dissenting in part:

Although I concur in Parts II and III of the majority opinion as well as much of Part I, I must dissent from the conclusion in Part I that Denholm may not appeal the pretrial order excluding evidence regarding lost royalties and loss of reputation because he accepted remittitur. The majority today adopts a sweeping and unnecessarily mechanistic view of which issues on appeal are sufficiently related to a remittitur order to trigger application of the longstanding rule that a plaintiff in federal court who accepts an order of remittitur may not appeal that order. See Donovan v. Penn Shipping Co., 429 U.S. 648, 650 , 97 S.Ct. 835, 837 , 51 L.Ed.2d 112 (1977) (per curiam). Persuaded by the Fifth Circuit’s decision in Lanier v. Sallas, 777 F.2d 321, 325 (5th Cir.1985), the majority opinion embraces an analysis that focuses on “whether each issue on appeal is a separate and distinct cause of action from the subject of the remittitur.” Majority Opinion at 360 (emphasis added). Unfortunately, this analysis oversimplifies both the Lanier decision and the nature of the events which occurred below in the instant case, leading to an absurd result.

I

Contrary to what the majority implies, the Fifth Circuit in Lanier never stated that only “separate and distinct cause[s] of action” from the subject matter of a remit-titur order may be appealed. Instead, the Fifth Circuit began its analysis with the uncontroversial proposition that in spite of a plaintiff’s acceptance of remittitur on one cause of action, he may appeal the district court’s ruling on a separate and distinct cause of action. See id. at 325 . 1 The court then proceeded to acknowledge “the slightly broader proposition that ‘[a] plaintiff may, however, appeal from other parts of the judgment,’ ” id. (quoting 6A J. Moore, Moore’s Federal Practice ¶[ 59.08[7], at 59-205 to 206 (1984)), without stating which formulation was preferable. Indeed, the Lanier court cited Ohio-Sealy Mattress Mfg. Co. v. Sealy, Inc., 585 F.2d 821 (7th Cir.1978), cert. denied, 440 U.S. 930 , 99 S.Ct. 1267 , 59 L.Ed.2d 486 (1979), as an example of the “broader proposition” regarding appealability, and then proceeded to distinguish rather than criticize it. La-nier thus cannot be read as establishing the appealability only of “separate and distinct cause[s] of action.”

Lanier acknowledged with apparent approval that the Sealy court “permitted a plaintiff who had accepted a 50% remittitur on the award of damages to appeal the district court’s denial of injunctive relief.” 777 F.2d at 325 . Thus, the situation in Sealy was similar to the situation in the instant case in that both involve a plaintiff suing on a single cause of action yet seeking two types, or measures, of damages. 2

The same could be said, of course, of the situation in Lanier itself, where a plaintiff suing on a legal malpractice theory sought *364 both compensatory and punitive damages. Yet this only means that the Lanier court did a rather poor job of distinguishing Sealy. The Lanier court stated that whereas in Sealy the question of injunctive relief on an antitrust theory could be decided separately from the question of damages on the same theory, the question of punitive damages could not be tried separately from the question of actual damages. Id. at 325 . The Lanier court’s conclusion was grounded on the fact that “the punitive-damage claim is dependent on establishing liability for compensatory damages and, in addition, establishing malice or some other aggravating circumstance.” Id. at 325-26 . From this the court concluded that the punitive damages claim “cannot be tried in isolation, and remand of that claim would necessitate retrial of the entire case.” Id. at 326 . Yet if this is so, then why can the issue of injunctive relief be decided apart from the issue of compensatory damages, since both require an initial finding of liability? 3 Surely the Sealy court was correct in starting with the given of antitrust liability. The Lanier court, by contrast, never explained why a jury would be unable to decide a claim for punitive damages in isolation, starting with an instruction to the effect that actual damages have already been found and hence the jury’s only inquiry is to consider the possibility of malice.

In any event, Sealy is much closer to the instant case than Lanier is. For both Sealy and this case concern a single theory of liability with two separate measures of damages, either of which can be awarded without regard to whether the other is awarded. Lanier , by contrast, concerns a single theory of liability with two interrelated measures of damages, one of which cannot be awarded unless the other is first awarded.

