Aliso: A Journal of Systematic and Evolutionary Botany

Volume 21 | Issue 1 Article 2

2002 Genetic Control of Self-Incompatibility in () Pungens Subsp. Laevis (Madiinae, ) Elizabeth A. Friar Rancho Santa Ana Botanic Garden

Tasha LaDoux Rancho Santa Ana Botanic Garden

Follow this and additional works at: http://scholarship.claremont.edu/aliso Part of the Botany Commons

Recommended Citation Friar, Elizabeth A. and LaDoux, Tasha (2002) "Genetic Control of Self-Incompatibility in Centromadia (Hemizonia) Pungens Subsp. Laevis (Madiinae, Asteraceae)," Aliso: A Journal of Systematic and Evolutionary Botany: Vol. 21: Iss. 1, Article 2. Available at: http://scholarship.claremont.edu/aliso/vol21/iss1/2 Aliso. 21(1). pp. 1-6 © 2002. by The Ran ch o Santa Ana Botanic Garden.C laremo nt. CA 9171 1-3157

GENETIC CONTROL OF SELF-INCOMPATIBILITY IN CENTROMADIA (HEMIZONIA ) PUNGENS SUBS? LAEVIS (MADIINAE, ASTERACEAE)

ELIZABETH A. F RI AR AND TASHA LxOoux Rancho Santa Ana Botanic Garden 1500 N. College Ave. Clarem ont, Cali}: 91711

ABSTRACT The presence of self-incompatibility was tested in Centroma dia pun gens subsp. laevis and (he genetic basis of the self-incompatibility respon se was ex plored using crossing studies. We performed full diallel cross ing experiments amo ng 10 ind ividuals from one natural population and four F , fam­ ilies. We observed a strong se lf-inco mpaubility resp onse in all indi viduals tested . with a significant difference in seed set between selfed and outcrosse d maii ngs . Most pairw ise rnat ings amo ng parental were co mpatible. with so me nonreciprocall y incom patible matin gs (i.e.. the rnatin gs were suc ­ cessful in one direction. but not the other), and only one recipro cally incompatible mating. The full diallel cross ing studies among sibs in the four F, families showed two major compatibility classes. Th ese results are consistent with a single-loc us spo rophytic self-incompatibility system in this spec ies. Key words: Asteraceae, Madiinae, self-incompatibility.

