obtained by selfing F x and F 2 plants in 1925 and 1926 respectively. Back crosses between the F! progenies and the parental strains were made in all cases, but many of them failed to fertilize owing to differences in time of maturity and a poor growing season. The ears in each generation were harvested as mature corn and later classified for kernel arrangement. Such classification included all the ears that were filled well enough so that the type of kernel arrangement could be determined with reasonable accuracy. Wherever the segregating progenies were separated into three or more phenotypic classes, the x 2 method was used for calculating the closeness of fit between the observed and expected frequencies. The probability values were taken from Elderton's tables as given by Pearson. 5 1929} INHERITANCE OF KERNEL ARRANGEMENT IN SWEET CORN 303 In the case of 3:1 segregations the probable errors were calculated by the formula P.E.= 0.6745 \/pqn, in which n is the total number of individuals, and p and q are the percentages, .75 and .25, corre- sponding to the ratios concerned. Review of Previous Work Halsted and Owen, 3 in crosses between Country Gentleman and several rowed types of sweet corn, observed "a strong preponderance of straight rows" in the progenies. In some, progenies they found only an occasional ear which was entirely zigzag, but many ears occurred in which the upper third was irregularly disposed while the remainder of the ear was rowed. East and Hayes 1 stated that the zigzag, or irregular, arrangement of kernels on the ears of Country Gentleman sweet corn is a dominant character due to a single genetic factor. They drew their conclu- sions from the behavior of the F x and F 2 progenies of a single irregu- lar ear which had been selfed. This selfed ear produced a progeny having approximately 3 normal to 1 irregular. This departure from the usual behavior of a heterozygous monohybrid, w y hen selfed, was explained as being due to "reversed" or "fluctuating dominance." A single progeny obtained by selfing a plant producing a normal ear gave all normal ears, which further led to the conclusion that the normal class was a homozygous recessive. In addition to the above type of irregularity there is also mentioned by East and Hayes 1 an- other kind of irregular kernel arrangement which they called "physio- logical fluctuations" which were found to be non-heritable. A con- fusion of these two types made it difficult to classify the segregates. The authors experienced the same difficulty in classifying their ma- terial. Emerson, 2 in reporting the results of a cross between dent corn (rowed) and pop corn (irregular), states that the arrangement of grains in regular rows is perhaps the dominant character. The segre- gation in the F 2 generation seemed to indicate that there is a single factor concerned. However, extreme fluctuations in the F t progeny, reaching as far as the irregular (zigzag) type, threw doubt upon the single factor hypothesis unless such fluctuations are regarded as the "physiological fluctuations" of East and Hayes. 1 EXPERIMENTAL RESULTS Progeny Segregations and Their Classification Crosses between Country Gentleman and Narrow Grain Evergreen produced F x progenies which approached the Narrow Grain Evergreen parent in type of rowing (Fig. 1). The rowed kernel arrangement must, therefore, be incompletely dominant over the irregular type. 304 BULLETIN No. 320 [February, lz 05 o O H W a & 02 a K PH H m 1 1 O5 O5 -H CO (N -l (M 4J U O PH'&, EE EPn '.'< Descrip phenotyp 1929] INHERITANCE OF KERNEL ARRANGEMENT IN SWEET CORN 305 s~ 5 iS , . .- e. .J? '58 &> "?r ^* O< *^ "^7 li ~ ~n g ;f g ;| ; : S p ~ ~i r 1 r- 'E ^i ?r ?r M C* c3 'E c. x~ - z. c, c. "*? S> brf' ^ rS rN rS >< X ^ ^ x ^ X 3*8 * JS? M -^ .i e N * "-I *"* M _ = 'E ?- 'E ^ ^ 'E ', '! ^, ^ fl 'E ix b- , 'E ' ^ .Q P-i E E 1 if '= -E g ? ^ xx 1 f f. M x g 9 f C fl fl c E c c a "o "b a'c" be'? ^ s? r d h o3 ^3 II U i^ w 5 S C3 rO r^- ^tf X ^ X tf %^~ o^ ~ ^ ~ cT ^H ^W ;;- 3 Ogj o X T-SS7-S =ve' ^ W g ^- ~ F -t _= 5 ,= :: !N _= S M "^ r ' ^ g TJ 2 ggccgo fcf; 3 ccO b-^- b ri 3 > X >"cX >"c> M ^- --io2c^o2M-SS P ,9 i F r 9 L 9 >5 i- = wi> bOco sZ o * o * o o Z g_ MO -- "" * * -* " "** r * -^ ^ *^ ^* - ^-* ^ ^, *"^ ?