heterozygous parents have been used to map QTLs in potato (Leonards-Schippers et al. 1994), loblolly pine (Groover et al. 1994) and Eucalyptus spp. (Grattapaglia et al. 1995). The possibility of using F, crosses to map QTLs in alfalfa has yet to be investigated. The previously published maps of alfalfa covered 659 cM (Kiss et al. 1993), 467 cM (Brummer et al. 1993), and 553 and 603 cM (Echt et al. 1994). The total lengths of the two maps obtained in this study suggest that the RFLP markers identified here cover only part of the alfalfa genome. However, the linkage information for PG-F9 together with the knowledge of linkage phase for alleles at these loci, will be useful for further studies aimed at verifying the mechanisms of 2n egg production in the PG-F9 mutant. From the Dlpartimento dl Blotecnologle Agrarle ed Amblentali, Unlverslta degll Studl di Ancona, Via Brecce Blanche, 60131 Ancona, Italy (Tavolettl and Veronesl) and the Department of Agronomy, University of Wisconsin, Madison, Wisconsin (Osborn). Research was conducted at the Department of Agronomy, University of Wisconsin, Madison, and partially supported by the Italian Ministry of Agriculture, Special Project on Forage Crops, Director Prof. P. Rotlll. Special thanks are due to Dr. Kim Kldwell and Doug Brouwer lor their assistance during the RFLP analysis. The authors also wish to thank the College of Agricultural and Life Science, University of Wisconsin. The Journal of Heredity 1996.87(2)

Kldwell KK and Osborn TC, 1993. Variation among alfalfa somaclones In copy number of repeated DNA sequences. Genome 36506-912. Kiss GB, Csanadl G, Kalman K, Kalo P, and Okresz L, 1993. Construction of a basic genetic map for alfalfa using RFLP, RAPD, Isozyme and morphological markers. Mol Gen Genet 238.129-137. Kublslak TL, Nelson CD, Nance WL, and Stlne M, 1995. RAPD linkage mapping In a longleaf pine X slash pine F, family. Theor Appl Genet 90:1119-1127. Lander ES, Green P, Abramson J, Barlow A, Daly MJ, Lincoln SE, and Newburg L, 1987. MAPMAKER; an interactive computer package for constructing primary genetic linkage maps of experimental and natural populations. Genomics 1:174-181. Leonards-Schippers C, Gleffers W, Schafer-Pregl R, Rltter E, Knapp SJ, Salamlnl F, and Gebhardt C, 1994. Quantitative resistance to Phytophthora infestans In potato' a case study for QTL mapping in an allogamous plant species. Genetics 137:67-77. Lincoln S, Daly M, and Lander E, 1992 Constructing genetic maps with MAPMAKER/EXP 3.0, 3rd ed. Cambridge, Massachusetts: Whltehead Institute. Liu Z and Furnler GR, 1993 Inheritance and linkage of allozymes and restriction fragment length polymorphisms In trembling aspen. J Hered 84:419-424 McCoy TJ and Blngham ET, 1988 Cytology and cytogenetics of alfalfa. In: Alfalfa and alfalfa Improvement (Hanson AA, Barnes DK, and Hill RR Jr., eds). Monograph 29. Madison, Wisconsin: American Society of Agronomy. Ott J, 1985. Analysis of human genetic linkage. Baltimore, Maryland: Johns Hopkins University Press; 104. Plomlon C, O'Malley DM, and Durel CE, 1995. Genomlc analysis in maritime pine (Pinus pinaster). Comparison of two RAPD maps using selfed and open-pollinated seeds of the same Individual. Theor Appl Genet 90: 1028-1034. Tavoletti S, 1994. Cytologlcal mechanisms of 2n egg formation In a diplold genotype of Medicago saliva subsp. falcata Euphytica 75:1-8.

