OPINION OF THE COURT

STAPLETON, Circuit Judge:

For the second time in just over five years, a group of New Jersey blueberry farmers (collectively, "Plaintiffs") appeals orders of the District Court granting summary judgment to defendant Novartis Crop Protection, Inc. ("Novartis") on Plaintiffs' claims for damage to their crops allegedly caused by use of a pesticide manufactured and distributed by Novartis.

The principal issues on appeal are: (1) whether Plaintiffs' claims of negligent misrepresentation/fraud, violation of the New Jersey Consumer Fraud Act ("NJCFA"), and failure-to-warn are preempted by the Federal Insecticide, Fungicide, and Rodenticide Act ("FIFRA"), a comprehensive regulatory statute that covers the use, sale, and labeling of pesticides; (2) whether Plaintiffs have made a sufficient showing that they relied on the alleged misrepresentations by Novartis to avoid summary judgment on their negligent misrepresentation/fraud and NJCFA claims; and (3) whether Novartis was entitled to summary judgment on Plaintiffs' strict product liability, design defect claim on the ground that Novartis owed no duty to test its pesticide when mixed with fungicides.

We conclude that because Plaintiffs' negligent misrepresentation/fraud and NJCFA claims are based on alleged misrepresentations in Novartis's marketing brochure, and that brochure does not qualify as "labeling" under FIFRA, those claims are not preempted. We further conclude that Plaintiffs, other than Plaintiff Indian Brand Farms, have tendered prima facie evidence of their reliance on Novartis's alleged written misrepresentations. Accordingly, we will vacate the District Court's grant of summary judgment as to the negligent misrepresentation/fraud and NJCFA claims as to all Plaintiffs except Indian Brand Farms and remand for further proceedings. We will affirm the District Court's grant of summary judgment to Novartis on these claims as to Indian Brand Farms.

Because Plaintiffs' failure-to-warn claim, if successful, would not result in a labeling requirement in addition to or different from those required by FIFRA, the failure-to-warn claim is not preempted, and further proceedings on that claim are required. Finally, we conclude that there is a genuine issue of material fact with respect to Plaintiffs' design defect claim, and summary judgment in Novartis's favor was inappropriate.

I. Background

A.

For several years, Plaintiffs treated their blueberry plants with two of Novartis's pesticides, Diazinon 50 WP and Diazinon AG500 ("50 WP" and "AG500," respectively). Before applying 50 WP and AG500, Plaintiffs would combine them with fungicides called Captan and Captec in a process known as tank mixing. Plaintiffs assert that tank mixing is a common practice that is well known in the industry, and while using 50 WP and AG500 in this manner, they experienced no crop damage.

In the spring of 1997,1 Plaintiffs purchased and began using Diazinon AG600 ("AG600"), a new pesticide produced and marketed by Novartis, and Plaintiffs tank mixed AG600 with Captan and Captec in the same manner they had tank mixed 50 WP and AG500. AG600 was indicated for use on sixty-two different plants, one of which was blueberries. Novartis did not recommend, however, on the product label or orally, that growers mix AG600 with Captan or Captec, and the product label warned purchasers that unintended consequences such as crop injury could result from the "presence of other materials, or the manner of use or application." App. at A545.

The AG600 product label is a twenty-one-page document that is divided into "Directions for Use" and "Conditions of Sale and Warranty." The "Conditions of Sale and Warranty" are contained in the first two paragraphs of the label, and the "Directions for Use" take up the remainder, containing multiple subparagraphs and charts covering various crop and pest types. The label does not specify that AG600 contains an inert ingredient called an ionic surfactant2 because the United States Environmental Protection Agency ("EPA") does not require manufacturers to identify the inert ingredients of a product on the label unless the agency has determined that a particular inert ingredient is of toxicological concern. See 7 U.S.C. § 136h(d)(1)(C); 40 C.F.R. § 156.10(g)(1) & (7); Labeling Requirements for Pesticides and Devices, 49 Fed. Reg. 37,960, 37,965 (1984) ("Because the identity of an inert ingredient is protected from disclosure by FIFRA sec. 10(d)(1)(C), a prerequisite for labeling identification of such ingredients is that the Agency make a finding that `disclosure is necessary to protect against an unreasonable risk of injury to health or the environment.'").

Novartis distributed advertising literature stating that AG600 was safer and more effective than its previous products. This literature was in the form of a seventeen-page, full-color, marketing brochure stating that AG600 had "[t]he same powerful product performance," "[i]ncreased safety to users and the environment," and promoted "[b]etter crop safety" with "equal performance." App. at A533-34. The marketing brochure contained no instructions for use of AG600. The brochure was distributed to, among others, product retailers and scientists at the Rutgers University Cooperative Extension, a part of the Rutgers New Jersey Agricultural Experiment Station. As a result of these marketing efforts, these scientists recommended AG600 to New Jersey blueberry farmers, both at a "twilight meeting" of farmers in May 19973 and via the May 29, 1997, edition of the Blueberry Bulletin, a newsletter published by Rutgers.

Contrary to these claims of crop safety, Plaintiffs contend that AG600, when mixed with the fungicides Captan and Captec, caused systemic injury to their blueberry plants, including blotches, depressions, spots on the plants, and even plant death. Plaintiffs allege that this injury was due to the presence of an ionic surfactant, which was not an ingredient of 50 WP or AG500. Plaintiffs allege that this ingredient, about whose inclusion in AG600 Plaintiffs were unaware, when mixed with the fungicides, caused the damage to their plants.

B.

Plaintiffs filed suit in the United States District Court for the District of New Jersey on May 7, 1999, seeking damages based on claims of strict liability under the New Jersey Products Liability Act ("NJPLA") (in that AG600 had a latent design defect and Novartis failed to warn that AG600 could be harmful to crops if mixed with a fungicide); negligence (in that Novartis was negligent in failing to test AG600 before putting it in the stream of commerce); negligent misrepresentation/fraud (in that Novartis marketed AG600 as controlling insects without having adverse effects on plants, when Novartis knew or should have known that this was false); breach of the NJCFA (in that Novartis deceptively represented that AG600 was safe to use on blueberry plants); and breach of express warranty (in that Novartis warranted that AG600 would conform to the chemical description on its label and would not injure plants).

Following discovery, Novartis moved for summary judgment, arguing, inter alia, that Plaintiffs' claims were preempted by FIFRA. The District Court agreed and granted the motion. Plaintiffs then appealed to this Court.

On appeal, this Court concluded, in light of the intervening case of Bates v. Dow Agrosciences LLC, 544 U.S. 431, 125 S.Ct. 1788, 161 L.Ed.2d 687 (2005), in which the Supreme Court clarified the scope of FIFRA preemption, that Plaintiffs' claims of strict product liability, negligent testing of AG600, and breach of express warranty were not preempted by FIFRA, and it reversed the District Court's grant of summary judgment on those claims for that reason. Mortellite v. Novartis Crop Protection, Inc., 460 F.3d 483, 489-90 (3d Cir. 2006). With respect to Plaintiffs' claims of negligent misrepresentation/fraud and violation of the NJCFA, to the extent that they were based on oral misrepresentations by Novartis, we ruled that they were not preempted by FIFRA. However, to the extent that these claims were based on written misrepresentations, we remanded them to the District Court, because the preemption issue with respect to these claims had not been fully briefed and argued on appeal. We remanded Plaintiffs' failure-to-warn claim as well, also because the issue of whether they were preempted by FIFRA was not fully briefed and argued on appeal.

