DIVERSITY OF PHYSICAL AND COOKING QUALITY CHARACTERS OF SELECTED TRADITIONAL AND IMPROVED RICE VARIETIES IN SRI LANKA

Several physical and cooking quality parameters of fifteen traditional and nine hybrid rice varieties were analyzed. All the tested parameters showed significance variations (p<0.05) irrespective of being traditional or hybrid varieties. The highest equivalent diameter was observed in Devaraddiri while lowest was recorded for Sudurusamba . The mean kernel volume values ranged from 4.08±0.01-20.54±0.13 mm 3 . The highest kernel length of 6.82 mm was observed in At 305 and the lowest value of 3.6 mm was observed in Suwandel . The highest surface area was reported in Devarradiri (5.54±1.07 mm 2 ) while the lowest value (2.35±0.27 mm 2 ) was reported in At 306. Bw 272-6B , Bg 358 , Bg 360 , Suwandel and Sudurusamba recorded high surface area to volume ratios than others varieties. Bulk density varied from 833.9±1.37 kg/m 3 to 754.2±0.918 kg/m 3 . Bold shape grains of Bg 352 , Bg 358 , Suwandel and Bw 267-3 had the higher bulk density than slender grains of Dikwee , Iginimitiya , Bg 94- 1, At 306 and At 405 . Porosity was lower in bold shape grains than in slender grains. Minimum and highest cooking time of 15.00±0.01 min and 31.00±0.24 min were recorded for At 405 and Herathbanda respectively. Slender rice varieties of Dikwee , Iginimitiya, Bg 94-1 , At 306 and At 405, showed a lower solid loss than other grains due to their comparatively low surface area.


Introduction
Rice (Oryza sativa L.) is the staple food crop in world which accounts for 21%, 14% and 21% of the global energy, protein and fat supply, respectively (Kennedy and Burlingame, 2003). It is widely cultivated throughout the world and has become the second most important cereal in the world after wheat in terms of cultivation (Jones, 1995). Rice occupied 989,950 ha of the total cultivated area in Sri Lanka, of which 305,314 ha were cultivated in Yala season and 684,636 ha in Maha season (Abesekara et al., 2008). About 0.88 million farmer families are engaged in paddy cultivation and in they produced 3.9 million MT (Yala season; 1,128,984 MT and Maha season; 2,716,961 MT) of paddy annually which fulfills the total domestic requirement (Hafeel et al., 2011). The per capita consumption of rice fluctuates around 114 kg per year depending on the price of rice, bread and wheat flour . At present cultivation of traditional rice varieties has become more popular due to consumer preference due to their high nutrient value and other functional properties of the rice grain. More than 400 different traditional ice varieties are growing in the country. These varieties are highly diversified among each other with respect to unique nutrient, cooking and eating characteristics (Abesekara et al., 2008).
Rice grain quality is mainly defined by their physical, milling, cooking and eating qualities as well as nutritional quality. Physical and engineering properties of rice grain are important in handling, planting, harvesting, threshing, cleaning, sorting and drying (Irtawange, 2000) and also storage and grain machineries. Principal axial dimensions of rough rice are useful in selecting sieve separators and in calculating power consumption during the rice milling process. Furthermore physical properties of rice grain can also be used to calculate the surface area and volume of kernels which are important for modeling of drying, aeration, heating and cooling of paddy.
Among many quality parameters of rice, cooking and eating qualities are the most important. Consumer preference mainly depends on its cooking and processing quality, which can be measured in terms of water uptake ratio, grain elongation during cooking, solid loss during cooking and cooking time (Bhattacharya and Snowbhagya, 1971). During cooking, water uptake and swelling of the rice granules occur due to water diffusive process (Oko et al., 2012). When water is at sufficiently high temperature (~95 o C), the starch undergoes a gelatinization process (Juliano, 1971). The gelatinization temperature, gel consistency and amylose content directly related to cooking and eating qualities (Little et al., 1958), which are mainly governed by the genotype as well as the environment.
The environmental conditions that affect the cooking and eating qualities of rice changes in the day temperature, day length and genotypeenvironment interaction (Jianrong et al., 2005).
Only few studies have been carried out on the physical and cooking quality characteristics of traditional rice varieties in Sri Lanka (Abesekara et al., 2008;Wickramasinghe and Noda, 2008). Nevertheless, local germplasm has not yet been fully evaluated for their physical, engineering and cooking qualities.
Differences in grain properties can result in a significant variation in the processing characteristics of the grain. Proposed study was carried out to evaluate the diversity of the physical and cooking qualities of rice grain of selected traditional and elite varieties important information for future rice breeding programs and designing of handling, processing, and packaging machinery for rice production.

