Japanese Journal of Ichthyology 魚 類 学 雑 誌 Vol.34, No.3 1987 34巻3号1987年

Identity of the Galaxioid Fishes of the consistent with a report of a single elongate, Genus Jenyns from transparent juvenile of an undetermined species Southern Chile of Aplochiton from waters around the shores of Lago Villarica in central/southern Chile (Mc-

Robert M. McDowall and Kazuhiro Nakaya Dowall, 1969a). Small juveniles, again of an (Received November 6, 1986) undetermined species of Aplochiton were collected from the sea in southern Chile (Puerto Aysen) by

The genus Aplochiton Jenyns is one of two Mr. Akira Zama and reported on briefly (Mc- genera in the family Aplochitonidae. There are Dowall, 1984). two species of Aplochiton recognised from southern So little is known of these fishes that the capture

South America and the Falkland Islands and, in of a small series of notably large Aplochiton in the addition, one species of Lovettia from Tasmania vicinity of Puerto Aysen is worth reporting and

(McDowall, 1971a). A close phylogenetic rela- assists somewhat in clarifying the taxonomic tionship between Aplochiton and Lovettia should status of these fishes. not be assumed, although both genera clearly belong to the assemblage of galaxiid-aplochitonid Material examined fishes that is characteristic of cool southern tem- perate fresh waters (McDowall, 1969b; Fink, : 2 individuals, HUMZ (Lab- 1984). The relationships between Aplochiton and oratory of Marine Zoology, Hokkaido University) Lovettia and their relationships with galaxiids are 095682, 095704, 21 January 1982, Fiordo Steele, Chile, not well enough understood to justify restructuring 48•K13'S, 73•K23'W, 361, 326.5mm SL. 1, SNP (PA) of these generic and familial arrangements al- (Servicio Nacional de Pesca (Puerto Aysen)) 76, 1 January 1982, Estero Ventisquero, Chile, 48•K15'S, though some authors present a contrived sim- 73•K35'W, 271mm SL. 1, HUMZ 095704, 1 January plification by placing all genera in 1982, Fiordo Ventisquero, Chile, 48°13'S, 327 mm SL.

(Nelson, 1972; Rosen, 1974). This does little to 1, Instituto de la Patagonica 46, 21 September 1977, clarify relationships. Estero Veto, Chile, 49•K32'S, 74•K30'W, 189mm SL.

There are several nominal species of Aplochiton, A. zebra: 4, HUMZ 095696-9, 1 December 1982, the principle binomina being: Lago Quetro, Chile, 48•K7'S, 73•K7'W, 236-284mm SL. Aplochiton zebra Jenyns, 1842: collected from In addition data from the study of McDowall the Falkland Islands. (1971a) were reconsidered. Aplochiton taeniatus Jenyns, 1842: collected from

Tierra del Fuego. Study methods Aplochiton marinus Eigenmann, 1928: collected from Estero Cutipai, near Valdivia, Chile. Study methods adopted were those consistently These fishes are poorly known. Their tax- in use for study of the southern freshwater salm- onomy was examined by McDowall (1971a) who, oniform fishes (Galaxiidae, Aplochitonidae etc.) somewhat tentatively, treated A. marinus as a most recently reviewed in McDowall and junior synonym of A. taeniatus, although this Frankenberg (1981). These methods, in general, action has not been accepted by Campos (1973), follow Hubbs and Lagler (1958); distinctive Bahamonde and Pequeno (1975), or Zama and features include: measurements taken point to

Cardenas (1984). Neither Campos nor Baha- point with needle point dividers to the nearest monde and Pequeno gave reasons for their posi- 0.5 mm; fin ray counts include segmented rays tion but Zama and Cardenas discussed differences (excluding non-segmented procurrent rays); ver- in colouration between A. marinus and A. taeniatus. tebral counts exclude hypural centra. Biologically these fishes are also little known.

