Abstract

Background and Aims The aims of this investigation were to highlight the qualitative and quantitative diversity apparent between nine diploid Fragaria species and produce interspecific populations segregating for a large number of morphological characters suitable for quantitative trait loci analysis.

Methods A qualitative comparison of eight described diploid Fragaria species was performed and measurements were taken of 23 morphological traits from 19 accessions including eight described species and one previously undescribed species. A principal components analysis was performed on 14 mathematically unrelated traits from these accessions, which partitioned the species accessions into distinct morphological groups. Interspecific crosses were performed with accessions of species that displayed significant quantitative divergence and, from these, populations that should segregate for a range of quantitative traits were raised.

Key Results Significant differences between species were observed for all 23 morphological traits quantified and three distinct groups of species accessions were observed after the principal components analysis. Interspecific crosses were performed between these groups, and F2 and backcross populations were raised that should segregate for a range of morphological characters. In addition, the study highlighted a number of distinctive morphological characters in many of the species studied.

Conclusions Diploid Fragaria species are morphologically diverse, yet remain highly interfertile, making the group an ideal model for the study of the genetic basis of phenotypic differences between species through map-based investigation using quantitative trait loci. The segregating interspecific populations raised will be ideal for such investigations and could also provide insights into the nature and extent of genome evolution within this group.

INTRODUCTION

Kellogg (1996) and Baum (2002) reviewed the experimental approaches that could be employed to explore the genetic changes underlying plant morphological evolution, through both functional genomics and classical genetics. Since the investigations of Bradshaw et al. (1995, 1998) a number of studies have employed map-based quantitative trait loci (QTL) analysis to increase the understanding of the genetic architecture of morphological diversity in plants (Vlot et al., 1992; Doebley and Stec, 1993; Fenster and Ritland, 1994; Fenster et al., 1995; Lin and Ritland, 1997), with the majority of these investigations focusing on morphological divergence between Mimulus species. As a model for undertaking such studies, a simple model plant system comprising a set of diploid species that display divergence in a large number of morphological traits, yet which remain inter-fertile so that mapping progenies can be raised, is required.

The genus Fragaria consists of approx. 20 species, with a base chromosome number of x = 7, that exist in four levels of ploidy. The genus has become economically important due to the world-wide cultivation of the hybrid octoploid species F. × ananassa (the cultivated strawberry) which produces large berries sold as dessert fruit and, to a lesser extent, the cultivation of the diploid ‘Alpine’ strawberry, F. vesca. The 12 diploid species of the genus occupy an impressive array of ecological niches and have adapted morphologically to their localized environments (Hancock and Luby, 1993), leading to many phenotypic differences between species. The diploid Fragaria also have a relatively small genome size (164 Mbp/C) that is roughly comparable with that of the model plant species Arabidopsis thaliana (Akiyama et al., 2001). This, together with their morphological diversity (Staudt, 1989), close relatedness (Potter et al., 2000) and interfertility (Bors, 2000) make the diploid Fragaria an excellent candidate model system for the study of phenotypic evolution through classical genetic approaches such as QTL analysis.

A number of monographs of the diploid Fragaria (Losina-Losinskaja, 1926; Staudt, 1999a), along with taxonomic reviews of many of the species within this group (Staudt, 1962, 1989, 1999b; Naruhashi and Iwata, 1988; Staudt and Dickoré, 2001) have been published, and thus they are a well-characterized species group. In addition, the extent to which the phenotype of F. vesca populations is controlled by environmental stimuli has been quantified (Hancock and Bringhurst, 1978; Jenson and Hancock, 1981), as has the quantitative variation at the morphological and molecular level within and between populations of F. vesca and the American octoploid species (F. chiloensis and F. virginiana) (Harrison et al., 1997, 2000).

However, no quantitative studies have been carried out to measure the extent of phenotypic variation within the diploid section of the genus. Whilst a number of phenotypes, such as perpetual flowering, non-runnering and yellow fruit mutants, have been described and have been shown to be under monogenic control (Brown and Wareing, 1965; Deng and Davis, 2001), the majority of variance in morphological traits observed within F. vesca is largely continuous (Staudt, 1999a) and therefore the phenotypic variance within and between diploid Fragaria species is likely to be under polygenic control. Thus an investigation to quantify the morphological variance that exists within and between the diploid Fragaria species could provide insights into the patterns of species divergence within this group and would permit the selection of morphologically diverse species for use in QTL mapping of morphological traits under polygenic control.

Fragaria vesca is a well-characterized species that has been the subject of many analyses at both the morphological and molecular levels and has a number of practical advantages for use as a parental species in the production of segregating populations for QTL mapping studies. Fragaria vesca is a self-compatible species, which facilitates the acceptance of pollen from both self-incompatible and self-compatible pollen donors (Lewis and Crowe, 1958; Evans and Jones, 1967), allowing a greater number of interspecific pollinations to be performed. Jones (1955) performed a number of interspecific crosses between diploid Fragaria species and reported successfully producing segregating back-cross (BC) populations between F. vesca and F. viridis. More recently, Bors (2000) performed an extensive series of 31 interspecific cross combinations between nine diploid Fragaria species and found that F. vesca was interfertile with five other diploid Fragaria species. In addition, it is a largely homozygous species with many lineages that are homozygous for a particular allele at the majority of molecular loci investigated, as demonstrated for both isoenzyme and microsatellite loci (Arulsekar and Bringhurst, 1981; Hadonou et al., 2004) and there are cultivars of F. vesca that have been well characterized and breed true to type from seed.

The purpose of this study was to determine the level of phenotypic variation between species accessions of nine diploid Fragaria species. Multivariate statistics were employed to characterize diploid Fragaria species with divergent morphologies. These data were then used to select morphologically divergent interfertile parents for the production of mapping populations for QTL analysis of the genetic basis of the heritable phenotypic differences that the species display. As F. vesca has been shown to be interfertile with a large number of other diploid Fragaria species, an accession of F. vesca subsp. vesca f. semperflorens ‘Baron Solemacher’ (FDP815) was chosen as the female parent for all interspecific crosses performed.

MATERIALS AND METHODS

Plant material

Nineteen accessions representing examples from nine different species of diploid Fragaria (including one previously undescribed species) were obtained from a number of germplasm repositories and private collections (Table 1), either as vegetatively propagated material in the form of runner plants, or as seed. Mother plants were established in a glasshouse under ambient conditions. All accessions were cytometrically tested to confirm their diploid status against known diploid, hexaploid and octoploid standards by Plant Cytometry Services, NL.

Table 1.

Fragaria species accessions tested in this study, their geographic origin and sources from which they were obtained

Plant material
 
EMR accession no.
 
Origin
 
Source*
 
Source accession no.
 
