Breeding barriers between the cultivated strawberry, Fragaria ananassa, and related wild germplasm

Transcription

1 Euphytica 136: , Kluwer Academic Publishers. Printed in the Netherlands. 139 Breeding barriers between the cultivated strawberry, Fragaria ananassa, and related wild germplasm Arias E. Marta 1,ElsaL.Camadro 2,JuanC.Díaz-Ricci 3 & Atilio P. Castagnaro 3,4, 1 Cátedra Anatomía Vegetal, Facultad de Ciencias Naturales e Instituto Miguel Lillo, Universidad Nacional de Tucumán (UNT). Miguel Lillo 205, 4000-Tucumán, Argentina; 2 Estación Experimental Agropecuaria (EEA) Balcarce, Instituto Nacional de Tecnología Agropecuaria (INTA)-Facultad de Ciencias Agrarias, Universidad Nacional de Mar del Plata (UNMdP). C.C. 256, 7620-Balcarce, Buenos Aires, Argentina; CONICET; 3 Instituto de Química Biológica Dr Bernabé Bloj, Facultad de Bioquímica y Farmacia, Instituto Superior de Investigaciones Biológicas (INSIBIO; CONICET-UNT), Departamento de Bioquímica de la Nutrición, UNT. Chacabuco 461, 4000-Tucumán, Argentina; 4 Sección Biotecnología, Estación Experimental Agroindustrial Obispo Colombres- Unidad Asociada al INSIBIO. CC N 9, 4101-Las Talitas, Tucumán, Argentina; ( author for correspondence; atilioc@fbqf.unt.edu.ar) Received 7 April 2003; accepted 12 January 2004 Key words: cross-incompatibility, Duchesnea, Fragaria, pollen-pistil compatibility, Potentilla, strawberries Summary Five-hundred interspecific and intergeneric crosses were performed among accessions of the wild strawberries Fragaria vesca (2x), Duchesnea indica (8x), Potentilla tucumanensis (2x) and 9 genotypes of the cultivated strawberry, Fragaria ananassa (8x), following an incomplete diallele mating design. Crosses between D. indica and F. ananassa produced many putative hybrids when D. indica was used as female but a few achenes and plants when used as male; therefore, pollen-pistil compatibility relations were analyzed by fluorescence microscopy in this direction of the cross. Of the genotypic combinations, 78.6% were incompatible at the stigma level and 17.2% at the first third of the style. Only 3.6% were pollen-pistil compatible and produced fruits with achenes (seven did not germinate or originated short-lived plants and nine produced normal plants). F. vesca F. ananassa crosses produced 35 hybrid achenes but only 14% germinated, yielding short-lived plants; histological analyses revealed that inviable seeds had less developed (or collapsed) endosperms and smaller embryos than control plump F. vesca seeds. P.tucumanensis was only used as male, with negative results. These species and genera are partially isolated by a complex system of pre- and post-zygotic barriers. Knowledge of their nature would allow the breeder to devise strategies to put the genetic variability available in the group into a useful form. Introduction The genus Fragaria and the closely related genera Duchesnea and Potentilla belong to the Rosaceae family, tribu Potentilleae. The taxonomy of the group is controversial and there is no agreement among authors regarding the number of species in each genus (Staudt, 1962, 1989; Darrow, 1966; Kalkmam, 1968; Mabberly, 2002). Wild strawberries include species of the three genera. According to Darrow (1966), the This paper is part of the first author s doctoral thesis species are assigned to four groups on the basis of their chromosome numbers (five of them are diploid, two tetraploid, one hexaploid and three octoploid, with x=7) and reproduce asexually either by stolons or apomixis and sexually by autogamy or allogamy. Diploid Fragaria species have a gametophytically controlled self-incompatibility system, whereas polyploids are self-compatible, possibly due to competitive interactions of different S alleles in diploid or higher ploidy pollen grains (Evans & Jones, 1967). The diploid Potentilla tucumanensis is a recently described species, endemic of the Tucumán region (Castagnaro et al.,

2 ; Zardini, 1999; Arias et al., 2001) and has not been studied yet from the reproductive point of view. Strawberries are perennial or annual herbs, some of them edible, with medicinal use. The cultivated strawberry, Fragaria ananassa Duch. (2n=8x=56), is an interspecific hybrid between the wild species F. chiloensis L. (2n=8x=56) and F. virginiana Duch. (2n=8x=56) (Darrow 1966). Intraspecific crosses of F. ananassa (8x) have extensively been used to obtain new cultivars (i.e. Pájaro, Chandler, Milsei Tudla, Camarosa, among others) with improved agronomic traits. However, this species is susceptible to fungal diseases and bacterial pathogens, among them, Colletotrichum spp., Xanthomonasfragariaeand