Campos (1969) described the location of spawning, Meristic and morphometric variation and the development of the eggs of A. taeniatus in Lago Llanquihue, a southern Chilean lake. Systematics of fishes depends heavily on the He wrote of accumulations of juveniles in the use of meristic and morphometric variation to littoral zone of Lago Llanquihue-possibly an clarify variation within and differences between indication of shoaling juvenile stages. This is species. The salmoniform fishes, the southern

•\ 377•\— 魚類学雑誌 Japan. J. Ichthyol. 34 (3), 1987 families like Aplochitonidae and Galaxiidae that were possibly diadromous. In one sample amongst them, are notoriously difficult in this (Lago Rinihue) vertebral number encompassed regard for several distinct reasons. 1. Meristic both ranges (59-65) suggesting that both dia- variation within populations tends to be high. dromous and non-diadromous stocks were present 2. Meristic variation between populations with in that lake. Campos (1974) demonstrated a the same life history strategy is also high. 3. similar broad range in vertebral number for Often there is clinal variation in features such as Galaxias maculatus (Jenyns) lower in that river vertebrae and fin ray numbers. 4. Furthermore, system which drains Lago Rinihue (Rio Valdivia- many salmoniforms have alternative life history Rio Calle Calle), although his bimodality did not strategies within species. Many species have both extend that far upstream. diadromous (sea migratory) and entirely freshwater Similar variability was also demonstrated for stocks and sometimes these stocks are sympatric A. taeniatus by McDowall (1971a) who described (Campos, 1974; McDowall, 1972, 1979; Northcote the following: Eye size decreased and snout length and Ward, 1985). The meristic features of di- increased with absolute size. Vertebral number adromous and non-diadromous stocks of a species was found to be bimodal with some populations may be distinctly different, eg. diadromous stocks at 62-66 and others at 69-73. The lower counts of Galaxias maculatus (Jenyns) may have 57 to were regarded as being from non-diadromous 61 vertebrae at the northern extremities of that stocks. Bimodality was also evident in gill raker species' range in New South Wales, Australia counts, but this feature subdivided the popula- (McDowall and Frankenberg, 1981) or up to tions into different subsets than those generated 60-66 in populations in far southern Chile by vertebral count. (McDowall, 1971b). Lacustrine populations have On the basis of vertebral count, the Rio Calle many fewer vertebrae, with 50-53 in northern Calle stocks (69-73) were different from all others Australian populations, but up to 54-60 in (62-66). Using gill rakers, however, the Rio southern Chile. Similar sorts of variation are Calle Calle fish grouped with most other popula- described for other galaxiids and also for Re- tions (17-23 gill rakers) fish from Lago Todos los tropinna (McDowall, 1979). Morphometric fea- Santos had higher counts (26-31) while those tures also tend to vary in relation to the above from Lago Llanquihue had fish belonging to both factors, differences between diadromous and non- subsets (21-23 and 30-34 gill rakers) . Thus diadromous stocks being particularly well re- various characters are varying in different ways cognised for characters like head length and eye between and within populations (in Rio Calle size. 5. In addition, allometric growth during Calle vertebral number is high gill raker number somatic growth of salmoniform fishes is widely low; in Lago Todos los Santos both are low; recognised in addition to which there is often ex- in Lago Llanquihue vertebral number is low but aggeration of some morphological features in gill raker numbers are both high and low). relation to the attainment of sexual maturity, In view of these sorts of variations, resolution especially in males. of the and the identification of the All these problems relating to between in- new series of specimens is not simple. Two dividual, within and between population, and plausible scenarios can be envisaged. One is within and between species variability do or can that there is a complex of species (or perhaps be expected to apply to Aplochiton. In view of subspecies) of Aplochiton in the rivers and lakes the meagre knowledge of the life history strategies of Patagonian South America and the Falkland of these fishes, resolution of their taxonomy has Islands, for which existing material is inadequate its difficulties. to allow resolution. Alternatively there are two McDowall (1971a) has already examined some (A. zebra, A. taeniatus) or perhaps three (A. of these questions for a limited array of Aplochiton marinus) species of Aplochiton, which have al- samples. In A. zebra, it was shown that eye size ternative within-species life history strategies decreases with growth. Vertebral numbers in (diadromous and non-diadromous) and that there the samples appeared to be bimodal with number are morphological differences within species as- being low (about 57-62) in populations that ap- sociated with each of the alternate life history peared to be landlocked and high (63-66) in those strategies. Material available for study and

