**2. A general view of genetic relationships among cultivated citrus species**

It is suggested that the cultivated citrus derived from the three true species, citron, pummelo, and mandarin (Barrett & Rhodes, 1976). These three species reproduce sexually and if different cultivars within the species are intermated, the progeny are similar to their parents. The other important types (orange, grapefruit, lemon, and lime) are believed to have originated from one or more generations of hybridization between these ancestral genera. Most of the cultivars of orange, grapefruit, and lemon are believed to have originated from nucellar seedlings or budsports. Currently citrus fruits have high level of morphologic variations and various fruit characteristic because of inter and intraspecific interaction (Fig 1). Consequently, the amount of genetic diversity within these groups is relatively low, in spite of there being many named varieties. Conversely, mandarins, pummelos, and citrons have higher levels of genetic diversity since many of the cultivars have arisen through sexual hybridization (The Citrus and Date Crop Germplasm Committee, USA, CDCGC, 2004).

Fig. 1. Citrus fruits have distinct fruit characteristics.

Relationships among citrus species was invastigated previously. In a recent study, citrus accessions were divided into two large groups. The first group included citron, lemon, lime, rough lemon, with a similarity value of 0.60 from other Citrus species (Uzun et al., 2009a). The similar results were reported in previous studies (Federici et al., 1998; Nicolosi et al., 2000; Barkley et al., 2006). Citron was one of the progenitor of lemons (Nicolosi et al., 2000; Gulsen & Roose, 2001). It was reported that lemons were thought to be natural hybrids of a citron and a lime (Scora, 1975; Barrett & Rhodes, 1976) or a hybrid of citron and sour orange (Nicolosi et al., 2000; Gulsen & Roose, 2001). Limes are apparently hybrids of citrons and papedas (Scora, 1975) or a tri-hybrid cross of citron, pummelo, and Microcitrus, and had the highest observed heterozygosity of all the taxonomic groups (Barrett & Rhodes, 1976; Nicolosi et al., 2000; Barkley et al., 2006). Rough lemon was reported as hybrid of mandarin and citron (Scora, 1975). *C. volkameriana* was clustered with rough lemon as in the RAPD and SCAR study, and reported as a hybrid between citron and sour orange (Nicolosi et al., 2000).This finding supports the citron as major progenitor of some commercial citrus cultivars such as all 'true' lemons, limes and rough lemon.

Sweet orange, mandarin, sour orange, pummelo and grapefruit nested in same large group in previous study (Uzun et al., 2009a). This group separated two subgroup at similarity level of 0.64. The first subgroup included sweet oranges, mandarins and sweet oranges were separated from mandarins at 0.78. Parental sweet orange tree was a hybrid of pummelo and mandarin (Scora, 1975; Barrett & Rhodes, 1976), which was later supported by Nicolosi et al. (2000). Barkley et al. (2006) suggested that sweet orange has a majority of its genetic makeup from mandarin and only a small proportion from pummelo. The second subcluster included pummelo, grapefruit and sour orange. In this subcluster, pummelos and grapefruits were separated from sour oranges with a similarity value of 0.68. Pummelos and grapefruits showed a similarity level of 0.83. Grapefruit was reported as a hybrid of pummelo and sweet orange (Barrett & Rhodes, 1976; Nicolosi et al., 2000), and all grapefruit cultivars originated from single parent through mutations (Corazza-Nunes et al., 2002). Pummelo was indicated as one of the 'true basic species' in cultivated Citrus (Barrett & Rhodes, 1976). On the other hand, sour orange was reported as a hybrid of mandarin and pummelo in previous studies (Barrett and Rhodes, 1976; Barkley et al., 2006; Abkenar et al., 2007).

'Rangpur' lime (*C. limonia*) and bergamot (*C. bergamia*) were nested in the same branch and closely related to sour orange (Uzun et al., 2009a). Sour orange was reported as a hybrid of mandarin and pummelo in previous studies (Barrett & Rhodes, 1976; Barkley et al., 2006; Abkenar et al., 2007). Low level of genetic variation was found among sour oranges (Uzun, 2009). On the other hand, there was no polymorphism in sour oranges based on leaf isozymes (Torres et al., 1978) and SSR markers (Luro et al., 2000). Torres et al. (1978) reported that 'Rangpur' lime is quite different morphologically and genotypically from limes and was listed under *C. reticulata*. Nicolosi et al. (2000) indicated that 'Rangpur' was a hybrid of citron and mandarin and clustered with the citrons. According to Barkley et al. (2006), Webber (1943) believed that rangpurs were more similar to mandarins therefore, the origin and parentage of the rangpurs has been unclear, but they have generally been classified with mandarins in most previous studies. Hodgson (1967) suggested origin of bergamot was obscure, but probably related to sour orange. This accession was identified as a hybrid of citron and sour orange (Nicolosi et al., 2000) and clustered with sour orange (Federici et al., 1998).

