THE INFLUENCE OF DIFFERENT HERBICIDE DOSES ON WEED INFESTATION OF WINTER TRITICALE CULTIVATED IN MONOCULTURE

The study was carried out in 2003–2005 in the Bezek Experimental Farm (University of Life Sciences in Lublin). The experimental fi eld was situated on light loamy sand soil. The phosphorus content in soil was high, in potassium medium, in magnesium low. The humus content was 1.2%. The experiment was carried out in randomized blocks in three replications. The aim of the investigation was to compare three doses of herbicides in winter triticale canopy (Janko cv., Woltario cv., Krakowiak cv.) cultivated in monoculture. The herbicides Atlantis 04 WG and Factor 365 EC were applied at full recommended doses (200 g ha-1), at doses reduced to 75% and 50%. The control was not treated with the herbicides. The weed infestation level was determined by means of the quantitative-weighting method at two dates: the fi rst one at the 6th week after herbicide application and the second one before harvest. The density of weed individuals was counted; the species composition and air– dry biomass of above-ground parts were estimated from the randomly selected areas of 1 m x 0.25 m at four sites of each plot. The density of weeds and weed air dry weight were statistically analysed by means of variance analysis, and the mean values were estimated with Tukey’s confi dence intervals (p=0.05). It was found that the number of weeds and air dry mass of weeds in the control were signifi cantly higher in comparison with the herbicide treated objects. The application of different doses of herbicides did not differentiate signifi cantly the weed infestation level in the winter triticale canopy. Viola arvensis, Matricaria maritima, Chenopodium album and Apera spicaventi were dominant species of weeds in the winter triticale canopy. The selection of cultivars did not infl uence the canopy weed infestation level.


INTRODUCTION
Weeds signifi cantly reduce yields of crop plants and affect yield quality.They compete with the crop plant for room, light, water and nutrients; they can even lead to complete choking of plants.For hundred of years, the fi ght against weeds has been mainly based on their control by using mechanical and agrotechnical methods.A breakthrough took place when chemical plant protection agents were discovered (A d a mc z e w s k i , 2000).
A reduction in herbicide doses allows to cut costs of weed control, to mitigate the risk of environmental contamination, as well as to reduce permissible levels of residues of biologically active substances in cereal products (R o l a et al. 1997; D o m a r a d z k i and R o l a , 2000; D o m a r a d z k i and S a d o w s k i , 2002).The application of a proper dose of a herbicide must take into account its weed-killing effectiveness in given conditions, the maintenance of selectivity for the protected plant and consumer safety (S t r e i b i g , 1989).
A reduction in herbicide doses also involves a change in the perception of the problem of weed control.The aim of this control should not be the total destruction of all weeds growing in a crop canopy but such reduction of their occurrence and weakening of the form so that they do not pose a threat to the crop plant (D o m a r a d z k i and R o l a , 1999).Proven et al. (1991) have found that the application of reduced herbicide rates is of special signifi cance at a low level of cereal canopy weed infestation.D a v i e s and W h i t i n g (1989), D o m ar a d z k i and R o l a (2001,2004), D o m a r a d z k i (2006) and K r a s k a ( 2006) have demonstrated that it is possible to reduce herbicide doses from 20% to 50% without a signifi cant reduction in yielding levels of winter and spring cereal crops, at the same time maintaining the required weed-killing effectiveness.They have also found that the application of herbicides at reduced doses gives the best effects in fi elds where weeds occur at early development stages.Very good results can be obtained on plantations with low weed intensity and in controlling weeds sensitive to a particular herbicide.

MATERIALS AND METHODS
The fi eld experiment was carried out in the years 2003-2005 in the Bezek Experimental Farm near Chełm, Poland, owned by the University of Life Sciences in Lublin.In the two-factor experiment conducted in randomized blocks in three replications, the effect of three herbicide doses in the winter triticale Tricicale rimpaui Wittm.canopy (Janko cv., Woltario cv., Krakowiak cv.) was compared.The herbicides were applied at full recommended doses, reduced to 75% and at doses reduced by half.Herbicide untreated plots were the control.The plot size was 12 m 2 .The results were presented as three-year means.
The experimental fi eld was located on incomplete podzolic soil lying on marl substrate with the granulometric composition of light and strong loamy sand.This soil is classifi ed as IVb evaluation class and good rye complex.Soil nutrient availability in terms of available phosphorus was high, in potassium medium, in magnesium low.Expressed in mg kg -1 , it was as follows, respectively: P 2 O 5 -171; K 2 O -120; Mg -22.The humus content was 1.2%.The soil reaction was slightly acid, and the pH in 1 mol KCl was 6.0.
Canopy weed infestation was determined twice using the quantitative-weighting method.The fi rst time was 6 weeks after herbicide treatment, the second time before winter triticale harvesting.The number of individuals, species composition and air-dry mass of above-ground parts of weeds were determined from the test areas surrounded by a frame of 1 m x 0.25 m in four randomly selected sites of each plot, in accordance with the recommendations given in the paper of B a d o w s k i et al. (2001).
The obtained results were statistically analysed by means of variance analysis.The mean values were compared by means of the least signifi cant differences using the Tukey test.

