ABSTRACT

The use of GMO herbicide-resistant oilseed rape varieties must be considered as a complete change in weeds control techniques in oilseed rape, since, at present time, weeds control is obtained mainly through pre-sowing and pre-emergence herbicides.

The results of this study gives some answers concerning the way to optimize the use of herbicide-resistant oilseed rape systems regarding efficacy and weeds control strategies. It clearly appears that the control of some weeds species remains difficult, even with the broad spectrum herbicides which are involved in these systems.

Weather interactions may be important on the efficiency of the weeds control strategies (number and dates of herbicides applications) using non persistent herbicides, as far as it may be more or less favorable to continuous emergence of weeds during autumn.

At last, the possibility to spray or not according to the actual situation of the field would induce needs of decision criteria for farmers .

INTRODUCTION

If accepted in western Europe countries, the introduction of GMO oilseed rape varieties for herbicide tolerance would be a major evolution in winter OSR weeds control techniques. At present time, winter OSR weeds control is very generally obtained using pre-sowing and/or pre-emergence herbicides with residual effects. Herbicides sprays after emergence are limited to the control of monocotyledons grasses and a limited number of dicotyledons weeds. A large scale use of post-emergence herbicides without residual effects constitutes a complete technical change, permitting to adapt the sprays according to present weeds species, climatic conditions and crop status.

The objectives of this 3 years study was to evaluate 3 herbicides tolerance systems in a practical way in order to achieve a better assessment of their potential technical interest and limits for OSR production in Western Europe.

MATERIAL AND METHODS

The study deals with 3 winter OSR varieties tolerant to 3 active ingredients: glufosinate, glyphosate and ioxynil-bromoxynil.

Glufosinate ammonium: is a contact broad spectrum herbicide, with some systemic action within leaves. It inhibits glutamine synthetase in the plants, with accumulation of ammonium ions and inhibition of photosynthesis. It has no residual effect. SL formulation 200ga.i. /l. Basic dose D in our experiments: 600g/ha.

Glyphosate: is a broad spectrum systemic herbicide, absorbed by the foliage, with rapid translocation throughout the plant. It acts on various enzyme systems, interfering with the formation of amino-acids. It is inactivated on contact with soil and has no residual effect. SL Formulation 360ga.i./l. Basic dose D: 720g/ha in our experiments.

Ioxynil-bromoxynil: ioxynil and bromoxynil are contact herbicides with some systemic activity, absorbed by the foliage with limited translocation. They inhibit photosynthesis and uncouples oxydative phosphorilation. Active mainly on polygonaceae, asteraceae, boraginaceae. Formulation: 200ga.i./l + 200ga.i./l. Basic dose D in our experiments: 300 g/ha + 300 g/ha.

Herbicides experiments were carried out during 3 consecutive seasons (1995/96, 1996/97, 1997/98) quite different considering the weather conditions, in different French regions, representative of the OSR cropping conditions. Glufosinate system was studied on a total of 14 experiments, whose location were declared to official services according to the present regulations, Glyphosate system on 12 experiments, ioxynil-bromoxynil system on 9 experiments. 5 additional experiments were carried out on classical OSR varieties (without regard to OSR destruction) to study the effect of single applications of the formulations on chosen floras, known as difficult to control.

All experiments were managed according to the CETIOM chart for GMO good experimentation practices and according to the recommendations of the French Commission des Essais Biologiques, following the OEPP standards, for the technical aspects.

Classical experimentations designs are made of 3 complete blocs, with adjoining non treated test plots. Since the objective was the evaluation of practical efficacies and visual selectivity, most of the experiments were not harvested.

Efficacies for each specy were evaluated by counting on the test plots and visual estimation of the efficacy or counting on treated plots.

The basic protocols of this study include:

- single herbicide application at a basic rate at 3 different dates, i.e. 3 different weeds/crop stages

- double applications of reduced doses (50% to 66% of the basic rate), the first application at autumn, the second one before or after winter.

RESULTS AND DISCUSSION

Selectivity

Very limited symptoms were observed in 1995/96, for the 3 systems, without incidence. These symptoms were not observed in 1997 and 1998 even at double rates.

