Plaintiff-appellant Joanne Kotler, individually and as administratrix, filed suit in federal district court against The American Tobacco Company (ATC), Philip Morris, Inc. (PMI) and Liggett Group, Inc. (Liggett) seeking money damages in consequence of her husband's death. Having tasted defeat below, Kotler appeals. We affirm.

Appellant's husband, George Kotler, died of lung cancer which was thought to have been caused by cigarette smoking. The decedent was born in 1931, started smoking in his youth, and thereafter regularly used Pall Malls, an ATC product, until sometime after 1965. He then switched primarily to L & Ms, a Liggett brand. After a few years, he defected to Benson & Hedges, a PMI product. In October 1984, lung cancer was diagnosed. Death followed in early 1986.

Premising jurisdiction on 28 U.S.C. Sec. 1332 (1988) and invoking Massachusetts substantive law, the plaintiff filed a 21-count complaint (seven counts per defendant). Broadly described, the complaint proposed three theories of liability (negligence, breach of warranty, and misrepresentation), each pleaded primarily and derivatively (as loss of consortium claims) against each defendant, thus accounting for 18 statements of claim. The breach of warranty claims, as pleaded, implicated both express and implied warranties and were based upon the Massachusetts version of the Uniform Commercial Code. See Mass.Gen.L. ch. 106, Secs. 2-314-2-318 (1988). The last count in each grouping was for punitive damages.

Following our decision in Palmer v. Liggett Group, Inc., 825 F.2d 620 (1st Cir.1987), the district court dismissed substantial segments of appellant's claims. See Kotler v. American Tobacco Co., 685 F. Supp. 15 (D.Mass.1988) (Kotler I ). Appellant conceded, and the district court found, id. at 18, that "those ... claims which allege inadequacy or failure to warn after January 1, 1966" were preempted by the Federal Cigarette Labeling and Advertising Act, as amended, 15 U.S.C. Secs. 1331-1340 (1988) (the Labeling Act). The district court ruled, however, that preemption did not apply to matters predating January 1, 1966, Kotler I, 685 F.Supp. at 18--a ruling leaving only ATC at risk in respect to such claims inasmuch as the decedent, to all intents and purposes, had smoked nothing but Pall Malls prior to 1966. The court's other rulings comprised a mixed bag; plaintiff's claims for misrepresentation and deceit, alleged to have occurred after 1965, were foreclosed, id. at 20, but her pre-1966 misrepresentation claims against ATC and her breach of warranty claims against all defendants, to the extent based on a design defect independent of the duty to warn, were adjudged not preempted by federal law. Id. at 19-20. In the district court's view, the plaintiff could make out a design defect claim by proving that the defendants' tobacco was "bad," i.e., that defendants' cigarettes contained toxic or carcinogenic ingredients in addition to natural tobacco. Id. at 20.

After discovery was substantially completed, Liggett and PMI each moved for summary judgment on the design defect claims. The court below, finding that Kotler had produced no evidence that the cigarettes in question were defective in a manner which went beyond the inherent characteristics of tobacco, granted the motions. See Kotler v. American Tobacco Co., Civ. No. 86-0810-S, memorandum and order (D.Mass. Nov. 21, 1989), reprinted in Kotler v. American Tobacco Co., 731 F. Supp. 50, 55-57 (D.Mass.1990).1 In roughly the same time frame, ATC moved for summary judgment on all surviving causes of action. Consistent with its rulings in favor of Liggett and PMI, the district court terminated the design defect claim. See Kotler II, 731 F. Supp. at 51-52. The court also granted summary judgment on the misrepresentation claim because appellant tendered no evidence to show that the allegedly tortious pre-1966 advertisements and statements were either health-related or untrue. See id. at 52. Finally, the court rejected Kotler's effort to impose strict liability. See id. at 52-54. Consequently, the case was allowed to go forward only against ATC and solely in respect of the pre-1966 claims for negligence and breach of warranty. See id. at 54-55. Trial began on February 13, 1990 and lasted for roughly three weeks. The court granted ATC's motion for a directed verdict with regard to breach of warranty, but sent the negligence claim to the jury. A defendant's verdict ensued.

With this preface, we turn directly to the issues presented on appeal, offering further facts and details when and as necessary to place particular assignments of error into realistic perspective.

Appellant insists that the court below was incorrect in ruling that the Labeling Act preempted her post-1965 claims of intentional misrepresentation. In response, appellees question our jurisdiction to entertain the argument; and tell us, in any event, that no error was committed.A. Appellate Jurisdiction.

Fed.R.App.P. 3(c) requires, inter alia, that a notice of appeal "shall designate the judgment, order or part thereof appealed from." The Supreme Court has construed Rule 3(c) as being mandatory and jurisdictional. See Torres v. Oakland Scavenger Co., 487 U.S. 312, 315-16, 108 S. Ct. 2405, 2408, 101 L. Ed. 2d 285 (1988). Nevertheless, in holding that the failure to name a party specifically in a notice of appeal constitutes a failure of that party to appeal, the Court took pains not to overrule Foman v. Davis, 371 U.S. 178, 181, 83 S. Ct. 227, 229, 9 L. Ed. 2d 222 (1962). Although Foman held that "the requirements of the rules of procedure should be liberally construed and that 'mere technicalities' should not stand in the way of consideration of a case on its merits," the Foman Court was addressing the "separate provision of Rule 3(c)" requiring that the judgments or orders appealed from be designated. Torres, 487 U.S. at 316, 108 S.Ct. at 2408. Thus, while all provisions of Rule 3(c) are mandatory and jurisdictional, the level of specificity required to satisfy the different provisions varies. Id.

This case implicates the looser Foman specificity standard rather than the unyielding core standard of Torres itself. We are faced with the threshold issue of whether the two notices of appeal which Kotler filed sufficiently designated the lower court's preemption ruling as a subject for appellate inquiry.2 The order in question was part and parcel of Kotler I, originally decided in March 1988 and entered on the docket in final form that May. Neither notice of appeal alluded specifically to the order.

We start with basics. The general rule is that "[a] mistake in designating a judgment or part of a judgment in the notice of appeal ordinarily will not result in loss of the appeal as long as the intent to appeal from a specific judgment can be fairly inferred from the notice, and appellee is not misled by the mistake." Kelly v. United States, 789 F.2d 94, 96 n. 3 (1st Cir.1986). In determining whether appellant's notices of appeal, despite their obvious failure to mention the May 1988 order, sufficiently demonstrated an intent to appeal that order, we are not limited to the four corners of the notices, but may examine them in the context of the record as a whole. See McLemore v. Landry, 898 F.2d 996, 999 (5th Cir.), cert. denied, --- U.S. ----, 111 S. Ct. 428, 112 L. Ed. 2d 412 (1990); FTC v. Hughes, 891 F.2d 589, 590 (5th Cir.1990); Kruso v. International Tel. & Tel. Corp., 872 F.2d 1416, 1423 (9th Cir.1989), cert. denied, --- U.S. ----, 110 S. Ct. 3217, 110 L. Ed. 2d 664 (1990); Brandt v. Schal Associates, Inc., 854 F.2d 948, 954-55 (7th Cir.1988).

