Comparative Evaluation of Water Deficit Tolerance Capacity of Extra-early and Early Maize Genotypes under Controlled Conditions

Maize is an important staple food crop in Tropical Africa including Nigeria. However, the production of the crop is constrained by inadequate soil moisture resulting from erratic rainfall distribution. There is therefore the need to breed and select for drought tolerant genotypes for production especially in the southern Guinea savannah ecology responsible for over 60% of maize production in Nigeria. Controlled experiments using potted plants were therefore conducted during the dry periods between November 2007 and April 2008. The study evaluated moisture deficit tolerance capacity of two maize maturity groups, consisting of 15 extra-early and 12 early genotypes along with two local checks, subjected to two moisture levels 25% (stressed) and 100% (unstressed) soil available moisture) determined gravimetrically. Crop establishment parameters (% germination and mean germination time (MGT), morphological growth parameters (number and area of leaves, plant height, flowering characteristics), physiological growth indices (leaf area index, crop growth rate, relative growth rate, net assimilation rate and leaf area ratio) were measured during growth. Yield components (harvest index, shelling percentage and number of kernels per cob) and grain yield were determined at harvest. The data were analyzed using the general model of ANOVA and significant means were separated by the Least Significance Difference (LSD) at 5% probability level. The results showed that there were no appreciable differences between the two maturity groups for most measured parameters. However, across the two groups, crop establishment parameters, morpho-physiological growth parameters, yield components and grain yield were significantly reduced by soil moisture deficit, while flowering characteristics were significantly delayed by soil moisture stress with significant variable genotypic responses. Grain yield reduction due to water stress was significantly related to drought susceptibility index (DSI) of the genotypes in the two maturity groups. Conclusively, whereas extra-early genotypes showed good yield potentials but poor drought tolerance which suggested poor yield stability and therefore may not be suitable for the southern Guinea savanna (SGS) ecology in the event of severe stress. However, early genotypes, though showed lower yield potential, had good yield stability and hence are promising genotypes for the SGS ecology.


Introduction
Maize (Zea mays L) is a staple food for a vast number of people around the world.Although the crop has its origin in a semi-arid area, is not a reliable crop for cultivation under dry land conditions with limited or erratic rainfall (Arnon, 1972).Maize is the third most important cereal crop after wheat (Triticum aestivum L) and rice (Oryza sativa L) in terms of production in the world (IITA, 2009).Many studies have shown that maize has low drought tolerance capacity due to its high transpiration surface and poor root system.In Nigeria, notwithstanding the efforts of breeders and agronomists to develop high yielding cultivars (Fakorede, et al, 2001;Olakojo and Iken, 2001;Olaoye, et al, 2009) as well as improved production packages (Abayomi, 2004), the yield of maize in farmers' fields throughout the production regions, most especially the savannah ecologies of Nigeria is generally low averaging < 1.5 t ha -1 .This low productivity has been attributed to low soil fertility and drought stress.Drought stress is a major abiotic factor that limits agricultural production (Nemeth, et al 2002;Chaves and Oliveria, 2004;Jaleel, et al, 2009;Golbashy, 2010), more importantly in the rain-fed areas of the world (Nikus, et al., 2004).This is probably due to inhibited cell expansion and reduced biomass production (Ashraf & Mehmood, 1990).Frequent drought stress in the largely rain-fed agricultural system is therefore a major constraint limiting maize production in Nigeria.
One strategy to reduce the effect of water stress on crop yield is to use drought tolerant species and cultivars (Carrow, et al, 1990).This assertion was later supported by Siddique, et al. (2000), who reported that for the purpose of crop production, yield improvement and yield stability under water stress conditions, development of drought tolerant varieties is the best option.Crop plants are usually under stress at one time or another and the plant species able to withstand such stresses have great economic potential (Bibi, et al., 2010).Previous reports on drought tolerance in crops in literature show that variability in genotypic responses to water stress do occur in crops, for example in wheat (Moinuddin, et al. 2005), peanut (Upodhyaya, 2005), barley (Rizza, et al. 2004), soybean (Hufsteler, et al, 2007) and cowpea (Abayomi and Abidoye, 2009).These studies revealed that varieties/ genotypes in each species differed from each other in their responses to water stress conditions, suggesting that drought tolerance in such species may be improved through breeding.