Efficacies of single applications

Efficacies at basic rates are generally very satisfying (see table 1) with high weeds control levels, equal or superior to classical pre-emergence references. Nevertheless some exceptions appear:

- geranium sp. are hardly controlled by glyphosate with a single application at a basic rate of 720 g a.i./ha. Results are better with glyphosate (600g a.i./ha) except for applications at the end of winter.

- viola arvensis is poorly controlled by any of the 3 herbicides at basic rates.

- cereals volunteer are not controlled by ioxynil-bromoxynil, which is selective of these crops.

Low efficacies of single applications are due to several causes:

- weeds growth stages: weeds are generally more difficult to control when they are developed and quite old. An example for ghyphosate on rapistrum rugosum in chart 1. On the same location, a similar behavior is observed with ioxynil-bromoxynil, but there is no dose effect of glyphosate for anthemis.

- weeds populations dynamics depend on species (some species, like viola a., are able to germinate all over the year, others preferentially at spring, or summer or autumn) and on rainfall periods. It may cause some difficulties to control complex floras by a single spray. Good examples are given by the efficacies of glyphosate and glufosinate on geranium species and viola a. in Burgundy in 1997 (table 2): the emergence of these species occurred quite lately (October) and progressively (chart 2).

- ``umbrella'' effect of OSR crop: young weeds covered by the leaves of OSR are not reached by the herbicide which appears inefficient (contact herbicides). This effect was observed on several trials, generally for late autumn sprays. The problem does not exist for early autumn sprays (rapeseed is still young and does not cover weeds plantlets). The fact that many rapeseed crops loose a part of their foliage during cold winter, due to frost and/or temporary nutrient deficiencies limits this problems for the sprays at date 3 (end of winter).

- effect of weather conditions on herbicides efficiencies: the 3 herbicides are contact herbicides, with important or limited systemic activities, and are more or less sensible to the physiological status or the plants, growing or not, specially depending on air temperatures.

The adaptation of doses must be considered according to weeds species, growth stages and weather conditions. Some doses effects for single applications of the 3 herbicides are shown in table 3, for example with glyphosate on geranium. The decreases for glufosinate efficacy at 2/3 of the basic dose are consequent for date 2, in quite cool conditions. A synthesis of the possibilities and risks of doses reductions according to our references is made on table 4.

CONCLUSION AND PERSPECTIVES

The characteristics of the active ingredients used in the present OSR herbicide tolerant varieties lead to the practical following advantages:

- the post emergence application allows to adapt the sprays according to the observed flora, and actual emergence of this flora, which depends on species and weather conditions. In poorly infested fields, observations will allow savings as number of sprays, active ingredients quantities and possibly costs.

- the active ingredients of the 3 systems have no residual effects and do not represent any risk for the following crops; in case of early failure of the OSR crop, sowing another crop is not so problematical.

The experiments show the limits of these systems too some species reveal to be quite resistant at the basic rates, such as viola arvensis for the 3 systems, and at a lesser extent geranium species (but much more competitive to OSR). Two questions will get answers in future: 1/ will others weeds species show similar behavior, 2/ could misuses or poor doses adaptation lead to weeds flora evolutions, specially in case the same active ingredients would be used on several crops of the rotation.

- several factors must be taken into account to obtain the best results: stage of the crop (and ``umbrella'' effect), weather conditions, stages of the weeds. A quite precise follow up of the crop will be necessary.

- the control of progressive emergence of weeds could require double sprays, since a single early spray will not control later emergence, and a single late spray could be less efficient on certain species (due to growth stages, OSR``umbrella''effect, weather conditions...). As emergence dynamics vary according to weeds species (spring, summer, autumn or indifferent), double sprays will be probably necessary for complex floras: some weeds with important growth during autumn (many brassicaceae) are better controlled by early sprays when other species are still to emerge (geranium, stellaria...). Farmers will have to control sometimes several weeds emergence waves, specially in case the OSR crop still does not cover completely the soil surface.

A regular follow up of the crops will be required to decide of the most appropriate time for the herbicides applications and to observe the possible new infestations.

Acknowledgments

This study has been carried out on the basis of results obtained on the experimental stations of CETIOM (Nancy, Dijon, St Florent sur Cher, Toulouse-En Crambade, Surgères).