In this case, the sufficiency of the notices of appeal is not easily proved or disproved. On the one hand, appellant's complete omission of any reference to the May 1988 order suggests abandonment of her quest; on the other hand, the papers contain intimations that appellant intended to fight the preemption ruling--indeed, in an earlier, untimely notice of appeal, now withdrawn, she mentioned it explicitly--and that appellees realized as much. While we tend to think that appellant fell short of meeting her burden in this respect, we hesitate to decide so bitterly contested a controversy by drawing conclusions about intent and reliance from an imperfectly delineated record. We take shelter instead under the familiar principle that where an appeal presents a difficult jurisdictional issue, yet the substantive merits underlying the issue are facilely resolved in favor of the party challenging jurisdiction, the jurisdictional inquiry may be avoided. See Norton v. Mathews, 427 U.S. 524, 530-32, 96 S. Ct. 2771, 2774-76, 49 L. Ed. 2d 672 (1976); Secretary of the Navy v. Avrech, 418 U.S. 676, 677-78, 94 S. Ct. 3039, 3039-40, 41 L. Ed. 2d 1033 (1974) (per curiam); In re Pioneer Ford Sales, Inc. 729 F.2d 27, 31 (1st Cir.1984). Because the district court's order dismissing appellant's post-1965 intentional misrepresentation claims was flawless, we leave the jurisdictional question unanswered.

B. The Scope of Palmer Preemption.

Giving due suzerainty to this circuit's existing precedent, appellant's post-1965 claims for intentional misrepresentation were, as the lower court ruled, plainly preempted by the Labeling Act. The Act provides:

No requirement or prohibition based on smoking and health shall be imposed under State law with respect to the advertising or promotion of any cigarettes the packages of which are labeled in conformity with the provisions of this chapter.

15 U.S.C. Sec. 1334(b).3 In Palmer, a panel of this court concluded that, though section 1334(b) does not explicitly pretermit common law liability claims, preemption is implicit in the Act as a whole. See Palmer, 825 F.2d at 625. In reaching this conclusion, Palmer found particularly helpful the "straightforward and unambiguous" language of the Act setting forth its purpose: the balancing of "health protection" and "trade protection." Id. at 626. Because this equipoise was so intricately crafted, it would be "inconceivable that Congress intended to have that carefully wrought balance of national interests superseded by the views of a single state, indeed, perhaps of a single jury in a single state.... To permit the interposition of state common law actions into a well-defined area of federal regulation would abrogate utterly the established scheme of health protection as tempered by trade protection." Id.

Appellant challenges the hegemony of Palmer on two grounds. First, she reads Palmer as not banning intentional misrepresentation claims. Alternatively, she maintains that Palmer must be reconsidered in light of English v. General Electric Co., --- U.S. ----, 110 S. Ct. 2270, 110 L. Ed. 2d 65 (1990). Neither point is persuasive.

1. The Applicability of Palmer. The theory underlying Kotler's misrepresentation claims was that, in the post-1965 era, the cigarette companies' purposeful misrepresentations undermined the effectiveness of the aposematic statements required by the Labeling Act. Fleshing out her claims, Kotler averred that the appellees, while promoting their products as desirable, sexy, exciting, sophisticated, and glamorous, misled people into believing that they were safe. She further alleged that the appellees knew or should have known the grave dangers of cigarette use, yet failed adequately to communicate information about these perilous effects beyond the federally mandated warnings. Instead, appellees' advertising was cleverly constructed to vitiate the efficacy of those warnings.

Palmer does not explicitly foreclose the maintenance of such an action; the only claims that Palmer, on its facts, held to be preempted were state-law claims of inadequate warning. Palmer, 825 F.2d at 629. But the Palmer court's reasoning, fairly read, necessarily extends to claims regarding the propriety of advertising, at least insofar as the advertising is attacked on grounds related to the interconnection between smoking and health. After all, Palmer characterized the federal regulatory program established by the Labeling Act as "conspicuously fail[ing] to provide any role for the states in the area of cigarette labeling and advertising," id. at 628; and, in framing the preemption inquiry, Palmer included the propriety of advertising as one component. See id. at 621. The court's analysis leads inexorably to the conclusion that when a state tort claim is based on the propriety of cigarette advertising, the claim is preempted.

Following the threads embedded in the Palmer tapestry erases any genuine doubt. The primary method by which the Labeling Act achieves its purpose of balancing health protection with trade protection is by requiring warnings on cigarette packaging and advertisements. If this equipoise will be set askew by state tort claims branding federally mandated warnings as inadequate, id. at 626, then the balance will equally be tipped by state tort claims that posit improper advertising as undermining the adequacy of federally mandated warnings. Accord Pennington v. Vistron Corp., 876 F.2d 414, 421 n. 9 (5th Cir.1989); Stephen v. American Brands, Inc., 825 F.2d 312, 313 (11th Cir.1987) (per curiam); Cipollone v. Liggett Group, Inc., 789 F.2d 181, 187 (3d Cir.1986), cert. denied, 479 U.S. 1043, 107 S. Ct. 907, 93 L. Ed. 2d 857 (1987). At bottom, appellant's misrepresentation claims are just another species of state tort claims which, like those outlawed in Palmer, are founded on the adequacy of, and ultimately founder upon, the federal warnings. See Cipollone v. Liggett Group, Inc., 893 F.2d 541, 582 (3d Cir.1990) (holding intentional misrepresentation claims to be preempted by the Labeling Act). The district court properly determined such claims, and all others "which allege post-1965 failure to warn or which otherwise necessarily depend upon a showing that the labelling or advertising used by defendants was inadequate or improper," Kotler I, 685 F. Supp. at 20, to be within Palmer 's preemptive sweep.4

2. Palmer's Continuing Vitality. We turn next to appellant's assault on the wisdom and continued vitality of Palmer. At the outset, we reiterate that panels of this circuit are bound by the decisions of prior panels unless there is supervening authority compelling a contrary result. See Jusino v. Zayas, 875 F.2d 986, 993 (1st Cir.1989) (citing cases). Appellant argues that English, --- U.S. ----, 110 S. Ct. 2270, handed down by the High Court after we decided Palmer, constitutes such a supervening precedent. We disagree.

English held that a state-law claim for intentional infliction of emotional distress was not preempted by section 210 of the Energy Reorganization Act under either an "occupation of the field" or "actual conflict" preemption analysis. The Court ruled that there was no field preemption where the claim's effect on the federal regulatory scheme would be "neither direct nor substantial enough to place [it] in the preempted field," id. 110 S.Ct. at 2278; and that maintenance of the claim did not actually conflict with the statute, id. at 2279-80. We are at a loss to see how these holdings call Palmer into legitimate question.

