Other formats

    Adobe Portable Document Format file (facsimile images)   TEI XML file   ePub eBook file  

Connect

    mail icontwitter iconBlogspot iconrss icon

Zoology Publications from Victoria University of Wellington—Nos. 33 and 34

L. Gnathophis habenatus ?longicaudatus. Text-figs. 2, A-F; 5, A-C

L. Gnathophis habenatus ?longicaudatus. Text-figs. 2, A-F; 5, A-C.

Material Examined. C.S.I.R.O. Division of Fisheries and Oceanography (Cronulla) Collection (85 specimens,: 5.5mm total length, 7.0, ca. 10.0, 10.8, ca. 12.0, Warreen Station 3/40, 32° 47′ S., 152° 40′ E, 11/1/40, N100 (100 cm. net), o (oblique tow), 0-200m; ca. 6.0, 6.2, 15.8, St. 31/40, 34° 08′ S., 151° 32′ E., 25/4/40, N100, O, 200m; 6.2, 15.7, St. 31/40, 34° 08′ S., 151° 32′ E, 25/4/40, N70, O, 200m; 8.3, St. 104/38, 34° 03.5′ S , 151° 39′ E, 15/12/38, N200, H (horizontal tow), 100m, 30mins (minutes); 8.9, St 7/40, 30° 13′ S., 153° 33′ E, 14/1/40, N70, V (vertical tow), 250-0m; 9.5, 10.6, 12.5, 14.9, Discovery St 2724, 34° 05′ S., 151° 29′ E, 9/10/50, N100B, 108-0m; 10.2, 84.5, Warreen St 132/39, 30° 15′ S., 153° 33′ E., 5/5/39, N100, H, 100m, 60mins; 11.2, St. 24/40, 34° 08.5′ S., 151° 36.5′ E, 27/2/40, N70, O, 200m; 11.3, 17.0, St 127/39, 33° 18′ S., 152° 13′ E., 3/5/39, N70, H. 100m, 60mins; 13.2, 15.0, 16.9, 20.3, 23.0, St 127/39, 33° 18′ S., 152° 13′ E., 3/5/39, N100, H, 100m. 60mins; 14.0, St. 22/42, 39° 46′ S., 148° 36′ E, 18/4/42, N70, H, 50-0m; 14.6, Discovery St 2723, 34° 08′ S., 151° 33′ E, 9/10/50, N100B, 93-0m; 14.7, 23.1, 23.5, 23.5, 24.3, 24.3, 25.2, 25.3, 26.1, 31.1, 31.6, 32.4, 65.5, 67.3, 74.6, 79.0, 82.7, 85.0, 87.1, Warreen St unknown, ca. 34° 04′ S., 151° 15′ E., ?/39, no other data; 15.0, St 40/39, 35° 06′ S., 150° 51′ E, 25/1/39, N70, H, 25m, 30mins; 19.8, 20.8, 27.5, St 129/39, 32° 34′ S., 152° 55′ E., 3/5/39, N100, H, 100m, 60mins; 21.2, off Eden, Victoria, 9/3/48, V, 46-0m; 21.9, 27.2, 27.4, probably Port Hacking, New South Wales, ?/1 /43; 26.8, St 3/49, 32° 03′ S, 114° 53.5′ E, 7/4/49, N70, H, surface, 10mins; 30.0, St 118/39, 37° 10′ S., 150° 24′ E., 17/4/39, N70, H, 100m, 30mins; 37.0, 37.5, 45.0, 55.0, 64.5, St 26/40, 32° 34′ S, 152° 55′ E, 20/4/40, N100, O, 0-200m; 37.5, Discovery St 2718, 33° 55′ S., 153° 02′ E., 9/10/50, N100B, 117-0m; 38.8, Warreen St 27/40, 31° 08′ S., 153° 05′ E., 21/4/40, N70, V, 50-0m; 46.3, 58.2, 62.3, 81.2, St. 205/39, 34° 03′ S, 151° 11.5′ E, 4/8/39, N200, H, surface, 15mins; 57.3, 96.7, St 37/48, 28° 48′ S., 113° 36′ E., 2/9/48, N70, H, surface, 15mins; 62.0, Derwent Hunter St 16/52, Port Fairy wide, 27/8/52, N70, V, 730-0m; 65.4, Warreen St 224/39, 35° 03′ S., 151° 09′ E, 4/10/39, N200, H, surface; 78.8, St 195/39, 24° 21′ S., 153° 22′ E., 7/7/39, N100, O, 0-200m; 81.8, St 64/38, 34° 04.5′ S., 151° 14′ E., 10/11/38, N200, H, 25m, 30mins; 82.9, St 64/38, 34° 04.5′ S., 151° 14′ E, 10/11/38, N100, H, surface, 30mins; 84.1, St 214/39, 36° 14′ S., 150° 25′ E., 3/9/39, N70, O,200m; 85.2, 86.9, St 138/39, 30° 56′ S., 153° 21′ E., 16/5/39, N100, H, 100m, 60mins: 89.0, Discovery St 2710, 37° 06′ S, 150° 48′ E., 6/10/50, N100B, 84-0m; 90.4, Warreen St 36/48, 28° 48′ S., 113° 47′ E 1/9/48, N70, H, surface, 15mins; 109.9, St. 302/39, 37° 29′ S., 150° 20′ E., page 26
Text-fig. 2 (A-G).—L. Gnathophis habenatus ?longicaudatus (Ramsay & Ogilby, 1888), 90.4mm total length, I.F.O. (Cronulla), Warreen St. 36/48, Western Australia, 1/9/48. Fig. A—Lateral view to show position of major vertical blood vessels and intestinal pigment. Fig. B—Lateral view of head. Fig. G—Lateral view of caudal region to show pigment on tips of hypurals and anal bases. Figs. D-F—Gnathophis habenatus longicaudatus (Ramsay & Ogilby, 1888), 60.8mm total length, Western Australian Museum No. P5240, Western Australia, 16/5/62. Fig. D—Lateral view to show posterior dorsal origin, ventral pigment and fan-like caudal. Fig. E—Lateral view of head to show dorsal and ventral iris pigment. Fig. F—Ventral view of abdomen to show pigment spots.