F. daltoniana FDP001 Unknown EMR, UK FDP001 
F. iinumae FDP201 Hokkaido, Japan NCGR, USA CFRA1008.000 
F. sp. FDP301 Utar Pradesh, EMR, UK FDP301 
F. nilgerrensis FDP401 Yunnan, China IPK, Germany 94056-12.K 
F. nilgerrensis FDP405 Guizhou, China NCGR, USA CFRA1188.000 
F. nilgerrensis FDP406 Hubei, China NCGR, USA CFRA1610.000 
F. nipponica FDP501 Japan IPK, Germany 98134-09.K 
F. nipponica FDP502 Japan AFFRC, Japan na 
F. nubicola FDP601 Pakistan IPK, Germany 94056-33.K 
F. pentaphylla FDP701 China IPK, Germany 94059-01.P 
F. pentaphylla FDP702 China IPK, Germany 99116-02.K 
F. vesca subsp. vesca FDP801 Hokkaido, Japan AFFRC, Japan na 
F. vesca subsp. vesca FDP804 Finland NCGR, USA CFRA438.000 
F. vesca subsp. vesca FDP805 Probably UK EMR, UK FDP805 
F. vesca subsp. americana FDP806 NH, USA University of New Hampshire, USA na 
F. vesca f. semperflorens FDP807 Cultivar ‘Yellow Wonder’ University of New Hampshire, USA na 
F. viridis FDP901 Dohna, Germany IPK, Germany 92046-01.P 
F. viridis FDP902 Gotland, Sweden IPK, Germany 96148-01.P 
F. viridis FDP903 Tian-Shan, China IPK, Germany 94056-13.K 
Plant material
 
EMR accession no.
 
Origin
 
Source*
 
Source accession no.
 
F. daltoniana FDP001 Unknown EMR, UK FDP001 
F. iinumae FDP201 Hokkaido, Japan NCGR, USA CFRA1008.000 
F. sp. FDP301 Utar Pradesh, EMR, UK FDP301 
F. nilgerrensis FDP401 Yunnan, China IPK, Germany 94056-12.K 
F. nilgerrensis FDP405 Guizhou, China NCGR, USA CFRA1188.000 
F. nilgerrensis FDP406 Hubei, China NCGR, USA CFRA1610.000 
F. nipponica FDP501 Japan IPK, Germany 98134-09.K 
F. nipponica FDP502 Japan AFFRC, Japan na 
F. nubicola FDP601 Pakistan IPK, Germany 94056-33.K 
F. pentaphylla FDP701 China IPK, Germany 94059-01.P 
F. pentaphylla FDP702 China IPK, Germany 99116-02.K 
F. vesca subsp. vesca FDP801 Hokkaido, Japan AFFRC, Japan na 
F. vesca subsp. vesca FDP804 Finland NCGR, USA CFRA438.000 
F. vesca subsp. vesca FDP805 Probably UK EMR, UK FDP805 
F. vesca subsp. americana FDP806 NH, USA University of New Hampshire, USA na 
F. vesca f. semperflorens FDP807 Cultivar ‘Yellow Wonder’ University of New Hampshire, USA na 
F. viridis FDP901 Dohna, Germany IPK, Germany 92046-01.P 
F. viridis FDP902 Gotland, Sweden IPK, Germany 96148-01.P 
F. viridis FDP903 Tian-Shan, China IPK, Germany 94056-13.K 
*

AFFRC, National Institute for Vegetable Research, Morioka, Japan; EMR, East Malling Research, East Malling, UK; IPK, Institute of Plant Genetics and Crop Plant Research, Fruit Gene-Bank, Dresden-Pillnitz, Germany; NCGR, National Clonal Germplasm Repository, Corvallis, USA.

Trial plants

Ten 6-week-old clonally propagated plants of each accession, derived either from runners or by dividing the crowns of the mother plant, were used to establish a field trial. Two replicate plots of each accession were planted in July 2002 in a single row at a spacing of 60 cm on a 1-m-wide bed covered with Mypex mulch. Plants were allowed to establish and overwinter naturally, during which time all runners and dead plant matter were removed periodically.

Character measurement

A total of 23 morphological traits (Table 2) representing both sexual and asexual reproductive traits, vegetative plant characteristics and characters relating to the yield and vigour of the plants were measured quantitatively over the course of one growing season. Qualitative assessments of the plants were done using at least two accessions of each species and were made on trial plants at East Malling, UK, and on additional plant material held at the IPK Fruit Gene Bank, Dresden-Pillnitz, Germany. A range of morphological traits was recorded including vegetative, floral and fruit characteristics, plant habit, general vigour and any other unusual characteristics observed.

Table 2.

Characters measured for analysis of variance and principal component analysis

Trait
 
Grand mean
 
Mean range
 
d.f.
 
s.e.m.*
 
Flower diameter (FD) 21·21 14·4–28·2 0·428 
Receptacle diameter (RD) 4·027 2·7–5·6 0·100 
Flower ratio (FD : RD) 5·431 3·8–8·8 0·152 
Petal number 5·559 5–7·4 0·131 
Leaflet length (LL) 57·37 35·2–80·7 1·026 
Leaflet width (LW) 40·19 27·7–55 0·874 
Leaflet ratio (LL : LW) 1·439 1·1–1·8 0·018 
Leaflet length to widest point 26·38 16·2–37·3 0·717 
Leaflet area 15·48 6–25·6 0·509 
Longest petiole length 211·63 133·6–300·8 3·927 
Petiole diameter 2·372 1·4–3·8 0·043 
Total number of stolons 24·95 10·8–44·8 1·136 
Stolon internode length 17·91 8·2–29·6 0·588 
Stolon diameter 1·262 0·8–2·4 0·040 
Berry length (BL) 15·144 11·3–19·7 0·351 
Berry width (BW) 16·088 12·3–20·7 0·326 
Berry ratio (BL : BW) 0·950 0·7–1·5 0·018 
Berry weight 1·509 0·6–2·6 0·079 
Calyx diameter 39·29 26·9–50·4 0·755 
Peduncle length 107·2 28·2–254·2 5·7 
Primary pedicel length 42·63 0–106·2 3·552 
Flowers per inflorescence 7·4 1–14·8 0·467 
Total number of flowers 72·4 1·6–238·4 6·59 
Trait
 
Grand mean
 
Mean range
 
d.f.
 