To be sure, the Palmer panel eschewed a characterization of its preemption finding as either field occupation or direct conflict, instead inquiring whether the state-law claim "disturb[ed] too much the congressionally declared scheme." Palmer, 825 F.2d at 626. Be that as it may, the panel necessarily engaged in the English inquiry of whether the challenged state-law claims directly and substantially affected the federal regulatory scheme.5 Moreover, English was premised on no new doctrine, being "strongly suggested" by the Court's earlier decision in Silkwood v. Kerr-McGee Corp., 464 U.S. 238, 104 S. Ct. 615, 78 L. Ed. 2d 443 (1984). See English, 110 S. Ct. at 2278. The Palmer court carefully distinguished Silkwood in terms of the gaping differences between the Labeling Act and the Atomic Energy Act (at issue in Silkwood ),6 and relied heavily on the clear statutory language regarding preemptive intent found in the Labeling Act, determining that the asserted state-law claims threw a monkey wrench into the Act's balancing mechanism.

That the English Court found a totally different type of tort claim not preempted under a totally different statutory scheme sheds little or no light on the correctness of Palmer 's core holding. It is far more important, we suggest, that English and Palmer applied compatible legal standards. Conceptually, these cases do not conflict. We are, therefore, bound to follow our prior precedent (which is closely in point).

Appellant challenges the entry of summary judgment on her design defect claims. In gauging the district court's stewardship of Fed.R.Civ.P. 56(c), our review is plenary. See Mack v. Great Atlantic and Pacific Tea Co., 871 F.2d 179, 181 (1st Cir.1989). Furthermore, in elucidating state law for purposes of this diversity action, we must keep in mind that "[o]ur function is not to formulate a tenet which we, as free agents, might think wise, but to ascertain, as best we can, the rule that the state's highest tribunal would likely follow." Porter v. Nutter, 913 F.2d 37, 40-41 (1st Cir.1990) (quoting Kathios v. General Motors Corp., 862 F.2d 944, 949 (1st Cir.1988)). It is not a diversity court's proper place to torture state law into strange configurations or precipitously to blaze new and unprecedented jurisprudential trails. See id.; see also Ryan v. Royal Ins. Co., 916 F.2d 731, 744 (1st Cir.1990).

The district court considered a design defect claim not preempted by the Labeling Act if the claim was independent of an alleged failure to warn. Accord Kyte v. Philip Morris, Inc., 408 Mass. 162, 556 N.E.2d 1025, 1029-30 (1990) (denying defendant's summary judgment motion inasmuch as plaintiffs' design defect "claim [was] not based on the proposition that all cigarettes are inherently defective"). It initially concluded that appellant, on the pleadings, had stated such a claim under Massachusetts law by averring that the tobacco was "bad." Kotler I, 685 F. Supp. at 19-20. Later on, the court granted brevis disposition because appellant failed to develop proof that appellees' products were defective in some way apart from their inherent characteristics; appellant, in short, had "insufficient evidence as a matter of law to show that the defendants' tobacco was not 'good tobacco.' " Kotler II, 731 F. Supp. at 55; see also id. at 51-52.

On appeal, Kotler does not harangue the district court's assessment of the evidence. Rather, she takes exception to the court's synthesis of Massachusetts law, urging that liability for a design defect is cognizable even absent any showing that the product was defective in a way above and beyond its inherent characteristics. Put succinctly, appellant's argument is that, when one balances risk against utility, cigarettes per se are so unreasonably dangerous as to be actionably defective.

Since Massachusetts does not recognize actions in strict liability for products, see Mason v. General Motors Corp., 397 Mass. 183, 490 N.E.2d 437, 442 (1986), a risk/utility approach cannot be shoehorned into a strict liability espadrille. See Mavilia v. Stoeger Indus., 574 F. Supp. 107, 109 (D.Mass.1983). Plaintiff's argument must, therefore, stand or fall in terms of breach of warranty--a remedy intended, in Massachusetts, "to be fully as comprehensive as the strict liability theory of recovery ... adopted by a great many other jurisdictions." Back v. Wickes Corp., 375 Mass. 633, 378 N.E.2d 964, 968 (1978). The Massachusetts Supreme Judicial Court (SJC) has described the state's law of warranty as being "congruent in nearly all respects with the principles expressed in Restatement (Second) of Torts Sec. 402A (1965)." Id. at 969; see also Hayes v. Ariens Co., 391 Mass. 407, 462 N.E.2d 273, 277 (1984).7 The pivotal question, then, is whether, in the absence of a non-inherent defect, breach of warranty can be found based solely on a risk/utility analysis. We think this question must be answered in the negative.

Liability for breach of warranty is premised on the principle that "[t]he seller warrants that the product is 'fit for the ordinary purposes for which such goods are used.' " Hayes, 462 N.E.2d at 277 (quoting U.C.C. Sec. 2-314(2)(c)). Thus, a breach of warranty can occur if either (1) the product is defectively designed, or (2) foreseeable users are not adequately warned of the dangers associated with its use. Id. For claims of the first type--design defect claims--Massachusetts requires a plaintiff to show that the manufacturer has breached its "duty to design products 'so that they are reasonably fit for the purposes for which they are intended.' " Marchant v. Dayton Tire & Rubber Co., 836 F.2d 695, 698 (1st Cir.1988) (quoting Smith v. Ariens Co., 375 Mass. 620, 377 N.E.2d 954, 957 (1978)). This does not mean a product must be risk-free; rather, it means that any risk created mut be reasonable. See id. In this context, reasonableness is a flexible concept. To evaluate the adequacy of a product's design, the factfinder should consider, inter alia:

the gravity of the danger posed by the challenged design, the likelihood that such danger would occur, the mechanical feasibility of a safer alternative design, the financial cost of an improved design, and the adverse consequences to the product and to the consumer that would result from an alternative design.

Back, 378 N.E.2d at 970 (citation omitted). Weighing these competing factors requires an inquiry essentially the same as the negligence inquiry. See id.

Here lies the rub. We are aware of no Massachusetts case in which liability attached in the absence of evidence that some different, arguably safer, alternative design was possible. In a design defect case premised on negligence, the existence of a safer alternative design is a sine qua non for the imposition of liability. See, e.g., Colter v. Barber-Greene Co., 403 Mass. 50, 525 N.E.2d 1305, 1310-11 (1988); Uloth v. City Tank Corp., 376 Mass. 874, 384 N.E.2d 1188, 1193 (1978). It follows, we think, that a design defect case premised on breach of warranty is, in Massachusetts, similarly dependent on proof of the existence of a safer alternative design--a design which reasonably could, or should, have been adopted. Cf. Collins v. Ex-Cell-O Corp., 629 F. Supp. 540, 543 (D.Mass.1986) ("the feasibility of alternative and possibly safer designs is a relevant inquiry in the warranty context" under Massachusetts law), aff'd mem., 815 F.2d 691 (1st Cir.1987).

