Discussion

This study provides descriptions of two new predatory amphipods of the genus Princaxelia from hadal depths. Princaxelia malohi sp. nov. collected from 8,200–8,350 m water depth in the Tonga Trench, and Princaxelia kahat sp. nov. collected from between 8,000–8,964 m in the Mariana Trench. These descriptions increase the number of formally described species in the genus to seven and provide both morphological diagnoses and mitochondrial DNA barcodes for 16S and COI, where amplification was successful. The addition of genetic data is particularly important for Princaxelia, for which sequence data remains limited. At present, only P. marianaensis has been described with an associated COI sequence, although additional unidentified Princaxelia sequences are available from broader amphipod genetic studies (Jażdżewska and Mamos 2019; Jażdżewska et al. 2026). The new sequences generated here therefore provide an important reference point for future work on the taxonomy, diversity, and biogeography of Princaxelia and other deep-sea pardaliscid amphipods.

Morphological taxonomy

The taxonomy of Princaxelia remains difficult to resolve because several described species are based on incomplete or poorly preserved material. No complete type specimens are available for P. stephenseni, P. abyssalis, or P. magna. For P. stephenseni, only damaged mouthparts remain, while P. abyssalis is represented only by dissected appendages. The repository of the type specimen of P. magna was not stated in the original description, and previous attempts to locate it were unsuccessful (Lörz 2010). As a result, the descriptions of these species remain incomplete, despite the additional notes and revisions provided by Lörz (2010). This uncertainty is particularly relevant for Princaxelia kahat sp. nov., which closely resembles the published description and illustrations of P. magna in several morphological features. However, some illustrations of P. magna are unclear and key diagnostic information is missing, preventing confident assignment of the Mariana Trench material to that species. The uncertain status of the P. magna holotype, together with dubious reports of this species from the Tonga and Japan trenches (Beliaev and Brueggeman 1989), means that Princaxelia kahat sp. nov. is here treated as a distinct species rather than as a range extension of P. magna.

Scanning electron microscopy images also remain rare for deep-sea amphipods and is especially limited for Pardaliscidae. Martin et al. (1993) reported scale-like structures across the body surface of a pardaliscid amphipod and suggested that similar structures may occur more widely within the family. In contrast, SEM examination of Princaxelia malohi sp. nov. showed a lack of prominent surface scales and instead have a smooth exoskeletal surface (Fig. 6). A faint surface pattern is visible only along the ventral margin of the gnathopod dactylus in Princaxelia malohi sp. nov. (Fig. 6D). Although the taxonomic value of this character is currently uncertain, the presence or absence of surface sculpturing may become useful as SEM is more widely applied to deep-sea amphipod taxonomy.

Biogeography

Current distribution records suggest that Princaxelia may be more widespread and more diverse than formal taxonomy presently indicate. Although P. abyssalis has been reported from eight trench systems, it has only been formally described from the Kermadec Trench. Reports from other trench systems generally lack sufficient morphological detail or physical specimens to confirm these identifications and may instead represent undescribed species. This interpretation is supported by the absence of recent confirmed records matching the description of P. abyssalis from those regions, and was speculated in (Jamieson et al. 2022). Additional evidence for a wider distribution of the genus comes from unidentified specimens resembling Princaxelia observed in video footage from the Java Trench at 5,760–6,957 m (Jamieson et al. 2022), and from a damaged specimen collected from the Diamantina Fracture Zone at 7,009 m (Jamieson and Weston 2023). Continued sampling, particularly when paired with high-resolution imaging and DNA sequencing, will be necessary to determine whether these records represent known species, range extensions, or additional undescribed taxa.