Following remand, Novartis filed several summary judgment motions, and in a series of rulings, the District Court once again granted summary judgment to Novartis on all of Plaintiffs' claims. The District Court concluded that, to the extent Plaintiffs' claims of negligent misrepresentation/fraud and violation of the NJCFA were based on written misrepresentations by Novartis, they were preempted by FIFRA, because the AG600 marketing brochure qualified as "labeling" under the statute. Alternatively, the District Court concluded that even if these claims were not preempted, summary judgment was appropriate because "the record reflects Plaintiffs never received and/or relied upon any written representations outside of the product label before purchasing and using" AG600. Indian Brand Farms, Inc. v. Novartis Crop Protection, Inc., No. 99-2118, 2007 WL 4571087, at *7 n. 5, 2007 U.S. Dist. LEXIS 94443, at *23 n. 5 (D.N.J. Dec. 20, 2007). To the extent Plaintiffs' negligent misrepresentation/fraud and NJCFA claims were based on alleged oral misrepresentations by Novartis, the District Court ruled that "there is no evidence that any oral representations were made by Novartis regarding the use of Diazinon AG600 and/or relied upon by Plaintiffs." Id. at *7, 2007 U.S. Dist. LEXIS 94443, at *24. Regarding Plaintiffs' failure-to-warn claim, the District Court concluded that success on this claim "would impose a labeling requirement in addition to the requirements set forth in FIFRA," and thus this claim was preempted. Id. at *9, 2007 U.S. Dist. LEXIS 94443, at *28. With reference to Plaintiffs' design defect claim,4 the District Court concluded that tank mixing of AG600 was not a reasonably foreseeable use of the product, and it was not "practical, feasible, and reasonable, as a matter of law, to require [Novartis] to have tested its product in combination with every fungicide for use on all plants."5 Indian Brand Farms, Inc. v. Novartis Crop Protection, Inc., No. 99-2118, slip op. at 11 (D.N.J. Oct.10, 2008).

Plaintiffs once again appeal.6

II. Preemption and the Alleged Waiver of Efficacy Jurisdiction

We first briefly address an overarching contention of Plaintiffs. They insist that there can be no FIFRA preemption of any form of crop damage claim because Congress granted the EPA authority to waive its jurisdiction over pesticide efficacy issues,7 and the EPA has opted in favor of exercising that authority. It follows, in Plaintiffs' view, that there is no federal regulation of product efficacy and, accordingly, no preemption. We cannot agree for a number of reasons.

First, this case does not involve the efficacy of AG600. Plaintiffs do not allege that it fails to perform in the manner intended with respect to targeted pests; rather, they complain about plant damage, which in FIFRA terminology is damage to the "environment." 7 U.S.C. § 136(j) ("The term `environment' includes ... all plants...."); 40 C.F.R. § 158.130(e)(1) ("The information required to assess hazards to nontarget organisms is derived from tests to determine pesticidal effects on ... plants."); Kuiper v. Am. Cyanamid, 131 F.3d 656, 664 (7th Cir.1997) (concluding that "corn is a plant and falls within this definition"); Etcheverry v. Tri-Ag Serv., Inc., 22 Cal.4th 316, 93 Cal.Rptr.2d 36, 993 P.2d 366, 375 (2000) (stating that with regard to crop damage claims, "the EPA's waiver of the submission of efficacy data is irrelevant, since plaintiffs complain of phytotoxicity, not inefficacy").

Moreover, Congress did not authorize the EPA to waive its jurisdiction over efficacy issues, and the EPA has not done so. The authority referred to (see footnote 7, supra) is authority only to waive "data requirements pertaining to efficacy" when passing on an application for initial registration of a pesticide. If there is cause for concern about crop damage, initially or thereafter, the EPA will respond by requiring additional data and, if appropriate, changes in label warnings.8 It is thus not the case that the EPA no longer regulates efficacy.

Finally, we held when this case was first before us, based upon Bates, that Plaintiffs' failure-to-warn claim was preempted if it imposed a labeling requirement different from or in addition to the labeling requirement imposed by FIFRA. Plaintiffs' broad sweeping waiver of efficacy jurisdiction argument is inconsistent with that holding and, accordingly, with the law of the case.

III. The Negligent Misrepresentation/Fraud and NJCFA Claims

A.

As noted, the District Court held that Plaintiffs' claims of negligent misrepresentation/fraud and violation of the NJCFA are preempted by FIFRA. We turn to that issue. FIFRA sets up the basic system of pesticide regulation in the United States, and it covers, inter alia, the use, sale, and labeling of pesticides. FIFRA requires a manufacturer seeking to register a pesticide to submit to the EPA "a complete copy of the labeling of the pesticide, a statement of all claims to be made for it, ... any directions for its use," and "a full description of the tests made and the results thereof upon which the claims are based, or alternatively a citation to [supporting] data that appear in the public literature." 7 U.S.C. § 136a(c)(1)(C), (F). "The EPA will register the pesticide if it determines that the pesticide is efficacious and will not cause unreasonable adverse effects on humans and the environment, and that its label complies with FIFRA's prohibition on misbranding." Mortellite, 460 F.3d at 488. Under FIFRA, a pesticide is "misbranded" if its labeling contains statements that are "false or misleading in any particular," the pesticide's labeling does not contain directions for use which are "necessary for effecting the purpose for which the product is intended," or "the label does not contain a warning or caution statement which may be necessary ... to protect health and the environment." 7 U.S.C. § 136(q)(1).

Importantly for present purposes, FIFRA provides that a "State may regulate the sale or use of any federally registered pesticide or device in the State, but only if and to the extent the regulation does not permit any sale or use prohibited by this Act." 7 U.S.C. § 136v(a). Additionally, it provides that states "shall not impose or continue in effect any requirements for labeling or packaging in addition to or different from those required under this Act." 7 U.S.C. § 136v(b). Thus, the states have joint control with the federal government in regulating the sale and use of pesticides with only the exception of the EPA's exclusive supervision of labeling. Novartis does not contend that the state law relied upon by Plaintiffs permits a sale or use prohibited by FIFRA. Accordingly, the preemption issue before us turns not on whether Novartis's brochure is related to the sale or use of a pesticide, but rather on whether it constitutes labeling within the meaning of FIFRA. Moreover, even if the brochure constitutes labeling, the law relied upon by Plaintiffs is not preempted unless it imposes a requirement or "requirements ... in addition to or different from those required under" FIFRA. Id.; Bates, 544 U.S. at 444, 125 S.Ct. 1788.

