Climatic conditions requirements of maize germplasm for flowering in the rainforest Agroecology of Nigeria

The number of days from planting to flowering in maize (Zea mays L.) is of interest to maize breeders because of its importance in the selection of appropriate hybrid parents. Highly significant interaction of planting dates with varieties for flowering traits have been observed during the early and late cropping seasons in the rainforest agro-ecology of South West Nigeria. This makes the use of flowering dates as indicators of maturity unreliable. Therefore the objectives of this study were to evaluate 100 maize varieties for flowering traits and determine the climatic factors influencing the interaction of the environments with days to flowering (GxE) in the rainforest ecology of South West Nigeria. One hundred maize varieties were evaluated during the late and early cropping seasons of 2007/2008 and 2008/2009. Significant differences were observed among the varieties for flowering traits (days to 50% tasseling, anthesis and silking). There was also significant variety x season interaction mean squares. In the early season, TZEE-WSRBC5, TZEEPOPSTRCo and 97TZEE-Y-2C1 with 47-53 days to full flowering were the earliest to flower while Oba-Super II and ACR96DMR-LSR W with 64-71 days to flowering were the latest to flower. In the late seasons, 2004TZEE-WPOPSTRC4, TZEEPOPSTRCo, SINETEE-WSR and TZE-WPOPDTSTRC4F2 were the earliest to flower (42-47 days) while BUSOLA STR, TZLCOMPCO, 9021-18STR and Oba super II were the latest (61-68 days). Flowering interval was shorter in the late than the early season regardless of the maturity group with temperature as main climatic factor influencing flowering in this ecology.


INTRODUCTION
Maize (Zea mays L.) is the second most important food crop, behind cassava (Manihot esculenta Crantz) in Africa, and is grown in a wide range of environments ranging from Niger's northern Sahel to Ethiopia's highlands and the converted forest lands of Sierra Leone.The popularity of maize among African farmers grew slowly until the early part of the 20th century after which it has increased rapidly (African Crops, 2000).The reproductive stages of maize are based on appearance of the male flower (tassel producing anthers), female flower (silk) and developmental changes in the kernels.Therefore, time to flowering commonly used in research include days to 50% tasseling, anthesis and silking.The number of days from planting to flowering is a trait of interest to maize breeders because of its importance in selecting appropriate hybrid parents and for its role in the utilization of unadapted germplasm (Koester et al., 1993).
The environments found in the tropical and sub tropical locations are extremely diverse (Whiteman, 1985) with highly variable effects on flowering in maize.. Bonhomme et al. (1994) reported that anthesis rather than silking was more stable and reliable among the flowering traits because of more pronounced environmental stress effects on silking.Oluwaranti et al. (2008) observed highly significant interaction of planting dates with varieties for flowering traits in maize during the early and late cropping seasons at Ile-Ife, a typical tropical rainforest location in Nigeria.
Temperature is one of the most important factors that determine plant growth, development, and yield.Accurate summary of plant temperature response in plants is a prerequisite to successful crop systems modelling and application of such models to crop management (Wiekai and Hunt, 1999).Thermal models involving a framework of three cardinal temperatures were used to study crop development with particular reference to maize (Xinyou et al., 1995).The cardinal temperatures were the base (T b ), the optimum (T o ) and the ceiling (T c ). Maize is a warm weather crop and is not grown in areas where the mean daily temperature is less than 19°C or where the mean of the summer months is less than 23°C (Jean, 2009).Dauda (1992) reported that Crop Heat Unit (CHU) closely predicted days to silking than other flowering traits in a study conducted at Obafemi Awolwo University Teaching and Research Farm.
Photoperiodicity has been defined as physiological reaction of organisms to the length of day or night and was found to occur in plants and animals (Photoperiodism, 2009).Photoperiodic flowering plants are classified as long-day plants or short day plants, although the regulatory mechanism is actually governed by the hours of darkness, not the length of the day.A long-day plant requires fewer than a certain number of hours of darkness in each 24 h period to induce flowering while a short day plant flowers when the night is longer than a critical length.There are also day-neutral plants that do not initiate flowering based on photoperiodism at all, they flower regardless of the length of the night.The day-neutral plants may initiate flowering after attaining a certain overall developmental stage or age, or in response to alternative environmental stimuli, such as vernalization (a period of low temperature), rather than in response to photoperiod (Photoperiodism, 2009).Russell and Stuber (1983) observed that in maize photoperiod effect on days to tassel initiation (DTI) and total leaf number (TLN) was considerably greater than that of temperature.Photoperiod x temperature interactions were significant only for DTI.Total leaf number (TLN) was concluded to be better than DTI for measuring effects of photoperiod on the duration of vegetative growth in maize.
The role of photoperiod x temperature interaction in controlling flowering processes has been one of the key research problems in the understanding of the growth of some crops such as maize and soybean [Glycine max (L).Merr] (Jean, 2009;Cregan and Hartwig, 1984;Egli et al., 1989;Mcblain et al., 1987;Sinclair et al., 1991;Wang et al., 1987).Therefore the objectives of this study were to evaluate 100 maize varieties for flowering traits and determine the climatic factors influencing the interaction of the environments with days to flowering (GxE) in the rainforest ecology of South West Nigeria.

