[ホーム発表論文>本文]

吉田総合案内へ戻る.
本頁は"Plant Prod. Sci. 3:246-253,2000"に掲載したものです. 図表一部省略. Fig.5試験管とポットの比較 

Screening of Al-tolerant Sorghum by Hematoxylin Staining and Growth Response

Anas and Tomohiko Yoshida
(Faculty of Agriculture, Kyushu University, Fukuoka 812-8581, Japan)
Abstract: Hematoxylin staining of root (hematoxylin staining method) and the growth response to Al added to the soil in pots (growth-response method) were used to select Al-tolerant sorghum. Twenty-two sorghum genotypes were screened by the hematoxylin staining method, and twelve genotypes selected were evaluated again by the growth-response method. Five genotypes (G4, H11xC8, Real 60, SPA2, and SPAD) showed consistent Al tolerance by both screening methods. Genotype C9xH13 was susceptible to Al. A significant correlation (r = 0.666**) was observed between the score of hematoxylin staining method and that of relative root length (RRL). The value of RRL also showed a significant correlation with the score of hematoxylin staining (r = 0.622*). The roots of genotypes that were not stained by hematoxylin tended to have a long root in the soil with Al added.

Key words: Al-tolerant, Growth response, Hematoxylin staining, Sorghum.

Received 2 August 1999. Accepted 2 February 2000. Corresponding author: T. Yoshida (yoshidat@agr.kyushu-u.ac.jp, fax +81-92-642-2819).
Abbreviations : RRL;relative root length, RSL;relative shoot length

Aluminum (Al3+) toxicity is considered as a major constraint for the production of maize, sorghum, and rice in acid soil. Al damages the root apex (root cap) and inhibits root hair growth, resulting in nutrient deficiency and leaf disorder (Delhaize and Ryan, 1995; David et al., 1997; Chang et al., 1998; Godbold and Jentschke, 1998). Approximately 68% of America’s tropical land, 38% of Asia’s tropical land and 27% of Africa’s tropical land are classified as acid soil (Martinez and Estrella, 1999; Delhaize and Ryan, 1995). Acid soil covers approximately two-thirds of the world’s humid tropics (Sanchez, 1976). Most of soils in Latin American planted with sorghum are acidic (Baligar et al., 1993).
In general, the Al-screening technique can be classified into laboratory screening and field screening. Laboratory screenings methods include screening of plants with solution-soaked paper and solution culture methods (Konzak, et al., 1976), soil-petri dish method (Hill et al., 1989), and screening in pots in a greenhouse (Baligar et al., 1989). For sorghum, screening in the field (Duncan, 1988; Flores et al., 1988; Miller et al., 1992), and in pots or nutrient solution (Furlani and Clark, 1981; Boye and Marcarian, 1985; Gourley et al., 1990; Miller et al., 1992) are commonly used for selection of Al-tolerant genotypes.
A rapid screening method is needed to select a large number of new genotypes or new inbred lines in plant breeding, such as solution-soaked paper, solution culture and soil-petri dish methods used to evaluate Al-tolerant sorghum. All of these rapid screening techniques use the response to Al of the rate of seedling germination and root development. However, the method using such growth responses would curtail the accuracy of screening. Detection systems not dependent on the rate of seedling or root development, would greatly improve the success of screening procedure (Konzak et al., 1976).
Screening by using hematoxylin staining of seedling roots (hematoxylin staining method) which requires less time and simpler pH management than the other methods, is very useful for selection or screening a relatively large population in breeding program. Measurement of Al tolerance is based on the staining pattern of the root. The hematoxylin staining method is a very common technique for the evaluation of Al-tolerance in wheat (Polle et al., 1978; Takagi et al., 1981; Wallace et al., 1982) and barley (Minella and Sorrells, 1992), but there have been no reports on the use of hematoxylin staining methods in the screening of Al-tolerant sorghum.
For selection of genotypes tolerant to Al, a precise screening technique to evaluate sensitivity of plants to Al is needed. Therefore, the results of the hematoxylin staining method should be compared with other screening methods.
Field screening for Al tolerance would be the best approximate for selecting Al-tolerant plants. In practice, however, reliable ranking of tolerance in the field screening is difficult because the Al concentration in soil may not be uniform and because environmental factors interact with soil Al to mask the expression of Al tolerance (Champbell and Carter, 1990). Screening by using the growth response to Al added to the soil in pots at in a greenhouse (referred to as growth-response method hereafter) may be superior in this respect.
It is important to compare the laboratory screening methods with the field screening methods. There was a correlation between the performance of sorghum in the greenhouse study and grain yield in the field (Baligar et al., 1989). The plants that showed severe reduction of shoot or root weight in a greenhouse study showed also low grain yield in the field. There was also a similar genotype response to Al-induced stress in nutrient solution and to acid-soil stress in the field (Duncan et al., 1983).
The objectives of this research were to screen Al-tolerant sorghum by using the hematoxylin staining method and to compare the results with those obtained by the growth-response method mentioned above.