In Bates, a group of Texas peanut farmers alleged that a newly marketed pesticide manufactured and distributed by defendant Dow severely damaged their crops. The pesticide bore a label stating that use of the pesticide was "recommended in all areas where peanuts are grown," but the farmers alleged that Dow should have known that the pesticide would stunt the growth of peanuts in soils with pH levels of 7.0 or greater. Bates, 544 U.S. at 435, 125 S.Ct. 1788. When the farmers applied the pesticide to their western Texas farms, where the soil typically has a pH level of 7.2 or higher, crops were damaged. The farmers brought claims of strict product liability, negligence, fraud, breach of warranty, and violation of the Texas Deceptive Trade Practices Act, Tex. Bus. & Com.Code Ann. § 17.01, et seq. In the course of determining whether these claims were preempted by FIFRA, the Supreme Court articulated the following two-part test:

Id. at 444, 125 S.Ct. 1788 (italics in original).

The Supreme Court in Bates: (1) held that plaintiffs' claims of strict product liability, negligent testing, and breach of express warranty were not preempted, because "[n]one of these common law rules requires that manufacturers label or package their products in any particular way;" (2) concluded that plaintiffs' claims under the Texas Deceptive Trade Practices Act were not preempted, to the extent that statute might provide a remedy for the breach of an express warranty; and (3) remanded the case to the Court of Appeals to determine whether, consistent with the Supreme Court's reasoning, plaintiffs' fraud and failure-to-warn claims were preempted. Id. at 444-47, 125 S.Ct. 1788.

FIFRA defines "label" as "the written, printed, or graphic matter on, or attached to, the pesticide or device or any of its containers or wrappers." 7 U.S.C. § 136(p)(1). FIFRA defines "labeling" as:

7 U.S.C. § 136(p)(2). It is undisputed that the AG600 marketing brochure was not "on, or attached to" AG600, and so it cannot qualify as a "label." It is equally unquestionable, though, that the brochure qualifies as "all other written, printed, or graphic matter." Given that the brochure is not referenced on the AG600 "label," and that there is no other writing accompanying the product which references it, the question of whether the brochure qualifies as "labeling" thus comes down to whether the brochure was "accompanying" AG600, as that term is used in the statute. The case law on the meaning of "accompanying" in this and similar contexts is sparse but helpful.

The Federal Food, Drug, and Cosmetic Act of June 25, 1938, 21 U.S.C. § 301, et seq., defined "labeling" as "all labels and other written, printed, or graphic matter (1) upon any article or any of its containers or wrappers, or (2) accompanying such article." Kordel v. United States, 335 U.S. 345, 347-48, 69 S.Ct. 106, 93 L.Ed. 52 (1948). The term "label" was defined as "a display of written, printed, or graphic matter upon the immediate container of any article." Id. at 348 n. 2, 69 S.Ct. 106. The Supreme Court was asked in Kordel to determine whether this definition of "labeling," similar to that in FIFRA, covered sales literature that was not distributed with a drug. The Court declined to read "accompanying such article" as limited to materials "accompanying such article in the package or container." Rather, it was the content of the materials, not their physical proximity, that controlled:

Id. at 350, 69 S.Ct. 106.

The Court acknowledged that Congress had not intended that the labeling provisions of the Act regulate drug advertising generally. It held, however, that advertising matter was nevertheless "labeling" where it "performs the same function as it would if it were on the article or on the containers or wrappers." Id. at 351, 69 S.Ct. 106. The Supreme Court concluded that the particular material before it was "labeling" because it instructed the ultimate users how to use the drugs:

Id. at 348, 69 S.Ct. 106.

Justice Black's dissenting opinion agreed with this aspect of the Court's decision. He summarized the Court's holding as follows:

Id. at 352, 69 S.Ct. 106.

The decision of the Second Circuit in New York State Pesticide Coalition, Inc., v. Jorling, 874 F.2d 115 (2d Cir.1989), appears to be the only federal appellate decision which speaks directly to the meaning of "accompanying" in the context of FIFRA. It too focuses on the content of material alleged to be labeling, rather than the manner of its distribution, and on whether it instructs the ultimate user on how to use the product.

In Jorling, the Court resolved the question of whether a "New York law, designed to assure public awareness that poisonous chemicals are being utilized," created "labeling" requirements under FIFRA, and was therefore preempted. Id. at 116. The New York law at issue required all commercial pesticide applicators to: (1) "enter into a written contract with the owner of the premises where extermination is to occur;" (2) "provide a list of the chemicals to be applied along with any warnings which appear on the pesticide's... label;" (3) "give the prospective purchaser a notification `cover sheet' which provides further warnings and safety information;" (4) post signs on the perimeter of the premises, "instructing persons not to enter the area for a 24 hour period;" and (5) in some instances "notify the public in newspapers of prospective use over large tracts." Id. at 116-17.

The plaintiffs in the case, a coalition of pesticide applicators, argued that the New York notification requirements constituted "labeling" under FIFRA and were therefore preempted because those provisions required additional "written, printed, or graphic matter" which "accompan[ies] the pesticide or device at any time." Id. at 118-19. The Jorling Court disagreed, concluding that even though these notification materials would be "present in some spatial and temporal proximity" to the pesticides, the materials would not "accompany" the pesticides, as that term is used under FIFRA. Id. at 119. As the Jorling Court advised, "`[l]abeling' is better understood by its relationship, rather than its proximity, to the product." Id. The Court instructed that "FIFRA `labeling' is designed to be read and followed by the end user." Id. It then concluded:

Id. at 119-20.

While, as the District Court stressed, these precedents do rule out physical proximity to the product as the controlling factor, they also speak persuasively to the necessity of constraining the scope of "accompanying" if Congress's intent is to be served. As the Jorling Court stressed, "Congress explicitly preserved the states' right to regulate the `sale and use' of pesticides while reserving `labeling' to federal control." Id. at 118. The labeling provisions of FIFRA were thus clearly not intended to regulate sales literature generally and the legal obligations that can arise therefrom. Congress's objective was much narrower. It sought to impose uniformity of labeling throughout the country, Bates, 544 U.S. at 452, 125 S.Ct. 1788,9 and to protect the integrity of that uniform labeling. Jorling, 874 F.2d at 119.

Novartis's AG600 marketing brochure cannot be read as providing a supplement to the AG600 label. Its function is to point out the advantages of the new product to wholesalers and retailers, as well as farmers. Importantly, it contains no instructions for the use of AG600. If we were to construe the term "labeling" as including the AG600 brochure, then all sales and marketing materials would necessarily be included within the scope of that term. We are confident that such was not the intent of Congress.

In light of the foregoing, the District Court erred when it concluded that Novartis's marketing brochure qualified as "labeling" under FIFRA. Plaintiffs' claims of negligent misrepresentation/fraud and violation of the NJCFA are not preempted by FIFRA.

B.

The District Court, however, as an alternative basis for its summary judgment ruling on the claims of negligent misrepresentation/fraud and violation of the NJCFA, concluded that "the record reflects Plaintiffs never received and/or relied upon any written representations outside of the product label before purchasing and using" AG600. Indian Brand Farms, 2007 WL 4571087, at *7 n. 5, 2007 U.S. Dist. LEXIS 94443, at *23 n. 5. Thus, the District Court ruled that even if these claims were not preempted by FIFRA, and even if the statements in the marketing brochure were fraudulent misrepresentations, Plaintiffs could not prevail because they failed to show that they relied on the statements to their detriment. We are unpersuaded by this alternative analysis.

