MEPS 485:223-234 (2013)  -  DOI: https://doi.org/10.3354/meps10341

ABSTRACT: Feeding habits of 3 hagfish species were investigated along a depth gradient (~50 to 900 m) in New Zealand using nitrogen (δ¹⁵N) and carbon (δ¹³C) stable isotopes. Neomyxine biniplicata had the lowest mean δ¹⁵N value (14.2‰), followed by Eptatretus cirrhatus (14.9 ‰) and Eptatretus sp. 1 (15.8‰). Neomyxine biniplicata (~50 m depth) was characterized by (1) relative low lipid content in muscles and (2) consistent body condition index which together with its trophic position indicated that this species probably acquires its food by active predation, supplemented by opportunistic scavenging. Eptatretus cirrhatus (48 to 912 m) and Eptatretus sp. 1 (290 to 922 m) had similar morphology, but their δ¹⁵N signature indicated that they were feeding on slightly different trophic levels. For Eptatretus sp. 1, the combination of (1) variable lipid content, indicating phases of feeding and fasting, (2) decreasing body condition index with depth, indicating less regular feeding at depth, (3) increasing δ¹⁵N with depth and (4) decreasing δ¹³C signature with depth, pointed towards a feeding behaviour specialized in scavenging on large but rare falls of high-level predators such as whales, sharks or bony fishes. On the other hand, E. cirrhatus was characterized by (1) less variable lipid content, (2) a body condition index not influenced by depth, (3) δ¹⁵N values decreasing with depth and δ¹³C values constant across its depth range, which is likely to indicate a more opportunistic and mobile feeding behaviour on a range of prey.

KEY WORDS: Fish · Ecology · Nitrogen · Carbon · Body condition index · Lipid content · Eptatretus · Neomyxine

Full text in pdf format Supplementary material

Cite this article as: Zintzen V, Rogers KM, Roberts CD, Stewart AL, Anderson MJ (2013) Hagfish feeding habits along a depth gradient inferred from stable isotopes. Mar Ecol Prog Ser 485:223-234. https://doi.org/10.3354/meps10341 Export citation Share:    Facebook - - linkedIn