An estimated ~176M was spent on agricultural chemicals in Australia in 1980 and, if one adds to that the cost of application equipment and labour, it can be seen that chemical farming in Australia today is really big business (Table 1). It should, however, be remembered that business is a two-sided affair and whilst this enormous upsurge in pesticide usage has no doubt gladdened the hearts of the chemical companies it has generally had an even greater benefit on productivity and profitability at the farm level.

Table 1. Expenditure on agricultural chemicals in Australia, 1980. (A.V.C.A. Chemical Audit) Pearce (1981)

Category	Australia
Herbicides	$ 94.30 M.
Insecticides	$ 47.80 M.
Fungicides	$ 15.29 M.
Others	$ 19.03 M.
Total	$176.42 M.

In this paper I intend to deal mainly with herbicides - the agricultural chemicals most generally used on farms in this region today. Indeed, there has been no greater impact on our crop production during the last decade than that of the herbicide revolution. The ability to control weeds through the use of chemicals has directly or indirectly resulted in radically different tillage systems, intensified cropping sequences, and expanded crop area, more timely sowing and higher yields. Similar trends have occurred throughout the world and one can read of the changes in farming systems that have evolved. In the UK, for example, a major swing to autumn planting of barley, compared to the normal spring planting, was facilitated by changed weed control and minimum tillage.

Whilst increased usage of herbicides has clearly benefited crop production immensely, it is already evident that total reliance on chemical weed control will be a mistake that could lead to a spiral of tolerant weeds, more persistent herbicides and ultimately excessive residues in our soils. Community attitude to all pesticide usage is hardening and even at farmer field days it is now commonplace for growers to question widespread chemical application. For herbicides to survive the 1980’s unscathed, a careful approach must be made to develop weed management systems that integrate the planned use of herbicides with limited cultivations possibly in the presence of crop residues, sound crop rotations, suitable planting methods and other management strategies aimed at reducing weed problems.

In addition to the development of these integrated control systems, there is a need for much more data on the economics of weed control as the farmer is faced not only with an ever- increasing range of herbicides, but also increasingly expensive herbicides. He therefore needs to know when spraying is necessary and how much chemical should be applied.

Clearly the major problem facing the cereal grower in the 1980’s is increasing costs, and as costs and prices are largely out of his control, he can only beat these cost rises through increased productivity. Improved weed control is an essential part of that drive for increased productivity.

The Herbicide Revolution

It is estimated that at least 65% of the wheat crop in Australia is now sprayed with herbicides each year. This compares with about 25%, 6 to 10 years ago. The growth area in the herbicide market has been in the selective control of annual grasses, particularly ryegrass, and this is well illustrated by research data on the herbicide market in Western Australia (Table 2). The figures from WA are chosen for reasons of convenience -but similar trends are evident in both New South Wales and Victoria.

Table 2. Area of cereal crops treated and cost of the herbicides used, Western Australia 1980 and 1981. (G. Pearce, 1981)

	1980	1981
Herbicide Use	Area (ha)	Cost	Area (ha)	Cost
Wild oats	140 000	$ 2.41 M	175 000	$ 3.01 M
Ryegrass	694 000	$ 5.12 M	1 345 000	$ 7.52 M
Pre-planting paraquat/diquat	230 000	$ 2.03 M	496 000	$ 2.73 M
Broadleaved weeds (post-emergence)	1 907 000	$ 5.57 M	2694 000	$ 7.60 M
Total	2 971 000	$15.13 M	4 710 000	$20.86 M