s.e.m.*
 
Flower diameter (FD) 21·21 14·4–28·2 0·428 
Receptacle diameter (RD) 4·027 2·7–5·6 0·100 
Flower ratio (FD : RD) 5·431 3·8–8·8 0·152 
Petal number 5·559 5–7·4 0·131 
Leaflet length (LL) 57·37 35·2–80·7 1·026 
Leaflet width (LW) 40·19 27·7–55 0·874 
Leaflet ratio (LL : LW) 1·439 1·1–1·8 0·018 
Leaflet length to widest point 26·38 16·2–37·3 0·717 
Leaflet area 15·48 6–25·6 0·509 
Longest petiole length 211·63 133·6–300·8 3·927 
Petiole diameter 2·372 1·4–3·8 0·043 
Total number of stolons 24·95 10·8–44·8 1·136 
Stolon internode length 17·91 8·2–29·6 0·588 
Stolon diameter 1·262 0·8–2·4 0·040 
Berry length (BL) 15·144 11·3–19·7 0·351 
Berry width (BW) 16·088 12·3–20·7 0·326 
Berry ratio (BL : BW) 0·950 0·7–1·5 0·018 
Berry weight 1·509 0·6–2·6 0·079 
Calyx diameter 39·29 26·9–50·4 0·755 
Peduncle length 107·2 28·2–254·2 5·7 
Primary pedicel length 42·63 0–106·2 3·552 
Flowers per inflorescence 7·4 1–14·8 0·467 
Total number of flowers 72·4 1·6–238·4 6·59 
*

All trait variance significant Fpr < 0·001.

Floral induction and crossing

F1, F2 and backcrosses (BC) were performed using the parental lines listed in Table 3. With the aim of inducing flowering, 16-week-old parental lines that had been clonally propagated either from runner plants or by dividing the crowns of the mother plants were potted into compost in 2-L plastic pots and placed in a growth room at 15 °C with a photoperiod of 8 h for 4 weeks. The plants were then transferred to an unlit cold-store maintained at 2 °C for a further 4 weeks, after which time they were returned to a glasshouse maintained at 18 °C with a photoperiod of at least 12 h. The plants were kept under these final conditions throughout the length of their flowering period.

Table 3.

Crosses between diploid Fragaria species and the fertility of the progeny raised

Female parent
 
Male parent
 
Type of cross
 
No. of crosses
 
No. of viable seeds
 
Mean seeds per cross
 
No. of seeds sown
 
Germination (%)
 
Survival (%)
 
Fertility
 
F. vesca 815 F. daltoniana 001 F1 156 52 20 11 73 Plants not flowered 
F. vesca 815 F. iinumae 201 F1 – – – No viable seed 
F. vesca 815 F. sp. 301 F1 203 51 20 16 88 Plants not flowered 
F. vesca 815 F. nilgerrensis 401 F1 207 51·8 30 40 50 Plants not flowered 
F. vesca 815 F. nipponica 501 F1 148 49·33 50 62 All plants died 
F. vesca 815 F. nubicola 601 F1 84 42 25 76 100 Plants flowered 
F. vesca 815 F. pentaphylla 702 F1 136 45·3 30 57 94 Plants flowered 
F. vesca 815 F. viridis 903 F1 172 57·3 50 92 98 Plants flowered 
FDP815 × 601 FDP815 × 601 F2 227 75·7 20 98 93 Plants flowered 
F. vesca 815 FDP815 × 601 BC 106 53 20 85 94 Plants flowered 
F. vesca 815 FDP815 × 903 BC 488 61 20 85 71 Plants flowered 
Female parent
 
Male parent
 
Type of cross
 
No. of crosses
 
No. of viable seeds
 
Mean seeds per cross
 
No. of seeds sown
 
Germination (%)
 
Survival (%)
 
Fertility
 
F. vesca 815 F. daltoniana 001 F1 156 52 20 11 73 Plants not flowered 
F. vesca 815 F. iinumae 201 F1 – – – No viable seed 
F. vesca 815 F. sp. 301 F1 203 51 20 16 88 Plants not flowered 
F. vesca 815 F. nilgerrensis 401 F1 207 51·8 30 40 50 Plants not flowered 
F. vesca 815 F. nipponica 501 F1 148 49·33 50 62 All plants died 
F. vesca 815 F. nubicola 601 F1 84 42 25 76 100 Plants flowered 
F. vesca 815 F. pentaphylla 702 F1 136 45·3 30 57 94 Plants flowered 
F. vesca 815 F. viridis 903 F1 172 57·3 50 92 98 Plants flowered 
FDP815 × 601 FDP815 × 601 F2 227 75·7 20 98 93 Plants flowered 
F. vesca 815 FDP815 × 601 BC 106 53 20 85 94 Plants flowered 
F. vesca 815 FDP815 × 903 BC 488 61 20 85 71 Plants flowered 

Pollen was collected from male parental lines and stored in a desiccator at 4 °C until required. Female parents were emasculated at the ‘white bud’ stage and the emasculated receptacle was covered with one half of a gelatine pill capsule to protect it from rogue pollen. The emasculated receptacles were pollinated after 24, 48 and 72 h and the pill capsules were replaced after each pollination. Once the receptacles had begun to swell, the pill capsules were removed and the berry was allowed to develop normally. Seeds from crosses were collected and stored at 4 °C until they were sown. All seeds were germinated by surface sowing on a 3-cm-deep layer of Levington's F2 seed compost in 70 % humidity at 18 °C for 4 weeks, or until 80 % of the seeds had germinated. Self-pollination experiments were performed on all species accessions by placing a pollination hood over plants that had initiated flowers in the glasshouse. Open flowers were self-pollinated with an artist's paint brush each day until the flowers set fruit or the receptacles became blackened.

Fragaria vesca subsp. vesca f. semperflorens ‘Baron Solemacher’ (FDP815) was used for all F1 crossing experiments as the female parent. FDP815 was used in preference over the ‘semperflorens’ variety included in the morphological investigation, as it is homozygous for a recessive mutation of a nuclear-encoded chlorophyll gene that gives the plants a pale green leaf phenotype. This allowed any rogue ‘selfed’ seedlings to be identified at the cotyledon stage.

Data analysis

The means of each of the 23 quantitative characters recorded were compared to test for significant differences at both the species and accession levels using a hierarchical general analysis of variance (ANOVA), with species accessions nested within species, using Genstat 6 for Windows (VSN, UK). Principal components analysis (PCA) was used as it summarizes patterns of correlations among observed variables and reduces a large number of observed characters to a smaller number of derived variables or components. It is designed to find linear combinations of the original variables that maximize the observed variance (Tabachnick and Fidell, 1989). It therefore allows the variance within all the phenotypic traits to be considered simultaneously, and so those taxa that were maximally separated in multidimensional space were selected as parents for the proposed investigation of heritable phenotypic diversity. A PCA was performed using 14 mathematically unrelated morphological traits selected from the 23 traits recorded (Table 4). These were chosen to encompass all aspects of the plants' morphology. The data were standardized by subtracting the accession mean from the grand mean for each character and dividing by the trait standard deviation to produce a standardized character matrix. The analysis was performed using MVSP for Windows (MVSP v3.1; http://www.kovcomp.co.uk/mvsp/) and the first three principal components were plotted using the graphical interface of MVSP.