Materials and Methods

1. Genotype
Twenty-two sorghum genotypes were evaluated for their sensitivity to Al by the hematoxylin staining method. The materials consisted of parental lines, inbred lines and Al-tolerant lines. Parental lines and inbred lines were obtained from the germplasm collection of Crop Science Laboratory, Kyushu University and Chuugoku National Agricultural Experiment Station, Japan. ICRISAT (India) and USDA (USA) kindly provided tolerant lines. Real 60, SPA2 and SPAD (tolerant lines) have been screened under 60%-80% Al saturation at Carimagua, Matazul and Chilichau-Columbia by ICRISAT (Reddy, personal communication). Table 1 lists the genotypes and origin.
Twelve genotypes selected by the hematoxylin staining method were evaluated again for acid soil tolerance in a pot: G3, G4, G6, G7, G8, H11xC8, C9xH11, C8xD12, C9xH13, Real 60, SPA2, and SPAD. These genotypes were chosen for a wide range of Al tolerance. Based on the data obtained by hematoxylin staining method, SPA2 and C9xH11 were selected as representative tolerant and susceptible genotypes, respectively. Although C9xH11 genotype was not the most susceptible genotype with the hematoxylin staining method, it was chosen because it had also early maturity character in the field (Can and Yoshida, 1999). Genotype PI 533869 was evaluated only by the culture on acid soil, because the number of seeds was very limited. In addition, the growth of PI 533869 genotype was very poor during the preliminary test by the hematoxylin staining method.

2. Hematoxylin staining method
A hematoxylin staining method based on the technique of Polle et al. (1978) with some modifications was used. Levels of Al concentration of 0, 17.87, 35.73, 53.60, and 71.46 ppm Al (0, 74, 148, 222, and 296 mM) were designed. Al was added to the distilled water from 0.1 M AlCl3・6H2O stock solution. The choice of these Al concentrations was based on the sorghum screening for Al tolerance in nutrient solution (Furlani and Clark, 1981).
Seeds were germinated directly on the planting tray in the plastic container filled with distilled water and kept in an incubator in darkness at 25oC for 29 hours on February 20, 1999. The tray was covered with thin clear plastic plate. Planting trays (19.3cm x 25.3cm) had 560 holes (0.7cm x 0.7cm) to place the seeds. A nylon fabric screen was glued onto the tray. After the radicle had completely emerged, water in the plastic container was replaced with nutrient solution (4.0 mM CaCl2・H2O, 6.5 mM KNO3, 2.5 mM MgCl2・6H2O, 0.1 mM (NH4)2SO4, 0.4 mM NH4NO3) that was adjusted to pH 4.0 with 0.25 M HCl. Seedlings were grown for 31 hours on the above nutrient solution in the growth chamber under the light at 25oC. The surface of the planting tray with the seedlings was covered with a thin clear plastic plate. The seedlings were then grown for 17 hours on the nutrient solution with 0 to 296 mM Al added without plastic cover and kept in the growth chamber in the light at 25oC. The planting tray was then placed on distilled water for 30 min and placed in a container with 0.2% hematoxylin solution for 15 minutes. Prior to evaluation of staining, seedlings were washed with distilled water for 30 minutes. Stained root tips of six seedlings per genotype were then evaluated for scoring. The hematoxylin staining was scored based on the staining of roots as shown in Table 2.

