MATERIALS AND METHODS

The field experiments were conducted during 1997-1998 and 1998-99 at the Agricultural Research Institute of Maragheh, Iran (latitude 37°15’ N, longitude 46°22’ E and 1720 m above mean sea level). The maximum and minimum temperature averaged as 36.7 and 4.8°C, respectively. The field soil was loamy-clay. During the crop growing season of 1997-98, cumulative pan evaporation and rainfall were 1223 and 346.3 mm, respectively. During 1998-1999 these factors were 1133 and 201.7 mm, respectively (Fig. 1).

Wheat was sown during the early days of the Oct. of two cropping years on six rows 5.5 m long and 20 cm apart (5.5x1.2 m² plots). Sixteen treatments were laid out in completely randomized blocks with a split-split plots design with four replications. Treatment combinations comprised 2 cropping years, 2 levels of tillage methods and 4 levels for seeding rates of wheat. Tillage methods including TM1 = without mold board and TM2 = minimum tillage with plot size of 30x5.5 m², tillage practices were done by a tractor with power of 65 hp. Tillage practices were done in the fall of year whose land was fallow.

Fig. 1:	Occurred rainfall (mm) during two cropping years. cumulative rainfall was 346.3 and 201.70 mm in the two cropping years

Seeding rates treatments including SR1 = 100, SR2 = 125, SR3 = 150 and SR4 = 175 kg ha^-1 with plot size of 7x5.5 m². N up to 30 kg ha^-1 in the form of urea was applied to the soil in the sowing time and spring. P up to 60 kg ha^-1 was applied to the soil in the sowing time. The crop was managed well with respect to weed and disease control. Plant height was measured by 15 plant and spike length, number of grain in the spike and 1000 grain weight was measured by 15 spikes at each plot. After maturity for border effect, 50 cm from the both sides of plots was omitted, therefore plots in 4.5x1.2 m² were harvested for grain. The collected data during the experiment was analyzed according to RCB design and mean values compared with Least Significant Differences (LSD) test (Steel and Torrie, 1980).

RESULTS AND DISCUSSION

The grain yield is significantly affected by year at 5% level of probability. The acquired grain yield in the 1st and 2nd years averaged as 1755.313 and 642.594 kg ha^-1. It seems the difference was originated from rainfall that occurred as 346.3 and 201.70 mm in the consecutive two years. Analysis of data (Table 1) revealed that interaction of tillage method and seed rate had a significant (p = 0.07) effect on grain yield. The grain yield resulted from TM2xSR3 (=1329 kg ha^-1) and TM1xSR2 (= 1291 kg ha^-1) had the fist position relative other yields.

Maximum and minimum yield acquired from TM2xSR3 (=1329 kg ha^-1) and TM2xSR2 (=1097 kg ha^-1). This finding is differ with those find by Rajput et al. (1989) and Bhatnager et al. (1991) that have reported maximum grain yield is obtained with increase in seed rate. The obtained yield from TM1xSR3, TM1xSR4 and TM2xSR4 and on the other hand TM1xSR1, TM2xSR1 and TM2xSR2 are in similar groups (Fig. 2). For acquiring the maximum yield minimum tillage with seed rate of 150 kg ha^-1 or without mold board operation with seed rate of 125 kg ha^-1 can be recommended.

Table 1:	Mean squares and significance levels from the analysis of variance on grain yield and some yield components of dryland winter wheat
**,*, + and ns: Significantly different at 1, 5 and 10% and nonsignificantly different at 10% level of probability, respectively

Fig. 2:	Mean value of grain yield (kg ha-1) versus applied seed rate (kg ha-1) and soil tillage methods

A Weibull model were worked out for measured mean value of grain yield (kg ha^-1) versus applied seed rate (kg ha^-1) with applied soil tillage methods of TM1 and TM2, by multiple regression and the least squares procedure (Moghaddam, 1999). Acquired-models were as follow:

where GY, SR and TM are grain yield (kg ha^-1), applied seed rate (kg ha^-1) with soil tillage methods. TM1 and TM2 are treatments of without mold board and minimum tillage, respectively.