Table 4.

Factor loadings of the 14 characters for the first three principal components and the percentage variance accounted for

Trait
 
Axis 1
 
Axis 2
 
Axis 3
 
Flower diameter 0·26 0·04 
Receptacle diameter 0·11 −0·22 0·33 
Leaflet width −0·11 −0·33 −0·4 
Leaflet length to widest point −0·16 −0·16 −0·53 
Longest petiole length −0·26 −0·22 0·2 
Petiole diameter 0·54 −0·37 −0·24 
Total number of stolons −0·04 0·28 −0·27 
Stolon internode length −0·07 −0·44 −0·14 
Stolon diameter 0·4 −0·37 0·07 
Berry length 0·04 0·11 0·10 
Calyx diameter 0·12 −0·14 0·4 
Peduncle length −0·41 −0·19 −0·09 
Number of flowers per inflorescence −0·19 −0·31 0·21 
Total number of flowers −0·38 −0·25 0·22 
Percentage 36·29 25·77 12·16 
Cumulative percentage 36·29 62·05 74·21 
Trait
 
Axis 1
 
Axis 2
 
Axis 3
 
Flower diameter 0·26 0·04 
Receptacle diameter 0·11 −0·22 0·33 
Leaflet width −0·11 −0·33 −0·4 
Leaflet length to widest point −0·16 −0·16 −0·53 
Longest petiole length −0·26 −0·22 0·2 
Petiole diameter 0·54 −0·37 −0·24 
Total number of stolons −0·04 0·28 −0·27 
Stolon internode length −0·07 −0·44 −0·14 
Stolon diameter 0·4 −0·37 0·07 
Berry length 0·04 0·11 0·10 
Calyx diameter 0·12 −0·14 0·4 
Peduncle length −0·41 −0·19 −0·09 
Number of flowers per inflorescence −0·19 −0·31 0·21 
Total number of flowers −0·38 −0·25 0·22 
Percentage 36·29 25·77 12·16 
Cumulative percentage 36·29 62·05 74·21 

RESULTS

Qualitative assessment of accessions

Figure 1 shows the morphological diversity displayed by the eight positively identified species of diploid Fragaria included in this investigation. Table 5 summarizes the findings of the qualitative assessment of these eight species and highlights the key morphological differences observed for the vegetative and reproductive traits studied. Whilst many qualitative differences were observed in almost all traits studied, contrasts were most easily recognized in the floral and fruiting traits. Fragaria iinumae and F. nilgerrensis had, in general, a stout, more robust phenotype that contrasted with the more gracile habit of many of the other species and, in general, plants of F. vesca, F. viridis and F. nilgerrensis had more fecund phenotypes. In addition, accessions of F. daltoniana, F. nipponica, F. nubicola and F. viridis displayed a deciduous, or cryptophytic, nature and F. iinumae displayed an annual vegetative habit. The diploid Fragaria accession, labelled as F. sp. 301, was not included in the qualitative analysis as it was not attributed to any of the currently described diploid species of Fragaria.

Table 5.

A summary of the qualitative morphological characteristics assessed in eight diploid Fragaria species

Species
 
Plant and stolon habit
 
Foliage
 
Inflorescence structure
 
Flowers
 
Fruit
 
F. daltoniana Small; low-growing; short petioles; few hairs on all plant parts; deciduous habit. Small waxy leaflets; prominent petiolules Single flowered; fully erect in flower and fruit; flowers above foliage; peduncles absent Small; ovate, prominently veined and widely spaced petals; calyx clearly visible; round receptacle; large flat anthers; short stamens Conic to cylindrical; bright pink skin; purple to black achenes; white flesh; woolly texture; no aroma 
 Stolons deep red; sympodial     
F. iinumae Small; stout plants; thick leathery petioles and stolons; annual habit, only daughter plants surviving the winter. Glabrous leaves with large serrations; short petiolules Few flowers per inflorescence; few inflorescences per plant; short peduncles Medium-sized; long slender petals, always more than five; small receptacles; large numbers of anthers; short stamens Conic; bright red skin; soft watery flesh; bright yellow achenes; unpleasant, blackcurrant-like flavour; acidic 
 Stolons pinky-red; thick and leathery; sympodial     
F. nilgerrensis Robust; vigorous; long, thick leathery petioles; thick pubescence on all plant parts; evergreen habit. Thick, leathery leaves; leaflets almost round; deep, prominent veins; distinct petiolules Complex inflorescence, supporting many flowers; thick peduncles, erect to semi-erect in fruit Medium sized; prominent, large round receptacles; narrow petals do not overlap; calyx clearly visible; large flat anthers Globose-conic fruit; white skin; deeply pitted; firm flesh; small brown achenes; highly aromatic, peach-like aroma; clasping to spreading calyx 
 Stolons deep-red; thick and leathery; sparsely produced; sympodial     
F. nipponica Small; compact deciduous habit. Stolons pinky-red; slender and filiform; support a large number of daughter plants; monopodial Many leaves that fold upwards with a bluish hue; insignificant petiolules Simple inflorescence; peduncles droop when in fruit Medium-sized; very small receptacle; almost circular petals; long prominent outer filaments Short wedge-shaped berries; pinky-red skin; pleasant aroma like that of cultivated strawberry; long prominent calyx 
F. nubicola Small; not vigorous; few leaves; short petioles; remontant flowering habit; deciduous habit. Light-green leaflets, becoming darker around veins; short, almost absent petiolules Simple inflorescence; few inflorescences per plant; peduncles clothed in thick pubescence; lie along ground in fruit; remontant flowering habit Large; prominently veined petals small receptacle Globose, necked berries; flattened on top; dark wine-red skin; tightly clasping calyx 
 Stolons deep red; sympodial     
F. pentaphylla Compact; densely leafed; wavy petiolesStolons deep red; monopodial Leaflets shiny, almost leathery; petioles support tertiary leaflets Relatively simple inflorescence; slender peduncles; flowers above foliage Large; large, contorted, wrinkled petals Long conic berries, almost rectangular; knobbly appearance; pink to orangy-red skin; sunken achenes.White fruited form has larger, rounder berries 
F. vesca Robust and vigorous; long, gracile petioles; evergreen habit; americana subsp. is in general smaller, more gracile and less vigorous Leaflets large; often curled under at edges Complex inflorescence; peduncles stand erect in flower; americana subsp. has characteristically long pedicels and very short peduncles Small; round, prominently veined petals; large receptacle; small anthers held on short filaments;Subsp. americana has smaller flowers Long-conic to globose; usually bright red; distinctive, powerful aroma; large achenes raised or in shallow pits; clasping to reflex calyx 
 Stolons green to brown; many slender, filiform stolons; sympodial     
F. viridis Slender and erect; long, slender petioles; deciduous habit. Yellowish leaflets; petiolules almost absent Complex inflorescence; peduncles above leaves in flower but lie along ground in fruit; remontant flowering habit Large; medium sized receptacles; large, overlapping petals Oblate to globose berries; pale green skin with a red blush; firm flesh; acidic apple-like aroma; very large achenes 
 Stolons pinky-red; monopodial     
Species
 