Reference* Blngham ET, 1991. Registration of Isogenlc populations of diplold and tetraplold alfalfa, W2xlso-1 and W4xlso1. Crop Scl 31:496. Brummer EC, Bouton JH, and Kochert G, 1993. Development of an RFLP map In diplold alfalfa. Theor Appl Genet 86329-332. Echt CS, Kldwell KK, Knapp SJ, Osborn TC, and McCoy TJ, 1994. Linkage mapping In diplold alfalfa (Medicago saliva). Genome 37:61-71. Felnberg AP and Vogelsteln B, 1983. A technique for radlolabelHng DNA restriction endonuclease fragments to high specific activity. Anal Blochem 132:6-13. Grattapaglia D, Bertoluccl FL, and Sederoff RR, 1995. Genetic mapping of QTLs controlling vegetative propagation in Eucalyptus grandis and £ urophylla using a pseudo-testcross strategy and RAPD markers. Theor Appl Genet 90-933-947. Grattapaglia D and Sederoff R, 1994. Genetic linkage maps of Eucalyptus grandis and Eucalyptus urophylla using a pseudo-test-cross: mapping strategy and RAPD markers. Genetics 137:1121-1137. Groover A, Devey M, Fiddler T, Lee J, Megraw R, Mltchel-Olds T, Sherman B. Vujclc S. Williams C, and Neatle D, 1994. Identification of quantitative trait loci Influencing wood specific gravity In an outbred pedigree of loblolly pine. Genetics 138:1293-1300. Hemmat M, Weeden NF, Manganaris AG, and Lawson DM, 1994. Molecular linkage map for apple. J Hered 85: 4-11. Kldwell KK and Osborn TC, 1992. Simple plant DNA Isolation procedures. In: Plant genomes: methods for genetic and physical mapping (Beckmann J and Osborn TC, eds). Dordrecht, Netherlands: Kluwer Academic; 113.

1 7 0 The Journal of Heredity 199687(2)

Veronesl F, Marianl A, and Tavoletti S, 1988. Screening for 2n gamete producers In diplold species of the genus Medicago. Genet Agrar 42:187-200. Received May 1, 1995 Accepted November 8, 1995 Corresponding Editor Norman F. Weeden

Linkage of Semidwarf Phenotype to Interchange Homozygosity in Pearl Millet K. Uma Devi, P. S. R. L. Naraslnga Rao, and M. Krishna Rao In pearl millet [Pennisetum glaucum (L.) R. Br. = P. typhoides Burm. S&H; P. americanum(L.) K. Schum], an interchange homozygote bred true for semidwarf phenotype. It showed monogenic recessive pattern of inheritance in sib crosses; the inheritance pattern was examined in nonsib crosses using two unrelated inbred lines as female parents. Far fewer semidwarfs than the 25% realized in the sib crosses were recovered in the F2's of the nonsib crosses Tall interchange homozygotes

were recognized in the F2 based on the analysis of selfed and crossed progeny of the tall structural homozygotes segregating in the F2. The results appeared best explained by considering that the dwarf phenotype is controlled by duplicate genes where one of the genes was close to the interchange breakpoint. The observed difference in the segregation pattern of the dwarf phenotype in sib and nonsib crosses would be attributed to a difference in the gene pair located away from the interchange breakpoint both the parents in the sib crosses were homozygous recessive at this locus, whereas the nonsib parent is homozygous dominant In pearl millet, a semidwarf mutant homozygous for chromosomal interchange 36 has been described (Koduru and Krishna Rao 1984). The dwarf phenotype served as a marker for the interchange. There was no evidence of gametic or zygotic selection against the interchanged chromosomes in crosses of the dwarf interchange homozygote with the tall plants of the inbred from which it was isolated (sib mating). Such an interchange stock would be useful in linkage studies should the pattern persist in different genetic backgrounds. Therefore, nonsib crosses were carried out.

