|
|
||||||||
Centre UdL-IRTA, Rovira Roure 191, 25198 Lleida, Spain
* Corresponding author (jaume.lloveras{at}irta.es)
Received for publication November 19, 2002.
| ABSTRACT |
|---|
|
|
|---|
Abbreviations: TKW, one thousand-kernel weight
| INTRODUCTION |
|---|
|
|
|---|
In Europe, most published research comes from the north-central part of the continent. This area is characterized by long growing seasons and long and mild grain-filling periods. In Belgium and northern France, the optimum spike densities for wheat are 475 to 500 spikes m2 and are normally obtained with a target density of 200 plants m2 at the end of winter (Laloux et al., 1980; Gate, 1995). In northern Ireland, the highest grain yields were achieved with seeding rates of between 50 and 100 seeds m2 (Easson et al., 1993). A review of trials conducted in the UK by Gooding and Davies (1997) showed a normal seeding rate of 60 kg ha1 in September, much lower than the conventional recommendations for some Mediterranean areas (Lopéz Bellido, 1991).
According to Paulsen (1987), the recommended seeding rates ranged from 67 seeds m2 in the dryland plains to 400 seeds m2 in the eastern regions of the North America, with 200 seeds m2 as the most widely recommended rate in many areas of the USA. This basic rate can increased by 50% for irrigated conditions.
Most research on population density effects on crop yield shows increases up to a plateau value at moderate densities and a significant reduction in production only at very high densities (Holliday, 1960; Donald, 1963). Puckridge and Donald (1967) reported that grain yield was relatively unaffected by seeding rates above 33 seeds m2 in a study conducted in a southern hemisphere Mediterranean environment, in Australia. Yields fell only 25% at the exceptionally high rate of 1078 seeds m2.
Optimum seeding rate is strongly influenced by sowing date, which in turn is largely governed by the climate and the requirements of a rotation. In general, the higher the latitude, the cooler the summer temperatures, the longer the cropping period, and the earlier the drilling (Gooding and Davies, 1997). Late seeding dates normally result in higher seeding rates because a delay in sowing normally reduces individual plant growth and tiller production (Gooding and Davies, 1997; Satorre, 1999).
Past research shows a variation in the responses of different wheat varieties to density (Faris and De Pauw, 1981; Gate, 1995; Couvreur et al., 1999; Wiersma, 2002). Varieties have different abilities or plasticities to compensate for low or high plant populations by modifying the number of tillers and consequently the number of spikes per square meter, the number of kernels per spike, or the grain weight. Recently developed hybrid wheat may require lower seeding rates than the traditional varieties (Couvreur et al., 1999).
As discussed above, there has been a significant amount of research concerning the effects of seeding rates on wheat in central Europe and in the USA and Canada. However, little has been published about the relationships between grain yields and optimum seeding rates of Mediterranean type of wheat in irrigated Mediterranean regions (Acevedo et al., 1999). These areas are characterized by dry, hot summers alternating with temperate winters and high year-to-year variation in rainfall (Acevedo et al., 1999). Mediterranean wheat is characterized by strong photoperiod sensitivity or by slight vernalization requirements (Slafer and Whitechurch, 2001) and has a faster rate of vegetative development, fewer leaves, and hence fewer tillers (Loss and Siddique, 1994). Moreover, the moderate winter temperatures may accelerate each stage of development and affect seeding rates. The objective of this study was to determine the optimum seeding rates and their relationships with crop yield of Mediterranean wheat in irrigated Mediterranean cropping systems.
| MATERIALS AND METHODS |
|---|
|
|
|---|
|
The experiments were seeded on 24 November and 1 December 1999 at Site 2 and Site 1, respectively, and on 21 November (Site 1) and 21 December (Site 2) in 2000. Mean temperatures and rainfall for the 19992000 and 20002001 wheat growing seasons (NovemberJuly), and long-term averages are presented in Table 2. Crops were irrigated three times in the spring, every year. A total of 150 mm of water was applied between 14 February and 1 March, 9 and 14 April, and 2 and 5 May, according to the irrigation turns of the area.
|
Four wheat varieties were seeded: Gazul and Rinconada hard red spring wheat; Anza, a traditional soft spring wheat of CIMMYT origin (López and Serra, 1996; AETC, 1999); and Balsamina, a hybrid wheat of Spanish origin. All these wheat varieties are of Mediterranean types. Anza, with an average height of 75 cm, was the shortest variety and Balsamina (95 cm) the tallest. Seed lot viability was high (95%). Mean seed weight varied between 36.4 g per thousand-kernel weight (TKW) for Anza in the year 2000 to 50.8 g for Balsamina in 1999. Seed sowing rates were 10% higher than the target densities.
