DENIAL OF PETITION FOR PANEL REHEARING:

The panel issued its opinion affirming the district court on January 12, 2018 (the Opinion). People for the Ethical Treatment of Animals v. Miami Seaquarium , 879 F.3d 1142 (11th Cir. 2018). A petition for panel rehearing and rehearing en banc was filed on February 2, 2018. Having considered the arguments raised in the petition for panel rehearing, the panel adheres to its Opinion. Although the petition does not cause us to change our conclusion, the panel believes a few of Appellants' arguments warrant a response.

Appellants object to the Opinion's summary discussion of Lolita's injuries. As an initial matter, Lolita presents a unique case because she: (1) is of advanced age at 51, having surpassed the median life expectancy for wild, female Southern Resident Killer Whales; 1 (2) has received medical care for approximately 48 years 2 and continues to receive medical care; (3) has already been subject to an unsuccessful federal challenge to the conditions of her captivity; and (4) has no realistic means for returning to the wild without being harmed. Id. at 1145-46 . At her advanced *1309 age, Lolita cannot be expected to be free of health problems.

The outcome here avoids tying the hands of future courts in cases involving younger, healthier animals who may be faced with different circumstances. It also avoids signaling unintentionally that an animal's age and the level of medical care it receives are determinative. Put simply, based on the undisputed evidence and the unique circumstances of this case, the panel determined that no reasonable fact finder could conclude that Lolita's injuries present a "threat of serious harm" sufficient to trigger liability under the ESA. The Opinion further reflects the panel's determination that the law would be better served by announcing the "threat of serious harm" rule, without defining its contours, and allowing district courts the flexibility to apply that rule to future circumstances with which they are presented.

In any event, some injuries noted by Appellants would, given the record evidence, plainly fail to meet the "threat of serious harm" standard. For example, Appellants cite the "rakes" inflicted when the Pacific white-sided dolphins with whom Lolita shares her tank scrape her with their teeth. But cetaceans, including orcas and dolphins, undisputedly rake each other in the wild. And Appellants' own expert rated Lolita's rakes as a three to four on a scale from one to ten, with ten being the most raked orca observed in the wild and one being the least raked orca observed in the wild. So, not only is Lolita suffering fewer rakes than the average wild orca, she is receiving care to make sure her rakes heal. Given that evidence, Lolita's rakes are not serious enough to illustrate that the conditions of her captivity amount to "harm" or "harass[ment]" in violation of the ESA.

The Opinion also faithfully applied the Supreme Court's decision in Babbitt v. Sweet Home Chapter of Communities for a Great Oregon , 515 U.S. 687 , 698, 704-06, 115 S.Ct. 2407 , 132 L.Ed.2d 597 (1995). Appellants read Babbitt as holding that noscitur a sociis should not be used to interpret the ESA's prohibition against the "tak[ing]" of any endangered or threatened species. See 16 U.S.C. § 1538 (a)(1)(B). We disagree.

In Babbitt , the Supreme Court was urged to affirm a court of appeals decision that used noscitur a sociis to invalidate a regulation that interpreted "harm" as covering actions that "indirectly" impact endangered or threatened animals ( e.g. , habitat destruction). 515 U.S. at 697-99 , 115 S.Ct. 2407 . The court of appeals had applied noscitur a sociis to support its conclusion that "harm" only includes actions that "directly" impact endangered or threatened animals, reasoning "that 'harm' must refer to a direct application of force because the words around it do." Id. at 701 , 115 S.Ct. 2407 . The Supreme Court concluded that the court of appeals incorrectly applied noscitur a sociis to give the term "harm," which could have been read as covering indirect action, "essentially the same function as the other words in the definition [of 'take']," which the court of appeals understood as covering only direct action. Id. at 702 , 115 S.Ct. 2407 . Babbitt therefore stands for the proposition that noscitur a sociis should not be used to interpret the ESA's prohibition against "tak[ing]" in a manner that would deprive a term in the definition of "take" of standalone meaning. In other words, the Supreme Court in Babbitt did not indicate that noscitur a sociis cannot be applied in cases involving the ESA's prohibition against "tak[ing]" or in the analysis of the ESA's definition of "tak[ing]." The Supreme Court merely held that the manner in which the court of appeals had applied noscitur a sociis in that case was incorrect.

*1310 As the Opinion explains, the application of noscitur a sociis in this case does not deprive "harm" of independent meaning. "Harm" brings acts that "injure" by "caus[ing] hurt or damage" within the ESA's ambit. The term "harass" brings a different category of action within the ESA's purview-namely, "annoying, vexatious, and troubling conduct." The Opinion applied the canon to determine what degree of "harm" or "harass[ment]" was actionable by looking to the eight surrounding terms in the definition of "take." The ultimate holding that either sort of conduct must "pose a threat of serious harm" does not collapse "harm" and "harass[ment]" into one another. Instead, it clarifies that acts constituting either "harm" or "harass[ment]" are not actionable unless they meet a threshold level of severity-any actionable "harm" must be serious and any actionable "harass[ment]" must pose the threat of serious harm.

