**4. Genetic diversity in mandarin**

Mandarin was considered as one of the true citrus species (Barrett & Rhodes, 1976) and this idea supported by following researches (Nicolosi et al., 2000; Barkley et al., 2006; Uzun et al., 2009a). Mandarin group has great amount of cultivars and some of them originated from hybridization and the others derived from mutation. So, in the mandarins obtained from hybrid origin there was clear genetic variation. On the other hand, low level of diversity observed in the cultivars occurred by mutation such as Satsuma and Clementine groups (Breto et al., 2001; Barkley et al., 2006; Uzun et al., 2011a).

In the recent study carried out with SRAP markers, mandarins separated two large groups at similarity level of 0.79 (Uzun, 2009). Satsuma and Clementine mandarins was nested in different groups. 'King' mandarin (*Citrus nobilis* Loureiro) late maturing and has large fruit cultivar found closely to Satsuma group. It was also reported by Coletto Filho et al. (1998) that genetic similarity among mandarins was over 0.77 and Satsuma and 'King' nested in the same gruop. Nicolosi et al. (2000) found that Satsuma and King closely related according to their RAPD and SCAR data. 'Kara' reported as a hybrid from 'Owari' Satsuma X 'King' (Hodgson, 1967), grouped closely to Satsuma mandarins according to SRAP data (Uzun, 2009). Low level of genetic variation found in most Satsumas and similarity level of nearly 50 accessions was over 0.98. They separated several group including 2-15 accessions and there was no genetic differences into the each groups. In the another study, there was no variation among the 16 Satsuma mandarin and it was notified these genotypes obtained from mutations. In the same way, Fang and Roose (1997), found no differences in five Satsuma cultivars and Barkley et al. (2006), reported nearly all cultivars in the Satsuma group originated from mutations and they had same genetic construction. Hodgson (1967), classified Satsuma as seperate group in the mandarins and notified that Satsuma naming as *Citrus unshiu* Markovitch. Also same researcher reported Satsuma was a nonstable group and lots of cultivars and genotypes had been occurred by variations in this group.

In citrus many economically important genotypes are obtained from hybridization. In the mandarin group there are lots of hybrid accessions derived from mandarin x mandarin, mandarin x pummelo (as tangelo), mandarin x orange (as tangor) or mandarin x tangelo. According to SRAP data, all tangelos and tangors closely related to mandarin instead of orange and pummelo (Uzun, 2009). On the other hand, 'Ellendale', 'Ortanique', 'Mandora', 'Lake' tangelo, 'Orlando' tangelo (Fig. 3.), 'Thornton', tangelo, 'Seminello' tangelo, 'Sampson' tangelo and 'Robinson' and 'Nova' mandarins (both cultivars are result of Clementine X Orlando) grouped closely. 'Ortanique' and 'Mandora' were nearly identical and these cultivars were showed as synonym (Cottin, 2002). On the other hand, 'Ortanique' was reported as a natural hybrid between orange and mandarin (Hodgson, 1967). Also 'Robinson' and 'Nova' shared same parents had the high level of similarity. 'Dancy' mandarin separated clearly from other mandarins and nested alone in the dendrogram obtained by SRAP data (Uzun, 2009). According to Hodgson (1967), this cultivar classified as a species by Tanaka (1954) and originated from India. 'Tankan', 'Ponkan', 'Minneola' tangelo (Fig. 3.), 'Batangas', 'Swatow' nucellar and 'Fuzhu' grouped closely. 'Ponkan' and 'Batangas' were notified as a synonym (Cottin, 2002). Besides, 'Tankan' was reported as hybrid of mandarin and orange (Coletta Filho et al., 1998). Other cultivars have hybrid origin that 'Fremont', 'Kinnow' and 'Murcott' were nested together. 'Fremont' obtained from crossing of 'Clementine' X 'Ponkan' and 'Murcot' known as hybrid of mandarin and orange (Hodgson, 1967). 'Sunburst', 'Fairchild', 'Encore' and 'Bower'clustered in same small group (Uzun, 2009). Also 'Sunburst' and 'Fairchild'found as closely related and clustered same group in other study (Barkley et al., 2006). 'Fairchild' was a hybrid of 'Clementine' X 'Orlando' and 'Encore' was a hybrid of 'King' X 'Willowleaf' (Hodgson, 1967). 'Fortune' as a hybrid between 'Clementine' and 'Dancy' (Hodgson, 1967), was found more related to 'Clementine' instead of 'Dancy' (Barkley et al., 2006; Uzun, 2009). 'Lee' as 'Clementine' hybrid clustered closely to 'Clementine' than other 'Clementine' hybrids such as 'Nova' and 'Robinson' (Uzun, 2009).