The initial number of plants per unit area was estimated by counting plants along 50-cm sections of three rows in each plot when the plants were between the one- and two-leaf stage. The number of spikes per unit area was estimated by counting spikes just before harvest along 50-cm sections of three rows in each plot. Fifteen consecutive spikes per plot were harvested for the determination of grains per spike and TKW. Three measurements of plant height were taken from each plot to the base of the spike. Lodging was evaluated visually, using a 0 to 10 scale, with a value of zero when there was no lodging and 10 when the crop was 100% lodged.
Grain was harvested in Gimenells on 27 and 29 June in 2000 and 2001, respectively, and on 4 July in both years in Palau. In all trials, the plots were harvested using a 1.5-m-wide Nurserymaster Elite (Wintersteiger, Ried, Austria) plot combine. The grain moisture level was measured (GAC II, Dickey-John, Auburn, IL, USA) in a 300-g sample from each plot and grain yield adjusted to 140 g kg1 moisture.
The trials were located in a different area of their respective fields in each growing season, and the treatments were randomized every year. The experimental design was a completely randomized block with four replications.
The results were subjected to analysis of variance considering four environments (2 sites x 2 yr). Site 1 in 2000 and Site 1 in 2001 were Environments 1 and 2, respectively, whereas Site 2 in 2000 and Site 2 in 2001 were Environments 3 and 4, respectively. Single degree-of-freedom linear and quadratic contrasts were used to analyze the density effects. The interactions were studied using the slice statement of the GLM procedure of SAS (SAS Inst., 1989).
| RESULTS AND DISCUSSION |
|---|
|
|
|---|
|
Seeding rates affected grain yield (Table 3), with significant linear and quadratic trends. Therefore, the highest grain yields in each environment were obtained with the highest seeding rates, with plant numbers of between 371 and 508 plants m2. These seeding rates are generally higher than those reported in the areas of north-central Europe with longer growing seasons where the maximum yields are normally obtained with a sowing density of about 200 to 280 seeds m2 (Laloux et al., 1980; Ellen, 1990; Easson et al., 1993; Lock, 1993; Gate, 1995; Gooding and Davies, 1997; Couvreur et al., 1999). The optimum rates of this experiment are quite similar to the 400 seeds m2 in the eastern and some western regions of the North America (Joseph et al., 1985) but much higher than 200 seeds m2, which is the most widely recommended rate in many areas of the USA (Paulsen, 1987).
The response curves of the grain yields with increasing seeding rates for each environment are presented in Table 4. Only in Site 1-2001 with an excellent growing season, did the yield response to plant density show a quadratic response, which is frequently reported in high-producing areas (Holliday, 1960; Qi-Yuan et al., 1994). The maximum yield in this environment was reached at a seed density of 348 plants m2.
|
The density x environment interaction observed was mainly due to the different response curves in each environment. In three of the four environments, yields increased linearly with increasing seeding rates although with a different slope. However, in Site 1-2001, with an excellent growing season, the response was quadratic (Fig. 1) (Table 4).
|
Balsamina obtained the highest grain yields at the highest seeding rates in three environments although according to other authors, hybrid wheat may need lower seeding rates than traditional wheat (Couvreur et al., 1999). Anza, a high-tillering variety, also gave the highest yields with the highest seeding rates in two of the four environments.
