Previous PageTable Of ContentsNext Page




Plant architecture and growth characteristics of fine grain and aromatic rices and their relation with grain yield

R.K. Dutta, M.A. Baset Mia and Sakina Khanam

Crop Physiology Division, Bangladesh Institute of Nuclear Agriculture, PO Box 4, Mymensingh, Bangladesh


In Bangladesh, there are thousands of local landraces of rice (Kaul et al., 1983), many of which are either fine grain or aromatic types. The breeding of high-yielding varieties (HYVs) is an essential component of the strategy to overcome food deficiency. However, HYVs cover only one-third of the whole rice cropping area (BBS, 1998) and this coverage is not likely to improve in the near future because of the shortage of inputs such as water and fertilizers. Recently developed hybrid rice varieties may take the place of existing HYVs because hybrid rice has about a 30 percent yield advantage over conventional pure line varieties (Yuan, 1998). However, the area given to cultivation of local rice varieties is not likely to change for some time, thus it is vital to improve the varieties of local landraces, which cover two-thirds of the cropping area. The yield potential of local rices varies from 2.5 to 3.5 tonnes/ha during the aman season (BBS, 1997). It is well known that local rices are mostly non-responsive to nitrogen fertilizer and that yield reductions are caused by the lodging characteristics of some varieties. In such varieties, certain architectural defects cause significant problems with assimilate translocation. Although the nutritional quality, palatability, taste, cooking quality and consumer demand are higher for local rices than for HYV rices, the determining factors for production are yield per unit area and farmers' profit. Nevertheless, the higher prices and export quality of local fine and aromatic rices warrant their higher production - hence the need to increase the yield of local varieties. It will be necessary to improve the nitrogen assimilation, photosynthesis and assimilate partitioning of local varieties, and employ modern breeding techniques to improve their architecture. The first step is to identify the existing architectural and physiological defects. The two experiments described in this article were conducted in order to discover the architectural and physiological defects of local rices under Bangladeshi spatial and temporal conditions.


Experiment 1

Two experiments were conducted for the study. The first was conducted from June to December 1994 at the Bangladesh Institute of Nuclear Agriculture (BINA) Farm at Mymensingh. Twenty-seven local cultivars, two exotic cultivars (Basmati and KL-5) and two varieties (BR-5 and BINASAIL) released by the Bangladesh Rice Research Institute, Joydebpur, Gazipur, and BINA, Mymensingh, were used for the study. The local varieties were Nizersail, Chinigura, Masuri, Basmati, Kaskhani, Balagura, Mugy, Chiniatab, Bhogbalam, Kalajira, KL-5, BR5, Madhumala, Rhadhunipagal, Chitaragusail, Binnafool, Tilkafuly, Rasulbhog, Madhumadab, Kataribogh, Badshabhog, Zabsiri, Ragusail, Binni, Biroi, Ukunimadhu, Zinghasail and Balam. The design of the experiment was randomized complete block with three replications. The plot size was 2 m x 2 m and the plants were spaced at 20 cm x 15 cm. Fertilizers were applied at 70-30-20 kg/ha as N-P2O5-K2O in the form of urea, triple superphosphate and muriate of potash, respectively. Thirty-day-old seedlings were transplanted at one seedling per hill on 1 August 1994. Four hills from each plot were harvested at flowering stage to measure the leaf area index (LAI) and leaf area ratio (LAR), and at maturity stage to collect data on number of tillers, flag leaf angle and area and total dry matter (TDM). Physiological characteristics such as LAR and LAI were calculated from the mean data of TDM and leaf area. Five promising varieties of different morphology (Basmati, Badshabhog, Kalazira, Nizersail and Binasail) were studied for architectural parameters such as internode length, leaf sheath length, flag leaf area and angle. At maturity stage, ten hills were harvested for data on yield and yield components. The data were analysed statistically and the mean values were evaluated by standard deviation.

Experiment II

This experiment included some coarse (BR3, BR11 and BR22) and some fine (Pajam, Nizersail and Badshabhog) rice varieties with the objective of undertaking comparative grain development studies. The experiment followed the same method and cultural practices as in Experiment I. Twenty seeds were collected every three days starting from the day of anthesis and their weight was measured using a microbalance. The data gained on individual seed weights were plotted against time for seed development.