3. Growth-response method
The homogenous soil “Kumiai Kokuryuubaido” (Zennou, Japan) was used for pot experiment. The “Kumiai Kokuryuubaido” contains micronutrients, heat dried humus and fertilizer (0.28 N Kg-1, 0.28 P Kg-1, 0.28 K Kg-1). About 350 g soil in a pot 9.5 cm in diameter and 9 cm in depth, was mixed with 100 mL AlCl3・6H2O solution at various concentrations (8.32, 16.5, and 24.9 mM) to prepare the soil with different acidity.
The Al solution was adjusted to around pH 3 to 3.5 by HCl. The soil mixed with Al solution was kept for 24 hours and then the seeds were sown on June 22, 1999, in a greenhouse. The greenhouse temperature during the experiment ranged from 25oC to 28oC in the day time and 23oC to 25oC at night. Five seeds were sown per pot and thinned to four plants after seedling emergence. Three days after sowing, 40 mL of distilled water was added to keep the soil moist. Seven days after planting, 40 mL of AlCl3・6H2O solution was added to the soil.
In a preliminary experiment, Al at various concentrations (11.18 to 671.00 ppm Al) was added to the soil to determine the Al concentration that gives the widest range of toxic symptoms in the root. The soils containing 224 ppm, 447 ppm, and 671 ppm Al (8.32, 16.5, and 24.9 mM) gave significant differences in the root symptoms (data not shown), and used for screening in this experiment. Baligar et al. (1993) used dark red latosol soil with 2, 41 and 64 % Al saturation to observe the growth and uptake parameter of sorghum. Flores et al. (1988) used 40% and 60% Al saturation of acid ultisol soil to determine the effects of Al saturation on the growth and yield of sorghum. De Sousa (1998) used acid soil (pH 4.2 to 4.9) that contained 25 to 42.7 mM Al for classification of Al tolerance of wheat cultivars. On the other hand, the Al at the concentration of 0 to 296 mM was usually added to the complete nutrient solution for the selection of tolerant sorghum or physiological investigation (Gourley et al., 1990; Tang and Keltjens, 1995; Furlani and Clark, 1981).
Plants were harvested 15 days after sowing, roots were rinsed with stream flow of water tap, and data for the longest root length (from the base of the stem to root tip) and shoot length (from the base of stem to the longest tip of the longest leaf) was measured. The relative root length (RRL) (the longest root length in soil with Al / the longest root length in soil without Al) x 100 was scored as shown in Table 2.
To achieve a greater degree of precision for genotype comparison than for Al concentration response, we used the split plot experimental design with genotype assigned to the sub plot and Al-concentration to the main plot.

Results and Discussion
1. Hematoxylin staining method
Six seedlings per genotype with well-developed roots 1.5 cm or longer were chosen for scoring in the hematoxylin staining test. Hematoxylin staining of root is used to determine Al-tolerant genotype (Table 3). Although the length of seedling roots varied either in the solution with or without Al, Polle et al. (1978) found no relationship between the staining response of genotype and the length of seedling roots in the hematoxylin staining test. In addition, the seedlings were treated with Al for a short time.
The score ranged from 1 to 5. Eight genotypes showed the score for 4 or 5 (susceptible); seven genotypes showed score 3 (intermediately tolerant); and six genotypes showed score 1 or 2 (tolerant, Table 3). Variation in the hematoxylin score from complete staining to no staining was also observed in wheat, barley and pearl millet (Takagi et al., 1981; Minella and Sorrells, 1992; Yoshida and Shigemune, 1999). Minella and Sorrells (1992) observed difference in staining pattern of root to select 37 barley genotypes, and determined the loci number and allelic relationships among barley genotypes of diverse origins.
Roots of G4 and SPA2 genotypes were not stained at any Al levels, indicating that they are very tolerant (Table 3). Polle et al. (1978) and Takagi et al. (1981) reported that the roots of a tolerant wheat genotype were not stained with hematoxylin. This may be mainly due to the high pH of the cell wall in a tolerant plant. The high pH immobilizes Al and thus protects the plants from Al-toxicity (Ownby, 1993; Andrade et al., 1997) (Fig. 1).
All genotypes from ICRISAT (Real 60, SPA2, SPAD) showed tolerant conformity with the score of hematoxylin staining. These genotypes were classified by ICRISAT as tolerant genotypes, and their roots were not stained in a solution even with 53.60 ppm Al (Table 2 and Fig.1). In contrast, roots of G2, G8, and G9-1 genotypes exhibited complete staining at all Al levels and were classified as very susceptible genotypes.
G3, G6, G7, H11xH13, H11xD12, C8xD12, and TxxH13 were classified as genotypes with intermediate tolerance. Their roots exhibited staining either at 53.60 or 71.46 ppm Al. Although we did not study the segregation pattern here, crossing between the parents of intermediate tolerance (G7 and G3) resulted in a susceptible progeny (C9xH13). Boye and Marcarian (1985) reported that predominantly additive genetic effects with some degree of dominance controlled the Al tolerance trait in sorghum, while Gourley et al. (1990) reported that tolerance to Al-toxicity in sorghum was inherited as a dominant character. Genotype C9xH13 had high yield performance in the field (Can, personal communication) but was shown to be susceptible to Al toxicity.