There was a variety x density interaction (Table 3). This interaction was probably due to Rinconada, which showed a quadratic response to density, whereas the other three varieties in general gave the highest yields at the highest seeding rates (Fig. 2) (Table 5). One of the reasons for the different overall response of Rinconada could be the lodging observed for this variety in Site 1-2001, which went from a rate of 0.7/10 for the lowest seeding rate to 8.5/10 at the highest (data not shown).
|
|
The variety x environment interaction observed was probably due to the different adaptation of the varieties to the growing conditions, and consequently, their relative yields differed in each environment. In three of them, Balsamina produced the highest yields, as stated above, whereas Anza also produced the lowest yields in three environments. Rinconada, a tall variety, was the least productive in Site 1 and generated the highest yields in Site 2-2001. Since this was the late-seeded environment, the response of Rinconada may suggest a better adaptation to the late-seeding conditions.
The results of our experiments show that the optimal seeding rates vary from year to year. In general, they are higher than those reported in the USA and in areas of north-central Europe with longer growing seasons and grain-filling periods. The results suggest that lower seeding rates may be adequate under excellent growing conditions.
Yield Components
Spikes per Unit Area
The average number of spikes per square meter varied depending on the environment, density, and variety (Table 3). In our study, the number of spikes per square meter increased linearly with seeding densities, and the number of spikes per square meter went from 391 and 384 spikes m2 at 175 and 150 plants m2, respectively, to 487 spikes m2 for the target density of 500 plants m2, which is an increase of about 44% in the number of spikes per square meter (Fig. 1).
As reported by other authors (Faris and De Pauw, 1981; Hay and Walker, 1989; Joseph et al., 1985; Gate, 1995), an increase in plant population density is almost invariably associated with a continuous increase in spike population density across a very wide range of seeds rates and with a progressive decrease in the number of fertile tillers per plant. In our experiments, the number of spikes per plant went from an average of 1.05 spikes plant1 at 500 plants m2, which means that the plants are almost uniculum, to 2.07 spikes plant1 at 150 and 175 plants m2, which is an increase of about 100% in the average number of spikes per plant when seeding rates are decreased to the lowest densities.
The relative rankings of the four varieties for the amount of spikes per square meter were not the same in all environments. Anza produced the highest number of spikes per square meter followed by Rinconada, Balsamina, and Gazul in two environments (Site 1-2000 and Site 2-2001), whereas in the other two, Site 1-2001 and Site 2-2000, the ranking was Anza > Gazul > Rinconada > Balsamina.
In this experiment, Anza produced an average of 486 spikes m2, Rinconada 422 spikes m2, Gazul 397 spikes m2, and Balsamina 395 spikes m2. The number of spikes per plant was significantly different for Anza (2.77 spikes plant1) than for the rest of the varieties, with 1.96, 1.92, and 1.83 spikes plant1 for Rinconada, Balsamina, and Gazul, respectively. These results are similar to those of Gate (1995), who reported that the number of spikes per unit area generally varied with variety.
The mean number of spikes per unit area ranged from 375 spikes m2 in Site 1-2000 to 492 spikes m2 in Site 1-2001 (Fig. 1). These spike densities are generally lower or much lower than the 500 or 600 spikes m2 reported for many varieties in Belgium (Laloux et al., 1980), France (Gate, 1995), the Netherlands (Ellen, 1990), the UK (Hay and Walker, 1989), and the USA (Joseph et al., 1985; Norwood, 2000). Only in Site 1-2001, with an excellent growing season, was the number of spikes produced over 500 spikes m2.
The effects of seeding density on the number of spikes m2 differed according to the environments studied. They ranked site 1-2001 > site 2-2001 > site 2-2000 > site 1-2000, for seeding rates going from 150 to 400 plants m2. However, at the highest seeding density of 500 plants m2 this relationship was not maintained in site 2-2001. Also, although the number of spikes m2 increased linearly with seeding rates, the slope of the straight lines or the rate of increase was not the same in all environments (Table 4).
The number of spikes per square meter increased linearly with seeding rates for all varieties (Table 5). However, the environment x variety interaction was possibly due to the fact that not all varieties produced similar amounts of spikes per square meter at each location. Anza was the variety that produced the highest number of spikes at each location, but the ranking of the other varieties, which produced similar amounts of spikes per square meter, varied according to the location.
The lower numbers of spikes per square meter compared with reports from north-central Europe and the USA are probably related to the production features of Mediterranean wheat and production systems. Mediterranean types of wheat normally have a lower tillering capacity and a shorter growing season (Slafer and Whitechurch, 2001). These varieties grown under relatively high temperatures have fast development rates and low tiller rates (Loss and Siddique, 1994; Satorre, 1999).