Rice grains
Fifteen traditional and nine hybrid rice varieties (Table 1) were grown at the Rice Research and Development Institute, Bathalagoda in Yala season 2013/2014 under similar management practices. Seeds were collected at the harvest maturity and shade dried up to 12-14% (dry basis) of moisture content. Seed samples of 150 g were dehulled (Satake THU 358, Japan) and polished to a level of 8±2% of bran removal (Satake THO 50, Japan) from the total weight.

Physical properties of rice
Length (L), thickness (T) and width (W) of the milled rice kernels were measured using a micrometer. The measurements were taken from randomly selected 10 gains from each variety in triplicate. Size and shape was classified according to the method described by Cruz and Kush (2000). The equivalent diameter (Dp) considering a prolate spheroid shape for a rough rice grain was calculated using equation 1 (Mohsenin, 1986). Grain volume (V) and surface area (S) of brown rice were calculated using equations 2 and 3 (Varnamkhasti et al., 2008). (2) (3) Sphericity (Ø) which is the ratio of surface area of a sphere having the same volume as grain to the surface area was determined using in equation 4 (Mohsenin, 1986).
In order to determine kernel weights 100 grains sample selected randomly from each cultivar and the samples were weighed separately (Vidal et al., 2007). The average 100 kernel weight was obtained from five replicates. Sample of weighed rice grains were filled into a graduated measuring cylinder and the volume occupied were recorded to determine the bulk density (g/ml). True grain density was also measured using kerosene displaced method (Singh et al., 2005). Porosity (%) was calculated using the equation 5 (Mir et al., 2013).

Cooking properties of rice
Cooking properties of the rice samples were determined using standard procedures (Batcher et al., 1956;Singh et al., 2003 and. About of 2 g of rice sample was cooked in 20 ml of distilled water using a boiling water bath. The minimum time required to cook was estimated by removing few cooked kernels at regular intervals. Few kernels were pressed between two glass slides until no white core was noticed. The time taken to disappear the white core was considered as the minimum time of cooking. The gruel that obtained during the cooking was transferred into a 50 ml beaker and was evaporated in an oven at 110 °C. The solid obtained after drying were weighed and the percentage of the gruel solid loss was calculated. The rest of the cooked samples was weighed (±0.1 mg) and the water uptake ratio which was the ratio for final cooked weight to uncooked rice weight was calculated (Singh et al., 2003 and. Length of 10 randomly selected cooked rice kernels were measured and was divided by the length of uncooked rice to determine the grain elongation percentage.

Data analysis
Data obtained were subjected to one-way analysis of variance (ANOVA) followed by Duncan's New Multiple Range Test (DNMRT) to determine the statistical differences among varieties at a significance level of p≤0.05. Statistical analysis was done using SAS software package version 6.1 (SAS, 1990).

Physical properties of rice
Grain physical dimensions among the selected rice varieties (Table 1) varied with respect to the grain size and shape classes. The highest kernel length of 6.82 mm was observed in At 305 and the lowest length of 3.6 mm was observed in Suwandel. Kernel length of many varieties varied from 5 mm to 6 mm while kernel width varied from 2.00 to 2.50 mm. Based on the L of kernels, all selected traditional rice varieties were fallen either into short (≤5.50 mm) or medium (5.51-6.60 mm) grain size classes (Dela Cruz and Khush, 2000). But, there was no any medium size (L=5.51-6.6) rice variety was found in the tested hybrid rice varieties. Varieties At 306, At 405, Bg 94-1, Dikwee and Inginimitiya showed a L:W ratio >3 indicating that these grains are slender in grain shape. According to the grain size class, Dikwee and Inginimitiya varieties could be categorized into "medium" grain size where At 306, At 405 and Bg 94-1 were categorized as "long" grain size class. Therefore long-slender grains of At 306, At 405 and Bg 94-1 can be identified as basmati type rice. Out of 24 rice varieties, there were 5 bold, 5 slender and 14 medium grain shapes were observed. Most of the tested traditional rice varieties belong to medium grain shape category except Suwandel, Dikwee and Inginimitiya (Table 1). Significant differences (p<0.05) were observed in the equivalent diameter among the tested rice varieties ( Table 2). The highest equivalent diameter of 3.36±0.17 mm was observed in Devaraddiri while lowest value of 1.98±0.01 mm was in Sudurusamba.  (Table 2). Except Sudurusamba other others varieties were categorized as "bold" grain shape (Table  1). The sphericity values of the tested rice varieties were in the range of 33.7-64%. Significant differences (p<0.05) were also observed among the sphericity values of tested rice varieties (Table 2). In contrast to bulk density, higher porosity values were observed for slender rice varieties while lower values were observed for bold shape varieties.

