Schneider v Hoechst Schering Agrevo Pty Ltd [2000] FCA 154 (29 March 2000)

Last Updated: 29 March 2000

FEDERAL COURT OF AUSTRALIA

Schneider v Hoechst Schering Agrevo Pty Ltd [2000] FCA 154

PRACTICE AND PROCEDURE - cases heard together but not consolidated - evidential difficulties.

REPRESENTATIVE PROCEEDINGS - damaged wheat crops - reduced yield at harvest.

NEGLIGENCE - product liability - manufacturers and distributors of agricultural herbicide - failure to warn- causation - foreseeability - sufficiency of trials on the herbicide.

CAUSATION - a combination of unknown factors -distinction between "essential condition" and "legal cause" - "but for" test - role of expert evidence in determining causation.

Trade Practices Act 1974 (Cth) ss 52, 70, 71 and 74

Sale of Goods Act 1924 (NSW) ss 18 and 19

Fair Trading Act 1987 (NSW) s 42

McGhee v National Coal Board [1972] UKHL 7; [1973] 1 WLR 1 referred to

March v Stramare [1991] HCA 12; [1991] 171 CLR 506 followed

Bennett v Minister of Community Welfare [1992] HCA 27; [1992] 176 CLR 408 followed

Commonwealth v McLean [1997] 41 NSWLR referred to

Chappel v Hart [1998] HCA 55; [1998] 156 ALR 517 referred to

Tubemakers of Australia Ltd v Fernandez (1976) 50 ALJR 720 referred to

Dahl v Grice [1981] VR 513 applied

DAVID LOUIS SCHNEIDER v HOECHST SCHERING AGREVO PTY LTD, IAMA LTD

NG 374 of 1997

JIM MASMAN, LORETTA MASMAN, JAMES MASMAN and JON MASMAN v CASTLEREAGH MACQUARIE COUNTY COUNCIL, HOECHST SCHERING AGREVO PTY LTD

NG 931 of 1997

ELWIN JOHN EDWARDS and SCOTT ANTHONY EDWARDS v BAYEBB PTY LTD, HOECHST SCHERING AGREVO PTY LTD

NG 932 of 1997

WILLIAM LUDLOW CHAPMAN and JEAN CHAPMAN v WICKMAN RYAN PTY LTD, WESFARMERS DALGETY LTD, HOECHST SCHERING AGREVO PTY LTD

NG 77 of 1998

JUDGE: MATHEWS J

PLACE: SYDNEY

DATE: 29 MARCH 2000

IN THE FEDERAL COURT OF AUSTRALIA
NEW SOUTH WALES DISTRICT REGISTRY	NG 77 of 1998

BETWEEN:	WILLIAM LUDLOW CHAPMAN & JEAN CHAPMAN Applicants
AND:	WICKMAN RYAN PTY LTD First Respondent/First Cross-Claimant/Third Cross-Respondent
	WESFARMERS DALGETY LTD Second Respondent/Second Cross-Claimant/Third Cross-Respondent
	HOECHST SCHERING AGREVO PTY LTD Second Cross-Respondent/Third Cross-Claimant
JUDGE:	MATHEWS J
DATE OF ORDER:	29 MARCH 2000
WHERE MADE:	SYDNEY

THE COURT ORDERS THAT:

1. The Application be dismissed.

2. The cross-claims be dismissed.

3. The Applicants are to pay the Respondents' and Cross-Respondents' costs of the proceedings.

Note: Settlement and entry of orders is dealt with in Order 36 of the Federal Court Rules.

IN THE FEDERAL COURT OF AUSTRALIA
NEW SOUTH WALES DISTRICT REGISTRY	NG 374 of 1997

BETWEEN:	DAVID LOUIS SCHNEIDER Applicant
AND:	HOECHST SCHERING AGREVO PTY LTD First Respondent/Cross- Respondent/Cross-Claimant
	IAMA LTD Second Respondent/Cross-Claimant/Cross-Respondent

	NG 931 of 1997
BETWEEN:	JIM MASMAN, LORETTA MASMAN, JAMES MASMAN & JON MASMAN Applicants
AND:	CASTLEREAGH MACQUARIE COUNTY COUNCIL Respondent/Cross-Claimant
	HOECHST SCHERING AGREVO PTY LTD Cross-Respondent

	NG 932 of 1997
BETWEEN:	ELWIN JOHN EDWARDS & SCOTT ANTHONY EDWARDS Applicants
AND:	BAYEBB PTY LTD Respondent/Cross-Claimant
	HOECHST SCHERING AGREVO PTY LTD Cross-Respondent

	NG 77 of 1998
BETWEEN:	WILLIAM LUDLOW CHAPMAN & JEAN CHAPMAN Applicants
AND:	WICKMAN RYAN PTY LTD First Respondent/First Cross-Claimant/Third Cross-Respondent
	WESFARMERS DALGETY LTD Second Respondent/Second Cross-Claimant/Third Cross-Respondent
	HOECHST SCHERING AGREVO PTY LTD Second Cross-Respondent/Third Cross-Claimant

JUDGE:	MATHEWS J
DATE:	29 MARCH 2000
PLACE:	SYDNEY

REASONS FOR JUDGMENT

1 This case is about damaged wheat crops. In 1996, a number of wheat crops in the Walgett/Coonamble area of north-western New South Wales showed signs of severe stress. The yield was adversely affected at harvest. Indeed, in a few cases, harvest was not attempted as it would have been uneconomical. A common feature in relation to many of these crops was that they had been sprayed with a chemical herbicide called Puma S. One of the farmers, Mr Schneider, commenced a representative proceeding against the manufacturers of Puma S, Hoechst Schering Agrevo Pty Ltd (`Hoechst') on behalf of all affected farmers. He also commenced proceedings against IAMA Limited (`IAMA') from whom he had purchased the Puma S.

The Nature of the Proceedings

2 The proceedings are not without their complexity. Mr Schneider's representative action against Hoechst was brought on behalf of all wheat-growers in the north-west of New South Wales who claim losses arising out of the application of Puma S to their crops. There are ten identified members of this class, and an unspecified number of wheat-growers who remain unidentified. A number of heads of negligence were raised against Hoechst. In addition, it was claimed that Hoechst had engaged in misleading or deceptive conduct in contravention of s 52 of the Trade Practices Act 1974 (Cth) (`the Trade Practices Act'). The proceedings against IAMA were said to be brought on behalf of a sub-group of the represented parties, namely those who had purchased Puma S from IAMA. These proceedings are based on negligence, breach of contract, breach of implied conditions under ss 70 & 71 of the Trade Practices Act and ss 18 & 19 of the Sale of Goods Act 1924 (NSW) (`the Sale of Goods Act'), breach of an implied warranty under s 74 of the Trade Practices Act, and engaging in misleading or deceptive conduct contrary to s 52 of the Trade Practices Act and s 42 of the Fair Trading Act 1987 (NSW) (`the Fair Trading Act').

3 The matter does not end there. Both Hoechst and IAMA have filed cross-claims against each other in the proceedings commenced by Mr Schneider. IAMA has alleged that Hoechst made negligent misrepresentations as to the suitability of Puma S for its declared use, and that it contravened s 52 of the Trade Practices Act. Hoechst claims that IAMA was in breach of an implied term of the agreement between them whereby it was to provide agronomic advice to Mr Schneider and other purchasers of Puma S.

4 Six further proceedings were commenced by individual wheat-growers, being some of the represented parties in the proceedings brought by Mr Schneider against Hoechst. All of these individual proceedings were brought against the companies or organisations from which the growers had purchased Puma S. The statements of claim in all of these cases were based on the same causes of action as were raised by Mr Schneider in his action against IAMA. In each of these proceedings, the respondent distributor cross-claimed against Hoechst, which in turn cross-claimed against the distributor, again raising the same issues as those which had been raised in the Schneider case.

5 Before the matter came to hearing, the cross-claims between the distributors and Hoechst had been settled in three of the seven proceedings, namely those commenced by Messrs Schneider, Masman and Edwards. In each of those cases, Mr Harrison of Senior Counsel and Mr McNally appeared on behalf of both the manufacturer, Hoechst, and the relevant distributor. Notices of discontinuance of Hoechst's cross-claims were later filed in those proceedings. In the four remaining cases, the cross-actions between Hoechst and the distributors remained on foot, and the distributors were separately represented. Three of these cases, those commenced by Messrs Fester, Todd and Lyons, involved the same distributor, Ag-N-Vet Services Pty Ltd (`Ag-N-Vet') which was represented at the hearing by Mr Insall. In the case commenced by Mr Chapman, the distributor was represented by Mr Ryan of Senior Counsel.

6 Between the end of the hearing and the delivery of judgment, the three cases involving Ag-N-Vet, namely the actions commenced by Messrs Fester, Todd and Lyons, were settled. Accordingly, no orders will be made in relation to these cases. The fact that they have been settled does not affect the issues to be discussed in this judgment, as each of the applicants in the settled actions was a represented party in the Schneider proceeding against Hoechst.

7 All proceedings were, by consent, heard together. Indeed, given the nature of the issues involved, it would have been unthinkable to conduct separate hearings in relation to any of them. However, no application had been made before the hearing to consolidate the proceedings. As a result, the evidence adduced in some cases was not admissible in others. This caused a number of evidential and logistic difficulties during the hearing. The problems came to a head during the evidence of the plaintiff's principal expert, Professor Kennedy, who had based his overall findings upon the statements of all wheat-growers who had furnished statements in the combined proceedings. Both Mr Ryan and Mr Insall urged that these findings were not admissible against their clients as many of the statements which Professor Kennedy had used as a basis for his opinions had not been, and could not have been, tendered in evidence against them. This proposition was so manifestly correct that Mr Moore, who appeared with Mr Rowe for all the applicants, sought an order consolidating the proceedings so that all evidence could be admissible in all proceedings, thus circumventing these evidential difficulties. But by this time the hearing was well into its second week, and although an order of consolidation might well have been appropriate before the commencement of the hearing, it would have caused significant prejudice and inconvenience at that late stage. Accordingly, the application was rejected. The result is that some of the evidence in the case is available for use in some only of the proceedings. In particular, certain evidence tendered against Hoechst is inadmissible against the distributors, and much of the evidence relating to discussions between applicants and the distributors is inadmissible against Hoechst.

8 This situation gave rise to numerous well-based objections during the course of the hearing. It threatened to cause something of a logistic nightmare during the writing of this judgment, because of the constraints it imposed upon the use of certain pieces of evidence. Fortunately, these difficulties have turned out to be more apparent than real. They have not in any way affected the outcome of these proceedings.

9 Three weeks were set aside for the hearing of this matter. At the outset, Mr Rowe, on behalf of the applicants, suggested that the issues of liability and quantum might be split, and that the primary hearing should be restricted to matters relevant to liability. The respondents successfully objected to this course, and the hearing thus proceeded on all issues. However, on the last day of evidence, which was well into the third week of the hearing, several difficulties arose in relation to the foreshadowed evidence of a witness whom Hoechst proposed to call on the issue of damages. It was acknowledged that some time would be needed to rectify these problems. The only way to accommodate to this situation was to defer further evidence and submissions on the issue of damages until a later date. Accordingly, all final submissions were restricted to the question of liability. It was agreed that, should the applicants succeed on this issue, the proceedings would need to be reconvened at a later date in order to deal with damages. As it transpires, this course will not be necessary.

Outline of Judgment

10 This judgment will deal with the following matters:

* the general nature and function of Puma S;

* the evidence relating to the 1996 wheat crop of each of the eight wheat-growers whose losses have been raised in these proceedings;

* the conclusions of the various experts who gave evidence in the case;

* an outline of the trials carried out on Puma S before the 1996 season;

* outline and discussion of the legal and factual issues involved in the case.

Puma S - General Description

11 I shall be describing later the mode of operation of Puma S and the history of its manufacture and sale. For present purposes it is sufficient to say that Puma S is generally classified as a post-emergent selective herbicide. In other words, it is designed to eradicate weeds which have emerged during the growth of wheat crops, without affecting the wheat itself. It contains an active ingredient, fenoxaprop-p-ethyl (`FPE') and a `safener' fenchlorazole-ethyl (`FCE'). Both weeds and wheat crop take up the active ingredient at much the same time, which sometimes leads to a yellowing in the wheat. However, the wheat metabolises the active ingredient at a faster rate than the weeds, a process which is accelerated by the safener. The result is that FPE is detoxified within the wheat within a short time, so that no long term damage is suffered by the crop. The weeds, however, do not detoxify in this manner and Puma S is apparently regarded as very effective in the eradication of a number of weeds, particularly black oats (sometimes described as "wild oats") and winter phalaris.

12 It was the emergence of black oats amongst the 1996 wheat crops which led to the application of Puma S by most of the represented growers. Black oats, on all accounts, is a particularly virulent weed. If left unchecked, a black oats infestation can significantly reduce the yield of a wheat crop, sometimes to the extent of rendering harvest uneconomical. Attempts can be made to control the weed prior to the sowing of a crop, by applying a systemic herbicide such as Round-Up. This was done by some of the growers involved in this case, but the weed re-emerged during the growth season. When this occurs, a grower must choose between leaving the weed unchecked or spraying with a post-emergent selective herbicide such as Puma S. This chemical is, as I have indicated, generally very effective in eradicating black oats. It did so in all the crops involved in these proceedings when it was applied at the recommended rate. In general, therefore, no complaint is made that Puma S failed to meet the purpose for which it was intended. It was the allegedly adverse effect of the chemical on the wheat crop itself which gave rise to these proceedings.

13 Although Puma S is not generally considered a dangerous product to use, it is recommended that it not be sprayed on weeds or crops which are under stress. (I shall be referring later to the warnings, or "directions" on the label.) Whether Puma S damages wheat as well as weeds is one of the issues raised in this case, and it would be premature to comment on it here. But there is clear evidence that the effectiveness of the herbicide will be reduced if it is sprayed on weeds which are under stress. A failure to achieve a `kill' will generally indicate that the weeds were under stress when sprayed (assuming that the spraying was properly carried out and that the herbicide was applied at an appropriate rate). There is also evidence that if a crop is under stress, the weeds will also be stressed. Accordingly, it is a contra-indication for spraying with Puma S that the crop be under stress.

The 1996 Wheat Crop

14 Evidence of the 1996 wheat crop in the Walgett/Coonamble area was given by various wheat-growers, and by others who were in a position to observe the conditions during that year. No real issues of credibility were raised in relation to any of these witnesses. It was suggested that some might have been mistaken in relation to some of their observations, but all were generally accepted as witnesses of truth.

15 I should start by briefly saying something about the growth of wheat in the north-west of New South Wales. Wheat is extensively grown in the Walgett/Coonamble area. Being a winter crop, it is generally resistant to cold conditions, although it can be vulnerable to frost at certain points of its growth. Rainfall in the Walgett/Coonamble area is generally quite low. Accordingly, excessive soil moisture is rarely a problem, but lack of moisture sometimes can be. The soils tend to be self-mulching black soils, which are generally favourable for cereal crop growth.

16 To put the 1996 year into context: 1994 was a drought year in the Walgett/Coonamble area, and no wheat was sown by any of the represented wheat-growers. By contrast, 1995 was a good year, with generally high yields at harvest. However, some crops suffered infestations of black oats, requiring the application of Puma S during the growth stage. This was successful and the weeds were apparently eradicated with no adverse effect upon the crops themselves. Certainly no complaints were received about damage caused by Puma S in 1995, notwithstanding that it was widely used throughout the Walgett/Coonamble area.

17 On all accounts the 1996 year commenced well. Sowing by represented growers commenced on various dates between 13 May and the middle of June. However, all of them suffered an infestation of black oats in at least some of their crops. These crops, or part of them, were therefore sprayed with Puma S. In nearly every case the chemical was effective in eradicating the weeds. However, the wheat crops also appeared to be adversely affected. In all cases a lower than expected yield was obtained at harvest. This, according to the applicants, must have been caused by Puma S. The respondents deny this, and claim that the damage must have been caused by a combination of unusual and presently unexplained features of the 1996 growth season.

18 It will be necessary here to describe the events which occurred during the 1996 wheat growing season, in relation to each of the growers who gave evidence in the proceedings. I shall do so in the sequence in which they gave their evidence. I shall endeavour to refer to areas and weights by metric measure, but this is often not possible. The growers still tend to refer to `acres' not hectares, and to the wheat yield in terms of `bags to the acre'. There are twelve bags in a tonne.

Schneider: `Ashantee'

19 The first grower to give evidence was David Schneider, the applicant in the representative proceedings against Hoechst. He and his father, Louis Schneider, own and operate a property, `Ashantee', 32 kilometres south of Walgett. Wheat has been grown on `Ashantee' since about 1970, although not in drought years. For many years, the wheat growing on Ashantee has been done by way of share farming arrangement. In 1996, the share farmer was Claude Whillock. In 1995, only 350 acres of `Ashantee' was sown to wheat. Barley was sown in the balance of the available area. In 1996, Mr Whillock decided to sow wheat over the whole available area of 1,247 acres. Sowing commenced on 27 May 1996, and finished in early June.