When a large number of genotypes is available for screening against any stress condition, availability of a technique, which could rapidly and efficiently identify the varieties is important (Iqbal, et al, 2010).Although maize is susceptible to water deficit, there is a marked genotypic variations in root density, morphological and physiological characteristics in the crop, while Farhad, et al (2011) also reported that plant adaptation to drought involves both morphological and physiological alterations.It has been observed that water resources for agriculture are decreasing due to increase in demand for irrigation and other non-agricultural water uses (Bacon, 2004).As maize crop requires about 400 to 600 mm of water during its life cycle (Singh, 1991), water availability therefore imposes strong and recurring demand for screening maize genotypes for drought tolerance (Bohnert, et al., 1995).In their continual effort to improve maize production in the West and Central African sub-regions, the International Institute of Tropical Agriculture (IITA) and the West and Central Africa Maize Collaborative Research Network (WECAMAN) under the drought tolerant maize (DTMA) programme released some drought tolerant (DT) genotypes of different maturity groups for evaluation under rain-fed conditions.It was therefore the objective of the present study to screen some extra-early and early maize DT genotypes for moisture deficit tolerance under controlled conditions in a southern Guinea savannah ecology of Nigeria.

Experimental Location, Design and Treatment Application
The study was conducted during the dry periods between November 2007 and April 2008 at the crop pavilion of the Department of Agronomy, Faculty of Agriculture, University of Ilorin (8 o 39' N, 4 o 35'E) in the southern Guinea savannah ecology of Nigeria using potted plants.The study designed as a factorial experiment, evaluated the responses of two maize maturity groups (consisting of 15 extra-early (60-75 days) 12 early (75-90 days) genotypes) obtained from the IITA, Ibadan and two local checks, Afo and DMR-STR-Y (Table 1) to two soil water levels (25% (stressed) and 100% (unstressed) soil available moisture determined gravimetrically (Kramer, 1983).The factorial combinations of the two factors were replicated three times.Ten litre capacity plastic pots perforated at the bottom were filled with 10 kg top soil and thereafter laid out according to the randomization plans.Prior to planting, each pot was moistened with enough water to achieve the desired soil moisture content (25 or 100 % soil available moisture) following the randomization plans.Ten seeds of each genotype, treated with a fungicide, Apron Plus 50 DS (10% Metalaxy, 6% Carboxin and 14% Ferothiocarp) were planted in each pot at a depth of 2.5 cm.The resultant seedlings were later thinned to four per pot at two weeks after planting (WAP).Two of the seedlings were tagged for the collection of non-destructive data (plant height, number of leaves per plant, flowering traits), while the other two plants were harvested at 4 and 6 WAP for the collection of dry matter.Inorganic fertilizer  was applied at a rate equivalent to 80 kg N/ha in two split applications at 2 and 6 WAP.Weed control inside and around the pots were done by regular hand pulling of emerging weeds.
Yield components (harvest index (HI) = grain yield/biological yield; shelling percentage (SP) = grain weight/cob weight x 100 and number of kernels per cob) and grain yield were determined at harvest.To assess the drought tolerance of genotypes, percent yield reduction (PRED) was calculated as PRED = (Yp -Ys) /Yp x 100, while drought susceptibility index (DSI) was estimated using the expression (1 -Ys/Yp)/SI, where Ys = grain yield under stress; Yp = yield at normal soil moisture; SI = stress intensity = 1-mean yield of all genotypes under stress/mean yield of all genotypes at normal soil moisture, as described by Golabadi, et al (2006).

Statistical Analysis
All data were analyzed using the general model of the Analysis of Variance (ANOVA) with Genstat Discovery 3 and significant means were separated by the Least Significant Difference (LSD) at 5% probability level.