Results and discussion

The results presented in Table 1 showed that the varieties have been grouped as Group I, Group II and Group III depending on the grain yield. Group I (Binasail, Chinigura, Masuri, Basmati, Kaskhani, Balagura, Mugy, Chiniatab, Bhogbalam, Kalajira, KL-5), representing comparatively higher yields of 2.97-3.96 tonnes/ha, was associated with a lower number of tillers, lower percentage of sterile tillers, higher number of grains per panicle, lower percentage of sterile grains, lower LAR, lower flag leaf angle, higher flag leaf area and higher harvest index. Group III (Kataribogh, Badshabhog, Zabsiri, Ragusail, Binni, Birai, Ukunimadhu, Zingasail, Balam), representing lower yields ranging from 2.10 to 2.51 tonnes/ha, was associated with a higher number of tillers, higher percentage of sterile tillers, lower number of grains per panicle, higher percentage of grain sterility, higher LAR, higher LAI, higher flag leaf angle, lower flag leaf area and lower harvest index. Group II (Nizersail, BR5, Madhumala, Radhunipagal, Chitaragusail, Binnafool, Bashfool, Tilkafuly, Rasulbogh, KL-5, Madhumadab), representing medium yields ranging from 2.54 to 2.93 tonnes/ha, showed intermediate values for the above-mentioned characteristics. However, plant height, TDM, 1 000 grain weight and crop duration were similar among the groups.

Grain yield, total dry matter and yield components of aromatic fine rice varieties and correlation with seed yield


Group I

Group II

Group III


Plant height (cm)

145.90 ± 5.97

142.7 ± 14.72

142.70 ± 16.25


No. of tillers/hill

8.63 ± 0.84

9.28 ± 1.00

10.14 ± 0.90


Sterility of tillers (%)

19.90 ± 5.97

28.30 ± 9.36

30.30 ± 11.66


No. of panicles/hill

6.78 ± 0.92

6.47 ± 1.07

7.32 ± 0.70


No. of grains/panicle

114.60 ± 39.20

97.77 ± 22.08

84.97 ± 29.51


Grain sterility (%)

19.07 ± 5.28

28.30 ± 9.36

29.90 ± 11.68


TDM (g/hill)

26.82 ± 3.59

24.35 ± 1.73

25.64 ± 3.12


LAR (cm2/g)

25.40 ± 5.79

26.33 ± 10.13

34.30 ± 9.37



2.15 ± 0.77

1.98 ± 0.55

2.41 ± 0.42


Flag leaf area (cm2/hill)

28.60 ± 3.97

25.74 ± 3.91

25.71 ± 3.72


Flag leaf angle (degree)

31.81 ± 8.88

41.36 ± 13.41



Harvest index (%)

39.10 ± 6.02

35.20 ± 5.10



Crop duration1 (days)

145.60 ± 5.66

144.38 ± 5.26



1 000 grain weight (g)

15.71 ± 4.46

15.59 ± 2.63

15.78 ± 4.08


Protein (%)

9.26 ± 0.56

9.28 ± 0.00

9.25 ± 0.35


Note: Group I represents 2.97 - 3.96 tonnes/ha; Group II represents 2.54 - 2.93 tonnes/ha;
Group III represents 2.10 - 2.51 tonnes/ha.
1 Seed to seed.

Although the photosynthetic potentials of the three groups were similar, as reflected by their TDM production, grain yield was higher in Group I because of the harvest index. In this group, yield was associated with lower LAR, higher flag leaf area, lower flag leaf angle, lower percentage of sterile tillers and grains.

Data presented in Table 2 show that the first, second and third internodes of the rice plant may not determine its height. The lengths of the third internode of different varieties are not widely different. The height of the plant was found to be dependent on the number of internodes and their individual lengths (Guevarra and Chang, 1965). Leaf sheath length was always higher than the corresponding internodes and similar among varieties. First, second and third leaf from basipetal succession among different varieties were similar and only higher in Badshabhog. It is evident that the correlation of seed yield with plant height is weak (r = 0.27). Thus, height may be reduced and stem thickness may be increased for non-lodging and better translocation of assimilates.

The data relating to grain growth presented in Figure 1 show that the rates of grain growth for all coarse grain varieties (BR3, BR11 and BR22) were higher than those for the fine grain varieties (Nizersail, Pajam and Badshabhog). The duration of grain filling was identical in all varieties. BR3, with the largest grain, had the highest rate of grain development and Badshabhog, with the smallest grain, had the lowest. All varieties showed a lag phase of six days and a linear phase of 12 days; grain maturation took another nine days.

Figure 1
Grain growth pattern of some rice cultivars

From the data presented in Tables 1 and 2 it is quite evident that plant architecture has an impact on yield achievement and that there is ample scope for modification. The length of the second and third leaf may be reduced to some extent to check drooping and mutual shading. Flag leaf area may be increased, which may provide more recent photosynthate for grain filling. Flag leaf angle may be reduced to make it more erect for reducing mutual shading and greater exposure to light through abaxial and adaxial surfaces. Stem thickness may be increased and plant height may be reduced for non-lodging and better translocation.