20 On all accounts, the growing season commenced well on `Ashantee'. This was the opinion of the growers, and it was shared by Colin Saunders, an insurance agent with a great deal of experience in wheat growing. Mr Saunders' advice was regularly sought by the Schneiders and by other growers in the area. However, by mid-July it was apparent that black oats were growing up through the wheat crop. On about 22 July 1996, Mr Whillock asked Dugald Spenceley, an agronomist employed by IAMA, to have a look at the crop. It was apparent, Mr Whillock said, that in addition to black oats there was also phalaris and rape. Mr Spenceley advised that Puma S be sprayed at the rate of 600 millilitres per hectare, together with another chemical, Decoy, to combat the rape. On 24 July 1996, Mr Whillock purchased the requisite quantity of Puma S and Decoy from IAMA. Either that or the following day he commenced spraying the chemicals onto the wheat crop. Instead of applying Puma S at 600 millilitres per hectare, as advised, he reduced the rate to 470 millilitres per hectare. He did this, he said, because the chemical was very expensive, and he believed that there was a margin built into the recommended application rate to compensate for inexperienced sprayers.

21 The spraying was completed on 6 August 1996. By this time, Mr Whillock noticed that sections of the crop which had first been sprayed were showing signs of yellowing. He assumed that this was only a temporary setback. However this did not turn out to be the case. Indeed, according to Mr Saunders, the crop continued to deteriorate. Mr Saunders first observed the damage to the crop on 25 August 1996. He described it as different from anything that he had ever seen in his many years of experience with wheat crops. He was so concerned that he returned to `Ashantee' and other affected properties in the area three times between 4 and 20 September 1996.

22 The wheat crop on `Ashantee' was harvested between 12 and 23 November 1996. It yielded 2.7 bags (0.225 tonnes) per acre with a protein content of 11 per cent. Both David Schneider and Mr Whillock had anticipated, before the crop deteriorated, that it would yield at least eight bags to the acre with a minimum of 13 per cent protein. The total yield from the crop was 285.02 tonnes. In 1995, the yield from the 350 acres under wheat had been 188.02 tonnes. This, according to my calculations, represents a decrease in yield from 0.54 tonnes to an acre in 1995, to 0.25 tonnes per acre in 1996.

23 It is relevant to consider the evidence as to the nutritional content of the soil where the `Ashantee' crop was grown, and any unusual climatic features of the 1996 season. As to the soil, it was not the practice of the Schneiders to obtain regular soil analysis. However, after the failure of the 1996 wheat crop, Mr Whillock submitted a sample of soil from the "Silo" paddock, where the wheat had been grown, for analysis. The soil was said to be high in nitrate nitrogen, but low in sulphur. The comment in relation to sulphur was:

"This surface sample is low, but will assume that there is adequate sulfur at depth. If you suspect a deficiency, suggest test strips at 10 kg/ha S."

24 The zinc levels were also said to be low. On the basis of these results, an agronomist attached to Agritech Laboratory Services Pty Ltd commented:

"Sulphur levels are extremely low and any fertiliser program should incorporate sulphur."

and

"Zinc will need to be added with the planting fertiliser."

25 As to the prevailing climatic conditions, the evidence of all growers is that the factor which most affects the health of wheat crops is the moisture content of the soil, particularly inadequate moisture. Mr Saunders described the 1996 season as generally a good season, with adequate moisture and good rains in late August. Mr Louis Schneider had kept a chart of the rainfall at `Ashantee', which showed that during the months of July and August, there was a total of 247 points of rain, with a further 170 points in September and October. In 1995, which was considered a good season, only twenty-one points fell in July and August, and 323 points in September and October. Mr Whillock was questioned about this during his evidence. He said that in August and September 1996 there was insufficient rain on `Ashantee'. He was living in Toowoomba at the time and returned to `Ashantee' in early to mid October. At that time he observed the crop to be "pretty ordinary". He attributed this to stress from lack of moisture. By that time, he said, the moisture profile, which had been good at the time of sowing, had "deteriorated to an unacceptable level". (T 189) It was this which led him, later in the season, to revise downwards his estimate, for insurance purposes, of the anticipated crop yield.

26 In January 1997, following complaints by Mr Schneider, Dugald Spenceley of IAMA came and inspected the remnants of the crop at `Ashantee'. He later submitted a report, dated 9 January 1997, in which he made the following comments:

"In both paddocks there was a high incidence of Crown Rot (pink mould at base of stem).

There were also signals of frost damage. Stems were pinched and scarred around the 2^nd and 3^rd nodes. Paddocks also had very low levels of tillering, indicating the possibility of low nutrition, crown rot or chemical damage. Paddocks have been harvested which makes it very difficult to assess apparent damage."

27 Mr Louis Schneider disagreed with Mr Spenceley's comments about crown rot. As he put it: "we never had crown rot before and we've never had it since". (T 158)

28 Mr Schneider's complaint led to correspondence with Hoechst. On 11 March 1997 Mr John Ashby, Hoechst's Field Development Manager, wrote to Mr Schneider and Mr Whillock in the following terms:-

"1996 Wheat Crop - Yield

Dear Messrs Schneider & Willock,

We thank you for the query which you have made regarding the alleged yield reductions on your property in the 1996 winter cereal season.

As you are undoubtedly aware, AgrEvo has received a small number of other queries from your area regarding similar instances. As our company is zealous in its environmental policy and always anxious to assist its customers, we have conducted extensive investigations in the area to attempt to ascertain the cause of such instances. Such investigations have included obtaining reports from distinguished consultants. We have now received a final report from the last of these consultants. We have correlated it to all the other information received by us.

From the information received, it has become apparent that the alleged crop effects you experienced cannot be attributed to Puma S but were caused by an interaction of climatic, agronomic and geographical variables. The sum of the reports is that there appears to have been an interaction of low nutrient status in the soil, frosts and low moisture levels at or following application of products to the crops with some incidence of various diseases. Such interaction has apparently affected the vegetative growth in the manner exhibited on your property.

The interaction of products with such factors as frosts and moisture stress following application cannot be predicted and are beyond control. However, the effects from such events can be minimised by adhering strictly to label statements of correct timing, correct application and the minimisation of other stress components such as crop nutrition and disease.

Many years of experience with Puma S (both in trials, as part of product development, and commercial usage) have demonstrated the high level of selectivity of Puma S in wheat. These have included comprehensive crop screening trials by both private consultants and State Departments of Agriculture which have evaluated Puma S for crop damage ratings and yield. Such trials have been conducted over a wide range of wheat varieties, over numerous climatic conditions and at application rates up to 1.4 L/ha (ie twice the maximum recommended label rate). The sole symptom arising from the application of Puma S, which is well within agronomic knowledge, is an occasional, minor, transient yellowing up to four weeks following application. Such yellowing quickly grows out and does not significantly affect yield in what are weed free sites.

We therefore conclude from the information in my possession that the alleged crop as exhibited at your property is atypical given all previous experience with Puma S (fenoxaprop-p-ethyl). Such experience extends to other products within the same chemical group (fops/dims - clodinafop, tralkoxydim, diclofop, etc.) Under these circumstances, our Company cannot accept any liability for the damage allegedly suffered by you.

We thank you for your query and suggest that in the future you follow the STAR program (enclosed) and adhere to label recommendations.

If we can be of any further assistance please contact me.

Yours sincerely,

JOHN F ASHBY

Field Development Manager"

29 The partners in `Ashantee' and Mr Saunders were all cross-examined about the possible effects of frost on the 1996 `Ashantee' crop. There is evidence that a severe frost at the time of flowering can significantly set back a wheat crop and reduce its yield. In the case of `Ashantee', this would have occurred in about mid to late September. Both David and Louis Schneider denied that there was any frost at that time which could have affected the crop. In any event, Mr Saunders said that the damage to the crop did not resemble any frost damage that he had ever observed. Mr Saunders commented that frost gets wrongly blamed for a great deal of crop damage. Wheat being a winter crop is not, he said, highly susceptible to frost.

Munns: `The Ridges' and `Pollieboomi'

30 Peter Geoffrey Munns is part owner of `The Ridges' and `Pollieboomi', near Baradine in north-west New South Wales. Mr Munns is one of the represented growers in Mr Schneider's proceedings against Hoechst. He is not involved in any individual proceedings against distributors. Mr Munns presents as a progressive, astute farmer who has been concerned for some time to observe best practice in order to obtain optimum yield from his crops. In 1996, he planted 485 hectares or 1,198 acres to wheat over both properties. One hundred and thirty-five acres of this were in the `David Dell' paddock in `The Ridges'. This paddock had not been cropped since 1993. In 1994 it was grazed with livestock. In 1995 it was intermittently grazed between chemical spraying. On three occasions between the spring of 1995 and March 1996, `David Dell' was sprayed with Round-Up and on two occasions with 24 D Ester. Before sowing in 1996, the paddock was pre-planted with Urea, a nitrogen-based fertiliser. On 16 May, the paddock was sown with wheat. In June 1996, Mr Munns noticed a high infestation of black oats in the `David Dell' paddock, together with some phalaris, turnip weed, and thistle. He discussed his problem with Mr Jackson of Coonabarabran Fertiliser Distributors, and they agreed that the crop should be sprayed with a combination of Puma S at a rate of 500 millilitres per hectare, and Lontrol at 300 millilitres per hectare. The spraying took place on 25 June 1996.

31 A few days after spraying, Mr Munns noticed that the wheat colour had changed from green to yellow and there was a noticeable lack of growth. A few days later, on 1 July 1996, he observed what he described as a blue oat mite problem. Blue oat mites are insects, unrelated to black oats, which cause wheat to become stressed and the leaves to turn yellow. Left untreated, they can lead to a yield loss of up to 80 per cent of the crop. Mr Munns immediately contacted Mr Jackson and the same day the crop was sprayed with a chemical called Strike Out. The blue oat mites were eradicated within approximately twenty-four hours. However, the crop did not recover as Mr Munns had anticipated it would. He described it as looking anaemic, stunted in its growth, and very thin on the ground. Mr Munns was unable to locate the cause of the problem. He checked the soil moisture with a moisture probe, and found "an excellent amount of sub-soil moisture". (T 227) In September, he decided to strategically graze the crop, and he placed twenty sheep per acre into `David Dell' for three days. His experience was that this aggressive grazing often helped a struggling crop to recover. On this occasion the wheat crop made a slow partial recovery after the grazing but still provided an inadequate yield.

32 The wheat harvest took place in December 1996. `David Dell' yielded 3.4 bags (0.28 tonnes) per acre on Mr Munns' crop. The previous yield had been 10.2 bags (0.85 tonnes) per acre. A nearby paddock, `Trudi Dell', yielded 14.5 bags (1.2 tonnes) per acre. `Trudi Dell' had similar soil composition to `David Dell' and had received much the same treatment during 1996, including being subject to a blue oat mite infestation which had been treated with Strike Out. However, it had not been sprayed with Puma S.

33 1996 was not the first year that Mr Munns had used a selective post-emergent fertiliser on his crops. He had previously sprayed with Puma S or its predecessor Puma, without any ill effects.

Todd: `Combogolong Woolshed' and `Kimo'

34 Dennis Todd owns four properties in the Walgett/Coonamble area, including `Combogolong Woolshed' and `Kimo'. These two properties are approximately 35 kilometres south of Walgett, and are near `Karuka' (Mr Lyons' property). In 1996, Mr Todd employed Alexander Hewson and Lesley Bruce to work on his properties. In addition to their wages, they and their wives received a percentage of the proceeds from the crop.

35 In 1996, four of Mr Todd's paddocks were planted to wheat. On `Combogolong Lake' paddock, which was on `Combogolong Woolshed', approximately 1,000 acres were assigned to wheat. The remaining three paddocks were on `Kimo'. They were:

* the `Kimo Lake' paddock -- approximately 171 hectares or 425 acres;

* the `Brigalow' paddock -- approximately 405 hectares or 1,000 acres; and

* the `Ridge' paddock -- approximately 141 hectares or 350 acres.

36 Sowing took place between 13 and 19 May 1996. Earlier in May, all paddocks had been sprayed with systemic herbicides being either Round-Up or a combination of Gleam and Round-Up.

37 According to Mr Todd, the germination was excellent and the crop came away vigorously. However, in the middle of June he noticed that all paddocks, apart from the `Ridge', had black oats in them to a varying extent. He contacted Ian Elliott, an agronomist employed by Ag-N-Vet at Coonamble, who agreed to inspect his crop. On 16 July 1996, Mr Elliott and Mr Bruce inspected the various paddocks. Mr Elliott found a combination of phalaris and turnip weed in the `Kimo Lake' paddock, and a high level of black oats in the other paddocks, apart from the `Ridge' paddock. He telephoned Mr Todd and recommended that the paddocks be sprayed with Puma S at rates varying between 600 and 700 millilitres per hectare, together with a herbicide called Eclipse in the `Kimo Lake' paddock.

38 Spraying duly commenced on 17 July. The `Kimo Lake' Paddock was sprayed as recommended. During the spraying of the `Brigalow Paddock', the rigs ran out of Puma S. Mr Todd attempted to purchase more of the herbicide, but was told by Mr Elliott that he had run out. Instead, he was supplied with a herbicide called Topik. I shall be saying more about this herbicide later. Suffice it to say at this stage that, like Puma S, it is a post-emergent selective herbicide. It is manufactured by a competitor of Hoechst and is more expensive than Puma S.

39 Spraying was completed on 20 July 1996. Neither Mr Todd nor Mr Bruce know which part of the `Brigalow' or `Combogolong Lake' paddocks were sprayed with Puma S, and which parts were sprayed with Topik.

40 Within about ten days it was evident that the application of Puma S and Topik had effectively eradicated the black oats. However, the wheat had also been knocked back. Damage was apparent in the three paddocks which had been sprayed with the chemicals. There was no observable difference between areas sprayed with Puma S and those sprayed with Topik. Some peripheral areas had been sprayed with neither, as the spraying rigs could not fit around trees or telegraph poles. In these areas the wheat was healthy in appearance and, in contrast with the sprayed areas, the wheat heads, when they grew, were found to be full of grain. The situation was so bad in the `Combogolong Lake' paddock that Mr Todd decided to cut the wheat for hay. Although this was not his usual practice, he regarded it as the most economically viable way of dealing with this crop. Wheat must be cut for hay before it ripens, and Mr Todd had a window of only a few days in which to do this. He managed to cut and bale 350 acres of `Combogolong Lake' paddock between 6 and 11 November 1996. This resulted in 450 bales of hay.

41 The harvesting of the rest of the wheat crop commenced in the second week in November. The `Ridge' paddock yielded about eight or nine bags (about 0.7 tonnes) per acre with a protein level of approximately 17 per cent. The remaining 650 acres of `Combogolong Lake' yielded one and a half bags (about 0.125 tonnes) per acre with a protein level of 11 per cent. A decision was made to harvest only about one quarter of the `Kimo Lake' paddock as the wheat in the remainder had virtually no grain in the head. The quarter of the paddock which was harvested yielded less than one quarter of a bag (0.2 tonnes) per acre with a protein level of approximately 12 per cent. Mr Todd's evidence did not indicate the yield of the `Brigalow' paddock. However, Mr Bruce said that the combined yield of the three paddocks sprayed with Puma S averaged between one and one and a half bags (approximately 0.104 tonnes) per acre. Left unharvested, at Mr Todd's direction, were strips of both the `Combogolong Lake' and `Brigalow' paddocks which had not been sprayed with Puma S or Topik and which were visibly thicker and healthier than the remaining crop.

42 For the `Ridge' paddock to have a higher yield than the other three paddocks was, according to Mr Todd, one of the anomalies of the 1996 season. Part of the paddock had been scalded by wind erosion, and it generally produced considerably lower yields than the other paddocks. Before 1996, the yield of the `Ridge' paddock had, according to Mr Todd, averaged approximately two thirds of the yield of the other paddocks on a per acre basis.

43 On 20 December 1996, following complaints by Mr Todd, a representative of Hoechst, Denis Tomlinson, visited and inspected his properties. Mr Saunders was also present. Mr Tomlinson has a Bachelor's degree in Agricultural Science, and a background in the growth of cereal crops. His general observations were as follows. In the `Brigalow' paddock he could find no distinction between the areas apparently sprayed with Puma S and Topik. However, there was a significant difference between the sprayed and unsprayed areas. The number of tillers (stems per plant) was much the same in both areas. However, the number of grains in the heads of the plants differed significantly between the sprayed and unsprayed areas. The unsprayed areas had full heads of grain, but in the sprayed areas Mr Tomlinson described a typical plant as having one or two tillers with normal grain whilst the remaining five or six tillers produced normal looking heads but they contained no grain. Mr Tomlinson observed a moderate infection of crown rot and some head deformation which he thought to be the result of frosting. In the `Kimo Lake' paddock, Mr Tomlinson observed greater tiller density in the unsprayed areas compared with the sprayed areas. There was some frost damage, he said, in the wheat heads in the unsprayed areas. In the sprayed areas, 90 per cent of the heads had no grain in them. Finally, in the `Combogolong Lake' paddock, Mr Tomlinson observed that the wheat heads in the sprayed areas looked healthy and normal but contained no grain. The unharvested part of the crop, he said, had very severe root or crown disease. Mr Saunders, who was present during this inspection, did not agree with Mr Tomlinson's assessment as to infestations of crown rot.