Pseudomonas solanacearum (Maas, 1998). On the other hand, Fragaria vesca L. (2n=2x=14) and Duchesnea indica (Andr.) Focke (2n=8x=56) have edible fruits and, with Potentilla tucumanensis (2n=2x=14), are valuable sources of genetic determinants for disease resistance and stress tolerance. These three species grow naturally in Tucumán province, Argentina, and are of interest in the national breeding program (Ontivero et al., 2000). Direct crosses between octoploid and lower ploidy species are often unsuccessful, although crosses between octoploid cultivated strawberries with F. vesca (2x) and F. moschata Duch. (6x) were reported to produce viable hybrids with partial seed set (Bauer, 1976; Trajkovski, 1982; in Luby et al., 1991; Hancock et al., 1990). Three decaploid cultivars, Annalie, Spadeka, and Sara, were produced using colchicine-doubledf. vesca in successive crosses with F. ananassa (8x) (Bauer & Bauer, 1979; Trajkovski, 1988; in Luby et al., 1991). Other approaches for transferring genes from the lower ploidy species to the cultivated octoploids rely on the use of colchicine or gametes with the unreduced chromosome number (2n gametes) to derive decaploids, nanoploids or synthetic octoploids to be used as bridges (Darrow, 1966; Evans, 1977; Hancock et al., 1990, Trajkovski, 1982). Intergeneric crosses were reported by different authors. Wolf (1908) obtained hybrids between D. indica (8x) and Potentilla reptans L. (2x) that only grew vegetatively. Mangelsdorf & East (1927) reported the first successful intergeneric Fragaria Potentilla cross that produced two short-lived seedlings; they also obtained hybrids between F. vesca (2x) and D. indica (8x) that were weak and died before reaching the adult stage. Jones (1955) obtained 639 seedlings in crosses between F. vesca (2x) and Potentilla spp. (2x), none of which survived. Ellis (1962) reported the production of hybrids with different ploidy levels in crosses between species of Fragaria and Potentilla, most of which died and some of the few that survived were sterile. In crosses between Fragaria moschata (6x) and Potentilla fruticosa L. (2x), Asker (1970) obtained a small number of hybrids whereas Macfarland Smith & Jones (1985) obtained only nine plants from a large number of seeds; five of these plants died before flowering and the others were vigorous but sterile (four had the expected 4x chromosome number and the rest were aneuploids). In contrast, Bringhurst & Barrientos (1973) reported fertile 10x hybrids between F. chiloensis (8x) and Potentilla glandulosa (Lindl.) Rydb. (2x). In the unsuccessful experiments, the causes of the failure in the production of fertile intergeneric hybrids were not studied. Abdullah & Hermsen (1993) also made crosses between F. ananassa (8x) and P. fruticosa (2x), followed by embryo rescue. They obtained 154 plants from 1001 in vitro cultivated embryos, 55 of which were able to grow autotrophically. Fortythree out of 50 of the growing plants proved to be hybrids based on their morphology. Yuhua Li et al. (2000) carried out interspecific crosses between F. ananassa Honcoye (8x) and F. vesca (2x) from Changbaishan, Jilin, China, and obtained 167 seedlings (84 hybrid and 83 apomictic). As the rate of viable hybrid achenes was low, they analyzed the possible incompatibility barriers. They observed irregular elongation of pollen tubes in the style, no further growth of the embryo beyond the globular stage and the absence of endosperm. In nature, external and internal barriers can hinder or prevent gene flow between related taxa (Hadley & Openshaw, 1980). Examples of external barriers to genetic exchange are the physical separation of populations in time or space, the adaptation of populations to specific ecological niches or combinations of barriers that produce discontinuity among populations. External barriers are usually reinforced by internal barriers, which are those that reside within plant tissues and can prevent hybrid formation by acting at the pollen-pistil (pre-zygotic) and/or the embryo and/or the endosperm levels (post-zygotic) or, if hybrids are formed, by producing hybrid weakness, sterility or breakdown in segregating generations. In interspecific and intergeneric crosses between F. ananassa (8x) and the related species F. vesca (2x), Potentilla tucumanensis (2x) and D. indica (8x), in which F. ananassa and F. vesca were used as females, we observed incompatibility problems, revealed by either lack of fruit formation or low seed set, with unvi-