Effects Genotypes and Moisture Deficit on Crop Establishment and Morphological Growth Parameters
The effect of maturity group was significant for percent germination under moisture deficit in favour of the extra-early genotypes, while there was no significantly difference in the percent germination of the two groups under normal soil moisture.However, MGT was similar for the two maturity groups under both soil moisture conditions (Table 2).Germination percent was higher with the extra-early genotypes, while MGT was lower with the group.Across the genotypes of the two groups, the effect of soil moisture levels was not significant for the two crop establishment parameters.There were significant variations among genotypes of each maturity group for crop establishment parameters.Among the extra-early genotypes, percent germination was highest with G11 and least with G2, while the highest and the lowest germination percent were obtained with G17 and G26 respectively among the early genotypes.Overall, germination was faster but slightly higher under normal soil moisture condition than under soil moisture deficit.Significant genotype x moisture level observed for the percent germination revealed that while moisture level had no significant effect on germination percent of most genotypes across the two maturity groups, the parameter was significantly reduced in G27 and increased in G28 by soil moisture deficit.3 show that the effect of maturity group was significant for the morphological growth parameters.The number of leaves per plant and leaf area were higher with the extra-early genotypes, while the plant height was higher with the early genotypes.The effect of soil moisture deficit was significant for the plant height and leaf area, but not for the number of leaves per plant.The number of leaves per plant and consequently leaf area were significantly higher in the extra-early than in the early genotypes.However, the extra-early genotypes were significantly shorter than the early genotypes.The number of leaves was not significantly different between the well-watered and the stressed plants, even though the leaf area was significantly higher in the well-watered plants.Plant height was also significantly higher in the well-watered plants than in the stressed plants.The interactive effects of maturity group x soil moisture deficit was significant for the number of leaves per plant and leaf area, but not significant for the plant height.Results in Table 3 also show that the plant height and leaf area were higher with the well-watered than in the stressed plants.

Results in Table
The number of leaves and plant height were not significantly different among genotypes, while the LA varied significantly among them.Leaf area was significantly highest (3232 cm 2 ) with an extra-early genotype, G5, and lowest (1241 cm 2 ) with G24, an early genotype, which value was however not significantly lower than the values obtained for G17, G18, G20, G22 and G25 (Table 3).
Both the tassel and silk emergences were not significantly influenced by the maturity group, but the anthesis and ASI differed significantly between the two groups (Table 4).While the anthesis was significantly earlier in the extra-early group, ASI was significantly smaller with the early group.Table 4 also shows that all flowering traits were significantly delayed by soil moisture deficit.The genotypes in the two groups varied significantly for most flowering traits, except ASI which values were not significantly different among genotypes of the two groups.Significant maturity group x moisture deficit effect observed for anthesis, silk appearance and ASI revealed that anthesis date was significantly longer with the early than the extra-early genotypes in well-watered plants, while there were no appreciable differences between the groups in stressed plants.Silk appearance was significantly delayed by soil moisture deficit in the extra-early genotypes, while the treatment has no significant effect on silk appearance of the early genotypes.Consequently, ASI was significantly increased by moisture deficit in extra-early genotypes, while there were no significant differences between the well-watered and stressed early maize genotypes.Significant moisture level x genotype effect obtained for the LA revealed that while LA was not significantly influenced in 5 genotypes, G10 (extra-early), G18, G22, G23 and G24 (early), the parameter was significantly reduced by soil moisture deficit in all other genotypes (Table 4).