In New Jersey, a successful claim of fraud requires proof of five elements: "(1) a material misrepresentation of a presently existing or past fact; (2) knowledge or belief by the defendant of its falsity; (3) an intention that the other person rely on it; (4) reasonable reliance thereon by the other person; and (5) resulting damages." Gennari v. Weichert Co. Realtors, 148 N.J. 582, 691 A.2d 350, 367 (1997). "Negligent misrepresentation is ... [a]n incorrect statement, negligently made and justifiably relied upon, [and] may be the basis for recovery of damages for economic loss... sustained as a consequence of that reliance." H. Rosenblum, Inc. v. Adler, 93 N.J. 324, 461 A.2d 138, 142-43 (1983).

Regarding the reliance element, it is enough for a plaintiff to show "indirect reliance." The New Jersey Supreme Court has explained this concept in the following manner:

Kaufman v. i-Stat Corp., 165 N.J. 94, 754 A.2d 1188, 1195 (2000) (citing Judson v. Peoples Bank & Trust Co., 25 N.J. 17, 134 A.2d 761 (1957); Metric Inv., Inc. v. Patterson, 101 N.J.Super. 301, 244 A.2d 311 (App.Div.1968)).

"[T]o state a claim under the [NJCFA], a plaintiff must allege each of three elements: (1) unlawful conduct by the defendants; (2) an ascertainable loss on the part of the plaintiff; and (3) a causal relationship between the defendants' unlawful conduct and the plaintiff's ascertainable loss." N.J. Citizen Action v. Schering-Plough Corp., 367 N.J.Super. 8, 842 A.2d 174, 176 (App.Div.2003). The proscribed unlawful conduct includes, inter alia, "any unconscionable commercial ... fraud, false pretense, [or] misrepresentation" and "the knowing ... omission of any material fact with intent that others rely upon such ... omission." N.J. Stat. Ann. § 56:8-2. While the required "causal relationship" may be shown under some circumstances without evidence that would satisfy the reliance requirement of common law fraud, evidence of the kind of indirect reliance which satisfies the common law requirement would clearly satisfy the causal relationship requirement of the NJCFA. See Varacallo v. Mass. Mut. Life Ins. Co., 332 N.J.Super. 31, 752 A.2d 807, 817 (App.Div.2000).

Plaintiffs concede that "there was no credible evidence that [they] directly relied on the brochure" before purchasing and applying AG600. Pl. Reply Br. at 16. Plaintiffs contend, though, that they have made a sufficient showing of indirect reliance to avoid summary judgment, given the evidence regarding the recommendations of AG600 by Rutgers scientists at a "twilight meeting" in May 1997, and the Blueberry Bulletin that was published on May 29, 1997. We agree with Plaintiffs as to all Plaintiffs except Indian Brand Farms.

Dr. Sridhar Polavarapu of Rutgers testified that he relied on the marketing brochure in recommending to the farmers at the twilight meeting that they purchase and use AG600 on blueberries:

App. at A428-30. Further, Dr. Polavarapu testified that he relied on the marketing brochure in writing the Blueberry Bulletin article recommending the use of AG600 on blueberries:

Id. at A432-35.

Plaintiffs, in turn, certified and/or testified that they relied upon the "twilight meeting" recommendation of Dr. Polavarapu and the Blueberry Bulletin in purchasing and mixing AG600.10 In light of this evidence, Plaintiffs have made a sufficient showing that they indirectly relied upon the AG600 marketing brochure in making their decision to purchase or apply AG600.

Novartis raises an issue, however, based on the dates of Plaintiffs' purchase and use of AG600, as to whether Plaintiffs could have indirectly relied on the brochure. Novartis points out that most Plaintiffs purchased or used AG600 prior to the May 29, 1997, publication date of the Blueberry Bulletin. See footnote 2, supra. However, this does not resolve whether the two Plaintiffs who clearly purchased and used AG600 after May 29, 1997—Wm. Cappuccio & Sons and Clark Farms—did so in reliance on the Blueberry Bulletin. Nor does it negate the possibility that Plaintiffs relied on Dr. Polavarapu's recommendation at the earlier "twilight meeting." The testimony in the record indicates different dates for the "twilight meeting,"11 but construing this testimony in a light most favorable to Plaintiffs, Plaintiffs (with the exception of Indian Brand Farms) have demonstrated that a genuine issue of material fact exists as to whether the farmers indirectly relied on the representations in the marketing brochure in their decision to purchase and mix AG600.

The problem for Plaintiff Indian Brand Farms is that it purchased AG600 on April 21, 1997, and it applied AG600 on May 10, 1997, before either the "twilight meeting" or the publication of the Blueberry Bulletin. Thus, Indian Brand Farms is the only plaintiff that is temporally excluded from having indirectly relied on the AG600 marketing brochure.

In light of the foregoing, the District Court's grant of summary judgment to Novartis on the claims of negligent misrepresentation/fraud and violation of the NJCFA based on written representations will be reversed as to all Plaintiffs except Indian Brand Farms. The District Court's grant of summary judgment on these claims as to Plaintiff Indian Brand Farms will be affirmed.

Our discussion of the District Court's alternative holding regarding Plaintiffs' claims of negligent misrepresentation/fraud and violation of the NJCFA has thus far been limited to claims based on written representations. As noted in Part I.B above, the District Court also granted summary judgment to Novartis on the negligent misrepresentation/fraud and NJCFA claims to the extent they were based on alleged oral misrepresentations. We will affirm this grant of summary judgment, because we agree with the District Court that Plaintiffs have failed to "identif[y] with any certainty any oral representations made by Novartis [regarding the use of AG600] that they relied on." Indian Brand Farms, 2007 WL 4571087, at *7, 2007 U.S. Dist. LEXIS 94443, at *24.

IV. The Failure-to-Warn Claim

Plaintiffs seek to impose liability on Novartis for failing to warn of dangers to their crops from tank mixing AG600 with the fungicides Captan and Captec. Plaintiffs contend that FIFRA requires labels to contain sufficient warnings, see 7 U.S.C. § 136(q)(1), and, accordingly, that their failure-to-warn claim would only impose labeling requirements equivalent and parallel to those of FIFRA, not labeling requirements in addition to or different from those of FIFRA.