Grains per Spike
The number of grains per spike was affected by environment, density, and variety, with significant density x environment, variety x environment, and variety x density interactions (Table 3).
Seeding density significantly affected the number of grains per spike, with significant linear and quadratic trends. On average, the number of grains per spike went from 32.8 for the density of 500 plants m2 to 41.0 and 40.4 grains for the densities of 175 and 150 plants m2 respectively, which is a production of 24% more grains per spike at the lowest densities. As reported by other authors, lowering seeding rates normally increases the number of grains per spike (Hay and Walker, 1989; Gate, 1995; Gooding and Davies, 1997).
In our experiments, the environment x density effects for the number of grains per spike were due to the nonsignificant effect of the density on the number of grains per spike in Site 1-2001, the one that had excellent growing conditions (Fig. 1), possibly because the conditions allowed all the spikes to develop their potential.
The variety x density interaction was due to the lower rate of decrease of the number of grains per spike, with increasing density, of Gazul compared with the other varieties (Table 5; Fig. 3), showing that the number of grains per spike for Gazul was less affected by density. With regard to the variety x environment interaction, Balsamina, the hybrid wheat, produced the highest number of grains per spike in any environment, whereas the ranking of the other varieties varied with locations. In all environments, Balsamina had a higher spike size.
|
Grain Weight
Kernel weight was significantly affected by environment and variety but not by seeding rate (Table 3). The average TKW for the varieties was 46.33 g for Rinconada, 44.88 g for Balsamina, 44.68 g for Gazul, and 38.18 g for Anza. These figures are in the range of the values reported by other authors (Gate, 1995; Lloveras et al., 2001). It is known that TKW tends to be characteristic of a variety, and there are large differences between varieties even under good conditions (Hobbs and Sayre, 2001).
Anza had the lowest grain weight in any environment, but the relative grain weight of the other varieties varied according to the environment. The lowest grain weight, 33.61 g, was obtained in Site 2-2001, with a shorter growing period, whereas the highest, 48.75 g, was observed in Site 2-2000 (Fig. 1). For the grain weight, the ranking among environments was Site 2-2000 > Site 1-2001 > Site 1-2000 > Site 2-2001, the latter being the one with the late seeding dates. However, in Site 2-2001, grain weight was affected by density (Table 4). It has been reported that, at least in the conditions of north-central Europe, mean grain weight is normally less variable than the other yield components in seeding rate trials, provided that the crop is not subject to lodging (Hay and Walker, 1989).
| CONCLUSIONS |
|---|
|
|
|---|
The plant densities giving the highest yields are at least 400 to 500 plants m2 for most of the varieties studied. These densities are normally higher than those recommended in many non-Mediterranean areas of Europe and USA, showing that seeding rates reported in these regions are not appropriate for our irrigated Mediterranean conditions. This is probably due to the low tillering ability of the wheat and the moderate winter temperatures of the Mediterranean conditions, which may accelerate each stage of development. Consequently, higher seed densities are needed for a complete ground cover. The hybrid wheat Balsamina produced the highest average yields of 5842 kg ha1, particularly in normal growing seasons.
| ACKNOWLEDGMENTS |
|---|
| REFERENCES |
|---|
|
|
|---|
s guide: Statistics. Version 6.03. SAS Inst., Cary, NC.This article has been cited by other articles:
![]() |
A. S. Lithourgidis, K. V. Dhima, C. A. Damalas, I. B. Vasilakoglou, and I. G. Eleftherohorinos Tillage Effects on Wheat Emergence and Yield at Varying Seeding Rates, and on Labor and Fuel Consumption Crop Sci., March 27, 2006; 46(3): 1187 - 1192. [Abstract] [Full Text] [PDF] |
||||
| ||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
| HOME | HELP | FEEDBACK | SUBSCRIPTIONS | ARCHIVE | SEARCH | TABLE OF CONTENTS |
| The SCI Journals | Crop Science | Vadose Zone Journal | |||
| Journal of Natural Resources and Life Sciences Education |
Soil Science Society of America Journal | ||||
| Journal of Plant Registrations | Journal of Environmental Quality |
The Plant Genome | |||