INTRODUCTION and the diploid genotype of the pist illate parent . Cross­ The majority of flowering plants are hermaphroditic, ing ability in sporophytic systems, on the oth er hand, having both male and female floral organs. The clo se is controlled by the diploid genotype of each parent. proximity of the sexual organs allows for self-polli­ Thus, a successful cro ss can only occur when the phe ­ nation by pollen transfer within the flower. While self­ notype , which may be different than the dipl oid ge­ pollination is common among plants, man y mecha­ notype due to dominance interactions between allel es, nisms that promote outcrossing have evol ved , impl y­ of each parent is different. Sporophytic SI is known in ing that there is an evolutionary advantage in avoiding six families, including Asterace ae, Betulaceae, self-fertilization (Richards 1997; Steinbachs and Hol­ Caryophyllaceae, Convolvula ceae, Stercul iace ae, but si nge r 1999). Among the many mechanisms that pro­ has been extensively studied at the molecular level in mote outcrossing, including dioecy, monoecy, distyly, only one family, Brassicaceae (Kao and McCubbin and other morphological adaptations, self-incompati­ 1996 ). bility (SI) is of interest for several reasons. First, SI is The genetic control of the SI respon se is relatively a genetically regulated self-recognition system. Sec­ eas y to characterize by crossing studies of full-sib fam­ ond, the genetic and molecular bases of SI are under­ ilies. In any F , family segregating for four alleles, sin­ stood in some, but not all, systems. Third, SI has a gle-locus GSI and SSI can be distin gui shed by the wide phylogenetic distribution in the Angiosperms, cross ing behavior of the siblings. One key character­ with several different mechanisms, indicating multiple istic of GSI sys tems is the presence of semi-compati­ evolutionary origins (Charles worth and Charlesworth ble crosses. In GSI , if the parents share one allele, 50 % 1979; de Nettancourt 200 I; Lundqvist 1990a. b; Rich­ of pollen grains will be capable of effecting pollina­ ard s 1997). tion . Gametophytic systems are also characterized by Although the phenomenon of SI has been obser ved the presence of four equally common cross ing type s for over 200 yea rs (Kolreuter 1763) the genetics con­ among the F , individuals and the absence of nonrecip­ trolling it were not discovered until the 1920's (Ea st rocal crosses (Goodwillie 1997; Richard s 1997 ). In and Mangelsdorf 1925 ; Prell 1921) and not until the SSI systems. there are no semi-compatible crosses. but last 50 years have there been exhaustive pollination there may be two, three or four crossing type s among studies to elucida te the mechanisms of the different F, individuals, due to dominance relationships amo ng forms of SI. Self-incompatibility systems can be cat­ alleles . Also. nonreciprocal crosses are common in SSI egorized into two basic types, gametophytic SI (GSI) sibships, because of dominance relationships amo ng and sporophytic SI (SSI). Gametophytic SI is wide­ alleles (Richards 1997). spread in the Angiosperms, occurring in at least 15 Se ver al cross ing studies within Asteraceae support families (Richards 1997; Newbigin et al. 1994). Cross­ the presen ce of sing le- locus SSI. the first studies being ing success under gametophytic control is determined of Cosmos bipinnatus (Crowe 1954), Crepis foetida by the haploid genotype of the individual pollen grain (Hughes and Babcock 1950), and Parthenium argen- 2 Friar and LaDoux ALISO tatum (Gerstel 1950). Additional studies within the the natural population and generate F] families for fur­ family include: Hymenoxys acaulis var. glabra ther study. To ensure control of pollen movement, disk (DeMauro 1993), Aster furcatus (Les et al. 1991), Se­ florets were removed from heads to be used as female necio squalidus (Hiscock 2000a), Calotis cuneifo lia parents prior to anthesis . All pollinations were per­ (Davidson and Stace 1986), Helianthus spp. (Desroch­ formed after all ray florets had opened and the stig­ ers and Rieseberg 1998), and Rutidosis leptorrhyn­ matic lobes had reflexed. At that point. heads of the choides (Young et al. 2000). However, several recent male parent that were presenting pollen were brushed studies have suggested a breakdown of the SSI system, onto the stigmatic lobes. This procedure was followed or the presence of additional loci controlling the re­ for both cross- and self-po llinations. Each cross-pol­ sponse, suggesting that the SSI system within Aster­ lination was performed at least three times. Each self­ aceae may not be as simple as previously thought (Les pollination was performed at least five times. In ad­ et al. 1991; DeMauro 1993; Hiscock 2000b). These dition, two heads per individual were simply bagged results serve as reminders that exhaustive studies are to determine the amount of seed set without pollen necessary in order to get a complete picture of the SI movement. Heads were removed from the plants after mechanism within any given family. Here we pre sent 3-4 weeks and the number of filled ray achenes count­ pollination data for (Hooker & ed . The percentage of seed set was calculated using Arnott) Greene subsp. laevis (D. D. Keck) B. G. Bald­ the number of ray florets and the number of filled win, which will be used as the model system for on­ achenes. Filled achenes could be easily distinguished going molecular and genetic studies of SI in Astera­ based on their plump appearance and indurate achene ceae subtribe Madiinae. wall. Unfilled achenes were thin and papery. Several achenes of each type were dissected to verify the cor­ MATERIALS AND METHODS relation between achene appearance and embryo de­ velopment. Plant Material