This herbicide revolution was, I believe triggered off by the release of trifluralin for annual grass control during the early 1970’s and its subsequent widespread usage from about 1975 onwards. The success of trifluralin was not only a reflection of the major yield losses caused by annual weeds (particularly ryegrass) throughout the cereal belt, but was also a marketing masterpiece by Elanco showing both Government and Industry how to sell a new concept. The usage of trifluralin had a significant influence on cereal production, both good and bad. On the debit side, farmers paid greater attention - in retrospect too much attention - to the preparation of a finely worked seedbed. On loam soils in south-eastern Australia this has had a deleterious effect on soil structure, not only at the surface but also in the sub-surface where there is evidence of compaction that was probably aggravated by the repeated passage of tractor wheels. In the Mallee, concern is also being expressed that the working of soils for trifluralin usage is creating an increased wind erosion hazard. However, when the case for and against trifluralin is considered, the benefits of effective weed control at low cost still heavily outweigh these disadvantages at least in the short-term. In addition, the quickened demise of soil structure as a direct result of trifluralin usage has heightened farmer concern for soil structure - a noteworthy and highly desirable spinoff - and firmly re-established the pursuit of minimum tillage systems.

The availability of new herbicides has certainly had a major impact on farmers’ ability and willingness to reduce their level of cultivation. Whilst paraquat and diquat have been with us for many years, the marketing of the ‘Sprayseed’ formulation and the addition of ‘Roundup’ to the broadacre market have made for more effective and economical weed knockdown. I estimate that within a 100km radius of Rutherglen the area of direct drilled crops has risen from approximately 1000 ha in 1978 through 4000 ha in 1979 to approximately 50,000 ha last year. Again, data available from Western Australia confirms this tremendous increase in the use of chemicals for reduced tillage systems. The figures for ‘Sprayseed’ are shown but one should also remember that similar trends in ‘Roundup’ usage would also be apparent, particularly here in the eastern States.

The development of herbicides to control grassy weeds without the need for soil incorporation was the key to overcoming the weak link in minimum tillage, which was to consistently get similar in-crop weed control to that obtained in a normally cultivated crop. A number of herbicides have helped to fill the bill, but none more so than Hoegrass (R). Its effective control of annual ryegrass and wild oats in wheat, barley, triticale, lupins and other crops is now well documented. This year Glean(R) is added to the armoury, allowing even more flexibility in the management of cropping paddocks.

Table 3. Area direct dri.lled using paraguat/diguat, Western Australia. (G. Pearce, 1981)

Year	Area planted (ha)
1971	10,000
1972	11,000
1973	28,000
1974	70,000
1975	22,000
1976	22,000
1977	25,000
1978	40,000
1979	100,000
1980	230,000
1981	496,000

¹Trade name Sprayseed

There has also been another interesting spin-off as farmers have become more familiar with the direct drilling technique and that is an increasing usage of the ‘straight-in’ system where seeding is the first operation carried out and all weed control is attained either in the preceding year or by ‘in-crop’ selective herbicides.

One potential problem that is arising as new and better herbicides are developed is the danger that the longstanding marriage between cultivation and weed control will become another statistic in the divorce court. Our concern for the 1980’s must be to ensure that the marriage continues to the extent that we still maintain flexibility in our attitude to crop protection, indeed to our whole farming system - integrated management programmes are the name of the game if we are all to be in business in 10 or 20 years’ time.

Chemicals in the Management Team

Chemicals are an important member of the management team and like all good team players are more effective when giving support to, and receiving support from, other management techniques. Complementary techniques include the strategic use of grazing, rotations, crop and pasture nutrition and cultivation.

Grazing/Cultivation

The role of grazing animals in supporting the performance of herbicides should not be forgotten. A classical older example was the prevention of ryegrass seeding in the pasture year prior to cropping by using heavy grazing or hay cutting, this reduced ryegrass population in the following crop (Table 4) and increased subsequent grain yields (Table 5).

Table 4. Effect of pasture management and ploughing implement on ryegrass density.

Treatment	Cropping year
	1st	2nd	3rd	4th
	plant s/in
Untreated pasture	75.Oa*	135.4a	325.Oa	291.3a
Heavily grazed	16.2bc	46.4b	194.Ob	202.4b
Cut for hay	ll.7c	39.4b	147.lb	190.3b
Burnt	25.2b	51.2b	184.2b	178.Ob
Mouldboard plough	20.2a	52.9a	189.2a	194.2a
Disc plough	30.8b	67.9b	218.7b	233.lb

Table 5. Effect of pasture management and ploughing implement on wheat grain yields.