RESULTS
The weed infestation level of winter triticale canopy was signifi cantly modifi ed by the application of herbicides.At both dates of weed infestation assessment, it was found that there was a signifi cantly lower number of dicotyledonous and monocotyledonous weeds and their total number as well as smaller air-dry mass of weeds in the herbicide-treated plots compared to the control in which no herbicides were applied (Tab.1).
The selection of triticale cultivars did not signifi cantly change the canopy weed infestation level at both assessment dates.However, a tendency towards the occurrence of a larger total number of weeds in the canopy of 'Woltario' cv. was noticeable before the harvest (Tab.2).
The largest number of weed species at both dates of weed infestation assessment was found in the control compared to the herbicide-treated plots (Tab.3, Tab.4).In the herbicide-treated plots, those in which a half of herbicide doses was applied were characterised by the greatest species diversity, whereas those where the full recommended dose was applied had the lowest species diversity.At the fi rst date of weed infestation assessment, the applied herbicides completely eliminated the following weed species, compared to the herbicide untreated control: Chenopodium album, Cerastium holosteoides, Conyza canadensis, Spergula arvensis, Sonchus arvensis, Anchusa arvensis and Raphanus raphanistrum.At the same time, they reduced to a large extent the occurrence of Matricaria maritima and Apera spica-venti (Tab.3).But before winter triticale harvest, the used herbicides eliminated Spergula arvensis, Gypsophila muralis and Anchusa arvensis.At the same time, out of the species occurring in great numbers, they markedly reduced the number of Chenopodium album, Matricaria maritime and in the fi rst place Apera spica-venti (Tab.4).
From the group of dicotyledonous species, Viola arvensis, Matricaria maritima, Veronica arvensis and Chenopodium album occurred most numerously in the    canopy of all winter triticale cultivars at both dates of weed infestation assessment.But among monocotyledonous species, the most numerous were Apera spicaventi and Echinochloa crus-galli (Tab.5, Tab. 6).At the fi rst assessment date, 24 weed species, including 19 dicotyledonous and 5 monocotyledonous ones, were found in the canopy of 'Janko' cv., whereas there were 22 dicotyledonous and 4 monocotyledonous species in the canopy of the other cultivars (Tab.5).In the canopy of winter triticale 'Woltario' cv., 30 weed species, including 25 dicotyledonous ones, were found at the second assessment date, whereas in the canopy of the cultivars 'Janko' and 'Krakowiak' there were 28 species in each, including 24 and 23 dicotyledonous ones, respectively (Tab.6).