2. Growth-response method
The primary effect of Al toxicity was the restriction of root growth. Visual symptoms of Al toxicity in roots became clearer as the Al concentration in the soil increased. Injured roots did not branch normally, and were shorter than the roots grown in soil with lower Al concentrations. Roots tended to become swollen with a stubby appearance in response to Al (Fig. 2).
Baligar et al. (1993) reported 93% and 52% reduction of shoot and root growth in sorghum at 64% Al saturation of natural acid soil (Typic Haplorthox). Shuman et al. (1990) found that 50 % Al-saturated topsoil and 40% Al-saturated subsoil were toxic to sorghum. On the other hand, Flores et al. (1988) reported that 40% or less Al saturation of ultisol soil or less gave no impact on the growth of sorghum. The sorghum growth decreased sharply at 148 mM Al in complete nutrient solution (Ohki, 1987). Champbell and Carter (1990) observed 30% reduction of soybean growth at 55% Al saturation (amended with 35 mmol kg-1 Al) and, 39% reduction in the solution culture with 7 to 15 mM Al.
The solubility of the Al compound and severity of their toxic effect on the plant are influenced by many chemicals and physical factors such as pH, oxidation-reduction potential, the composition of clay minerals, organic matter and exchangeable cation concentration. Dong et al. (1999) reported that soil pH was the major factor that controls Al3+ availability and uptake of Al from soil into tea plants. The Al3+ ion will be dominant when the soil pH is less than 5.0. Because Al toxicity changes with soil pH, we also measured the soil pH at the end of experiment. The soil pH was still under 4.7 in all pots at the end of our experiment, indicating that Al3+ in the soil was available to the plant throughout the experiment.
Organic matter in the soil can reduce the toxicity of Al to the plants (Shuman et al., 1990). Kapland and Estes (1985) reported that the critical level of Al in an Alfalfa pot experiment increased as the soil organic matter level increased from 6.6 to 81.6 g kg-1. In this experiment, we needed high concentration of Al (8.32, 16.5, and 24.9 mM) to obtain significant effect on root growth among genotypes. This may be mainly due to the chemical and physical factors and content of organic matter in the soil used.
RRL of all genotypes decreased in response to the increase in Al concentration (Fig 3). Negative values of correlation coefficients between RRL and Al concentration were observed for all genotypes (Table 4). Inhibition of root elongation is widely recognized as Al stress symptoms in plants (Delhaize and Ryan, 1995; Gallego and Benito, 1997; Martinez and Estrella, 1999). Tang and Keltjens (1995) reported that Al toxicity was expressed by the direct damage of roots (stubby and discolored root) with a concomitant reduction in specific root length.
Utilizing shoot length as a criterion of selection for Al toxicity is not reliable, because relative shoot length (RSL) of G6 did not show a significant correlation (r = - 0.523) with Al concentration, and some genotypes showed a weak correlation with Al concentration. Shoot growth measurements often produced contradictory results compared with total root length measurements (Bushamuka and Zobel, 1998).
The 447 ppm Al was used as an index of Al tolerance. Analysis of variance showed that the effect of Al concentration was not significant in this experiment (Table 5). It may be due to the use of the split plot design for data analysis. With split plot design, the precision for the measurement of the effects of the main-plot factor (Al concentration) was sacrificed to improve that of the subplot factor (genotype) (Gomez and Gomez, 1984). Therefore, evaluation of Al-tolerant genotypes at four levels Al concentrations would probably give greater precision for genotype comparison than for response to Al concentration.
Analysis of variance also revealed a significant effect of interaction between the genotype and Al concentration (G x C), indicating the different response of genotypes with the Al concentration. Tan and Keltjesns (1995) reported that plant species and genotypes of the same species often differ in sensitivity to acid soil. They also reported that root damage well reflected genetic differences in response to aluminum at a high acidity.
The mean RRL at different Al concentrations showed that the 447 ppm Al gave the widest range of toxic response among genotypes. Eight genotypes were tolerant, three intermediately tolerant and two susceptible (Table 3). G3, G4, G7, Real 60, SPA2, and SPAD were significantly more tolerant than G6, G8, C9xH13 and C9xH11 (Table 6). RRL of plants grown with 224 ppm Al and 671 ppm Al also showed Al toxicity symptoms but differences among genotypes were difficult to distinguish. The means RRL was not significant in the soil with 224 ppm Al, while, in the soil with 671 ppm Al, only one genotype showed a significant difference in RRL (Table 6).
The score of hematoxylin staining showed a significant correlation (r = 0.666**) with that of RRL (Fig.4). Genotypes that showed tolerance with the hematoxylin staining method showed also tolerance with the growth-response method (Table 3). Polle et al. (1978) reported that the results of the hematoxylin test were in good agreement with those of the acid soil-screening method in wheat. Takagi et al. (1981) also reported that scoring of Al tolerance of wheat by the hematoxylin staining method strongly coincided with the scoring of Al tolerance in nutrient solution with Al.
A significant effect was also shown in the correlation between RRL and the score of hematoxylin staining (r = 0.622*) (Fig. 5). The roots of genotypes that were not stained by hematoxylin also tended to have long roots in the soil with Al added. On the other hand, the RSL showed no significant correlation with the score of hematoxylin staining (Fig. 6). This is in agreement with the fact that shoot length was not correlated with Al concentration in the growth-response method.