The R² of models indicating that fitted models explain all of the variability in the grain yields under conditions of conducted experiments in two cropping years. The grain yield from developed models compared with measured values and displayed in Fig. 3. The agreements were satisfactory. Bhatnager et al. (1991) reported that grain yield increased with the in crease in the seed rate from 100 to 200 kg ha^-1. This research finding is agreed with those find by Bhatnager et al. (1991).

The effect of tillage method and interaction of year with seed rate on 1000 grain weight is statistically significant at 5 and 10% level, respectively. Maximum and minimum 1000 grain weight acquired from plots applied Y1xSR1 (=37.88 g) and Y2xSR1 (= 27.45 g) treatments. The obtained 1000 grain weight from applying Y1xSR2 and Y2xSR3 and on the other hand SR1 to SR4 in the 2nd year are in similar groups (Fig. 4). This finding is consistent with those reported by Khan et al. (2002).

Analysis of data (Table 1) revealed that spike length was significantly (p = 0.05) affected by interaction of year and seed rate, while non significant effect was observed due tillage methods.

Fig. 3:	Variation of mean value of grain yield (GY in kg ha-1) versus applied seed rate (SR in kg ha-1) with soil tillage methods. TM1 and TM2 are treatments of without mold board and minimum tillage, respectively

Fig. 4:	Mean value of 1000 grain weight (g) versus applied seed rate during cropping years of 1997-98 and 1998-99

Mean values indicated that maximum spike length were taken by those plots which ware sown in the 1997-98 year at the rate of 100 kg ha^-1 while minimum spike length were recorded in plots that sown at the 175 kg ha^-1 in the 2nd year. In general, plots sown in the 2nd year that grouped in similar class produced low spike length relative plots sown in the 1st year.

Fig. 5:	Mean value of spike length (cm) versus applied seed rate (kg ha-1) during cropping years of 1997-98 and 1998-99

Fig. 6:	Mean value of grains per spike versus applied seed rate (kg ha-1) during cropping years of 1997-98 and 1998-99

Fig. 7:	Mean value of spike m-2 versus applied seed rate (kg ha-1) during cropping years of 1997-98 and 1998-99

Spike lengths produced by plots sown at the rate of 125 and 150 kg ha^-1 are in similar groups (Fig. 5)

Table 1 demonstrated that interaction of year and seed rate had a significant (p≤0.05) effect on grains spike^-1, while non significant effect was observed due tillage methods. Mean values indicated that, maximum grains spike^-1 (= 21.422) were taken by those plots which ware sown in the 1997-98 year at the rate of 100 kg ha^-1 while minimum grains spike^-1 (= 7.296) were recorded in plots that sown at the 175 kg ha^-1 in the 2nd year (Fig. 6). Grains spike^-1 produced by plots sown in the 1^st year at the rate of 125, 150 and 175 kg ha^-1 are in similar groups. Also, grains spike^-1 produced by plots sown in the 2nd year at the rate of 100 and 150 kg ha^-1 are in similar groups. This result is consistent with those find by Khan et al. (2002).

Analysis of data (Table 1) revealed that interaction of year and seed rate had a significant (p≤0.05) effect on spike m^-2, while non significant effect was observed due tillage method. Mean values indicated that, maximum spike m^-2 (= 507.6) were taken by those plots which were taken by those plots which ware sown in the 1998-99 year at the rate of 175 kg ha^-1 while minimum grains spike^-1 (= 301.00) were recorded in plots that sown at the 100 kg ha^-1 in the 2nd year. Spike m^-2 produced by plots sown in the two consecutive years at the rate of 100 and 125 kg ha^-1 are in similar groups (Fig. 7).

Analysis of data (Table 1) indicated that cropping years had a significant effect on plant height, while non significant effect was observed due tillage methods, seed rate and its interaction. Khan et al. (2002) found that seed rate had significant effect on plant height. The finding in the present study is differing with those reported by Khan et al. (2002). Mean values indicated that, maximum and minimum plant height were as 71.144 and 42.094 cm taken by those plots which were sown in the 1997-98 and 1998-99.

CONCLUSION

The influence of tillage methods and seed rates on grain yield and some yield components of dryland winter wheat were studied. The present study supports the following conclusions:

ACKNOWLEDGMENTS

We thank the Iranian Agricultural Engineering Research Institute (IAERI) and Agricultural Research Institute of Maragheh for supporting this research.