The red fox (Vulpes vulpes) is a mesocarnivore species that exploits opportunistically a wide range of prey items that are consumed as a function of their abundance and availability (Díaz-Ruiz et al., 2013). Fish are an unusual prey group in the diet of red foxes that is occasionally reported in dietary studies (e.g., Basuony et al., 2005; Wagnon & Serfass, 2017), but we do not know if the red fox obtains fish from scavenging or active hunting. Here, we report what may be the first known case of several fish hunted by a red fox. We raise different reasons to explain the observed novel behavior and discuss its ecological implications.

While doing field work in the surroundings of the Valuengo reservoir in southern Extremadura (Spain; 38.294845° N, −6.674353° W), we observed an adult male red fox catching European carps (Cyprinus carpio) on the shore (Figure 1 and Videos S1–S3). The observation took place on 24 March 2016, between 1:18 and 2:51 PM. During this time period, until it noticed our presence and left the area, the individual hunted 10 medium-sized carps (~1 kg) out of 12 observed attempts, a capture success rate of 83%. The male fox approached the water's edge, where the carps spawned their eggs and while they were distracted by the frenzy of their reproduction (Figure 1 and Video S5), and jumped into the water to catch the fish (Figure 1 and Videos S1–S3). After each capture, the fox moved away ~20–30 m from the shore to leave, hide, or bury the captured prey that could be understood as a potential caching behavior, presumably for later consumption (Video S3). On one occasion, the fox took one fish and left the area and moved to a scrubby area further away. The observation distance was ~100 m with binoculars and recorded by a reflex camera with a telephoto lens. This allowed us to observe how the fox never consumed any substantial part of the captured prey. However, on several occasions we observed that the fox seemed to eat small parts of the prey (e.g., Videos S2 and S3). These parts might be some eggs from the pregnant female carps, but the lack of direct vision prevented us to unequivocally confirm this activity. After 52 min, a female fox (easily identifiable by docked tail) appeared from the same area, took a large carp that the male had hunted and carried it toward the scrubland (Video S4) without any interference from the male. The purpose of these captures may have been to supply prey for the female and the pups, because the observations occurred during the known breeding season of red foxes in Iberian Mediterranean habitats (Zapata et al., 1997). However, we did not observe any pups, so this assumption remains unconfirmed.

Our observation has several implications for canid behavioral ecology. First, this observation shows the ability of this species to catch fish in their environment and confirms fish as a food item that can be consumed as a fresh capture and not only opportunistically as carrion (Basuony et al., 2005). This behavior has been described previously for gray wolves (Canis lupus) in the USA (e.g., Darimont et al., 2003; Gable et al., 2018), but not documented for red foxes. But the most novel aspect of the observation was the fact that large numbers of fish were caught in such a short period of time and with a high effectiveness, documenting (photographically and by video) for the first time this hunting behavior. This behavior may not be an isolated behavior associated with a single individual, and probably other individuals can do this as well. This could imply that this opportunistic behavior is an intrinsic characteristic of foxes as consequence of ephemeral high-prey availability, similar to that described for gray wolves and salmon in British Columbia and Alaska (Darimont et al., 2003; Stanek et al., 2017). Notwithstanding, perhaps not all red fox individuals are able to develop this hunting behavior so efficiently (capture success rate >80%) suggesting that this could be a learned behavior based on previous experience (Darimont et al., 2003; Stanek et al., 2017).