It is also evident that fertility of tillers and grains is the basis for higher yield. For local rice varieties, higher LAR and LAI lead to lower harvest index. Thus, LAR and LAI should be proportional to make it optimum. The panicle should be as large as possible.

The results of the experiments show that for local rice varieties, the photosynthetic surface as indicated by higher LAR and LAI is not a limiting factor. Rather, the partitioning of assimilate from stem and leaf to grains may be a problem. The photosynthetic potential may be manipulated by reducing the crop duration to ensure a pre-flowering boost of the crop. In this way assimilate partitioning might be improved. The experimental results showed weak correlation of crop duration with yield. The duration of grain filling in coarse and fine grain rices is identical; however, the rate of grain filling in different sized grains is different and higher in coarse grain varieties (Figure 1). This means that reducing the duration of the reproductive phase is not possible. However, reducing the duration of the vegetative phase up to panicle initiation is possible through the manipulation of plant architecture and growth potentials.

Scientists at the International Rice Research Institute (IRRI) have proposed a new model for indica-japonica-type rice improvement with an ambitious yield attainment of 11-15 tonnes/ha. The model emphasizes a minimum of three to four tillers per hill, a sturdy root system, a dark-green leaf colour and better vascular system (That and Tran, 1989). Dutta et al. (1997) have made suggestions to overcome the limitations of nutrients and photosynthesis for yield improvement in indica-japonica-type rice. They emphasized an efficient root system, erect and narrow top canopy leaves, proper distribution of phenomenal utility and high genetic potential for the synthesis of soluble protein. However, proposals emphasizing the architectural improvement of local indica type rices have been meagre. The literature generated so far, especially on the architectural model of indica-japonica-type varieties, may not hold true for indica-type rice. In our experimental results, LAI and LAR have not been positive characteristics for yield (i.e. structural carbohydrate in the leaf is not beneficial for grain filling in local rices). Rather, flag leaf area (r = 0.88) and angle may be more related (r = 0.90) to recent photsynthate synthesis and translocation for grain filling and higher grain fertility.

It is a logical conclusion that reducing the length of leaf of the first, second and third nodes may help reduce mutual shading and in turn enhance assimilate partitioning from those leaves. For indica-japonica types, Takeda, Oka and Agata (1983) observed that high-yielding rice varieties had higher LAI and greater LAR and consequently produced more dry matter. Huang and Miao (1982) observed that higher leaf area, longer panicles and more grains per panicle were associated with higher grain yield. Sun (1985) observed that differences in leaf types were reflected in photosynthesis and yield. He added that varieties with large, thick and erect leaves resulted in higher yield. Burondker et al. (1990) reported a significant positive correlation with pre- and post-flowering LAR, crop growth rate and relative growth rate with seed yield in rice. Tsai (1984) suggested that increased LAI either before or after heading could increase rice grain yield. However, Swain et al. (1986) observed a complicated phenomenon of dry matter production, photosynthetic rate and yield. They reported that flowering and LAI were associated negatively with photosynthetic rate but positively with dry matter content and yield. Photosynthetic rate at flowering was negatively associated with yield. They made suggestions for the best combination of photosynthetic rate and LAI. Yoshida (1972) came to the conclusion that in tropical rice the relationship of LAI and yield is not a direct one. Salam et al. (1990) reported the architectural cause of yield improvement in Binasail, a mutant. They noted a more erect and wider leaf and higher total leaf area and grain per panicle for Binasail. Mia et al. (1994) reported on the basis of studying some rice lines developed by crosses that yield was associated with higher pre-flowering CGR in short duration rice crops and with higher post-flowering CGR in long-duration rice crops and suggested that rice yield improvements may be possible by enhancing the photosynthetic rate before anthesis. From the experimental results and literature it may be suggested that for local indica rice, plant height and leaf area should be reduced and stem thickness should be increased to some extent to avoid lodging and mutual shading. The number of fertile grains should be increased; this may be done by enhancing the assimilate translocation to the grains. In that case, recent photosynthesis of the flag leaf may be helpful. The flag leaf should be broad and erect; lower leaves should be short, narrow and non-droopy. The percentage of fertile tillers should be as high as possible.

It may be concluded that there is scope for improving the indica-type fine grain and aromatic rice varieties by altering their agricultural make-up with the aid of modern breeding techniques.