~*\ T diate, 19 zigzag. GO ?^ ^^ *^ O ^ ! C 1 ! "* >. 7^5 ~~ ^. re ," :** ^T r*'- o S c 1 'S. 'S. S, 'E E 'E 'E 'E 'E 'E 'E. 'E. IM tfl S. h-i 5-i p_ z. i. PH CL, PH T3 1 'S. 'E 'E 'E 'E'E 'E 'S. 'E 'E 'E 'E'E & PH ^ p.si.a, ^^ p, -E PH PH A I O 00 i. 1 03 S* S g * g J g 1 i 2 2 2 2 3c>> BOAEIXN No. 320 [.February, There was some variation, however, in the degree of rowing among the F! progenies from the different crosses. The dominance of the rowed kernel arrangement, as indicated by the authors' experiments, is in accordance with the observations of Halsted 3 and Emerson. 2 F Progenies. Much of the parental material was found to be heterozygous. It may be noted that two or more types of F 2 segregation occurred in each cross. Crosses 1003. 1004, 1005, and 1022 (Table - ~ 18, and 22) gave rise to both dihybrid and mono- hybrid ratios, while Cross 1015 (Table 23) produced two types of monohybrid segregations. Obviously in each of these crosses the F. plants which were selfed were not all of the same genetic composition. In Table 2 are listed the crosses and the F x factorial formulae ne - sary to account for the various - -legations obtained. (Where there was only one type of segregation, it may be assumed that the F t plants were alike genetica Since only 1 and IS plant were used in making each cro- least one of the parental plants must have been heterozygous. Altho all the strains used as parents had been previously inbred for two generations, many of these by subsequent inbreeding proved to be heterozygous. In columns 3 an-, .ble 2 are given the pedigree numbers of the parental plants used hi each cross and their probable genetic composition. Where the same parental strain was used in two or more crosses, individual plants were used in each cross shown by the last figure of the pedigree number. The Fj open-pollinated progenies from the crosses mentioned above, with two exceptions as shown in column 5 of Table 2. were rowed and fairly uniform. Each of the F T plants from these ere- must have contained both the Pi, and Pi, factors, for the "rowed" kernel arrangement, either one or both being heterozygous, as shown in column 2. In Cross 1002, where the F x generation contained only the Pi, factor, the open-pollinated progeny was intermediate. Cross 1008 produced an F x progeny which was much more variable than the rest and seemed to give a segregation of 1 rowed: 2 intermediate: 1 zigzag. An F. segregation of this type might be obtained if the row Grain Evergreen parent contained the factors Pi a pi x Pi, pi, and the Country Gentleman parent was homozygous for pi t and pi,. The F, progeny would be expected to contain the following four types in approximately equal numbers: Pij pi x Pi, pi,, Pi l pi x pi 2 . pi 2 , pi x pi. Pi, pij. and pi, pi, pi 2 pi,. Since the F, classification was made before the various types of rowing were well under- stood, it is probable that the Pi, pi, Pi 2 pi 2 type was classed as * f rowed" while the Pi, pi! pi, pi, and pi, pi. Pi, pi 2 types, combined, made up the intermediate class. All the Fj plants from which F 2 progenies were grown must have been of the same genetic composi- tion since only one type of F 2 segregation was obtained (see Table fMfl " :f_EU> H. A r5- ** y. til'" _OJLV 25). The remaining crosses. Xoe. 1002. 1006, 1010, and 1018, were >:~ : 7 !'-_, r i5 In Table 3 are fisted the numbers of tine inbred strains from wtodb IDC pir-ir--.^ p.;ir.-.~ -^-re ^I^::cC- -..:*-.-:: -sriiiL -Jr.- composition of each. It will be noted that* with one exception, the Narrow Graim dvcxgieen strains wise hflnMteygyms. Wanow Graui rgreen strain 207, together vith all the Country Gentleman :: 1 1. i :':.'j.'",r 3 i; z:"r 0" r'rT'I ITTc'l '". T !