**1. Introduction**

Rice serves as the main staple food crop of nearly half of the world's population and it is obvious that genetic improvement of rice cultivars play an important role in the rice production for fulfilling ever increasing food demand. Rice is the staple

food which occupies 34% (0.77/million ha) of the total cultivated area in Sri Lanka and currently produces 2.7 million tonnes of rough rice annually and satisfies around 95% of the domestic requirement [1].

The rice genus *Oryza* L. consists of *ca.* 21 wild and two cultivated species distributed in Asia, Africa, Australia, and the America [2, 3] and these species have been categorized into ten different genome types, such as six diploids (AA, BB, CC, EE, FF, and GG) and four allotetraploid species (BBCC, CCDD, HHJJ, and HHKK) [4, 5]. Wild rice spices are important in rice breeding programs because, these species comprise traits of agronomic interest, for example, the resistance and tolerance to biotic and abiotic stresses [2, 6–8]. However, due to the sterility barriers, most of the *Oryza* germplasm is of limited use in rice breeding programs [8, 9]. Genetic resources of the AA- genome group also referred to as the *Oryza* complex, have long been a focal point of the rice breeders.

The *Oryza sativa* complex includes eight diploid species [2] and the Asian cultivated rice consists of main subspecies, *O. sativa* ssp. *Indica* and *O. sativa* ssp. *Japonica* [10–12] are of Asian origin and globally cultivated today. The two presumed wild progenitors; the perennial *O. rufipogon* (**Figure 1**) is distributed throughout tropical Asia and Oceania, whereas the annual *O. nivara* is distributed in tropical continental Asia (**Figure 2**). Another cultivated species in the genus, *O. glaberrima*, was parallelly domesticated in West Africa where it is endemic [2]. There are two additional wild species also endemic to Africa, *O. barthii* and *O. longistaminata*. The former is the annual wild progenitor of *O. glaberrima*, while the latter is a perennial, rhizomatous and partially self-incompatible grass species [13].

In Sri Lanka, the genus *Oryza* consists of two species complexes, the *O. sativa* complex that includes the AA genome species, the *O. officinalis* complex which includes the CC genome species [3, 14] and a single species *O. granulate* (GG) [15, 16]. The two complexes, *O. sativa* complex and *O. officinalis* complex are both pan tropical and have very similar overall distribution. However, only AA genome species have been cultivated and domesticated. It appears that *O. officinalis* complex species do not have the attributes that make them attractive or likely to cultivate. Of the five wild rice species reported in Sri Lanka, (*O. nivara*, *O. rufipogan* (AA); *O. eichengeri*, *O. rhizomatis* (CC); and *O. granulate* (GG)), *O. rhizomatis* grows in partially shaded areas/grass lands and has been reported only in Sri Lanka and hence considered endemic to Sri Lanka [17, 18].

**Figure 1.** O. rufipogon *(a) panicle (b) growing in a periodically drying temporary ponds.*

**Figure 2.**

O. nivara *(a) panicle (b) growing along the border of a canal in Sri Lanka.*

*O. rhizomatis* is one of the species of the *O. officinalis* complex (**Figure 3**). The taxonomy of *O. officinalis* complex in Sri Lankan has been puzzling due to insufficiency of satisfactory herbarium specimens and the living plant materials. As an attempt of resolving the problem of the morphological variation in the complex, Biswal and Sharma [19] retracted the name *O. collina* and considered this taxon to be synonymous with *O. eichingeri*. Thus, Biswal and Sharma [19] agreed with both Bor [20] and Tateoka [14] that *O. eichingeri* is the sole representative of *O. officinalis* complex in Sri Lanka (**Figure 4**). *O. offocinalis* in Sri Lanka grows in both shaded and open habitats, whereas *O. eichingeri* grows in the shade of forests in Uganda [21]. However, taxonomists were not able to give much weight to the habitat of this taxon since field notes are generally infrequent.

The new collections make known clear morphological and habitat differences in *O. eichingeri* and it is a larger taxon which occurs in the drier habitats in Sri Lanka [2]. This larger rhizomatous taxon has previously been called *O. latifolia* and *O. officinalis*. *O. latifolia* is a large non-rhizomatous tetraploid from South and Central America with broader leaves and whorled panicle branches. *O*. *officinalis* which usually has rhizomes, has smaller spikelets, shorter palea tip, more branches of approximately equal length from the lowest panicle node, and spikelets inserted away from the base of primary branches. *O. officinalis* is also genetically different from this Sri Lankan taxon with which it can form sterile hybrids. However, Sri Lankan taxon belongs to the same genome group as both *O. officinalis* and *O. eichingeri*, which is CC [22].

There are two diploid CC genome species in Sri Lanka, *O. eichingeri* and *O. rhizomatis* [17, 19]. Previously *O. collina* was the name used for Sri Lankan germplasm of the *O. officinalis* complex [23]. *O. collina* has been used for both *O. eichingeri* and *O. rhizomatis*. However, *O. rhizomatis* is readily distinguished from *O. eichingeri* by its larger plant stature and rhizome formation. *O. rhizomatis* appears to be intermediate between *O. officinalis* and *O. eichingeri*. Analysis of the nuclear and chloroplast genome of *O. rhizomatis* by RFLP and SSR reveals that *O. rhizomatis* differed from *O. eichingeri* and *officinalis* [24–26].

The nomenclature and the taxonomy of the elements of these complexes have been studied and nomenclatural changes have been suggested and certain de novo species was described to disentangle the problem within the complexes. Due to this reason, the exact number of wild rice species in Sri Lanka becomes uncertain and

*Cereal Grains - Volume 2*

**Figure 3.**

*(a) Panicles (b) Rhizomes (c) well-spread rhizome submerged in water of* O. rhizomatis*. (d)* O. rhizomatis *in open spaces in the dry zone (Anuradhapura District), Sri Lanka.*

detailed studies specially, on morphological, anatomical, and molecular aspect of the Sri Lankan wild rice are needed for the delimitation of *Oryza* complexes in Sri Lanka.

Several recent studies demonstrated that the reliance on single source of information possibly misleading the results in the phylogenetic inferences due to analytical inconsistency and biological processes [27, 28]. The inconsistencies among the

*Characterization of Wild Rice -* Oryza *Species Complexes in Sri Lanka DOI: http://dx.doi.org/10.5772/intechopen.97244*

**Figure 4.** *Panicles of* O. eichingeri *in open spaces in the forest in the dry zone, Sri Lanka.*

phylogenies have become one of the most common problems during the reconstructing molecular phylogenetics using different datasets, such as individual genes. Studies carried on the genome-wide markers have witnessed new phylogenetic reconstructions that use large quantities of genome-wide markers to illustrate former controversies on evolutionary relationships at all taxonomic levels [27–31]. In general, a gene tree does not necessarily reflect a species tree, even if the orthology of marker genes are clearly identified and employed. Therefore, many genetic markers, including unlinked loci with extensive functional representation as well as intergenic genomic regions, are needed to comprehensively track organismal history. Such a robust phylogeny will build a foundation for future insights into rice genome evolution.

Therefore, there is a need to delimit the *Oryza* species complexes in Sri Lanka using morphological, anatomical, and molecular information. The objectives of the present study are to enumerate the number of species within each *Oryza* complex (*O. sativa* complex and the *O. officinalis* complex) in Sri Lanka and characterization of species and species complexes with evidence generated from morphological, anatomical, and molecular studies.