3 141 able seeds. To detect the internal hybridization barriers that were possibly operating, we analyzed pollen-pistil compatibility relations and the development of embryo and endosperm in crosses; the results of these analyses are reported in this communication. Materials and methods During , 18 accessions of the following wild strawberry species were used in crosses: D. indica (15), F. vesca (2) and P. tucumanensis (1). Plants were collected in the field and placed in a greenhouse. At bloom and depending on the availability of flowers, they were crossed with nine cultivars of F. ananassa (Table 1) following an incomplete diallele mating design. In order to avoid self-pollination of self-compatible accessions we protected the flower after emasculation during at least three days before the pollination. Emasculation control experiments were conducted with the self-compatible accessions used to rule out self-pollination due to emasculation mishandling. All commercial and wild genotypes were reproduced by stolons except P. tucumanensis that does not produce runners. Since RAPD analyses showed genomic differences among different Pájaro accessions (García et al., 2002), two of them (P.111 and P.11) were included in the crossing work. P. tucumanensis was only used as male because it has very small flowers and is, therefore, difficult to emasculate. Closed blossoms of the female progenitors and mature fully expanded flowers of the male progenitors were used; flowers were emasculated prior to pollination. Part of the crosses were carried out in September- November of ; the rest, in which pollenpistil compatibility relations were studied, were carried out in September-November Compatibility relations Sixty to 200 styles of each pollinated flower were analyzed. The styles were removed from the plants 48 hs after pollination, fixed in FAA (1:1:8, v/v/v, formaline, glacial acetic acid, ethanol 80 ) and stored at 4 C. Ovaries were left on the plant for fruit and seed formation and an aqueous solution of 4 ppm 2 4 D (Dionne 1958) was applied to them to prevent their premature abscission. Fixed styles were rinsed with tap water, treated with an 8N NaOH solution for 4 hs, rinsed again with tap water, stained with a 0.1% aniline blue solution in 0.1 M potassium phosphate (Martin, 1958), mounted in a drop of glycerol on a glass slide, squashed with a cover slip and observed by fluorescence microscopy (Olympus optical BX40 F4 CO. LTD). Berries were harvested at maturity, approximately 21 days after pollination. Achenes were extracted; the plump ones were treated with concentrate HCl during 5 10 min and washed six times with distilled water under sterile conditions in a laminar flow. These achenes were germinated on MS medium (Murashige & Skoog, 1962). Embryo and endosperm development Twenty-one days old achenes from the F2 (2x) F. Chandler Mza (8x) cross were fixed with 3% glutaraldehyde for 6 h at 4 C and then post-fixed overnight with 1% osmium tetroxide in 0.1 M phosphate buffer (ph 7.2). The samples were dehydrated with a graded series of ethanol, ending with 100% acetone, and then embedded in Spurrïs medium (Spurr, 1969) and polymerized overnight in a 60 C oven. Cross-sections (0.5 µm) were obtained with an ultramicrotome and stained with toluidine blue (Richardson et al., 1960) before visualization with a light microscope to analyze embryo and endosperm development. Sections of normal achenes of F. vesca, used as controls, were also observed by light microscopy and Scanning Electronic Microscopy (SEM). Pollen viability Pollen viability was estimated indirectly by staining a sample on a glass slide with acetocarmine glycerol jelly (Marks, 1954) and observing under a light microsocope. Round pollen grains with red cellular walls and cytoplasm were considered viable. Results Crosses During September-November of , 500 crosses were carried out among two genotypes of wild F. vesca (F1, F2), different genotypes of the cultivated strawberry and two species of related genera, D. indica and P. tucumanensis. In September-November of , crosses between D. indica F. anannassa were made in both

