Asian Journal of Agriculture and Rural Development, 3(5) 2013: 290-295 
 

290 
 

 

Screening Drought Tolerance Criteria in Maize 

 

 

Masoud Kiani 

Department of Plant Breeding, Collage of Agriculture, Azad University, Kermanshah Branch, Iran 

 

Abstract 

Six pure lines of maize were tested in a randomized complete block design with three replication 

under irrigated and rainfed conditions. Genetic variation was found between the genotypes for yield 

potential (Yp) stress yield (Ys), tolerance index (TOL), geometric mean productivity (GMP), 

harmonic mean (HM) and stress tolerance index (STI). Stress tolerance index was corrected using a 

correction coefficient (Ki) and thus a modified stress tolerance index (MSTI) was introduced as the 

optimal selection criterion for drought-tolerant genotypes. The results of three-D plotting indicated 

that the most desirable genotype for irrigated and rainfed conditions was the genotype K1515, for 

non-stressed conditions K18 and for stress conditions K104/3, K760/7 and K126/10.  

Keywords: Maize, Drought tolerance, Biplot, Modified stress tolerance index 

 
Introduction1  

 

The improvement of drought tolerance has 

been defined as a desirable breeding 

objective in crops such as maize (Clark et 

al., 1992). Drought tolerance in native plant 

species is often defined as survival, but in 

crop species it should be defined in terms of 

productivity (Passioura, 1983). The 

definition of drought tolerance as the ability 

of plants to grow satisfactorily when exposed 

to water deficits has little direct applicability 

                                                 
1 Corresponding author’s Email: 

   m_kiani1359@yahoo.com  

 

to either quantifying or breeding for the 

character in crop species (Clark et al., 1992). 

In the absence of an understanding of the 

special mechanisms of tolerance the 

quantification of drought tolerance should be 

based on the grain yield under dry conditions 

(Fischer and Maurer, 1978). It is worthwhile, 

therefore, to look at the methods that have 

been used to quantify tolerance. Several 

selection criteria are proposed to select 

genotypes based on their performance in 

stress and non-stress environments 

(Fernandez, 1992). Rosielle and Hamblin 

(1981) defined stress tolerance as the 

difference between grain yield in stress (Ys), 

and non-stress (Yp) environments, and mean 

 

 
 

Asian Journal of Agriculture and Rural Development 
 
 
 

journal homepage: http://aessweb.com/journal-detail.php?id=5005   

mailto:m_kiani1359@yahoo.com


Asian Journal of Agriculture and Rural Development, 3(5) 2013: 290-295 
 

291 
 

productivity (MP) as the average of Yp  and 

Ys. Fischer and Maurer (1878) proposed a 

stress susceptibility index (SSI). Fernandez 

(1992) defined a new stress tolerance index 

(STI). 

 

Genotypes can be categorized into four 

groups based on their performance in stress 

and non-stress environments: genotypes 

which express uniform superiority in both 

stress and non-stress environments (Group 

A); genotypes which perform favourably 

only in non-stress environments (Group B); 

genotypes which yield relatively well  

only in stress environments (Group C) and 

genotypes which perform poorly in both 

stress and non-stress environments  

(Group D). The optimal selection criteria 

should distinguish Group A from the other 

three groups (Fernandez, 1992). 

 

The objectives of the current experiment, 

carried out in the Agricultural research 

Station of Dezful, Iran the year 2000, were 

(i) the screening of quantitative criteria of 

drought tolerance, (ii) the introduction of a 

new drought tolerance index and (iii) the 

identification of drought tolerant genotypes. 

 

Materials and Methods 

 

Six pure lines of maize, namely K104/3(1), 

K760/7(2), K1515(3), K18(4), K19(5) and 

K126/1(6), were cultivated in a randomized 

complete block design with three 

replications under two different 

environments (irrigated and rainfed) in the 

Agricultural Research Station of Dezful, 

Iran. From each pure line 50 seeds were 

selected and single seeds were sown in 5m 

rows with 20×75cm plant to pant and row to 

row distances, respectively. The minimum 

and maximum temperatures at the station 

were 5.6°C and 54.6°C, respectively. The 

average rainfall in 2000 was 250mm and the 

region was arid. The chemical properties of 

the soil in the experiment were reported as: 

 