DISCUSSION
At the fi rst date of weed infestation assessment, the applied herbicides reduced the number of weeds from 68.6 to 80.1% and their air-dry mass from 84.0 to 93.0%, compared to the herbicide untreated control.The total number of weeds estimated before triticale harvest was smaller by 66.7 to 77.9% compared to the control, and air-dry mass of weeds from 87.5 to 94.0%.Worth noting is the fact that in the plot with the herbicide dose reduced to 75%, at the second assessment date, both the number and air-dry mass of weeds were slightly smaller than in the other herbicide treated plots (Tab. 1).Differences between the herbicide treated plots were within the statistical error.Research conducted by S t a r c z e w s k i and Ż ą d e ł e k (2000) has shown that full doses of preparations reduced airdry mass of weeds, depending on the triticale cultivar, from 46% to 54% compared to the control.
The selection of winter triticale cultivars did not signifi cantly affect the canopy weed infestation level.At the fi rst date of weed infestation assessment, the effectiveness of action of different herbicide doses, compared to the control, was from 9.2% to 42.0% in relation to Viola arvensis, Veronica arvensis from 59.3% to 74.6%, Matricaria maritima from 92.9% to 96.0%, Capsella bursa-pastoris 97.7% and Apera spica-venti from 86.3% to 99.5%.The effi cacy of control of the most numerous weed species before triticale harvesting was 94.4% for Chenopodium album, Matricaria maritima from 89.1 to 68.8% and Apera spica-venti from 96.6% to 99.7%.
In their experiment, R o m e k and D z i e n i a (2000) confi rm that decreased herbicide doses resulted in reduced effectiveness of control of Apera spica-ven-ti.They found in their research that the application of a full dose of the herbicides Protugan 500 SC, Lentipur 80 WP and Chisel 75 DF reduced Apera spica-venti, on the average, by 71% to 77%.In turn, a reduction in the dose of the tested herbicides by 50% resulted in reduced effectiveness of control of the abovementioned weed by 16% to 31%.Higher effectiveness of Apera spica-venti reduction was noted by D o m a r a d z k i et al. (2003), when applying Arelon Dispersion 500 SC at a full dose and at doses reduced to 75%, 50% and 25%.
The preparation applied at a full dose reduced to a high extent (99-100%) the occurrence of common windgrass, whereas lowered doses reduced Apera spicaventi to a slightly smaller extent (from 93% to 99%).W h i t i n g et al. (1991) have found that it is possible to reduce the occurrence of dicotyledonous species and Apera spica-venti in cereals by the application of herbicides at reduced doses from 30 to 60%.Also D a vi e s and W h i t i n g (1989), when applying herbicides containing isoproturon, did not fi nd any differences in the effectiveness of control of Apera spica-venti between the full dose and the dose reduced to 50%.
A d a m c z e w s k i and P r a c z y k (1999) claim that Apera spica-venti poses the greatest risk to winter cereals among monocotyledonous weeds.It has found confi rmation in the control of the experiment in question.
D o m a r a d z k i ( 2006) is of opinion that the main criterion deciding about the herbicide rate should be, in the fi rst place, the condition and degree of weed infestation of a fi eld.A weaker effect of reduced herbicide doses should be expected in winter cereals, in particular in triticale and barley, where these weeds are more advanced in development due to early sowing of these species.Moreover, lower doses have a weaker effect especially in cases of very high weed intensity in a crop canopy.

CONCLUSIONS
1.The weed infestation level of winter triticale canopy, measured both in terms of the number of individuals and air-dry mass of weeds, was not signifi cantly differentiated by the applied doses of herbicides.It indicates the possibility of reducing herbicide doses in the winter triticale canopy without the risk of increasing the weed infestation level.A tendency towards the occurrence of a larger number of individuals and higher air-dry mass of weeds was only found in the plots with the herbicide dose reduced by half compared to the other herbicide treated plots.2. Before winter triticale harvesting, a larger number of weed species was found in all the plots assessed.
Viola arvensis, Matricaria maritima, Chenopodium album and Apera spica-venti were the dominant species in the winter triticale canopy at both dates of weed infestation assessment.3.At both dates of weed infestation assessment, the applied herbicides reduced the occurrence of Matricaria maritima, Chenopodium album and Apera spica-venti to the largest extent.4. The selection of cultivars did not signifi cantly affect the weed infestation level of winter triticale canopy.
F e l e d y n -S z e w c z y k and D u e r (2005), P a r yl a k et al. (2006) and K r a s k a (2006) draw attention to the possibility of reducing the weed infestation level in cereals by selecting an appropriate cultivar.
Weed infestation elements in canopy of x Triticale rimpaui Wittm.per 1 m 2 in dependence on herbicide doses, mean fi gures in the years 2003-2005.Weed infestation elements in canopy of x Triticale rimpaui Wittm.per 1 m 2 in dependence on cultivars, mean fi gures in the years 2003-2005.
*A -control (without herbicides) B -full dose of herbicides C -75% dose of herbicides D -50% dose of herbicides Table1

Table 3
Species composition and the number of weeds per 1 m 2 of x Triticale rimpaui Wittm.canopy at the fi rst date weed infestation assessment in dependence on herbicide doses, mean fi gures in the years2003-2005.

Table 4
Species composition and the number of weeds per 1 m 2 of x Triticale rimpaui Wittm.canopybefore harvest in dependence on herbicide doses, mean fi gures in the years2003-2005.

Table 5
Species composition and the number of weeds per 1 m 2 of x Triticale rimpaui Wittm.canopy at the fi rst date weed infestation assessment in dependence on cultivars, mean fi gures in the years2003-2005.

Table 6
Species composition and the number of weeds per 1 m 2 of a x Triticale rimpaui Wittm.canopybefore harvest in dependent on cultivars, mean fi gures in the years2003-2005.