•\ 378•\— McDowall and Nakaya: Galaxioids from Chile knowledge of the life histories of Aplochiton Identification species do not permit absolute clarification of these hypotheses. Material in the present study is easily divided As knowledge of within-species morphological into two taxa (see Table 1). variation related to similarly alternative life history Differences are quite clear in body depth, shape strategies in several species of Galaxias (McDowall, of the caudal peduncle, the sizes of the dorsal 1972) and Retropinna (McDowall, 1979; Northcote and anal fins, head length and depth, eye diameter, and Ward, 1985) is entirely consistent with the mouth size and other features. The four more types of variation discussed above for Aplochiton slender large-mouthed fish (Fig. 1) are typically we see no reason to modify the taxonomic ar- heavily covered on the upper sides and back with rangement of Aplochiton proposed by McDowall small round spots, and in some there is a general (1971a) in which two species A. zebra and A. peppering of small widely spaced melanophores. taeniatus were recognised and redescribed. They have 72-73 vertebrae. When alive the fish

Table 1. Morphometric and meristic variation in Aplochiton species (figures given as percentages of denominators in ratios).

•\ 379•\— 魚類学雑誌 Japan. J. Ichthyol. 34 (3), 1987

Fig. 2. Aplochiton zebra, 284mm SL, Lago Quetro, Chile. were bright silvery olive, and darker on the back. marinus from Chile. Zama and Cardenas (1984) These fish are consistent with A. taeniatus re- described one specimen (SNP (OA)76 of the present described by McDowall (1971a) from locations at study) as A. marinus on the basis that the specimen low elevations close to the sea. had distinct dark spots on the body, pointed The four somewhat (although not much) stouter snout, longer maxillary, smaller eye, slender, fish (Fig. 2) have much smaller mouths, are much well-developed teeth in each jaw. These are all darker in colour being blotchily black/olive on feature characteristics of and described for A. the back and upper sides with a series of narrow taeniatus (McDowall, 1971a) apart from the vertical chevron shaped bands across the sides. spotting on the back. McDowall, however, noted When alive they were dark black olive on the that he had seen no freshly captured specimens. back, paling to silvery olive on the sides and The syntypes of A. marinus (CAS 51275) do not belly. These fish have 64-66 vertebrae. They are have spotting on the back now, either, and we do consistent with those redescribed by McDowall not regard this character as having taxonomic (1971a) as A. zebra again being similar to fish significance in the present analysis. taken from lower elevation localities close to the Allometric growth. Morphometric data from sea. the new specimens are, in general, consistent with In spite of the fact that the specimens presently those published by McDowall (1971a) for both under discussion are far larger than any previously species except that eye diameter is smaller and reported, they would have been identified as A. snout length greater. These two differences are taeniatus and A. zebra using McDowall's (1971a) however consistent with allometric growth pat- key. terns previously described for Aplochiton. In Status of A. marinus. Although McDowall Figs. 3 and 4 data from the present study (cross) (1971a) treated A. marinus as a junior synonym of are added to those from McDowall (1971a). A. taeniatus, Campos (1973) and Bahamonde and Size. The present collection comprises by far Pequeno (1975) without comment, listed A. the largest examples of Aplochiton on record. To

•\ 380•\ McDowall and Nakaya: Galaxioids from Chile

• data from McDowall (1971a)

Fig. 3. Variation in eye size with growth in Aplochiton zebra (•œ, from McDowall, 1971a; x , present study).

• data from McDowall (1971a)

Fig. 4. Variation in eye and snout size with growth in Aplochiton taeniatus (•œ, from McDowall, 1971a; •~

, present study).