Effects of Maize Genotype and Soil Moisture Deficit on Physiological Growth Indices
The two maturity groups were not significantly different for all physiological growth traits (Table 5), with the exception of LAI which was significantly higher in the extra-early than in the early genotypes (Table 6).However, all indices varied significantly (p<0.001)among genotypes and were significantly (p<0.01)reduced by soil moisture deficit, except RGR which values were similar for all genotypes and was not significantly (p>0.05)influenced by soil moisture (Table 5) and LAR which was significantly increased by moisture deficit (Table 6).Significant maturity group x moisture deficit for CGR showed that while the extra-early had significantly higher CGR value than the early genotypes in well watered condition, the earlier group showed significantly lower CGR than the later under moisture deficit condition (Table 5).Significant moisture level x genotype effect for CGR showed that moisture deficit significantly reduced CGR in most genotypes of the two maturity groups, except in 2 extra-early (G4, G10) and 7 early (G18, G19, G21, G22, G23, G25 and G26) genotypes which were not significantly influenced by soil moisture deficit (Table 5).Similar significant moisture level x genotype effect for LAR revealed that while LAR was significantly increased by soil moisture deficit in 5 extra-early (G2, G3, G5, G9 and G15) genotypes and 3 early (G18, G20 and G28), the parameter was not significantly influenced by soil moisture level in other genotypes (Table 6).

Effects Genotype and Soil Moisture Deficit on Yield Components and Grain Yield
The yield components of genotypes under normal and moisture deficit conditions are presented in Table 7, while grain yields under both moisture conditions, yield reduction due to moisture deficit and drought susceptibility index of genotypes are presented in Table 8.The harvest index was significantly (p<0.05)higher in the extra-early than in the early genotypes, while the shelling percent and number of kernels per cob were not significantly different for the two maturity groups (Table 7).All yield components were significantly reduced by soil moisture deficit across the genotypes of the two maturity groups.The values of HI and SP significantly varied among the genotypes of each maturity group, which however showed similar values for the number of kernels per cob (Table 7).Significant moisture level x genotype effect showed that HI was significantly reduced by moisture deficit in 7 extra-early (G8, G12, G13, 14, G15 and check 1, G16) and 2 early (G20 and check 2, G29) genotypes, while the parameter was not significantly affected by moisture level in other genotypes.Grain yield across moisture levels was insignificantly higher in the extra-early than in the early genotypes (Table 8).Table 8 also shows that grain yield was highest with G10 and lowest with G1 under moisture deficit.However, with adequate soil moisture, grain yield was highest with G8 and still lowest with G1.Significant maturity group x moisture deficit effect obtained for grain yield showed that grain yield was significantly higher in the extra-early than in the early genotypes when plants were well-watered, while there were no significant difference in grain yield of the two groups under soil moisture deficit (Table 8).Among the genotypes, percent grain yield reduction was highest with an extra-early (G8) and lowest with G22, an early genotype (Table 8).The results on percent grain yield reduction were similar to those of drought susceptibility index (DSI) which showed that DSI was higher with extra-early genotypes than in the early genotypes, and highest and lowest with G8 and G22 respectively, hence a significant positive relationships between the two parameters and / or negative relationships between grain yield under stress and drought tolerance index (Figure 1).

Discussion
Speed and quantum of germination are important prerequisites for the success of stand establishment in crop plants.Proper germination of planted seeds, the rate and degree of the subsequent seedling establishment are of great importance in determining both yield and time of maturity (Brigg and Aytenfisu, 1979).However, soil moisture deficit affects seed germination (Abayomi and Wright, 1999;Etejere, 2004), while the effects of moisture stress on seed germination had been shown to be crop species and genotype dependent (Abayomi, 1992).The results of the present study showed that germination percent and speed of germination (MGT) were significantly reduced under moisture deficit condition in all maize genotypes belonging to the two maturity groups evaluated.Significant maturity group x moisture deficit effect showed no significant differences in germination percent of the two maturity groups when adequately watered, while the extra-early genotypes germinated significantly better than the early genotypes when they were water-stressed.These results are in line with the early reports (Smith and Hoveland, 1986;Abayomi and Saliu, 1997;Abayomi and Wright, 1999) which showed that water stress affected germination by delaying its onset, slowing its rate and by decreasing the final germination percentage.The results also showed significant differences in germination between the two maturity groups and within the genotypes in consonance with the reports of Ashraf and Abu-Shakra (1978) who showed that the ability of seeds to germinate at low soil moisture was dependent on crop species and genotypes.The speed of germination was also significantly reduced (high MGT) by soil moisture deficit in the present study.This was similar to the reports of earlier workers (Abayomi and Mobolaji, 1995;Abayomi and Saliu, 1997;Abayomi and Wright, 1999;Smith, et al., 1989) who showed significant increase in MGT due to low available soil moisture.The decrease in germination rate as shown by high MGT with soil moisture deficit observed in the present study may be due to reduced water uptake by the germinating seeds (Abayomi, 1992).The overall results of germination tolerance of soil moisture stress show that crop establishment was better with extra-early than the early genotypes.Among the genotypes, G11 (extra-early) showed the best establishment, while G29 (local check 2) show the least in terms of germination percentage and speed of germination (MGT).