In this case, no such proof was advanced. Appellant's design initiative rested not on the contention that appellees' cigarettes could have been made safer, but solely on the considerably different contention that the cigarettes were defective because the risks inherent in their consumption, "good" tobacco notwithstanding, outweighed their social utility. Under the decided cases, more was required.

We decline appellant's invitation to expand Massachusetts law past its present borders and into hitherto unexplored terrain. It is illogical to say that a product is defective in its generic form when "defect" has historically been measured in reference to the availability, or at least the feasibility, of safer alternatives. As the court below noted, Kotler II, 731 F. Supp. at 53, the caselaw furnishes no encouragement for so venturesome a step. Accord Viola v. American Brands, Inc., 85-2496-WD, 1989 WL 234525, slip op. at 2 (D.Mass. July 14, 1989); Herlihy v. R.J. Reynolds Tobacco Co., 85-3888-MA, slip op. at 5, 1988 WL 73434 (D.Mass. June 27, 1988); see also Roysdon, 849 F.2d at 236 (defendant's cigarettes must present risks greater than those known to be associated with smoking in order to be defective under Tennessee law); Restatement (Second) of Torts, Sec. 402A, comment (i) ("Good tobacco is not unreasonably dangerous merely because the effects of smoking may be harmful."); cf. Kyte, 556 N.E.2d at 1029-30. Summary judgment was appropriately granted with respect to the design defect claims.

Appellant has challenged jury selection on three grounds: (1) improper administration of peremptory challenges, (2) abuse of discretion in excusing jurors for cause, and (3) unfairness in the provision of jury lists. We find these objections meritless.

A. Peremptory Challenges.

In a federal civil suit, each side must be permitted three peremptory challenges. 28 U.S.C. Sec. 1870 (1988). Neither the statute nor the district court's local rules specify how peremptory challenges are to be administered, thus leaving the formulation of appropriate procedures to the trial judge's discretion. See United States v. Morris, 623 F.2d 145, 151 (10th Cir.), cert. denied, 449 U.S. 1065, 101 S. Ct. 793, 66 L. Ed. 2d 609 (1980); United States v. Haldeman, 559 F.2d 31, 80 (D.C.Cir.1976) (per curiam), cert. denied, 431 U.S. 933, 97 S. Ct. 2641, 53 L. Ed. 2d 250 (1977). The court below allowed such challenges to be cast alternately, with a replacement juror being drawn immediately upon the challenge's exercise.8 We review what transpired.

First, the judge questioned potential jurors for bias and entertained challenges for cause. When the dust settled, six presumptively impartial jurors (Nos. 49, 126, 167, 197, 209 and 245) were in the box. The panel was then exposed to peremptory challenges. The appellant struck Juror 126 who was replaced by Juror 26. ATC struck Juror 167 (replaced by Juror 188). At that juncture, appellant's attorneys passed on a further strike, stating that they were satisfied with the aggregate. ATC thereupon excluded Juror 209. The replacement juror, No. 144, was challenged by appellant. Juror 12 was seated. ATC then struck Juror 188 and Juror 308 was called.

When appellant tried to unseat this newest talesman, claiming she had not been given the opportunity to use all three of her peremptories, the court blocked the attempt on the ground that passing on the second round of strikes (appellant instead announcing her satisfaction with the provisional panel) was tantamount to waiving one challenge. Appellant's counsel did not contest the court's reasoning; rather, asserting that they were unaware of this use-it-or-lose-it rule, they asked "the Court's indulgence that both sides be given one additional challenge just on the basis of fairness." Over ATC's objection, the court obliged. Appellant used its additional challenge to oust Juror 308. After a replacement was drawn, Juror 245 was removed by ATC. Juror 292 was then seated without objection. Once the jury of six was constituted, alternate jurors were selected (at which time additional peremptories were allotted).

On appeal, Kotler claims that it was error for the court to deem her inaction during the second round of strikes as waiving an unused peremptory challenge. The argument, essentially, is that the use-it-or-lose-it rule impermissibly denies a party the full benefit of the three peremptories allowed by statute. Alternatively, Kotler theorizes that, even if this "deemed waiver" constituted a permissible means of administering the empanelment, she was not adequately placed on notice of the practice. Because the district court granted appellant's remedial request, however, we have no occasion to dwell on these assertions. We explain briefly.

When the problem surfaced, appellant suggested a remedy--and the court granted it in full. Appellant then used her extra challenge to eliminate the unwanted venire person, Juror 308. There is nothing in the record to indicate that any seated member of the jury was unsatisfactory to appellant or that she would have employed yet another peremptory challenge, if she possessed one, to remove any other juror. Such a showing is required for reversal. See Goldstein v. Kelleher, 728 F.2d 32, 38 (1st Cir.) ("In the absence of any indication that the jury selected was unsatisfactory to plaintiff at the time chosen, we are unwilling to reverse the verdict of a jury that to all appearances was disinterested, competent and suitable."), cert. denied, 469 U.S. 852, 105 S. Ct. 172, 83 L. Ed. 2d 107 (1984).

While it is true that ATC in effect acquired an extra challenge, and used it, appellant cannot belatedly assign error on this score. After the district court accepted appellant's proposed anodyne, no objection to the court's granting of an additional challenge to ATC was interposed. It is, therefore, too late for appellant to cry foul. See Anderson v. Cryovac, Inc., 862 F.2d 910, 918 (1st Cir.1988) (a party's "active advocacy" of a faulty procedure subsequently adopted by the judge waives the point for purposes of appeal); Real v. Hogan, 828 F.2d 58, 62 (1st Cir.1987) (contemporaneous objection needed to preserve claim that peremptory challenges were used on a discriminatory basis); United States v. Griffin, 818 F.2d 97, 100 (1st Cir.) ("a party cannot sleep upon perceptible rights, but has an obligation to alert the district judge to error-in-the-making"), cert. denied, 484 U.S. 844, 108 S. Ct. 137, 98 L. Ed. 2d 94 (1987); United States v. Gonzalez Vargas, 585 F.2d 546, 547 (1st Cir.1978) (per curiam) (appellants proposed a certain compromise on peremptory challenges and therefore acquiesced in it). Appellant can hardly come before this court now, after an adverse jury verdict, and complain that the handling of peremptory challenges was unfair when the district court granted appellant precisely the relief that she requested.9

B. Challenges for Cause.

Our review of the empaneling shows that the trial court took great care in determining whether potential jurors could impartially evaluate evidence on the issues in the case. The court skillfully balanced the need for open-mindedness with a recognition that jurors need not come to a trial as entirely empty vessels, bereft of any sliver of an opinion touching upon the issues to be decided. See United States v. Angiulo, 897 F.2d 1169, 1182 (1st Cir.), cert. denied, --- U.S. ----, 111 S. Ct. 130, 112 L. Ed. 2d 98 (1990); United States v. Brooklier, 685 F.2d 1208, 1223 (9th Cir.1982) (en banc), cert. denied, 459 U.S. 1206, 103 S. Ct. 1194, 75 L. Ed. 2d 439 (1983); see also Irvin v. Dowd, 366 U.S. 717, 723, 81 S. Ct. 1639, 1643, 6 L. Ed. 2d 751 (1961) ("It is sufficient if the juror can lay aside his impression or opinion and render a verdict based on the evidence presented in court.").