Text-fig. 2 (A-G).—L. Gnathophis habenatus ?longicaudatus (Ramsay & Ogilby, 1888), 90.4mm total length, I.F.O. (Cronulla), Warreen St. 36/48, Western Australia, 1/9/48. Fig. A—Lateral view to show position of major vertical blood vessels and intestinal pigment. Fig. B—Lateral view of head. Fig. G—Lateral view of caudal region to show pigment on tips of hypurals and anal bases. Figs. D-F—Gnathophis habenatus longicaudatus (Ramsay & Ogilby, 1888), 60.8mm total length, Western Australian Museum No. P5240, Western Australia, 16/5/62. Fig. D—Lateral view to show posterior dorsal origin, ventral pigment and fan-like caudal. Fig. E—Lateral view of head to show dorsal and ventral iris pigment. Fig. F—Ventral view of abdomen to show pigment spots.

page 27 15/11/39, N100, O, 200m; 115.0, 116.4, St 55/38, 39° 58′ S, 148° 57′ E, 16/10/38, N200, H, 100m, 30mins; 124.9, St 55/38. 39° 58′ S, 148° 57′ E., 16/10/38, N100, H, 100m, 30mins; 138.2, St 284/39, 42° 35′ S., 148° 38′ E, 12/11/39, N100, O, 200m.

Australian Museum Collection (7 specimens,: ca. 69.0, Aust. Mus. regd. no. IB.3786, Budgewoi Beach, N.S.W., 17/7/57; 71.8, Aust. Mus. regd. no. IA.2478, Manly, N.S.W., 1907: 72.5, Aust. Mus. regd. no. IA.7047, Palm Beach, N.S.W., 12/9/36; 76.3, Aust. Mus. regd. no. I. 3250, Dee Why, N.S.W.; ca. 77.5, Aust. Mus. regd. no. IB.4503, Collaroy Beach, N.S.W., 27/8/59; 78.0, Aust. Mus. regd. no. IA.3765, Bellingen, N.S.W., State Fisheries Dept; 78.8, Aust. Mus. regd. no. IB.2586, Newport Beach, N.S.W., 15/10/50.