Relationships among citrus species was invastigated previously. In a recent study, citrus accessions were divided into two large groups. The first group included citron, lemon, lime, rough lemon, with a similarity value of 0.60 from other Citrus species (Uzun et al., 2009a). The similar results were reported in previous studies (Federici et al., 1998; Nicolosi et al., 2000; Barkley et al., 2006). Citron was one of the progenitor of lemons (Nicolosi et al., 2000; Gulsen & Roose, 2001). It was reported that lemons were thought to be natural hybrids of a citron and a lime (Scora, 1975; Barrett & Rhodes, 1976) or a hybrid of citron and sour orange (Nicolosi et al., 2000; Gulsen & Roose, 2001). Limes are apparently hybrids of citrons and papedas (Scora, 1975) or a tri-hybrid cross of citron, pummelo, and Microcitrus, and had the highest observed heterozygosity of all the taxonomic groups (Barrett & Rhodes, 1976; Nicolosi et al., 2000; Barkley et al., 2006). Rough lemon was reported as hybrid of mandarin and citron (Scora, 1975). *C. volkameriana* was clustered with rough lemon as in the RAPD and SCAR study, and reported as a hybrid between citron and sour orange (Nicolosi et al., 2000).This finding supports the citron as major progenitor of some commercial citrus

Sweet orange, mandarin, sour orange, pummelo and grapefruit nested in same large group in previous study (Uzun et al., 2009a). This group separated two subgroup at similarity level of 0.64. The first subgroup included sweet oranges, mandarins and sweet oranges were separated from mandarins at 0.78. Parental sweet orange tree was a hybrid of pummelo and mandarin (Scora, 1975; Barrett & Rhodes, 1976), which was later supported by Nicolosi et al. (2000). Barkley et al. (2006) suggested that sweet orange has a majority of its genetic makeup from mandarin and only a small proportion from pummelo. The second subcluster included pummelo, grapefruit and sour orange. In this subcluster, pummelos and grapefruits were separated from sour oranges with a similarity value of 0.68. Pummelos and grapefruits showed a similarity level of 0.83. Grapefruit was reported as a hybrid of pummelo and sweet orange (Barrett & Rhodes, 1976; Nicolosi et al., 2000), and all grapefruit cultivars originated from single parent through mutations (Corazza-Nunes et al., 2002). Pummelo was indicated as one of the 'true basic species' in cultivated Citrus (Barrett & Rhodes, 1976). On the other hand, sour orange was reported as a hybrid of mandarin and pummelo in

previous studies (Barrett and Rhodes, 1976; Barkley et al., 2006; Abkenar et al., 2007).

'Rangpur' lime (*C. limonia*) and bergamot (*C. bergamia*) were nested in the same branch and closely related to sour orange (Uzun et al., 2009a). Sour orange was reported as a hybrid of mandarin and pummelo in previous studies (Barrett & Rhodes, 1976; Barkley et al., 2006; Abkenar et al., 2007). Low level of genetic variation was found among sour oranges (Uzun, 2009). On the other hand, there was no polymorphism in sour oranges based on leaf isozymes (Torres et al., 1978) and SSR markers (Luro et al., 2000). Torres et al. (1978) reported that 'Rangpur' lime is quite different morphologically and genotypically from limes and was listed under *C. reticulata*. Nicolosi et al. (2000) indicated that 'Rangpur' was a hybrid of citron and mandarin and clustered with the citrons. According to Barkley et al. (2006), Webber (1943) believed that rangpurs were more similar to mandarins therefore, the origin and parentage of the rangpurs has been unclear, but they have generally been classified with mandarins in most previous studies. Hodgson (1967) suggested origin of bergamot was obscure, but probably related to sour orange. This accession was identified as a hybrid of citron and sour orange (Nicolosi et al.,

cultivars such as all 'true' lemons, limes and rough lemon.

2000) and clustered with sour orange (Federici et al., 1998).

#### **3. Genetic diversity in orange**

In the cultivated citrus, sweet orange (*C. sinensis* L. Osbeck) originated as a natural hybrid between mandarin and pummelo (Barrett & Rhodes, 1976), showed low level of genetic diversity according to lots of previous studies (Luro et al., 1995; Novelli et al., 2000; Novelli et al., 2006; Uzun, 2009). It is notified that most of sweet oranges obtained by mutation from one ancestor tree. So despite of differences in morphological characters, genetic variation of sweet orange was low (Fang & Roose, 1997).