We note at the outset that prior to the Supreme Court's decision in Bates, several of our sister Courts of Appeals held that failure-to-warn claims based on inadequate labeling were pre-empted by FIFRA, on the reasoning that "[i]n order to prevail on... failure to warn claim[s], [plaintiffs] would have to prove that [product] labels [approved by the EPA] contained insufficient information and that different labels were warranted," and so "[a]warding damages on the[se] ... claim[s] would therefore be tantamount to allowing the state[s]... to regulate pesticide labeling indirectly, an action which is specifically prohibited by § 136v(b)." Taylor AG Indus. v. Pure-Gro, 54 F.3d 555, 560 (9th Cir.1995); see also Bice v. Leslie's Poolmart, Inc., 39 F.3d 887, 888 (8th Cir.1994) (holding that failure-to-warn claims are preempted by FIFRA because "actual agency approval eliminates any possible claims under state tort law for failure to comply with federal [labeling] requirements") (internal quotation omitted); MacDonald v. Monsanto Co., 27 F.3d 1021, 1025 (5th Cir.1994) (holding that "the express language of FIFRA clearly indicates that Congress intended that the federal act preempt conflicting state law, including state common law tort claims" of failure-to-warn); Ark.-Platte & Gulf P'ship v. Van Waters & Rogers, Inc., 981 F.2d 1177, 1179 (10th Cir.1993) ("To the extent that state tort claims in this case require a showing that defendants' labeling and packaging should have included additional, different, or alternatively stated warnings from those required under FIFRA, they would be expressly preempted."); Worm v. Am. Cyanamid Co., 5 F.3d 744, 748 (4th Cir. 1993) ("The [plaintiffs'] argument that their state law claims are based on duties not inconsistent with those imposed by FIFRA has no merit ..., [because] to argue that the warnings on the label are inadequate is to seek to hold the label to a standard different from the federal one."). However, Bates introduced a different analysis of FIFRA preemption, one that compels us to depart from this pre-Bates precedent.

As we have earlier explained, the Bates Court made clear that failure-to-warn claims were not preempted unless they would impose a requirement "in addition to or different from" those required by FIFRA. Bates, 544 U.S. at 447, 125 S.Ct. 1788. It stressed that it was thus endorsing a "`parallel requirements' reading of § 136v(b)" which preserved state law duties that are consistent with those imposed by FIFRA, whether or not state law provides a remedy that FIFRA did not provide.12 Id. The Court noted that "[p]rivate remedies that enforce federal misbranding requirements would seem to aid, rather than hinder, the functioning of FIFRA." Id. at 451, 125 S.Ct. 1788.

Also, as earlier noted, the Bates Court, because of insufficient briefing, remanded the case to the Court of Appeals to determine whether the plaintiffs' failure-to-warn claim would impose any requirements in addition to or different from the requirements under FIFRA. In doing so, it provided guidance which we find helpful here. First of all, the remand established that mere inconsistency between the duty imposed by state law and the content of a manufacturer's labeling approved by the EPA at registration did not necessarily mean that the state law duty was preempted.13 We must look to the requirements imposed by FIFRA. Accordingly, the Court suggested that, on remand, the Court of Appeals look to whether the failure-to-warn claim was "not equivalent to FIFRA's misbranding standards." Id. at 453 n. 27, 125 S.Ct. 1788. If equivalency is found between the claim and the statutory text, the Court should determine whether there are "any EPA regulations that further refine those general standards in any way that is relevant to petitioners' allegations." Id.14 "To the extent that EPA [has] promulgate[d] such regulations ..., they will necessarily affect the scope of preemption under § 136v(b)." Id. at 453 n. 28, 125 S.Ct. 1788.

FIFRA's misbranding provisions require "warning[s] or caution statement[s] which may be necessary ... to protect health and the environment." 7 U.S.C. § 136(q)(1)(G). The "term `environment' includes water, air, land, and all plants and man and other animals living therein...." 7 U.S.C. § 136(j); Kuiper, 131 F.3d at 664; Etcheverry, 93 Cal.Rptr.2d 36, 993 P.2d at 375. The NJPLA imposes liability on a manufacturer where "the product causing the harm was not reasonably fit, suitable or safe for its intended purpose because it... failed to contain adequate warnings or instructions." N.J. Stat. Ann. § 2A:58C-2. This provision has been interpreted as consistent with Section 2 of The Restatement (Third) of Torts: Product Liability, i.e., liability is imposed for inadequate warnings "when the foreseeable risks of harm posed by the product could have been reduced or avoided by the provision of reasonable... warnings ... and the omissions of the ... warnings renders the product not reasonably safe." In the context of this case, this does not appear to us to impose a duty inconsistent with or in addition to the duty imposed by the text of the warning provisions of FIFRA's misbranding requirements. Moreover, Novartis does not purport to have identified any duty imposed by New Jersey law that does not come within this statutory text. Nor has Novartis identified any EPA regulations that "further refine those general standards in any way that is relevant" to Plaintiffs' allegations. Bates, 544 U.S. at 453 n. 27, 125 S.Ct. 1788. And we have found none.15 Rather, Novartis's response to Plaintiffs' "parallel requirements" interpretation of the state law and FIFRA is an EPA "Notice" of "Revised Policy on Label Claims for Tank Mixing" issued in January 1982. As its title indicates, the revised policy is directed to applications for registration where the proposed label, unlike that of AG600, claims that the product is suitable for tank mixing. Under the revised policy, the "EPA will usually approve tank mix label claims without supporting compatibility and residue data if" certain specified conditions are met, including:

Pesticide Registration (PR) Notice 82-1 (Jan. 12, 1982).

While this policy revision was not applicable to AG600, Novartis finds it to be of controlling significance because its discussion of the "Background" of the policy revision contains the following statements:

Id. (emphasis added).

For Novartis, this statement establishes that the EPA regards FIFRA as imposing no label requirements for a warning of an unreasonable risk of plant damage from tank mixing so long as no tank mixing claim is made. Novartis therefore concludes that Plaintiffs' failure-to-warn claim imposes a labeling requirement not required by FIFRA. We are not persuaded.

Bates teaches that there is a strong presumption against preemption of state law:

Bates, 544 U.S. at 449, 125 S.Ct. 1788. Given this admonition, we would be most reluctant to base a preemption holding on a background observation of the kind relied upon by Novartis.

More importantly, however, this observation does not relate to the labeling requirements for manufacturers. Rather, it is addressed to what applicators have been permitted to do where the manufacturers' label makes no claim concerning tank mixing. The duty of manufacturers under FIFRA is to avoid misbranding, and that duty is not limited to the claims they make for their products. Indeed, the fact that the EPA does not share Novartis's view of the limitation on its duty to warn is apparent from the fact that, after Plaintiffs' crop damage was called to the attention of the EPA, it required that a warning of risk of tank mixing be added to the AG600 label, even though the label continued to contain no claim of tank mixing.

The NJPLA imposes a requirement for a warning of risk to property which is consistent with Section 2 of the Restatement (Third) of Torts: Products Liability. That requirement is thus not an extraordinary one. It is consistent in scope with the generally accepted commercial expectation. Moreover, we note that FIFRA expressly endorses a concept quite similar to New Jersey's duty to warn of risks associated with objectively foreseeable uses. Section 136a(c)(5) provides in part as follows:

7 U.S.C. § 136a(c)(5) (emphasis added). We find it significant that Congress found it advisable to include the provisions of (D) in addition to the "intended function" provisions of (C).

Given that Congress in FIFRA imposed a generalized duty to include in one's labeling any warning statement necessary to protect plant life and the fact that the EPA has not seen fit to narrow that duty, we find no basis for concluding that New Jersey law imposes a duty to warn different than or in addition to the scope of the requirement imposed by FIFRA. The District Court's judgment regarding Plaintiffs' failure-to-warn claim will be reversed.16

V. The Design Defect Claim

The final question before us is whether Novartis was entitled to summary judgment on Plaintiffs' design defect claim. The NJPLA provides that:

N.J. Stat. Ann. § 2A:58C-2. Plaintiffs insist that AG600 was "designed in a defective manner," because it contained the ionic surfactant that Plaintiffs allege caused the damage to their crops when AG600 was mixed with certain fungicides.

