Appellants Building Industry Association, et al., sought review of the Fish and Wildlife Service's listing of various fairy shrimp species as endangered or threatened. They now challenge a district court decision denying their motion for summary judgment. Determining that we have jurisdiction, we affirm the denial.

In those regions of California with Mediterranean climates, one finds shallow depressions called "vernal pools" that fill with rainwater in fall and winter only to evaporate in spring. In these pools reside numerous indigenous aquatic invertebrates that have evolved to survive in the pools' variable environmental conditions. In 1992 the Fish and Wildlife Service proposed to list as endangered species five tiny crustaceans resident in California's vernal pools: the vernal pool fairy shrimp, Conservancy fairy shrimp, longhorn fairy shrimp, California linderiella, and vernal pool tadpole shrimp (collectively, "fairy shrimp"). The proposed rule specified actual and threatened destruction of vernal pools as a justification for the listing.1

After a comment period, the Service withdrew the proposal to list the California linderiella. It listed vernal pool fairy shrimp as threatened and the three remaining species as endangered. Though the Endangered Species Act (ESA) requires the Service to designate "critical habitat[s]" for listed species "to the maximum extent prudent and determinable," the Service declined to make designations on the ground that so doing would put the listed species at risk of vandalism.2

The Service's decision then began its long and bumpy journey to appellate review. Appellants challenged the listing decision in the district court, asserting violations of the ESA, the Administrative Procedure Act (APA), and the Constitution. Along with contesting on various grounds the general validity of the listing decision, appellants took issue with the Service's failure to designate critical habitats. In response to cross-motions for summary judgment, the district court in July 1997 granted summary judgment to the Service on all claims except the critical habitat claim. On that claim, the court held that the failure to designate critical habitat was arbitrary and capricious and remanded to the Service either for designation or for explanation why designation was not prudent.3

While the critical habitat remand was pending, the district court certified the listing claims under Rule 54(b). An appeal of that portion of the decision followed, but we dismissed for lack of jurisdiction. The listing claims and the critical habitat claim arose out of the same body of law and fact, meaning that the listing appeal would be intertwined with a possible critical habitat appeal. With the threat of such a piecemeal appeal looming, and without an explanation from the district court as to why such an approach was desirable, we held that we lacked jurisdiction.4

In March 1999, the district court reviewed additional record citations the Service had provided in support of its conclusion that critical habitat designations would be imprudent. It ruled that these materials did not adequately support the Service's conclusion, vacated the Service's decision not to designate critical habitats, and remanded once again. According to the district court, the decision constituted a final judgment on all claims.

Appellants brought a second appeal of the listing decision. We ordered the parties to address whether a final decision existed because the critical habitat remand was still before the Service, which had not itself appealed. In an attempt to resolve any possible jurisdictional infirmity, appellants decided to abandon litigation of the critical habitat claim, leaving for resolution only the listing claims, which were clearly final. To that end, appellants moved the district court to amend its March 1999 judgment remanding to the Service or in the alternative for leave to amend their complaint to delete the critical habitat claim. The district court determined that due to the pending appeal it had no jurisdiction to consider the motion; it also indicated that if it were to regain jurisdiction, it would deny the motion to amend the judgment but grant leave to amend the complaint.5 Soon thereafter appellants moved this court to dismiss their appeal, which we did. The district court then granted appellants' motion to strike from their complaint the critical habitat claim, the only claim on which they had prevailed. Appellants immediately brought this appeal, their third attempt to gain review of the district court's dismissal of the listing claims. We once again ordered the parties to address our jurisdiction.

Appellants allege numerous errors in the district court's decision. They argue that the rule's heavy reliance on a study, the so-called "Simovich study," not made available during the comment period violated the APA, as did the rule's enumeration of fairy shrimp populations in terms of vernal pool complexes rather than individual pools. They further contend that the listing was not supported by the best available scientific data, as required by the ESA, and that the Service misapplied its own policy on independent peer review. The nonfederal appellees, public interest groups that intervened below ("intervenors"), argue that we lack jurisdiction because no final judgment exists. We address that argument before reaching the merits.

Intervenors' jurisdictional argument is subtle. Absent appeal by the agency, an order remanding to an agency for further proceedings is not an appealable final decision even where the district court dismisses the case. See NAACP v. United States Sugar Corp., 84 F.3d 1432, 1436 (D.C. Cir. 1996). Before amendment of the complaint, therefore, the critical habitat remand pending before the Service prevented our assertion of appellate jurisdiction over the listing claims. Accordingly, appellants amended their complaint to omit the critical habitat claim. Though the 1997 decision was not final at the time it was entered, under the doctrine of cumulative finality the dismissal of the only claim that survived that decision retroactively rendered it final and appealable. See 15A Charles Alan Wright, Arthur R. Miller, & Edward H. Cooper, Federal Practice & Procedure 3914.9, at 631-42 (2d ed. 1992); cf. Sacks v. Rothberg, 845 F.2d 1098, 1099 (D.C. Cir. 1988) (per curiam).

Intervenors argue, however, that the district court abused its discretion in allowing amendment of the complaint after judgment. If that were so, the March 1999 decision remanding to the Service would still be in force, and there would be no final judgment for this court to review. Ordinarily postjudgment amendment of a complaint under Rule 15(a) requires reopening of the judgment pursuant to Rule 59(e) or 60(b). See Cassell v. Michaux, 240 F.2d 406, 407-08 (D.C. Cir. 1956). This prevents litigants from resurrecting claims on which they have lost. Cf. Firestone v. Firestone, 76 F.3d 1205, 1207-08 (D.C. Cir. 1996) (per curiam). But that concern is absent here: appellants prevailed on the claim in question, and dropped it only so that they might appeal dismissal of other claims.