In recent study carried out using large amount of orange showed that there was high level of genetic similarity in oranges (Uzun, 2009). Similarity level of 250 orange accessions varied between 0.86 and 1.00. 'Chironja' was the most distant cultivar with 0.86 of similarity because of it derived from zygotic origin. This cultivar considered as a hybrid of orange and grapefruit and originated to Puerto Rico. Also it has large fruit and light yellow rind color (Fig. 1) (Hodgson 1967). Ambersweet had zygotic origin also separated clearly from other oranges. It is notified that this cultivar was a hybrid between a genotype obtained 'Orlando' tangelo X Clementine mandarin and unknown oranges (Jackson & Futch, 2003). Genetic similarity of all of other oranges was over 0.98 and some of them were identical. In this group there were many common orange cultivars and clones such as, many 'Washington Navel', 'Valencia' , 'Moro', 'Shamouti', 'Pineapple', 'Parson Brown', 'Salustiana', 'Sanguinello', 'Tarocco' and 'Yafa' clones introduced from other countries or selected in Turkey. On the other hand, lots of Turkish orange cultivars and clones for example, 'Agma', 'Alanya Dilimli', 'Dortyol Yerli', 'Kozan Yerli', 'Sultanhisar Yerli', also were existed in that group. Same results also reported in other studies. Barrett & Rhodes (1976) notified variations in orange, lemon, grapefruit and lime based on mutations occurred on one ancestor tree. Roose (1988), reported it was difficult to distinguish cultivars originated mutations using isozyme markers. Low level of polymorphism in orange also found with ISSR (Fang & Roose, 1997), SSR (Luro et al., 2000; Novelli et al., 2006), SRAP (Uzun et al., 2009a). On the other hand, no variation found in studied oranges in some researchs (Orford et al., 1995; Qing-Qin et al., 2007).

Orange cultivars are classified into four groups: common, low acidity, pigmented and navel oranges (Hodgson, 1967, as cited in Novelli et al., 2006). It is indicated despite the existence of substantial diversity among cultivated genotypes in respect of morphological, physiological and agronomic traits, very little DNA variation has been detected using DNAmarkers (Novelli et al., 2006). Same researcher found low level of genetic polymorphism among 41 orange cultivars. Similarity level of oranges varied between 0.96 and 1.00 and most of them were identical. They notified that sweet oranges have a narrow genetic basis and that most morphological characters originated through mutations, and clonal propagation of sweet oranges is the case for the majority of citrus species ((Herrero et al., 1996; as cited in Bretó et al., 2001). Fang & Roose (1997) used ISSR markers to differentiate 41 samples of orange belongs to three groups, Valencia, blood and navel based on fruit traits. All of these cultivars found almost the same ISSR fingerprints. This notified as majority of sweet orange cultivars derived from a single ancestor by mutation. However, some cultivar distinguished from others. Among the seven Valencia orange cultivars, only 'Midknight' differed from the other Valencias. Among the blood oranges, four of the five cultivars showed unique fingerprints for 1-3 fragments which distinguished them from all other sweet orange cultivars. Also 9 of 21 navel oranges had unique fingerprint patterns. Two Parent 'Washington' and 'Navel' samples obtained from different locations differed. It is explained as only case in which replicate samples of the same cultivar from different locations had different ISSR fingerprint patterns. It is indicated this result suggests that mutation occurred in at least one of them although horticultural traits are not known to between them (Fang &Roose, 1997). In other study, it was found identical microsatellite profiles at 9 out of 10 SSR loci among analyzed orange cultivars and clones (Hvarleva et al., 2008). For one locus 'Frost Valencia' and 'Shekeriko' acidless local Cyprus orange were discriminated from 'Shamouti', 'Jaffa', 'Valencia long' and 'Aematousiki' oranges. Researchers also investigated high level of similarity of genotypes, cultivars and clones is in contrast with the observed phenotypic variability among them (Fig. 2), indicating that the local cultivars were possibly derived through mutations which are not detectable by the used SSRs or they are clones of the same original cultivars. This is in accordance with the view that most of the orange cultivars were derived through mutations which affect mostly fruit traits.

Fig. 2. Fruit image of 'Chironja' (top) and 'Moro' (bottom) orange cultivars showed phenotypic variation among oranges( from Uzun, 2009).

explained as only case in which replicate samples of the same cultivar from different locations had different ISSR fingerprint patterns. It is indicated this result suggests that mutation occurred in at least one of them although horticultural traits are not known to between them (Fang &Roose, 1997). In other study, it was found identical microsatellite profiles at 9 out of 10 SSR loci among analyzed orange cultivars and clones (Hvarleva et al., 2008). For one locus 'Frost Valencia' and 'Shekeriko' acidless local Cyprus orange were discriminated from 'Shamouti', 'Jaffa', 'Valencia long' and 'Aematousiki' oranges. Researchers also investigated high level of similarity of genotypes, cultivars and clones is in contrast with the observed phenotypic variability among them (Fig. 2), indicating that the local cultivars were possibly derived through mutations which are not detectable by the used SSRs or they are clones of the same original cultivars. This is in accordance with the view that most of the orange cultivars

were derived through mutations which affect mostly fruit traits.

Fig. 2. Fruit image of 'Chironja' (top) and 'Moro' (bottom) orange cultivars showed

phenotypic variation among oranges( from Uzun, 2009).