Conclusion
Six genotypes (G4, H11xC8, H11xH2, Real 60, SPA2, and SPAD) showed very tolerant and tolerance to Al by hematoxylin staining test, and six genotypes (G3, G4, G7, Real 60, SPA2, and SPAD) by the growth-response method. Five genotypes (G4, H11xC8, Real 60, SPA2, SPAD) showed tolerance to Al toxicity by both screening methods. On the other hand, C9xH13 showed susceptibility to Al by both methods. These facts show that a similar evaluation for Al tolerance can be obtained by these two methods.
We can use either one of these methods to select Al-tolerant plants. For evaluation of a large number genotypes, the hematoxylin staining method is more convenient than the growth-response method. The hematoxylin staining procedure is simple and requires little space and labor. However, screening Al-tolerant sorghum to Al toxicity in the field (natural acid soil) is still needed, especially for comprehensive information about yield and growth of the tolerant plants. The growth-response method can be used as a preliminary test for field test.

References

Andrade, L.R.M., Ikeda, M. and Ishizuka, J. 1997. Localization of aluminum in root tip tissues of wheat 
 varieties differing in aluminum tolerance. J. Fac. Agr., Kyushu Univ. 41:151 -156.
Baligar, V.C., Dos Santos, H.L., Pitta, G.V.E., Filho, E.C., Vasconcellos, C.A. and Filho, A.F.D.C.B. 1989. 
  Aluminum effects on growth, grain yield and nutrient use effeciency ratios in sorghum genotypes. Plant Soil 
 116:257 - 264.
Boye-Goni, S.R. and Marcarian, V. 1985. Diallel analysis of aluminum tolerance in selected lines of grain  
  sorghum. Crop Sci. 25:749 - 752.
Bushamuka, V.N. and Zobel, R.W.1998. Maize and soybean tap, basal and lateral root responses to a 
  stratified acid, aluminum-toxic soil. Crop Sci. 38:416 - 421.
Can, N.D. and Yoshida, T. 1999. Breeding for high yield, early maturity and short plant height in sorghum: 
  heterosis and combining ability. Proc. Int. Symp."World Food Security", Kyoto: 129-132. 
Champbell, G.K.A. and Carter, Jr. T.E. 1990. Aluminum tolerance in soybean: I. Genotypic correlation and 
  repeatability of solution culture and greenhouse sceening method. Crop Sci. 30:1049 -1054.
Chang, Y.C., Ma, J.F. and Matsumoto, H. 1998. Mechanisms of Al-induced iron clorosis in wheat (Triticum 
  aestivum L.). Al-inhibited biosynthesis and secretion of phytosiderophore. Physiologia Plantarum 102:9 -15.
David, J., Simon, G., Paul, L., Stephen, H. and Leon, K. 1997. Effect of aluminum oxidative, anaerobic and 
  mechanical stress on cytoplasmic CA2+ homeostasis in roots hairs of Arabidopsis thaliana. Tetran-Agricultural 
  Research Servive Home page-US.
Delhaize, E. and Ryan, P.R. 1995. Aluminum toxicity and tolerance in plants. Plant Physiol. 107:315 - 321.
De Sousa, C. N. A. 1998. Classification of Brazillian wheat cultivars for aluminium toxicity in acid soils. Plant 
  Breeding 117:217-221.
Dong, D.,  Xie, Z., Du, Y., Liu, C. and Wang, S. 1999. Influence of soil pH on aluminum availability in the soil 
  and aluminum in tea leaves. Commun. Soil Sci. Plant Anal. 30:873 - 883.
Duncan, R.R., Clark, R.B. and Furlani, P.R. 1983. Laboratory and field evaluations of sorghum for response to 
  aluminum and acid soil. Agron. J. 75:1023 - 1026.
Duncan, R.R. 1988. Sequential development of acid soil tolerant sorghum genotypes under field stress 
  conditions. Commun. Soil Sci. Plant Anal. 19: 1295 - 1305.