Treatment	Cropping year
	1st	2nd	3rd	4th
	t/ha
Untreated pasture	1. 86a~	2.13a	l.38a	l.25a
Heavily grazed	2.38b	2.58b	l.50a	1.39a
Cut for hay	2.33b	2.48b	1.58a	l.35a
Burnt	2.23b	2.67b	1.57a	1.39a
Mouldboard plough	2.34a	2.54a	l.57a	l.41a
Disc plough	2.06b	2.39b	l.45b	l.28b

* Postscript letters refer only to the comparison between pasture management or between ploughing treatments within the one year. Values followed by the same letter do not differ (P 0.05).

Grazing is also critical to the performance of Sprayseed and to a lesser extent Roundup (.R). Where weeds are grazed from the break of the season lower chemical rates can be used for effective weed control.

In more recent work we have looked at the interactions of cultivation and herbicide usage on two important weeds - wild radish and sorrel. In a preliminary experiment it was shown that the density of wild radish could be greatly reduced following deep burial of the seed by mouldboard ploughing, but as these deeply buried seeds retained their viability (ability to germinate) for at least 4 years, we became more interested in tillage systems that could help to reduce wild radish in the longer term, Cultivation treatments evaluated were scarifying to a depth of 7cm, mouldboard ploughing to 15cm, and direct drilling of the crop following no prior cultivation or following an ‘early tickle’ - a shallow cultivation to about 3cm.

Wild radish seedlings emerging in the crop were counted just prior to using an early post-emergence herbicide (bromoxynil + MCPA, or Sencor-T) for their control, and again just prior to the application of 2, 4-D to control late germinating seedlings.

At both times of assessment the ploughed plots had less wild radish than the other treatments, whilst the wild radish density was greatest on the direct drilled plots (Table 6).

Table 6. Effect of cultural treatments on the density of wild radish in the following wheat crop.

Treatment	Wild radish seedlings (no/m2)
	Site 1		Site 2
	July 24	Sep.18	July 24	Sep.18
Scarified 7 cm	119 b	13 b	39 b	7 a
Mouldboard ploughed 15 cm	41 a	3 a	14 a	1 a
Direct drilled	202 c	25 c	106 b	20 a
Scarified 3 cm + direct drilled	97 b	17 bc	32 b	6 a

1 Values followed by different letters are significantly different (P=0.05) (Comparisons within columns only.)

Because of the high germination where the crop was direct drilled and the more rapid loss of viability of wild radish seed near the surface, success cropping with direct drilling and the effective in-crop use of herbicides should result in a marked reduction in the number of viable radish seeds in the soil. This is an example where the combination of direct drilling and herbicide usage now may eventually reduce the need for chemical usage in the future.

Sorrel - a weed becoming more prevalent in both crops and pastures in this region - poses a rather different problem, however, and highlights the need for a flexible approach to weed control. There are herbicides on the market that will give some control of sorrel, but as it spreads both as a perennial and by prolific seed production this control is often not satisfactory. We have preliminary evidence that minimum tillage, particularly after Sprayseed, can encourage the spread of sorrel compared to cultivation, or cultivation plus the use of herbicides (Table 7). More work needs to be carried out but there is an indication at least, that if you have a significant sorrel problem in a paddock, adapt your management to include strategic cultivation, rather than hunt for a more potent herbicide mixture.

Table 7. Effect of various tillage and chemical treatments on sorrel populations in the following year, Rutherglen 1981. (G.R. Code & T.G. Reeves unpublished data.)