Chart 1 Efficacy of glyphosate on anthemis (ANTSS)et rapistrum rugosum (RASRU) for different growth stages of the weeds in trial 97DCT31015

Table 1 : Efficacies on the OSR weeds flora with a single herbicide spray

	GLYPHOSATE (2l/ha pf)			GLUFOSINATE (3l/ha pf)			OXYNILS (1.5l/ha pf)
OSR growth stage	B4	B6	C2	B4	B6	C2	B4	B6	C2
ammi majus	100 (1)	97 (1)		100 (1)	85 (1)	86 (1)	-	-
anthemis sp.	100 (1)	99 (1)	99 (1)	99 (1)	99 (1)	100 (1)	99 (1)	89 (1)	50 (1)
Arabidopsis thaliana			95 (1)	-	-		-	-
bromus sp	100 (1)	25 (1)	83 (1)	99 (1)	100 (1)	42 (1)	40 (1)	0 (1)	0 (1)
capsella bursa pastoris	100 (1)	93 (1)		-	-		-	-
papaver rhoeas		-	93 (1)	91 (2)	97 (2)	90 (2)	-	85 (1)	28 (1)
galium aparine		-	100 (1)	-	-		-	-
geranium molle	80 (1)	80 (1)	70 (1)	98 (1)	98 (1)	70 (1)	25 (1)	0 (1)	0 (1)
geranium pusillum	61 (2)	73 (2)	50 (2)	85 (2)	93 (2)	6 (1)	41 (1)	96 (1)	0 (1)
geranium sp	67 (3)	75 (3)	57 (3)	92 (4)	96 (4)	38 (2)	33 (2)	-	0 (1)
Lamium amplexicaule	100 (1)	100 (1)	100 (1)				100 (1)	90 (1)
medicago sativa	95 (1)	100 (1)	99 (1)	99 (1)	97 (1)	85 (1)	25 (1)	0 (1)	0 (1)
matricaria sp	98 (1)	100 (1)	94 (3)	92 (3)	99 (3)	100 (3)	98 (1)	49 (2)	33 (1)
mercurialis annua	95 (1)	87 (2)		73 (1)	85 (1)		80 (1)	100 (1)
sinapis arvensis	100 (3)	76 (3)	100 (1)	98 (3)	90 (3)		96 (2)	100 (3)	70 (1)
viola arvensis	40 (2)	67 (3)	77 (5)	48 (6)	40 (6)	48 (4)	100 (1)	48 (3)	39 (3)
rapistrum rugosum	100 (1)	80 (1)	70 (1)	94 (1)	20 (1)	98 (1)	99 (1)	92 (1)	78 (1)
raphanus raphanistrum	-	-		100 (2)	94 (2)	32 (1)	-	11 (3)
Wheat volunteersr	100 (2)	100 (2)	87 (3)	100 (1)	58 (1)	76 (1)	-	-
barley volunteers	95 (2)	89 (2)	100 (2)	72 (4)	92 (4)	55 (4)	9 (3)		0 (2)
OSR volunteers	100 (2)	86 (2)	93 (3)	100 (2)	87 (2)	100 (2)	92 (2)	57 (2)	50 (1)
Sisymbrium officinale	100 (1)	100 (1)	98 (1)	98 (2)	99 (2)	0 (1)	98 (2)	45 (2)	0 (1)
stellaria media	98 (1)	100 (1)	88 (2)	64 (2)	99 (2)	100 (1)	92 (1)	35 (1)
veronica persica		-	89 (1)	100 (1)	100 (1)	100 (1)	-	57 (1)	74 (1)
veronica sp		-	89 (1)	100 (1)	100 (1)	100 (1)	-	57 (1)	74 (1)
alopecurus myosuroides		-	94 (1)	-	-		-	-

OSR Growth stages: B4 = 4 leaves / B6 = 6 leaves / C2= end of winter, stem growth.

() : number of references

Very sensible : efficacy >= 95%

Sensible : 85% <= efficacy < 95%

Medium- Irregular : 70% <= efficacy<85 %

Resistant : efficacy< 70%.