Flores, C.I., Clark, R.B. and Gourley, L.M. 1988. Growth and yield traits of sorghum grown on acid soil at 
  varied aluminum saturation. Plant and Soil. 106:49-57.
Furlani, P.R. and Clark, R.B. 1981. Screening sorghum for aluminum tolerance in nutrient solution. Agron. J. 
  73:587 - 594.
Gallego, F.J. and  Benito, C. 1997. Genetic control of aluminium tolerance in rye (Secale cereale L.). Theor. 
  Appl. Genet. 95:393 - 399.
Godbold, D.L. and Jentschke, G. 1998. Aluminium accumulation in root cell walls coincides with inhibition of 
  root growth but not with inhibition of magnesium uptake in Norway spruce. Physiol. Plant. 102:553 - 560.
Gomez, K.A. and Gomez, A.A. 1984. Statistical Procedures for Agricultural Research. John Wiley & Sons, 
  New York. 85 -129.
Gourley, L.M., Rogers, S.A., Gomez, C.R. and Clark,R.B. 1990. Genetic aspects of aluminum tolerance in 
  sorghum. Plant Soil 123:211 - 216.
Hill, P.R., Ahlrichs, J.L. and Ejeta, G. 1989. Rapid evaluation of sorghum for aluminum tolerance. Plant Soil 
  114:85 - 90.
Kaplan, D.I. and  Estes, G.O. 1985. Organic matter relationship to soil nutrient status and aluminum toxicity in 
  alfalfa. Agron. J. 77:735 - 738.
Konzak, C.F., Polle E. and Kittrick J.A. 1976. Screening several crops for aluminum tolerance. In Wright, M.J 
  and Ferrari, A.S. (ed) Proceeding of Workshop on Plant Adaptation to Mineral Stress in Problem Soils 
  (Beltsville, MD) Cornell Univ. Press, Ithaca, New York. 311 - 327. 
Martinez, J.M.D.F. and Estrella, L.H. 1999. Advance in the understanding of aluminum toxicity and the 
  development of aluminum tolerant transgenic plants. In Donald L.P. ed. Advances in Agronomy Vol.66. 
  Academic Press, San Diego-USA. 103 -120. 
Miller, D.R., Waskom, R.M., Duncan, R.R., Chapman, P.L., Brick, M.A., Hanning, G.E., Timm, D.A. and 
  Nabors, M.W. 1992. Acid soil stress tolerance in tissue culture-derived sorghum lines. Crop Sci. 32:324 - 327. 
Minella, E. and Sorrells, M.E.1992. Aluminum tolerance in barley: genetic relationships among genotypes of 
  diverse origin. Crop Sci. 32:593 - 598.
Ohki, K. 1987. Aluminum stress on sorghum growth and nutrient relationships. Plant Soil 98:195 - 202. 
Ownby, J.D.  1993. Mechanisms of reaction of hematoxylin with aluminum-treated wheat roots. Physiol. Plant. 
  87:371 - 380.
Polle, E., Konzak, C.F. and Kittrick, J.A. 1978. Visual detection of aluminum tolerance levels in wheat by 
  hematoxylin staining of seedling root. Crop Sci. 18:823 - 827.
Sanchez, P.A. 1976. Properties and Management of Soils in The Tropics. John Wiley & Sons, New York.
Shuman, L.M., Ramseur, E.L. and Duncan,R.R. 1990. Soil aluminum effects on the growth and aluminum 
  concentration of sorghum. Agron. J. 82:313 - 318.
Takagi, H., Namai, H. and Murakami, K. 1981. Evaluation of the hematoxylin staining method for detecting 
  wheat tolerance to aluminum. Japan J. Breed. 31:152 -160*.
Tang, K. and Keltjens, W.G. 1995. Analysis of acid-soil stress in sorghum genotypes with emphasis on 
  aluminium and magnesium interactions. Plant Soil 171:147 -150.
Wallace, S.U., Henning, S.J. and Anderson, I.C.1982. Elongation, Al concentration, and hematoxylin staining 
  of aluminum-treated wheat roots. Iowa State J. of Research. Vol. 57, No.1: 97 -106.
Yoshida, T. and Shigemune, A. 1999. Mass selection for drought, salt, aluminum tolerance and heritabilities in 
  pearl millet. Rep. Kyushu Br. Crop Sci. Soc. Japan, 65:54 - 56**.