Fig. 3. 'Minneola' (top) and 'Orlando' (bottom) tangelos have same parentage (hybrid of Duncan grapefruit and Dancy mandarin) and contributed to the parentage of the such mandarin.

and 'Bower'clustered in same small group (Uzun, 2009). Also 'Sunburst' and 'Fairchild'found as closely related and clustered same group in other study (Barkley et al., 2006). 'Fairchild' was a hybrid of 'Clementine' X 'Orlando' and 'Encore' was a hybrid of 'King' X 'Willowleaf' (Hodgson, 1967). 'Fortune' as a hybrid between 'Clementine' and 'Dancy' (Hodgson, 1967), was found more related to 'Clementine' instead of 'Dancy' (Barkley et al., 2006; Uzun, 2009). 'Lee' as 'Clementine' hybrid clustered closely to 'Clementine' than other 'Clementine' hybrids

Fig. 3. 'Minneola' (top) and 'Orlando' (bottom) tangelos have same parentage (hybrid of Duncan grapefruit and Dancy mandarin) and contributed to the parentage of the such

mandarin.

such as 'Nova' and 'Robinson' (Uzun, 2009).

Fig. 4. Dendrogram showed relationships among 42 Clementine accessions based on SRAP markers (from Uzun et al., 2011a).

Clementine (*Citrus clementina* Hort. ex Tan.) was classified as a *Citrus* species (Tanaka, 1977). Currently, this species is one of the most important mandarin hybrid especially in the Mediterranean countries due to its good fruit quality and flavour, high yield, easy peeling. A lot of Clementine clones with high quality and different maturity time were obtained from clonal selection and most of them registered as new cultivars. Bud mutations often arise in Clementine, as it is the case also for orange and Satsuma mandarin, which are generally detected by the growers in branches of trees showing altered horticultural traits, such as maturity and flowering time, or fruit characteristics (Breto et al., 2001). Contrasting with this diversity for agronomic traits, very low genetic variability has been found in cultivated citrus (Fang & Roose, 1997; Federici et al., 1998; Coletta-Filho et al., 1998; Luro et al., 2000; Breto et al., 2001; Corazza-Nunes et al., 2002, Uzun et al., 2011a). According to recent study carried out with 42 Clementine accession using SRAP markers genetic similarity of Clementine mandarins over the 0.96 (Uzun et al., 2011a). As a Turkish selection 'Clementine A 64' seperated from other accessions. The rest of 41 accessions divided two groups at 0.99 (A and B). Group A consisted of six Clementine accessions' selected in Turkey and all of them were identical (Fig. 4). Group B included 35 foreign Clementine accessions and 34 of them were identical. Only 'Clementine SRA 87' was distinguished from others in this group.

In a study (Uzun et al., 2011a) genetic diversity of Clementine accessions was very low and most of them were indistinguished. Seven accessions originated in Turkey were separated from foreign accessions. Six of seven Turkish selection were identical. Turkish and other accessions grouped based on their geographic origin whereas accessions originated from other countries such as SRA series from France except 'Clementine SRA 87', 'Fino' from Spain and 'Algerian Tangerina' from Algeria, were indistinguished. It can be explained that Turkish accessions have low level of polymorphism due to long period of cultivation in Turkey without influence of foreign cultivars. It is reported previously that there was limited number of polymorphism in Clementine based on RAPD data and it is suggested Clementines were genetically similar (Russo et al., 2000). Breto et al. (2001), found low level of polymorphism, distinguished only two accessions and notified Clementines are vegetatively propagated and the new cultivars are obtained after careful selection of spontaneous somatic mutations. Luro et al. (2000) speculated that the microsatellites could not distinguish mutation-derived species such as sweet and sour orange, whereas polymorphism was detected among lemon and citron cultivars.

#### **5. Genetic diversity in lemon and relatives (citron, rough lemon,**  *C. volkameriana***)**

Citron that major progenitor of some commercial Citrus cultivars such as all true lemons and rough lemon was reported as one of the "basic" true Citrus species and (Barrett & Rhodes 1976; Gulsen & Roose 2001). Lemon (*C. limon* (L.) Burm. f.) was accepted as a species by two important taxonomic systems (Swingle & Reece 1967; Tanaka, 1977), but it has been reported as a hybrid by other studies (Barrett & Rhodes 1976; Torres et al., 1978; Herrero et al., 1996). Besides, lemon was notified as a hybrid of citron and sour orange (*C. aurantium* L.) in recent studies (Nicolosi et al. 2000; Gulsen & Roose 2001). Most lemons have highly similar morphological and biochemical characters, and some are reported to have originated by mutation from a single parental lemon tree. Rough lemon (*Citrus jambhiri* Lush) was reported to be closely related with the citrons in previous studies (Federici et al., 1998; Nicolosi et al., 2000; Barkley et al., 2006; Pang et al., 2007) and was also reported as a hybrid of mandarin and citron (Scora 1975; Nicolosi et al., 2000; Barkley et al., 2006). *Citrus volkameriana* was reported as a hybrid between citron and sour orange (Nicolosi et al., 2000).