Materials and Methods The dwarf interchange homozygote line carries the accession number IP 12781 at ICRISAT (Patancheru 502324 India). It was isolated as a segregant in the inbred line IP457 with a normal tall phenotype. The line IP 12781 was crossed as female with IP 457 (its tall sib). The F2 of this sib cross was examined to see if the inheritance pattern of the semidwarf phenotype conformed to the 3:1 ratio observed by Koduru and Krishna Rao (1984) nine generations earlier. The two nonsib lines used were Vg 272 and IP 482. For convenience in crossing, genie male sterile segregants in these two lines were chosen for females. Individuals within F2 families were classified as (1) tall or dwarf, (2) interchange heterozygote or structural homozygote (for interchanged or normal chromosomes), and (3) male fertile or male sterile. The interchange heterozygotes have sparse seed set (—50%) on open pollinated ears, while the structural homozygotes have full seed set (>90%). Hence, they were classified as semisteriles (Sst) and fertiles (Fr), respectively. The chromo-

Table 1. Segregation for the semidwarf phenotype associated with homozygoslty for an interchange in sib and non»ib matlngs involving the 3-6 interchange homozygote in pearl millet

Tall

Cross X dw Interchange homozygote

Sib crosses IP 457 X IP 12781IP 457 X IP 12781* Nonslb crosses Vg 272 (mjms) X IP 12781' Selfing of tall Fr Fj plant obtained from the above cross

IP 482 (msms) ControlVg 272 (Inbred)

X

IP 12781'

Family

Germination percentage

Seeding survival percentage

53-62 1 2 3 4

Sst

Fr

Ratio

X1

1.07 2.77

355 94

332 89

1:1 1:1

0.48 0.14

15:1

0.0002

556

533

1:1

0.48

— 15:1 15:1

003 0.21

151

1.13

105

1:1

0.72

—

—

—

—

—

dw.

Ratio

527 147

160 36

3.1 3:1

85-99

1,021

68

80 66 60 64 182

0

F, F, F, F,

X1

Tails

—

F,

49-63

83-90

—

48-65

78-100

—

4 1 0 16 —

—

93 —

" Data from Koduru and Krishna Rao (1984). • Cross performed 15 generations after the mutant has been Isolated along with other nonslb crosses (The dwarf Interchange homozygote now given the accession number IP 12781 was isolated from the Inbred IP 457 In 1975." and * therefore represent sib crosses.) ' Pooled data from six F, families in two seasons. ' Pooled data from two replicates of an F, family in two seasons. • For comparison of germination and seedling survival percentage. Fr = Plant with >9O5fc seed set (structural homozygote' homozygous for normal or interchanged chromosomes); Sst - Semlsterile plant with ~ 5056 seed set (Interchange heterozygotes).

somal status of the F2 plants as inferred from seed set was confirmed by meiotic examination of pollen mother cells (PMC) in random samples of the F2 male fertile plants. The classification into male fertiles and male steriles was based on anther phenotype: the male steriles have thin, white, or yellow undehisced anthers (Krishna Rao and Uma Devi 1983), and the male fertiles, including the Sst (interchange heterozygotes), have normal plump dehiscing anthers. The proportion of dwarf plants (which would indicate interchange homozygosity) in F2 families resulting from nonsib crosses was found to be much less than the 25% observed in the sib crosses made by Koduru and Krishna Rao (1984). Burton and Fortson (1966) noted that the expression of dwarf phenotype in some pearl millet genotypes is influenced by the genetic background. To check if the underrepresentation of the dwarf interchange homozygote class in F2 families from nonsib crosses was due to failure of expression of the dwarf phenotype, 10 tall Fr (structural homozygote) F2 plants were crossed to the plants of the inbred line Vg 272 with normal chromosome complement. The progeny of these crosses were cytologicaJly examined (were the tall parent an interchange homozygote, then all its progeny from the cross would be interchange heterozygotes showing association of 3 or 4 chromosomes in PMC). The selfed progeny of four tall Fr plants (F3) were also examined.