Treatment in 1981	Sorrel present April 1982
	Rosettes/m2	Seedlings/m2
Convential cultivation	132	49
Conv. cultivation & dicamba/2, 4-D in crop	77	47
Direct drill (after Sprayseed)	257	168
Direct drill (after Sprayseed) & dicamba/2, 4-D in crop	174	88
Direct drill (after Sprayseed & dicamba) & dicamba/2, 4-D in crop	180	114
Direct drill (after Roundup & dicamba) & dicamba/2, 4-D in crop	60	316

But sorrel raises an even more fundamental question - why has it become widespread, is it just a problem in itself, or is it the symptom of some other deficiencies in our farming system? In the case of sorrel we believe that a decline in the productivity of subclover-’based pastures, together with increased cropping, has contributed to the increase in sorrel. Accordingly, we have commenced work in a heavily infested paddock where lime, deep ripping and trace elements are being used to promote vigorous pasture growth. Combinations of improved soil management plus herbicide usage will be used to hopefully cure not only the visible symptom - sorrel - but also the underlying causes, which we consider could be soil acidity, soil compaction and phosphorus deficiency.

Rotations

The use of biologically sound rotations, including the use of grain or crop legumes, is the keystone to successful farming in this area. Rotating crops such as wheat and lupins for example, helps to maintain soil fertility and break crop disease cycles, and is one of the reasons that we do not often have to use fungicides or insecticides in broadacre crop production. Rotating crops also gives the opportunity to rotate chemical usage and therefore handle specific weed problems in different parts of the rotation. For example, in work at Rutherglen we are investigating the effect of different rotations on wild radish populations. The rotations include combinations of pasture/wheat, lupins/wheat and peas/wheat. All with and without chemical control.

Table 8. Effects of rotation and herbicides on wild radish populations, Rutherglen 1981.

Rotation		Wild Radish plants/in2
1980	1981
p	P	11
P	W unsprayed	398
P	W sprayed	NIL
P	L unsprayed	497
P	L sprayed	219

P = Annual pasture, W = Wheat, L = Lupins

The clear message from this work is that chemical control measures should be concentrated in the wheat crops of a wheat- wheat-lupin rotation as effective chemical control in lupins is not yet possible. Similarly, emergence of wild radish in pasture is too low to permit any substantial reduction in seed numbers where pasture and wheat are rotated. In 1982 a peas- wheat rotation is included as selective radish control is possible in peas using metribuzin, Tribunil (R) or MCPA. Whilst lupins are a superior grain legume to peas in terms of their effects on subsequent wheat yields, in the short-term at least it may be necessary to grow peas if wild radish is too serious a problem in lupins. In the longer term we expect to gain better control of radish in lupins by the possible use of atrazine/ simazine combinations or SSH 0860, a new herbicide.

At one time we were often forced to put paddocks back to pasture because weeds, usually ryegrass, had built up to proportions that could not be controlled by cultivation. Chemicals have largely removed that problem, but they have not removed the need to preserve soil fertility, soil structure and crop hygiene through the use of biologically sound rotations. Herbicides are an important factor in helping us to adopt these rotations.

Getting the best out of herbicides

By definition, a weed of crops has some significant adverse effect on the production of that crop. Where the effect is physically visible - for example, a weed that causes harvesting difficulties such as skeleton weed or wild radish, or a weed that results in contaminated produce such as vetch - a decision on control is relatively easy to make. We are often concerned with a more difficult situation that involves the loss of crop yield associated with weed infestation and for the grower this often boils down to deciding when the expenditure on weed control measures - normally spraying - is worthwhile.

This important management decision often involves up to 20% of total variable costs for crop production and is influenced by a number of major factors. When one recalls that about 65% of the wheat crop is annually treated with herbicides, these factors assume great importance.

The questions that need to be answered are, what is the effect on yield of a weed population in the crop, or when does it pay to spray? This question can also be posed from the environmental aspect - how many weeds can be tolerated in a crop before it is necessary to use chemical control?

In answering these questions there are many factors to be considered, including weed species, crop species and cultivar, crop and weed densities, management and environment.