Western Australian Museum Collection (89 specimens,: Egg with developing prelarva, Accession No. P5306, Lancelin, 37 miles west of West End, Rottnest Island, Western Australia, 12/11/62 (0450–0520 hrs), larval net in 37 metres; 9.0, P5278, 40mls, 3/7/62 (0200), In, 37m; 15.1, P5305, 44mls, 3/7/62 (0300–0330), In, 110m; 15.9, P5302, 34mls, 3/7/62 (0030–0100), N70, surface; 18.1, P5304, 44mls, 3/7/62 (0300–0330), In, 110m; 26.3, P5164, 48mls, 24/1/62 (0015–0045), N70, surface; 26.4, P5160, 49mls, 1/8/61 (0500–0530), N70, surface; 26.5, P5162, 50mls, 6/10/61 (0000–0030), In, 27m; 30.3, P5243, 44mls, 9/5/62 (0515–0545), N70, surface; 30.8, 33.2, 76.8, P5170, 50mls, 20/6/61 (2300–2330), In, 91m; 34.7, P5175, 39mls, 20/6/61 (2115–2145), In, 55m; 37.1, P5288, 43mls, 7/6/62 (0140–0210), In, 37m; 42.5, P5307, 44mls, 3/7/62 (0300–0330), N70, surface; 43.3, P5152, 21mls, 23/8/61 (0000–0030), In, 110m; 43.4, P5163, 35mls, 13/7/61 (1900-1930), N70, surface; 45.3, P5301, 42mls, 3/7/62 (0230–0300), N70, surface; 45.8, P5303, 44mls, 3/7/62 (0300–0330), In, 110m; 46.9, P5300, 42mls, 3/7/62 (0230–0300), N70, surface; 51.8, P5290, 38mls, 3/7/62 (0100–0130), In, 37m; 51.8, P5294, 42mls, 3/7/62 (0230–0300), N70, surface; 54.2, P5281, 44mls, 3/7/62 (0300–0330), In, 110m; 55.2, P5232A, 45 mls, 9/5/62 (0100–0130), in, 37m; 55.4, P5311, 42mls, 3/7/62, in, 37m; 60.8 (glass-eel), P5240, 21mls, 16/5/62 (0050–0120), In, 37m; 67.0, P5153, 52mls, 31/7/61 (0230–0300), N70, surface; 67.3, P5310, 42mls, 3/7/62, In, 37m; 68.2, P5312, 39mls, 7/6/62 (0500–0530), In, 37m; 70.0, P5227, 18mls, 4/4/62 (0200), N70, surface; 70.7 P5283, 42mls, 3/7/62, In, 37m; 80.0, 103.0, 107.5, P5179, 50mls, 5/7/61 (2050–2120), In, 27m; 80.5, P5284, 38mls, 7/6/62 (0530–0600), In, 37m; 81.4, P5293, 38mls, 3/7/62, N70, surface; 82.5, 86.7, 89.5, 91.9, P5173, 50mls, 15/11/61 (0200–0230), In, 110m; 82.5, P5148, 51mls, 15/11/61 (0120–0150), In, 55m; 83.0, P5238, 20mls, 4/4/62 (0430), In, 37m; 83.2, P5161, 40mls, 13/12/61 (0010–0040), In, 37m; 83.9, P5158, 47mls, 12/9/61 (0200–0230), In, 55m; 84.0, P5149, 35mls, 13/7/61 (1900-1930), In, 27m; 85.5, P5286, 45mls, 7/6/62 (0005–0035), In, 37m; 86.9, P5235, 50mls, 9/5/62 (0030), In, 37m; 88.1, P5249, 50mls, 9/5/62, In, 37m; 88.5, P5245, 40mls, 9/5/62 (0100–0130), In, 37m; 92.5, P5157, 38mls, 13/7/61 (1945-2015), N70, surface; 93.0, P5250, 50mls, 9/5/62, In, 37m; 100.6, P5231, 20mls, 4/4/62 (0330), In, 110m; 110.5, P5174, 53mls, 13/7/61, In, 55m.