Perhaps the difference, is there is one, between Lanier and Sealy is that in Lamer the evidence that would be presented in a claim for punitive damages would necessarily consist of largely the same evidence that had already been considered with regard to actual damages, whereas the important factors in a claim for injunctive relief in Sealy would have to go much further, encompassing traditional equitable considerations. Houghton Mifflin’s attorney focused on this difference at oral argument; she explained Lanier as establishing that when consideration of an issue will require presentation of the same evidence that was considered in the issue that was the subject of the remittitur order, the new issue cannot be appealed. 4 But even if this reading of Lanier is correct, surely this case more closely resembles Sealy than it does Lanier , for Denholm is seeking to introduce entirely new evidence about royalties and reputation. He is not, as was the case with the appellant in Lanier , attempting to have the jury draw inferences of malice from the very same evidence that was considered in the action for compensatory damages.

II

Once the majority’s misguided reliance on Lanier is placed to one side, it is easy to see why the “separate and distinct cause of action” requirement works a pointlessly harsh result. The district judge in this case granted appellees’ motion in limine to exclude the presentation of Denholm’s evidence concerning damages from lost royalties and loss of reputation. At trial, Den-holm made an offer of proof regarding lost royalty income, which the judge denied. At this point the judge specifically informed the jury that evidence of lost royalties or lost reputation would not be admitted. The judge’s jury instructions also admonished the jury not to award damages for potential lost royalties. The only evidence actually presented at trial of Den-holm’s damages was testimony to the ef- *365 feet that the value of the time Denholm had devoted to the project was $20,000. Nonetheless, the jury awarded Denholm $100,000, an amount that the district court reduced on appellees’ motion for remittitur to $20,000.

Surely the jury erred, since it had not been presented with any evidence whatsoever regarding Denholm’s damages apart from the testimony regarding $20,000. Yet I fail to see how the majority can fairly characterize the in limine motion as directly related to the remittitur order accepted by Denholm. Although the remittitur order awarded a new trial generally “on the issue of damages[ ] unless plaintiff accepts a remittitur of the verdict and judgment,” only by wilfully blinding itself to context can the majority disclaim that this order necessarily referred only to what had been presented at trial: damages for the value of Denholm’s time spent preparing his book series. The subject of the in limine order — whether Denholm could collect damages for lost royalties and loss of reputation — involved the same underlying breach of contract cause of action that the jury ultimately was called upon to decide, but it obviously represented an entirely different measure of damages from the reasonable-value-of-time-spent measure submitted to the jury at trial. Consequently, Judge Hauk’s remittitur order represented a decision that the jury’s verdict did not conform to the evidence introduced at trial regarding the value of Denholm’s time; it did not decide anew whether lost royalties and/or loss of reputation damages were awarda-ble. 5

In short, the remittitur order concerned a breach of contract trial whose parameters were strictly limited on the question of damages to the reasonable value of plaintiff’s time spent on the project. As such, the trial necessarily decided only two things: 1) defendants had breached their contract with Denholm; and 2) the value of Denholm’s time spent on the project was $100,000. The court’s remittitur order, by reducing the $100,000 to $20,000, itself necessarily also addressed only the value of Denholm’s time. The order had nothing to do with lost royalty or loss of reputation damages. 6 By concluding otherwise simply because the remittitur order was broadly worded, the majority inexplicably ignores the circumstances that led to the issuance of the order.

The majority posits a bright-line rule which is easily applied. Unfortunately, this is about, all that the rule has to recommend itself. For it leaves no way for Den-holm to appeal the district court’s grant of the in limine motion in a reasonable amount of time. In the typical remittitur situation, the plaintiff has the option of having a second trial, where he may presumably offer better witnesses and more convincing evidence and hence win a favorable jury verdict that the court will not disturb. At the very least, the possibility looms that while the second jury might award an amount identical to that awarded by the first jury based upon similar evidence, the court might at that point take pause and allow the award to stand. But in the instant case, the judge made a pretrial decision to exclude certain types of evidence. It is obvious that he would exclude the identical evidence at a retrial. Thus, under the “separate and distinct cause of action” rule, the only way for an *366 appellant like Denholm ever to obtain appellate review of the initial in limine order would be to suffer through a second trial that he knows will not help him obtain recovery for the only real damages he has ever alleged, and then appeal. 7 This strikes me as a waste of time, money, judicial resources, and a disincentive to acceptance of remittitur. Cf. Aaro, Inc. v. Daewoo Int’l (Am.) Corp., 755 F.2d 1398 , 1401 n. 6 (11th Cir.1985). Denholm does not argue here that the reasonable value of the time that he devoted to the mathematics project should properly exceed $20,000. Instead, he argues the separate issue that he was unable to present evidence to the jury concerning lost royalties and reputation damage.