Figure 2.
Variation of solid loss in cooking water with L:W ratio among selected rice varieties in Sri Lanka. Vertical lines indicate the mean ±SD.
Water uptake ratio also showed significance differences (p<0.05) among the selected rice varieties (Table 3). Lowest water uptake ratio of 2.13±0.01 was reported for Dular while highest water uptake of 3.40±0.17 was reported for Bg 352. The values for grain elongation during cooking ranged from 14.96±0.22 mm to 64.09±2.42 mm recorded for Kahatawee and Bg 352 respectively.

Physical properties of rice
Physical appearance of the rice kernels varies according to its size and shape which is highly important for rice marketability. It is important to classify the grains based on their shape to determine the extent of off-size in market grades (Varnamkhasti et al., 2008) which is also important for milling. The size and shape of rice grain is determined by L, W and L:W ratio. Long, medium and short grain size classes depend on the kernel L of 6.61-7.50 mm, 5.51-6.60 mm and ≤5.50 mm, respectively (Cruz and Khush, 2000). Nevertheless, slender, medium and bold grain shape classes are subjected to the L:W ratios of >3, 2.1-3 and ≤2 respectively. Grain physical dimensions results of this study are in compatible with Samal et al. (2014) who reported that grain length and width of the traditional aromatic rice varieties is in the range of 4.90-12.41 mm and 1.80-3.50 mm respectively. Sphericity values of traditional and hybrid rice reported in this study was within the range of 33-64%. These values are in accordance with the previous data within the range of 32%-100% reported by Mohsenin (1986). The value reported for the bulk density in the present study was similar to the findings of Bhattacharya et al. (1972).
Bulk density was independent from the rice length, width and weight but inversely related to L:W ratio or with respect to the grain shape. Round grains showed the higher bulk density and slender grains showed the lowest, as described by Correa et al. (2007). The knowledge of bulk density is useful to design grain storage silos and hoppers for grain handling during processing (Nalladulai et al., 2002).

Cooking properties of rice
Minimum cooking time values reported in this study was within the previously reported rang of 13-25 min (Oko et al., 2012). The variation in the cooking time among different varieties could be related to the gelatinization temperature of rice. It has been found that gelatinization temperature is directly proportional to the cooking time of rice (Bhattacharya and Snowbhagya, 1971). However, At 405, At 306, Bg 352, Mawee, and Pokkali that fall into high gelatinized temperature class (Abesekara et al., 2008;Rebeira et al., 2014) (Abesekara et al., 2008).
Solids loss of rice during cooking is considered as a parameter of cooking quality attribute (Julliano, 1971). Low solid loss contents in the gruel can be attributed to the slender rice varieties having comparatively smaller surface area.
These observations were comparable to those reported by Hirannaiah et al. (2001). Rice varieties with higher amylose content are prone to leaching out solids into the cooking water as starch granules during cooking (Juliano, 1971). According to Rebeira et al. (2014), Herathbanda, Wannidahanala, Suduheeneti, Pokkali, Sulai and Bw 267-3 are categorized as high amylose rice and are shown to have higher solid loss into the cooking water. Frei and Becker (2003) reported that rice with high amylose content tends to absorb more water during cooking. According to the results high amylose rice varieties such as Bg 352, Bg 358, Herathbanda, Pokkali, Dikwee, and Kaluheeneti showed higher water uptake while At 405, At 306, sudurusamba and Bg 360 which was categorized as low amylose group (Abesekara et al., 2008) showed low water up take ratio. The extent of water absorbed by rice during cooking is considered as an economic quality as it gives some estimate of volume increase during cooking.
Elongation of rice grains during cooking is dependent basically on rice variety and degree of milling (Mohapatra and Bal, 2006). Expansion of size in length-wise without increase in width during cooking is considered as a highly desirable quality characteristic rice grain. Water uptake is directly related to taste and consistency of rice and thus high water uptake indicates good quality (Shittu et al., 2012, Verma et al., 2013. The rice varieties which absorbed high quantity of water during cooking are soft varieties and these varieties which absorb low amount of water during cooking considered as hard rice varieties.

Conclusion
This investigation has showed that the physical properties and cooking related characteristics of the different varieties of Sri Lankan rice vary significantly according to the size classes. The wide variation of bulk density, porosity, water uptake ratio and grain elongation were recorded for rice varying from short to long varieties. Regardless of traditional or hybrid improved slender grain rice varieties showed desirable cooking characteristics and better physical properties for milling.