Since this paper was written the following additional 36 specimens became available: 59.5, P5535, 22mls, 20/9/62 (0215), In, 37m; 52.0, P5540, 44mls, 10/10/62 (0230), N70, surface; 34.6, P5541, 43mls, 10/10/62 (0330), In, 110m; 39.7, P5542, 42mls, 10/10/62 (0415), In, 37m; 35.0, 37.5, 39.4, 42mls, 10/10/62 (0415), N70, surface; 37.1, P5544, 41mls, 10/10/62 (0500), N70, surface; 85.9, P5546, 32mls, 24/10/62 (0010), N70, surface; 79.1, P5547, 33mls, 24/10/62 (0110) In, surface; 47.3, 53.7, 70.6, 71.2, 85.3, 85.4, P5550, 43mls, 7/11/62 (0010), In, 37m; 43.5, 46.4, P5551, 44mls, 7/11/62 (0100), In, 37m; 50.8, 52.8, 70.8, 84.1, page 28 P5552, 45mls, 7/11/62 (0140), In, 37m; 37.8, 42.3, 47.5, 51.6, 59.2, 91.7, P5553, 44mls, 7/11/62 (0230), N70, surface; 61.9, 64.2, P5554, 7/11/62, In, 110m; 75.4, P5555, 42mls, 7/11/62 (0440), In, 37m; 47.5, 52.6, 71.1, P5558, 41mls, 15/11/62 (0100), In, 37m; 85.0, P5559, 40mls, 15/11/62 (0320), In, 110m; 57.8, P5560, 39mls, 15/11/62 (0410), In, 37m.

Centre d'Océanographie de l'Institut Français d'Océanie Collection (5 specimens,: 60.4, IFO Station P As 8, 23° 01′ S., 158° 05′ E., 9/5/58 (0637hrs), 0.5m net, no. 2 mesh, 2 oblique tows in ca. 300 metres; 71.2, 85.1, St 7-6, 22° 35′ S., 166° 16′ E., 26/7/62, 5ft midwater trawl (Isaacs-Kidd), H, ca. 50m; 79.1, St G 26, 23° 17′ S.. 165° 44′ E., 4/4/62, 3ft midwater trawl (Isaacs-Kidd), H, ca. 16m; 107.3, St 7-3, 22° 35′ S., 166° 16′ E, 17/7/62, 5ft midwater trawl (Isaacs-Kidd) H, ca. 35m.

New Zealand Collections (5 specimens,: 98.0, VUW Zoo. Dept. Coll., probably Cook Strait, no other data; 111.1, Dom. Mus. No. 1834, 1ml north of Stephen's Island, Cook Strait, 100 fathoms, 17/12/54; 111.5, Cant. Mus. Coll., Grey River, Westland, about a mile from river mouth, 24/10/54; 114.0, Dom. Mus. No. 2527, Port Underwood; 116.9, VUW Zoo. Dept. Coll., Queen Charlotte Sound, no other data.

Description. 191 specimens: total lengths 5.5mm-138mm, myomeres ca. 116–131, dorsal fin-rays 121–215, anal fin-rays 96–173. Description made from a specimen which has just begun metamorphosis, DFO (Cronulla) Coll. Warreen St 36/48 (measurements in mm) : total length 90.4, head 4.6, snout 1.5, eye 1.0, upper jaw 1.9, postorbital 2.4, branchial aperture 0.6, pectoral 1.1, snout-vent 79.2, predorsal 68.3, depth just before eye 1.9, depth at pectoral origin 3.6, depth at midpoint between pectoral and vent 10.5, depth at anal origin 7.1. Branchiostegal
Text-fig. 3.—Gnathophis habenatus, frequency of myomeres in leptocephali and adults from Western Australia, Eastern Australia and New Zealand.

Text-fig. 3.—Gnathophis habenatus, frequency of myomeres in leptocephali and adults from Western Australia, Eastern Australia and New Zealand.

and pectoral rays not formed, dorsal rays before level of vent 109, total dorsal rays 163, anal rays 136, caudal rays 5 + 4. Teeth 1 + 1 + V + 9 over 1 + X + 3. Myomeres 97 + 26 = 123. a–d = 33. Vertical vessels at 12, 37, 41. Anterior margin of gall-bladder at 37.
Body relatively elongate, much compressed except along the head, not deep, the maximum depth 8.5 in total length, reducing more gradually in front of the midpoint. Head short, about 19 in total, distinct from trunk; snout sharply conical, about one-third of head, its dorsal profile barely convex; anterior nostril a short tube directed forwards, with a simple rim; posterior nostril widely open, page 29
Text-fig. 4.—Gnathophis habenatus, variation in number of preanal myomeres (expressed as a percentage of the total) with total length in 141 leptocephali.