4 142 Table 1. Genetic materials used in crosses and their origin Genetic material Genus and species Type of material Origin Wild accession Site of Collection in Tucumán province P1 Potentilla tucumanensis Seed a Potrero F1 Fragaria vesca Clone Villa Nougués F2 Fragaria vesca Clone Taficillo D.H.M Duchesnea indica Clone Horco Molle D. Duchesnea indica Clone Bosque de Alisos D.22 Duchesnea indica Clone Taficillo D.25 Duchesnea indica Clone Taficillo D.24 Duchesnea indica Clone San Pedro de Colalao D.26 Duchesnea indica Clone Potero de Trancas D.30 Duchesnea indica Clone Taficillo D.31 Duchesnea indica Clone Taficillo D.32 Duchesnea indica Clone Potero de Trancas D.78 Duchesnea indica Clone Potero de Trancas D.V.P.M Duchesnea indica Clone Villa Padre Monti D.R Duchesnea indica Clone Raco D. a. Duchesnea indica Clone Raco D.S.J Duchesnea indica Clone San Javier D.V.N Duchesnea indica Clone Villa Nougués Cultivated accession Producer in Province F. Pájaro11 F. ananassa Clone Salta F. Pájaro111 F. ananassa Clone Mendoza F. Camarosa F. ananassa Clone Mendoza F. Enzed Donna F. ananassa Clone Mendoza F. Rosalinda F. ananassa Clone Mendoza F. Sweet Charlie F. ananassa Clone Buenos Aires F. Gaviota F. ananassa Clone Mendoza F. Milsei Tuddla F. ananassa Clone Mendoza F. Chandler Mza F. ananassa Clone Mendoza a Obtained from one seedling because P. tucumanensis is an annual plant. directions (results not shown). When the female progenitor was a cultivated genotype, the only successful cross was F. Pájaro 111 D. 25, which produced one fruit with five viable achenes; one of these achenes originated a plant but, since pollen germination and growth was not analyzed, it could not be established if its origin was sexual or asexual (apomictic). The rest of the crosses among other genotypes were unsuccessful or produced only one or two achenes that were not viable. However, in the reciprocal cross, in which the female progenitors were different genotypes of D. indica, 993 achenes were obtained, 42% of which germinated and only 4% gave rise to plants that reached maturity. The latter are being evaluated by morphological and molecular characters to confirm their hybrid condition. The combination F2 F. Chandler Mza produced thirty-five achenes. Half of them were used for germination and the rest for histological studies. Germination experiments showed that the achenes were either not viable or gave rise to plants that died shortly after germination. The reciprocal cross did not produce any achenes. After numerous unsuccessful tries to emasculate the flowers, the wild genotype P. tucumanensis was crossed only as male and no fruits were obtained. The results of the crossing work in 2000 are presented in Table 2. Achenes were produce in only five out of 27 interspecific and intergeneric genotypic

5 143 Table 2. Number of pollinated flowers, fruiting receptacles and achenes, and pollen-pistil compatibility relations in crosses carried out in 2000 Cross #Pollinated #Fruiting #Achenes per Compatibility Female male flowers receptacle fruiting relations receptacle F.Camarosa D c-1 F.Camarosa D c-1 F.Camarosa D a,1 a c-2 F.Camarosa D. H. M a c-2 F.Camarosa D. R c-1 F.Camarosa D. S. J , 16c-2, c-3 F.Camarosa D c-1 F.Camarosa D. V. P.M c-1 F.Camarosa D c-1 F.Camarosa D. V. N c-1 F.Camarosa D c-1 F.Camarosa D c-1 F.Milsei tuddla D. R c-1 F.Milsei tuddla D c-1 F.Milsei tuddla D c-1 F.Milsei tuddla F c-2 F.Rosalinda D c-1 F.Rosalinda D c-1 F.Sweet Charlie D c-1 F.Sweet Charlie D. H. M. 5 1 c 1 b c-1 F.Sweet Charlie D c-1 F.Sweet Charlie D. a c-1 F.Enzed- Donna F1 2 1 c 2 b c-2 F.Gaviota D c-1 F.Gaviota D. S. J c-1 F.Gaviota D c-1 F.Gaviota D. H. M c-1 Control F.Camarosa F.Milsei tuddla , 200, 225 c-4, c-4, c-4 a died shortly after germination; b aborted; c dry fruiting receptacle; c-1, inhibition of pollen grain in the stigma; c-2, inhibition of pollen tube growth in the first third of the style; c-3, normal pollen tube growth down to the end of the style; c-4, control combination. combinations. The number of achenes per fruiting receptacle varied between one and 16 in those combinations and between 165 and 225 in the intraspecific one (F. camarosa F. Milsei tuddla). The only successful intergeneric cross between the cultivated genotypes and Duchesnea was F. Camarosa D.S.J.. It produced sixteen achenes in a single fruiting receptacle; nine of these gave rise to plants that reached maturity and the others either did not germinate or gave rise to plants that died shortly after germination (Figure 1a, b). The other two combinations with F. Camarosa as the female progenitor and genotypes D.22 and D.H.M. of Duchesnea as males produced four achenes, but the plants died after germination. In the F. Enzed Donna F1 combination, two achenes were obtained but aborted and in the cross F. Milsei Tudla x F1 no achenes were obtained (Table 2). Pollen viability According to these results, we have investigated whether the low yield of achenes production could be attributed to pollen viability. During springtime (blooming season) D. indica (8x) and F. ananassa

6 144 Figure 1. Hybrid plants between F. ananassa (8x) var. Camarosa Duchesnea indica (2x). a. Normal development; b. Abnormal development of root and leaves. Bar = 1 cm.