At first glance, our observation might fit with a case of “surplus killing”, because the red fox invested large efforts in killing a large number of fish that it apparently did not immediately consume (Kruuk, 1972). The observation took place during carp spawning, which in reservoirs occurs in large groups on shallow banks (Doadrio, 2002), making them more vulnerable to predation. The fact that it occurred at the moment of maximum availability and vulnerability of the prey concurs with previous hypotheses about the triggers of surplus killing (please refer to Kruuk, 1972; Lincoln & Quinn, 2019; Short et al., 2002; Wiesel, 2010). However, in some cases apparent surplus killing may be consistent with the optimal foraging theory, and may be considered as an adaptive behavior. For instance, this occurs when one or more individuals kill a large number of prey that they do not consume but are exploited by conspecifics of the same social unit (Kruuk, 1972) or when there is a selective consumption of optimal carcass discarding low quality ones (Darimont et al., 2003; Lincoln & Quinn, 2019). The fact that the female, most likely breeding (Zapata et al., 1997), and most likely to be a partner of the fox that hunted the fish, took part of the catch without the male preventing it (Video S4), indicates that the male was capturing and caching prey to feed the family group (i.e., female and pups) that at that time was highly dependent on the male (Macdonald, 1979). Therefore, it could be understood that there is a use for and optimization of the resource, at least partially, so that a behavior could be occurring with the aim of obtaining a large amount of necessary resources with little effort, which would fall within the optimal foraging theory (Stephens & Krebs, 1986). A similar behavior has been described for arctic foxes (Vulpes lagopus) that cache large amounts of bird eggs during the reproduction season to feed their pups (Clermont et al., 2021; Giroux et al., 2012). Additionally, the male fox seemed to feed on some eggs from the female carps (e.g., Videos S2 and S3), which would be consistent with a selective consumption of the carcasses and discarding lower quality tissues, behavior previously described for brown bears (Ursus arctos) and gray wolves under high availability of salmon in British Columbia and Alaska (Darimont et al., 2003; Lincoln & Quinn, 2019). Accordingly, we hypothesize that the observed behavior follows the optimal foraging theory, giving less support to surplus killing.

It is of key importance that these types of field observations be reported and published because these behavioral phenomena are very difficult to observe in the wild. Furthermore, the uncommon nature of these behaviors makes it difficult to test hypotheses in nature using designed experiments. Therefore, the accumulation of this kind of evidence will contribute to the development of more robust studies that will lead to a better understanding of these behaviors and their ecological implications (Lincoln & Quinn, 2019).

红狐 ( Vulpes vulpes ) 是一种中等食肉动物物种，它会机会性地捕食范围广泛的猎物，这些猎物会根据其丰度和可用性而被消耗掉（Díaz-Ruiz 等人，2013 年 ）。鱼是红狐饮食中不寻常的猎物群，偶尔会在饮食研究中报道（例如，Basuony 等人，  2005 年；Wagnon 和 Serfass，  2017 年），但我们不知道红狐是否从食腐动物或主动狩猎。在这里，我们报告了可能是第一例已知的几条鱼被赤狐猎杀的案例。我们提出不同的理由来解释观察到的新行为并讨论其生态意义。

在埃斯特雷马杜拉南部（西班牙；38.294845° N，-6.674353° W）的 Valuengo 水库周围进行实地考察时，我们观察到一只成年雄性红​​狐捕捉欧洲鲤鱼（Cyprinus carpio) 在岸上 (图 1 和视频 S1–S3)。观察发生在 2016 年 3 月 24 日下午 1:18 到 2:51 之间。在此期间，直到它注意到我们的存在并离开该区域，在观察到的 12 次尝试中，该个体捕获了 10 条中型鲤鱼（约 1 公斤），捕获成功率为 83%。雄性狐狸接近水边，鲤鱼在那里产卵，当它们因繁殖的狂热而分心时（图 1 和视频 S5），然后跳入水中捕捉鱼（图 1 和视频 S1-S3 ). 每次捕获后，狐狸都会离开海岸约 20-30 米，以离开、隐藏或掩埋捕获的猎物，这可以理解为潜在的缓存行为，大概是为了以后食用（视频 S3）。有一次，狐狸带了一条鱼离开了这个区域，搬到了更远的灌木丛中。使用双筒望远镜观察距离约为 100 米，并使用带有长焦镜头的反光相机记录。这使我们能够观察到狐狸是如何从不吃掉捕获的猎物的任何实质部分的。然而，有几次我们观察到狐狸似乎只吃猎物的一小部分（例如，视频 S2 和 S3）。这些部分可能是怀孕的雌性鲤鱼的一些卵，但由于缺乏直视，我们无法明确确认这一活动。52 分钟后，同一区域出现了一只雌性狐狸（可以通过断尾很容易辨认），在没有任何雄性干扰的情况下，带着雄性猎杀的一条大鲤鱼并将其带到灌木丛中（视频 S4）。这些捕获的目的可能是为雌性和幼崽提供猎物， 1997 年）。然而，我们没有观察到任何幼崽，所以这个假设还没有得到证实。