44 Mr Todd described the 1996 wheat growing season as generally a good season. He said that on 29 August 1996 about two inches of rain fell on his properties and that there was excellent soil moisture during the crucial flowering period. As to frosts, Mr Bruce conceded that there had been some severe frosts earlier in the year, but said that these were too early to cause the damage later observed in the crop. Both he and Mr Todd denied that there had been any severe frosts during or after the flowering period, which is the time when wheat plants are most susceptible to frost damage.

45 1996 was not the first year in which Mr Todd had used Puma S. In 1993, part of his `Kimo Lake' paddock was sprayed with Puma S and later produced a lower than anticipated yield. Mr Todd attributed this to the effect of the black oats on the crop before spraying. He denied that he had been alerted by this incident to the possibility that Puma S might have an adverse effect on wheat crops. In the 1995 season, according to Mr Bruce, black oats appeared in part of the wheat crop. Puma S was applied to those areas. The chemical was effective in eliminating the weed and did not appear, according to Mr Bruce, to have any adverse effect on the crop itself.

Lyons: `Karuka'

46 Thomas Lyons lives at Gulargambone and owns, in partnership with other family members, a 10,000 acre property `Karuka'. `Karuka' is thirty-five to forty kilometres south of Walgett and is next to `Ashantee'.

47 In preparation for sowing his 1996 wheat crop, Mr Lyons arranged for Round-Up, Garlon and Ester to be applied to 1,370 acres where the previous year's wheat crop had been grown. Later, on 22 and 28 May, he resprayed much of the area with Round-Up. This was done on the advice of Ag-N-Vet's agronomist, Ian Elliott. Over the years, Mr Lyons had come to rely on Mr Elliott's advice, particularly in relation to chemical selection and application.

48 In early June 1996, 1,460 acres of `Karuka' were sown to wheat. Mr Lyons also arranged for Urea, a nitrogen-based fertiliser, to be introduced with the seed in a band across a paddock known as `Creswick'.

49 Mr Lyons' wheat crop started strongly. Mr Saunders, who visited `Karuka' on 3 July 1996, commented that the crop was ahead of other crops in the area. Indeed, he described 700 acres of the `Karuka' crop as `magnificent'.

50 However, about two weeks later, Mr Lyons observed that there was a noticeable infestation of black oats and some phalaris in the wheat crop. He telephoned Ian Elliott who, in the company of Dennis Tomlinson, came and inspected the crop. Mr Elliott's assessment was that the black oats infestation was so substantial that, if left untreated, the damage to the crop would probably render it unharvestable. He recommended to Mr Lyons that he apply Puma S to all affected paddocks at a rate of 600 millilitres per hectare, together with Ally and Agral for the other weeds. Spraying took place on 17 July in accordance with Mr Elliott's recommendations. In all, 1,200 acres were sprayed. Two paddocks were left unsprayed: `Creswick E' which was more healthy than the other areas, and `Creswick C', a small area of forty acres across a bore drain from the remaining areas.

51 About the end of July Mr Lyons noticed a slight yellowing of the crop. He assumed that this was the immediate after-effect of the chemical and that the crop would shortly recover. However, by 10 August he considered that the crop had, if anything, deteriorated. It had the appearance of a crop which was under moisture stress or suffering from lack of nutrition. However, he said that there had been good falls of rain and the strip on `Creswick' where Urea had been applied was no different from the rest.

52 In mid-September 1996, Ian Elliott returned to `Karuka' and inspected the wheat crop in the presence of Thomas Lyons' son, Adrian. Mr Elliott observed that the appearance of the crop varied according to its cropping history. Areas which had been cropped over a long period appeared to be more affected than newer areas. This, he said, suggested some nutritional disorder. There was also a question as to whether moisture stress or other adverse factors might have affected the crop. According to Mr Elliott, the possibility that Puma S might have caused the problem did not occur to him, nor was it raised by Mr Lyons as a possibility.

53 One feature which both Mr Lyons and his son Adrian commented upon was that in an area of approximately 150 acres which had not been sprayed with Round-Up in May, a considerable amount of self-sown or "volunteer" wheat had grown up, having been seeded by the previous year's grain. At the time it was sprayed with Puma S this self-sown wheat was in advance of the sown wheat by approximately twelve to sixteen days. This volunteer wheat did not exhibit the same damage as the sown wheat, and remained generally healthy.

54 At the beginning of November 1996 Adrian Lyons contacted Graham Callaghan, an independent agronomist based at Coonamble. Mr Lyons was concerned to ascertain what was wrong with the crop at `Karuka'. On 7 November Adrian Lyons and Mr Callaghan inspected the crop at various points. They noted that the self-sown wheat had performed to expectations. The sown wheat, by comparison, was spindly and retarded and there were very few grains in the head. Mr Lyons pointed out the 40 acres of `Creswick C', across the bore drain, where the plants were normal and healthy. He asked Mr Callaghan to explain the difference. Mr Callaghan said: "I reckon that the frost has got 50 to 60% of the crop but here there are some scattered trees and maybe they kept the frost off. Because you sowed the whole crop late it was more susceptible to frost".

55 Mr Callaghan did not give evidence in the proceedings. Nor did another agronomist, Ian Herbert, who was described as the "district agronomist", who also inspected the crop at `Karuka' during September 1996.

56 Harvest took place between 15 and 21 November. The grain from the 1,200 acres sprayed with Puma S was collected together with that from 220 acres of the unsprayed areas, and no differential record was kept as to the yield of each. Some of the sprayed areas were not harvested as there was no grain in the heads. The total yield from the 1420 acres was 268 tonnes. This, averaged over the whole area, represented a yield of 0.19 tonnes per acre with a protein level of 12.5 per cent. The yield from the forty acres on the other side of the bore drain was, according to Adrian Lyons, forty tonnes, the equivalent of one tonne per acre, with a protein level of 13 per cent. Although the matter is not beyond dispute, I accept that this area was not sprayed with Puma S.

57 Puma S does not appear to have been used on `Karuka' before 1996. However, since then Wildcat, Puma S's successor, has been used on the property, without any apparent ill effects upon the crop.

Fester: `Merriwee' and `Redbank'

58 `Merriwee' and `Redbank' are adjoining properties, approximately fifty kilometres north of Coonamble and thirty to forty kilometres to the south-east of Mr Lyons' `Karuka'. `Merriwee' has been owned by the Fester family since 1977; `Redbank' was leased by them in 1990. Neil Fester and his family have been involved in wheat-growing in the area for decades. In 1996, four paddocks in the two properties were sown to wheat. They were:

* `Tracy Farm' paddock - 278.43 hectares or 688 acres;

* `Tammy' paddock - 275.2 hectares or 680 acres;

* `Creek 2' paddock - 250 hectares or 617 acres; and

* `Oaklands' paddock - 202 hectares or 499 acres.

59 The wheat was sown on the first two paddocks between 16 and 24 May 1996 and on the other paddocks a little later. No herbicide had been sprayed preparatory to sowing. Despite a history of black oats in the area, Mr Fester had decided to treat them, if they emerged, by spraying after sowing with a post-emergent selective herbicide.

60 In about the middle of June 1996, Mr Fester noticed black oats and other weeds in the crop on `Tracey Farm'. He contacted Kevin Dickey of Ag-N-Vet in Coonamble, who recommended that the crop be sprayed with Puma S at a rate of 500 millilitres per hectare, combined with Ally for the other weeds. In late June 1996, Mr Fester's wife Dianne sprayed the `Tracey Farm' paddock with Puma S. However, because Puma S was an expensive chemical, Mr Fester decided to reduce its rate of application to 450 millilitres per hectare. He then made a mathematical miscalculation, which resulted in a rate of spray of only 360 millilitres per hectare.

61 In July 1996 Mr Fester observed an infestation of black oats in the `Tammy' paddock. In the second half of July he sprayed this paddock with Puma S at the recommended rate of 500 millilitres per hectare, together with Ally. However, he had only enough Puma S to spray 500 acres, and left the remaining 180 acres unsprayed.

62 There is some uncertainty as to the subsequent appearance of the crops in the two sprayed paddocks. Mr Fester's father, Arthur Fester, who has been involved in wheat farming for nearly six decades, described damage to the plants in both paddocks. It was not similar to any frost damage he had ever seen, he said. Indeed it was not like any damage that he had seen in his many years of wheat-growing. On the other hand, Neil Fester said in evidence that the wheat in both paddocks looked "all right", although the plants contained little leaf, and that it was only at harvest that he realised that there was a serious problem.

63 Harvest on the `Tracey Farm' paddock commenced in the first week in November. However, it immediately became obvious to Mr Fester that there were more black oats than wheat grains going into the bin. Not only had the black oats not been controlled by the spray, but the heads of the wheat crop had not filled out. He cancelled the remainder of the harvest on that paddock. He telephoned Ian Elliott at Ag-N-Vet in Coonamble and asked him to come and inspect the crop. However, there was a delay before Mr Elliott arrived and in the meantime Mr Fester commenced harvesting the crop on the `Tammy' paddock. The 180 acres of this paddock which had not been sprayed with Puma S was not harvested as the black oats had choked out the wheat crop. The remaining 500 acres of the `Tammy' paddock' yielded approximately three-quarters of a bag (0.06 tonnes) per acre. Mr Fester had anticipated that both paddocks would yield six bags (0.5 tonnes) per acre, this being the average for his property. Subsequently, the `Creek Two' paddock was harvested and yielded, as anticipated, six bags (0.5 tonnes) per acre. The `Oaklands' paddock, where there was poorer soil and where a lower yield was anticipated, produced four bags (0.33 tonnes) per acre.

64 On about 12 November 1996 Ian Elliott and Denis Tomlinson came and inspected the `Tracey Farm' paddock. They observed a significant infestation of black oats. Mr Fester's complaint at that time was that the spray had been ineffective in dealing with the weeds. They did not inspect the `Tammy' paddock.

65 Mr Fester said that there were no adverse frosts in 1996 and the soil had good moisture content. He had used Puma S's predecessor, Puma, in previous years with no ill effects upon his wheat crop.

Edwards: `West Common'

66 Alwyn Edwards, in partnership with other family members, has a ninety-nine year lease over `West Common', a 150 hectare property about two kilometres west of Coonamble. On 21 May 1996, 370 acres on `West Common' was sown to wheat. Mr Edwards described the moisture profile as `excellent' at the time of sowing, as there had been good rains through the first five months of the year.

67 Mr Edwards first noticed black oats in his crop at the end of June 1996. He took no action, hoping that the wheat would outgrow the black oats as had happened in previous years. However, by August it was apparent that the black oats infestation was increasing to the extent that it was necessary to seek advice as to their control. He arranged with Darren Fitzgerald, the agronomist employed by Dalgety's in Coonamble, to inspect his crop. Mr Fitzgerald observed that Mr Edwards' wheat crop was infested with wild oats, turnip weed, wire weed, and Paterson's curse. He assessed that the crop was not under stress, despite the weed infestation. He recommended to Mr Edwards that he spray the crop with Puma S at the rate of 500 millilitres per hectare combined with Ally for the other weeds.

68 Spraying took place on 11 August 1996 in accordance with Mr Fitzgerald's advice. However, only 340 acres was sprayed with Puma S. Mr Edwards elected to leave the additional 30 acres unsprayed rather than pay the considerable cost of a further twenty litres of the chemical.

69 Later in August, Mr Edwards noticed a serious deterioration in his wheat crop. It was yellow, he said, and parts of the plants were dying. He asked Mr Fitzgerald to inspect the crop which he subsequently did. Mr Lyons said that Mr Fitzgerald told him that the crop was suffering from "yellow leaf spot", and that a bit of rain would bring the crop back. Mr Fitzgerald denied that this conversation took place. Indeed, he said that he did not see Mr Edwards' crop at all between spraying and harvest.

70 In spite of subsequent rains, the crop did not improve. Harvest took place on 15 and 16 November 1996. Mr Edwards arranged for the grain from the unsprayed thirty acres to be collected separately from that of the sprayed areas. The thirty acres yielded 1.96 tonnes to the hectare (0.793 tonnes to the acre) with a protein level of 13.7 per cent. The sprayed area yielded 0.7 tonnes to the hectare (0.283 tonnes to the acre) with a protein level of 12.7 per cent. Mr Edwards had predicted that the crop would produce approximately eight or nine bags (approximately 0.7 tonnes) per acre. The 1995 yield had been 0.639 tonnes per acre.

71 Mr Edwards had not used Puma S before 1996. There is no evidence of his having used it or any similar chemicals since.

Masman: `Combogolong'

72 `Combogolong' is a 12,000 acre property approximately 50 kilometres south of Walgett and is owned by members of the Masman family. Wheat cropping on the property has been carried out over a long period, increasing from 300 acres in 1976 to 3,000 acres in 1997.

73 Preparatory to the sowing of the 1996 wheat crop, the relevant areas were sprayed with Round-Up and Glean. Four paddocks were then sown to wheat in late May 1996. They were:

* `One Thousand Acre' paddock - 400 hectares or 1,000 acres;

* `Plain' paddock - 277 hectares or 685 acres;

* `Over the River' paddock - 80 hectares or 200 acres; and

* `New' paddock.

74 Towards the end of July the principal grower, Jim Masman, observed that there was an unusually large weed infestation in his crop. This consisted mainly of phalaris with some black oats. He telephoned Ian Kelly, an agronomist with the Castlereagh Macquarie County Council. Mr Kelly came to `Combogolong' in early August and inspected the crop. He recommended, in relation to the `Thousand Acre' paddock and the `Plain' paddock, that Puma S be applied at the rate of 700 millilitres per hectare.

75 On 9 August 1996, Puma S was sprayed at the recommended rate on both paddocks. The weed infestation in the `New' paddock did not merit spraying. Mr Masman had not shown the `Over the River' paddock to Mr Kelly as he regarded the cost of treating its infestation as outweighing any expected benefits.

76 Not long after spraying, two inches of rain fell. Mr Masman went to inspect the crop and was surprised to find it looking yellow and unhealthy. It did not improve, and in October Mr Masman decided that the wheat in the two sprayed paddocks was not worth harvesting. This was because the heads of wheat contained either no grain or very little grain. This was in contrast to unsprayed areas, around bore drain and trees, where Mr Masman observed that the wheat had grown normally.

77 As to the two unsprayed paddocks, a number of head of cattle later got into the `New' paddock. By the time Mr Masman realised that this had happened, the crop had been so extensively damaged as to render it not worth harvesting. Accordingly, the only unsprayed paddock which was harvested was the `Over the River' paddock. This paddock had experienced a yield loss through competition from black oats and phalaris. It produced an average of 6.6 bags (0.55 tonnes) per acre.

78 Mr Masman made a number of telephone calls to Denis Tomlinson of Hoechst about the state of his crop. On 20 December 1996, Mr Tomlinson came to `Combogolong'. He observed that in the sprayed paddocks, approximately 50 per cent of the wheat heads contained no grain, and others had as few as one or two grains per head. In the `Over the River' paddocks, there was also a deficiency of grain in the heads but not as severe as in the sprayed paddocks. Mr Tomlinson did not give Mr Masman his opinion as to the cause of the problem. However in a subsequent letter dated 11 March 1997 from Hoechst to Mr Masman it was suggested that the damage to his crop was caused by an interaction of climatic, agronomic and geographical variables. The letter continued:

"there appears to have been an interaction of low nutrient status in the soil, frosts and low moisture levels at or following application of products to the crop with some incidence of various diseases. Such interaction has apparently affected the vegetative growth in the manner exhibited on your property".

79 In the Masman case the respondent adduced evidence from neighbouring wheat-growers as to their crop history in 1996. Their experiences were quite different from those of Mr Masman. The first witness was Doug Lyons, the manager of two properties in the Combogolong area. At the recommendation of Ian Elliott of Ag-N-Vet, Mr Lyons had sprayed extensively with Puma S following an infestation of black oats. Subsequently, adverse climatic conditions, presumably frost, caused considerable damage to his crops. The unsprayed areas, according to Mr Lyons were worse affected than the sprayed areas. The problem did not manifest itself in any visual difference between the crops. However a close inspection revealed that there was no grain in the wheat heads in certain areas.

80 The other wheat-grower, Scott Harris, owns and manages three properties in the Combogolong area, two of them adjacent to the Masman property. In 1996, he sowed approximately 11,000 acres of wheat. As a result of later infestations of black oats he sprayed approximately 4,000 acres of his crop with Puma S. He observed no subsequent damage to the wheat crop apart from the "normal temporary setback that occurs after spraying". In one paddock of about 900 acres only half was sprayed with Puma S as the other half had fewer black oats. At harvest, the yield from the sprayed half was approximately double that of the unsprayed half.

81 Mr Harris described a generally low wheat yield in 1996. He said that this had nothing to do with the application of Puma S. One paddock which was adjacent to the Masman property yielded virtually no wheat at all. It emerged in cross-examination that this crop had been sprayed with Puma S. Nevertheless he appeared to be asserting that the problems with this crop were not related to the chemical, but to climatic conditions of the 1996 season.