1971, there were records of A. taeniatus up to Geographical distribution. The earlier study by

240mm LCF (Ringuelet et al., 1967) and of A. McDowall (1971a) showed two major large dis- zebra to 240mm LCF (McDowall, 1971a). The junct areas of distribution for A. zebra and A. present collection raises maximum sizes of these taeniatus, viz. a northern area between about species to 384mm LCF for A. taeniatus and Valdivia and Puerto Montt (39-42•‹S) and a far

304mm for A. zebra. southern area, roughly between Puerto Natales

•\ 381•\— 魚類学雑誌 Japan. J. Ichthyol. 34(3), 1987

Fig. 5. Distribution of Aplochiton zebra and A. taeniatus. Open symbols: from McDowall, 1971a; closed symbols: present study. Circles: A. zebra; stars: A. taeniatus. Arrows indicate general area from which additional material in present study originated. and Ushuaia at the far tip of South America Distribution maps for Galaxias maculatus and

(50•K-56•KS). A. zebra was also reported from the G. platei show similar huge gaps in known oc- Falkland Islands. Two major areas of absence currence in this same areas (McDowall, 1971b) are those between latitudes 42 and 50•KS along the but Zama and Cardenas (1984) again show their Chilean (western) coastline, as well as the entire presence at Aysen and vicinity, at intermediate Argentinian (eastern) coastline of Patagonia. latitudes. Lack of records clearly is due to The present specimens and those reported on lack of collection. Absence of Aplochiton along by Zama and Cardenas (1984) provide localities the Atlantic coast of Patagonia is consistent in all between these two disjunct areas in Chile (48- known collections (although G. maculatus and 49•KS) and it can probably safely be suggested that G. platei both occur there). This distributional both species are present along the entire Chilean gap seems more likely to be real. coastline. (There has been virtually no systematic Life history strategy. McDowall (1971a, 1984) collection of fishes in this remote and inaccessible has suggested that the life histories of at least area.) some populations of Aplochiton are diadromous.

•\ 382•\ McDowall and Nakaya: Galaxioids from Chile

The reports of small juveniles from the sea McDowall, R.M. 1984. Southern Hemisphere fresh- (McDowall, 1984) indicate this and suggest that water salmoniforms: development and relationships. amphidromy is involved (Myers, 1949; McDowall, Pages 150-153 in H. G. Moser et al., eds. Ontogeny 1987). The finding of these very large adults of and systematics of fishes. Spec. Publ., Amer. Soc. Ichthyol. Herpetol., 1, 760pp. both A. zebra and A. taeniatus, in localities close McDowall, R.M. 1987. The occurrence and dis- to the sea, and their high vertebral counts offers tribution of diadromy in fishes. Amer. Fish. Soc. circumstantial support for this hypothesis. The Symp., 1: 1-13. lake spawning population of A. taeniatus which McDowall, R.M. and R.S. Frankenberg. 1981. The Campos (1969) studied is interpreted as a lake galaxiid fishes of Australia. Rec. Austr. Mus., 33 (10): 443-605. limited stock of that species. Myers, G.S. 1949. Usage of anadromous, catadro- mous and allied terms for migratory fishes. Copeia, Acknowledgments 1949(1): 89-97. Nelson, G.J. 1972. Cephalic sensory canals, pitlines We are grateful to Mr. Akira Zama for in- and the classification of esocoid fishes, with notes formation and the loan of specimens, and to on galaxiids and other teleosts. Amer. Mus. Nov., Chile-Patagonia Expedition Party of Hokkaido (2492): 1-49. University for collection of specimens. Northcote, T.G. and F.J. Ward. 1985. Lake resident and migratory smelt Retropinna retropinna (Rich- ardson) of the lower Waikato River system, New Literature cited Zealand. J. Fish Biol., 27 (1): 113-129. Ringuelet, R.A., R.H. Aramburu and R.A. Aramburu. Bahamonde, N. and G. Pequeno. 1975. Peces de 1967. Los peces Argentinos de agua dulce. Com- Chile-lista sistematica. Mus. Nac. Hist. Nat., ision de Investigacion Scientifica, La Plata, 602 pp. Chile, Publ. Occ., 21: 1-20. Rosen, D.E. 1974. Phylogeny and zoogeography of Campos, H. 1969. Reproduccion del Aplochiton salmoniform fishes and relationships of Lepido- taeniatus Jenyns. Bol. Mus. Nac. Hist. Nat., galaxias salamandroides. Bull. Amer. Mus. Nat. Chile, 29: 207-221. Hist., 153: 267-325. Campos, H. 1973. Lista de peces de aguas continen- Zama, A. and E. Cardenas. 1984. Introduction into tales de Chile. Not. Mens. Mus. Nac. Hist. Nat., Aysen Chile of Pacific Salmon. No.9. Descriptive Chile, 19(198-9): 3-14. catalogue of marine and freshwater fishes from the Campos, H. 1974. Population studies of Galaxias Aysen Region, southern Chile, with zoogeographical maculatus (Jenyns) (Osteichthyes: Galaxiidae) in notes on the fish fauna. Servicio Nacional de Chile with reference to the number of vertebrae. Pesca, Chile and Japan International Co-operation Stud. Neotrop. Fauna, 9: 55-76. Agency, 75 pp. Fink, W.L. 1984. Basal euteleosts: relationships. Pages 202-206 in H.G. Moser et al., eds. Ontogeny (RMM: Fisheries Research Division, Ministry of and systematics of fishes. Spec. Publ., Amer. Soc. Agriculture and Fisheries, Christchurch, New Zealand; Ichthyol. Herpetol. 1, 760pp. KN: Laboratory of Marine Zoology, Faculty of Hubbs, C.L. and K.F. Lagler. 1958. Fishes of the Great Lakes region. Univ. of Michigan, Ann Fisheries, Hokkaido University, 3-1-1, Minato-machi, Arbor, 213pp. Hakodate 041, Japan) McDowall, R.M. 1969a. A juvenile of Aplochiton