我们的观察对犬类行为生态学有几个启示。首先，这一观察显示了该物种在其环境中捕鱼的能力，并证实鱼是一种可以作为新鲜捕获物食用的食物，而不仅仅是机会性地作为腐肉食用（Basuony 等人，2005 年 ）。这种行为之前已在美国的 灰狼 ( Canis lupus ) 中描述过（例如，Darimont 等人， 2003 年；Gable 等人，  2018 年）), 但没有记录红狐狸。但观察中最新颖的方面是在如此短的时间内捕获了大量的鱼，而且效率很高，这是第一次记录（通过照片和视频）这种捕猎行为。这种行为可能不是与单个人相关的孤立行为，其他人也可能会这样做。这可能意味着这种机会主义行为是狐狸的内在特征，这是短暂的高猎物可用性的结果，类似于不列颠哥伦比亚省和阿拉斯加的灰狼和鲑鱼的描述（Darimont 等人，2003 年；Stanek 等人，  2017年 ）). 尽管如此，也许并非所有红狐个体都能如此有效地发展这种狩猎行为（捕获成功率 >80%），这表明这可能是一种基于以往经验的习得行为（Darimont 等人，2003 年；Stanek 等 人，  2017 年）。

乍一看，我们的观察可能符合“剩余杀戮”的情况，因为红狐投入大量精力杀死了它显然没有立即吃掉的大量鱼 (Kruuk, 1972  )。观察发生在鲤鱼产卵期间，在水库中，鲤鱼在浅岸上成群结队地出现 (Doadrio, 2002 )，使它们更容易受到捕食。事实上，它发生在猎物的最大可用性和脆弱性时刻，这与之前关于过量杀戮触发因素的假设一致（请参阅 Kruuk，1972 年；Lincoln & Quinn，2019 年；Short 等人，2002 年；  Wiesel，  2010年）). 然而，在某些情况下，明显的剩余杀戮可能与最佳觅食理论一致，并且可以被视为一种适应行为。例如，当一个或多个个体杀死大量猎物时会发生这种情况，这些猎物他们不食用但被同一社会单位的同种动物剥削 (Kruuk, 1972)，或者当有选择性地食用最佳尸体丢弃低质量 尸体时（Darimont 等人，  2003 年；Lincoln 和 Quinn，2019 年）。事实上，雌性最有可能繁殖（Zapata 等人，  1997), 并且很可能是捕鱼的狐狸的伙伴, 在没有雄性阻止的情况下捕获了一部分渔获物 (视频 S4), 表明雄性捕获并缓存猎物以喂养家庭群体 (即雌性)和幼崽)，当时高度依赖雄性 (Macdonald,  1979 )。因此，可以理解的是，至少部分地使用和优化资源，以便可以发生一种行为，目的是通过很少的努力获得大量必要的资源，这将落在最佳范围内觅食理论 (Stephens & Krebs,  1986 )。北极狐（ Vulpes lagopus）也有类似的行为) 在繁殖季节储存大量鸟蛋以喂养幼崽（Clermont 等人，  2021 年；Giroux 等人，  2012 年）。此外，雄性狐狸似乎以雌性鲤鱼的一些卵为食（例如，视频 S2 和 S3），这与选择性食用尸体和丢弃低质量组织的行为一致，之前描述过棕熊的行为（Ursus arctos ) 和灰狼在不列颠哥伦比亚省和阿拉斯加的鲑鱼大量供应下（Darimont 等人，2003 年；Lincoln 和 Quinn，  2019 年）。因此，我们假设观察到的行为遵循最佳觅食理论，对剩余杀戮的支持较少。

报告和发表这些类型的实地观察非常重要，因为这些行为现象在野外很难观察到。此外，这些行为的不常见性使得很难使用设计的实验来检验自然界中的假设。因此，此类证据的积累将有助于开展更有力的研究，从而更好地了解这些行为及其生态影响（Lincoln & Quinn，2019 年 ）。