~ 'i Frrn Y?: ^ as :T Lvaiasa 3)7. F--:>-i ^-rr-r-r. Nb . : * >": sri ues. Twculy-tiro F 2 pragenies firoam asera different i:/.:c'-i Li:'_r^-7 IT.:.: :;~_r iifiin:: zr-: '_;: .:::'_:: .. mehi were distinetlly lowed; G^onp n,, in which the kem- distinetfy rowed than in Group I; Group HI. an inter- ; vhidi ^ ndttei r:~ -re gave a ratio of 217:5.8:6-4:0:9 fTabies 1 and 4), or ap- 3:6:6:1 ratio. This tjpka% dihrbrid ratio led to the at kenod arrangement is doe to the interaction of at 308 BULLETIN No. 320 [February, This method of separating the F, progenies into four phenotypic classes did not prove satisfactory, owing to the lack of uniformity in the individuals under Groups II and III. It was found that Group II could be further subdivided into two classes based on the contin- uity of the rows. Group III was also divided into two subgroups, one approaching the rowed type and the other resembling the zigzag type. Each subgroup was further separated into two classes. The F 2 material was therefore classified into eight phenotypic classes as shown in Table 1 and Figs. 2 to 9. Groups I and IV may be readily distinguished. In Group I the rows are clearly defined and continuous from butt to tip. All the em- bryos face the tip of the ear. In Group IV the rows appear to be entirely absent due to the zigzag arrangement of the kernels, which have a typical "shoepeg" form in contrast with the flattened kernels in Group I. The chief difference between Groups I and II is that the rows are less regular in Group II (see Figs. 2, 3, and 4). Subgroup III-A (Figs. 5 and 6) resembles Group II, while Subgroup III-B (Figs. 7 and 8) tends more toward the zigzag type (Fig. 9) in Group IV. When Figs. 3 and 4 are compared, however, with Figs. 5 and 6 it will be noted that the chief difference is in the slight offsetting of the kernels. Altho Subgroup III-B resembles Group IV, it cannot be included with the latter owing to the traces of rowing. Modifications in the Genetic Expression of Rowing East and Hayes 1 mention two kinds of irregular (zigzag) kernel arrangements in sweet corn. The first is a "physiological fluctua- tion" which is not inherited, while the second is "a definitely inherited character, or possibly a set of characters." The first type of irregu- lar kernel arrangement will always be encountered in sweet corn cul- tures. The cause lies in the development of less than the normal number of kernels. Frequently only the butts and tips of the ears are affected in this way often to the extent of being entirely bare. Less frequently the spikelets which fail to develop kernels are scattered over the entire ear. It is obvious that any condition which prevents the development, of the entire complement of kernels on the ear will impair the genetic expression of kernel arrangement. In the case of a genetically rowed type, the spaces left vacant by undeveloped kernels will tend to be filled by those remaining. This probably gives rise to the physiologically irregular type of East and Hayes. 1 The genetically pure zigzag type of kernel arrangement is modified in a similar manner. The kernels likewise spread into the vacant spaces and thus lose their "shoepeg" form, but more serious still the lower flower of the spikelet often fails to develop in scattered areas, thus giving the ear a partially rowed appearance. These and other modi- fications which obscure the genetic expression of rowing lead to errors IXHKKITANCK OK IvKHNKI. AliK A Nt iKM KN T IN S \\KKF CliKN 309 in classifying the individuals in a given progeny. Such errors become cumulative within a large population of numerous progenies and are conlined mainly to the rowed classes as will he shown later. It is often impossible to properly classify inbred strain.