Chapman: `Capelle'

82 `Capelle' is owned by William Chapman and his wife Jean. It is a 4,500 acre property about 25 kilometres south of Walgett. It has been used for grain production since its acquisition by the Chapmans in 1976. Mr Chapman presented as one of the most thorough, progressive and astute of the growers who gave evidence before me. In 1994, he and fifteen other landowners formed the Walgett Sustainable Agricultural Group Trust which secured the employment of an agronomist, Greg Rummery. Since then he has devised and implemented a "total farm plan" which incorporates a program for the management of weeds, disease and pests. Commencing in 1994, Mr Chapman used Puma S to control infestations of black oats. Until 1996, he experienced no adverse effects from the chemical.

83 In 1996, three paddocks on `Capelle' were sown to wheat. They were:

* `Number 3' paddock - 440 hectares or 1,090 acres;

* `Number 5' paddock - 160 hectares or 400 acres; and

* `Letterbox Numbers 1 and 2' - 400 hectares or 1,000 acres.

Sowing commenced in the last week of May and continued until early June. According to Mr Chapman the moisture profile at that time was good.

84 In early July 1996, Mr Chapman observed black oats in part of his crop. Mr Rummery recommended that they be sprayed with Puma S. On 19 and 20 July 1996, Puma S was applied to half of `Number 3' paddock at the rate of 500 millilitres per hectare. Another broadleaf herbicide was applied at much the same time. On about the 15 August 1996, the other half of `Number 3' paddock was sprayed with Puma S at the same rate.

85 The Puma S used on `Capelle' was purchased from Wesfarmers Dalgety at Walgett. There is a real issue as to whether Mr and Mrs Chapman were the `purchasers' or the `consumers' of the chemical, given that the parties to the transaction were Wesfarmers Dalgety and the Walgett Tyre Service, of which Mr Chapman was the proprietor. The resolution of that issue, however, is unnecessary here.

86 When Mr Chapman inspected his crop in late September or early October he found that some of the heads were small and stunted, there was poor tillering on the plants and the stems appeared pinched. The crop did not improve. Harvest took place between 7 and 24 November 1996. Mr Chapman described the yields as "disastrous", except for `Number 5' paddock. This paddock yielded an average of seven bags (0.58 tonnes) to the acre with a protein level of approximately 16 per cent. The remaining paddocks, `Number 3' and the `Letterbox' paddocks, yielded between them 2.7 bags (0.225 tonnes) to the acre. Mr Chapman did not keep separate records for these paddocks, but his observation was that the yield from `Letterbox' was better than that from `Number 3' paddock. However, in evidence he described the yield from all paddocks apart from `Number 5' as very bad; there was little difference between them, he said.

87 After the completion of the harvest Mr Chapman telephoned Peter Smith, the merchandise manager of Wesfarmers Dalgety, and complained about the yield loss suffered by his crop. Mr Smith arranged for Michael Skipper to inspect the crop. Mr Skipper is an accounts manager employed by Hoechst with extensive experience in the agriculture business. On 16 December 1996 Mr Skipper attended `Capelle' with Mr Chapman and the agronomist Greg Rummery. The first paddocks they inspected were the `Letterbox' paddocks. According to Mr Skipper, Mr Chapman told him that `Letterbox No2' had been sprayed with Puma S during the 1996 season. The paddock had already been harvested, but Mr Skipper noticed a "vast difference" between the stubble in the two `Letterbox' paddocks. The remnants of the crop in `Letterbox Number 2' were much thinner than those in `Letterbox Number 1'. Mr Skipper then went on to the `Number 3' and `Number 5' paddocks. However there were no unharvested plants and he was unable to make any meaningful observations about the state of the crop in these paddocks.

88 Mr Chapman, in a statement in reply, conceded the correctness of Mr Skipper's account of their visit to the `Letterbox' paddocks. He was cross-examined about this by Mr Ryan. It was suggested to him that neither of the `Letterbox' paddocks had been sprayed with Puma S in 1996. Mr Chapman said that he thought that part of these paddocks had been sprayed that year. However when shown his records for 1996 he conceded that this could not have been correct. He also agreed that there was, as Mr Skipper described it, a "dramatic visual difference" between the stubble in the two `Letterbox' paddocks. It was this which had led him to assume, wrongly as it transpires, that `Letterbox Number 2' had been sprayed with Puma S. Interestingly, Mr Chapman said that part of the `Number 3' paddock had been sprayed with Puma S in 1995, as had the whole of another paddock, `Number 1', with no ill effects to the crop.

89 Mr Chapman acknowledged that there was a drop in yield between 1995 and 1996 throughout his property. By way of illustration, paddock `Number 1', which was sprayed with Puma S in 1995, yielded 1.05 tonnes per hectare that year, compared to 0.41 tonnes in 1996. In paddock `Number 3' the yield dropped from 3.54 tonnes per hectare in 1995 to 0.39 tonnes in 1996. In paddock `Number 5' the yield dropped from 3.5 tonnes per hectare in 1995 to 1.36 tonnes in 1996. Finally, in `Letterbox Number 1', being the only `Letterbox' paddock under wheat in 1995, the yield dropped from 2.99 tonnes per hectare in that year to 0.41 tonnes in 1996.

90 Mr Chapman was questioned about this. He said that it was accepted throughout the district that the 1996 yield would be substantially less than the previous year. When asked to elaborate he said:

"Well it all relates to moisture content, whether it would be available as stored moisture and/or what was precipitation. So, I mean, we have methods which we use to determine what we consider is good potential or lesser potential by the moisture availability by what's stored in the ground, and obviously what is received in the form of rain enhances your potential from then on."(T 605)

91 Later in his evidence Mr Chapman confirmed that all wheat-growers in the area had the same experience over these two years.

92 Mr Chapman was unusual amongst the represented growers in that the soil from his cropping paddocks was regularly submitted for analysis. Tests conducted in February 1997 showed that none of the relevant paddocks was deficient in nitrogen or sulphur. Similar results had been obtained in previous years.

93 Mr Chapman was questioned about the effect of frost upon wheat crops. He said that the most vulnerable time is at flowering. There is a period of two or three days when the head is emerging when frosts can seriously affect a crop's yield. Until that time, wheat plants are not particularly vulnerable to frost. One of the purposes of late planting is to reduce the chance of frosts occurring during those vulnerable few days. The formation of the head normally occurs, he says, between the last week of August and the second week of September. It requires a temperature of minus three degrees, approximately three inches above the ground surface, to cause any severe damage. He expressed the same view as Mr Saunders, namely that frost is often wrongly blamed as contributing to yield loss. He said that to his knowledge there were no severe frosts which could have affected the 1996 crop.

What caused the damage? The experts' view

94 It is indisputable that significant damage was sustained by the wheat crops of each of the growers who gave evidence. The real issue is the role, if any, played by Puma S in causing the damage. This was the subject of considerable evidence from experts called on behalf of both the applicants and the respondents.

95 The applicants obtained reports from Professor Ivan Kennedy, Professor in Agricultural and Environmental Chemistry at the University of Sydney. Professor Kennedy's background and experience amply qualified him to express opinions on all aspects of this matter. Professor Kennedy did not suggest that Puma S was the sole cause of damage suffered by the wheat farmers. Such a contention could not be sustained, given the evidence, which I shall mention later, of the very extensive use of Puma S on wheat crops in the area compared to the small number of crops where damage was sustained. At one time there was a suggestion that the Puma S used by the represented growers might have come from a defective batch. However, information provided by Hoechst apparently established that this was not the case. Accordingly, Professor Kennedy's starting point was, of necessity, that "[t]he fact of spraying with Puma S alone cannot be the basis of deciding that crops were damaged". (p 6 of Professor Kennedy's second report)

96 The principal expert called by Hoechst, Dr Christopher Preston, agreed with this assessment. Dr Preston is a lecturer in weed management in the Department of Crop Protection at the University of Adelaide. His major focus of interest relates to the biochemistry, genetics and management of herbicide resistance in weed species. He agreed that Puma S was unlikely to be the sole source of damage to the represented growers' wheat crops and suggested that a combination of circumstances contributed to the damage. This was generally in accord with Professor Kennedy's views. Indeed, there was considerable consistency between the views held by Professor Kennedy and Dr Preston on many aspects of this case. To a large extent the differences between them related to details rather than substantial issues, although the details in some cases were potentially significant.

97 Professor Kennedy's `main thesis' (to use his words) is that the damage to the crops was related to a nutritional deficiency in the soil, specifically a deficiency in sulphur levels. Nitrogen was also mentioned by Professor Kennedy as an essential element in the detoxification of Puma S. However it is acknowledged that nitrogen deficiency is unlikely to have played a role in the damage caused to these crops. Soils in this area are not known to be nitrogen deficient. Further, two of the growers had added Urea, a nitrogen-based fertiliser, to at least part of their crops which later sustained damage.

98 Various suggestions were made as to other factors which might have contributed to the crop damage. Principal amongst these were:

* low temperatures;

* frosts subsequent to spraying; and

* moisture stress.

99 I shall deal with each of these factors in turn. However, it is appropriate at this stage to say something more about the operation of Puma S. It was described in a general way earlier in this judgment, but a more detailed exposition will be necessary in order to comprehend some aspects of the evidence.

How Puma S Works

100 Puma S was released commercially for the wheat season in 1993. It is known as an APP herbicide. Its active ingredient is fenoxaprop-p-ethyl (FPE). Once inside a plant, the herbicide is de-esterified into an acid form. As such it inhibits fatty acid biosynthesis in the plant. Puma S is primarily absorbed through plant leaves. The inhibition of fatty acid biosynthesis in leaves is not fatal to plants. It is when this inhibition occurs within the meristem of the plant that the plant will die. The effect of Puma S upon target weeds was described by Mr John Ashby, Hoechst's Field Development Manager in the following terms:

"The first symptoms in the target weeds appear two to three days after application with a cessation of growth. As the growth ceases in the target weeds, new leaves or secondary roots are no longer formed. The first visual symptoms of Chlorosis (a change in leaf colour) begins at 10-14 days in young leaves and the necrosis (death of plant tissue) spreads over the leaves and shoots. Slowly the whole weed dies which may take from four to eight weeks depending on climatic conditions." (statement dated 30 October 1998)

101 As to the effect of Puma S upon wheat: in wheat, unlike target weeds, the herbicide will be metabolised by the plant cells in the leaf, ensuring that only small quantities will reach the meristem. In other words, wheat will detoxify the chemical more rapidly than weeds, so that the herbicide will have been neutralised before it reaches an area where any systemic damage can be caused. The addition of a safener, in this case FCE, serves to accelerate the detoxification process.

102 In the meantime, before the FPE is reduced to a neutral level, wheat plants suffer what has been described as a "setback period" or a "knockback period". This was described by Professor Kennedy as a "brief setback", which lasts whilst detoxification takes place. During this phase, he said, the effect of other stresses, such as disease, frost and nutrient deficiencies, might be accentuated. Dr Preston agreed with this assessment. However it must be assumed that, in the normal course of events, these stresses have effects which are temporary only and do not lead to permanent damage. For a very large number of trials were conducted on Puma S before 1996. These encompassed a wide range of growth and climatic conditions as I shall discuss later. In virtually no case was any permanent damage to crops or any significant loss of yield observed after the application of Puma S, notwithstanding that in some cases frosts were known to have occurred after spraying. If stressful events occurring soon after spraying generally led to permanent (as opposed to transitory) damage, it is virtually inconceivable that this would not have been revealed through the trial process.

103 In many cases wheat will show no outward signs of setback, as the various trials of Puma S illustrated. However Dr Preston said that if you look "very carefully", you will see that there is in fact "a short set back of growth". (T 922) In other cases, a yellowing of the leaves occurs. Dr Preston was asked whether this yellowing might indicate a greater degree of setback, but he said that the process was more complex than this proposition suggested. The yellowing, he said, could be caused by a number of factors. As he put it, "I think it would be hasty to assume that because you have got yellowing you have got decreased detoxification". (T 923) He agreed with Professor Kennedy that, during the setback period, the wheat plant, being already under stress, might be more susceptible to other forms of stress.

104 The yellowing of the crop can, according to Darren Fitzgerald, an agronomist employed by Wesfarmers at Coonamble, last for up to 4 weeks after spraying. The applicants have relied upon this as indicating that the setback period might last for as long as this. However, in the light of Dr Preston's evidence, this connection cannot be drawn. Indeed both Professor Kennedy and Dr Preston described the setback period as normally being a "brief" or "short" one only. Moreover Mr Ashby said that the yellowing often follows the detoxification process. From his experience and the documentation he had read, he said "the detoxification process occurs within two to three, maybe four days, whereas these transient effects [the yellowing] are seen some seven to ten days later". (T 965)

105 With this background I return to the various factors suggested by the experts as possibly contributing to the damaged wheat crops in 1996.

Sulphur Deficiency

106 Professor Kennedy concedes that Puma S, properly formulated and applied, is not normally considered a dangerous product to use. To the contrary, he says that there is ample worldwide evidence that the active ingredient in Puma S, combined with a safener, provides a successful technology for controlling weeds. His principal thesis, as already mentioned, relates to the role of sulphur in the detoxification of FPE in wheat crops. Both the natural detoxification process and the operation of the safener are, he says, dependent upon the glutathione content of wheat, this being a substance containing sulphur. Accordingly, he says, the process of detoxification is directly related to the sulphur level in the soil. In his third report he concludes that "acute sulphur or nitrogen deficiency would almost certainly cause failure of safening action".

107 Professor Kennedy commented that most Australian soils where wheat is grown receive regular applications of superphosphate, which corrects sulphur deficiency as well as phosphorus deficiency. However, soils in north-western New South Wales, which are known to be marginal or deficient in sulphur, are not phosphorus deficient and are therefore not generally treated with superphosphates. Sulphur deficient soils can only be rectified by the application of fertilisers, Professor Kennedy said, unlike nitrogen which can be added to the soil through the planting of certain legumes. He conceded in evidence that he had no information as to actual fertiliser use by the affected growers, and said he was relying upon his general awareness of fertiliser usage in the area. He referred to the soil analysis from the `Silo' paddock on `Ashantee' as supporting his thesis. It will be remembered that this analysis, conducted in February 1997, indicated a sulphur deficiency in the soil. Professor Kennedy also referred to analyses from `Eurie Plains'. However, no direct evidence in relation to that property was adduced in these proceedings, with the result that this material cannot be used in support of Professor Kennedy's thesis.

108 The evidence in this case has thrown up a number of apparent hurdles to Professor Kennedy's "sulphur deficiency" hypothesis. The first is that wheat sprayed with Topik on two of Mr Todd's paddocks exhibited damage which was indistinguishable from that on wheat which was sprayed with Puma S. Professor Kennedy initially sought to explain this by saying that the safener in Topik involves the same processes as that in Puma S, and its detoxification process is similarly dependent upon adequate sulphur levels in the soil. He later conceded that the processes are not the same. Nevertheless, Topik requires some sulphur for its detoxification processes, although not as much as Puma S. Professor Kennedy surmised that the sulphur deficiency in Mr Todd's soil might have been so severe as to have impaired the detoxification process in both herbicides. This suggestion is consistent with Mr Todd's description of crop damage occurring after the application of Puma S in 1993, but it is difficult to see how it stands with Mr Bruce's account of Puma S being sprayed in 1995 with no damage to the crop.

109 Dr Preston agreed that in a "heavily sulphur deficient plant" the process of protein synthesis might be reduced after application of either Topik or Puma S, with a consequent effect upon plant detoxification. However, he disagreed with Professor Kennedy's thesis that the spraying of Puma S on sulphur deficient soils was the likely cause of damage in these cases. He referred particularly to the numerous trials of Puma S conducted throughout Australia on widely divergent soils. I shall be referring to these trials later. Dr Preston pointed out that sulphur deficient soils are not restricted to the Walgett area. Soils in the Riverina area, the area around Toowoomba, and much of the Western Australian wheat belt are also, he said, known to be sulphur deficient. All these areas were included in the Puma S trials, but none of them showed a reduction in wheat yield after spraying with the herbicide. Dr Preston also pointed out that some of the growers had previously sprayed Puma S on the same soils on previous occasions without damaging their crops. This happened in the case of Mr  Chapman and Mr Todd and possibly also Mr Munns.

110 It is difficult to reconcile Professor Kennedy's thesis with the fact that soil analyses on Mr Chapman's property showed adequate levels of sulphur. Mr Harrison SC asked Professor Kennedy about this in a general manner:

"If it be the case that some farmers had damage but no sulphur deficiencies, others had damage but had sulphur deficiencies and the damage appeared the same, that would cast your hypothesis into some doubt, would it not?---Yes, it would." (T 651)

111 Later, the questioning turned to Mr Chapman's case:

"Let me give you a specific example. Mr Chapman is a farmer in these proceedings. He says he sowed to wheat in 1996, used Puma S, observed damage, had soil tests done which became available in January of 1997 which showed that he had no sulphur deficiencies, indeed no nutrient deficiencies, in his paddock"---Well, I would - as I said the soil analyses can be inherently variable.

Let's assume for the moment that they were accurate?---Okay, let's say that there was a good average concentration for the whole field, yes.