*	In Japanese with English summary.
** In Japanese.

Fig 1. Staining of a root tip of a susceptible and a tolerant genotype by the hematoxylin staining method.
Fig 2. A root in soil with Al at a high concentration.
Fig 3. Decrease of relative root length (RRL) in response to increased Al concentrations determined by the 
growth-response method.
Fig 4. Correlation between the score of hematoxylin staining and that of relative root length (RRL). **Significant 
at the 1% level. ◆ G6, G8, C9xH11, C9xH13, H11xC8, C8xD12; ■■ G4 and SPA2; ▲▲ Real 60 and SPAD; 
●● G3 and G7.


Fig 5. Correlation between the score of hematoxylin staining and relative root length (RRL) in the soil containing 
447 ppm Al. * Significant at the 5% level.
Fig 6. Correlation between the score of hematoxylin staining and relative shoot length (RSL) determined in the 
soil containing 447 ppm Al.  ●● G7 and C8xD12, ◆ G3, G4, G6, G7, C9xH11, C9xH13, H11xC8, Real 60, 
SPA2, and SPAD.

Table 1. List of genotypes and their origin.
-------------------------------------------------------------------------- 
Genotypes             Name	                                     Origin 
--------------------------------------------------------------------------
Parental lines      G2, G3, G4, G8, G9-1	                   Collection of Crop Sci. Lab., Kyushu Univ.	
	           G6, G7	                                 Chugoku Natl. Agric. Exp. Station, Japan.	
Inbred lines         C9xD12, C9xH11 (high yield),          Breeding lines, Crop Sci. Lab., Kyushu
                         C9xH13, C9xH2, H11xC8 (high        Univ., selected from the cross mentioned
                         yield), H11xH2, H11xH13,H11xD12,  in the name.	
                         C8xD12, TxXH13, H13xD12	
 Tolerant lines      Real 60, SPA2, SPAD 	        ICRISAT (INDIA)	
	            PI 533869	                                   USDA (USA)	
---------------------------------------------------------------------------
Note: SPA2;SPA2 94039B,  SPAD;SPAD 940006B.

Table 2. Scoring of hematoxylin staining at various Al concentrations 
and the relative root length (RRL) in acid soil containing 447 ppm 
in relation to Al tolerance. 
------------------------------------------------------------------
Staining pattern of root tips	RRL (%) 	 Score	 Remark	
----------------------
Al concentration (ppm)				
17.87	35.73	53.60	71.46	447 ppm Al
------------------------------------------------------------------		
NS	NS	NS	NS	79.2 ≦ 91.9	1	Very tolerant
NS	NS	NS	S	66.5 ≦ 79.2	2	Tolerant
NS	NS	S	S	53.8 ≦ 66.5	3	Intermediate
NS	S	S	S	41.1 ≦ 53.8	4	Susceptible
S	S	S	S	0  ≦ 41.1	5	Very susceptible
-------------------------------------------------------------------- 
Note: NS;not stained, S;stained.

Table 3.	Aluminum tolerance of sorghum genotypes evaluated by hematoxylin staining and growth response method.
--------------------------------------------------------------------
No      Genotype Score of      Score     No           Genotype         Score of        No
of                      hematoxylin  of         of                                    hematoxylin  of
plant                  staining       RRL    plant	                       staining       RRL
--------------------------------------------------------------------
1	G2 	5	-	12	H11xC8               2	2	
2	G3 	3	1	13	H11xH2	             2	-	
3	G4	1	1	14	H11xH13	             3	-	
4	G6	3	4	15	H11xD12         	3	-	
5	G7	3	1	16	C8xD12     	3	3	
6	G8	5	3	17	TxXH13	             3	-	
7	G9-1	5	-	18	H13xD12     	4	-	
8	C9xD12 	4	-	19	Real 60     	2	1	
9	C9xH11	4	3	20	SPA2       	1	1	
10	C9xH13	4	5	21	SPAD       	2	1	
11	C9xH2	4	-	22	PI 533869	-	2	
---------------------------------------------------------------------
SPA2;SPA294039B,  SPAD;SPAD940006B, score of hematoxylin staining was based on staining pattern of 
root; score of growth-response was based on RRL at 447 ppm Al.