Text-fig. 4.—Gnathophis habenatus, variation in number of preanal myomeres (expressed as a percentage of the total) with total length in 141 leptocephali.

page 30 oval; eye oval, the greatest diameter vertical, but horizontal diameter contained about twice in snout; cleft of mouth oblique, extending to beneath anterior edge of pupil; teeth similar in size and formula to those of Anguilla but directed a little more anteriorly; branchial aperture small, barely curved. Pectoral fin short, a little less than twice in postorbital, with a thick, fleshy base but a thin web; fin-rays not sufficiently well-developed to be counted. Dorsal fin low but with a distinct anterior margin over the 64th myomere. Anal fin similar but with the basal structures not so well developed. Caudal fin conspicuous, partly separated from posterior tips of dorsal and anal fins.

Pigmentation in formalin in the form of a minute, black, crescentic patch on the anterodorsal aspect of the eyeball and a larger, conspicuous brown to black patch below the posteroventral margin of the iris; a series of three round melanophores on each side of the throat, each with a compact, central dark area and a lighter surrounding area; a paired ventral somatic series of about 95 rounded melanophores following the course of the intestine from the ninth segment, thence on each segment to the vent; a series of minute, stellate, more diffuse melanophores on the bases of the anal fin-rays with an average of about two spots for every three rays; a row of darker, more conspicuous spots on the tips of the hypurals; a few spots on the bases of the terminal dorsal rays; a few deep spots on the posterior part of the spinal cord and vertebral column.

Vertical blood-vessels to the intestine numerous, occurring at segments 12, 20, 23, 27, 29, 31, 37, 41, but of these only those at 12, 37 and 41 are conspicuous. Gall-bladder easily seen, its anterior margin at myomere 37.

Remarks. Whitley (1937, pp. 8–9, fig. 3) describes a leptocephalus from Maroubra, Sydney (Aust. Mus. regd. no. IA.6893) which he refers to Gnathophis habenatus longicaudatus. I have examined this specimen and find that it agrees well with Whitley's description but that it is a typical muraenid leptocephalus showing little similarity other than in number of myomeres to the larvae of G. habenatus. It has a short head, no pectoral, a rounded tail and pigment in a series of minute, rounded spots ventral to the gut anterior to the gall-bladder but dorsal to the gut between the gall-bladder and the vent, with pigment spots on the jaws, branchial region and occiput. There is no crescentic black patch below the iris and the eye is round, not oval. The Maroubra specimen is described by Whitley as agreeing in myomere number, 127, with a practically metamorphosed specimen of G. habenatus longicaudatus from Cape Everard, Victoria (Aust. Mus. regd. no. IA.6263). The Maroubra specimen does not now have a silver iris, a characteristic feature of late Gnathophis larvae, but this may have faded. As I have yet to examine in detail the muraenid leptocephali from the east Australian area I cannot finally identify this specimen, but there is little doubt that it belongs to the Muraenidae. The Cape Everard specimen, which I have not seen, is quite probably referable to G. habenatus longicaudatus in view of its area of capture.

Although there are otherwise no observable morphological differences between leptocephali of G. habenatus collected from western Australia through to eastern Australia, New Caledonia and New Zealand, there are slight differences in the average number of myomeres in each group. Text-fig. 3 shows that east Australian larvae have slightly more numerous myomeres than do those from the west and also New Zealand adults of G. habenatus habenatus, the average numbers being 125.6, 122.3 and 123.2 respectively. The five New Caledonia larvae have an average of 126.6 myomeres and are therefore included in the east Australian group. The differences shown may indicate three distinct populations, one spawning off west Australia, one off New South Wales, and one in New page 31 Zealand waters, but this subject will be discussed in more detail later. The two peaks occurring in the frequency diagram of east Australian larvae may possibly indicate a difference in total number of myomeres with sex but this has not been confirmed in the adult specimens of G. habenatus habenatus from Wellington Harbour.

Dakin & Colefax (1940, p. 204, figs. 289 and 290) describe and illustrate the eggs and larvae of two eel species taken off the coast of New South Wales. The larvae have over 150 myomeres, and there is no pectoral fin even in a larva of 11.2mm. These characters rule out the identification of these species with the gnathophid eels considered in this paper.