I would adopt the view that plaintiffs may appeal from “an order completely unrelated to the remittitur order,” id. at 1401 , in a functional sense; see 6A J. Moore & J. Lucas, Moore’s Federal Practice, ¶ 59.08[7], at 59-204 to -205 (2d ed. 1989) (although “a plaintiff in federal court may not appeal from a remittitur order he has accepted,” he may “appeal from other parts of the judgment.”) (emphasis added). Consequently, I would allow Denholm to appeal the ruling on the in limine motion.

. Indeed, a narrow reading of Call Carl, Inc. v. BP Oil Corp., 554 F.2d 623, 627 (4th Cir.), cert. denied, 434 U.S. 923 , 98 S.Ct. 400 , 54 L.Ed.2d 280 (1977), reveals that the Fourth Circuit, presented squarely with an attempted appeal concerning an entirely separate cause of action (antitrust) from the cause of action which was the subject of the remittitur order (fraud and deceit), simply held that such an appeal was permissible. The court did not state that only separate causes of action would suffice.

. The majority relies on the fact that Denholm’s complaint sought generic breach of contract compensatory damages. Majority Opinion at 360. But such reliance amounts to pointless formalism and in the end proves nothing. For even if Denholm could only be properly said to have sought one type of damages, a review of the record makes it abundantly clear that the district court treated various categories of evidence on this "single” damage claim differently. This is all that should matter under the functional analysis I propose. See infra part II.

. Admittedly, section 16 of the Clayton Act, which covers injunctive relief, only requires impending, as opposed to actual, antitrust injury; yet the Sealy court, see 585 F.2d at 844 , inferred from the jury’s damages verdict that an actual antitrust violation had already occurred.

. Significantly, appellee's counsel never referred to Lanier as establishing an absolute rule requiring "separate and distinct causefs] of action."

. The judge admonished the jury that it not award such damages, and indeed no evidence was admitted during trial regarding lost royalties and loss of reputation. These facts affirmatively rebut any assumption that the jury considered improper measures of damages. We can thus only infer that the jury miscalculated the value of the time Denholm spent developing the mathematics program. Even if the jury disobeyed the judge and implicitly awarded damages for lost royalties and/or loss of reputation in its verdict, the remittitur order did not decide that such damages could not be awarded; instead, the order at most reflected the earlier, pretrial decision that such damages could not be awarded.

. The situation in this case is to be contrasted with a situation in which a plaintiff is allowed to introduce evidence at trial to prove lost royalty or loss of reputation damages yet thereafter accepts remittitur. In such a situation, the re-mittitur' order obviously covers the entire cause of action. Here, by contrast, the in limine order removed certain damages from consideration before the trial started.

. Actually, the plaintiff may first seek review of the district court’s order granting a new trial. See Oltz v. St. Peter’s Community Hosp., 861 F.2d 1440, 1451 (9th Cir.1988). Yet in determining whether the district court abused its discretion by ordering a new trial, we would surely be unable to review the propriety of the initial grant of the in limine motion, unless the district court granted a new trial expressly on the basis that it erred in granting the in limine motion in the first place. See id. at 1451-52 . For we would review the grant of a motion for a new trial for " ‘abuse of discretion as to each ground upon which the court based its determination.'" Id. at 1452 (quoting Peacock v. Board of Regents, Etc., 597 F.2d 163, 165 (9th Cir.1979)) (emphasis added). And in the case now before us it is certain that the district court threatened to grant the motion for a new trial solely because the jury's verdict on damages for the value of Denholm’s time did not comport with the evidence presented at trial.