arise in Clementine, as it is the case also for orange and Satsuma mandarin, which are generally detected by the growers in branches of trees showing altered horticultural traits, such as maturity and flowering time, or fruit characteristics (Breto et al., 2001). Contrasting with this diversity for agronomic traits, very low genetic variability has been found in cultivated citrus (Fang & Roose, 1997; Federici et al., 1998; Coletta-Filho et al., 1998; Luro et al., 2000; Breto et al., 2001; Corazza-Nunes et al., 2002, Uzun et al., 2011a). According to recent study carried out with 42 Clementine accession using SRAP markers genetic similarity of Clementine mandarins over the 0.96 (Uzun et al., 2011a). As a Turkish selection 'Clementine A 64' seperated from other accessions. The rest of 41 accessions divided two groups at 0.99 (A and B). Group A consisted of six Clementine accessions' selected in Turkey and all of them were identical (Fig. 4). Group B included 35 foreign Clementine accessions and 34 of them were identical. Only 'Clementine SRA 87' was distinguished from others in

In a study (Uzun et al., 2011a) genetic diversity of Clementine accessions was very low and most of them were indistinguished. Seven accessions originated in Turkey were separated from foreign accessions. Six of seven Turkish selection were identical. Turkish and other accessions grouped based on their geographic origin whereas accessions originated from other countries such as SRA series from France except 'Clementine SRA 87', 'Fino' from Spain and 'Algerian Tangerina' from Algeria, were indistinguished. It can be explained that Turkish accessions have low level of polymorphism due to long period of cultivation in Turkey without influence of foreign cultivars. It is reported previously that there was limited number of polymorphism in Clementine based on RAPD data and it is suggested Clementines were genetically similar (Russo et al., 2000). Breto et al. (2001), found low level of polymorphism, distinguished only two accessions and notified Clementines are vegetatively propagated and the new cultivars are obtained after careful selection of spontaneous somatic mutations. Luro et al. (2000) speculated that the microsatellites could not distinguish mutation-derived species such as sweet and sour orange, whereas

polymorphism was detected among lemon and citron cultivars.

**5. Genetic diversity in lemon and relatives (citron, rough lemon,** 

Citron that major progenitor of some commercial Citrus cultivars such as all true lemons and rough lemon was reported as one of the "basic" true Citrus species and (Barrett & Rhodes 1976; Gulsen & Roose 2001). Lemon (*C. limon* (L.) Burm. f.) was accepted as a species by two important taxonomic systems (Swingle & Reece 1967; Tanaka, 1977), but it has been reported as a hybrid by other studies (Barrett & Rhodes 1976; Torres et al., 1978; Herrero et al., 1996). Besides, lemon was notified as a hybrid of citron and sour orange (*C. aurantium* L.) in recent studies (Nicolosi et al. 2000; Gulsen & Roose 2001). Most lemons have highly similar morphological and biochemical characters, and some are reported to have originated by mutation from a single parental lemon tree. Rough lemon (*Citrus jambhiri* Lush) was reported to be closely related with the citrons in previous studies (Federici et al., 1998; Nicolosi et al., 2000; Barkley et al., 2006; Pang et al., 2007) and was also reported as a hybrid of mandarin and citron (Scora 1975; Nicolosi et al., 2000; Barkley et al., 2006). *Citrus volkameriana* was reported as a hybrid between citron and sour orange

this group.

*C. volkameriana***)** 

(Nicolosi et al., 2000).

In a recent study genetic diversity in citron, lemon, rough lemon and *C. volkameriana* group was carried out using SRAP and SSR markers. They evalutaed 56 accessions were evaluated (Uzun et al., 2011b). Similarity level of citrons to other accessions were ~0.70. Four citron accessions that 'Buddhas Hand' (fingered citron, Fig. 5) , 'Etrog' and two Turkish selections were distinguished clearly (Fig. 6). Similarly, Gulsen and Roose (2001), reported that similarity level of citron and lemon-rough lemon group was 0.65 based on their ISSR data. On the other hand, according to Uzun et al. (2011b), genetic similarity among lemons and rough lemon-*C. volkameriana* group was 0.80. Rough lemons and *C. volkameriana* were closely related. At the same way, *C. volkameriana* was clustered with rough lemon as in the RAPD (Luro et al., 1992) and SCAR (Nicolosi et al., 2000) based studies.