MATERIALS AND METHODS
One hundred maize varieties of different maturity groups were evaluated for flowering traits (days to 50% tasseling, anthesis and silking) at the Teaching and Research Farm (TRF) of Obafemi Awolowo University (7°28'N 4 o 33'E and 244 m above sea level), Ile-Ife during late and early cropping seasons of 2007/2008 and 2008/2009.These maize varieties were supplied by the Maize Breeding Programme of IITA, Ibadan, Nigeria.The climatic data used for this study were supplied by Nigerian Meteorological Agency (NIMET), Oshodi Lagos and the Geo-Spatial Laboratory of IITA, Ibadan.
In all plantings, a tripple lattice design with three replications and ten incomplete blocks in each replication was used.Plots were two rows, 5 m long each, with intra-and inter-row spacings of 0.5 and 0.75 m, respectively.Each plot was bordered by a common variety.Three seeds were planted per hill and the plants were thinned to two at the three-leaf stage, giving a population of 53,333 plants/ha.Seeds were treated with Apron plus prior to planting to control damage by soil-borne diseases and insect pests.Apron plus, a systemic fungicide and insecticide also controls downy mildew which is endemic at the site.Ploughing and harrowing were done before laying out of the experimental field.Fertilizer NPK was applied at a total rate of 180 kg N, 90 kg P2O5 and 90 kg K2O/ha in two splits; first at three weeks after planting and finally at five weeks after planting.Weeds were controlled with primextra, which contained atrazine (2-chloro-4-(ethyl amino)-6-isopropylamino-striazine) and alachlor (N-(1-methyl-2-methoxy-ethyl)-2-ethyl-8methyl-chloroacetanilide) as active ingredients.The herbicide was applied the day after planting at the rate of 5 L/ha.Weeds were also controlled by hand weeding as necessary after the crop had established.Dates when 50% of the plants in a plot attained tasseling (TS), anthesis (ANTH) and incipient silk extrusion (SILK) were recorded and expressed as days after planting (DAP).
Analysis of Variance (ANOVA) using the General Linear Model (GLM) procedures of SAS version 9.2 (SAS Institute, 2003) was computed for each flowering trait recorded in each trial and combined trials for all cropping seasons.Means of the days to flowering traits for the varieties were separated with Least Significant Difference at 0.05 level of probability.Furthermore, correlation between each flowering trait and climatic factors (temperature, photoperiod and heat units) was carried out for all the trials and the combined trials.This was followed by simple linear, multiple linear and quadratic regressions of mean values the varieties (N=100) for each trait on temperature, heat units and

RESULTS
In the ANOVA combined across four environments, significant (p=0.01)mean squares were obtained for environment, variety, replication, block within replication and environment x variety interaction (Table 1).The CV was about 4 to 4.5% while the R 2 was about 88 to 89% for each flowering trait.Means of days to silking for five earliest and latest varieties in this study are presented in Table 2.The five earliest varieties in 2008 early cropping season took between 47 and 50 days after planting to full flowering while it took 50 to 54 days after planting for the earliest varieties to full flowering in 2009 early cropping season.Varieties TZEE-WSRBC 5 was the earliest to silk with 47 days after planting in 2008 early season while TZEEPOPSTRC 0 was the earliest to show silk extrusion in 2009 early season.TZEE-WSRBC 5 , TZEEPOPSTRC 0 and 97TZEE-Y-2C 1 were observed to be among 5 earliest varieties to full flowering in both 2008 and 2009 early cropping seasons.Similarly, in the late seasons of 2007 and 2008, TZEEPOPSTRC 0, 97TZEE-Y-2C1 and TZEE-WSRBC 5 were also observed to be among the five earliest to silk with 45-57 days after planting.TZPB-SRPROL.F1C3, BR9943-DMRSR and TZLCOMP.1C6 were among the latest varieties to silk with 66 -71 days after planting in 2008 and 2009 early seasons while in the late seasons of 2007 and 2008, Oba Super II, TZB-SR and MASYN VAR3F2 were among the latest to silk with 64 -71 days after planting (Table 2).

Correlation analysis
Highly significant positive correlation coefficients (rvalues) were observed between all the flowering traits and the climatic factors (temperature, photoperiod and heat units) in the individual trials except for the correlation between photoperiod and days to 50% tasseling which was negative but not significant in the 2007 late season trial (Table 3).The correlation between days to silking and heat units was also negative but highly significant in this trial.Therefore, increases in the values of these climatic factors delayed the expression of each of the flowering traits.

Regression analysis
In 2007 late season trial, the regression analyses of days to flowering on the climatic factors for the 100 varieties consistently indicated that the variations observed in days to flowering were explained by the variation in temperature (Table 4).For the three types of regression models performed for each flowering event, the coefficient of determination (R 2 ) for temperature per se or temperature plus photoperiod were above 90% (Table 4).However, there were no increases in the R 2 -value due to the addition of photoperiod in the multiple regression models, indicating that the effect of photoperiod on flowering was negligible during this season.Similar trends were observed during the 2009 early cropping season (Tables 5).Although the other two climatic variables appeared to be as important as the temperature itself, the changes in R 2 values were not consistent.Addition of photoperiod in the multiple regression or quadratic term had little or no improvement for the fit of the temperature model.