The single perfected objection taken by Kotler illustrates the flimsiness of her argument on this point. She urges that it was error for the trial judge, who originally indicated that all smokers would be excluded for cause, to seat a juror who smoked "a couple of cigarettes per week." The objection borders on the whimsical.10 The juror was closely questioned about her views and the judge made a reasoned determination that her casual smoking did not make her a "smoker" in the sense that she might be partial on the issues in the case.

Appellant also berates the judge for consistently excluding those jurors who stated that their views on the relationship between smoking and lung cancer would not be changed regardless of the testimony. Passing the lack of contemporaneous objections, suitably focused, the criticism is baseless. Trial lawyers are by nature a suspicious lot and see ghosts under virtually every bed. That is the lawyer's prerogative--but expecting the judge to credit such spectral visions asks too much. We think it fanciful to premise harmful error on the judge's decision to give credence both to jurors who stated they could be impartial on this point and to others who stated they could not. While a judge ought not blindly to accept jurors' assurances or reflexively to brush aside counsel's qualms, there must be strong evidence that the judge was hoodwinked or his judgment grossly in error before an appellate court will interfere. See Angiulo, 897 F.2d at 1183; United States v. McNeill, 728 F.2d 5, 10 (1st Cir.1984).

To sum up, a trial court enjoys wide discretion in ruling on challenges for cause. See Dennis v. United States, 339 U.S. 162, 168, 70 S. Ct. 519, 521, 94 L. Ed. 734 (1950); Real, 828 F.2d at 62; Brooklier, 685 F.2d at 1223. In the absence of manifest juror prejudice, we will not set aside a judge's actions in empaneling a jury which he reasonably considers to be suitable and impartial. See McNeill, 728 F.2d at 9. So here.

C. Provision of Jury Lists.

Appellant challenges the lower court's decision to grant both parties access to jury lists on the Friday before the Monday on which the empanelment took place. Appellant raised an objection to this decision at the beginning of the Monday session. The judge ascertained that neither side had made any attempt to contact potential jurors. ATC did admit, however, that its agents drove through some neighborhoods to get a better "feel" for the jurors' backgrounds.

We need not linger over this buzznacking. Appellant's argument, made without meaningful citation of authority, boils down to the notion that providing the lists was unfair not because anything improper occurred, but because ATC alone had the financial backing to make use of the jury roster. The contention is frivolous. It is rare that opposing parties have the same wherewithal with which to pursue their causes--yet nothing in our system of jurisprudence guarantees that all litigation budgets must be equal. Trials are not rendered unfair in any cognizable sense merely because of disparity in the parties' resources. See, e.g., Caroline T. v. Hudson School Dist., 915 F.2d 752, 757 (1st Cir.1990) ("[t]here is no constitutional prohibition on permitting one party to employ litigation techniques which the other party cannot afford or chooses not to use, so long as those techniques are used for legitimate purposes").

The most formidable weapon in appellant's armamentarium involves the trial court's removal of an issue from jury consideration.

A. What Transpired.

As mentioned above, plaintiff's case went to trial only against ATC and only on the causes of action for breach of warranty and negligence stemming from failure to warn (pre-1966). During trial, the district court granted a directed verdict in ATC's favor on the warranty claim. The judge--who simultaneously refused to direct on the negligence claim--ruled that, to prevail in warranty, appellant had to prove that failure to warn was the sole proximate cause of the harm. He believed that no reasonable jury could so conclude.

Appellant argues that the instructed verdict was improperly granted because the judge imposed too great a burden on her. She importunes that Massachusetts law does not require a product to be the sole proximate cause of a plaintiff's harms in order to ground a breach of warranty suit. We concur. Under Massachusetts law, a warranty plaintiff need only prove proximate cause in the ordinary sense.11 See, e.g., Fernandes v. Union Bookbinding Co., 400 Mass. 27, 507 N.E.2d 728, 734 (1987) (stating the proximate cause requirement for breach of warranty); Ferragamo v. Massachusetts Bay Transp. Auth., 395 Mass. 581, 481 N.E.2d 477, 484 (1985) (allowing recovery in warranty despite fact that jury found plaintiff guilty of comparative negligence); Barry v. Stop and Shop Cos., 24 Mass.App.Ct. 224, 507 N.E.2d 1062, 1065 (1987) (proximate cause in design defect case must be analyzed "in the customary or usual sense"). ATC, to its credit, conceded at oral argument in this court that the trial judge misperceived the law and erred in directing a verdict on this basis.

B. The Reversible Error Standard.

Recognition of the mistake does not end our inquiry. Ordinarily, error demands reversal only if it "affect[s] the substantial rights of the parties." Fed.R.Civ.P. 61. In this case, appellant's negligence claim was submitted to the jury and a general verdict in ATC's favor eventuated. Not surprisingly, ATC argues that the court's intervention on the warranty count, though erroneous, was inconsequential; the jury's verdict on the negligence count necessarily foretold that ATC would also have prevailed on the warranty claim. Phrased in more neutral terms, if the instructions that the jury should have received on breach of warranty were materially equivalent to, or subsumed by, the instructions actually received on negligence, then the failure to submit the warranty count to the jury was rendered harmless by the verdict on the negligence count. Cf. Allen v. Chance Mfg. Co., 873 F.2d 465, 469 (1st Cir.1989) (error in jury instructions mandates retrial only if the error could have affected the outcome of the jury's deliberations); Flaminio v. Honda Motor Co., 733 F.2d 463, 467-68 (7th Cir.1984) (no error in withholding instruction which, "if given, would not have put before the jury a concept of strict liability distinct from the negligence instruction the judge did give").

C. The Legal Framework.

As diversity jurisdiction demands, we turn to Massachusetts law to determine if, notwithstanding its verdict on the negligence count, the jury might have found for appellant on the breach of warranty count. Because the warranty action in Massachusetts is akin to an action in strict liability in many other jurisdictions, see Allen, 873 F.2d at 466 n. 2, we can, when we encounter hypoplasia in local decisional law, look profitably for guidance in the caselaw of these other jurisdictions. See Back, 378 N.E.2d at 969 ("we find the strict liability cases of other jurisdictions to be a useful supplement to our own warranty case law"); see also Anderson v. Owens-Illinois, Inc., 799 F.2d 1, 2 (1st Cir.1986); see generally Ryan, 916 F.2d at 732 (where state law indistinct, diversity court should consider the caselaw of other jurisdictions).

