**2. Germplasm and breeding research of loquat**

Loquat is mainly self-compatible, self-incompatibility has only been found in a few varieties (Chen & Chu, 2008). The traditional propagation from seed has provided a range of varieties adaptable to different environments and planting regions. *Ex situ* germplasm collections have been established in China, Japan and Spain (Zheng, 2007; Badenes et al., 2009). Among these collections, the Chinese collections possess highest diversity. There are more than 1000 accessions described in the various Chinese germplasm collections (Zheng, 2007). The largest Chinese collection is located in Fuzhou, with a total of more than 250 accessions. The major European loquat germplasm collection is held at Instituto Valenciano de Investigaciones Agrarias (IVIA) of Spain. The IVIA collection includes more than 74 accessions, of which 49 accessions are Spanish varieties. In addition, Italy and Greece also have small collections of mainly local germplasm (Lin, 2003).

Understanding of genetic diversity in loquat germplasm has been greatly improved through the application of molecular marker technology (He et al., 2011). Using internal transcribed spacer (ITS) region of the 18S–5.8S–26S nuclear ribosomal cistron, Li et al. (2009) analyzed 15 *Eriobotrya* accessions and six close related species. Result of cluster analysis suggested that *E. malipoensis* had closer relationship with *E. japonica* Lindl. This result was further supported by Yang et al. (2009) who analyzed *Eriobotrya* taxa using AFLP markers. However, Zhao et al. (2011) also used ITS region of the 18S–5.8S–26S nuclear ribosomal cistron and assessed the phynogenetic relationship among the *Eriobotrya* species represented by 25 loquat cultivars and seven wild taxa. Their result suggested that the evolution order of the studied taxa was *E. bengalensis* f*. angustifolia*, *E. prinoides var. dadunensis*, *E. prinoides* (*E. bengalensis*), *E. dayaoshanensis* and *E. japonica* Lindl. And *E. bengalensis* f*. angustifolia* was found the likely ancestral taxa of *E. japonica* Lindl.

The intraspecific genetic diversity in loquat has been analyzed by various molecular markers as well. Soriano et al. (2005) assessed genetic relationships among 40 loquat accessions that originated from different countries, including accessions from the European loquat germplasm collection maintained at IVIA in Valencia, Spain. Thirty pairs of microsatellites previously identified in *Malus* × *domestica* (Borkh.) were used. The expected and observed heterozygosities were 0.46 and 0.51 on average, respectively, showing a high level of out crossing behavior in loquat germplasm. However their result also indicates a low degree of genetic diversity in the set of loquat accessions analyzed, in comparison to other members of the Rosaceae family.

He et al. (2011) used apple SSRs that are transferable to loquat and assessed the level of genetic diversity within loquat. They used 39 SSRs transferable to loquat and genotyped 54 loquat accessions from Japan, Spain, and four Chinese provinces, as well as two wild species (*E. prinoides* var. *Daduheensis* and *E. prinoides*). In total they identified 155 different alleles with a mean value of 3.38 per locus, and the mean observed heterozygosity was 0.47. These values indicate a high degree of genetic diversity in the set of Chinese loquat accessions analyzed. Cluster analysis grouped the accessions into cultivated and wild loquats. The

2000; Cai 2000; Vilanova et al. 2001; Soriano et al. 2005; Dong 2008; Qiao 2008; Watanabe et al. 2008; Gisbert et al. 2009; Yang 2009). These studies significantly enhanced our understanding about the distribution and structure of genetic diversity in loquat germplasm

Loquat is mainly self-compatible, self-incompatibility has only been found in a few varieties (Chen & Chu, 2008). The traditional propagation from seed has provided a range of varieties adaptable to different environments and planting regions. *Ex situ* germplasm collections have been established in China, Japan and Spain (Zheng, 2007; Badenes et al., 2009). Among these collections, the Chinese collections possess highest diversity. There are more than 1000 accessions described in the various Chinese germplasm collections (Zheng, 2007). The largest Chinese collection is located in Fuzhou, with a total of more than 250 accessions. The major European loquat germplasm collection is held at Instituto Valenciano de Investigaciones Agrarias (IVIA) of Spain. The IVIA collection includes more than 74 accessions, of which 49 accessions are Spanish varieties. In addition, Italy and Greece also

Understanding of genetic diversity in loquat germplasm has been greatly improved through the application of molecular marker technology (He et al., 2011). Using internal transcribed spacer (ITS) region of the 18S–5.8S–26S nuclear ribosomal cistron, Li et al. (2009) analyzed 15 *Eriobotrya* accessions and six close related species. Result of cluster analysis suggested that *E. malipoensis* had closer relationship with *E. japonica* Lindl. This result was further supported by Yang et al. (2009) who analyzed *Eriobotrya* taxa using AFLP markers. However, Zhao et al. (2011) also used ITS region of the 18S–5.8S–26S nuclear ribosomal cistron and assessed the phynogenetic relationship among the *Eriobotrya* species represented by 25 loquat cultivars and seven wild taxa. Their result suggested that the evolution order of the studied taxa was *E. bengalensis* f*. angustifolia*, *E. prinoides var. dadunensis*, *E. prinoides* (*E. bengalensis*), *E. dayaoshanensis* and *E. japonica* Lindl. And *E. bengalensis* f*. angustifolia* was found the likely