Plant and stolon habit
 
Foliage
 
Inflorescence structure
 
Flowers
 
Fruit
 
F. daltoniana Small; low-growing; short petioles; few hairs on all plant parts; deciduous habit. Small waxy leaflets; prominent petiolules Single flowered; fully erect in flower and fruit; flowers above foliage; peduncles absent Small; ovate, prominently veined and widely spaced petals; calyx clearly visible; round receptacle; large flat anthers; short stamens Conic to cylindrical; bright pink skin; purple to black achenes; white flesh; woolly texture; no aroma 
 Stolons deep red; sympodial     
F. iinumae Small; stout plants; thick leathery petioles and stolons; annual habit, only daughter plants surviving the winter. Glabrous leaves with large serrations; short petiolules Few flowers per inflorescence; few inflorescences per plant; short peduncles Medium-sized; long slender petals, always more than five; small receptacles; large numbers of anthers; short stamens Conic; bright red skin; soft watery flesh; bright yellow achenes; unpleasant, blackcurrant-like flavour; acidic 
 Stolons pinky-red; thick and leathery; sympodial     
F. nilgerrensis Robust; vigorous; long, thick leathery petioles; thick pubescence on all plant parts; evergreen habit. Thick, leathery leaves; leaflets almost round; deep, prominent veins; distinct petiolules Complex inflorescence, supporting many flowers; thick peduncles, erect to semi-erect in fruit Medium sized; prominent, large round receptacles; narrow petals do not overlap; calyx clearly visible; large flat anthers Globose-conic fruit; white skin; deeply pitted; firm flesh; small brown achenes; highly aromatic, peach-like aroma; clasping to spreading calyx 
 Stolons deep-red; thick and leathery; sparsely produced; sympodial     
F. nipponica Small; compact deciduous habit. Stolons pinky-red; slender and filiform; support a large number of daughter plants; monopodial Many leaves that fold upwards with a bluish hue; insignificant petiolules Simple inflorescence; peduncles droop when in fruit Medium-sized; very small receptacle; almost circular petals; long prominent outer filaments Short wedge-shaped berries; pinky-red skin; pleasant aroma like that of cultivated strawberry; long prominent calyx 
F. nubicola Small; not vigorous; few leaves; short petioles; remontant flowering habit; deciduous habit. Light-green leaflets, becoming darker around veins; short, almost absent petiolules Simple inflorescence; few inflorescences per plant; peduncles clothed in thick pubescence; lie along ground in fruit; remontant flowering habit Large; prominently veined petals small receptacle Globose, necked berries; flattened on top; dark wine-red skin; tightly clasping calyx 
 Stolons deep red; sympodial     
F. pentaphylla Compact; densely leafed; wavy petiolesStolons deep red; monopodial Leaflets shiny, almost leathery; petioles support tertiary leaflets Relatively simple inflorescence; slender peduncles; flowers above foliage Large; large, contorted, wrinkled petals Long conic berries, almost rectangular; knobbly appearance; pink to orangy-red skin; sunken achenes.White fruited form has larger, rounder berries 
F. vesca Robust and vigorous; long, gracile petioles; evergreen habit; americana subsp. is in general smaller, more gracile and less vigorous Leaflets large; often curled under at edges Complex inflorescence; peduncles stand erect in flower; americana subsp. has characteristically long pedicels and very short peduncles Small; round, prominently veined petals; large receptacle; small anthers held on short filaments;Subsp. americana has smaller flowers Long-conic to globose; usually bright red; distinctive, powerful aroma; large achenes raised or in shallow pits; clasping to reflex calyx 
 Stolons green to brown; many slender, filiform stolons; sympodial     
F. viridis Slender and erect; long, slender petioles; deciduous habit. Yellowish leaflets; petiolules almost absent Complex inflorescence; peduncles above leaves in flower but lie along ground in fruit; remontant flowering habit Large; medium sized receptacles; large, overlapping petals Oblate to globose berries; pale green skin with a red blush; firm flesh; acidic apple-like aroma; very large achenes 
 Stolons pinky-red; monopodial     
Fig. 1.

Plants of eight diploid Fragaria species in flower showing their range of morphological variation: (A) FDP001 Fragaria daltoniana; (B) FDP201 F. iinumae; (C) FDP401 F. nilgerrensis; (D) FDP501 F. nipponica; (E) FDP601 F. nubicola; (F) FDP702 F. pentaphylla; (G) FDP901 F. viridis; (H) FDP804 F. vesca subsp. vesca.

Fig. 1.

Plants of eight diploid Fragaria species in flower showing their range of morphological variation: (A) FDP001 Fragaria daltoniana; (B) FDP201 F. iinumae; (C) FDP401 F. nilgerrensis; (D) FDP501 F. nipponica; (E) FDP601 F. nubicola; (F) FDP702 F. pentaphylla; (G) FDP901 F. viridis; (H) FDP804 F. vesca subsp. vesca.

Quantitative results

The results of the hierarchical general analysis of variance revealed significant differences for all traits evaluated at both the accession (F ≥ 6·19, P ≤ 0·001) and species (F ≥ 35·2, P ≤ 0·001) levels (Table 2). Whilst significant differences were observed for all traits, there was not a uniform spread of variation and groupings or classes could be observed for many traits.

Principal components analysis

The PCA variable loadings, percentage and cumulative percentage variance for the first three principal components are given in Table 4. The first three principal components accounted for 74·2 % of the total variance observed. The first principal component (PC1) accounted for 36·3 % of the total variance, and had high contributing factor loadings from petiole diameter, peduncle length, stolon diameter and total number of flowers. The second principal component (PC2) had high contributing factor loadings from stolon internode length, stolon and petiole diameter, leaflet width and number of flowers per inflorescence and contributed 25·8 % of the total variation. The third principal component (PC3) accounted for 12·2 % of the total variation, with high factor loadings for leaflet length to widest point, leaflet width, calyx diameter and receptacle diameter.