That throws your hypotheses into a bit of a spin, doesn't it?---For the critical factor of sulphur yes it does, but that doesn't mean to say that because it's a

multi factorial mechanism that's involved that it involves nitrogen, sulphur, sunlight, water and a failure in any one of those particular factors can diminish the extent of safening. So that you can't say that any one factor is specific and critical. So I wouldn't regard that as a scientist that it means that that the sulphur factor doesn't operate in some cases.

To the extent that - - -?---So on the basis of that information or that proposition you've just put I wouldn't immediately discount the hypothesis as being operative.

Just take a step at a time. I want you to understand that according to Mr Chapman's soil analyses his paddocks were not nutrient deficient in any respect?---Yes.

And yet so he says he suffered the same type of damage following the use of Puma S that it seems others complain of, do you understand?---Yes.

Your nutrient deficiency analysis whether it be sulphur or combinations of sulphur and nitrogen and other factors does not accommodate a loss that he apparently observed on his property, does it?---No, no." (T 669)

112 As this passage makes clear, Professor Kennedy was not suggesting that sulphur deficiency was the only factor that might have led to damage on the sprayed crops. His opinion, with which other experts essentially agreed, was that it must have been a combination of factors. The principal other factors which were suggested have already been mentioned. I shall now proceed to discuss them.

Low Temperatures

113 Professor Kennedy in his report expressed the view that shading by clouds and/or cold temperatures would decrease the rate of detoxification in wheat plants. All chemical reactions increase, he said, by a factor of two for each additional ten degrees celsius. Dr Preston agrees with this proposition.

114 It is difficult to know what measure of temperature is relevant in this respect. Professor Kennedy said that the critical measure was the average temperature over the period that metabolism is taking place. Dr Preston agreed that "intensely low temperatures" would reduce the rate of chemical processes within sprayed plants, including the rate at which the active ingredient inhibits biosynthesis and the rate of induction from the safener. He was asked whether the mean daily temperature would be the appropriate measure. He replied:

"I would suspect, and if I were to do this, that the mean minimum would be a much better indicator of what's happening. The reason I say that is because plants that are chilled often take a significant amount of time to come out of that chilling and the difference between 19 and 21 degrees may not be nearly as significant as the difference between, say, two degrees and freezing." (T 928)

115 Later in his evidence, Dr Preston described the kind of stresses which are likely to damage wheat plants and lead to a decline in yield, regardless of whether the plants have been sprayed with herbicides. The two principal causes of stress, he said, are lack of moisture and low temperatures. He was asked whether low temperatures after the application of Puma S can retard the chemical processes, including the safening process, so as to lengthen the period of setback. Dr Preston responded:

"Well lower temperature can give the appearance of setback in two ways, one of which is that if in fact you do get reduced metabolism ..., there might be more herbicide available in the plant and that may hold it back for longer. The alternative is that plants recover from the setback period by actually growing out of it and if temperatures are insufficient for them to grow rapidly they will give the appearance of an extended setback because of that." (T 942)

116 The Walgett/Coonamble area generally enjoys mild and sometimes warm winter days with quite cold nights. Sometimes the nights are extremely cold, with temperatures occasionally falling as low as minus four degrees celsius. Meteorological records from Walgett, Coonamble and Baradine were tendered by the applicants. Unfortunately, the Baradine records are incomplete and acknowledged to be `suspect'. The applicants, in an aide-memoire, suggested that the Coonabaraban records might be appropriate for properties at Baradine. However Baradine is between Walgett and Coonamble, and is a long way from Coonabarabran. Accordingly, the only available records are from the Walgett/Coonamble areas. The Coonamble records show in July 1996 the minimum temperature dropped below zero on two occasions. It fell between zero and two degrees celsius on seven other occasions that month. In August, there were four nights when the temperature dropped below zero. On 14 and 15 August the temperature was respectively minus 3.1 and minus 3.4 degrees. On a further six nights the temperature was between zero and two degrees celsius. At no time in September did the temperature drop below zero. There were three occasions when it was between zero and two degrees.

117 Walgett is slightly warmer than Coonamble. At Walgett Airport there were two occasions in July when the temperature dropped below zero, and a further six occasions when the minimum temperature was between zero and two degrees celsius. In August there were two occasions when the temperature dropped below zero, and a further eight when it was between zero and two degrees. The lowest temperature recorded during this time was minus 2 degrees celsius. The coldest temperature in September was zero degrees celsius. Interestingly, only a few days earlier, the minimum temperature had been 15.5 degrees with a maximum temperature for that day of 34.5 degrees.

118 It is difficult to see where this material takes the applicants' case. As Dr Preston pointed out, numerous trials of Puma S were carried out over a large variety of climatic conditions in Australia, including some very cold areas in the southern part of the country. In none of these trials was Puma S shown to have an adverse effect upon the yield of crops. Moreover a very large quantity of Puma S was sprayed on wheat crops in the Walgett/Coonamble area during the 1996 season. All were subject to similar climatic conditions. It follows that low temperatures cannot on their own have caused the damage observed in the affected crops. If they had any part to play, it can only have been in combination with other factors.

Frosts subsequent to spraying

119 There is clear evidence that severe frosts, at least at the time of flowering, can cause serious damage to wheat crops. The grain does not form in the heads, thus depriving the crops of yield and sometimes making them unharvestable. There was also a suggestion that frosts might damage the stems of wheat plants, thus cutting off the circulation from the main plant and causing subsequent yield loss.

120 I have already referred to evidence given by some of the represented growers when questioned about whether their crop damage might have been caused by frost. It is fair to say that on the issue of possible frost damage, there is a conflicting body of opinion from the growers on the one side and the respondents' witnesses on the other. Not one grower, when asked about frosts, conceded that there were any frosts in the 1996 wheat growing season which had the potential, either because of their timing or their severity, to adversely affect their crops. On the other hand, many of the respondents' representatives attributed at least part of the damage which they observed to frosts. Mr Fitzgerald, the Wesfarmers representative at Coonamble, said that there was extensive frost damage in the area during that year. He and other agronomists in the area, he said, were conducting crop monitoring processes for their clients. His clients would have covered approximately 120,000 acres in 1996. Of these, he said, "a good 50 to 60%" were frost damaged. (T 768)

121 It would, I think, be unproductive to detail the evidence of each of the affected growers as to the possibility of frost damage on their crops. As already indicated, each of them denied this possibility, given the conditions of the 1996 season.

122 I have already mentioned the minimum temperatures in the Walgett/Coonamble area during August and September 1996. There were some very cold nights. However the evidence is unclear as to how low the temperature must be before a frost will occur. One would, as a matter of logic, expect it to be at 0^oC, but nature does not always follow logic. Ian Elliot said: "My recollection of frost is anything below 2^o celsius." (T 1043) On the other hand, Mr Chapman said that a temperature of minus 3^oC at approximately three inches above the surface of the ground is required before a frost will cause "any severe damage" to a wheat crop. (T 635) These statements are not necessarily inconsistent. There is abundant evidence that wheat, being a winter crop, is generally not susceptible to frost. It requires a severe frost before any serious damage can be occasioned to a wheat crop.

123 With the background of this evidence I return to the meteorological evidence relating to Walgett and Coonamble. In the light of the evidence that wheat is most susceptible to frosts during flowering, which is likely to occur in September or October, or possibly as early as August, I have restricted my attention to those months. Between 8-14 August there were a series of very cold nights in both Walgett and Coonamble. In Walgett they ranged from 2^oC to 0.6^oC, and in Coonamble from 2.6^oC to minus 1.1^oC. Again, there were four days between 20-24 August when the minimum temperature in Walgett was either -1^oC or -2^oC. On 20 and 21 August at Coonamble the temperature fell to -3.1^oC and -3.4^oC respectively. According to Mr Chapman's evidence, this would have been sufficient to cause a damaging frost in that area. On the other hand, the evidence indicates that flowering generally takes place later than this. September, not surprisingly, was much warmer. In Walgett there was only one day, 22 September, when the temperature fell to 0^oC. In Coonamble on 22 and 23 September the temperature was respectively 0.2^oC and 1.4^oC. That was the last occasion in 1996 that any temperature of less than 2^o was recorded.

124 Steve Halton, who sprayed Mr Masman's crop with Puma S, described a "big frost" in the area on 18 September, which he described as unusual. But this does not accord with the meteorological records, which show minimum temperatures of 11^o and 13 ^o at Coonamble and Walgett on that day. It is possible that Mr Halton was mistaken as to the date, for it was certainly very cold a few days later, on 22 September, but this is an inadequate basis for making any finding on this matter.

125 Dr Preston, in his report dated 15 January 1999, had this to say about the effects of frost upon wheat:

"Wheat is most sensitive to frost damage at anthesis and a frost event at time other than anthesis may have a much more limited effect on yield than a frost at anthesis. The effect of frosting is to reduce grain filling in the head. Symptoms can range from mild, such as reduced grain size and grain deformity, to total loss of grain set. At a casual glance, crops might look to be satisfactory; however, at harvest yield will be a little to much lower than expected. This is consistent with the comments of a number of the farmers and sharefarmers on the condition of their crops at harvets [sic].

Even if frosting was the ultimate source of damage, it is still not clear why parts of crop that were not treated with Puma S escaped severe frost damage. I am unable to offer a firm opinion on this in the absence of additional information; however, these patches must have flowered either before or after the damaged parts of the crop."

126 Mr Moore suggested to Dr Preston in cross-examination that the ability of wheat plants to withstand frost might be dependent on the metabolism of the plant. Dr Preston answered:

"The major issue of frosting to wheat is at flowering and that is not necessarily dependent upon the metabolism of the plant but in fact what stage the plant is at when the frost occurs." (T 925)

127 Later Dr Preston was asked whether a root crop would be less able, during the setback period which follows spraying by Puma S, to tolerate frost. He responded:

"Well, one would assume that a plant that was under stress from one situation may be more susceptible to other types of stress.

Including frost?---Including frost." (T 926)

128 Professor Kennedy surmised that low temperatures might have had had the effect of reducing the rate of herbicide detoxification within wheat plants, thereby extending the setback period after spraying and increasing the period during which the plants might have been susceptible to damage by frost. Another possibility mooted during the hearing was that the growth of sprayed crops was set back after the application of Puma S, thus delaying the time when they came into flower and were most susceptible to frost. In other words, when a severe frost occurred, the unsprayed crops had already passed the crucial flowering stage, whereas the sprayed crops remained within the period of vulnerability. This hypothesis has the advantage that, unlike many others, it can explain the differential damage which was observed between sprayed and unsprayed areas of some of the affected crops.

129 There is a real possibility that frost contributed to the damage observed on the represented wheat growers' crops. However the process by which this might have occurred is difficult to assess. And it is likely that other factors also contributed to the damage. Frosts tend to be quite widespread phenomena, albeit with local variations. Given the very extensive use of Puma S in the Walgett/Coonamble area in 1996, one would expect there to have been more reports of differential damage after spraying with Puma S if the sole additional factor had been the occurrence of frosts. In addition, it is difficult to discount the evidence of witnesses, such as Mr Saunders, that the damage to the crops did not resemble any frost damage they had previously seen.

Low Moisture

130 Moisture stress is, on all accounts, the prime determinant of the health and ultimately the yield of wheat crops. In the north-west of New South Wales excess moisture is rarely a problem. In common with much of Australia the difficulty, if it occurs, arises from a lack of moisture in the soil.

131 Adequate moisture at the time of sowing is obviously important to enable full germination. In the present case sowing took place on various dates in May and June 1996. It was the evidence of all growers that at that time there was good moisture content in the soil and that the wheat, after sowing, germinated quickly and came up strongly.

132 Puma S's label advises that application of the chemical to weeds or crops under stress due to, inter alia, dry conditions should be avoided. The evidence is clear that none of the affected crops was exhibiting stress due to moisture deprivation at the time they were sprayed with Puma S. This was the evidence of all the growers as well as that of the various respondents' representatives who inspected the crops prior to spraying. In any event, the evidence indicates that lack of rain in the winter months, June, July and August, will be unlikely to lead to moisture stress in crops. There is little evaporation at this time and, even if there is no rain, there is usually some daily precipitation by way of dew. It is the moisture content of the soil in September, and particularly in October that is, on the evidence, much more significant to the yield of wheat crops.

133 There is considerable discrepancy between the evidence of the various growers as to the adequacy of the rainfall during the latter part of the 1996 wheat growing season. A number of them said that the good moisture profile which prevailed at the time of the sowing lasted throughout the season. Amongst these were Mr Edwards, Mr Lyons and Mr Fester. Mr Bruce, the manager of Mr Todd's properties, said that there were good follow-up rains during the growth season. He provided monthly totals of rainfall between June and October. On the other hand, Mr Whillock, Mr Schneider's sharefarmer, said that the crop on `Ashantee' was showing signs of stress from lack of moisture by October 1996. Mr Halton, who sprayed Mr Masman's property with Puma S on 9 August 1996, described Mr Masman as saying to him "if it doesn't rain in two weeks then we won't get a crop". Mr Masman was questioned about this during his evidence. He conceded that "we would have enjoyed more moisture" or, as he put it on another occasion, "a fall of rain would have helped". (T 464 and T 465) Nevertheless, he insisted that there was adequate moisture in the soil and denied making the statement attributed to him by Mr Halton. Mr Chapman agreed in his statement that there was below average rainfall over the whole district during the relevant part of the 1996 season. The wheat that year was planted on favourable rains, he said in his evidence, but later moisture was below average.

134 A possible explanation for these apparent inconsistencies emerges when one examines the records of 1996 rainfall from various sources. Notwithstanding that the general topography of the land in the Walgett/Coonamble area is essentially similar, it is a very large area and there are significant variations in rainfall within it. The applicants tendered the meteorological records from Walgett and Coonamble. In addition there was evidence from Mr Schneider as to the daily rainfall on `Ashantee' in 1996 and from Mr Bruce as to the monthly rainfall on Mr Todd's properties between July and October of that year. The following table sets out the rainfall during those months from each of those sources. (In the case of `Ashantee' Mr Schneider's figures have been converted from points to millilitres).

135 1996	136 Ashantee	137 Todd	138 Walgett	139 Coonamble
140 July	141 30	142 40	143 47.6	144 5.5
145 August	146 33	147 33.5	148 38.4	149 44.7
150 September	151 13	152 15	153 35.6	154 36.6
155 October	156 30.3	157 23.5	158 45	159 58.2
160 Total:	161 106.3	162 112	163 166.6	164 194.5

135 According to the evidence, 1995 was a better wheat season than 1996. The yields in 1996 were generally below those of 1995, regardless of whether the crops were sprayed with Puma S. Mr Chapman, it will be remembered, said that this was the general experience in the area. He attributed it to lack of moisture content in the soil in 1996. But this is not borne out by the available records. I do not have Mr Todd's figures for 1995, but I do have those for `Ashantee' as well as the meteorological records from Walgett and Coonamble. They show that between July and October 1995 the following rainfalls were recorded (Again Mr Schneider's figures for `Ashantee' have been converted from points to millilitres).

136 1995	137 Ashantee	138 Walgett	139 Coonamble
140 July	141 5.3	142 8.8	143 12
144 August	145 0.0	146 0.0	147 0.6
148 September	149 57.1	150 36.9	151 38.4
152 October	153 25.2	154 36.2	155 18.2
156 Total:	157 87.6	158 81.9	159 69.2

136 These charts indicate startling variations between rainfalls recorded at different parts of the affected areas and also between 1995 and 1996. Surprisingly, they also indicate a significantly higher rainfall in 1996 than in 1995. Perhaps the difference is that in September - a crucial month for wheat crops - the 1995 rainfall was generally higher than in 1996.

137 It is difficult to know what to make of all this material. It would appear that rainfall levels in 1996 were generally adequate. However the variables are such that there may well have been some areas where there was inadequate moisture in the soil towards the end of the season. It is unlikely that Mr Chapman, who is an exemplary farmer, or Mr Whillock, would have made this concession had it not been so.

Other factors?

138 Other factors have been mentioned as possibly contributing to the damage sustained by the growers' crops. A deficiency in nutrients other than sulphur or nitrogen was mooted by the experts but was considered unlikely. The proposition that there was a chemical failure of the safener in the Puma S sprayed on affected crops was, as already mentioned, effectively eliminated.

139 Nor does it appear that the wheat variety was a relevant factor in the yield loss suffered by represented growers. A number of wheat varieties were used, both in sprayed and unsprayed and in damaged and undamaged areas. No pattern of damage was observable according to the variety used. Professor Kennedy said that there was no particular variety or cultivar of wheat that was likely to be more susceptible than others to damage from Puma S.

140 There is evidence that an infestation of black oats in a wheat crop can set the crop back and reduce its ultimate yield even though the infestation is later eradicated by a post-emergent herbicide. In other words, the weed, before eradication, can so impair the growth and reduce the vigour of wheat plants as to cause permanent damage. The extent to which this occurs will vary according to the extent of the infestation and the stage at which it is eradicated. It is likely that this occurred in relation to a number of the represented growers' crops. However it cannot explain the phenomenon, described by some growers, whereby differential damage was sustained by crops within the same areas, according to whether they were sprayed or not.