Value Soil properties 

0.04ds/m 
7.87 

0.48% 
2.55ppm 
6.7ppm 
101ppm 
1.31ppm 
3.17ppm 
0.32ppm 

E.C. 
PH 

O.C. 
Mn 
P 
K 
Cu 
Fe 
Zn 

 

Each plot consisted of 4 rows, the two 

middle rows being planted with the tested 

genotypes and the two lateral rows with the 

genotype SC 704 to eliminate the border 

effect. Ten competitive plants were 

randomly selected from each entry in both 

the irrigated and rainfed treatments and the 

yield potential (Yp) and stress yield (Ys) 

were recorded. Using Yp and Ys the 

following quantitative indices of drought 

tolerance were calculated: 

 

Tolerance index (TOL) and mean 

productivity (MP) (Rosielle and Hamblin, 

1981) : 

 

 
2

SYPY
andMPSYPYTOL


  

 

Harmonic mean (HM) (Zahravi, 1999): 

 



Asian Journal of Agriculture and Rural Development, 3(5) 2013: 290-295 
 

292 
 

 

SP

SP

YY

YY
HM






2
 

 

Stress susceptibility index (SSI) 

(Fischer and Maurer, 1978): 

)/(1;
)/(1

ps
ps YYSI

SI

YY
SSI 


  

        

Where SI is stress intensity and Ys and Yp 

are the means of all genotypes under stress 

and irrigated conditions, respectively. 

 

Geometric mean productivity (GMP) and 

stress tolerance index (STI) (Fernandez, 

1992; Kristin et al., 1997): 

 

2)(

)()(
,))((

p

ps
sp

Y

YY
STIYYGMP


  

 

Modified stress tolerance index (MSTI): 

 

2

2

22

2

1,
s

s

p

p
i

Y

Y
andK

Y

Y
KSTIkMSTI   

 

Where ki is the correction coefficient. 

 

Analysis of variance mean comparison, 

correlation analysis and three-dimensional 

plotting were done using the MSTAT-C and 

SPSS statistical softwares. 

 

 

 

 

Results and Discussion 

The results of analysis of variance (Table 1) 

for various quantitative criteria of drought 

tolerance showed highly significant 

differences for all the indices except SSI, 

indicating the presence of genetic variation 

and the possibility of selection for drought 

tolerant genotypes based on  Yp, Ys, TOL, 

GMP, HM and STI. 

 

Genetic variation between maize genotypes 

was reported for yield by Bolanos and 

Edmeades (1996) and Morris et al. (1991), 

for drought resistance by Vasal et al. (1997) 

and Banziger et al. (1997) and for Yp, Ys, 

TOL, MP, GMP, SSI, HM and STI by 

Ahmadzadeh (1997) and Afarinesh (2000). 

The estimates of stress tolerance attributes 

(Table 2) indicated that the identification of 

drought-tolerant genotypes based on a single 

criterion was contradictory. For example, 

according to TOL, the desirable drought-

tolerant genotype was K104/3(1), while 

according to STI the most desirable drought-

tolerant line was K1515(3). Moreover, MP 

failed to distinguish the group A and group B 

genotypes, while TOL and SSI failed to 

distinguish between group C and group A 

(Fernandez, 1992). To determine the most 

desirable drought tolerance criteria, the 

correlation coefficient between Yp, Ys and 

quantitative indices of drought tolerance was 

calculated (Table 2). 

 

 

 

 



Asian Journal of Agriculture and Rural Development, 3(5) 2013: 290-295 
 

293 
 

Table 1: Analysis of Variance for Different Indices of Drought Tolerance Irrigated and 

Rainfed Conditions 

Mean square           
d.f. 

Source of 
variation STI SSI HM GMP MP TOL Ys Yp 

0.83** 
0.03 
0.03 

0.60 
0.79 
0.38 

4262** 
68.9 

122.9 

4145** 
11.07 
76.90 

4232** 
26.18 
70.70 

766** 
1421* 
415.4 

3382** 
631 

418.8 

5480** 
534 

127* 

3 
2 
10 

Genotypes 
Replication 
Error 

  *, ** significant at the 5% and 1% probability level, respectively 

 