To succeed under a strict liability design defect theory in New Jersey, "a plaintiff must prove that (1) the product was defective; (2) the defect existed when the product left the hands of the defendant; and (3) the defect caused the injury to a reasonably foreseeable user." Jurado v. W. Gear Works, 131 N.J. 375, 619 A.2d 1312, 1317 (1993). "Because this case involves a design defect, as distinguished from a manufacturing defect, plaintiff must show specifically that the product `is not reasonably fit, suitable and safe for its intended or reasonably foreseeable purposes.'" Id. (quoting Michalko v. Cooke Color & Chem. Corp., 91 N.J. 386, 451 A.2d 179, 183 (1982); Suter v. San Angelo Foundry & Mach. Co., 81 N.J. 150, 406 A.2d 140, 149 (1979)).

"The decision whether a product is defective because it is `not reasonably fit, suitable and safe' for its intended purposes reflects a policy judgment under a risk-utility analysis" that "seeks to determine whether a particular product creates a risk of harm that outweighs its usefulness." Id. Under this analysis, a manufacturer is not liable for damages where a person misuses the product, unless that misuse was "objectively foreseeable." See Port Auth. of N.Y. & N.J. v. Arcadian Corp., 189 F.3d 305, 314 (3d Cir.1999) (stating that the "unforeseeable misuse of a product may not give rise to strict liability," because "where `the use of the product is beyond its intended or reasonably anticipated scope,' an injury resulting from that use is `not ... probative of whether the product was fit, suitable, and safe'") (quoting Suter, 406 A.2d at 144). Thus, the first step of the risk-utility analysis is for the jury to "determine whether the plaintiff used the product in an objectively foreseeable manner." Jurado, 619 A.2d at 1319.

Novartis contends that tank mixing, as a matter of law, was not a reasonably foreseeable use of AG600. We conclude that there is a genuine issue of material fact with respect to that issue. The evidence in the summary judgment record would support a finding of fact that Plaintiffs' "misuse" of AG600 was objectively, that is reasonably, foreseeable. First, the economics and utility of tank mixing make it almost inevitable that such mixing will occur. See, e.g., App. at A702 (testimony of Plaintiff Gene Martinelli) ("If you apply more than one chemical and it is compatible, that's one less time you have to go through the fields."), A886 (Expert Report of Dr. Carl Whitcomb) ("The tank mixing of insecticides with ... fungicides is common practice as it allows the farmer to address several problems with one spray application ... thereby saving time and equipment expense...."). Second, several Plaintiffs testified that tank mixing was a well-known and common practice among farmers. See, e.g., App. at A204 (Plaintiff Russell Franceschini) (stating that the prior versions of Diazinon "and Captan has been mixed for years"), A708 (Plaintiff Anthony Melora) ("[W]e always—I always mixed the fungicide with insecticide."). Third, several Plaintiffs testified that pesticide dealers and Rutgers personnel recommended tank mixing AG600 with fungicides, which indicates industry practice. See, e.g., App. at A115 (Plaintiff Joyce Cappuccio), A126-27 (Plaintiff Michael DiMeo), A165 (Plaintiff Anthony DiMeo), A216-17 (Plaintiff Joseph Martinelli). Finally, Novartis's own representatives testified that they were aware that farmers frequently mixed pesticides and fungicides. See, e.g., App. at A716-17, A723-24. This evidence is sufficient for a jury to conclude that tank mixing pesticides and fungicides was a reasonably foreseeable practice.

Novartis relies upon this Court's decision in Arcadian in arguing that tank mixing was not a reasonably foreseeable use of AG600. In Arcadian, the plaintiff sought to hold the defendants, manufacturers of fertilizers, liable for massive damage resulting from the 1993 terrorist bombing of the World Trade Center in New York because the terrorists used the defendants' fertilizers in constructing the explosive device. Arcadian, 189 F.3d at 308. One of plaintiff's claims was a products liability claim under New Jersey law, and he argued that defendants owed a duty to plaintiff, even though "there is no allegation that the fertilizer products were dangerous in and of themselves." Id. at 314. With reference to whether building an explosive was a foreseeable use of the fertilizer, we first explained that in New Jersey, foreseeability means objective foreseeability, which:

Id. at 315 (quoting Oquendo v. Bettcher Indus., Inc., 939 F.Supp. 357, 361 (D.N.J. 1996) (quoting Brown v. U.S. Stove Co., 98 N.J. 155, 484 A.2d 1234, 1241 (1984))). We then agreed with the District Court's conclusion that:

Id. at 35, 484 A.2d 1234 (quoting Port Auth. of N.Y. & N.J. v. Arcadian Corp., 991 F.Supp. at 402-03).

Arcadian holds that not all misuses that would be perceived as possible by a reasonable seller are reasonably foreseeable misuses that can lead to seller liability. Virtually all misuses are foreseeable as possibilities. Given that the ultimate objective of the risk-utility analysis is to determine whether a particular product creates a risk of harm that outweighs its usefulness, all possible misuses clearly cannot be the basis for liability on the part of the seller. This does not mean, however, that misuses which a reasonable seller would believe likely to occur in the normal course of events need not be taken into account. Such misuses are "objectively foreseeable" and must be considered. Here, unlike in Arcadian, there is a genuine issue of material fact as to whether a reasonable manufacturer in Novartis's position would have anticipated that Plaintiffs' mixing was likely to happen in the normal course of events.17

After determining that the plaintiff used the product in an objectively foreseeable manner, the next step of the risk-utility analysis requires the fact-finder to determine whether the reasonably foreseeable risk of harm posed by the reasonably foreseeable use of the product could have been reduced or avoided by a reasonable alternative design. Lewis v. Am. Cyanamid Co., 155 N.J. 544, 715 A.2d 967, 980 (1998). In making this determination, the jury is, of course, called upon to assess what risks of harm were reasonably foreseeable.

On remand from this Court's prior decision in this case, the District Court granted summary judgment to Novartis on Plaintiffs' defective design claim because it determined that the risk of harm to Plaintiffs' crops posed by AG600 when mixed with certain fungicides was not foreseeable. It concluded that it could not "find it practical, feasible, and reasonable, as a matter of law, to require the Defendant to have tested its product in combination with every fungicide for use on all plants." Indian Brand Farms, No. 99-2118, slip op. at 11 (D.N.J. Oct. 10, 2008). However, while it is correct to say that a manufacturer in Novartis's position has no duty to test every possible combination of ingredients for every possible misuse of a pesticide, it is not correct to say that a manufacturer is never chargeable with knowledge of the risks of harm that reasonable testing would have revealed. The issue boils down to what a reasonably prudent manufacturer would have done in the way of testing before introducing the product to the market. See Restatement (Third) of Torts: Products Liability § 2 cmt. m (stating that "a seller bears responsibility to perform reasonable testing prior to marketing a product" and "is charged with knowledge of what reasonable testing would reveal," and so "[i]f testing is not undertaken, or is performed in an inadequate manner, and this failure results in a defect that causes harm, the seller is subject to liability," if there was a reasonable alternative design;)18 Feldman v. Lederle Labs., 97 N.J. 429, 479 A.2d 374, 387 (1984) (stating that "a reasonably prudent manufacturer will be deemed to know of reliable information generally available or reasonably obtainable in the industry or in the particular field involved").