- corn because their weakness leads to the indeterminate or anomalous genetic expression of rowing. Abnormalities in cob growth. Mich as fasciations, have a similar effect. ANALYSIS OF THE INHERITANCE OF KERNEL ARRANGEMENT Segregations in the F Progenies The genotypes expected in the F, generation on the basis of the two-factor hypothesis and the proportionate number in each are shown in Table 1. Pi, Pi, Pi, Pi, and Pi, Pi, Pi, pi, could not be classified separately and are assumed to be phenotypically the same. The twenty-two F, progenies mentioned above were >eparated into the eight classes shown and are summarized in Table 4. The agreement between the observed and expected numbers in Table 4 is not close. It will be noted that the observed frequencies in the three-rowed classes are less than the expected. On the other hand. in the four intermediate classo (Table 4) the observed number ex- ceeds the expected. Reference to the individual F, progenies (Tables 16 to 22 inclusive) indicates that the observed frequencies in the classes mentioned vary nearly always in the same direction. Thus the deviations in Table 4 are really due to a series of cumulative errors which are without doubt due to the obscuring effects of non- heritable modifying factors. The segregations in Table 4, in view of the large deviations, do not by themselves .substantiate the two-factor hypothesis for the arrangement of kernels on the ear, but when taken in conjunction with Tables 16 to 22 inclusive it is evident that such an interpre- tation is the one most closely in accord with the facts. In addition to the twenty-two families referred to above. siv F, progenies gave monohybrid ratios in the F, generation. The- summarized in Tables 5, 6, 7, and 8. The progenies in Table 5 give a 3:1 ratio. The dominant pheno- type is distinctly rowed, whereas the recessive is intermediate and more nearly rowed than zigzag. The recessive class in no way re- semble- the true Country Gentleman type. 1 .lions of this type were secured by selling individuals having a Pi, Pi, Pi, pi, genetic composition. Three sel ated in this way. are shown in Tables 16, 17, and 23. In the four above-mentioned tables, the data clearly fit a 3:1 expectancy. 310 BULLETIN Xo. 320 [February, 5 a 1 =: Z x s - r- XS > 1 - = ;- SS gi - : - PN - c M -- ^ t* - &. o - > c. j r- r. -. -s. - >~ r r-i ri r. x x r. x r! -- ii 'Sfi re IT: C: ^ re - r- ^ ~- t 1 1 1 1 1 c. Is ~ ': n :? ^j -^ *i ji ~ ~i r. O II r^ t^ r : ~. - :-) ~ re re t^ re i i i i - - ~. ~ *^i *- *"i re ~~ ~~ . .Z" w - T" ^ =. L^ fi r"i re - r>- t- - . r - > i c. 5. L r . . - i~ re M re ~ ^xr3-e X X x x r r- r. M 1 t^ x II - ^ r. ?i if re ' re - re : 1 1 c. SK ~\ f ~ ~. ~ -- d a > i II .I x T-l 1 --_-- Sffi 1-* I-H II II "x ~ c s ~ t : - r ~ * '~ T. re c: c r r. y^ n ? i ue r i i ~ ue - re re - - ~ - 14 r-t re re "5' = Jj t.~ mM * V ' ' . s .- M >, o ^ ^< 2 S :- = ^3 "3 ~ * / o -* re -e v; x n - ~ - 3S St 3S 3E iJ C x - x /=; ''^~ r" i 1929] INHERITANCE OF KERNEL ARRANGEMENT IN SWEET CORN 311 ft -- ffi -- i - 'Jl Ll . . . 'L. L 'I. H. . . . ' L _ '=-'- . . . ftffi -z. fi'S :- ^- - J-. *3U3f GO ?! ~f. - * i 1 i I D T z. n. ' ^^ V 3 Sffi ; ; iCffi '- ^~ ---- ~ :-. - :: - E t H c Sffi cs _ _ -** G 3 e X ~ -. M C^^H t-1 ul " I - 1 Niiiiilicr Of 1 1-H 1 1 CO J* "z< . 1 i I r- 1 31 -- 312 BULLETIN No. 320 [February, TABLE 6. SUMMARY OF THE F 2 PROGENIES FROM SELF-POLLINATED PiipiiPi 2 Pi2 FI PLANTS Cross No. Number of progenies Total PiiPii Pi 2 Pi 2 Piipii Pi 2 Pi 2 piipii Pi 2 Pi 2 1004. . 1 128 34 60 34 1005 1 369 93 185 91 1015 3 567 138 283 146 1022 1 54 13 26 15 Total 6 1118 278 554 286 Expected. . . Deviation. . . 1118 279.5 -1.5 559.0 -5.0 279.5 6.5 x 2 = .2039 P > .6065 1 1 Values of P are not given in Elderton's tables when x 2 is less than 1. value of P is .6065 when x 2 = 1 -0000. The TABLE 7. SUMMARY OF THE F 2 PROGENIES FROM SELF-POLLINATED Piipiipi 2 pi 2 FI PLANTS Cross No. Number of progenies Total PjiFii Pi 2 pl2 Piipii pi 2 pi 2 piipii pi 2 pi 2 1002 2 299 64 154 81 Expected. . . Deviation. . . 299 74.8 -10.8 149.5 4.5 74.8 6.2 X 2 = 2.2087 P = .3377 TABLE 8. SUMMARY OF THE F 2 PROGENIES FROM SELF-POLLINATED piipiiPi 2 pi 2 FI PLANTS Cross No. Number of progenies Total piipii Pi 2 Pi 2 piipii Pi 2 pi 2 piipii pi 2 pi 2 1008 5 684 173 349 162 Expected. . . Deviation. . . 