Appellants respond (and the Service agrees) that because the remand was not a final judgment, no motion under Rule 59 or 60 was necessary to amend the complaint to abandon claims they no longer wished to pursue. We agree. The general requirement of a Rule 59 or 60 motion prior to postjudgment amendment is employed to serve the judicial policy "favoring finality of judgments and the expeditious termination of litigation." See 6 Charles Alan Wright, Arthur R. Miller, & Mary Kay Kane, Federal Practice & Procedure 1489, at 694 (2d ed. 1990). Neither goal is served by requiring a Rule 59 or 60 motion here. Indeed, it would be passing strange if in order to secure appeal of the claims on which they lost appellants were forced to litigate to finality claims on which they preliminarily prevailed and that they now wish to abandon.6

As noted, the rule relies heavily on the Simovich study, which was released after the proposal and which the agency received only during the comment period. The study was therefore not among the materials published for public comment. Appellants argue that the Service's failure to seek comment on the study violated the APA.

It is not disputed that the Service placed great weight on the Simovich study. It is cited frequently in the rule, which touted it as "[s]cientifically credible." 59 Fed. Reg. at 48,141. The Service concedes that the study is "the first long-term multidisciplinary study" and "the most scientifically based and well-documented professional study" of California vernal pools ever attempted, that it is "more comprehensive than any previous study," and that "the final rule relied substantially on the findings in the Simovich study."

The Service nonetheless contends that it was not required to publish the Simovich study for public comment, and we agree. The APA generally obliges an agency to publish for comment the technical studies and data on which it relies. See Solite Corp. v. EPA, 952 F.2d 473, 484 (D.C. Cir. 1991) (per curiam). But to avoid "perpetual cycles of new notice and comment periods," Ass'n of Battery Recyclers v. EPA, 208 F.3d 1047, 1058 (D.C. Cir. 2000), a final rule that is a logical outgrowth of the proposal does not require an additional round of notice and comment even if the final rule relies on data submitted during the comment period. See Int'l Fabricare Inst. v. EPA, 972 F.2d 384, 399 (D.C. Cir. 1992) (per curiam); Solite, 952 F.2d at 484. Such is the case here. The Simovich study, while the best available, only confirmed the findings delineated in the proposal. In relying on it, the Service "did no more than provide support for the same decision it had proposed to take." Int'l Fabricare, 972 F.2d at 399. Essentially, the proposal advanced for comment a hypothesis and some supporting data. The Simovich study provided additional support for that hypothesis--indeed, better support than was previously available--but it did not reject or modify the hypothesis such that additional comment was necessary. See Solite, 952 F.2d at 484.

Appellants next object to the rule because it counts fairy shrimp populations by the number of vernal pool complexes, not the number of individual vernal pools, in which they reside. See 59 Fed. Reg. at 48,137. (A pool complex is a group of individual pools that, due to their proximity, are susceptible to the same threats.) Appellants insist that the proposal never put the public on notice of the "complexes methodology" or of the definition of complexes.7 But the use of complexes to measure fairy shrimp populations was no surprise: the proposal itself used the term five times. See 57 Fed. Reg. at 19,856, 19,858, 19,859. Moreover, nothing in the final rule's use of complexes constituted a deviation from the proposed rule. The proposal posited danger to existing fairy shrimp populations, which were discussed in terms of both individual pools and pool complexes; consistent with the proposal, the final rule found danger to existing fairy shrimp populations, which were measured--most accurately, according to the rule--in terms of pool complexes. The final rule's measurement of populations solely in terms of complexes, after the proposal's uncommitted use of both methodologies, was a tightening of the rule's reasoning, but it was nonetheless a logical outgrowth of the proposal. Appellants have not pointed to any way in which the sharpened focus on complexes changed the rule's reasoning or conclusion.

Appellants also claim that methodological flaws in the Simovich study and other relied-upon authorities mean that the rule was not based on the "best scientific and commercial data available," as required by 16 U.S.C. 1533(b)(1)(A). Yet as the district court noted, appellants "have pointed to no data that was omitted from consideration." 979 F. Supp. at 903. Assuming that studies the Service relied on were imperfect, that alone is insufficient to undermine those authorities' status as the "best scientific ... data available." Appellants misread 1533(b)(1)(A): the Service must utilize the "best scientific ... data available," not the best scientific data possible. The Service may not base its listings on speculation or surmise or disregard superior data, cf. Bennett v. Spear, 520 U.S. 154, 176 (1997); City of Las Vegas v. Lujan, 891 F.2d 927, 933 (D.C. Cir. 1989), but absent superior data--and appellants point to none--occasional imperfections do not violate 1533(b)(1)(A).

Finally, we reject appellants' claim that the listing's validity is undermined by its failure to comply with the Service's peer review policy. To be sure, the listing was not subjected to the present peer review procedure, which requires that "during the comment period" the Service obtain three independent specialists' opinions on the merits of the decision and reprint them in the listing. See Peer Review Policy Statement, 59 Fed. Reg. 34,270 (July 1, 1994). But the current peer review policy came into force 22 months after the close of the fairy shrimp comment period. Appellants point out that a March 1995 letter from a Service official to Congressman Richard Pombo stated that "[i]n conformance with policy (59 FR 34270), the Service sought scientific peer review of the listing proposal." Letter from George T. Frampton, Jr., Assistant Secretary for the Service, to Hon. Richard Pombo 1 (March 10, 1995). That statement may have misled the congressman as to the Service's compliance with the specific peer review procedures promulgated in 1994, but the listing was in fact subject to peer review that was intense though less formal than is currently required. In any event, appellants suggest no basis on which the letter would render the later-enacted policy statement retroactively binding on an already-concluded comment period.8

* * * *

The denial of appellants' motion for summary judgment is

Affirmed.