The discovery of Princaxelia kahat sp. nov. provides the first evidence of more than one Princaxelia species occurring within a single trench system. Both P. marianaensis and Princaxelia kahat sp. nov. are known from the Mariana Trench, but they occur in markedly different settings. Princaxelia marianaensis was collected from cold seep habitats at the Shinkai Seep Field at 5,683 m (Tomikawa et al. 2021), whereas Princaxelia kahat sp. nov. was collected from boulder-field habitats below 8,000 m. This suggests possible niche partitioning by depth, habitat type, or both. However, the true distributions and ecological ranges of these species remain poorly constrained, and substantially more material will be required to test whether these patterns reflect habitat specialization, depth stratification, or under sampling.

The Atlantic record of P. stephenseni remains especially intriguing. This species is currently the only representative of Princaxelia identified from the Atlantic Ocean and occurs at a comparatively shallow depth relative to most other members of the genus (Stephensen 1931). This may indicate that additional Princaxelia species remain undiscovered in Atlantic deep-sea ecosystems, or that the genus is influenced by temperature as well as depth, with shallower occurrences possible in colder polar or subpolar environments. At present, however, the limited material available for P. stephenseni prevents strong conclusions about its relationship to other species in the genus.

Behaviour and ecology

In situ video footage from the Tonga Trench provides new insight into the behaviour and ecology of Princaxelia. Similar predatory behaviour to what has been captured in these observations was previously observed in baited video footage from the Tonga Trench, where the individuals were identified as P. abyssalis (Jamieson et al. 2012). Given the new material described here, those observations may instead represent Princaxelia malohi sp. nov. The observed behaviour is consistent with an active benthopelagic predator that exploits dense aggregations of scavenging amphipods at baited deployments.

The locomotory behaviour of Princaxelia also contrasts with previous suggestions that pardaliscid amphipods are primarily pelagic (Bousfield 1979). In the available footage, individuals were repeatedly observed sitting on the sediment with pereopods 5–7 outstretched, gliding over the sediment using pereopods 3 and 4 with assistance from the pleopods, and swimming above the seafloor. Pereopods 6 and 7 did not appear to be used directly for locomotion. During free swimming, Princaxelia adopted a distinctive posture in which pereopods 3 and 4 were tucked beneath the pereon, pereopod 5 was angled upward from the ischium, and pereopods 6 and 7 trailed behind the body (Fig. 16 C, D). This combination of sediment-associated ambush behaviour and short-distance swimming is similar to the predatory strategies described for morphologically comparable amphipods in the family Eusiridae (Klages and Gutt 1990).

Across all available footage, Princaxelia individuals were observed feeding on other amphipods. At baited deployments, they appeared to exploit the high local abundance of scavenging amphipods, particularly Hirondellea sp. in the Mariana Trench (Fig. 16E). However, previous gut-content analysis indicates that members of the genus may also consume, or incidentally ingest, sponge and echinoid material (Kamenskaya 1981). This suggests that while amphipods may be the primary observed prey in baited-camera settings, the broader diet of Princaxelia may be more diverse than video observations alone indicate.

The relatively high abundance of Princaxelia in some hadal observations suggests that these predators may play an important ecological role in trench food webs (Dasgupta et al. 2024). In the Tonga Trench at 8,200 m, up to five individuals were observed simultaneously, while in the Mariana Trench at 8,098 m, up to seven individuals were observed at the same baited deployment. High local abundance is also supported by epibenthic sledge material from the Kuril–Kamchatka Trench, where Princaxelia individuals made up approximately 60% of the amphipods collected at 9,427 m (Brandt et al. 2016). These observations indicate that Princaxelia may be locally abundant where prey densities are high, particularly around baited food falls or natural aggregations of scavenging amphipods.

Despite their role as conspicuous predators of amphipods, Princaxelia may also be important prey for larger hadal predators. Individuals have been observed being consumed by snailfish through suction feeding at upper hadal depths (Dasgupta et al. 2024). This predator-prey relationship is notable because fishes appear to reach their lower depth limit at approximately 8,300 m, close to the depth range occupied by several Princaxelia species, including the two new species described here. Below this depth, where fishes are absent, Princaxelia may occupy a particularly important predatory role within hadal amphipod communities.