Table 2: Estimates of stress tolerance attributes from potential yield and stress yield data for 

maize genotypes 

K2STI K1STI STI SSI HM GMP MP TOL Ys Yp Lines 
11.8 
12.3 
22.6 
0.29 
3.91 
1.83 

9.4 
9.9 

34.7 
2.5 

10.2 
1.5 

0.19 
0.19 
0.30 
0.05 
0.14 
0.07 

0.27 
0.31 
1.32 
1.09 
1.9 

0.65 

50.1 
50.8 
62.8 
22.1 
44.4 
32.6 

50.2 
50.8 
64.1 
26.7 
44.8 
32.5 

50.2 
50.9 
65.5 
32.7 
46.5 
32.8 

3.6 
3.9 

25.7 
20.9 
29.8 
6.2 

48.4 
48.9 
42.9 
14.2 
31.6 
31.2 

51.9 
52.9 
78.4 
50.3 
61.5 
33.9 

K104/3 
K760/7 
K1515 
K18 
K19 
K126 

 

Table 3 showed that MP and STI had a 

positive significant correlation with Yp and 

Ys; thus, MP and STI were better predictors 

of mean Yp and mean Ys than the other 

indices. However, MP fails to distinguish 

between group A and group B, while STI is 

estimated based on GMP; the rank 

correlation between STI and GMP is thus 

equal to 1 (Rosielle and Hamblin, 1981; 

Fernandez, 1992). The higher the value of 

STI for a genotype, the higher its stress 

tolerance and yield potential. The stress 

intensity value is also incorporated in the 

estimation of STI. Therefore, STI is expected 

to distinguish group A from group B and 

group C. This result was in close agreement 

with the findings of Fernandez (1992), 

Maroufi (1998), Imamjomah (1999) and 

Farshadfar et al. (2001). 

 

In the STI index, Yp
2 is a constant value, 

while the square root of the multiplication of  

Yp and Ys is the geometric mean of a 

genotype under stress and non-stress 

condition. For this reason a pair of numbers 

with different natures may have the same 

geometric mean. For example, the geometric 

mean for the data pairs 1 and 12, 2 and 6, 

and 3 and 4 is 3.46, while these data, if 

related to the yield of the genotypes, have 

clearly different natures. This problem arises 

in the stress tolerance index (STI) and hence 

decreases its efficiency in distinguishing 

group A genotypes from the other group 

(Naderi et al., 1999). 

 

To improve the efficiency of STI a modified 

stress tolerance index (MSTI) was calculated 

as ki STI, where ki is a correction coefficient 

which corrects the STI as a weight. 

Therefore, k1 STI, k2 STI are the optimal 

selection indices for stress and non-stress 

condition, respectively. Considering Yp and 

Ys as dependent and  k1 STI, k2 STI and STI 

as independent variables, the contribution of 

k1 STI to Yp in relation to STI was R2=0.817, 



Asian Journal of Agriculture and Rural Development, 3(5) 2013: 290-295 
 

294 
 

while the contribution of  STI to Yp was  

R2=0.65. the contribution of k2 STI to Ys was 

R2=0.78, while that of STI was R2=0.72. 

thus, k1 STI, and k2 STI are better predictors 

of Yp and Ys respectively, in non-stress 

environments. 

  

Table 3: Correction Coefficients between Yp , Ys and Drought Tolerance Index 

K2STI K1STI STI SSI HM GMP MP TOL Ys Yp    
0.74 
0.88* 
0.03 
0.95** 
0.95** 
0.94** 
-0.12 
0.97** 
0.90* 
1 

0.90* 
0.66 
0.42 
0.92* 
0.87* 
0.83* 
0.28 
0.92** 
1 
- 

0.82* 
0.88* 
0.15 
0.99** 
0.99** 
0.97** 
0.034 
1 
- 
- 

0.50 
-0.34 
0.96** 
0.086 
0.014 
-0.08 
1 
- 
- 
- 

0.69 
0.96** 
-0.07 
0.97** 
0.99 
1 
- 
- 
- 
- 

0.77 
0.92* 
0.10 
0.99* 
1 
- 
- 
- 
- 
- 

0.85* 
0.86* 
0.20 
1 
- 
- 
- 
- 
- 
- 

0.65 
-0.29 
1 
- 
- 
- 
- 
- 
- 
- 

0.46 
1 
- 
- 
- 
- 
- 
- 
- 
- 

1 
- 
- 
- 
- 
- 
- 
- 
- 
- 

Yp 
Ys 
TOL 
MP 
GMP 
HM 
SSI 
STI 
K1STI 
K2STI 

*,** significant at the 5% and 1% levels of probability respectively 

 

Using STI, k1 STI and, k2 STI as the optimal 

selection criteria the most desirable 

genotypes for irrigated and rainfed 

conditions was K1515 (STI = 0.30, and k1 

STI = 34.7 and k2 STI = 22.6). A three-

dimensional plot between Yp, Ys and STI 

(Fig.1) was used to distinguish the group A 

genotypes from the other three groups (B, C 

and D) (Fernandez, 1992; Farshadfar et al., 

2001). In this case the most desirable 

genotype for irrigated and rainfed conditions 

was K1515, for non-stress conditions K18 

and for stress conditions K760/7 and 

K126/1. 