7 145 (8x) showed the best flower production and high pollen viability (90%). During summer, the estimates of pollen viability were, on average, 90% in D. indica (8x) and 30% in F. ananassa (8x). In this season, few flowers and many stolons were produced. Pollen-pistil compatibility relations Crosses were classified according to: i) pollen grain germination, ii) pollen tube growth and iii) production of plump achenes per fruit (Table 2). The pollen-pistil compatibility relations observed could be classified into four groups: (c-1) Inhibition of pollen grain germination in the stigma. Pollen grains adhered to the stigma, but few of them germinated (Figure 2a); the receptacle showed a dry aspect. This group included 78.6% of the genotypic combinations shown in Table 2. (c-2) Inhibition of pollen tube growth in the first third of the style in 17.8% of the combinations. Many pollen grains germinated on the stigma but only a few pollen tubes grew down to the first third of the style (Figure 2b). The green receptacle was turgid and some ovaries were prominent but fruitless. (c-3) Normal pollen tube growth down to the end of the style in 3.6% of combinations (Figure 2c, d). In 40 out of 200 pistils analyzed, one to six irregular pollen tubes were observed growing along each pistil. Some of them were arrested in the first third of the style whereas others were arrested in the first two-thirds of the style; occasionally few pollen grains germinated but did not grow down the style or only a few tubes reached the ovaries. These were the only inter-specific and intergeneric genotypic combinations that produced red fruits with achenes. (c-4) Control combination (F. Camarosa F. Milsei Tudla). Many pollen grains germinated on the stigma and pollen tubes grew down to the base of the style (Figure 2e). Red fruits with many achenes were obtained. Post-zygotic incompatibility The interspecific F. vesca (2x) F. ananassa (8x) cross produced thirty-five putative hybrid achenes that had a low percentage of germination (14%). In transversal microtome sections, it was observed that most hybrid achenes had small embryos at the globular stage and poorly developed endosperms whereas the remaining achenes did not have endosperm and embryos were absent or very small (Figure 3a). Normal achenesfromthef. vesca F. vesca cross, in contrast, had the typical Dicotyledoneae embryo with two cotyledons and well-developed endosperm (Figure 3b). The histology of the pericarp of hybrid and normal achenes was similar, except that normal achenes had thicker pericarps than the hybrid achenes (Figure 3c, d). Discussion In all interspecific F. ananassa (8x) Duchesnea indica (8x) and F. ananassa (8x) Fragaria vesca (2x) crosses analyzed, pollen-pistil incompatibility barriers were observed, most frequently at the stigma level (Table 2). These barriers were not complete for a few achenes were, nevertheless, obtained. In the reciprocal D. indica (8x) F. ananassa (8x) crosses, on the other hand, many achenes (and forty putative hybrids) were obtained. Likewise, in crosses between F. vesca (2x) and F. ananassa (8x), pollen tube growth was arrested in the first-third of the style and only two achenes were obtained, but aborted. In the reciprocal direction, thirty-five achenes were obtained, indicating that pre-zygotic barriers were not acting but, since these achenes were either non-viable or gave rise to weak plants, the possible action of post-zygotic barriers had to be assumed. The phenomenon of pollen-pistil compatibility in one direction of a given cross and incompatibility in the reciprocal direction was described as unilateral incompatibility (UI) by Lewis & Crowe (1958). According to these authors, pollen grains from selfcompatible (SC) species are inhibited in styles of self-incompatible ones (SI), preventing the incorporation of self-compatibility genes in allogamous species (and the exposure of subvital or lethal alleles to natural selection). Later, several authors observed UI in other combinations of species (SC SC, SI SI and SC SI) (Abdalla & Hermsen, 1972) and also in both directions of a given cross (bilateral incompatibility or BI). In a first approach to the problem, some authors proposed that the S-locus was involved in the crossincompatibility reaction and assumed that this locus had a dual function: prevention of self-fertilization when the individuals involved in a cross were genetically close and cross-fertilization when they were genetically distant. A second approach disregards the possibility of the S-locus being involved in the crossincompatibility reaction because its structure has to be very complex to account for its dual function (re-

8 146 Figure 2. Pollen germination observed by fluorescence microscopy in F. ananassa (8x) var. Camarosa D. indica (2x) (Taficillo) cross. a. Inhibition of pollen grain germination and arrest of pollen tube growth in stigma (x400); b. Pollen tube arrested in the first third of the style (x200); c. Compatible pollen tube growth along the style (x400); d. Pollen tube growth until the end of the style (x400); f. Compatible control combination (F. ananassa (8x) var. Camarosa F. ananassa (8x) Milsei tuddla) (x400).