The percentage germination of seed and percentage survival of seedlings in F2 families from nonsib crosses were compared with the values from an inbred line (control) to check if embryo lethality (expressed as failure to germinate) or seedling mortality contributed to the underrepresentation of the dwarf interchange homozygote class. The joint segregation of male sterility and interchange was calculated to test if linkage existed between the male sterility gene and the interchange breakpoint. Results The inheritance pattern of the semidwarf phenotype did notfitthe 3:1 ratio as it did in the sib crosses. The frequency of the dwarf plants fell far short of the expected 25% (Table 1). The ratio of the Sst (interchange heterozygotes) to Fr (structural homozygotes) in the F2 was 1:1 (Table 1). The germination and seedling survival percentages of these F2 families were similar to the values in the control (Table 1). The decrease in the frequency of interchange homozygotes (as judged from the dwarf phenotype) is thus only apparent. If there was a real decrease in the number of interchange homozygotes, the 1:1 ratio of Sst to Fr would not have been realized. The segregation pattern of the dwarf phenotype in the F2 from nonsib crosses had a good fit to 15:1 ratio, indicating a two gene (duplicate factor) control for plant height (Table 1). In such a case, tall interchange homozygotes are expected in

the F2. The progeny analysis of the tall Fr (structural homozygote) plants did bear this out. Two of the four selfed progeny of the tall Fr F2 plants (F3) that were grown out had a 15:1 segregation for tall to semidwarf (Table 1). One of the 10 families of the crossed progeny of the tall Fr F2 plants to inbred Vg 272 that were grown out and analysed consisted of plants that all had interchange chromosome configurations in PMCs. The tall Fr plant involved in this cross and the two tall Fr plants that had 15:1 segregation for dwarf phenotype are thus interchange homozygotes but with a tall phenotype. The joint segregation of male sterility and interchange in crosses involving both Vg 272 and IP 482 revealed the absence of linkage between the male sterility gene and the interchange breakpoint (Table 2). The male steriles segregating in the inbreds Vg 272 and IP 482 have different cytological expression but the male sterility genes in both mutants are allelic (Krishna Rao and Uma Devi 1989). Discussion The results indicate that unlike interchange homozygotes in the source genotype, interchange homozygotes in the nonsib crosses were not always dwarf; tall interchange homozygotes were also realized. Two possible explanations are considered in the following: 1. The dwarf phenotype of the interchange homozygote has been attributed to a dwarfing gene very closely linked to

Brief Commurocations 1 7 1

Table 2. Linkage of the male sterility gene (m») to the breakpoint of the 3-6 Interchange (t) In pearl millet Chromosome and genetic

Cross

of F,

Vg 272 NN msms

NN or It Msfit Ms-

NN or tt msms Nt msms Segregation ratio 3:1 (Ms-jnsms) 1:1 (NN + tt:Nt) 3:3:1:1*

X

IP 12781- tt MtMs

IP 482 NN msms

x

149 148 38 65

39 29 8 13

0.12 1.68 7.41

0.09 0.28 2.71

IP 12781 tt Msms

• Pooled data of two F, families. 6 For NN or tt Ms-JitMs-:NN or tt msms-.Nlmsms. N = Normal chromosome; t = Interchanged chromosome Monogenic pattern of Inheritance of male sterility was noted In some of the families; these alone were employed In the linkage test.

the interchange breakpoint allowing no crossing over in the region between the gene and the interchange breakpoint (Koduru and Krishna Rao 1984). If that were the case, (a) the two events, that is mutation of the gene (controlling height) and the interchange (chromosomal mutation), could have occurred independently with the mutated gene being close to the interchange breakpoint; (b) alternatively, the gene mutation could be a consequence of chromosomal mutation, assuming a twoin-one event. If chromosome breakage occurred within or just adjacent to the gene (controlling height), its reattachment to the broken end could take place after damage (e.g., deficiency; Gustaffson et al. 1971), resulting in the origin of a mutant allele, a dwarfing gene. Whatever the origin of the dwarfing gene, its expression (in pearl millet) may be subject to the influence of the background genotype (Burton and Fortson 1966). The good fit of segregation pattern of the dwarf phenotype in F2 and F3 to a 15:1 ratio indicates duplicate genes with one of the genes being close to the interchange breakpoint. The tall sib plants of interchange homozygote line must have been double recessive (just as the dwarf interchange homozygote itself) at the locus away from the interchange breakpoint on the same or on a different chromosome. Therefore, monogenic segregation for the dwarf phenotype was observed in crosses involving sibs. The tall nonsib male sterile parents involved in the present crosses must have been homozygous dominant at both loci. Hence, the F2 ratio