684 171.0 2.0 342.0 7.0 171.0 -9.0 X 2 = .6404 P > .6065 The F 2 progenies shown in Table 6 segregate in a 1:2:1 ratio, only 25 percent being distinctly rowed. The recessives are intermediate but differ from those in Table 5, inasmuch as they are more nearly zigzag than rowed. They do not have the true zigzag arrangement, however. The data in Table 6 are the total of six selfs on plants having a genetic composition of Pi x pi x Pi 2 Pi 2 . The data of individual F 2 progenies are shown in Tables 17, 18, 22, and 23. INHERITANCE OF KKRM :;MKNT IN S \VKKT Cmcx 313 The data in the five tables mentioned above agree fairly well with the expectancy based on a 1:2:1 ratio. In Tables 7 and 8 the F, recessive classes were identical, being true zigzag. The dominant classes, however, bore no > blance to each other, indicating that they were segregating for dii: factors. F 2 segregations of the type shown in Table 7 could have arisen only by selfing plants with a genetic composition of Pi, pi, pi, pi... Likewise, the F 2 segregations in Table 8 were the result of selfing plants with a pi,, pi x PL pi, factorial composition. In both Tables 7 and 8 the fit is fairly close to expectancy for 1:2:1 ratios. The data of the individual progenies are given in Table 24 ' 10021 and in Table 2~ X 10081. The F, segregations in Tables 5, 6, 7, and 8 can be best explainer! by assuming that kernel arrangement is due to two factors. That these are by no means equal is shown by comparing the progenies in Tables 5 and 8, both of which segregate for the factor Pi, pi,. The factors Pi : P^ or Pi 1 pi l must be present in order to produce rowing. The progenies in Tables 6 and 7 are segregating for the Pi r pi l factor. The expression of rowing in a genotype Pi, Pi, pi, pi, (Tables 5 and 7) is much stronger than in the pi x pi t Pi 2 Pi, genotype in Tables 6 and 8. Accordingly the factor Pi, is more necessary for the complete expression of rowing than the factor Pi 2 . Segregations in the Back-Cross Progenies FJ plants from four of the progenies were successfully back-en to the double recessive parent. Eleven progenies were obtained which without exception segregated into a 1:1:1:1 ratio, as shown in Table 9. The data for the individual families are given in Tables 26 to 29 inclusive. In the five above-mentioned tables the data clearly fit an expected 1:1:1:1 ratio. Segregations in the F 3 Progenies A large number of selfs were made on the F, progenies. The F 3 plants had greatly decreased vigor. This, combined with an unfavor- able season in 1927, gave rise to low yields and caused the 1< many progenies. In addition, many of the ears were poorly filled, which fact made it difficult to classify them, especially in the rowed classes (Groups I and II i . Owing to such conditions the obscuring effect which has been mentioned previously came into play, causing a deficiency in the rowed classes shown in Table 10. By referring to the individual families in Tables 30 and 31, it will be found that the deficiencies are cumulative. In Table 32 one F 3 progeny shows a deficiency in the Pi x Pi x Pi 2 Pi 2 class, whereas the next two classes are slightly in ex- cess of the expected. As these two phis deviations are small, they are of little importance. 314 BULLETIN No. 320 [February, a O >H SJ H fc t> O O w K H w O h K Q PQ O PH O K O '&,'&, GO CO CD t- T 1 Tf O5 1C T-H 1 1 1 1C GO IN T-H COIN COCO oooo Number of progenies I-H IN COIN T-H o o o co * ic GO oooo T I T 1 t-H T 1 "o x PH 'a'S, '5,'S. - T-H i-H 'S.S * OO5 1-H T 1 qo COC35 * COIN : i i-HiCCO 1-H CIC IN rl S'ta i-H O Tt< C<1 rH 00 * O5 1C Tf C GO 1-H 1-H T-H "o x HWP - 1929} IMIKIUTANCE OF KERNEL ARRANGEMENT IN SWEET CORN 315 '5/5. e, c. ','. '5.S ' B 'S.E z 1 'Sffi ciT 'S/3. 1 '5. M E'S. ' ' E E a K Pn'S. S'S. 10 cooo rH 1-H CM COM Tf OiO t* 3 ss o - U 02 c.^, s i PH?H fa 00 w s H O S's, ss * CCU5 (NO-* ? c o fe H ss ss in 10 i i-t j-4 1 1 fa O 5 ^K "a H O5 to cc 5 ^ O i ( 0000 oo CM CM I : 11. SUMM Number of progenies tH f-l W r}t . a H & O III Total Expected . Deviation. : : : I CM oo *CM ^ : : : ss 00 CO 1 EE SS O5 H CO u 1015 Expected Deviation 316 BULLETIN Xo. 320 [February, TABLE 13. SUMMARY OF THE F 3 PROGENIES FROM SELF-POLLINATED Piipiipi 2 pi2 F2 PLANTS Cross No. Number of progenies Total PiiPii pi 2 pi 2 Piipii pi 2 pi 2 piipii pizpij 1003.. 