Here, the evidence would permit a finding of fact that a reasonably prudent manufacturer would have appreciated the significant risk of crop damage from the use of the ionic surfactant and that a reasonable alternative with less risk was available. Plaintiffs produced two experts, Dr. James Witt and Dr. Carl Whitcomb, who testified that the presence of the ionic surfactant in AG600 should have raised a "red flag" that the pesticide would be physically incompatible with fungicides. See App. at A764, A885-87. Novartis contends that there is a distinction between physical compatibility and whether the mixture would result in plant damage, but Dr. Witt specifically stated that the incompatibility caused by the ionic surfactant "can cause ... crop damage," id. at A885, and Dr. Whitcomb specifically stated that "it appears that the incompatibility and in turn the damage to the blueberry crops resulted from a compound(s) added to the [AG600] other than the primary active ingredient." Id. at A886. In addition, Dr. Pavlis of Rutgers testified that while physical incompatibility does not necessarily result in crop damage, "it's a good indication," because "[i]f you get a problem in the jar [testing for compatibility], then you most likely will get a problem in the field." Id. at A367.

In addition, there is evidence that such incompatibility could have been discovered by conducting a simple bottle-shaking test (in which a pesticide and fungicide are mixed in a bottle to test compatibility), id. at A376, A765, and Dr. Witt testified that there was no need to test every combination of pesticides and fungicides for stability, only combinations commonly used in the areas designated for sale. Id. at A764-65.

In contrast, the record contains no expert testimony to support Novartis's claim that it would have required testing of almost 3,000,000 combinations of pesticides and fungicides in order to determine that there was a significant risk of crop damage. In addition, there is evidence that Novartis did not conduct any compatibility testing at all before marketing and distributing AG600. See id. at A748, A753. It was only after the damage to the blueberry crops was reported that Novartis tested the pesticide with twenty-five other chemicals and found that several of the mixes were in fact incompatible. Id. at A895.

The evidence of record raises a genuine issue of material fact as to whether the risk of harm to Plaintiffs' crops was foreseeable, and whether such risk of harm could have been reduced or avoided by a reasonable alternative design, i.e., a pesticide not containing an ionic surfactant.19 Accordingly, the District Court erred when it granted summary judgment to Novartis on Plaintiffs' design defect claim.

VI. Conclusion

We will affirm the judgment of the District Court in Novartis's favor on Indian Brand Farms' claims of negligent misrepresentation/fraud and violation of the NJCFA. We will also affirm the judgment in Novartis's favor on Plaintiffs' negligent misrepresentation/fraud and NJCFA claims to the extent that they rely on oral representations by Novartis. In all other respects, the judgment of the District Court will be reversed, and this case will be remanded for further proceedings on all of Plaintiffs' other claims.

HARDIMAN, Circuit Judge, dissenting in part.

I am pleased to join Judge Stapleton's opinion for the Court in most respects. Unlike my colleagues, however, I would hold that Novartis had no duty to test AG600 for tank-mixing compatibility with the fungicides used by Plaintiffs. I therefore would affirm the District Court's summary judgment on Plaintiffs' design defect claim under the New Jersey Products Liability Act. And because it follows a fortiori that Novartis had no duty to warn Plaintiffs about the results of a test it had no legal duty to conduct, I would affirm the District Court's summary judgment on Plaintiffs' implied failure-to-warn claim as well. Accordingly, I respectfully dissent from Parts IV and V of the Court's opinion.

I.

The extensive factual background and tortuous procedural history of this case are aptly recounted in Judge Stapleton's thorough opinion, so I shall briefly mention only a few points relevant to my dissent.

Plaintiffs' fourth amended complaint includes two claims brought pursuant to the New Jersey Products Liability Act (NJPLA), N.J. Stat. Ann. § 2A:58C-1 et seq. First, Plaintiffs allege that AG600 was defectively designed because it contained an undisclosed ionic surfactant that harmed Plaintiffs' crops when "tank mixed" with certain fungicides regularly used by Plaintiffs, such as Captan and Captec. Although Novartis did not recommend—and in fact warned against—combining AG600 with fungicides during the application process, Plaintiffs contend that Novartis had a duty to test AG600 for adverse interactions with other agricultural chemicals before distributing it. Second, Plaintiffs aver that AG600 was defective because Novartis failed to warn of the dangers inherent in tank mixing the pesticide with Plaintiffs' fungicides.

In Parts IV and V of its opinion, the Court holds that the District Court erred in granting Novartis summary judgment on both of Plaintiffs' NJPLA claims. Because I believe Plaintiffs' practice of combining AG600 with various fungicides during the tank mixing process was not objectively foreseeable to Novartis, I disagree.

A.

I begin with Plaintiffs' claim for defective design. Under the NJPLA, a plaintiff asserting a claim for defective design must prove by a preponderance of the evidence that "the product causing the harm was not reasonably fit, suitable or safe for its intended purpose because it ... was designed in a defective manner." N.J. Stat. Ann. § 2A:58C-2. When a product is used for something other than its specifically intended purpose, a plaintiff nonetheless may prevail on a design defect claim under the NJPLA by demonstrating that such use was "reasonably foreseeable" to the manufacturer. Jurado v. Western Gear Works, 131 N.J. 375, 619 A.2d 1312, 1317 (1993) (citations and quotation marks omitted). Because it is undisputed that Novartis designed AG600 as a stand-alone pesticide and did not recommend applying it in conjunction with fungicides, Plaintiffs bear the burden of demonstrating that their tank mixing practice was a reasonably foreseeable misuse of the pesticide. Id. at 1317-18.1

To determine whether the misuse of a product was reasonably foreseeable under the NJPLA, we apply an objective test. Id. at 1317. We do not ask whether a manufacturer was aware of previous instances in which its product had been similarly misused because such evidence "tends to show only subjective foreseeability," which is irrelevant to the objective foreseeability analysis. Port Auth. of New York and New Jersey v. Arcadian Corp., 189 F.3d 305, 315 (3d Cir.1999) (quoting Oquendo v. Bettcher Indus., Inc., 939 F.Supp. 357, 363 (D.N.J.1996)); see also Brown v. U.S. Stove Co., 98 N.J. 155, 484 A.2d 1234, 1241 (1984). Instead, whether the misuse of a product is objectively foreseeable—and thereby imposes a duty on the manufacturer to take steps to ensure that the product is safe for that use—is ultimately "a question of fairness and public policy." Arcadian, 189 F.3d at 315. For that reason, we have emphasized that "[f]airness, not foreseeability alone, is the test" for reasonable foreseeability under the NJPLA. Id. at 316 (quoting Kuzmicz v. Ivy Hill Park Apartments, Inc., 147 N.J. 510, 688 A.2d 1018, 1020 (1997)).