Internode, leaf sheath and leaf length of different fine and aromatic rice varieties


Internode length

Leaf sheath length

Leaf length

















































































LSD 0.05














BBS (Bangladesh Bureau of Statistics). 1997. Statistical Yearbook of Bangladesh. Dhaka, Ministry of Planning, Government of the People's Republic of Bangladesh.

BBS (Bangladesh Bureau of Statistics). 1998. Statistical Yearbook of Bangladesh. Dhaka, Ministry of Planning, Government of the People's Republic of Bangladesh.

Burondkar, M.M., Chavan, S.A., Jadav, B.B. & Birari, S.P. 1990. Physiological difference in yield of early rice varieties J. Maharashtra Agric. Univ. 13(3): 343-344.

Dutta, R.K., Mia, M.A.B., Lahiri, B.P. & Salam, M.A. 1997. Assessment of grain yield and quality improvements in rice by modern breeding techniques in Bangladesh and projections of future rice improvements. IRC Newsletter 46: 63-70.

Guevarra, A.B. & Chang, T.T. 1965. Internode elongation in rice varieties of reduced rice stature. Philippine Agrist., 49(1): 23-24.

Huang, P.S. & Miao, B.S. 1982. A study of high yielding indica rice cultivar in North Jiangus. J. Nanjang Agric. Coll., 1: 12-21.

Kaul, A.K., Khan, M.R.I. & Munir, K.M. 1982. Rice quality: a survey of Bangladesh germplasm. Joydebpur, Dhaka, Bangladesh Rice Research Institute.

Mia, M.A.B., Salam, M.A., Chowdhury, S.I., Islam, A.F.M.S. & Dutta, R.K. 1994. Morpho-physiological studies in relation to yield potential in rice. Bangladesh J. Nuclear Agric., 10: 45-50.

Salam, M.A., Ali, K. & Chowdhury, S.I. 1990. Morpho-physiological aspects of Binasail, for its improved yield potential over the parent Nizersail. Bangladesh J. Nuclear Agric., 5&6: 15-21.

Sun, X.C. 1985. Studies on the classification of leaf type in rice and its relationship with photosynthesis. Scientia Agricultura Sinica, 4: 49-55.

Swain, P., Pattanaik, R.K., Nayak, S.K. & Murty, K.S. 1986. Relationship between photosynthetic rate, growth components and yield in elite rice varieties. J. Nucl. Agril. Biol., 15(1): 18-22.

Takeda, T., Oka, M. & Agata, W. 1983. Characteristics of dry matter and grains production of rice cultivars in warmer part of Japan. I. Comparison of dry matter production between old and new types of rice. Japanese J. Crop Sci., 52(3): 299-306.

That, T.T. & Tran, D.V. 1989. Prospect for a new generation of high yielding rice varieties. IRC Newsletter, 2: 15-19.

Tsai, J.C. 1984. Studies on the relationships between leaf area index and yield in rice plants. J. Agric. Res. China, 33(2): 109-116.

Yoshida, S. 1972. Physiological aspects of grain yield. Ann. Rev. Plant Physiol. 23: 437-468.

Yuan, L.P. 1998. Hybrid rice development and use: innovative approach and challenges. IRC Newsletter, 47: 7-15.

Structure des plantes et caractéristiques de la croissance des grains
fins et aromatiques par rapport aux rendements

Les rendements supérieurs des grains fins locaux et des riz aromatiques s'expliquent par la présence réduite de talles et de talles stériles, par des grains/panicules plus nombreux, par le faible pourcentage de grains stériles, à un angle différent de la feuille paniculaire et à un indice de récolte plus élevé. Pour les riz locaux, la structure des plantes peut être modifiée dans une certaine mesure par le raccourcissement de la longueur de la deuxième et de la troisième feuilles et de la hauteur pour éviter le port pleureur, l'ombrage mutuel et la verse. Les panicules devraient être aussi gros que possible.

Arquitectura de las plantas y características del crecimiento del arroz
de grano fino y aromático y su relación con los rendimientos

El aumento de los rendimientos de los arroces locales de grano fino y aromáticos se relacionó con el número inferior de retoños, el porcentaje más bajo de retoños estériles, el número mayor de granos/panícula, el porcentaje inferior de granos estériles, menor coeficiente de superficie foliar, ángulo de hoja apical más bajo e índice más alto de recolección. En los arroces locales, se podría cambiar en cierta medida la arquitectura de las plantas acortando la longitud de la segunda y tercera hojas, para impedir la inclinación, la sombra mutua y el encamado. La panícula deberá ser lo más grande posible.

Previous PageTop Of PageNext Page