Table 4. Correlation coefficients between root length (RRL) or relative 
shoot length (RSL) and Al concentration of 13 genotype sorghum 
determined by the growth-response method.
-------------------------------------
Genotype	Root length	Shoot length
-------------------------------------
G3	- 0.843 **	- 0.617 *
G4	- 0.873 **	- 0.859 **
G6	- 0.984 **	- 0.523
G7	- 0.819 **	- 0.995 **
G8	- 0.936**	- 0.936 **
H11xC8	- 0.924 **	- 0.958 **
C9xH11	- 0.969 **	- 0.934 **
C8xD12	- 0.968 **	- 0.993 **
C9xH13	- 0.969 **	- 0.854 **
Real 60	- 0.867 **	- 0.972 **
SPA2	- 0.950 **	- 0.605 *
SPAD	- 0.834 **	- 0.877 **
PI533869	- 0.931 **	- 0.878 **
-------------------------------------
*, **; significant at 5% and 1% level, respectively.

Table 5. Analysis of variance for relative root length (RRL) and 
relative relative shoot length (RSL) determined by the growth-
response method.
------------------------------------------------
Source	            df	Means square
                                     --------------------------
		               RRL	    RSL	
------------------------------------------------
Replication ( r )	1	  164.934	     135.387
Al level ( C )	3 	35529.999    16882.172
  Error a    	3	  135.864	     268.862
Genotype ( G )	12	  503.238**  947.498**
G x C	              36	  275.501**  523.770**
  Error b	              48	  128.119	     220.754
------------------------------------------------
**; significant at 1% level. 

Table 6. Means of relative root length (RRL) and relative shoot length (RSL) of 13 genotype 
determined by the growth-response method.
-----------------------------------------------------------------------------
Genotype	                 RRL (%) 		                RSL (%)	
              ----------------------------------    ---------------------------------
	                                                Al concentration (ppm)	
	0	224	  447	    671	     0	224	 447	   671
------------------------------------------------------------------------------	
G3	100 a 100.910 a	85.830 ab    4.505 b  100 a 110.805 ab	 121.670 a  32.120 b
G4	100 a  80.865 ab	82.500 abc 13.160 b  100 a  73.405 cd	  84.120 abc 32.860 b	
G6	100 a  62.135 b	49.585 de   18.430 b  100 a  57.635 d    112.220 ab  30.910 b
G7	100 a  99.500 a	91.900 a	     8.470 b  100 a  82.200 a-d  60.190 d   31.855 b
G8	100 a  99.380 a	64.215 b-e  17.145 b  100 a  95.780 abc  76.495 cd  37.590 b
H11xC8	100 a  84.155 ab	73.880 a-d  15.330 b  100 a	 94.060 abc  72.835 cd  39.990 b
C9xH11	100 a  87.430 a	54.530 de      7.115 b  100 a	 90.895 a-d  76.765 cd  32.090 b
C8xD12	100 a  87.270 a	58.910 cde   12.710 b  100 a	 78.050 bcd  60.400 d    28.945 b
C9xH13	100 a  92.075 a	41.135 e	      5.660 b  100 a	 78.340 bcd  86.880 bcd 32.910 b
Real 60	100 a  89.295 a	83.805 abc   11.260 b  100 a	 86.375 a-d  75.040 cd   46.785 b	
SPA2	100 a  97.360 a	86.185 ab     67.600 a  100 a	114.615 a    102.235 abc 118.385 a	
SPAD	100 a  96.360 a	90.345 ab     17.365 b  100 a	111.805 ab    76.525 cd   47.285 b	
PI 533869100 a 100.920 a	68.635 a-d   26.675 b  100 a	114.310 a      58.715 d    38.520 b	
-------------------------------------------------------------------------------
Means followed by a common letter in the column are not significantly different at the 5% level by DMRT.


以上