Figure 2 illustrates the accession distribution in the first three principal components and clearly shows the separation of the 19 accessions studied. Accessions of F. vesca and F. nilgerrensis were maximally separated by the morphological variance summarized by PC1, with accessions FDP807 (F. vesca) and FDP401 (F. nilgerrensis) being most widely separated on this axis due to the larger flowers, thicker stolons and petioles, and fewer stolons per plant of F. nilgerrensis, amongst other characters. Whilst separation of other species accessions by PC1 was less than for F. vesca and F. nilgerrensis, the total variance summarized in the first three principal components between F. vesca and accessions of other species was consistently high. The analysis separated the accessions into three distinct groupings, one containing members of the species F. vesca, one containing members of F. nilgerrensis and a third grouping of all other species accessions. The diploid Fragaria accession identified as F. sp. 301 that was not included in the qualitative analysis was compared quantitatively with the other species accessions within the trial, and was shown to be quantitatively similar to F. pentaphylla FDP702.

Fig. 2.

Principal component analysis scatter diagram based on the correlation of 14 quantitative morphological variables in 19 accessions of nine diploid Fragaria species.

Fig. 2.

Principal component analysis scatter diagram based on the correlation of 14 quantitative morphological variables in 19 accessions of nine diploid Fragaria species.

According to the PCA, accessions of all other species were quantitatively distinct from accessions of F. vesca, with accessions of F. nilgerrensis most distinct. Therefore crosses between F. vesca and accessions of the eight other species with which F. vesca produced fertile F1 progeny would potentially allow the production of segregating populations for QTL analysis. Crosses were therefore performed on to F. vesca 815 using representative accessions of the eight other diploid species included in this investigation as pollen donors.

Floral induction and pollination

Of the nine species that were included in this trial, only accessions of the species F. nipponica, F. nubicola, F. pentaphylla, F. vesca, F. viridis and F. sp. 301 responded to the floral induction treatment. Accessions of F. nilgerrensis flowered after overwintering in a glasshouse environment and so pollen was collected and self-compatibility could be determined from these plants. The F. daltoniana and F. iinumae accessions, however, did not produce flowers in the glasshouse environment, and thus pollen from these accessions was collected from plants from the field experimental trial but the self-compatibility status of these species could not be assessed.

The outcomes of the crosses performed, the number of seeds produced per cross, the percentage seed germination and percentage seedling survival, as well as the fertility of hybrid progeny raised, are summarized in Table 3. Self-pollination experiments showed that accessions of F. nipponica, F. nubicola, F. pentaphylla, F. viridis and F. sp. 301 were all self-incompatible, and F. nilgerrensis and F. vesca were self-compatible. The fertility status of F. iinumae 201 and F. daltoniana 001 could not be determined from this study as the experimental plants had not responded to the artificial floral induction procedure.

F1 crosses were successfully performed between F. vesca and seven other diploid Fragaria species. Of these, progeny raised from three crosses (using F. nubicola, F. pentaphylla and F. viridis as pollen donors) flowered during the course of this experiment. Of the others, three failed to flower and all plants of the fourth died at the cotyledon stage (Table 3). From the three fertile F1 progenies, F2 and BC populations were raised from two (F. vesca 815 × F. nubicola 601 and F. vesca 815 × F. viridis 903). Germination and survival rates for the three progenies raised (F. viridis BC, F. nubicola BC and F2) were comparable when 20 seeds were sown (Table 3) and all segregated for the recessive yellow leaf mutation in the expected 3 : 1 or 1 : 1 Mendelian ratios (χ2 = 0·2–0·8, P = 0·37–0·65).

DISCUSSION

Both qualitative and quantitative analyses were performed on eight of the 12 diploid Fragaria species known and an accession of a putative species nova (F. sp. 301) was included in the quantitative assessments. Despite a number of monographs (Losina-Losinskaja, 1926; Staudt, 1999a) and taxonomic reviews (Staudt, 1962, 1989, 1999b; Naruhashi and Iwata, 1988; Staudt and Dickoré, 2001) there have been no previously reported quantitative studies comparing the diploid Fragaria, and this study highlights clear quantitative morphological differentiation between the species. Examples of the remaining four described diploid Fragaria species were not included in this investigation because living material was unavailable for study.

The analyses revealed some previously unreported qualitative morphological characteristics and the quantitative analyses revealed significant variation between species for all traits quantified. The PCA partitioned the accessions studied into three distinct groupings, and on the basis of these groupings, crosses were made with the eight other species on to F. vesca which was shown to be significantly quantitatively distinct from all other species in this investigation. From the interspecific crosses performed between these parental accessions, it was possible to raise three interspecific populations that should segregate for a large number of morphological traits.

Qualitative and quantitative assessment of morphological characters

The present investigation has shown that clear morphological differences exist between all eight described species for almost all traits studied (Fig. 1 and Table 5). Indeed, many contrasts in fecundity, vigour, productivity and size can be recognized between the different species and the present observations agree with previously published descriptors of the eight described species investigated (for review, see Losina-Losinskaja, 1926). In addition, this evaluation has highlighted some characteristics that have not been reported previously: the deciduous habit of F. daltoniana, F. nipponica, F. nubicola and F. pentaphylla; the apparent annual nature of F. iinumae (where only stolon daughter plants that had established in the summer season survived the following winter in the field); and the remontant flowering habit of both F. nubicola and F. viridis. It should be noted that all plants were grown under identical environmental conditions and therefore the key morphological features of interest identified for the eight different species may not be characteristic of these species when the plants are observed in their natural environments. However, since the plants were observed in a common environment, the morphological variation that was observed can be attributed to genetic divergence between the species accessions under investigation.

For all 23 of the traits measured there were significant differences between the 19 species accessions. However, not all comparisons between all accessions for all traits were significant. When plotted in three-dimensional space (Fig. 2), three distinct groups of species accessions were recognized. The differences observed in the data and summarized in the PCA showed that accessions of F. vesca were quantitatively similar to each other, but distinct from all other species accessions. Similarly, F. nilgerrensis accessions were quantitatively different from other species accessions, whilst the variance between accessions of the other seven species (including F. sp. 301) was much lower, as summarized by the first three principal components. Whilst resolution of these species accessions through the quantitative assessment was low with the number of accessions studied here, qualitative differences, such as the bright pink fruit and waxy leaflets of F. daltoniana, the thick leathery petioles and stolons of F. iinumae, and the tertiary leaflets and wavy petioles of F. pentaphylla, allowed clear, unambiguous identification of all species investigated.

Interspecific crossing experiments

Fragaria vesca and F. nilgerrensis, two evolutionarily divergent species (Potter et al., 2000), were maximally separated by PC1, with a smaller degree of separation on PC2 and PC3, and thus a cross between these species was the initial choice from which to raise a mapping progeny. However, F1 plants from this cross were all stunted, displaying a compact, dwarf habit and failed to flower. The F. vesca × F. nilgerrensis hybrids of Jones (1955) that were likewise stunted, failed to flower over the course of two growing seasons. The production of segregating F2 or BC populations from this F1 progeny was therefore not possible.