9 Figure 3. Transversal section of hybrid and normal achenes of F. vesca (2 ). a. Abnormal embryo in hybrid achenes F. vesca (2x) F. ananassa (8x) (x400); b. Normal embryo of F. vesca (2x) with two cotyledons and endosperm (x400). SEM (Scanning electronic microscopy) of pericarp of a F. vesca fruit (2x). c. Pericarp with layers of parenchyma, elongated cells and fibers of a normal achene; d. Detail of elongated cells with helicoidally thickening walls of a normal achene. Bar = 100 µm. em embryo, per pericarp. 147

10 148 cently, a rather simple structure has been proposed for the locus by Stone & Goring, 2001), pollen tube arrest does not occur at a single site in the style but rather at various sites along it and in the stigma and both, UI and BI, are also observed in genera that do not possess a self-incompatibility system. The first to propose the action of specific genes independent of the S-locus in tuber-bearing Solanum species (which also possess a gametophytic self-incompatibility system) were Grun & Radlow (1961) and, later, Grun & Aubertin (1966), and Martin (1961, 1964). Hogenboom (1973, 1979), based on his studies in Lycopersicon, proposed the theory of incongruity. According to this theory, a series of genes in the pollen grains are involved in the control of pollen tube penetration along the style. These genes interact on a one to one basis with a series of genes in styles that set barriers to pollen tube penetration. Fertilization takes place only when there is congruency between the penetration and the barrier signals. Incongruity would, thus, be the result of evolutionary divergence. The presence of such systems of cross-incompatibility (CI) or incongruity could account for the results obtained in tuber-bearing Solanums by many authors, in both intra-specific and interspecific crosses (Grun & Aubertin, 1966; Hermsen & Sawicka, 1979; Camadro & Peloquin, 1981; Sala, 1993; Masuelli & Camadro, 1997; Camadro et al., 1998; Erazzú et al., 1999; among others). Camadro & Peloquin (1981) proposed a genetic model with dominant CI genes in styles that prevent fertilization by pollen grains carrying specific complementary dominant genes. The model, which assumes segregation for both type of loci, accounts for the results of both bilateral and unilateral incompatibility in interand intra-specific crosses and also accommodates the results obtained in Asparagus (Marcellán & Camadro, 1996). Post-zygotic barriers (hybrid inviability and hybrid weakness) were observed in the F. Camarosa (8 ) D.S.J. (8x) and Fragaria vesca (2x) F. ananassa (8x) crosses and their reciprocals (in which pre-zygotic barriers had been observed but that had, nevertheless, yielded a few achenes). The hystological study of F. vesca (2x) F. ananassa (8x) achenes revealed that the development of embryos and endosperm was poor, and in some serial microtome sections, embryos or endosperm were not observed. Similarly, in F ananassa (8x) F. vesca (2x) crosses, Yuhua Li et al. (2000) reported abnormalities not only in pollen grain germination and tube growth, but in embryo and endosperm development as well. Poor embryo development or abortion can be the result of unfavorable interactions between the genomes of both species in the embryo or to early abortion of the endosperm that leads to embryo starvation. But further investigations are needed to clarify the cause. With respect to the endosperm, various hypotheses have been put forward since 1930 to explain its development in interploid crosses. These hypotheses span from the need of a given relation among the ploidies of embryo, endosperm and maternal tissue to the need of a genetic balance in the endosperm between factors contributed by the female and the male parent to this tissue, which accommodates some unexpected results in tuber-bearing Solanums (in Johnston et al., 1980). However, none of them can accommodate the results obtained by several authors in strawberries (i.e. Scott, 1951; Darrow, 1966; Bringhurst, 1990), in which the function of gametes with reduced or unreduced chromosome numbers and the expected ploidy of the resulting hybrid cannot be predicted from the ploidies of the parents. The same holds true for the results of this study, although chromosome counts could not be carried out when achenes were obtained. The type of barriers acting in crosses in which P. tucumanensis was used as male could not be studied because fruits were not obtained from the crossing work. In addition, the numerous difficulties encountered when manipulating this species prevented the study of pollen-pistil compatibility relations. In summary, pre- and post-zygotic barriers were detected in the interspecific crosses studied. These barriers are not complete, therefore, it is possible to incorporate related germplasm into the cultivated pool. For the process to be efficient, further research is needed to clarify the basis of the detected barriers in order to devise strategies to circumvent them. Acknowledgements The authors thank Dr D. Micelli, from Laboratorio de Biología, Facultad de Bioquímica, UNT, for the use of optical microscopy and S. Salazar for excellent technical assistance in the crosses. This work was partially supported by the following grants: PICT 7227 and 7229 from Agencia Nacional de Promoción Científica ytecnológica (ANPCyT) and from CIUNT 26/D209.