As scientists we are interested in understanding the relationship between plants growing together, but as farmers the more fundamental question is to know whether the cost of spraying will be paid for by extra return from the crop. Greg Wells (1978)used the following calculation for an economic threshold response:

cost of herbicide + cost of application + return on capital invested = minimum value of increased crop yield.

In order to determine whether the removal of a weed infestation in a cereal crop will economically increase crop yield, the most important relationships for practical consideration are those that have been derived between the density of various weed species and crop yield. Different experimental and statistical techniques have been used by various workers, but they are all suitable for determining a critical weed density at which the cost outlayed for herbicide is returned in terms of extra grain yield, The magnitude of this density varies from weed to weed and also with crop yield potential, herbicide cost and the price of the crop in question. Table 9 shows critical densities for nine annual weeds in wheat.

Other important factors to be accounted for when considering the economics of spraying include the species, cultivar, density and spatial arrangement of the crop. In this regard wheat and rapeseed, for example, have similar capacities to compete with annual ryegrass and both are more competitive than lupins (Reeves 1981).

Amongst the cereals, I believe that oats are generally more competitive than barley, wheat or triticale. Choice of wheat cultivar can also influence the yield reduction caused by a given weed density (Reeves and Brooke 1977) and wheat crops of lower densities are more greatly affected by weeds. However, probably the most critical factor in determining the economic success of controlling annual weeds in cereals is time of weed removal.

Critical weed competition occurs within 3 to 6 weeks of wheat emergence, 4-8 weeks of rapeseed emergence and 8-12 weeks of lupin emergence. In other words, annual weeds such as rye- grass should be removed from wheat very early in the growing season whilst in lupins removal can, if necessary, be delayed without significant yield loss.

Another factor contributing to more effective weed control is the increasing use of herbicide mixtures, particularly where older, cheaper chemicals can be combined with newer materials in order to reduce costs but not performance. Chemicals such as dicamba, diuron and MCPA have all increased in usage as a result of this philosophy. There is also a constant search to find new uses for existing chemicals (simazine in lupins and trifluralin in wheat were examples) and new materials for new problems. Quite often we have seen a new product control an old problem - the use of Glean (R) for soursob control in cereals and ryegrass control in oats are two good examples of this.

More work is needed on the economics of weed control in new situations and in the light of these new developments. Critical weed densities need to be determined for mixed weed populations, for additional weeds (.e.g. sorrel) and additional crops Ce.g. triticale).

The Future for Chemical Farming in Australia

The use of chemicals, particularly herbicides, will continue to be a significant factor in increasing the efficiency and productivity of crop production in Australia. However, whether we ever move to the complete programmes of chemical crop protection currently practised overseas is debatable. There, herbicides, fungicides and insecticides are regularly used on crops and to facilitate these frequent spraying operations crops are sown with missed rows or ‘tramlines’ where the sprayer regularly operates.

In Australia we place more reliance on rotations and plant breeding for disease control and, to a lesser extent, insect control and this I believe is to our overall advantage. It is clear that the community wants to see less chemical usage in farming rather than more, and at least part of the reason for this is a problem of communication. We need continuing effort to educate both users and bystanders of both the socio-economic benefits of chemicals and the risks involved if they are misused. I quote from an article by J.T. Snelson which aptly describes the change in emphasis that needs to be continued:

“The pesticide industry is based on technology. The technologist has, through error, thought that public relations were relations with the users of his products. The pesticide industry has not realised that public relations are just that - relations with the general public. The result is a gigantic credibility gap with the public.

The average consumer believes that food comes from the supermarket. The farmer is nothing more than a subsidised, far-off person who requires tax money to live.”

Whilst these are sobering thoughts, they were written about 10 years ago, and certainly efforts have been made to bridge that credibility gap. Much more effort is still needed.