141 Some of the respondents' representatives said that there were signs of crown rot amongst the stubble of some of the damaged crops. But this was by no means a universal observation. The growers generally denied that their crops were affected by crown rot in such a way as to cause any yield loss. Even on the respondents' evidence, crown rot could not have been responsible, either alone or in combination with Puma S, for the whole of the damage sustained. At most it might have been one of the factors which affected the yield of some of the crops.

Trials conducted on Puma S

142 Puma S was released onto the commercial market in 1993. Commencing in 1987, the herbicide underwent numerous trials to measure both its efficacy in achieving weed control and its safety to wheat crops. These were respectively described as "efficacy" and "tolerance" trials.

143 Evidence relating to these trials was given by John Ashby and James Swain, both of whom were called on behalf of Hoechst. Mr Ashby, as Hoechst Field Development Manager was responsible for co-ordinating the trial work for Hoechst products within Australia. Mr Swain, an independent consultant with extensive qualifications and experience in the crop protection industry, was asked to comment on the adequacy of the trials conducted by Hoechst. Dr Preston also viewed the results of these trials, and gave his opinion on them.

144 Between 1987 and 1996, 254 efficacy trials were conducted on Puma S in Western Australia, South Australia, Victoria and New South Wales. As Mr Swain commented, these enabled Puma S to be evaluated over extensive geographic areas and over a wide range of environmental conditions, soil types, agricultural practices and wheat cultivars. The evaluation was most extensive in New South Wales, where the largest concentrations of target weeds (black oats and phalaris) are to be found.

145 In 203 of the efficacy trials the wheat was later harvested. Mr Ashby said that in every case the areas sprayed with Pumas S recorded yields equivalent to or greater than the untreated controls. According to Mr Swain, in 91 of these cases the yields in the Puma S areas were "significantly higher" than in the untreated areas.

146 In addition to the efficacy trials, a number of Puma S tolerance trials were conducted in weed-free conditions, at rates of up to 1.4 litres per hectare (double the highest recommended rate) with spraying taking place at different periods after sowing. The sprayed crops were regularly compared with unsprayed controls, and their eventual yields were also compared. The purpose of these trials was to determine whether Puma S, either alone or in combination with other herbicides, caused any significant damage to wheat crops or resulted in any reduction in yield.

147 The precise number of tolerance trials is difficult to determine. Mr Swain referred to 24, but Mr Ashby described many more. According to Mr Ashby's statement, in all but one year between 1987 and 1994, tolerance trials were conducted on between 1 and 29 wheat varieties. The total number of trials referred to in his statement was 105. In addition, the New South Wales Department of Agriculture conducted a number of tolerance trials at its research station at Wagga Wagga.

148 As to the result of these trials, in some cases a yellowing of wheat crop was observed within a short time after spraying. However these symptoms were generally transient and caused no reduction in yield. Although this is not entirely clear, it seems that the only occasion when any significant yield loss occurred (ie a loss of more than 5%) was in 1989 when the Wagga Wagga Institute recorded that the wheat cultivar "Sunbird", which had been sprayed with Puma S, yielded significantly less than an unsprayed control crop. This result was apparently a one-off event. Other trials relating the same wheat variety showed no reduction after the crop had been sprayed with Puma S. In his report Mr Swain described the Wagga Wagga result as an "aberration" which occurs from time to time, and which would not cause a prudent manufacture any concern unless it was repeated in subsequent trials. Mr Swain concluded that the reliable performance of Puma S over a wide range of geographical and environmental conditions had been demonstrated. He was satisfied that the product had been adequately evaluated.

149 Dr Preston had viewed the results of about 115 trials before he wrote his first report. In his second report, dated 15 January 1999, Dr Preston made the following comments:

"2. DID AgrEvo CARRY OUT SUFFICIENT TRIALS ON PUMA S?

I have viewed a further 177 trials carried out by AgrEvo staff between 1987 and 1996. These trials were carried out in five different states of Australia over a broad range of conditions. The overwhelming conclusion that could be drawn from these trials is that Puma S is safe for use on wheat crops. Where Puma S was used used[sic], it either outperformed, or performed as well as, products registered for similar purposes, such as Hoegrass, Grasp, Topik, and Achieve. On the few occasions where Puma S performed less well or higher levels of apparent crop phytotoxicity were observed, there was a heavy weed pressure of annual ryegrass. Puma S controls annual ryegrass poorly and the subsequent competitive effect of this weed would mean that yields were reduced compared to products such as Hoegrass. Mixtures of Puma S with some broadleaf herbicides, such as Igran, Jaguar and Tigrex, increased crop phytotoxicity to unacceptable, or near unacceptable levels. A mixture of Puma S with copper sulfate in one trial resulted in extensive phytotoxicity. One trial conducted in western Australia in 1987 indicated quite severe phytotoxicity (judging by the crop effect data provided); however, it is difficult to interpret this result as yields did not seem to be affected and the experimenter did not suggest the phytotoxicity results were unusual.

Having viewed these additional trials, it is my opinion that AgrEvo carried out sufficient trials to establish that Puma S was safe to use in wheat crops under most conditions.

It is significant that in two trials, S2/92 conducted in South Australia in 1992 and WA8/93 conducted in Western Australia in 1993, frost damage at flowering reduced yields by an estimated 60% and 50% respectively. In both of these trials, the Puma S treated plants yielded as well or better than untreated plants. These trials gave no indications that frosting following Puma S application would lead to excessive yield loss when compared with frosting of untreated plants."

150 The trials, as already mentioned, took place in all areas of mainland Australia where wheat is generally grown. A number were conducted in the north-west of New South Wales. For example, in 1989 three trials were conducted at Coonamble, two at Moree and two at Warialda. In 1990, one was conducted at Coonamble and two at Baradine. Between 1989 and 1995 a total of six trials were held at Coonamble, three at Moree, two at Baradine, two at Gunnedah and two at Quirindi. Many if not all of these trials were efficacy rather than tolerance trials. However, insofar as the yields from sprayed crops were compared with the yields of unsprayed crops, these trials provided data as to the potential damaging effect of Puma S upon wheat crops. As already mentioned, the evidence indicates that in all cases the yield of sprayed crops was equal to or greater than the yield from unsprayed areas.

151 Evidence as to the trials of Puma S is relevant to the issue of foreseeability. So also is evidence as to the extent of the use of the herbicide, once released onto the market, and of any complaints received by Hoechst about resultant damage to crops. In this regard, Mr Ashby estimated, from Hoechst's records, that approximately 939,000 hectares were sprayed with Puma S (based on an assumed application rate of 600 millilitres per hectare) during the period between 1993 and 1995. Before 1996, complaints had been received from only two growers, who claimed yield reductions in relation to three varieties of wheat during the 1993 season. One of these was in Moree, and the other in Eumungerie, both in New South Wales. Subsequent investigations as to the cause of the losses were inconclusive. According to Mr Ashby, the areas involved in these complaints constituted 0.01% of the total areas sprayed with Puma S between 1993 and 1995.

Applicants' Case against Hoechst

152 The success of Mr Schneider's representative action against Hoechst is pivotal to the outcome of these proceedings and will be discussed first. For although different issues are involved in the cases against the individual distributors, some crucial threshold issues are raised in the action against Hoechst.

153 It will be remembered that the action against Hoechst was based on negligence and on an alleged contravention of s 52 of the Trade Practices Act. Various breaches were alleged against Hoechst in the negligence claim. At the completion of the evidence, the applicants' case in negligence rested essentially upon two propositions: first, that Hoechst failed to conduct adequate tests on Puma S, given its knowledge of the scientific composition of the product; and secondly, that it failed to provide adequate warnings as to the possible adverse effects of Puma S upon wheat crops. I shall briefly set out the applicants' contentions in relation to each of these alleged breaches before returning to consider the crucial factors of causation and foreseeability.

Failure to conduct adequate tests

154 The applicants submit that it was incumbent upon Hoechst to conduct tests which were specific as to the effect of Puma S on wheat crops in conditions of sulphur deficient soils and low temperatures. Hoechst, as manufacturer, knew the scientific processes by which the herbicide and the safener operated. In particular it knew that the process of detoxification might be retarded in sulphur deficient soils or in low temperatures. This being the case, it was insufficient for Hoechst to undertake random trials in the hope that the conditions encountered during the trials might be representative of the conditions under which the product was likely to be used in the field. It should have conducted tests which were specific to these conditions, namely sulphur deficient soils and low temperatures. It should also have conducted tests in which wheat plants were likely to be subjected to external stresses, such as frosts, during the setback period after spraying.

Failure to warn

155 The applicants' final submissions refer to a number of claimed deficiencies in the warnings set out on the label attached to Puma S containers. In particular, the submissions assert that the warnings were inadequate in that they did not refer to a possible adverse effect of the chemical on wheat, or to any of the following matters:

* Wheat growing in sulphur deficient soil;

* Effect of cold temperatures;

* Length of knock back effect;

* Reduced capacity of plants to cope with stress during the knockback period;

* Reduced capacity of plants to cope with frost.

156 However, the pleadings in the case against Hoechst do not refer to any of these matters. The Further Amended Statement of Claim asserts that Hoechst should have warned the applicants that Puma S could "destroy or substantially reduce" a wheat crop. Given that no application was made to amend the pleadings, the applicants are now restricted to the warnings raised in their pleadings, as their counsel later conceded.

157 It is relevant to this issue to briefly describe the label attached to Puma S containers. The label commences with directions for use in the following terms:

"DIRECTIONS FOR USE: (For Use in All States)

Restraints: Do NOT apply if rainfall is expected within 4 hours.

Do NOT apply to weeds or crop under stress due to, for example, very dry, very wet, nutrient deficient frosty or disease conditions."

158 The label proceeds to specify the rate at which the herbicide should be applied (between 500-700 millilitres per hectare according to the weed involved and the time of application). It then describes a "withholding period" during which Puma S should not be applied. In the case of wheat, this is a period commencing ten weeks before harvest. Under "General Instructions" the following passage appears.

"Results are best under good growing conditions and application to weeds or crop under stress (eg due to continuous severe frosts, nutrient deficient, diseased, dry or waterlogged conditions) should be avoided."

159 The label contains detailed warnings relating to resistant weeds, directions relating to the method of application, and advice relating to the compatibility of Puma S with other herbicides. Protection of wildlife, protection of adjoining crops, and safety directions are all set out on the label. No further warnings are conveyed as to the possible effect of Puma S on wheat crops, other than as quoted above. The applicants, as already indicated, claim that the label was deficient in this regard.

160 Of the two breaches raised by the applicant, only the second - a failure to warn - is capable of giving rise to an action in negligence. The failure to give an adequate warning can, assuming the sufficiency of other evidence in the case, amount to a breach of duty to take reasonable steps to avert a foreseeable risk of injury. But the same cannot be said of a failure to conduct appropriate tests. The carrying out of further tests might have alerted Hoechst to the existence of previously unperceived risks arising from the use of Puma S. But it could not, on its own, have averted any risk of injury flowing from the use of the herbicide. Accordingly, I would take the applicants' contention that Hoechst failed to conduct adequate tests of Puma S as relevant to the issue of foreseeability but not as itself being an actionable breach.

161 Foreseeability is a major issue in this case. However before I turn to discuss it, there is a very significant question to be resolved, namely whether Puma S caused or contributed to the damage to the represented growers' wheat crops. This was one of the major factual issues in the case.

Did Puma S cause the damage?

162 This was described as the question of "causation" during the course of the hearing. It is an accurate description, but it is important to bear in mind the circumstances in which this issue arises. For in many negligence cases, the issue of causation arises in a slightly different context, namely during an inquiry as to whether the defendant's admitted breach of duty has caused or materially contributed to the injury suffered by the plaintiff. The applicants' counsel cited numerous cases relating to causation, but most of them were relevant to this somewhat different inquiry (McGhee v National Coal Board [1972] UKHL 7; [1973] 1 WLR 1, March v Stramare [1991] HCA 12; [1991] 171 CLR 506 ("March"), Bennett v Minister of Community Welfare [1992] HCA 27; [1992] 176 CLR 408 ("Bennett"), Commonwealth v McLean [1997] 41 NSWLR, Chappel v Hart [1998] HCA 55; [1998] 156 ALR 517).

163 In each of those cases, a breach of duty, and thus the foreseeability of injury, had already been established against the defendant. In these circumstances, the causal relationship in question was between the defendant's admitted breach and the plaintiff's injury. As Gaudron J said in Bennett at 420-421:

"generally speaking, if an injury occurs within an area of foreseeable risk, then, in the absence of evidence that the breach had no effect, or that the injury would have occurred even if the duty had been performed, it will be taken that the breach of the common law duty caused or materially contributed to the injury." (footnotes omitted)

164 The applicants have relied upon this line of reasoning in this case. They have suggested that they need only to establish that the injury to their crops was "not a farfetched or fanciful consequence" of Hoechst's failure to comply with its duty, namely to warn of the risks of using Puma S. "Without more," the applicants' submission continues, "an applicant will be entitled to succeed on the causation question, and be entitled to full damages as against the party who failed by act or omission to avoid the realistic risk of injury by compliance with the duty".

165 This line of reasoning might be appropriate in cases where foreseeability of risk of injury and a failure to warn of the risk have already been established. But one does not reach the question of foreseeability unless the applicants can establish a causal connection between the spraying of their crops with Puma S and the damage which was later sustained. If, as the respondents submit, the application of Puma S onto the applicants' wheat crops was not a causative factor in the subsequent damage to these crops, then that will be the end of the matter. No action of the respondents will be shown to have increased the risk of injury to the applicants.

166 It is for the applicants to establish, on the balance of probabilities, that the application of Puma S to their wheat crops was a relevant cause of the damage which subsequently occurred. They need not prove that it was the only cause. In this regard, I do not accept Mr Harrison's submission that Professor Kennedy's concession that Puma S alone could not have caused the damage to the crops was "sufficient to dispose of the applicants' case". If spraying with Puma S was a necessary ingredient in a combination of circumstances which led to the damage being sustained, then it is capable of being a `cause' of the damage. Whether it was so in fact is a different matter, which I shall be discussing shortly.

167 Nor do I accept Hoechst's submission that this link must be proved through expert evidence. It was submitted that this "is not a matter upon which the Court can pass judgment by reference to a series of facts without the benefit of expert opinion". The submissions then proceed to criticise Professor Kennedy's hypotheses, and to dismiss them as merely speculative. However, this submission does not accord with authority: Tubemakers of Australia Ltd v Fernandez (1976) 50 ALJR 720, Dahl v Grice [1981] VR 513 ("Dahl"). Both these cases involved medical issues, specifically the link between an initial injury and the subsequent occurrence of a serious or debilitating condition. In each case the medical evidence acknowledged the possibility of a link between the two, but was insufficient to establish the probability of such a link It was held in each case that it remained open for the tribunal of fact to find, on the balance of probability, that a causal link existed. In Dahl Gobbo J said at 522-523:

"The review of the authorities leads me to reject the appellants' argument that in matters of bodily health, even outside common experience, it is incumbent on a plaintiff to prove the causal connection to the requisite degree of probability by evidence from the expert. It is plain that in such matters the courts have recognized that a possible cause may be elevated to a probable cause. There are a number of reasons why it is undesirable that the opinion as to causal connection be stated in terms of probabilities. In the first place, this is the role of the tribunal of fact and the ultimate task rests with the judge or jury, as the case may be. Secondly, it is inadmissible in the ordinary course for an expert to give evidence in a form that takes up the very ultimate issue that is the responsibility of the tribunal of fact. Though there are many exceptions in practice to the general rule as to not asking questions that by their terms call for an answer to the ultimate issue, it is a rule that is soundly based in its endeavour to reserve to the tribunal of fact the actual responsibility for the resolution of the ultimate issue. A third consideration is that there is inevitably much difference in the views of expert witnesses as to what constitutes a probability as opposed to a possibility, whether in terms of a particular case or simply as a matter of logic. There is the obvious danger that an expert when asked to provide an opinion as to whether a causal link exists may do so in terms of scientific proof that may be altogether too exacting for the degree of satisfaction necessary in a legal proceeding.

.........

The problem was the subject of what was in my respectful opinion a clear and helpful analysis in the judgment of Mahoney, JA in Fernandez v Tubemakers of Australia Ltd [1975] NSWLR 190, at pp 198 and 199-200. His Honour said: "Problems of causality arise in different forms: thus, in one case it may be whether one of two possible causes can positively be excluded; in another case, it may be whether, of several possible causes, any one can be regarded as more probable than the other: cf Ramsay v Watson [1961] HCA 65; (1961), 108 CLR 642. In the present case, two questions emerged: (i) whether trauma of the relevant kind was a possible cause of the hand condition of which the plaintiff complained; and (ii) whether it was the actual cause of that condition in the present case."

168 This analysis is relevant in this case. Even Dr Preston concedes the possibility that Puma S may have constituted part of the chain of circumstances leading to the damaged wheat crops. Other causative factors must have been present, he says. Neither he nor anyone else is able to identify these factors with any degree of certainty. Professor Kennedy's thesis as to the interaction between Puma S and sulphur deficient soils has its difficulties, as I have already mentioned. The issue to be determined in these circumstances is whether, accepting that Puma S was a possible link in the chain of causation, the evidence establishes that it was the "actual cause" of the damage in these cases. This is a question of fact, to be determined "by applying common sense to the facts of each particular case" (per Mason CJ in March at 515).