F1 crosses were performed to F. vesca 815 with accessions of all seven other species; however, segregating F2 and BC populations were raised from just two. Three interspecific crosses of F. vesca (FDP815) pollinated by the self-incompatible species F. nubicola, F. pentaphylla and F. viridis were successful. All three pollen donor species were morphologically distinct from F. vesca, and were therefore all potentially useful as parents from which to produce a segregating progeny. Bors (2000) reported similar levels of seed set, germination and fertility to those in the current study for these three cross combinations and reported all three F1 hybrids to be fertile under open-pollinated conditions. However, he did not report on the habit and vigour of these hybrids, or the ease of production of either BC or F2 progenies. Jones (1955) reported successfully producing BC progenies using F1 (F. viridis × F. vesca) hybrids as pollen donors when back-crossed to F. vesca and also as females when back-crossed with F. viridis. He reported that the F. vesca × (F. vesca × F. viridis) progeny showed considerable variation in both morphology and vigour and reported the segregation of a number of morphological characters within the population, such as fruit colour, perpetual flowering and stolon habit.

The F1 hybrid seedlings produced from F. vesca × F. pentaphylla in this study were much less uniform than those of either the F. nubicola or F. viridis combinations, with plants ranging from stunted and weak seedlings that produced few flowers and runners, to healthy, vigorous, fecund phenotypes. The F1 progeny of this cross also had an unusual intermediate stolon habit that did not initiate clonal daughter plants. This meant that only single plants of each genotype were available for crossing experiments, and these did not initiate sufficient numbers of flowers to allow an adequate number of pollinations to be done. It was, therefore, decided to raise progeny from F1 populations that had either F. nubicola or F. viridis as a male parent.

Plants of five diploid Fragaria species were karyotyped by Iwatsubo and Naruhashi (1989, 1991) who found that F. daltoniana, F. nipponica, F. nubicola and F. vesca all displayed morphologically similar karyotypes that differed significantly from that of the phylogenetically distinct F. iinumae, which they reported as being probably the most divergent of the diploid Fragaria. Crosses between F. vesca and the three species to which it has been reported to be karyologically similar were all successful and thus the karyological divergence of F. iinumae from F. vesca, suggesting large chromosomal differentiation between the two species, most probably accounts for the failure of the cross F. vesca 815 × F. iinumae 201. As of August 2004, the F1 hybrids of F. vesca × F. daltoniana 001 and F. vesca × F. sp. 301 had established well, but had failed to flower. Seed of the cross of F. vesca × F. nipponica germinated well; however, all seedlings died at the cotyledon stage. This is consistent with the findings of Bors (2000) who postulated a post-zygotic lethal barrier to the success of this hybrid cross.

In accordance with the findings of Jones (1955), an F2 progeny could not be raised by selfing the F1 plants from the F. viridis cross. These findings are in accordance with other reports of the self-incompatibility of F. viridis (Staudt, 1953; Evans and Jones, 1967), and the present results which showed F. viridis did not set seed when selfed. Therefore, only a BC progeny was raised using the F. viridis F1 parent. However, it was possible to produce an F2 population using the F. nubicola × F. vesca F1 hybrid, indicating a possible breakdown in the operation of the self-incompatibility in this cross.

Germination and survival rates for the three progenies raised were high and all segregated for the pale green leaf mutation in the expected 3 : 1 (F2) or 1 : 1 (BC) Mendelian ratios. This indicated adequate pairing between chromosomes from both species within these interspecific crosses and no pre- or post-zygotic mechanisms affecting the allele transmission rate to the seedlings, at least for this gene.

Concluding remarks

A quantitative morphological comparison of eight of the 12 currently described diploid Fragaria species has been presented, along with an accession of a previously unreported species (F. sp. 301) that is morphologically similar to F. pentaphylla. Whilst highlighting clear morphological differentiation of F. nilgerrensis and F. vesca from the other diploid Fragaria species, the resolution of other species groups within this study based on morphometric evidence was less clear. Despite this, these species show clear qualitative differences and thus by referring to both quantitative and qualitative data, it was possible to adequately distinguish morphologically between all species examined.

By referring to the morphological data generated from the quantitative assessment when selecting parents for interspecific crosses, it has been possible to produce fertile interspecific populations of plants from combinations of parents that are morphologically divergent for a large number of traits. From the quantitative differences observed between the parental phenotypes of the crosses performed, the progenies would also be expected to segregate for a large number of quantitative morphological traits. They therefore constitute an ideal model for the study of the genetic basis of the heritable phenotypic differences displayed between the species, through QTL analysis.

Despite their morphological diversity, the diploid Fragaria are remarkably closely related (Potter et al., 2000), suggesting morphological differentiation and speciation may be the result of stochastic genetic change (such as chromosomal rearrangement) in addition to the gradual accumulation of mutation over time. Davis et al. (1995) provided tentative evidence of chromosomal translocation events since the divergence of F. vesca and F. viridis and thus, in addition to providing a model for QTL analysis of morphology, the interspecific populations presented herein offer the opportunity to study the nature and evolution of genome organization between the diploid Fragaria through comparative map-based investigations.

This research was supported by funds from the University of Reading Research Endowment Trust Fund, the East Malling Trust for Horticultural Research, the Worshipful Company of Fruiterers and DEFRA. D.J.S. is indebted to Ken Tobutt for his critical reviews of the manuscript and for many useful and enlightening discussions. The authors would also like to thank C. Campbell and K. Evans for reviewing the manuscript.

LITERATURE CITED

Akiyama Y, Yamamoto Y, Ohmido N, Oshima M, Fukui K.
2001
. Estimation of the nuclear DNA content of Strawberries (Fragaria spp.) compared with Arabidopsis thaliana by using Dual-step Flow Cytometry.
Cytologia
 
66
:
431
–436.
Arulsekar S, Bringhurst RS.
1981
. Genetic model for the enzyme marker PGI in diploid California Fragaria vesca.
Journal of Heredity
 
73
:
117
–120.
Baum DA.
2002
. Identifying the genetic causes of phenotypic evolution: a review of experimental strategies. In: Cronk QBC, Bateman RM, Hawkins JA. eds. Developmental genetics and plant evolution. London: Taylor and Francis, 493–507.
Bors RH.
2000
.A streamlined synthetic octoploid system that emphasises Fragaria vesca as a bridge species. PhD Thesis, University of Guelph.
Bradshaw HD, Wilbert SM, Otto KG, Schemske DW.
1995
. Genetic-mapping of floral traits associated with reproductive isolation in monkeyflowers (Mimulus).
Nature
 