As a general matter, the causes of action for negligence and breach of warranty in a Massachusetts products liability case are not identical. See Colter, 525 N.E.2d at 1313; Hayes, 462 N.E.2d at 275. Under a negligence theory, the primary focus is on the defendant's conduct. Breach of warranty, however, trains the focus mostly on the product. This difference in perspective is concretely reflected in the differing proofs a party must adduce to maintain (or defend against) claims under the two theories. In design defect cases, the difference between negligence and warranty may be especially stark, for a warranty action can be prosecuted without inquiring into the reasonableness of the manufacturer's behavior. Put another way, so long as the product is defective, the manufacturer is liable for breach of warranty, even if the manufacturer acted reasonably, indeed, carefully, in making the product.

Of course, some cases will do double duty; if a product defect is caused by the manufacturer's negligence, there is necessarily a breach of warranty. In the last analysis, then, a manufacturer's breach of its duty of care is, ceteris paribus, a sufficient, although not a necessary, condition to the imposition of liability for breach of warranty. The SJC has summed it up authoritatively:

A defendant in a products liability case in this Commonwealth may be found to have breached its warranty of merchantability without having been negligent, but the reverse is not true. A defendant cannot be found to have been negligent without having breached the warranty of merchantability.

While the Hayes rule applies to failure-to-warn cases as well as to design defect cases, the distinction between negligence actions and breach of warranty actions blurs considerably when the suit emanates from an alleged omission adequately to warn of a product's dangers. Accord Werner v. Upjohn Co., 628 F.2d 848, 858 (4th Cir.1980), cert. denied, 449 U.S. 1080, 101 S. Ct. 862, 66 L. Ed. 2d 804 (1981). In many instances, there is no need to treat the two theories differently at all. See Laaperi v. Sears, Roebuck & Co., 787 F.2d 726, 728-29 n. 1 (1st Cir.1986) (describing "factual questions regarding duty to warn and causation underlying both the negligence and warranty claims" as "essentially identical"); MacDonald v. Ortho Pharmaceutical Corp., 394 Mass. 131, 475 N.E.2d 65, 68 (similar), cert. denied, 474 U.S. 920, 106 S. Ct. 250, 88 L. Ed. 2d 258 (1985). Either way, liability is premised on the existence of a similarly formulated duty to warn:

"[A] manufacturer of a product, which the manufacturer knows or should know is dangerous by nature or is in a dangerous condition," is under a duty to give warning of those dangers to "persons who it is foreseeable will come in contact with, and consequently be endangered by, that product."

MacDonald, 475 N.E.2d at 68 (quoting H.P. Hood & Sons, Inc. v. Ford Motor Co., 370 Mass. 69, 345 N.E.2d 683, 688 (1976)); see also McIsaac v. Didriksen Fishing Corp., 809 F.2d 129, 133 (1st Cir.1987) (negligence theory); Anderson, 799 F.2d at 4 (warranty theory). The duty to warn, whether in negligence or warranty, embraces only those dangers of which the average consumer of the product would not be aware. See Anderson, 799 F.2d at 4 (no warranty duty to warn of generally known dangers); Laaperi, 787 F.2d at 730-31 (no negligence duty to warn where users ought to realize product's dangerous condition); Lowery v. AIRCO, Inc., 725 F. Supp. 82, 84 (D.Mass.1989) (no duty to warn under either theory where danger is readily apparent).

Lest we oversimplify, we add that, although the formulation of the duty to warn is similar in both negligence and warranty,13 the inquiry into breach of the duty can be quite dissimilar. For example, where an attempt has been made to warn of a particular hazard, the jury could conceivably conclude "that, although the manufacturer was not unreasonable and hence was not negligent in the warning provided, the warnings were still insufficient to render the product suitable for its foreseeable uses." Allen, 873 F.2d at 466 n. 2. The Allen example is a practical illustration of the principle that negligence focuses on the manufacturer's behavior while warranty focuses on the product's characteristics. The principle creates no distinction, however, in cases, like this one, which involve not an allegedly inadequate warning, but a total failure to warn at all.14 This is so because, given a duty to warn, an outright failure to warn is per se negligent, as well as being, by definition, a breach of warranty.15

We come now to the central question of whether there is any significant difference between the duty to warn under negligence as opposed to warranty. As we have seen, since the duty to warn under both theories is a function of, chiefly, what the manufacturer knows or should know about the danger necessitating the warning,16 the two theories often seem similar. In fact, other courts have concluded that the strict liability duty to warn is so largely informed by negligence principles as to be essentially identical to the negligence duty. See, e.g., Hull v. Eaton Corp., 825 F.2d 448, 454 (D.C.Cir.1987) (per curiam) (threshold question of duty to warn under District of Columbia law is the same in negligence and warranty cases); Gonzalez v. Volvo of America Corp., 752 F.2d 295, 300 (7th Cir.1985) (per curiam) ("This similarity exists because the language and concepts of reasonableness which determine unreasonable risk under Section 402A are the same concepts used in a negligence case. The principles of negligence law are particularly appropriate in Section 402A cases based on failure to warn.") (applying Indiana law); Flaminio, 733 F.2d at 466 ("the strict liability duty to warn, as usually formulated ... is hard to distinguish in practice from the duty to warn imposed by a negligence standard") (applying Wisconsin law).

We pause before putting full stock in these decisions. After all, we ourselves have said that, as applied in Massachusetts, section 402A imposes a standard higher than due care. See Anderson, 799 F.2d at 4 ("the seller is required to give warning against [a danger that is not generally known], if he has knowledge, or by the application of reasonable, developed human skill and foresight should have knowledge, of the presence of the ingredient and the danger") (quoting Restatement Sec. 402A, comment j). The Anderson court found this standard to be more rigorous than the negligence standard of reasonable care because it imputed to the manufacturer expert knowledge rather than merely trade knowledge (that is, the level of knowledge enjoyed by a prudent manufacturer in analogous circumstances). See id.; see also Back, 378 N.E.2d at 970-71 (discussing, in a design defect case, the difference between imputing expert knowledge to a manufacturer on the one hand, and holding a manufacturer on the other hand to the expectable knowledge of a reasonably prudent manufacturer in like circumstances).