DISCUSSION
There were significant differences among the hundred varieties of maize evaluated for flowering traits in this study.These results were expected because of the following reasons.First, the 100 varieties were of different maturity classes; extra-early, early, intermediate and late maturity.Second, they were bred for different purposes; tolerance to diseases such as downy mildew, streak, blight, and others, resistance to insect pests such as stem borer.Third, they were developed at different locations as indicated by their names.Four, the 100 varieties have different genetic backgrounds.Significant differences were also observed among the environments (seasons), where the varieties were evaluated in the early and late cropping seasons of this study.These results were also expected because the seasons are characterized by different temperature, sunshine hour, heat units received.The high coefficient of determination (R 2 values) obtained for all the flowering traits, in the combined trials is an indication of the reliability of the models.Likewise, low CVs obtained from this study showed that the observed days to flowering is quite stable and reliable.
Highly significant interactions of environments with varieties observed for the flowering traits were consistent with the findings of Oluwaranti et al. (2008), and this was the justification for this study.In other words, the use of flowering dates as indicators of maturity may not be reliable because of significant and relatively large GxE interaction.The results obtained for days to flowering in early and late cropping seasons of this study also corroborate with those of Fakorede (1985) in a study on the response of maize to planting dates at the Teaching and Research Farm of Obafemi Awolowo University.As part of a study carried out in 2001 late cropping season, days to 50% tasseling, anthesis and silking decreased from early planting to later plantings and increased thereafter because the amount of rainfall available for the crop decreased as planting was delayed (Oluwaranti et al., 2008).Likewise from the study, in 2002 the maize varieties flowered faster on planting dates that received larger amounts of rainfall than those which received smaller amounts.It was, therefore, concluded from the study that the higher the amounts of rainfall, the fewer the   Oluwaranti et al. (2008), however did not consider other climatic factors and therefore did not report the effect of temperature on flowering.Studies by Fakorede (1985) and Fakorede and Opeke (1985) examined the effects of several climatic factors on grain yield of maize in this location but did not include flowering traits.Therefore, the findings in the present study on the GxE interaction of flowering dates is important and warrant the analysis of climatic factors responsible for the interactions.

Effects of temperature, heat units and photoperiod on flowering traits of maize
Correlation analysis indicated that temperature, photoperiod and heat units are the major climatic factors affecting days to flowering in maize at this location.Days to flowering increased as temperature, heat units and photoperiod increased and decreased as these climatic factors decreased.Theoretically, delayed planting in the early season should reduce the number of days to flowering while in the late season, it should delay flowering.The trend observed in this study, however, showed that increased temperature, delayed the progress towards flowering thereby resulting in increased number of days to flowering.Regression models also indicated that photoperiod and heat units influenced flowering when considered singly.This is in corroboration with the findings of Shaykewich (1995) who observed that photoperiod and temperature had major effects on days to flowering in cereals and could be important sources of genotype × environment interaction.The combination of these variables with temperature was not important in determining flowering date of the maize varieties in the present study.Temperature was the main climatic factor contributing to the variation observed in the days to flowering in this study as addition of photoperiod and or heat unit did not increase this observed variation as shown by their R 2 values.

Conclusion
It was concluded that development rate from sowing to flowering (days to 50% tasseling, anthesis and silking) was affected mainly by temperature in the rainforest agro-ecology since use of photoperiod and accumulated heat units as additional parameters in predictive models did not improve goodness of fit of the models.

Table 1 .
Mean squares from the analysis of variance for the flowering traits of 100 maize varieties evaluated at the Teaching and Research Farm of Obafemi Awolowo University, Ile-Ife, during the late and early cropping seasons of2007/ 2008 and 2008/2009.

Table 2 .
Means of days to silking of 5 earliest and latest varieties evaluated at the Teaching and Research Farm of Obafemi Awolowo University, Ile-Ife, during the early and late seasons of 2008/2009 and 2007/2008 respectively.

Table 3 .
Pearson correlation coefficients (r-values)between the climatic factors and flowering traits for the individual and combined seasons of 2007 to 2009 at the Obafemi Awolowo University Teaching and Research Farm, Ile-Ife.
** significant at 0.01 level of probability.

Table 4 .
Simple linear, multiple linear and quadratic regression analyses of days to tasseling, anthesis and silking on temperature, photoperiod and heatunits of 100 maize varieties evaluated during the 2007 late cropping season at the Obafemi Awolowo University Teaching and Research Farm, Ile-Ife.

Table 5 .
Simple Linear, multiple linear and quadratic regression analyses of days to tasseling, Anthesis and silking on temperature, photoperiod and heatunits of 100 maize varieties evaluated during the 2009 Early cropping season at the Obafemi Awolowo University Teaching and Research Farm Ile-Ife.