The Opinion also aligns with Congress' intent in drafting the ESA: preventing extinction. Appellants contend that the Opinion ignores the intentionally broad language with which the ESA was drafted. On the contrary, the panel acknowledges that the ESA was drafted broadly. The point made in the Opinion is that the ESA's language is not so broad as to cover acts totally unrelated to the ESA's purpose of preventing extinction. Accounting only for the dictionary definitions of "harm" and "harass" would bring de minimis annoyances to endangered animals that bear no reasonable relationship to extinction within the ESA's coverage-a result inconsistent with the ESA's purpose.

"Harass," for example, means "to vex, trouble, or annoy continually or chronically." Webster's Third New International Dictionary 1031 (1986). Taken to its outer limits, this definition could be construed as covering Lolita's regular veterinary care which, despite being chronically annoying from her perspective, is life-prolonging. Surely Congress did not intend the ESA to illegalize veterinary care.

Alternatively, Appellants assert that our reading of the ESA conflicts with regulatory definitions. For example, 50 C.F.R. § 222.102 , which is entitled to deference, provides that "harm" "means an act which actually kills or injures fish or wildlife. Such an act may include significant habitat modification or degradation where it actually kills or injures fish or wildlife by significantly impairing essential behavioral patterns, including breeding, spawning, rearing, migrating, feeding, or sheltering." Appellants contend "these definitions provide the threshold that conduct must meet to violate the ESA." Petition at 7 (emphasis added). But the Opinion does not interpret the level of conduct required to fall within the ESA; it interprets the level of severity that "harm" or "harass[ment]" must meet to fall within the ESA.

Per the regulation, harm "may include significant habitat modification or degradation which actually kills or injures fish or wildlife by significantly impairing essential behavioral patterns." 50 C.F.R. § 222.102 (emphasis added). The regulation uses "significant" to describe the sort of activities that cause death or injury and "significantly" to describe the manner in which that death or injury occurs. The terms are not used to modify the phrase "to injure." Put differently, the regulation does not suggest injury must be "significant" to be actionable, it suggests only that "significant" habitat modification or degradation may constitute harm where it causes death or injury by "significantly" impairing essential behavioral patterns. The regulation says nothing about the level of severity to which the "injury" must rise to constitute actionable "harm." Therefore, the regulation does not answer the question posed by this case and is irrelevant to our analysis.

*1311 Examining the two key clauses in the regulation illustrates this point. First, we turn to the requirement that habitat modification or degradation be "significant." Even "significant" habitat modification would not be actionable if it did not injure or kill the animal in question by significantly impairing its behavioral patterns. Take, for example, an animal whose natural habitat is destroyed when the forest where it used to live is destroyed by logging. The habitat modification in that hypothetical is undeniably "significant." But imagine the animal, thanks to its adaptiveness, is neither injured nor killed-instead, it seamlessly relocates and proceeds with its life. That habitat modification, however "significant," would not fall within the ambit of the regulation because neither injury nor death resulted. Thus, the statement that habitat degradation must be "significant" does not answer the question of what degree of "harm" is actionable.

The back side of the regulation, which requires that the habitat modification or degradation "significantly" impair essential behavioral patterns, is similarly unhelpful. It clarifies that the regulation covers situations where an animal is injured or killed because its essential behavioral patterns are "significantly" impaired. The language of the regulation makes clear that it does not cover situations in which "significant" impairment of essential behavioral patterns occurs, but neither injury nor death results. Imagine, for example, an animal whose essential behavioral pattern is "significantly" impaired when it is forced to find a new breeding ground, but it succeeds in adapting without suffering any injury. That behavioral impairment, however "significant," would not be covered because no injury or death resulted. Further, the regulation cannot be reasonably read as indicating that "significantly impairing essential behavioral patterns" is itself "harm." If that were the case, there would have been no reason to specify that the impairment must actually kill or injure the animal in question. Thus, the statement that the modification of behavioral patterns must be "significant" does not answer the question of what degree of "harm" is actionable.

Finally, the standalone statement that the animal must be actually killed or injured is unhelpful. Harm is defined, in part, as "to injure." Therefore, the regulation does not answer the question of what degree of "harm" is actionable. Instead, it illustrates that "harm" need not be direct. "Harm" can, in some instances, be inflicted indirectly through habitat modification or degradation that meets the threshold conditions set forth in the regulation.

The petition for panel rehearing is therefore DENIED .