Myxine phantasma, Mincarone & Plachetzki & McCORD & Winegard & Fernholm & Gonzalez & Fudge, 2021

Mincarone, M. M., Plachetzki, D., McCORD, C. L., Winegard, T. M., Fernholm, B., Gonzalez, C. J. & Fudge, D. S., 2021, Review of the hagfishes (Myxinidae) from the Galapagos Islands, with descriptions of four new species and their phylogenetic relationships, Zoological Journal of the Linnean Society 192, pp. 453-474 : 467-469

MYXINE PHANTASMA SP. NOV.

GHOST HAGFISH

( FIGS 1, 2H, 4C, 5, 6E; TABLES 4–6)

Z o o b a n k r e g i s t r a t i o n: u r n: l s i d: z o o b a n k. org:act: 3A877CFC-03B5-40F8-8190-E90EEE79025C

Holotype: SIO 19-86, 510 mm, off NE Santa Cruz Island , 00°30’57.68”S, 90°10’17.90”W, 688 m depth,

Valeska Yamile, sta. G24, baited trap, Douglas Fudge et al., 11 June 2019, 07:59–11:10 h.

Paratypes: MCCDRS 9405 COI, 16S, 54 (178–426 mm) and SIO 19-83 COI, 16S, 2 (442–480 mm), taken with the holotype . MCCDRS 9398 COI, 16S, 1 (206 mm), off NE Santa Cruz Island , 00°29’21.27”S, 90°10’22.48”W, 789 m depth, Valeska Yamile, baited trap, Douglas Fudge et al., 29 May 2019 , 10:16–11:37 h. MCCDRS 9400 COI, 16S, 3 (187–365 mm), off NE Santa Cruz Island , 00°29’46.47”S, 90°12’00.73”W, 815 m depth, Valeska Yamile, baited trap, Douglas Fudge et al., 29 May 2019 , 09:21–12:39 h.

Diagnosis: Myxine phantasma differs from all congeners, except M. debueni Wisner & McMillan, 1995 from the Strait of Magellan, Chile, M. fernholmi Wisner & McMillan, 1995 from the Falkland Islands and M. garmani Jordan & Snyder, 1901 from Japan, by having six pairs of gill pouches and a 3/2 multicusp pattern of teeth. Myxine phantasma differs from these congeners by having: 18–24 prebranchial pores (vs. 27–29 in M. garmani ), 64–75 trunk pores (vs. 76 in M. debueni , 80–83 in M. fernholmi and 52–61 in M. garmani ), 6–10 tail pores (vs. 12–13 in M. garmani ), 90–106 total pores (vs. 113–121 in M. fernholmi ); and by having two bilaterally symmetrical nasal-sinus papillae in the dorsal surface of the nasal sinus (vs. one single nasal-sinus papilla in M. debueni ). In addition, M. phantasma has a totally unpigmented body, which makes its skin transparent in live specimens. This unique character has been not described for any other species of Myxine ( Wisner & McMillan, 1995; Mok, 2001; McMillan & Wisner, 2004).

Description: Body elongated, subcylindrical at prebranchial and branchial regions, laterally compressed at trunk and strongly compressed at tail. Rostrum triangular with rounded tip. Two conspicuous, bilaterally symmetrical nasal-sinus papillae in the dorsal surface of the nasal sinus. Eyespots absent. Three pairs of barbels on head: first two about equal in size (0.7–1.6% TL) and adjacent to opening of nasopharyngeal duct; third pair longer (1.3–2.2% TL) and immediately adjacent to mouth. Ventral finfold well developed (2–6 mm high), beginning within anterior 10% of trunk, extending backward to the cloaca. Caudal finfold thin, rounded, beginning immediately posterior to edge of cloaca, extending around tail to dorsal surface, ending about over cloaca.

Body proportions (in percentage of TL; description of the holotype followed by paratype in brackets): prebranchial length 23.5 (23.0–26.4); preventral length 24.1 (24.1–29.2); trunk length 65.7 (59.3–68.2); tail length 11.2 (9.0–12.4); body width at PCD 3.7 (2.1–3.8); body depth at PCD 4.7 (2.8–4.7); body depth including VFF 5.0 (4.2–6.4); body depth excluding VFF 4.2 (3.2–5.6); body depth at cloaca 4.3 (2.9–4.5); tail depth 4.3 (4.1–5.6).