2 81 19 48 14 1004 2 81 21 41 19 1005 2 130 33 72 25 1018 1 60 15 36 9 Total 7 352 88 197 67 Expected. . . Deviation. . . 352 88 176 21 88 -21 x 2 = 7.5171 P = .0240 TABLE 14. SUMMARY OF THE F 3 PROGENIES FROM SELF-POLLINATED piipiiPi 2 pi 2 F 2 PLANTS Cross No. Number of progenies Total piipii Pi 2 Pi 2 piipii Pi 2 piu piipii pi 2 pi 2 1004 3 214 60 97 57 1005 1 57 13 31 13 1018 3 235 60 112 63 Total 7 506 133 240 133 Expected . . . Deviation. . . 506 126.5 6.5 253.0 -13.0 126.5 6.5 x 2 = 1.3360 P = .5263 TABLE 15. SUMMARY OF THE F 3 PROGENIES FROM HOMOZYGOUS F 2 PLANTS Cross No.. Number of progenies Total PiiPii Pi 2 Pi 2 PiiPii pi 2 pi 2 fiipii i 2 Pi 2 1003. . 1 46 46 1004 2 112 112 1010 2 63 63 1003 2 85 85 1004 1 65 65 1005 1 70 70 1006 2 135 135 1018 1 87 87 1006 1 76 76 Monohybrid ratios such as were obtained in the F 2 also appeared in the F 3 generation. The summaries of these segregations are given in Tables 11, 12, 13, and 14. In Tables 11, 12, and 14 the fit is fairly close to expectancy but in Table 13 the deviations are somewhat 1929] INHERITANCE OF KKKNKI. AKK\N..KMK\T IN S\\ KI 317 FIG. 1. THE Fi PROI.KNY FROM A CK.'!'\ INHERITANCE OK KKKNKI. AKK\N<,KMKNT i\ > 319 FiQ.9 320 BULLETIN Xo. 320 [February, KK-TLTS OF CROSSES BETWEEN COUNTKV GENTLEMAN AND NARROW GRAIN EVERGREEN SWEET CORN SHOWN IN FIGS. 2 TO 9 Fig. 2. Distinctly rowed F- ears. The genetic composition is either Pi, Pi, Pi- Pi- or Pi, Pi, Pi 2 pi 2 . This type has been desig- nated us ( Iroup I. Fig. 3. Less distinctly rowed F- ears with rows continuous. Those form part of Group II. A- the rowing is continuous, the genetic composition assigned is Pi, pit Pi? Pi-. 1. I.e.-s distinctly rowed F- ears with rows not continuous. This type is less distinctly rowed than thai shown in Fig. 2. It falls in Group II along with Fig. 3 but differs from P"ig. 3 in that the rowing is not continuous. The assigned genetic composition is Pi, pi, Pi-., pi-. Fig. 5. Intermediate F- ears. This type of F 2 segregate falls into Group III-A. The kernel arrangement is intermediate but more nearly rowed than zigzag. The rowing is continuous. The assigned genetic composition is Pi, Pi, pi 2 pi 2 . Fig. 6. Intermediate F- ears. This type also falls into Group III-A. but it differs from Fig. 5 in that the rowing is not contin- uoiis. The a itiiH-d genetic 'Composition is Pi, pi, pi- pi-. Fig. 7. Interim-dial i I > are. This type belongs in Group III-B. The kernel arrangement differs from the types in Figs. 5 and 6 by being more nearly zigzag than rowed. The slight amount of rowing which appears is confined to butt or tip. The genetic composition is pi, pi. Pi- Pi-. Fig. 8. Intermediate F 2 ears. This type is also classified under Group III-B. It differs from Fig. 7 in the amount of rowing. only a slight trace appearing here. The genetic composition is pi, pi, Pi- Pi 2 . Fig. 9. Zigzag F- ears. This type has a true Country Gentleman zig- zag arrangement of kernels and belongs in Group IV. The genetic composition is pi, pi, pi 2 pi 2 . The ear at the right shows no trace of rowing but that at the left is somewhat doubtful. 19S9] INHERITANCE OF KERNEL ARRANCKMEXT IN S \VKKT 321 a 'a ' 'a co o a O S o s O w 'an. 'aa 1-H .32 'a O5 O <* CM 00 O CO ^H CM ^H 1-H CO ^H 00 CM 1-H O T 1 II O5OO5 OCO-* O i-H 1-H 1-H Tj* CO ^H- O O O JH 'a' 5. EE GO CO CO O CM CM ^H CO * CM CO I-H CO CO IOIO CM CM ffiffi" GO i 1 Tji O O rH t^- 1-H CM 1-H 1-H i-H GO GO CM -t 1 C^J O O -H "* 1 E'S. (N 00 CM * CO Tf CD i-H i-H * t^ co 11 3. t>- O t^- 1C CD ^^ O CO CD O ^H 10 ^ CM QOQO co co H 11 'S, t- <**' P^. 6 6 6 6 6 o ptj 1 be I a g>c> ffifc l- O5 ^-c CM CM CM C>4 1 1 1 1 1 1 1 CO CO CO CO CO CO CO 111 -1 gs sa . _- - i . - - - - ;l - . ' ' ' - - - _ 324 BULLETIN No. 320 [February, ffi s s XI '& '. 'S, 3 o i i o O O O a PH '.', i-H iH Tj< CD COOCO (N i-H '5/a GO -H 1-H(N OS CDC^l CO CO li 1-H 1-H *CD '=.'5. 2S O5OS CD CO | co co I E'a OH GC M 1-H * CO OGO^H 'c-'S. CO O coco 0000 ce re o OS 1C ss co >c 1-H? o O S o a u i-H (N H s 'S-'E. 