Chart 2: emergence dynamics of geranium p. and viola/ Burgundy 1997

Table 2: effects of germination dynamics / Burgundy 1997
efficacies% / non treated		Geranium	viola	geranium	viola
		dissectum	arvensis	pusillum	arvensis
	OSR stage	trial 21042	trail21042	trial 21026	trail21026
dose	at spray. time	glyphosate	glyphosate	glufosinate	glufosinate
1D	B4	70	15	68	50
1D	B6	80	62	90	8
2/3D	B6	78	52
D	C2	100	100
1/2D + 1/2D	B4 /C2	100	100	94	31
pre-emergenceclassical reference		100	90	78	68
weeds stage / at date 1 OSR B4)		2 leaves	?	2 leaves	3 leaves
weeds stage / at date 2 OSR B6)		8 leaves	8 leaves	rosette	rosette
weeds stage / at date 3 OSRC2)		rosette	flower	rosette	flower

Table 3: Effects of doses / exp C98DSR21042 / 1997/98				observations : 30 days after treatment
Single applications / species	bromus	geranium p.	viola a.	hordeum v.	papaver r.	galium a.	sonchus a.
nber of weeds on non treated plots	120-298	6-23	4-9	21-29	10-80	5-16	4-10
glyphosate dose D / date 1	100	100	100	100	-	-	100
glyphosate dose 1D/2 / date 1	94	61	100	55	91	33	100
glyphosate dose 2D / date 2	58	77	95	80	100	100	-
glyphosate dose D / date 2	25	65	100	78	100	(67)	-
glyphosate dose 2D/3 / date 2	42	59	93	97	100	-	100
glyphosate dose 2D / date3	98	53	100	100	-	-	-
glyphosate dose D / date 3	83	0	93	100	100	-	0
glufosinate dose D /date 1	99	100	100	45	-	-	100
glufosiante dose 1D/2 /date 1	81	92	100	0	85	69	100
glufosinate dose D /date 2	100	100	83	75	97	76	100
glufosinate dose 2D/3 / date 2	67	95	19	54	91	-	100
glufosinate dose D / date 3	42	41	47	0		100	-
iox-bromoxynil dose D / date 1	40	41	100	5	-	89	100
iox-bromoxynil dose 1D/ 2 / date 1	43	14	100	12	100	-	0
iox-bromoxynil dose D / date 2	0	96	52	0	95	93	100
iox-bromoxynil dose D/ 2 / date 2	0	82	54	0	100	80	50
iox-bromoxynil dose d / date 3	0	0	0	0	-	-	-
stade adventices à d1	3 leaves	2 leaves	2 leaves	6/8 leaves	2 leaves	2 stages	2 leaves
stade adventices à d2		5leaves	4 leaves		4 leaves	3stems	4/6 leaves
stade adventices à d3		rosette	rosette		rosette	3 stems	rosette
(x) : mean from irregular data

Table 4: Tests for doses réductions were achieved on classical OSR varieties in 1997/98, for single sprays. Tested doses were 70 and 50% of the basic dose.

TABLE 2

glyphosate (2l/ha pf)

(glufosinate (3l/ha pf)

(oxynils 1.5l/ha pf)

OSR Growth stage

B4

B6

C2

dose reduction

B4

B6

C2

dose reduction

B4

B6

C2

dose reduction

bromus sp

S

R

S

XX

S

S

R

XX

R

R

R

-

geranium pusillum

S

M

R

X

S

S

R

XX

R

S

R

XX

geranium molle

S

S

R

XX

S

S

M

XXX

R

R

R

-

viola arvensis

S

S

S

XXX

S

S

R

XX

S

R

R

XX

barley volunteers(1st trial)

S

M

S

XX

R

R

R

X

R

R

R

-

barley volunteers (2nd trial)

S

S

S

XXX

S

S

R

XX

R

R

R

-

barley volunteers (3rd trial)

S

S

S

XXX

R

R

S

X

R

R

R

-

sinapis arvensis

S

S

S

XXX

S

S

-

XXX

S

S

M

XX

medicago sativa

S

S

S

XXX

S

S

S

XX

R

R

R

-

myosotis arvensis

S

R

S

X

S

S

S

X

S

S

S

XX

Dose reduction: xxx : possible / xx : possible , but risks of irregular efficacy, depending on growth stages / x : dose reduction to be avoided / / S: Sensible: 85% <= efficacy < 95% / M: Medium- Irregular: 70% <= efficacy<85 %/ R: Resistant: efficacy< 70%.