Fig. 5. Fruit shape of 'Buddhas Hand' (fingered citron)

It is reported there was low level of polymorphism among most of lemons derived via clonal selection whereas higher genetic diversity was found in lemons which had hybrid origin (Uzun et al., 2011b). Genetic similarity of 45 lemons included both mutation and hybrid origin was notified between 0.80-1.00 (Fig. 6). The most distinct cultivars were 'Ponderosa' and 'Song Panache'. 'Ponderosa' was suggested as citron like fruits and notified as monoembryonic with large fruits and hybrid of citron and lemon (Hodgson 1967; Kahn et al., 2001). At the same way, 'Ponderosa' was the most distant cultivar from the other lemons based on leaf isozyme data (Torres et al., 1978). On the other hand, 'Song Panache' lemon have similar-shaped fruit as 'Ponderosa'. Another hybrid origin lemon 'Meyer' also separated from other lemons based on SRAP and SSR data (Uzun et al., 2011b). This cultivar was classified as lemon-resembling fruit by Hodgson (1967). Some Turkish lemons had hybrid origin such as 'Tuzcu 09 Aklimon' and its selections 'Tuzcu 896' and 'Tuzcu 897' was distinguished. These lemons were also found as distinct from others in previous studies (Aka-Kacar et al., 2005; Uzun et al., 2009b).

Interdonato' lemons was clearly separated from others and genetic similarity among 'Interdonato' lemons was very high caused by their mutation origin (Uzun et al., 2011b). This cultivar was reported as a hybrid between lemon and citron (Hodgson 1967; Gulsen and Roose 2000). "Interdonato" was found as apart from other lemons in previous studies carried out with different marker systems (Deng et al. 1995; Gulsen and Roose 2001).

Fig. 6. Dendrogram showed relationships among citron, rough lemon, *C. volkameriana* and lemons (a; citrons, b: C volkameriana, c: rough lemons, d: lemons; from Uzun et al., 2011b)

distinguished. These lemons were also found as distinct from others in previous studies

Interdonato' lemons was clearly separated from others and genetic similarity among 'Interdonato' lemons was very high caused by their mutation origin (Uzun et al., 2011b). This cultivar was reported as a hybrid between lemon and citron (Hodgson 1967; Gulsen and Roose 2000). "Interdonato" was found as apart from other lemons in previous studies

carried out with different marker systems (Deng et al. 1995; Gulsen and Roose 2001).

Fig. 6. Dendrogram showed relationships among citron, rough lemon, *C. volkameriana* and lemons (a; citrons, b: C volkameriana, c: rough lemons, d: lemons; from Uzun et al., 2011b)

(Aka-Kacar et al., 2005; Uzun et al., 2009b).

It is notified that although they were distinguished genetic similarity of most of lemon was very high (~0.92-1.00) (Uzun et al., 2011b). Similar knowledge also was allowed by Gulsen & Roose (2001) and they indicated most lemons originated from mutation. This group included some lemons from different countries such as 'Femminello', Zagara Bianca', Carruboro', 'Continella', 'Limoneira 8A', 'Santa Teresa', 'Verna', 'Monachello' and several from Turkey that 'Kutdiken', 'Yediveren', Italian Memeli', 'Lamas', 'Kibris'. There was high level of similarities between Turkish and other lemon accessions and there was no clear clustering according to their origin. Although most lemons studied were distinguished, diversity level among these lemons was low (Uzun et al., 2011b).

As an important citrus rootstock sour orange was investigated partly in terms of genetic diversity in some studies. Recently, Lombardo et al. (2011) studied genetic variability of eight sour oranges using ISSR markers. They found a very low level of genetic variability among the cultivars; 'Canaliculata' formed a separated cluster with orange, suggesting a probable hybrid origin derived from crossing between sour and sweet orange. On the other hand, six sour oranges shared the same ISSR fingerprinting pattern not allowing to genetically distinguish anyone of them, while morphologically, they are notably different for the peculiar traits of the fruit and/or leaves. 'Crispifolia' was closely related to the previous six cultivars, suggesting a common origin of the group. Authors notified that this very low or absent genetic variability could be explained on the basis that these particular characteristics depend from mutations that do not varythe DNA length between the simple sequence repeats.

High level of genetic similarity in sour orange also reported by other researchers. Barrett and Rhodes suggested variations in orange, lemon, grapefruit, sour orange and lime based on one ancestor tree. Torres at al. (1978) found any differences in 15 sour oranges accordingto their leaf isozyme data. Luro et al. (2000) found no polymorphism among 10 accession of sour orange with SSR markers. Recently, Uzun (2009) also found very narrow genetic diversity in sour oranges. Besides sour oranges had hybrid origin such as 'Australian' or 'Smooth Seville' were distinguished from other accessions.