Having paused, we proceed. We do not find Anderson at odds with the conclusion of other courts that the strict liability duty to warn is essentially a negligence duty. Given a state of knowledge, both theories only impose a duty if a danger is knowable or reasonably foreseeable. Anderson highlighted that the state of knowledge from which the determination of "knowability" is made can differ under the two theories; the case stands for the proposition that the state of knowledge is more broadly defined in a warranty case (viz., "expert" knowledge) than in a negligence case (viz., "trade" knowledge). Be that as it may, the Anderson court did not purpose to hold a manufacturer to a perfectly developed state of knowledge concerning the risks of his product at all times, whether scientifically knowable or not. The upshot is that, since even expert knowledge does not rise to the level of perfectly developed knowledge, a showing must still be made that the state of knowledge supports the conclusion that a certain risk is knowable--or what the law calls "reasonably foreseeable"--in order to catalyze a cause of action under either theory.

Anderson epitomizes our point. That case involved a suit against an asbestos company. On appeal, the unsuccessful plaintiff argued that the trial court erred in allowing state-of-the-art evidence in respect to the breach of warranty claim. Anderson, 799 F.2d at 1-2. Massachusetts law, plaintiff said, imputed to the seller full awareness of the risks of the product at all times, id. at 2--what we call above "perfectly developed" knowledge. Plaintiff placed heavy reliance on the words of the SJC:

A defendant vendor is held to that standard regardless of the knowledge of risks that he actually had or reasonably should have had when the sale took place. The vendor is presumed to have been fully informed at the time of the sale of all risks. The state of the art is irrelevant, as is the culpability of the defendant.

Hayes, 462 N.E.2d at 277. We rejected Anderson's plaint, recognizing a state-of-the-art defense and holding that there could be no liability under Massachusetts law for failing to warn of dangers which were not known to exist in the scientific community. See Anderson, 799 F.2d at 3-4; see also In re Massachusetts Asbestos Cases, 639 F. Supp. 1, 3-5 (D.Mass.1985) (later proceeding, Anderson, supra ); Collins, 629 F. Supp. at 543. We read Anderson as confirming that the duty to warn in a breach of warranty action is indelibly infused with negligence principles. Accord Moran v. Johns-Manville Sales Corp., 691 F.2d 811, 814 (6th Cir.1982) ("duty to warn attaches, not when scientific certainty is established, but 'whenever a reasonable man would want to be informed of the risk in order to decide whether to expose himself to it' ") (citation omitted) (applying Ohio law); Borel v. Fibreboard Paper Products Corp., 493 F.2d 1076, 1088 (5th Cir.1973) ("The requirement that the danger be reasonably foreseeable, or scientifically discoverable, is an important limitation of the seller's [strict] liability.... The requirement of foreseeability coincides with the standard of due care in negligence cases in that a seller must exercise reasonable care and foresight to discover a danger in his product and to warn users and consumers of that danger. ") (applying Texas law), cert. denied, 419 U.S. 869, 95 S. Ct. 127, 42 L. Ed. 2d 107 (1974); Basko v. Sterling Drug, Inc., 416 F.2d 417, 426 (2d Cir.1969) (risk must be knowable or reasonably foreseeable at relevant times) (applying Connecticut law); see also Brochu v. Ortho Pharmaceutical Corp., 642 F.2d 652, 657 n. 10 (1st Cir.1981) (negligence principles may be relevant in determining knowledge necessary to trigger duty to warn in warranty case brought under New Hampshire law).

"Knowledge" and "knowable," then, are the keys to this kingdom. When we use those terms in the present context, we signify that at a given point in time there may be vectors of information of varying magnitudes and directions regarding a particular product danger. Under a warranty theory, all of these vectors of information, so long as credibly proven, constitute the knowledge imputed to the manufacturer. Under a negligence theory, only those credibly proven vectors of information which (a) the defendant actually knows, or (b) the defendant, as a reasonably prudent manufacturer, should know, constitute the imputed knowledge. Whether a denominated danger is "knowable" is in turn a function of the particular set of imputed information vectors and is determined, under either theory, by inquiring whether, given the level of imputed knowledge, the existence of the danger was reasonably foreseeable. Accordingly, the difference in determining knowledge, not the difference in determining "knowability," is what, under Anderson, subjects the duty to warn to a standard higher than due care in a warranty milieu. See In re Asbestos, 639 F. Supp. at 3 n. 4; see also Anderson, 799 F.2d at 2 (in warranty action, "Massachusetts law requires a seller to warn only of reasonably foreseeable or scientifically discoverable dangers"). In the case at bar, where the manufacturer failed to warn at all of a danger and has not advanced affirmative defenses, the only relevant difference between negligence and breach of warranty would relate to the depth of knowledge that the law imputed to the manufacturer at a particular point in time.17

D. Fitting the Framework to the Case.

Against this backdrop, we turn to the record to determine whether the district court's erroneous withdrawal of the breach of warranty count from jury consideration was prejudicial. At trial, Kotler faced a formidable task to make out a prima facie case for failure to warn on either theory (negligence or breach of warranty). She had to prove, first, that ATC had a duty to warn of the dangers of cigarettes before 1966--a burden which required her to show that at some fixed point in time ATC, but not the average consumer, knew or should have known of the potential health hazards. Once the duty to warn was established, and it was shown that no warnings were given, it still remained for appellant to prove that the failure to warn caused decedent's illness and death. The causation inquiry entailed two parts, necessitating proof that (1) ATC's cigarettes were a cause in fact of the harm, and (2) ATC's failure to warn was a proximate or "legal" cause of the injuries. See generally Swift v. United States, 866 F.2d 507, 509-10 (1st Cir.1989) (explicating difference between actual causation and proximate causation under Massachusetts law). Each of these inquiries is the same whether done in connection with a negligence theory or a breach of warranty theory. See Lowery, 725 F. Supp. at 85. In addition, proof of what the average consumer knew at any set point in time is invariant with respect to the theory of liability, see supra note 16; and proof of what the decedent himself knew at any time is invariant, albeit germane with respect to proximate cause. See id. at 84-85. It is readily apparent, then, that the only relevant difference between the two theories on the facts of this case concerns the imputation of knowledge; after all, we can assume the jury's rationality--and a rational jury, to be consistent on these facts, would have to resolve all the relevant issues identically, whether under negligence or warranty, save only the issue pertaining to whether, pre-1966, the health hazards associated with smoking were knowable by ATC.

Taking all this into account, our scrutiny must be directed to those portions of the charge relating to the duty to warn. If that duty was posited in terms that met the more exacting warranty standard for imputed knowledge, then the court's failure to submit both counts to the jury was inconsequential and the verdict for the defendant necessarily betokened plaintiff's failure to establish a warranty case.18

Before examining the district court's instructions on knowledge to determine if those instructions were equivalent to the instructions due a properly advised jury on breach of warranty, it seems useful to review the instruction found proper in Anderson respecting knowledge vis-a-vis a claimed breach of warranty for failure to warn. There, the district court charged:

[T]he duty to warn extends only to such dangers--to such dangers or defects about which the manufacturer either actually knew or about which it reasonably should have known. Now should have known here means that a manufacturer is held to that level of knowledge which the experts in the particular industry had or in view of the state of medical and scientific knowledge in general should have had at any particular time.