'a 3. Tj '^ '2-2: 1C OC5 1-H 1 ( -H GO O 00 00 O5 CO O '^'^ ss 3 CC Fi Parent genotype 'S, S"d"7 *33 PH >, ' a M O i, "7- ft K O1 d ^ 1 1 1 GO CO 00 i i i i i i O O O Total Expected. . . . Deviation. . . . INHKKITAXCE OF KKRNKI. AKKANUKMKN i IN SUKKT CHUN 325 I '.', '3/5. T^IC^Tt< CO-* -r '7 I 1 1 1 : 'f.'5. '. t^ ooooo ~H C5 CO < 1 -H ccco 'as '~ '-^ * * COTj* CO >O CC M 1 1 1 1 '.'5. n ss rt M ^J CO 1 1 1 1 1 1 I 1 (N C^ re ?! lOO | 'l=r' (Cffi oo i> t - /: . 1 -J -H co ro | ''=. S ^ re -< c: t^l^ OcO I ' KS 3 g (M ic c: re co ec '" -: c; ~ * Tj< (N(M Fi Parent gcnol \ !> '- o 6 6 ^332- ffi ^. a -tiH b- O O oo -CLASSIFICATION FI Parent genotype 'a : _o_d 1 PS Total PhPii pispij 1'iipii piipii piipii pipi 1002-4 . . Pi,pi,pi 2 pij 156 39 75 A<> 1002-5 (do) 143 25 79 39 Total 299 64 1 54. 81 Expected 299 74 8 14Q 5 74 8 Deviation -10.8 4.5 6.2 x* = 2.2087 P = .3377 TABLE 25. CLASSIFICATION OF THE EARS iv FIVE F 2 PROGENIES FROM CROSS 1008 (pitpiipi 2 pi2 X PiipiiPi 2 pi?) Fi Progeny Xo. F, Parent genotype Total piipii PUPii piipii Piipi, piipii pijpij 1008-5. . . piipiiPi2pij 135 34 29 1008-9 (do.)" 160 44 76 40 1008-14 (do.) 156 38 79 39 1008-15 (do.) 162 38 85 39 1008-16 (do.) 71 19 37 15 Total . 684 173 349 162 Expected 684 171 342 171 Deviation 2 7 -9 X 2 = .6404 P> .6065 TABLE 26. CLASSIFICATION OF THE EARS IN A PROGENY OBTAINED FROM A BACK CROSS BETWEEN AN Fi PLANT OF CROSS 1003 AND COUNTRY GENTLEMAN Back cross progeny X". Total Piipii Pizpii Pi:pii pi 2 pit piipii Pi,pii piipii pizpij 1003B 79 24 19 18 18 Expected 79 19.8 19 8 19 8 ' 19.8 Deviation 4.2 -.8 -1.8 -1.8 x - = 1.2504 .7440 TABLE 27.- CLASSIFICATION OF THE EARS IN Two PROGENIES OBTAINED FROM BACK CROSSES BETWEEN F; PLANTS OF CROSS 1004 AND COUNTRY GENTLEMAN Back cross progeny Xo. Total Piipii Pitpij Piipii pijpij piipii Pijpii piipii pijpij 1004C.. 85 17 20 :, 23 1004D 90 21 23 26 20 Total 175 38 43 51 43 Expected 175 43.8 43.8 43.8 43.8 Deviation -5.8 -.8 7.2 -.8 l.'t.xox .57tiS 330 :T THX .'7.^ rx 5 re Pi:.::.i.x:T* '>?: iixiiz TV:*. SACS IUBI-C F E PtASTCS OT GBOBS 1O05 AX0 CoCXTmY GEXTLEMLAX _ i rt. aosc :>::> : :>? E I'>'"F 00 MB 73 a 15 is - i 15 : a 9 N 14 a 16 19 a i* 2'> 13 16 Tbtal r<; n ^ -- ' -. ' Ufl v* 5 1'j 5 M ' P = 39L Cka OP TUB EAZS ix Tvo PkoesEms OKXJOXED FBOM KICK CBOHEES BmrnmA F : PSJLSTS or CBOSS 101 S \~ ! '_';7>"T5.T rT V^ T'V < V .^^.l^ T ::*: a". t ~ i" .li'Tt f^,' fejj: ^ H :>B- ntoc a . 31 CO a 7 IUtf --, -2^0 ii: 4 -40 55 S3 2-0 -- .v=; r< 4.0 BqiertBi Dtontia m P> B65 cc :^ can be best explained on the baas of the two-factor hypothesis. The F, For the srgrpgation of: ,: Tables 32, 33, and 34 !.: Table 35 Pfi a pi, pL pi-i : Tables 30. 31, 32, and 33 pi^ pij PLpL : Tables 31. 32, and 33 Om ttiue batii* -lK-l. I'ixpcch-d . . Deviation. . 213 IN CO | (NM 1^ 1-H 1 ue 10 / ri M . . . 2 '.- * ~ - 8 II II A PH PH PH 1 1 (N i-l +1 co (N CO ^ CC rt< i-H c^ II II f "S II "x II OO 1 i"H ^ oo re re l-H i-H SS: g 'o. H S : '^ sS Ci 'S. '5. 'S. 'o, '. 7! ?t O ^H GO E'S, <* 05 i> ~ N II a & 1 EE Q^ ^1 Q E 03 H _ M S.E o EE CD H 02 PH ^ E'B, (M H^ +1 &H PH PH (N CD 02 ICI> 5 S EE EE co co I s - t- II _o 1C 00 TjH^H 3 ' W > w 1 "o i-H Q 1C 00 ^ ^H h H 1 1 T-H O SB o II 'a S- E w Cl KS E^- 02 r M Kn ~-< P4 2 H-l D E E u 1 & CO S ^ 8 ^ w ~a f-l(N PM (N (N 3 (N(N 22 __, O 3S. 00 00 1 1 1 1 O X HW 00 1 ( 1-H 334 BULLETIN No. 320 [February, ', a H 55 PQ s - c. C, . i _ S.S, "^ X H CXPL| 'S.S '3, iM OS CO i 1 1 1 1 i (NOO * Cl^H * * - ?r ^ *C^ " "^ ** PH D. PH O. ^ N a. a. V" 1 V" 1 S 'i'S KS re ccx OJ CO (M o>o -*xj< 1 oooo 1 A PH - *, ^ =. as f^iP-i co i i i i SS ss !M t^ 01 > M d 9g ^ & T^T O C2 C2 r-l i 1 ,-H 1 1 1 lO 10 1C i 1 i-H I I ooo I 1 1 1 1 ill 111 HWQ -t e. '2,'S, s _ ^ S.S, a as a 3, .r-- 4 H Is -co C M c 9r% M* _-* Pn'S, D Pn'S, "3 -O CO !> n /2 Pn'3. 3 - J> - a 0.0. 3 ~* 1^4 S PnPH BI d ^ e. 2; S, 4 3 3 SS SJ H H pTS, a 3 *4 Sffi : : : Q H 5 ss 3 s ss J 4 s 1 H Sgg 3 5 3 -1 -I -J H 5! .> H Q 3 F 2 Parent genotype if ) 3 a .