 

 

 

 

 

 

 

 

 

 

 

 

 

 

Fig 1: Three-dimensional Plot between Yp, Ys and STI 

 

 

 



Asian Journal of Agriculture and Rural Development, 3(5) 2013: 290-295 
 

295 
 

References 

Ahmadzadeh, A. (1997). Determination of 

the optimal drought tolerance index in 

maize. M.Sc. thesis. Tehran University, 

Iran. 

Afarinesh, A. (2000). Genetic analysis of 

drought resistance in maize using diallel 

cross. M.Sc thesis. Razi University, 

Kermanshah, Iran. 

Banziger, M., Edmeades, G. O. and Laffitte, 

H. R. (1997). Selection for drought 

tolerance increases in maize yield over a 

range of N levels. Crop Sci. 38: 120-127. 

Bolanos, J. and Edmeades, G. O. (1996). The 

importance of the anthesis-silking interval 

in breeding for drought tolerance in 

tropical maize. Field Crop Research, 48: 

65-80. 

Clark, J. M., Ronald, M. D. and Townley-

Smith, T. F. (1992). Evaluation of methods 

for quantification of drought tolerance in 

wheat. Crop Sci. 32: 723-728. 

Farshadfar, E., Ghanadha., M., Zahravi, M. 

and Sutka, J. (2001). Generation mean 

analysis of drought tolerance in wheat. 

Acta Agron. Hung. 49: 59-66. 

Fernandez, G. C. J. (1992). Effective 

selection criteria for assessing plant stress 

tolerance. In Proceedings of Symposium. 

Taiwan, 13-16 Aug. 1991. pp. 257-270. 

Fischer, R. A. and Maurer, R. (1978). 

Drought resistance in spring wheat 

cultivars: I. Grain yiel responses. Aus. J. 

Agric. Res. 29: 897- 912. 

Imamjomah, A., (1999). Determination of 

genetic distances by RAPD-PCR, 

evaluation of drought tolerance and 

analysis of adaptation in chickpea. M.Sc. 

thesis. Razi University, Kermanshah, Iran. 

Kristin, A. S., Serna, R. R., Perez, F. I., 

Enriquez, B. C., , Gallegos, J. A. A. 

Vallejo, P. R., Wassimi, N. and Kelly, J. 

D. (1997). Improving common bean 

performance under drought stress. Crop 

Sci. 37: 51–60. 

Maroufi, A. (1998). Chromosome 

localization of the genes controlling 

drought tolerance indices. M. Sc. Thesis. 

Razi University, Kermanshah, Iran. 

Morris, M. L., Belaid, A. and Byerlee, D. 

(1991). Wheat and barley production in 

rainfed marginal environment of the 

developing world. CIMMYT, Mexico, D. 

F. 

Naderi, A., Majidi, I., Dezfuli, A. H., Rezai, 

A. M. and Mohammad, G. N. (1999). 

Efficiency analysis of indices for tolerance 

to environmental stresses in field crops 

and introduction of a new index. Seed and 

Plant, 4: 390–402. 

Passioura, J. B. (1983). Roots and drought 

resistance. Agric.Water Manage. 7: 265-

280.  

Rosielle, A. A. and Hamblin, J. (1981). 

Theoretical aspects of selection for yield in 

stress and non-stress environments. Crop 

Sci. 21: 943–946. 

Vasal, D. K., Cordona, H. S., Beck, D. L. 

and Edmeades, G. O. (1997). Choices 

among breeding procedures and strategies 

for developing stress-tolerant maize 

germplasm. Proceeding of a symposium. 

March 25-25. CIMMYT, Mexico. Pp. 336-

347. 

Zahravi, M. (1999). Genetic analysis of 

drought tolerance in wheat. M.Sc. Thesis. 

Razi University, Kermanshah, Iran.