Anderson, 799 F.2d at 2. We upheld that instruction and believe it is equally apposite here. Cigarettes, like asbestos, do not pose open and obvious dangers. If such dangers were unknowable at some point in time, no effective warning about them could have been given. See id. at 4. Ergo, if the district court were to have ceded Kotler her full entitlement and charged the jury on breach of warranty, the appropriate charge, like the one approved in Anderson, would have resonated the tenets embedded in Restatement Sec. 402A (and particularly, comment j) rather than the language of the Hayes dictum, as plaintiff might wish.

Serendipitously, the court below, purposing to discuss negligence, gave a charge that was to all intents identical to the Anderson charge, instructing the jury to impute expert knowledge.19 Thus, while directing a verdict for ATC on the warranty claim, the court in effect took from ATC with the right hand what it had bestowed with the left, sending to the jury a claim fully equivalent to breach of warranty. Phrased another way, the charge as rendered, although it overstated the duty on negligence, contained all that appellant could properly have demanded on warranty. Hence, the erroneous allowance of the directed verdict was, in the end, harmless.

No different conclusion is required because the judge also instructed the jury about the manufacturer's reasonableness. A warranty instruction may contain such language when the knowledge that is imputed to the manufacturer about the existence of a danger is contradictory or uncertain, and an assessment must be made as to whether the existence of the danger was indeed "knowable." Cf., e.g., Moran, 691 F.2d at 814. That was precisely the case here; at trial, various experts testified about the existence of different studies at different times, discussing the possible connection between cigarettes and lung disease. The evidence presented a complex picture of conflicted, sometimes speculative, information. Against this mise-en-scene, an instruction to the effect that evaluation of this evidence centered around reasonableness was a proper statement of law, under warranty theory as well as under negligence theory.

In sum, then, the court below, on the negligence count, gave what amounted to a warranty instruction with respect to the imputation of knowledge. Since all other elements of the negligence and warranty claims were congruent, and the only meaningful distinction between the two related to the standard for imputing knowledge, this charge ensured that the jury's finding on the negligence count accurately and inevitably foretold what would have transpired if the warranty count had also been submitted to the jury. Thus, the lower court's erroneous refusal to send the warranty claim to the jury did not deprive the plaintiff of the chance to have the factfinders consider some other or different concept which they might have resolved in a way more favorable to her. It follows inexorably that appellant's substantial rights were not prejudiced.20 There is no cause for reversal.

Last and least is appellant's assertion that the district court used prejudicial language when instructing the jury. In particular, appellant castigates the court for characterizing the issue of causation as complex and difficult. We find this caterwauling to comprise much ado about very little.

"The charge must, of course, be considered as a whole, not in isolated bits and pieces." United States v. Cintolo, 818 F.2d 980, 1003 (1st Cir.), cert. denied, 484 U.S. 913, 108 S. Ct. 259, 98 L. Ed. 2d 216 (1987). After reviewing the entirety, we have difficulty fathoming how Kotler can seriously contend that the challenged snippets were prejudicial.21 On any fair reading of the remarks, taken in context, the trial judge was merely offering words of obligation, telling the jurors that the causation inquiry, though labyrinthine, must nonetheless be confronted conscientiously. Appellant's version--that the jury would probably infer from such comments that the causation inquiry was difficult because her proof was scanty--strikes us as chimerical.

Where, as here, the judge adhered to appropriate standards of fairness and impartiality, he was not required to "function as some bloodless automaton." United States v. Polito, 856 F.2d 414, 418 (1st Cir.1988). Having been scrupulous throughout the trial and the charge in respecting the jury's independence and domain, it was permissible for the presider to express empathy for the jurors' plight and concern for their steadfastness of purpose. Rather than seeming a sarcastic put-down of plaintiff's case, we think it likely that the court's comments comprised a useful motivational device. At any rate, we think it certain that these comments were not "outside the bounds of proper judicial conduct." Aggarwal v. Ponce School of Medicine, 837 F.2d 17, 22 (1st Cir.1988).

We need go no further. We have carefully considered appellant's myriad objections in light of the voluminous record, the forceful briefs, the sterling arguments, and the applicable law. Finding no reversible error, we conclude that the judgment below must be

Affirmed.

The basic issue is one of negligence, and negligence means a failure to use due care, failure to exercise a duty of due care to someone, namely the Plaintiff; in this case, George Kotler. A manufacturer has a duty to exercise due care with respect to the people who are likely to purchase its product. It must exercise due care for their safety.

Due care is the care that a reasonably prudent manufacturer should use under the circumstances that existed at the time to avoid reasonably foreseeable injury to users of its product. Due care is to be viewed under the circumstances which exist at the time which Mr. Kotler was using this Defendant's product. Due care does not require a manufacturer to meet a standard of perfection, but it does require that the manufacturer exercise reasonable care under all of the circumstances.

The manufacturer is charged with knowledge about his product which existed at the relevant time, not with information which may later be developed. The manufacturer, however, is charged with familiarity of all the available information, from whatever source, which bears on the safety of its product. It must take, has affirmative duty, to keep abreast of responsible literature on the subject.

It doesn't mean that if there is something that appears in the National Inquirer [sic] or some such paper that the manufacturer has to bend itself all out of shape in response. But if there is legitimate, responsive information, or rather, legitimate information from responsible sources which exists in the community, which exists in the medical, scientific or industrial community, then the manufacturer is charged with familiarity with all of that information, insofar as it bears on the safety of that manufacturer's product.

The manufacturer is not an insurer of its product, but it has a duty to exercise reasonable care to warn potential users of reasonably foreseeable hazards associated with the use of its product.

The key word here is "reasonable," and a consideration bearing on reasonableness is the seriousness of the risks that are involved, and, of course, the risk of death being more or less the ultimately serious risk. A test of reasonableness is not necessarily whether there has been conclusive evidence of the risk. If there is evidence which would induce a reasonable person to recognize the existence of a serious risk, then there is a duty on the part of the manufacturer to warn the consumer about that risk....

For the plaintiff to prevail, you have to be satisfied by a preponderance of the evidence that at some time prior to January of 1966 the Defendant had information which would oblige a reasonable manufacturer to give a warning that there was enough risk, and the relationship of the level of risk--rather, the level of information to the seriousness of the risk should have prompted a reasonable manufacturer to give a warning.

Also, it must be satisfied by a preponderance of the evidence that those risks were not generally recognized in the community, and the evidence on that is a very mixed bag, as you've heard, as to what was in the papers. There was a lot of reference to the statements of people that there wasn't any risk. You could infer, I suppose, from the fact somebody is saying we dispute this risk, that somebody else was saying there was a risk.