169 Professor Kennedy regards the difference between the sprayed and unsprayed areas of some growers' crops as providing the strongest argument that Puma S was, in combination with other circumstances, responsible for the reduction in yield. The applicants urge that this is an unanswerable proposition: that the evidence shows that in many of the represented growers' crops there was an observable difference between areas sprayed with Puma S and areas which were not, and that this difference was later reflected in a significant yield loss in the sprayed areas.

170 Hoechst and the other respondents concede this to have been the evidence in certain cases. However they say that this may have been caused by other factors. There are any number of conditions that can lead to crop variations, they say, not only between paddocks, but also within paddocks, in such a way as to affect the appearance and ultimately the yield of wheat crops. These conditions will include:

* Moisture profile

* Soil type

* Topography and therefore susceptibility to frosts

* Cropping history and fertiliser use affecting the nutritional status of soil

* Extent of weed infestation

* Other pests or diseases.

171 The respondents also urge, with considerable force, that the crops of the represented wheat growers, eight of whom gave evidence in the proceedings, constitute only a minute self-selected sample of the extensive wheat crops which were sprayed with Puma S in the north-west of New South Wales in 1996 and which apparently suffered no damage. By the same token, there were many other crops in the area which sustained damage from causes apparently unrelated to Puma S.

172 Dealing first with the proportion of Puma S users who now complain that the herbicide damaged their crops. At present there are ten represented growers who have been identified. Mr Harrison SC urges, with some force that, given the extensive advertising in the Walgett/Coonamble area relating to this case, and given the time which has now passed, it is unlikely that there will be a significant increase in this number. These ten growers constitute a minute proportion of wheat growers in the area. They also constitute only a tiny proportion of growers who sprayed their crops with Puma S. In this regard, Darren Fitzgerald said that Wesfarmers Dalgetty at Coonamble sold about 10,000 litres of Puma S in 1996. According to Ian Elliot, AG-N-Vet at Coonamble sold 15,000 litres that year. Applied at a rate of 600 millilitres per hectare, these two outlets alone provided sufficient Puma S to spray approximately 45,000 hectares or 110,000 acres in the Coonamble area. Nor were these the only suppliers in the Coonamble/Walgett area, as the pleadings in these proceedings indicate.

173 The respondents rely on evidence from a number of sources as to damage sustained in the Walgett/Coonamble area in 1996 from causes apparently unrelated to Puma S. This included evidence from Scott Harris (referred to earlier in relation to Mr Masman's property) that in one paddock of 900 acres, half of which was sprayed with Puma S, the sprayed half produced a considerably higher yield than the unsprayed half.

174 The respondents also rely on the evidence of Doug Lyons (also referred to in relation to Mr Masman's claim). Mr Lyons sprayed extensively with Puma S in 1996. His crop later sustained yield losses, which Mr Lyons attributed to climatic conditions, presumably frost. But the unsprayed portions of his crop were more affected than the sprayed areas.

175 There was general evidence that 1996 was not a good year for wheat growing in the Walgett/Coonamble area, particularly when compared with 1995. Darren Fitzgerald, an agronomist employed by Wesfarmers at Coonamble, commented in his statement that frost affected many growers in the area in 1996. He was asked about this by Mr Harrison SC and the following exchange took place:

"In terms of yield, did you observe there to be any effect on yield as the result of frost on those crops you observed in that year? --- Yes, definitely, there is, depending on the timing of the sowing, it sort of reflected on the frost damage on the crop, but I observe from like 2.5 tonne to the hectare crop right down to nothing, like it had, it looked like that visually but it didn't yield anything at all right up to crops that lost one or two bags, if you understand that." (T 768)

176 Mr Fitzgerald said that of 120,000 acres which his clients had under wheat in 1996, "a good 50 to 60%" were frost damaged. (T 768)

177 Ian Elliott of Ag-N-Vet Coonamble said that by August and September 1996 a number of crops in the Coonamble area were "very stressed". (T 1045)

178 Also relevant here is the evidence of Mr Chapman. Mr Chapman described the general wheat growing conditions in 1996 as much worse than in 1995. He attributed this to low moisture levels later in the season and said that the 1996 yield was adversely affected over a wide area. As to his own crops, it will be remembered that he mistakenly told Mr Skipper in December 1996 that his `Letterbox Number 2' paddock had been sprayed with Puma S. Mr Chapman believed this to have been the case at the time, because of the significant difference between the state of the crops in `Letterbox Number 1' and `Letterbox Number 2'. The latter had produced a much lower yield at harvest. Even after harvest there was a "dramatic visual difference" between the stubble in the two paddocks, according to Mr Skipper. As it transpires, neither `Letterbox' paddock had been sprayed with Puma S, so one must look elsewhere for the cause of damage in `Letterbox Number 2'. Some explanation might be found in the different cropping histories of the paddocks and accordingly in the nutritional status of the soil. But this does not detract from the force of the respondents' contention that factors other than Puma S were causing crop damage in 1996, and that the experiences of the eight represented growers who gave evidence must be seen as extremely isolated, particularly having regard to the very extensive areas which were under wheat in the district that year.

179 This is a very powerful argument. On the other hand, the applicants say that there is no explanation for the differential damage which was observed on some growers' crops, according to whether they were sprayed or not, other than that it was Puma S which made the difference. The following is a brief summary of the experiences of the eight growers who gave evidence, with an emphasis on factors relevant to causation, including any differences between the sprayed and unsprayed areas of their crops, and possible explanations for those differences.

180 Mr Schneider of `Ashantee' sprayed the whole of his wheat crop with Puma S in 1996. According there is no evidence of any differential effect on sprayed and unsprayed areas. On the other hand, there was a close temporal link between the spraying of the crops and the subsequent yellowing of the plants, from which they never recovered. Both Mr Schneider and Mr Saunders described the damage as different from any other damage, including frost damage, that they had ever seen.

181 Against this, the evidence presented a number of possible explanations for the damage, quite apart from spraying with Puma S. These included:-

* Residual damage from black oats. David Schneider said that 1996 saw the worst infestation of black oats in `Ashantee's' history. The spraying with Puma S did not achieve a 100% weed kill, he said. It will be remembered that the herbicide was sprayed at a lower rate than recommended: 470 millilitres instead of 600 millilitres per hectare.

* Low moisture. Claude Whillock said that he attributed the damage he observed in October 1996 to moisture stress.

* Nutritional deficiency. Subsequent soil analysis showed deficiencies in sulphur and zinc.

* Frost damage. In spite of the evidence of Mr Schneider and Mr Saunders, the possibility of frost damage cannot be discounted.

182 Mr Munns planted two adjoining paddocks with wheat on his property `The Ridges'. `David Dell' paddock was sprayed with Puma S after a significant growth of black oats emerged. It subsequently yielded 3.4 bags (0.28 tonnes per acre). A nearby paddock, `Trudi Dell', which had not been sprayed with Puma S, yielded 14.5 bags (1.2 tonnes per acre). Hoechst's counsel sought to explain this discrepancy by reason of an infestation of blue oat mites in `David Dell' after the spraying of Puma S. However there was a similar infestation in Trudi Dell, which was treated identically. The two paddocks had a similar cropping history and it is difficult to identify any factor which can explain the differential damage other than the fact of spraying. It is unlikely that any residual damage from black oats could account for the very significant difference in yield between these two paddocks.

183 Mr Todd of `Combogolong Woolshed' and `Kimo' planted four paddocks to wheat in 1996. One of them was sprayed with Puma S, and two were partially sprayed with Puma S and partially sprayed with Topik. The unsprayed paddock, which was generally of lower quality than the other three paddocks, produced a much higher yield than the others. Indeed only about one quarter of the paddock which was sprayed with Puma S was harvested. In the remaining two paddocks, `Combogolong Lake' and `Brigalow', the yield was very low indeed. More significantly, in both these paddocks there were a number of areas left unsprayed, as the spraying rig could not fit around trees or telegraph poles. There was a dramatic difference in the appearance of the wheat in the sprayed and unsprayed areas, as confirmed by Dennis Tomlinson, an agronomist employed by Hoechst. Mr Todd also produced a number of photographs which demonstrated this difference.

184 Mr Todd's evidence as to the difference between the sprayed and unsprayed portions of his crop was probably the most graphic of all represented growers who gave evidence. One must infer that it was the spraying of the crops which made the difference between the occurrence and the non-occurrence of damage. On the other hand, these differences were observed in the two paddocks which were sprayed with both Puma S and Topik. Mr Todd did not know which parts of these paddocks had been sprayed with either chemical, and there was no discernible difference to indicate this. Accordingly, whilst some of the differences in appearances and yield between sprayed and unsprayed areas of Mr Todd's crop were presumably attributable to spraying with Puma S, others must have been attributable to spraying with Topik. In other words, the damage was related to the application of a herbicide, being either Puma S or Topik.

185 Mr Lyons at `Karuka' sowed 1460 acres to wheat in 1996. 1200 acres were later sprayed with Puma S. This area later showed visible signs of crop damage. The subsequent yield from 1420 acres (being the 1200 acres which had been sprayed with Puma S and 220 acres which had not) was significantly lower than the remaining 40 acres which had not been sprayed.

186 This would appear to suggest that it was the spraying with Puma S which made the difference. Some explanations for this differential damage were posited during the evidence. They are:-

* Different cropping history and therefore different nutritional status. Adrian Lyons said that the unsprayed 40 acres had been farmed less intensively than most of the rest of the property. Ian Elliott said that the general appearance of the crop appeared to vary with its cropping history.

* Frost damage. Graham Callaghan, an independent agronomist, apparently attributed the damage to frost, saying that the 40 acre paddock was more protected than the damaged areas. However, Mr Lyons said that much of the 40 acres was some distance from any tree cover.

187 Whatever the position may be in relation to the 40 acre paddock, there is no real explanation for the observable damage to the 1200 acres of wheat which was sprayed with Puma S as opposed to the remaining 220 acres, which was apparently undamaged. It would therefore appear that the application of Puma S was in some way connected with the damage sustained by Mr Lyons' crops.

188 An interesting aspect of Mr Lyons' experience was that he had 150 acres of "volunteer" wheat which was more advanced than the balance of his crop, and which apparently remained normal and unaffected after being sprayed with Puma S. The significance of this matter will be discussed later.

189 Mr Fester of `Merriwee' sprayed one paddock, `Tracy Farm' at the rate of 360 millilitres per hectare, this being well below the recommended rate. He later sprayed 500 acres of the 680 acres `Tammy' paddock with Puma S at 500 millilitres per hectare. The 180 acres of `Tammy' paddock which were not sprayed were unharvestable, as black oats had choked out the wheat crop. Similarly, the harvest on the `Tracy Farm' paddock was cancelled because there were more black oats than wheat grains going into the bin. The only part of the two paddocks which was harvestable was the 500 acres which had been sprayed with Puma S at the recommended rate. Nevertheless, the yield from this 500 acres was lower than Mr Fester had anticipated, and was also lower than the yield of other, unsprayed paddocks. This is the basis of his claim.

190 Given the extensive infestation of black oats in the `Tracy Farm' and `Tammy' paddocks, it is at least possible that the reduced yield in the sprayed portion of `Tammy' paddock was attributable to the damage caused by black oats before spraying.

191 Mr Edwards of `West Common' had 370 acres under wheat in 1996. After black oats appeared in his crop Mr Edwards sprayed 340 acres with Puma S. The remaining 30 acres were left unsprayed as he could not afford to pay for the additional herbicide. His unsprayed 30 acres subsequently produced nearly three times the yield of the sprayed area. (0.793 tonnes compared to 0.283 tonnes to the acre). There is no suggestion of other factors which might have caused the difference between the yield of these two areas.

192 Mr Masman of `Combogolong' sowed four paddocks to wheat in 1996, two of which were subsequently sprayed with Puma S. A decision was taken not to harvest these paddocks as the wheat contained either no grain or very little grain. One of the two unsprayed paddocks was later damaged by livestock. The crop in the remaining paddock produced a reduced yield as a result of competition from black oats and phalaris, but was at least worth harvesting.

193 Probably the most significant aspect of Mr Masman's evidence was that, in the sprayed paddocks, where the wheat generally contained very little grain, there were unsprayed areas around bore drains and trees where the wheat had grown normally and contained full grain. To this extent, Mr Masman's evidence was similar to that of Mr Todd.

194 Mr Chapman of `Capelle' sprayed Puma S on one of the three paddocks he had under wheat in 1996. The yield from this paddock was subsequently much lower than that in one of his unsprayed paddocks. However it will be remembered that there was a similarly poor yield in part of another unsprayed paddock, namely `Letterbox Number 2'. This suggests that factors other than Puma S - possibly moisture stress - were responsible for the damage.

Discussion

195 The starting point, in my view, must be this: that given the general conditions of the 1996 wheat growing season, the mere fact that a wheat crop was sprayed with Puma S and later sustained a reduction in yield is insufficient, on its own, to implicate the herbicide in the damage. Something more must be shown in order to link Puma S with the subsequent damage to crops. This "something more" may be present in the cases of Messrs Munns, Todd, Lyons, Edwards and Masman, each of whom sustained differential damage in areas sprayed by Puma S, for which no adequate explanation has been given. In the remaining cases, those of Messrs Schneider, Fester and Chapman, that additional element is lacking. It could not be concluded, in my view, that spraying with Puma S was linked to the damage later sustained by these three growers' crops.

196 This gives rise to a further consideration. In the representative action against Hoechst (as opposed to the individual actions against the various distributors) the evidence of each of the eight growers is admissible and relevant in proving causation. In other words, when assessing the case of each individual grower, the experiences of the other seven growers can be taken into account. But the evidence relating to the crops of Messrs Schneider, Fester or Chapman cannot advance the other applicants' cases on causation, given the findings I have just made. More importantly, I do not see how the experiences of Messrs Munns, Todd, Lyons, Edwards or Masman (each of whom has shown a potential link between Puma S and the damage to his own crops) can be used to assist the cases of the other three growers.

197 We are thus left with five wheat growers whose crops apparently sustained differential damage on sprayed and unsprayed areas. Is this a sufficient basis to conclude that Puma S was a "cause" of the damage? In this context, three factors must be taken into account:

1. The fact that no one, experts or otherwise, can identify the process by which damage was sustained by these crops.

2. The fact that Puma S was used extensively in the Walgett/Coonamble area in 1996. So far only ten complaints have been made of subsequent damage to crops. In three of these cases no causative link has been established between the herbicide and the damage. It must therefore be assumed that the great majority of crops which were sprayed with Puma S suffered no ill effects which their growers attributed to the herbicide.

3. The fact that numerous crops in the area apparently sustained damage that year, probably because of moisture stress and/or frost.

198 It is clear that Puma S alone was not responsible for the damage to any of the represented growers' crops. This much was conceded by Professor Kennedy. The damage occurred as a result of a combination of conditions which remain unexplained to this day, and which may well have varied from crop to crop, or even within crops.

199 In my view the most likely explanation for the fact that some crops sustained damage in sprayed areas but not in unsprayed areas is this: that the application of the herbicide affected the subsequent stage of growth of the wheat plants so that they were at a particular stage of susceptibility when a later stressful event (or, more likely, a combination or series of stressful events) occurred. I include nutritional deficiency amongst possible stressful events, although this is in truth a condition rather than an event.

200 I do not think it likely that it was simply a case of external stresses occurring during the normal setback period after spraying. Had the process been as simple as this, the whole of Mr Lyons' crop would have been affected. However, as we know, the volunteer wheat, which was 12-16 days ahead of the rest of the crop, was not affected by Puma S, and remained relatively normal. In any event, as observed earlier in this judgment, if permanent damage was caused by stressful events occurring during the setback period, this would almost certainly have been revealed during the very extensive trials of Puma S. Yet the trials revealed no hint of this phenomenon.

201 It follows that the particular stage of growth at which the crop was sprayed must have been a relevant factor in the series of causative events.

202 It is not likely, however, that the crops were under stress at the time of spraying. Virtually all witnesses, growers and distributors' representatives alike, gave evidence to this effect. This is supported also by the fact that, in all cases where Puma S was applied at the recommended rate, a complete weed kill was achieved. The evidence, it will be recalled, indicates that if wheat plants are under stress at the time of spraying it is likely that the weeds will similarly be stressed and an incomplete kill will be achieved.

203 A very significant point is this: that it was not only crops sprayed with Puma S which displayed a susceptibility to subsequent stressful events (assuming the accuracy of this scenario). For we know that Topik had the identical effect on Mr Todd's crops. In some contexts, this might not be a matter of great significance. The manufacturer of a defective product cannot seek to escape liability merely because other products contain the same defect. But the issue we are presently concerned with is that of causation. And if, as occurred here, indistinguishable damage is sustained after applying two independently manufactured products, then it must be assumed that any causative element in one of the products is also present in the other. In other words, it is unlikely that it was a particular property of Puma S which resulted in susceptibility in the affected crops. Rather it was a property shared with Topik, and possibly also with other post-emergent herbicides of similar action.