It is in this assessment of the objective foreseeability of Plaintiffs' tank mixing of AG600 with fungicides where I part ways from my colleagues. The Majority marshals a considerable amount of evidence to support its conclusion that tank mixing was a reasonably foreseeable misuse of AG600. According to the Majority, individual plaintiff farmers, pesticide dealers, and Rutgers University crop-treatment scientists all testified that tank mixing pesticides with fungicides was a common and well-known industry practice among blueberry farmers. Furthermore, Novartis sales representatives themselves acknowledged a general awareness that farmers often engage in tank mixing.

Although such evidence suggests a subjective awareness on the part of Novartis and others that Plaintiffs would tank mix AG600 with different fungicides, it sheds no light on the question of whether this subjectively foreseeable misuse is objectively reasonable. For a duty to attach under the NJPLA, the misuse of a product must be objectively foreseeable to the manufacturer—a determination which, as explained above, turns on questions of fairness and public policy, not Novartis's subjective awareness of past tank mixing by Plaintiffs. See Jurado, 619 A.2d at 1317; Arcadian, 189 F.3d at 315. Because the deposition testimony cited by the Majority does not indicate whether it would be either fair or sound public policy to charge Novartis with responsibility for ensuring that Plaintiffs' practice of tank mixing AG600 with various fungicides was safe, that evidence would seem irrelevant to the question of whether Plaintiffs' misuse of AG600 was objectively foreseeable under the NJPLA.

Unlike the Majority, I am not convinced that Plaintiffs have carried their burden of demonstrating that their misuse of AG600 was reasonably foreseeable to Novartis. As the Majority's foreseeability analysis demonstrates, Plaintiffs can point to no relevant evidence in the record that suggests tank mixing was an objectively foreseeable misuse of AG600. Instead, Plaintiffs—like the Majority—rely almost entirely on evidence of subjective foreseeability. Because the burden of demonstrating reasonable foreseeability rests on Plaintiffs, see Jurado, 619 A.2d at 1317, I would affirm the District Court's conclusion that Novartis had no duty under the NJPLA to test AG600 for compatibility with various fungicides.

In contrast to Plaintiffs' failure to introduce relevant evidence on the issue of reasonable foreseeability, Novartis has cited persuasive evidence which strongly suggests that it would be neither fair nor prudent public policy to impose a duty on Novartis to test AG600 for compatibility with fungicides such as Captan and Captec. When viewed as a whole, this evidence compels the conclusion that Plaintiffs' practice of tank mixing was not an objectively foreseeable misuse of AG600.

Evidence suggests that the burden such testing would impose on Novartis would be substantial to say the least. The AG600 label indicates that the pesticide is approved for use not just on the blueberries that Plaintiffs raise but on 62 different types of plants. See App. at 548-574. Novartis has represented—and Plaintiffs have not disagreed—that there are approximately 98 different registered fungicides and 141 different registered insecticides that growers of those 62 types of plants could elect to tank mix with AG600. Appellee's Br. at 13-14, 27-28. This yields over 850,000 tank-mix/plant combinations that Novartis would be required to test for compatibility before marketing AG600.2 Evidence also indicates that such extensive testing would be time consuming: a single test conducted by Rutgers University scientists involving AG600, Captan, Captec, and blueberry plants took two full years to complete. See App. at 614-620.3

The Majority points to the deposition testimony of Dr. James Witt, who testified that Novartis need not test AG600 for every possible tank-mix combination. Instead, Dr. Witt opined that Novartis could have looked to "what some of the common and usual practices are, and the areas where [AG 600] is going to be sold and used" and limited its testing accordingly. App. at 764-765. According to the Majority, Dr. Witt's testimony suggests the testing burden is not nearly as onerous as Novartis claims.

With additional information, Dr. Witt's testimony could be a helpful means of evaluating the testing burden that the Majority's holding imposes on Novartis. Despite their burden of demonstrating the objective foreseeability of tank mixing, however, Plaintiffs have cited no additional evidence indicating where AG600 is commonly used and sold and no evidence suggesting what the "common and usual practices" of farmers who use AG600 actually are. Without such information, Dr. Witt's testimony provides us with no basis for concluding that the Majority's holding will require Novartis to test anything less than the approximately 850,000 potential tank-mix combinations discussed above. The notion that far fewer tests would be required is based, I believe, on the fallacy that the manufacturer knows in advance which of the many potential AG600/insecticide/fungicide/crop combinations the end user will choose.

In my view, the considerable burden that the Majority's holding will impose on manufacturers is unsound public policy. Requiring Novartis to test AG600—and, apparently, all other pesticides—in innumerable tank mixing combinations would stifle the development of agricultural pesticides and increase substantially their cost of production. This would, in turn, drive up the cost of food, since pesticide manufacturers and farmers would inevitably pass on at least a portion of their escalating costs to consumers. As a matter of policy, then, it seems both logical and prudent to lay the responsibility for ensuring tank-mixing compatibility on end users such as Plaintiffs, who have actual knowledge of the specific types of fungicides that they wish to combine with AG600 and the various crops they wish to treat with the mixture.

Nor would requiring Novartis to test AG600 for tank mix compatibility be particularly fair, either. Because the Environmental Protection Agency has not approved AG600 for mixing with fungicides such as Captan and Captec, any such mixture would be considered an "off-label" use of the pesticide. Federal law prohibits Novartis from recommending or marketing AG600 for off-label uses. See 7 U.S.C. § 136j(a)(1)(B). Furthermore, Novartis explicitly cautions farmers against mixing AG600 with other chemical substances. See App. at 545. Accordingly, the Majority's holding will require Novartis to undertake extensive and expensive testing to ensure that AG600 is fit to be used in a manner that it warns against and is explicitly prohibited from advocating.

In sum, the record evidence suggests that requiring Novartis to test AG600 for tank mix compatibility with fungicides such as Captan and Captec is both unfair and unsound as a matter of public policy. Accordingly, I would conclude that Plaintiff's misuse of AG600 in the tank mixing process was not objectively foreseeable to Novartis, despite Plaintiffs' evidence suggesting that the company was subjectively aware of the practice. Because Novartis had no duty to test AG600 for compatibility with the fungicides employed by Plaintiffs, I would affirm the District Court's summary judgment on Plaintiffs' defective design claim under the NJPLA.

B.

The District Court also granted summary judgment on Plaintiffs' failure-to-warn claim brought pursuant to the NJPLA after concluding that the claim was preempted. In Part IV of its thorough opinion, the Majority persuasively explains why this holding was error, and I agree fully with its preemption analysis.

Notwithstanding the District Court's flawed preemption reasoning, I do not believe it necessary to reverse the District Court's dismissal of Plaintiffs' failure-to-warn claim. For the reasons explained herein, I would hold that Novartis had no duty to test AG600 for tank mix compatibility with the fungicides used by Plaintiffs. It follows a fortiori that Novartis had no duty to warn Plaintiffs. Accordingly, I would affirm the District Court's summary judgment on Plaintiffs' implied failure-to-warn claim as well.