Jenyns. Copeia, 1969(3): 631-632. チ リ南 部 のAplochiton属 の 分 類 McDowall, R. M. 1969b. Relationships of galaxioid fishes with a further discussion of salmoniform Robert M. McDowall・ 仲 谷 一 宏 classification. Copeia, 1969(4): 796-824. 以 前 に は 分 布 が 確 認 さ れ て い な か っ た チ リ 南 部 か ら McDowall, R.M. 1971a. Fishes of the family AplochitonidaeのAplochiton zebra Jenynsと.A. Aplochitonidae. J. Roy. Soc. New Zealand, 1(1): taeniatus Jenynsの2種 を 報 告 す る.本 研 究 に 用 い ら 31-52. れ た 最 大 の 標 本 は 以 前 に 知 ら れ て い た 最 大 の 個 体(両 種 McDowall, R.M. 1971b. The galaxiid fishes of South と も に240mmLCF)よ りは る か に 大 き く,A.zebra America. Zool. J. Linn. Soc., 50(1): 33-73. で304mmLCF,A.taeniatusで384mmLCFで あ McDowall, R.M. 1972. The species problem in freshwater fishes and the taxonomy of diadromous っ た.魚 体 の 大 き さ,海 に 極 め て 近 い 所 か ら と れ て い and lacustrine populations of Galaxias maculatus る こ と,脊 椎 骨 数 が 多 い こ と な ど か ら彼 ら は 両 側 回 遊 (Jenyns). J. Roy. Soc. New Zealand, 2(3): 325-367. 型 の ラ イ フ サ イ ク ル を も っ て い る も の と考 え ら れ る. McDowall, R.M. 1979. Fishes of the family Re- tropinnidae (Pisces: Salmoniformes)-a taxonomic (McDowall:ニ ュ ー ジ ー ラ ン ド,ク ラ イ ス トチ ャ ー チ, revision and synopsis. J. Roy. Soc. New Zealand, 農 業 漁 業 省;仲 谷:041函 館 市 港 町3-1-1北 海 道 大 学 9 (1): 85-121. 水 産 学 部 水 産 動 物 学 講 座)

•\ 383•\