376
:
762
–765.
Bradshaw HD, Otto KG, Frewen BE, McKay JK, Schemske DW.
1998
. Quantitative trait loci affecting differences in floral morphology between two species of monkeyflower (Mimulus).
Genetics
 
149
:
367
–382.
Brown T, Wareing PF.
1965
. The genetical control of the everbearing habit and three other characters in varieties of Fragaria vesca.
Euphytica
 
14
:
97
–112.
Davis TM, Yu H, Haymes, KM.
1995
. Genetic linkage group structure in the diploid strawberry: evidence for chromosome differentiation between Fragaria vesca and F. viridis. In: Pritts MP, Chandler CK, Crocker TE, eds. Proceedings of the IV North American Strawberry Conference. Horticultural Sciences Department, University of Florida, 213–216.
Deng C, Davis TM.
2001
. Molecular identification of the yellow fruit color (c) locus in diploid strawberry: A candidate gene approach.
Theoretical and Applied Genetics
 
103
:
316
–322.
Doebley J, Stec A.
1993
. Inheritance of the morphological differences between maize and teosinte: comparison of results for two F2 populations.
Genetics
 
134
:
559
–570.
Evans WD, Jones JK.
1967
. Incompatibility in Fragaria.
Canadian Journal of Genetics and Cytology
 
9
:
831
–836.
Fenster CB, Diggle PK, Barrett SCH, Ritland K.
1995
. The genetic of floral development differentiating two species of Mimulus (Scrophulariaceae).
Heredity
 
74
:
258
–266.
Fenster CB, Ritland K.
1994
. Quantitative genetics of mating systems divergence in the yellow monkeyflower species complex.
Heredity
 
73
:
422
–435.
Hadonou AM, Sargent DJ, Wilson F, James CM, Simpson DW.
2004
. Development of microsatellite markers in Fragaria, their use in genetic diversity analysis and their potential for genetic linkage mapping.
Genome
 
47
:
429
–438.
Hancock JF, Bringhurst RS.
1978
. Inter-populational differentiation and adaptation in the perennial, diploid species Fragaria vesca L.
American Journal of Botany
 
65
:
795
–803.
Hancock JF, Luby JJ.
1993
. Genetic resources at our doorstep: the wild strawberries.
BioScience
 
43
:
141
–147.
Harrison RE, Luby JJ, Furnier GR, Hancock JF.
1997
. Morphological and molecular variation among populations of octoploid Fragaria virginiana and F. chiloensis (Rosaceae) from North America.
American Journal of Botany
 
84
:
612
–620.
Harrison RE, Luby JJ, Furnier GR, Hancock JF.
2000
. Differences in the apportionment of molecular and morphological variation in North American strawberry and the consequences for genetic resource management.
Genetic Resources and Crop Evolution
 
47
:
647
–657.
Iwatsubo Y, Naruhashi N.
1989
. Karyotypes of three species of Fragaria (Rosaceae).
Cytologia
 
54
:
493
–497.
Iwatsubo Y, Naruhashi N.
1991
. Karyotypes of Fragaria nubicola and F. daltoniana (Rosaceae).
Cytologica
 
56
:
453
–457.
Jenson RJ, Hancock JF.
1981
. Multivariate relationships among California strawberries.
Bulletin of the Torrey Botanical Club
 
109
:
136
–147.
Jones JK.
1955
.Cytogenetic studies in the genera Fragaria and Potentilla. PhD Thesis, Manchester University, UK.
Kellogg EA.
1996
. Integrating genetics, phylogenetics, and developmental biology. In: Sobral BS, ed. The impact of molecular genetics. Boston: Birkhauser, 159–172.
Lewis D, Crowe LK.
1958
. Unilateral interspecific incompatibility in flowering plants.
Heredity
 
12
:
233
–256.
Lin JZ, Ritland K.
1997
. Quantitative trait loci differentiating the outbreeding Mimulus guttatus from the inbreeding M. platycalyx.
Genetics
 
146
:
1115
–1121.
Losina-Losinskaja AS.
1926
. Revision critique du genre Fragaria.
Bulletin du Jardin Botanique USSR
 .
25
:
47
–88 [in Russian].
Naruhashi N, Iwata T.
1988
. Taxonomic re-evaluation of Fragaria nipponica Makino and allied species.
Journal of Phytogeography and Taxonomy
 
36
:
59
–64.
Potter D, Luby JJ, Harrison RE.
2000
. Phylogenetic relationships among species of Fragaria (Rosaceae) inferred from non-coding nuclear and chloroplast DNA sequences.
Systematic Botany
 
25
:
337
–348.
Staudt G.
1953
. Zur Selbsterilitat von Fragaria viridis Duch.
Naturwissenschaften
 
24
:
572
–574.
Staudt G.
1962
. Taxonomic studies on the genus Fragaria, typification of Fragaria species known at the time of Linnaeus.
Canadian Journal of Botany
 
40
:
212
–218.
Staudt G.
1989
. The species Fragaria, their taxonomy and geographical distribution.
Acta Horticulturae
 
265
:
23
–33.
Staudt G.
1999
.Systematics and geographic distribution of the American strawberry species. Taxonomic studies in the genus Fragaria (Rosaceae: Potentilleae). Botany Vol. 81. University of California.
Staudt G.
1999
. Notes on Asiatic Fragaria species: Fragaria nilgerrensis Schltdl. ex J. Gay.
Botanische Jahrbücher für Systematik, Pflanzengeschichte und Pflanzengeographie
 
121
:
297
–310.
Staudt G, Dickoré WB.
2001
. Notes on Asiatic Fragaria species: Fragaria pentaphylla Losinsk. and Fragaria tibetica spec. nov.
Botanische Jahrbücher für Systematik, Pflanzengeschichte und Pflanzengeographie
 
123
:
341
–354.
Tabachnick BG, Fidell LS.
1989
. Principal components and factor analysis. In: Tabachnick BG, Fidell LS. Using multivariate statistics, 2nd edn. New York: Harper Collins.
Vlot EC, Vanhouten WHJ, Mauthe S, Bachmann K.
1992
. Genetic and non-genetic factors influencing deviations from five pappus parts in a hybrid between Microseris douglasii and M. bigelovii.
International Journal of Plant Sciences
 
153
:
89
–97.

Author notes

1East Malling Research, New Road, East Malling, Kent ME19 6BJ, UK, 2Institute fur Pflanzengenetik und Kulturpflanzenforschung, Genbank Obst, Dresden-Pillnitz, Germany and 3School of Plant Sciences, University of Reading, Whiteknights, PO Box 221, Reading RG6 6AS, UK

Comments

0 Comments