Morphological growth characters of number of leaves per plant, leaf area and plant height were significantly reduced by soil moisture deficit.The reduction in leaf area as a result of water deficit is attributable to decreased rate of leaf initiation and expansion / and or increased rate of leaf senescence and leaf shedding (Legg, et al., 1979).Nesmith and Rochie (1992) and Fortis and Edward (1995) have reported that water deficit reduced leaf development and decreased leaf area expansion.The importance of the number of leaves to grain yield has been shown by Benti and Ranson (1993) who reported that grain yield is directly proportional to the number of leaves removed.The decrease in plant height by soil moisture deficit as observed in the study was in line with the report of Gavloski, et al. (1992) who observed that there was a decrease in maize plant height following the withdrawal of water from one or more section of the root system.The importance of plant height as a yield predictor has been shown by Abayomi (1992) and Hadjichristodoulou (1987) had earlier suggested that variations in plant height may become one of the causes of variation in crop yield.This suggests that genotype that can maintain good plant height under stress are likely to give good grain yield in line with the results of this study which showed that when water stress reduced plant height by 37.4%, grain yield was reduced by 59.28% in the extra-early genotypes.However, in the early genotypes, a reduction of 24.06% in plant height by moisture deficit resulted in a decrease of only 39.46% grain yield.Among the genotypes, two early genotypes (G21 and G24) showed good heights and hence good grain yields under stress.
The results of the effect of water stress on flowering traits showed that inadequate soil moisture significantly delayed tassel and silk appearance, anthesis and hence increased ASI.Flowering traits in maize have been shown to be significantly delayed under severe drought (Edmeades, et al., 1997).Similarly, Herrero and Johnson (1981) also reported that water stress during the reproductive period can increase the interval from silking to pollen shed which shortened the grain filling period.The results of this study were also in line with the observation of Mckenzie (2006) who reported that water stress during flowering causes several changes in plant development including delay in silk emergence and pollen shed, resulting in synchronous pollination, decreased kernel weight, delay in physiological maturity and a decrease in final yield.Edmeades, et al. (2000) also reported that water stress slows ear growth in relation to tassel growth resulting in an increase in anthesis to silking interval (ASI).
The results of the present study also show that most physiological growth indices measured were significantly reduced by inadequate soil moisture content across the genotypes.The reduction in net assimilation rate (NAR) was in agreement with the report of Boyle, et al (1991) who showed that low seed yield at low water potential can be accounted for by lack of assimilate supply at flowering.The results also showed that soil moisture stress reduced leaf area index (LAI), suggesting reduction in photosynthesis.It has been observed that photosynthesis can be reduced by water stress through reduction in leaf area, stomatal closure and decrease in the efficiency of carbon fixation process (Ludlow, 1975).Soil moisture deficit reduced all yield components and grain yield in maize genotypes across the two maturity groups.Girma, et al (2005) have reported reduction in HI as a result of moisture stress.Other workers have also reported reduction in yield components and grain yield due to water stress.O'Neil, et al (2004) reported that water stress during grain fill period caused decrease in kernel weight, earlier physiological maturity and decrease in final yield.Percent grain yield reduction due to soil moisture deficit was observed to be positively related to drought susceptibility index (Figure 1) in line with the report of Rizza, et al. (2004) who showed significant negative relationship between grain yield and water stress index of barley (Hordeum vulgare, L) genotypes.A lower grain yield reduction under stress suggests a higher yield stability which is more important than breeding and selection for absolute high yield under stress.