1 7 2 The Journal of Heredity 1996:87(2)

is 15 tall to 1 dwarf. In such a case, 25% of the F2 population are expected to be interchange homozygotes, but only one-fourth among them would be dwarf and the resttall. The tall interchange homozygotes on selfing are expected to either (a) breed true for the tall phenotype or (b) show a 15:1 segregation ratio for semidwarfism. Tall Fr plants of both types have been recognized (Table 1). Thus, the results seem to be best explained by considering duplicate factor control of plant height. However, the selfed progeny of the tall interchange heterozygotes (Sst) segregating in the F2 have not been analysed. Half of these F3 families (from interchange heterozygotes) are expected to show a 3:1 segregation for the dwarf phenotype if duplicate genes are involved. 2. Contrary to the idea of a mutant (dwarfing) gene arising in addition to (or as result of) translocation, the dwarf phenotype of the interchange homozygote could also be envisaged as a case of position effect variegation (PEV) with the expression of the locus governing height (unmutated) being affected by closeness to the interchange breakpoint (a situation similar to that observed in Oenothera (Catcheside 1939) and the rosy locus of Drosophila (Rushlow and Chovnick 1984). Position effect of genes controlling nonautonomous cell functions results in reduced expression of the gene instead of variegation (Rushlow and Chovnick 1984). In the present case, a reduced expression of the gene controlling height can lead to semidwarfism in interchange homozygotes. One can also visualize that inter-

change homozygotes arising in nonsib crosses are subject to suppression of position effect (in the altered genetic background) such that the unmutated locus is still proximal to the breakpoint but is derepressed. A dominant gene suppressing the position effect could have been introduced from the tall nonsib parent—hence, the absence of semidwarfism in threefourths of the segregating interchange homozygotes leading to a 15:1 ratio (of tails to semidwarfs) in F2. A modulation of position effect for achaete variegation depending on different parental genotypes was observed in Drosophila (Baker 1968; Noujdin 1944). The nonsib crosses of the interchange homozygote have thus revealed the influence of a second gene on the expression of semidwarf phenotype linked to interchange homozygosity in pearl millet. The second gene may control dwarfing (as described in the first explanation) or reverse position effect (as described in the second explanation). From the Department of Botany, Andhra University, \Tsaldiapatnam 530 003 (A.P), India. We thank the UGC Special Assistance Programme In cytogenetlcs to the Botany Department, A.U , for financial support. The Journal of Heredity 199687(2)

Reference* Baker KW, 1968. Position effect variegation. Adv Genet 14:133-169. Burton GW and Fortson JC, 1966. Inheritance and utilization of five dwarfs In pearl millet (Pennisetum glaucurn) breeding Crop Scl 6:69-72. Catcheside DG, 1939. A position effect In Oenothera. J Genet 38:345-352. Gustaffson A, Hagberg A, Pearson G, and Wlklund K, 1971. Induced mutations and barley Improvement. Theor Appl Genet 41:239-248. Koduru PRK and Krishna Rao M, 1984. Cytogenetlcs of a semidwarf phenotype In pearl millet, Pennisetum amencanum (L) Leeke. Can J Genet Cytol 26:272-278. Krishna Rao M and Uma Devi K, 1983. Variation In expression of genlc male sterility In pearl millet. J Hered 74:34-38. Krishna Rao M and Uma Devi K, 1989. Allellc relationship of four male sterility genes and nucleocytoplasmlc Interactions In the expression of male sterility In pearl millet. Pennisetum americanum (L.) Leeke. Theor Appl Genet 77576-580. Noujdin N, 1944. The regularities of heterochromatln Influence on mosalclsm. J Gen Blol (USSR) 5:357-388. Rushlow CA and Chovnick A, 1984. Heterochromatlc position effect at the rosy locus of Drosophila melanogaster, cytologlcal, genetic and biochemical characterization. Genetics 108589-602. Received January 19, 1995 Accepted August 28, 1995 Corresponding Author Prem P. Jauhar