Counts (description of the holotype followed by paratype in brackets): multicusp pattern 3/2; anterior unicusps 8 (7–10); posterior unicusps 7 (7–9); total cusps 40 (38–46). Prebranchial pores 19 (18–24); trunk pores 65 (64–75); tail pores 8 (6–10); total pores 92 (90–106).

Six pairs of gill pouches, with efferent branchial ducts on either side combined into a single external gill aperture posterior to the gill pouches. Gill aperture on the left side confluent with the pharyngocutaneous duct aperture. Dental muscle overlies the first pair of gill pouches. Ventral aorta not branched.

Colour (in life): body overall pinkish due mostly to the underlying muscle and blood vessels coloration; skin transparent, without pigmentation; muscles, gill pouches, liver and slime glands visible through the skin ( Fig. 4). Colour (in alcohol): body overall whitish to light beige. Specimens filmed in situ reveals a subtle blue tinge on the tail when illuminated with white light ( Fig. 6).

Distribution and habitat: Galapagos Islands: known from 62 specimens collected in three stations off northeastern Santa Cruz Island, between 688 and 815 m depth ( Fig. 1).

Etymology: From the Greek φάντασ Μ α, ghost, which refers to the transparent skin and lack of melaninbased pigmentation in this species. It is a noun in apposition.

PHYLOGENY

Our phylogenetic results are consistent with previous analyses of similar data from hagfishes ( Kuo et al., 2003; Kuo et al., 2010; Fernholm et al., 2013; Zintzen et al., 2015; Song & Kim, 2020). With just two loci and ~1 kb of sequence alignment data, we expected limited support for some nodes. For instance, our phylogenetic analyses do not resolve the position of the Neomyxine species , nor do they support several more recent bifurcations among several eptatretid clades. However, by focusing our attention on only those nodes that receive greater than 0.75 alrt and/ or 95% ultrafast bootstrap support (http://www. iqtree.org/doc), our results are sufficient to conclude the following regarding the evolutionary history of Galapagos hagfishes.

The eight species described thus far from the Galapagos ( Rubicundus lakeside , Eptatretus mccoskeri , E. bobwisneri , E. grouseri , E. goslinei , Myxine greggi , M. martinii and M. phantasma ) represent five distinct clades (four described here and Rubicundus previously described morphologically), which most likely suggests multiple, relatively recent colonization events of hagfishes to the Galapagos. Our data also suggest that species diversification may have occurred in the Galapagos following the introduction of the ancestor of M. martinii and M. greggi , as well as after the arrival of the ancestor of E. mccoskeri and E. goslinei . While the Galapagos endemics E. mccoskeri and E. goslinei are clearly differentiated morphologically, the molecular data do not have the power to resolve these two species, indicating an early stage of speciation.

OBSERVATIONS FROM BAITED REMOTE UNDERWATER VIDEO (BRUV)

Information provided by BRUV revealed a rich and apparently healthy benthic environment. We obtained video at depth of five Galapagos hagfish species ( Fig. 6). The only species that we caught but failed to film live was M. martinii . The video data provide us with information about the colour and behaviour at depth of hagfishes, as well as the nature of the habitat at each collection station, particularly those off north-eastern Santa Cruz (near Gordon Rocks) and off north-western Fernandina (Cabo Douglas). Bottom type near Gordon Rocks ranged from flat, sandy, naked bottom, covered by only thin sediments to a structured, irregular, rocky bottom, inhabited by a great variety of corals, shrimps, isopods, crabs, sea urchins, brittle stars and fishes. Among the fishes reliably identified from the video are: the broadnose sevengill shark Notorynchus cepedianus (Péron, 1807) (Hexanchidae) , recently reported for the first time in the Galapagos by Buglass et al. (2020), the mottled scorpionfish Pontinus clemensi Fitch, 1955 (Scorpaenidae) and the snailfish Paraliparis sp. (Liparidae) . Unidentified fishes of the families Ophichthidae , Macrouridae and Moridae were also observed. At Cabo Douglas, the bottom was a mix of rocks of different sizes and shapes, resulting in an irregular and structured habitat. Due to the BRUV design used near Cabo Douglas, most of the field of view was obstructed by the trap and the images of the fauna were limited. Nevertheless, it was still possible to observe some unidentified species of crabs ( Majidae and Xanthidae ) and fishes ( Moridae and Macrouridae ). The BRUV used near Gordon Rocks gave a less obstructed view of the bottom and fauna ( Fig. 6B–E).