F J generation.-Full diallel crossing studies were per­ Centromadia pungens subsp. laevis, formerly Hem­ formed on four full- sib families from the parental gen­ ironia pungens (Hooker & Arnott) Greene subsp. lae­ eration. The four families were chosen to encompass vis D. D. Keck, is a small annual plant of the South the range of crossing results from the parental gener­ Coast and Peninsular Ranges of Southern , ation. Two families represent offspring from a recip­ typically growing in grassland habitats. The inflores­ rocally successful mating in each direction (# I: l oX cences measure 4-6 mm and have female ray florets 3 <;;>; #3: 33 X I <;;», one family represents an additional and bisexual (functionally staminate) disk florets. reciprocally successful mating (#2: 23 X 7 <;;», and one Parental plants were derived from seed collected family represents the offspring from a non-reciprocal from the San Jacinto Valley in Riverside County, Cal­ mating (#4: 7 3 X I <;;>; Fig. 2). ifornia in early July 1996 (Michael Wall 146, RSA). These seeds were placed into long-term storage at the Analysis of Crossing Data Rancho Santa Ana Botanic Garden for future use in reintroduction and restoration projects (accession The mean number of filled achenes for self- and # 19178). Ten seeds from the RSABG seed storage pro­ outcross pollinations was compared using a one-tailed gram were randomly chosen and germinated on agar paired t-test as implemented in StatView 5.0.1 (SAS plates, then transplanted into a steri Ie potting mixture. Institute, Cary, NC). The correlation between self and Plants were grown in a growth chamber at 14 hour outcrossed seed set and male and female success for days, with daytime temperatures of 26°C and nighttime individual plants was also calculated using StatView temperatures of 12°C. Parental plants were numbered 5.0.1. 1-10. The crossing results in the parental generation were Subsequent generation plants were derived from also examined to determine the number of S-alleles controlled crosses among the parental plants. The present in the sample. S-genotypes were assigned to achenes were harvested when fully ripe and stored un­ each individual based on crossing phenotype. It wa s til use . Achenes were germinated and grown as above. assumed that individuals showing nonreciprocal in­ compatibility shared one S-allele, that dominance hi­ F 1 families were numbered 1-4. Individuals in each F family were identified by letters. erarchies between alleles were possible, and that all 1 individuals were heterozygous at the S-Iocus. Alleles were defined as recessive if, when shared, the cross Cross- and Self-Pollinations wa s nonreciprocally incompatible. For instance, the I Parental generation.-A full diallel crossing study X 7 cross is nonreciprocally compatible (i.e., 13 X wa s performed on the ten individuals from the wild 7 <;;> is successful, I <;;> X 70 is not). It was assumed, population to estimate the number of alleles present in therefore, that these two individuals shared an S-allele VOLUME 21. NUMBER I Self-incompatibility in Centromadia 3

120%

100% f--

80% r-- i ~6O ~

oW" ..~ ll.40% f--

20% -

o • • a., %--0-10% 10-20% 20-30% 30-40% 40-50%• 50-60%- 60·70% 70-80% 80-90% 90-100% Percent Seed Set Fig. I. Percentage of self and outcross matings resulting in various percentages of seed set for the parental generation. in common. Moreover, as the cross was successful distribution was used to assign compatibility types for when individual 1 was the male parent, it was assumed all matings. All matings were categorized as being that one allele in individual 1 was recessive in the compatible, incompatible or indeterminate based on pollen, and that this allele was shared with individual the percentage of filled achenes. Incompatible matings 7. The total number of alleles in the sample was es­ were defined as those that had a mean of less than timated from the S-genotype assignments. 10% filled achenes, compatible matings were those that had a mean of greater than 40% filled achenes, RESULTS AND DISCUSSION and indeterminate matings had between 10% and 40% filled achenes. These limits were defined to include the Parental Generation maximum number of outcrosses as compatible and al­ The mean number of ray florets scored per head was low for variation in the success of seed set per indi­ 13.8 (4.83 standard deviation), with a mean of 56.6% vidual. Only five crosses fell into the 10-40% range; (40.6%) seed set overall. The mean percentage seed it is unclear whether these represent "leaky" incom­ set for outcrossed heads was 65.9% (36.1 %). The patible crosses or compatible crosses with poor seed mean percentage seed set for selfed heads was 0.2% set. The distribution of matings was highly skewed, (1.1 %). Outcrossing resulted in significantly greater with most outcross matings having high percentages seed set than selfing for each plant (paired z-test, t = of seed set, and all selfed matings and a minority of -13.126, P < 0.000 I). There was no significant cor­ outcross matings having very low seed set. relation between outcross seed set and selfed seed set Centromadia pungens subsp. laevis appears to be for individual plants (r = -0.121, P = 0.75). There strongly self-incompatible. There is a significant dif­ was considerable variation in mean seed set (65.1 % to ference in the mean seed set between self and outcross 88.04%) and mean male success, defined as mean seed matings, with selfed matings resulting in virtually no set for an individual over all matings made as a pollen seed set. Outcross matings resulted in variable parent (52.7% to 91.2%). However, there was no cor­ amounts of seed set. The distribution of seed set for relation between success as a male parent and as a outcross matings is largely bimodal, with a large peak female parent for individual plants (r = 0.203, P = near 100% and a smaller peak near 0% seed set, with 0.59). Heads that had been bagged without any pollen few matings showing intermediate values of seed set. movement had no observed seed set. These results suggest that the majority of outcross mat­ A frequency histogram for the mean percent seed ings are compatible, with a minority that are strongly set for both self and outcross matings is shown in Fig. incompatible. I. Each bar represents the percentage of matings of Figure 2 shows the distribution of compatible and each cross type (i.e., self vs. outcross) with the re­ incompatible matings among the ten individuals from spective percentage of filled achenes. This frequency the natural population. Of the 90 possible outcross 4 Friar and LaDoux ALISO