According to Roland (1993), drought resistant genotypes are known for their higher yields under conditions of limited and uncertain rainfall.The results of the present study show that genotypes G2, G9, G10, G21 and G25 had higher grain yields under stress condition, suggesting that they are more water stress tolerant than the other genotypes.The focus of varietal response evaluation to environmental stress is to identify superior genotypes that perform relatively well under stress and could be used to replace the existing genotypes and / or as sources of genes for the production of inbred lines aimed at the development of stress tolerant genotypes (Olaoye, 2009).The above identified genotypes are therefore promising for the southern Guinea savannah and could replace the two local checks, Afo (G2, G9, and G10) and DMR-SRT-Y (G21 and G25), as well as being sources of genes for breeding programme for drought tolerant maize cultivars in the zone.However, it has been suggested that in an environment prone to severe drought conditions like the Guinea savannah ecology (SGS), the most suitable genotypes should maintain high yield under both favourable and stress conditions (Rizza, et al., 2004).Considering the results of the 29 genotypes evaluated in the present study, an extra-early genotype, G10, ranked best under stress and no stress conditions and may therefore be the best genotype for the SGS ecology.Although reduced with inadequate soil moisture, its yield was superior under both conditions.The results further show that five genotypes (G8, G11, G3 and G5, all extra-early genotypes), had higher yield potentials under favourable soil moisture condition, but were highly susceptible to drought stress.There is some agreement that a high yield potential is advantageous under moderate stress, while high drought tolerance may be more useful under severe stress (Voltas, et al, 1999;Panthywan, et al, 2002)., thereby suggesting that these five genotypes may not be suitable for the SGS ecology in the event of severe stress.However, three early genotypes (G22, G23 and G19), with lower yield potentials under favourable moisture condition, showed lower yield reductions under stress (Table 8), thereby suggesting good yield stabilities and are therefore promising candidates for the ecology.An extra-early genotype, G1, showed poor yield potential with adequate soil moisture and dismally low drought tolerance, is neither good for cultivation nor as a source of genes for drought tolerance improvement programme for the ecology.
Conclusively, two extra-early genotypes (G2 and G10) and two early genotypes (G21 and G25), which showed both high yield potentials under favourable moisture condition and good drought tolerance under moisture deficit, are promising as DT genotypes for the SGS ecology and can therefore replace Afo and DMR-SRT-Y (local checks), and as well serve as sources of genes for the improvement of three extra-early genotypes (G8, G11 and G16) with high yield potentials under favourable moisture condition, but low stability under moisture deficit; and three early genotypes (G19, G22 and G23) which look promising as DT genotypes, but showed low yield potentials under favourable moisture condition.Hybridization of these identified ten genotypes may result in the production of better DT genotypes well suited to the SGS ecology.
Table 1.The evaluated extra-early, early and local maize genotypes for responses to soil moisture deficit at Ilorin, southern Guinea savanna, Nigeria Table 2. Crop establishment of some maize genotypes under moisture deficit and irrigated conditions Table 3. Morphological growth characters of some maize genotypes under moisture deficit and well watered conditions Table 4. Tassel and silk appearances, anthesis and anthesis-silking interval (ASI) of maize genotypes under moisture deficit (STR) and well-watered (WWT) conditions Table 5.Net assimilation rate, crop growth rate and relative growth rate of maize genotype under moisture deficit (STR) and well-watered (WWT) conditions Table 6.Leaf area ratio and leaf area index of maize genotype under moisture deficit (STR) and well-watered (WWT) conditions Table 7. Harvest index, shelling percentage and number of kernels of maize genotypes under moisture deficit (STR) and well-watered (WWT) conditions Table 8.Grain yields under soil moisture deficit (Ys), normal soil moisture (Yp), yield reduction percent due to moisture deficit (PRED) and Drought Susceptibility Index (DSI) of maize genotypes of two maturity groups

Figure 1a .Figure 1b .
Figure 1a.Relationship between percent reduction in grain yield due to soil moisture deficit and drought susceptibility index in maize