204 With this background, I return to the crucial question, namely: can it be said that spraying with Puma S was, in a legal sense, a cause of the damage to the represented growers' crops? In my view, it cannot. In March, Mason CJ had this to say about the `but for' test at 515-516:

"Commentators subdivide the issue of causation in a given case into two questions: the question of causation in fact - to be determined by the application of the `but for' test - and the further question whether a defendant is in law responsible for damage which his or her negligence has played some part in producing: see, eg Fleming, Law of Torts, 7<sup>th</sup> ed (1987), pp 172-173; Hart and Honore, Causation in the Law, 2<sup>nd</sup> ed (1985), p 110. It is said that, in determining this second question, considerations of policy have a prominent part to play, as do accepted value judgments: see Fleming, p 173. However, this approach to the issue of causation (a) places rather too much weight on the `but for' test to the exclusion for the `common sense' approach which the common law has always favoured; and (b) implies, or seems to imply, that value judgment has, or should have, no part to play in resolving causation as an issue of fact. As Dixon CJ, Fullagar and Kitto JJ. remarked in Fitzgerald v Penn `it is all ultimately a matter of common sense' and `[i]n truth the conception in question [ie causation] is not susceptible of reduction to a satisfactory formula'.

That said, the `but for' test, applied as a negative criterion of causation, has an important role to play in the resolution of the question. So much was conceded by Dixon CJ, Fullagar and Kitto JJ in Fitzgerald v Penn in their discussion of the unreported decision of this Court in Skewes v Public Curator (Qld) (6 September 1954) where A and B were driving their vehicles at excessive speeds in conditions of poor visibility so that their vehicles collided. A was on his correct side of the road, B was not. A's negligence was not causative of injury. Their Honours pointed out that, had the action been tried by a jury, it would have been correct for the judge to instruct the jury `to ask themselves the question whether they were satisfied that the collision would not have taken place with the same results if driver A had been driving at a reasonable speed'. See also ICIANZ Ltd v Murphy; Duyvelshaff v Cathcart & Ritchie Ltd.

The commentators acknowledge that the `but for' test must be applied subject to certain qualifications. Thus, a factor which secures the presence of the plaintiff at the place where and at the time when he or she is injured is not causally connected with the injury, unless the risk of the accident occurring at that time was greater: see Hart and Honore, at p 122. As Windeyer J observed in Faulkner v Keffalinos:

`But for the first accident, the [plaintiff] might still have been employed by the [defendants], and therefore not where he was when the second accident happened: but lawyers must eschew this kind of "but for" or sine qua non reasoning about cause and consequence.'

The `but for' test gives rise to a well-known difficulty in cases where there are two or more acts or events which would each be sufficient to bring about the plaintiff's injury. The application of the test `gives the result, contrary to common sense, that neither is a cause': Winfield and Jolowicz on Tort, 13^th ed (1989), p 134. In truth, the application of the test proves to be either inadequate or troublesome in various situations in which there are multiple acts or events leading to the plaintiff's injury: see, eg Chapman v Hearse; Baker v Willoughby; McGhee v National Coal Board; M'Kew (......). The cases demonstrate the lesson of experience, namely, that the test, applied as an exclusive criterion of causation, yields unacceptable results and that the results which it yields must be tempered by the making of value judgments and the infusion of policy considerations. That in itself is something of an irony because the proponents of the `but for' test have seen it as a criterion which would exclude the making of value judgements and evaluative considerations from causation analysis: see Weinrib, `A Step Forward in Factual Causation', Modern Law Review, vol 38 (1975) 518, at p 530.

Novus actus interveniens

In similar fashion, the `but for' test does not provide a satisfactory answer in those cases in which a superseding cause, described as a novus actus interveniens, is said to break the chain of causation which would otherwise have resulted from an earlier wrongful act. Many examples may be given of a negligent act by A which sets the scene for a deliberate wrongful act by B who, fortuitously and on the spur of the moment, irresponsibly does something which transforms the outcome of A's conduct into something of far greater consequence, a consequence not readily foreseeable by A. In such a situation, A's act is not a cause of that consequence, though it was an essential condition of it. No doubt the explanation is that the voluntary intervention of B is, in the ultimate analysis, the true cause, A's act being no more than an antecedent condition not amounting to a cause. But this explanation is not a vindication of the adequacy of the `but for' test." (footnotes omitted)

205 As Mason CJ, Deane and Toohey JJ said in Bennett at 412-413:

"In the realm of negligence, causation is essentially a question of fact, to be resolved as a matter of common sense. In resolving that question, the `but for' test, applied as a negative criterion of causation, has an important role to play but it is not a comprehensive and exclusive test of causation; value judgments and policy considerations necessarily intrude. The inadequacy of the `but for' test has emerged in cases in which a superseding cause, amounting to a novus actus interveniens, has been held to break the chain of causation which would have otherwise resulted from an earlier wrongful act or omission. In those cases, though the earlier wrongful act or omission may have amounted to an essential condition of the occurrence of the ultimate harm, it was not the true cause or a true cause of that harm." (footnotes omitted)

206 The issue of causation in March and Bennett arose in a different context from here, as already discussed. In those cases, it was accepted that the defendant was in breach of its duty to the plaintiff. Accordingly, the issue was whether the acknowledged wrongful act of the defendant was a relevant cause of the subsequent damage, given the intervention of supervening events. The situation here is different. The spraying with Puma S was a neutral event insofar as any foreseeable risk of injury was concerned. (See my discussion shortly on foreseeability). In other words, the intervening events which were said in those cases to "break the chain of causation" did not have that role to play here, as there was no presumed chain of causation in the first place.

207 Accordingly, the issue here is not whether the negligent act of the defendant continued to play a causative role in the sustaining of damage, notwithstanding the occurrence of later, unforeseen events (as was the case in March and Bennett) but whether the applicants can establish that an act of the respondents, namely the manufacturing and distribution of Puma S, played such an integral role in the chain of causation leading to damage as to make it a legal `cause' of the damage. In my view, this has not been established. Spraying the crops with Puma S - or, it seems, with Topik - might well have set the scene whereby subsequent adverse conditions combined to damage the crops, but the herbicide itself was not a relevant cause of it. Even if one adopts the reasoning in March and Bennett, the spraying with Puma S can only have amounted to an "essential condition" of the occurrence of the damage to the five represented growers' wheat crops. It was not a true cause of that harm. The true cause or causes lay elsewhere, in the stressful conditions, whatever they were, that later beset those crops. The manufacturer and distributors of Puma S cannot be responsible in law for the consequences of those stressful conditions.

208 It follows that the applicants have failed to establish their case against either Hoechst or the individual distributors. However the issue of foreseeability requires some discussion. My finding on this issue was adverted to earlier, and I think I should explain, albeit briefly, how I reached this finding.

Foreseeability

209 The applicants say that Hoechst should have foreseen that Puma S might cause damage to wheat crops in low temperatures and/or in sulphur deficient soils; and that during the setback period after spraying, wheat might be particularly susceptible to external stresses, such as frosts. As such, it was incumbent upon Hoechst to warn prospective users of these dangers.

210 It cannot, in my view, be seriously suggested that the trials conducted in relation to Puma S between 1987 and 1995 put Hoechst on notice that there was any danger to wheat crops inherent in the use of Puma S. In the very many efficacy and tolerance trials conducted between those years there were very few indeed in which wheat which was sprayed with Puma S produced a lower yield than an unsprayed control. According to Mr Swain, the trials were sufficient to establish Puma S as an effective herbicide which was safe to use on all wheat varieties in all growing conditions.

211 Nor were the complaints received in relation to the use of Puma S before 1996 sufficient to put Hoechst on notice that there was a problem with this product. The 1993 complaints in relation to the three wheat varieties which were said to have been damaged by Puma S were the only complaints received by Hoechst before 1996. Investigations into these complaints failed to reveal whether it was Puma S or some other factor which was responsible for the damage to these crops. In any event, as Mr Ashby pointed out, these complaints related to 0.01% (or one ten thousandth) of the area sprayed with Puma S before 1996.

212 The essence of the applicants' claim under this head is that Hoechst failed to conduct adequate tests on Puma S. To quote the applicants' written submissions:

"The need for a manufacturer of products to conduct adequate testing and research before the product is placed on the market and to keep abreast of the current scientific knowledge to discharge the duty of care, is implicit in the requirement that a manufacturer has a duty to guard against dangers which he knew or ought reasonably to have known may injure the plaintiff".

213 I do not think that anyone would cavil at this proposition. The real question is how it is to be applied in this case. For on the face of it, the trials conducted in relation to Puma S, appear to have been thorough and comprehensive. However, the applicants say that these trials lacked specificity or direction. In their randomness, the applicants say, the trials were in danger of missing combinations of circumstances which Hoechst knew might affect the safety of Puma S on wheat crops. In particular, Hoechst knew that the detoxification of Puma S in wheat plants was likely to be retarded in sulphur deficient soils and/or in low temperatures. It also knew that during the setback period after spraying (which was likely to be longer in cold or sulphur-deficient conditions) wheat plants would be particularly susceptible to other stresses, particularly frosts. It was therefore incumbent upon Hoechst to conduct trials which were specifically directed towards these factors. It was also pleaded in the applicants' further amended Statement of Claim that Hoechst

"failed to properly or adequately test the herbicide to ensure that it was to specification and/or suitable for application in accordance with the label recommendations to wheat crops grown in the areas and district where the applicant and the represented parties conducted their wheat growing enterprises."

214 Dealing first with sulphur deficient soils: it is conceded by Hoechst that no trials of Puma S were specifically conducted in sulphur deficient soils. Mr Ashby, who was in charge of the trial program, was unaware that sulphur played an essential role in the detoxification of Puma S. This was one of the applicants' complaints, namely that Hoechst, which had this knowledge, did not transmit it to the person charged with trialing the herbicide. On the other hand, as Dr Preston pointed out, the trials of Puma S were conducted throughout Australia on widely divergent soils, including soils that are known to be sulphur deficient, such as the Riverina area, the area around Toowoomba, areas in Western Australia and the Walgett/Coonamble area itself. In none of these tests was there any significant reduction in yield. Moreover, even if trials had been conducted in sulphur deficient soils, there is nothing in the evidence to indicate that any reduction in yield would have been shown. For it is almost certain that other factors, in addition to or in substitution for sulphur deficiency, must have been involved in causing the damage to the applicants' crops, as Professor Kennedy eventually conceded. Indeed, Professor Kennedy's opinion that sulphur deficiency was at least partially responsible for the damage could not be elevated to anything greater than a "hypothesis", as he also conceded.

215 The applicants also suggest that trials should have taken place in very cold areas, so as to enable the length of the setback period to be assessed in these conditions. However it is clear from the charts produced by Mr Ashby that many trials of Puma S took place in areas which were likely to be much colder than the north-west of New South Wales. This includes areas in the south-west of New South Wales, in Victoria and South Australia. Moreover, in many of these trials - approximately 112 of them - frosts were experienced after the application of Puma S, with no resultant damage to crops. This being the case, there is, in my view, no substance in the applicants' suggestion that further trials should have been conducted which were specifically directed to these conditions. The conditions were inherent in many of the trials already conducted.

216 The applicants suggested in their pleadings that inadequate trials were conducted in the area where the applicants grew their wheat crops. However, as already noted, a number of trials were conducted in the north-western wheat growing region, some of them in Coonamble itself. Moreover, as Mr Harrison SC, pointed out, the use of Puma S in the Walgett/Coonamble area during the 1995 wheat growing season could itself be regarded as a massive trial of the herbicide. Puma S was used very extensively in the area that year. The temperatures in 1995 were generally similar to those in 1996. The soils, of course, were the same. Even some of the growers were the same. Mr Todd and Mr Chapman used Puma S on their crops in 1995, without any ill effects.

217 In short, there is nothing to support the proposition that, had Hoechst conducted trials which were specifically directed to the conditions suggested by the applicants, any further information would have been obtained which might have alerted Hoechst to the possibility of damage occurring as it apparently did in 1996.

218 It is often possible, with the benefit of hindsight, to identify the chain of causation which has led to the sustaining of damage by plaintiffs. Even in those cases, foreseeability will not be established unless it can be shown that the defendant should have been alerted, before the event, to the possibility of damage of that type occurring. In the present case, even hindsight cannot tell us what factors or combination of factors caused damage to the applicants' crops. Still less was there anything, before the 1996 season, which should have alerted Hoechst, as a prudent manufacturer, to the possibility that other factors, combined with the application of Puma S, might cause damage to wheat crops. The trials which were conducted were, in my view, entirely adequate in the circumstances. Accordingly, the applicants have failed to show that Hoechst should have foreseen the risk of its product being part of a chain of causation whereby damage was sustained by crops upon which it had been sprayed.

Conclusion

219 My findings on causation and foreseeability are sufficient to dispose of the representative action in favour of Hoechst. I should indicate in conclusion that there were further difficulties in the applicants' cases. One related to the terms in which any warning should have been couched. The applicants referred in their submissions to warnings which related to sulphur deficient soils, cold temperatures, the reduced capacity of plants to cope with stress during the knockback period etc. However the warning as set out in their pleadings, to which the applicants must be restricted, is a warning that Puma S could "destroy" or "substantially reduce" the yield of wheat crops. Before the 1996 wheat growing season, as already discussed, there was nothing to alert Hoechst, as a reasonable manufacturer, that Puma S might have this effect upon wheat crops. Moreover, given the destructive effect of black oats on wheat crops, and the effectiveness of Puma S in killing this weed, it is at least possible that such a warning, if heeded, might have caused greater damage to wheat crops than was sustained by the represented growers. A warning in such general terms was clearly not called for. It was always open to the applicants to seek to amend their pleadings so as to rely on a more specific warning. Their failure to do so probably reflects the fact that it is difficult, if not impossible, to devise a specific warning which meets the circumstances of this case, given that the factors causing damage are still unknown.

220 There is also an issue, as Hoechst's counsel points out, as to whether, even if a warning had been given, it would have been heeded by any or all of the represented growers. However in the circumstances is not necessary to explore that issue here. The applicants have failed to show that Hoechst was in breach of any duty of care which it owed to them. They have thus failed to establish negligence.

221 The claim that Hoechst engaged in misleading or deceptive conduct contrary to s 52 of the Trade Practices Act is, as Mr Rowe described it, based on "knowledge combined with silence". It is rare that silence will be sufficient to establish a contravention of s 52. However it is unnecessary to explore that issue here, for there was clearly no knowledge. The claims under the Trade Practices Act must therefore fail.

222 My findings on causation mean that the individual applicants' claims against the distributors from whom they purchased Puma S must inevitably fail. As Puma S was not a relevant "cause" of the damage to their crops, no breaches have been established under the Trade Practices Act, the Sale of Goods Act, or pursuant to any of the other counts relied upon in these proceedings.

223 I have given considerable thought to the consequential orders which will need to be made, particularly with regard to the various cross-claims. The dismissal of the application in each case will necessarily render the cross-claims otiose. Nevertheless, orders will need to be made dismissing the various cross-claims. As to the costs in the various proceedings, the applicant will pay the costs of both respondents in the representative proceedings commenced by Mr Schneider. In the separate proceedings commenced by Messrs Masman and Edwards, a settlement was reached, before the hearing, between the respondent and Hoechst as cross-respondent. These parties were jointly represented at the hearing. The only other case is that of Mr and Mrs Chapman against Wickman Ryan Pty Ltd and Wesfarmers Dalgety Limited. Hoechst is also involved in that case as cross-respondent and as third cross-claimant. In each of these three cases, an order that the applicants pay the respondents' and cross-respondents' costs of the proceedings will suffice to deal with the situation. No further order need be made as between the various respondents and cross-respondents. Accordingly, the formal orders I make are as follows:-

Schneider v Hoechst Schering Agrevo & Ors - NG 374 of 1997

1. That the Application be dismissed.

2. That the cross-claims be dismissed.

3. That the Applicant pay the Respondents' and Cross-Respondents' costs of the proceedings.

4. The persons bound by this judgment are the group members referred to in paragraph two of the Application (other than the persons who have opted out of the proceedings).

Masman & Ors v Castlereagh Macquarie County Council - NG 931 of 1997

Edwards & Anor v Bayebb Pty Ltd - NG 932 of 1997

Chapman & Anor v Wickman Ryan & Anor - NG 77 of 1998

1. That the Application be dismissed.

2. That the cross-claims be dismissed.

3. That the Applicants pay the Respondents' and Cross-Respondents' costs of the proceedings.

I certify that the preceding two hundred and twenty-three (223) numbered paragraphs are a true copy of the Reasons for Judgment herein of the Honourable Justice Mathews .

Associate:

Dated: 29 March 2000

Counsel for the Applicants in all proceedings:	JE Rowe and CE Moore
Solicitor for the Applicants in all proceedings:	Peter Long & Co
Counsel for the Respondents in NG374/97, NG931/97 and NG932/97:	I Harrison QC and G McNally
Solicitor for the Respondents in NG374/97, NG931/97 and NG932/97	Hunt & Hunt
Counsel for the Respondent in NG77/98:	D Ryan SC
Solicitor for the Respondent in NG77/98:	Andersen Legal
Dates of Hearing:	18 October - 5 November 1999
Date of Judgment:	29 March 2000