Table 9. Critical weed density for a common chemical control method for various weed species in wheat (wheat valued at $ 130 t -1)

Weed species		Weed control	Variable Cost	Wheat yield (weed free)	Critical weed density 1
Common name	Botanical name	Herbicide and rate	($/ ha)	(t /ha)	(no/ m2)
Annual ryegrass*	Lolium rigidum	Diclofop-methyl 0.375 kg ha-1	13.50	2.0 1.5 1.0	23 41 92
Threecornered** jack/Doublegee	Emex australis	MethabenzthiaZurOn 0.385 kg ha-1	7.00	1.4	15
Wild oats+	Avena fatua	Diclofop—rflethyl 0.56 kg ha1-1	21.00	2.0	7
Corn gromwell++	Buglossoides arvensis (Li thospernium arvense)	Terbutryne 0.275 kg ha-1	6.50	1.6	6
Wild turnip++	Brassica tournefortii	MCPA 0.23 kg ha-1	1.55	1.4	5
Deadnettle++	Latnium amplexicaule	Terbutryne 0.275 kg ha-1	6.50	1.6	55
Smallflower++ fumitory Yellow	Fumaria parviflora	Terbutryne 0.275 kg ha-1	6.50	1.5	70
burrweed++	Amsinckia calycina (A. hispida)	Terbutryne 0.275 kg.ha-1	6.50	1.7	12
Wild radish+++	Raphanus raphanistrum	Metribuzin! methabenzthiazuron .105/.42 kg ha-1	13.50	2.4	10

¹Density at which the cost of control equals the value of the additional wheat yield.

* Reeves (1976) ** Gilbey (1974) + Dew (1972) ++ Wells (1979) +++ Reeves & Code, unpublished data

The key weed control issues for the next decade will be:

1. Herbicide persistence, safety and the environment.

2. Herbicide application techniques.

3. Development of integrated weed management, especially in minimum tillage and stubble retention situations.

4. Control of new problem weeds, particularly perennials.

5. Development of further data on critical weed densities.

6. Development of herbicide mixtures.

To achieve results in these key areas there needs to be a rationalisation of the few weed research resources that we have. This could be best attained by allocating, where possible, research areas to individual research establishments regardless of State and/or organisational boundaries. This would be particularly applicable to the subjects of herbicide persistence and herbicide application techniques. These important topics need considerable expensive support for sophisticated equipment and resources. It is only logical that these resources would be best located at specialised research centres. Research into tillage machinery would also be amenable to this centre specialisation.

Most weed control and competition studies are obviously localised problems requiring research to be carried out at various centres and this should continue, but when we face the major tasks for weed control research in the 1980’s, an overall plan is required where scarce resources are rationally used and parochialism ploughed in. The next few years will see increasing community pressure on pesticide usage and increasing farmer pressure on the basic resource, the soil.

Integrated farming systems that combine the sensible use of chemicals with reduced tillage, sound rotations and overall improved soil management, will ensure that Australian farmers can further increase their contribution to feeding mankind.

References

1. Dew, D.A. (1972) Can.J.Plant Sci. 52:921

2. Donaldson, T.W. & Code, G.R. (1981) Proceedings 6th Australian Weeds Conference Vol. 1 p. 79

3. Gilbey, D.J. (1974) Aust. Journal Exp.Agric.An.Husb. 14:656 PEARCE, G.A. (1981) proceedings 6th Australian Weeds Conference Vol. 2 p.53

4. Reeves, T.G. & Smith, I.S. (1975) Aust.Journal Exp.Agric. An.Husb. 15:527

5. Reeves, T.G. (1981) Paper presented to combined seminar South Aust.Crop Sci.Soc. & Weed Sci.Soc. Adelaide, May 7th, 1981

6. Snelson, J.T. (1973) Pesticides Branch, D.P.I., Canberra P.B. 185

7. Wells, G.J. (1978) Proc. 1st Conf. Council Aust.Weed Sci.Soc. p.402

8. Wells, G.J. (1979) Weed Res. 19:185

9. Reeves, T.G. & Brooke, H.D. (1977) Proc. 6th Asian Pacific Weed Sci. Soc. Conference p. 167