69 79 I sv I 9~ 109 Tabl e I. All eli c assignments of individuals in the parental gen­ I 19 29 39 I 49 I I I eration based on inference from crossing phenotype. + -- +- 1 :-" 10' + + -'-f l Individual Alkk.. ,. 26 i + + + + + - + I 5,,51 , , 2 5 ,,5, 3d . + + + · + 1j1='il · I I I 3 5,,5 I I I 0 + I 4 40' I + + + + · + 57.5" · 5 5 1) .5/1) ------l-- 21 -,- 5 0' + + + + . ! + + + 6 5 " .5'2 , I . _ . 7 5/,5" , + 60' + I + · ~ + · + 8 5/4,5,; --.I , 9 5/,5, 70' + . t I + 10 . · · 51,5'6 l I - .: 8d + + ± . + + + I I 96 + · i + + the parental generation. The four families were chosen to represent the range of crossing results in the parental 100' + . .. I I · generation, including both reciprocally compatible and nonreciprocal matings. The sample sizes for the four Fig . 2. Diallel crossing results for the parental generation. The families vary due to differences in germination rate male parents are listed in rows, the female parents in columns. A among them. As the parental generation produced few .+' indi cates a compatible cross, a '-' an incompatible one. and a . ±. an indeterminate one . Light shading indicate s nonreciprocally matings with intermediate percentages of seed set, we incompatible crosses. Dark shading ind icates reciprocally incompat­ were able to score matings in the F, generation simply ible crosses. as compatible or incompatible. Therefore, statistics on the percentage of filled seeds were not recorded for

matings, 83 (92.2%) were compatible. Of the incom­ the F I generation. The results of the diallel crossing patible rnatings, most were nonreciprocally incompat­ experiments are presented in Fig . 3-6. ible (i.e ., compatible in one direction but not the re­ All families show two major compatibility groups, verse). Of the 45 pairs of plants, 40 were reciprocally with occasional aberrant results. Most of the aberrant compatible, four were nonreciprocally incompatible, results appear to be related to low seed set in a few and one was reciprocally incompatible (Fig. 2) . The individuals. For instance, in family #2, individuals 2D assignment of 16 5-alleles to the sample of 10 indi­ and 2G never set seed regardless of the male parent viduals explains all of the observed mating phenotypes used (Fig. 4). Family #4, the only F, family resulting (Table I). [n one incompatible cross (1 <3 X 9 «), it is from a non-reciprocal cross (Fig. 2; 7 <3 X I «), clear that individual 9 shares an allele with individual showed the most aberrant results. In the inferred allelic 2, but which allele cannot be determined, so we arbi­ compositions of each individual, it was assumed that trarily assigned allele 53 to individual 9 for purposes parental individuals I and 7 share an allele in common, of discussion. In addition, the se assignments assume either 5) or 52> which is recessive in the male parent.

that 51 is dominant to 53' but recessive to 5 13, and 52 The possibility of homozygous individuals resulting is recessive to 5 16 in pollen in order to explain the from this cross may explain the aberrant results. Al­ nonreciprocal matings. ternatively, the aberrant results may be due to poor These allele assignments, while relatively arbitrary, indicate that a sample of ten individuals from the nat­ ural population contains approximately 16 5-alleles, indicating significant genetic diversity at this locus. These results are typical for self-incompatible taxa, which typically show large numbers of alleles main­ l e d tained in relatively small populations (Lawrence 2000). By contrast, Brennan et al, (2002) found that a IB d natural population of Senecio squalidus had only six alleles in a natural population. This finding was ex­ 10 0' plained by the presence of a severe population bottle­ l Ed neck at the introduction of this to Britain. IF d F) Generation

Complete dial leI crossing experiments were per­ Fig. 3. Diallel crossing results for F, family # I (1 0 X 3 <:( ). An formed with four F) families derived from crosses in expl anation of symbols and shading is given in Fig. 2. VOLUME 21. NUMBER I Self-incompatibility in Centromadia 5

, 213 9 I 2C9 2H 4A 9 409 4139 41'9 4 (i ~ 419 - ----1------1 2A o" + -lA d + 1_ _ -"- 20 0 400 + ~ I T

2Go 41:0

28 0 4Ho +

2Cd 413 0'

4(' 0'