11 149 References Abdullah, J.S. & M.J. Hennerty, Intergeneric hybrids of Fragaria and Potentilla. Acta Hort 348: Abdala, M.M.F. & J.G.Th. Hermsen, Unilateral incompatibility: hypothesis, debate and its implications for plant breeding. Euphytica 21: Arias, M., J. Diaz Ricci & A. Castagnaro, Changes in the distribution range of Potentilla tucumanensis (Rosaceae), an endangered cryptic species. Bol Soc Argent Bot 36: Asker, S., An intergeneric Fragaria Potentilla hybrid. Hereditas 64: Bringhurst, R.S., Cytogenetics and evolution in American Fragaria. Hortscience 25: Bringhurst, R.S. & F. Barrientos, Fertile Fragaria chiloensis and Potentilla glandulosa amphiploids. Abstr Genetics 74, Univ. Calif. Davis, USA. Camadro, E.L. & S.J. Peloquin, Cross-incompatibility between two sympatric polyploid Solanum species. Theor Appl Genet 60: Camadro, E.L., A.L. Verde & O.N. Marcellán, Pollenpistil incompatibility in a diploid hybrid potato population with cultivated and wild germoplasm. Amer J Potato Res 75: Castagnaro, A., Diaz Ricci, J., Arias M. & P. Albornoz, A new southern hemisphere species of Potentilla (Rosaceae). Novon 8: Darrow, G.M., The Strawberry. Holt, Rinehart & Winston, New York. Dionne, L.A., ,4-Dichlorophenoxyacetic acid as an aid to seed production when widely separated Solanum species are crossed. Nature 181: 361. Ellis, J.R., Fragaria-Potentilla intergeneric hybridization and evolution in Fragaria. Proc Linn Soc London 173: Erazzú, E.L., E.L.Camadro & A.M. Clausen, Pollen-style compatibility relations in natural populations of the wild diploid potato species Solanum spegazzinii Bitt. Euphytica 105: Evans, W., The use of synthetic octoploids in strawberry breeding. Euphytica 26: Evans, W.D. & J.K. Jones, Incompatibility in Fragaria.CanJ Gen Cytol 9: García, M.G., M. Ontivero, J.C. Diaz Ricci & A.P. Castagnaro, Morphological traits and high resolution RAPD markers for the identification of the main strawberry varieties cultivated in Argentina. Plant Breed 121: Grun, P. & M. Aubertin, The inheritance and expression of unilateral incompatibility in Solanum. Heredity 21: Grun, P. & A. Radlow, Evolution of barriers to crossing of self-incompatible with self-compatible species of Solanum. Heredity 21: Hadley, H.H. & S.J. Openshaw, Interspecific and intergeneric hybridization. In: W.R. Fehr & H.H. Hadley (Eds.), Hybridization of Crop Plants, pp Am Soc Agron, Crop Sci Soc of America, USA. Hancock, J.F., J.L. Maas, C.H. Shanks, P.J. Breen & J.J. Luby, Strawberry (Fragaria spp.). In: J.N. Moore & J. R. Ballington (Eds.), Genetic Resources in Temperate Fruit and Nut Crops, pp Int Soc Hort Sci, Wageningen, The Netherlands. Hermsen, J.G.Th. & E. Sawicka, Incompatibility and incongruity in tuber-bearing Solanum species. In: J.G. Hawkes, R.N. Lester & A.D. Skelding (Eds.), Linn Soc Symp Ser 7: Hogenboom, N.G., A model for incongruity in intimate partner relationships. Euphytica 21: Hogenboom, N.G., Incompatibility and incongruity in Lycopersicon. In: J.G. Hawkes, R.N. Lester & A.D. Skelding (Eds.), The Biology and Taxonomy of the Solanaceae, pp Academic Press, London, U.K. Johnston, S.A., T.P.M. den Nijs, S.J. Peloquin & R.E. Hanneman, Jr., The significance of genic balance to endosperm development in interspecific crosses. Theor Appl Genet 56: 5 9. Jones, J.K., Cytogenetic Studies in the Genera Fragaria and Potentilla. Ph. D. thesis, University of Manchester, UK. Kalkmam, C., Potentilla, Duchesnea, andfragaria in Malasia (Rosaceae). Blumea 16: Lewis, D. & L.K. Crowe, Unilateral interspecific incompatibility in flowering plants. Heredity 12: Luby, J.J., J.F. Hancock & J.C. Cameron, Expansion of the strawberry germoplasm base in North America. In: A. Dale & J. Luby (Eds.), The strawberry into the 21st Century, pp Timber Press, Portland, Oregon. Maas, J.L., Compendium of Strawberry Diseases. 