The intraspecific genetic diversity in loquat has been analyzed by various molecular markers as well. Soriano et al. (2005) assessed genetic relationships among 40 loquat accessions that originated from different countries, including accessions from the European loquat germplasm collection maintained at IVIA in Valencia, Spain. Thirty pairs of microsatellites previously identified in *Malus* × *domestica* (Borkh.) were used. The expected and observed heterozygosities were 0.46 and 0.51 on average, respectively, showing a high level of out crossing behavior in loquat germplasm. However their result also indicates a low degree of genetic diversity in the set of loquat accessions analyzed, in comparison to

He et al. (2011) used apple SSRs that are transferable to loquat and assessed the level of genetic diversity within loquat. They used 39 SSRs transferable to loquat and genotyped 54 loquat accessions from Japan, Spain, and four Chinese provinces, as well as two wild species (*E. prinoides* var. *Daduheensis* and *E. prinoides*). In total they identified 155 different alleles with a mean value of 3.38 per locus, and the mean observed heterozygosity was 0.47. These values indicate a high degree of genetic diversity in the set of Chinese loquat accessions analyzed. Cluster analysis grouped the accessions into cultivated and wild loquats. The

around the world.

**2. Germplasm and breeding research of loquat** 

have small collections of mainly local germplasm (Lin, 2003).

ancestral taxa of *E. japonica* Lindl.

other members of the Rosaceae family.

cultivated loquat can be subdivided into three subgroups which generally reflect their geographic origin in China.

The Chinese loquat germplasm is traditionally classified according to the flesh color. Varieties can be classified as either "white flesh" or "orange flesh". The former accounts for about 30 percent of the total Chinese varieties. The texture of the white-fleshed loquat is generally more delicate and tender. Varieties in this group include 'Ruantiaobaisha' and 'Baiyu', both are the leading varieties of Zhejiang and Jiangsu province in China. Several Japanese varieties, such as 'ShiroMogi', canbe placed in the white flesh group. Orangefleshed varieties, such as 'Mogi', 'Tanaka', and 'Nakasakiwase' account for 95% of the total crop area of Japan. Spanish commercial production depends on four orange flesh varieties, including 'Algerie', 'Magdal', 'Golden Nugget', and 'Tanaka' with 'Algerie' varieties (Badenes et al., 2009).

Crop improvement, both through breeding for new varieties and selection of accessions from existing germplasm, have been carried out in China, Japan and Spain (Zheng, 2007; Shih, 2007; Badenes et al., 2009). However, so far the varieties used in production were mostly farmer selections made by growers in the local areas. Surveys and selection of germplasm accessions have been carried out in China (Zheng, 2007), Mediterranean countries, Turkey and Pakistan (Badenes et al., 2009). Seedless or fewer seeds is one of important objectives in today's fruit breeding programs. Loquat fruit has a relatively smaller fruit than other fruit tree species. The average weight of a loquat fruit is about 30 to 40 grams. Some large-fruited varieties can have a fruit size of 70 grams, and 34 usually with 3 to 4 large seeds. The low edible rate ( less than 70%, is an important quality constraint for fresh consumption of loquat (Lin et al, 2007b). Therefore, seedless loquat fruit is highly desired by consumers.

The classical approach to eliminate seeds in many crops is to produce triploids through ploidy manipulation (Ollitrault et al., 2008). And there are three routes to obtain triploidy genotypes, including use of nonreduced megaspore or microspore, crossing induced tetraploids with diploids and in vitro culture of nuclear tissue (Janick, 2011). Tetraploid varities can be spontaneously formed or induced by colchicines (Yahata et al., 2004). In 1978, Chinese researchers of Fruit Research Institute of Fujian Academy of Agricultural Science developed a tetraploid varities 'Min No.3' using colchicine-induced polyploidization. However the tetraploid variety was not well adopted due to the performance in other agronomic traits and quality. Between 1983 and 1985, researchers in the same institute obtained triploid loquat plants through endosperm culture. In early 1990s, the Southern Prefectural Horticulture Institure, Chiba Prefectural Agriculture and Forestry Research Center of Japan hybridized diploid variety "Nagasakiwase" 6 with tetraploidy variety "Tanaka", and developed the first triploid loquat variety "Kibou", as well as a series of triploidy seedlings. From 1997 to present, Liang and his team in the Southwest University of China have made significant progress in identification of natural triploid loquats (Guo et al., 2007).