2Fd 41'0

4G d' + + + Fig. 4. Diallel cros sing results for F, family #2 (2 0 X 7 2). An exp lanation of sy mbo ls and shading is given in Fig. 2. 410' + + male performance for individual 4E and poor female Fig. 6 . Diallel crossing results for F, fam ily #4 (7 0 x 12 ). An seed set for individual 4D. expl anation o f symbols and shading is give n in Fig . 2. The presence of nonreciprocal crossing results in both the parental and F 1 generations and two compat­ compatibility response. This pattern has been found in ibility classes in the F sibships are consistent with 1 several other self-incompatible species, including ere­ single locus sporophytic control of self-incompatibili­ pis foetida and Hyp ochoeris radicata in the Astera­ ty. These results are also con sistent with previous find­ ceae, several spec ies in the Brassicaceae, and at least ings throughout the phylogenetic span of Asteraceae, two species in the Caryophyllaceae (Lewis et al. 1988; which have shown single-locus sporophytic control of Lundqvist I990a, 1994; Hiscock 2000b ). SI (Gerstel 1950; Hughes and Babcock 1950; Crowe Recent results (Hiscock et al. 2003) suggest that the 1954; Hiscock 2000a). As 51 systems appear to be a self-incompatibility systems of the Asteraceae and family-level trait (with at least one exception in the Brassic aceae may not be homologous at the molecular Polemoniaceae; Levin 1993; Goodwillie 1997; La­ level, despite the similarities in physiology and inher­ Doux and Friar in prep.), all self-incompatible taxa in itance. The members of the Asteraceae that have been the Asteraceae may share a similar system (de Nettan­ investigated to date seem to share a similar single­ court 200 I). However, there was no evidence in these locus sporophytic SI syndrome, though with some im­ crossing results of the anomalous compatibility found plications of other modifying loci in addition to the by Hiscock (2000b) , who found fully or partially com­ locus of large effect. Once the gen es res ponsible for patible crosses in otherwise completely incompatible the SI res ponse in the Asteraceae are cloned, the mo­ groups. These results led him to postulate the presence lecular, developmental, and/or physiological homolo­ of an additional gamerophytic locus controlling the in- gies between the two families, and among taxa within the Asteraceae, can be addressed.

ACKNOWLEDGMENTS 3Ao The authors wish to thank Eric Roal son and Michael 30 0 Kinney for help in cro ssing experiments; Dylan Han­

31'0 non, Su san Jett and Michael Wall for assistance in seed germination and plant maintenance; and two anony­ l Go mou s reviewers for helpful comments. This work was supported by Rancho Santa Ana Botanic Garden. )Ho

310 LITERATURE CITED

BRENNA N. C .. HARRIS, TABAH , AN D HISCOCK . ~ Co' A. S. A. D. A. S. J, 2002. Th e population ge netics of sporophyiic se lf-inco mpa tibility

l Eo in Senecio squalidus L. (Asterace ae ) I: S a lle le diversity in a nat­ ural po pulation. Heredity 89 : 430-438. CHARLESWORTH. D.. AND B. CHA RLESWORTH. 197 9, The evolutionary Fig. 5. Diallel crossing results for F, family #3 (30 x 12 ), An genetics of sexual systems in flowering plants, Proc. RO\'. Soc. e xplanation of symbol s and shading is given in Fig. 2. London. Ser. B, Bioi. Sci, 205: 5 I3-530, 6 Fri ar and LaDoux ALISO