2nd Edition. APS Press, USA. Mabberly, D.J., Potentilla and Fragaria reunited. Telopea 9: Macfarland Smith, W.H. & J.K. Jones, Intergeneric crosses with Fragaria and Potentilla. I. Crosses between Fragaria moschata and Potentilla fruticosa. Euphytica 34: Mangelsdorf, A.J. & E.M. East, Studies on the genetic of Fragaria. Genetic 12: Marcellán, O.N. & E.L. Camadro, Self- and crossincompatibility in Asparagus officinalis and Asparagus densiflorus cv. Sprengeri. Can J Bot 74: Marks, G.E., 1954, An acetocarmine glycerol jelly for use in pollen fertility counts. Stain Technol 29: 277. Martin, F.W., Staining and observing pollen tubes in the style by means of fluorescence. Stain Technol 34: Martin, F.W., Complex unilateral hybridization in Lycopersicon hirsutum. Proc Nat Acad Sci 47: Martin, F.W., The inheritance of unilateral incompatibilty between two tomato species. Genetics 50: Masuelli, R.W. & Camadro E.L., Crossability relationships among wild potato species with different ploidies and endosperm balance numbers (EBN). Euphytica 94: Murashige, T. & F. Skoog, A revised medium for rapid growth and bioassays with tobacco tissues cultures. Physiol Plant 15: Ontivero, M., M. Arias, J. Díaz Ricci, J. Babot, P. Albornoz & A. Castagnaro, Analysis of genetic similarities among species of Fragaria, Potentilla and Duchesnea growing in the northwest of Argentina by using morphological, anatomical and molecular characters. Can J Bot 78: Richardson, K., L. Jarret & E. Finke, Embedding in epoxy resins for ultra thin sectioning in electron microscopy. Stain Technol 35: Sala, C.A., Incompatibilidad cruzada entre cinco especies tuberosas de Solanum (Solanaceae). Darwiniana 32: Scott, D.H., Cytological studies on polyploids derived from tetraploid Fragaria vesca and cultived strawberries. Genetics 36: Spurr, A.R., A low-viscosity epoxy resin embedding medium for electron microscopy. J Ultrastruct Res 26: Staudt, G., Taxonomic studies in the genus Fragaria. Typification of Fragaria species known at the Linnaeus. Can J Bot 40: Staudt, G., The species of Fragaria, their taxonomy and geographycal distribution. Acta Hort 265:

12 150 Stone, L.S. & D.R. Goring, The molecular biology of selfincompatibility systems in flowering plants. Plant Cell, Tissue & Organ Culture 67: Trajkovski, K., Hybrisidering med Fragaria, en litteraturestudie. Sveriges Lantbruniversitet, Balsgard, Verksmhetsbetattelse : Wolf, T.H., Monographie der Gattung Potentilla. Bibl Bot, Hett 71. Yuhua Li, Xiunying Hou, Liang Lin, Shisi Jing & Mingqin Deng, Abnormal pollen germination and embryo abortion in the interspecific cross, Fragaria ananassa F. vesca, as related to cross-incompatibility. J Japan Soc Hort Sci 69: Zardini, E.M., Rosaceae. In: F.O. Zuloaga & O. Morrone (Eds.), Catálogo de las Plantas Vasculares de la República Argentina II. Fabaceae-Zygofyllaceae (Dicotyledoneae), pp Missouri Botanical Garden Press. St. Louis, Missouri.