^ 4 a < be g H PH CO-* CO fe 1 1 1 1 1 I 1 1 1 1 J 1 1 1 1 ( 1 ( 1029] INHERITANCE OF KERNEL ARRANGEMENT IN SWEET CORN 335 that the latter, as a rule, produce more vigorous seedlings and larger plants altho the average number of days to maturity is about the same. Hybrids grown from crosses between selfed strains of Coun- try Gentleman and Narrow Grain Evergreen are usually more vigor- ous than Country Gentleman intravarietal crosses. On the other hand, crosses between Narrow Grain Evergreen selfed strains are more vigorous than either. Repeated observations of this kind indicate that the rowed arrangement of kernels is associated with more vigor- ous growth and larger gross yields than the zigzag arrangement. It is not improbable, therefore, that the double recessive, zigzag kernel arrangement is associated with one or more plant characters which may segregate in a like manner. PRACTICAL ASPECTS OF THE INHERITANCE OF ROWING The inheritance of rowing is of particular interest to the breeder of Country Gentleman sweet corn. Since the zigzag character is a double recessive, a considerable percentage of rowed ears is bound to reappear each year in open-pollinated cultures. Most of these rowed ears will probably fall within Class III-B and a few possibly within Class III-A in cultures which have been carefully selected for a number of years. In commercial strains the range of segregation will usually be much wider. It is very doubtful whether the breeder is justified in selecting only individuals of the p^ pi t pi 2 pi 2 phenotype. If the true zigzag arrange- ment is unduly emphasized, there is a possibility of reducing yields thru the inbreeding effect of close selection. The presence of pheno- types of pi x pi x Pi 2 Pi 2 and pi x pi Pi 2 pi 2 composition is not objec- tionable from the commercial viewpoint. It is barely possible that all ears falling within Groups III and IV might be shelled together advantageously thus maintaining the culture in a heterozygous con- dition. Altho the evidence available is inconclusive, nevertheless the slight amount of rowing thus introduced may be associated with in- creased plant growth and better seedling vigor. For the breeder of rowed varieties of sweet corn, the elimination of all but slight irregularities in rowing is a relatively simple matter owing to the incomplete dominance of rowing. It is probable that in spite of continued selection such genotypes as P^ Pi x Pi 2 pi 2 and Pi t pij Pi 2 Pi 2 will persist, but their presence does not detract from the value of the strain. 336 BULLETIN No. 320 [February, SUMMARY AND CONCLUSIONS 1. The rowed kernel arrangement in sweet corn is incompletely dominant over the zigzag arrangement, as shown by Fj progenies. 2. Dihybrid segregations into eight classes in the F 2 generation establish the presence of two factors for rowing. Pi, and Pi 2 . 3. This hypothesis is supported by monohybrid segregations in the F, and F 3 generations, of which no single progeny included both the distinctly rowed and the zigzag types. 4. Back crosses to the zigzag parent segregated into 1 Pi x . pi x Pi 2 pi 2 : 1 Pi x pii pi 2 pi,: 1 pi x p^ Pi, pi,: 1 p^ pi! pi, pi,. 5. Certain F 3 progenies proved homozygous for the intermediate types Pi x Pii pi, pi, and pi x pi x Pi, Pi,. LITERATURE CITED 1. EAST, E. M., AND HAYES, H. K. Inheritance in maize. Conn. Agr. Exp. Sta. Bill. 167, 1-142. 1911. 2. KMERSOX. R. A. The inheritance of certain "abnormalities" in maize. Rpt. Amer. Breeders Assoc. 8, 385-399. 1912. 3. HALSTED, BYRON D., AND OWEN, EARL J. Report of the botanist. 27th Ann. Rpt. N. J. State Agr. Exp. Sta. 369-514. 1906. 4. KEMPTON, J. H. Floral abnormalities in maize. U. S. Dept. Agr. Bur. Plant Indus. Bui. 278. 1913. 5. PEARSON, KARL. Tables for statisticians and biometricians. Cambridge, Eng- land. 1914. 6. STEWART, ALBAX. The pistillate spikelet in Zea mays. Science, n. s. 42, 694. 1915. 7. STRATTON, MILDRED E. The morphology of the double kernel in Zea mays var. polysperma. Cornell Univ. Agr. Exp. Sta. Mem. 69, 1-18. 1923. 8. WEATHERWAX, PAUL. Morphology of the flowers of Zea mays. Torrey Bot. Club. Bui. 43, 127-144. 1916. 9. WEATHERWAX, PAUL. The morphological basis of some experimental work with maize. Amer. Nat. 53, 269-272. 1919. "ERSITYOFILLINOIS-URBANA