C ROWE. L. 1954 . Incompat ibil ity in Cos mos bipinnatus . Herediry 8 : nebst Fortsetzungen I, 2 u. 3. 1761-1766. Ost wald 's Klassi ker, I-I I. Nr. 4 1. Engelmann . Le ipzig. 266 p. DAVIDSON, J. K.• AND H. M. STACE. 1986. Geneti cs of se lf-incom­ LAWRE NCE, M. J. 2000. Population ge netics of the homomorphic S I patibility in Calotis cuneifo lia. J. Heredity 77: 471 -472. polymorphism in flowering plants. Ann. Bot. 85A: 221-226. DEM AURO. M. M . 1993. Re lationship of breeding system to rarity LES, D . H., J. A .REINARTZ. AN D E. J. ESSELMAN . 1991. Genetic in the Lakeside Daisy (Hymenoxys aca ulis var. glab ra) . Cons . consequences o f rarity in Aster furcatus (Asteraceae), a threa te ned Bio/. 7: 542-550 . self-incom pa tible plant. Evolution 45 : 1641 -1 650. DE N ETIANCOURT. D. 200 I. Incompat ibility and incongruity in wild LEVIN , D . A. 1993. S- gen e po lymorp hism in Phlox dr ummondii. He­ and cultiva ted plants. S pringer-Verlag. Berlin. 322 p. redity 71: 193-198. DESROCHE RS. A. M .• AND L. H. RI ES EBERG. 1998. Mentor e ffec ts in LEWIS , D .,S. C. VERM A, AND M . I. ZUBER I. 1988. Garne to phy tic­ in w ild species of Helianthus (As teraceae). Amer. J. Bot. 85 : 770­ spo ro phy tic incompatibili ty in the Cruciferae, Raphanus sativus. 775 . Her edity 61 : 355-366. EAST, E. M.• AN D A . J. MANGELSDORF. 1925 . A new interpretation LUNDQVIST, A .I990a. On e-lo cus spor op hy tic S-gen e sys te m with of the hereditary behaviour of sel f-ste rile plants. Proc. Natl. Acad. traces of garnetophytic pollen control in Cerasrium arvense ssp. Sci. U.S.A. 11: 166-171. strictum (Caryophy llaceae). Her editas 113: 203-215 . GERSTEL, D . U. 1950. Se lf-incom pa tibi lity studies in guayule . Ge­ ---. I990b. The compl ex S-ge ne sys tem for co nt ro l of se lf-in­ nerics 35 : 482- 506. compa tibility in the butt ercup Ranun culus. Hereditas 113: GOODWILLI E. C. 1997. T he ge ne tic co ntro l of se lf-incompa tibility in 29-46. Linanthus parviflorus (Po lemoniaceae). Heredity 79 : 424-432. ---. 1994. 'S low ' and 'q uick' S-all el es without dominance in­ HISCOCK, S . J. 2000a. Se lf- inco mpa tibility in Sen ecio squalidus L. teraction in the sporo phytic one- loc us se lf-inco m pa tib ili ty sy ste m o f Srellaria holostea (Caryophy llaceae). Her editas 120 : 191- 202. (As te raceae). Ann. Bot. 85 A: 181-1 90. NEWBI GIN. E., M . A. ANDERSON, AN D A. E. CLARKE. 1994. Game­ ---. 2000b. Gen eti c co ntro l o f se lf-incompatibility in Se necio tophyt ic se lf-incom patibility in Nicotiana alata. pp. 5- 5 18. /n E. sq ua lidu s L. (As teraceae): a successful colo nizing species . Hered­ G. William s, A. E. C larke and R. B. Kn ox [eds.], Ge ne tic control ity 85 : 10-1 9. of self-incom patibility and reproduct ive development in flowering ---,S. M . McIN NIS. D . A. TABAH, C. A. HENDERSON, AN DA . C. plants. Kl uw er Aca dem ic, Boston. BRENNA N. 20 03. Spo ro ph ytic se lf- incom pa tibi li ty in Se necio PRELL, H. 1921 . Das Problem der Unfruchtbar keit, Naturwiss. Woch­ sq ua lidu s L. (A steraceae)-the search fo r S. J. Exp . Bio/. 54: 169­ enschr. N. F. 20 : 440-446. 174. RICHAR DS, A . J. 1997. Plant breeding syste ms . G. All en & Unwin. HUGH ES, M. B., AND E. B. BABCOCK. 1950. Self-incompatibility in London. 529 p. Crepis foetida L. subsp. rhoeadifolia Bieb. Schinz et Keller. Ge­ STEIN BACHS, J. E. , AND K. E. HOLSI NGER. 1999. Poll en tra nsfer dy­ nerics 35 : 570-588 . nam ics and the evolution of game to phytic se lf- incompatibility. J. KAo. T. -H., AND A . MCCUBBIN. 1996. How flo wering plant s d is­ Evol. Bioi. 12: 770-778. cr imi na te be tween se lf an d non-sel f po lle n to prevent in bree d ing . YOUNG A ., C. MILLER, E. G REGORY, AND A . LANGSTON. 2000. Spo­ Pro c. Nat l. Acad. Sci. U.S.A. 93: 1205 9-12065. rophytic se lf-incom patib ility in diploid and tetraploid races of Ru­ KOLREUTER, J. G. 1763. Vorla ufige Nachricht von ei nige n da s Ge­ tidosis lept orrhynchoides (Asteraceae). Ausrral. J. Bot. 48 : 667 ­ schlecht de r Pflanzen betre ffenden Versuc he n und Beobachtungen , 672.