A number of leptocephali have been described from other parts of the world which bear at least superficial resemblances to those of G. habenatus. The best known of these is L. Congermuraena mystax from the central and southern Atlantic Ocean. Except for the absence of the minute crescentic patch of black pigment above the eye (which in any case may easily be overlooked) and the greater number of myomeres (132–147) this species is strikingly similar in form and pigmentation to the leptocephali of G. habenatus. Lea's description (1913, pp. 18–21, figs. 12–15, pl. 3, nos. 1–2) of the leptocephali of the Atlantic species so closely agrees with the material described here that it seems likely that Congermuraena mystax (Delaroche) is a Gnathophis. Leptocephali of this genus are readily distinguishable from those of Ariosoma which have pigment present on the lateral surface before the lateral line on each myoseptum, a round eve and more numerous ventral spots; from those of the deep-water genus Pseudoxeno-mystax which have a relatively deep body, a round eye without black pigment below the iris and pigment on the anal bases; and from Conger which may have large spots on the lateral surface but always possesses distinct spots above the anal fin.

Indo-Pacific congrid leptocephali of the Gnathophis group are otherwise known only from several species described from the Pacific coast of Central America. Leptocephalus cingulus (Garman, 1899) has 131–133 myomeres but ventral pigment is absent. With this number of myomeres this species could possibly be the young of G. catalinensis; the absence of ventral pigment has been observed in specimens of L. Gnathophis habenatus which have been preserved for some years in formalin. Leptocephalus cinctus (Garman, 1899), listed by Ancona (1928, p. 106) as a possible synonym of L. Congermuraena mystax has 138 myomeres and a silver iris and may also be referable to G. catalinensis, but this is known from one specimen and there is no record of the range of vertebrae in this species. L. dentatus (Garman, 1899) and L. obtusus (Garman, 1899) both have pigment below the lateral line and thus do not fall readily into the Gnathophis group of leptocephali. They are probably larvae of Ariosoma gilberti (Ogilby, 1898) which inhabits the Pacific coast of Central America.

Growth and Metamorphosis (Text-figs. 4, 5, A-E). The 191 leptocephali of Gnathophis habenatus examined in the preparation of the above description range in total length from 5.5mm to 138.2mm. The collections include one egg with an almost completely developed embryo. About 150 of these specimens are leptocephali in various stages of metamorphosis in which the vent is in the process of movement towards its final position in the elver and the body undergoing contraction in length. Included in the latter group are two almost metamorphosed glass-eels, that is, elvers which have the general form of the adult and still have the pigment of the leptocephali. The material examined is sufficient, therefore, to give a very good indication of the morphological changes taking place during page 32 active growth of the leptocephalus and the metamorphosis to the elver. In these periods of development there are changes in (a) myomere shape, (b) the number of preanal myomeres indicating movement of the vent, (c) the number of dorsal and anal fin-rays, (d) the number and disposition of the teeth, (e) the condition of the nasal organ, (f) the distribution of pigment.

(a) Myomere Shape. At a length of below 10.0mm the myomeres are shallow V-shaped with the dorsal and ventral limbs of the future W-shaped myomere only just appearing in the myomeres at the middle of the body length. Post-anal myomeres have at best only a very slight V-shaped flexure. The tendency for the V to deepen and become W-shaped increases steadily until at about 40mm total length all myomeres are W-shaped, although the post-anal segments have very short dorsal and ventral limbs.

(b) Number of Preanal Myomeres (Text-fig. 4). During growth from about 5mm to 30mm in total length the vent undergoes a rapid actual movement from a position located at about myomere 55 posteriorly to myomere 100 in a total of about 125. The vent remains in this relative position until the total length reaches about 80mm when in most cases metamorphosis begins. There is then a rapid reduction in the length of the gut so that the vent moves forwards (or regresses) to reach a final position in the elver at about the 37th myomere. This movement is illustrated for the 154 specimens in Text-fig. 4. To eliminate as wide a variation as possible in the plot of preanal myomeres against total length due to variation in the total number of myomeres the number of preanal myomeres in each specimen was reduced to a percentage of the total number. It was also found that a plot of preanal length against total length produced a much wider scatter due to the relatively large errors made in measuring the preanal lengths especially of the smaller specimens, since these often could not be laid very flat and they were easily stretched beyond their normal lengths. It is important to note that, as the change in the number of preanal myomeres shows, there is an actual movement of the vent at first posteriorly and then after metamorphosis begins, anteriorly. This method of analysis eliminates the factors of differential growth rates of the preanal region or of the tail. As Text-fig. 4 shows, there is a great variation in the size of the leptocephalus at which metamorphosis begins. The usual length of the body at the onset of metamorphosis is between 70mm and 90mm, but many larvae reach much greater lengths without sign of metamorphosis. The significance of the latter is discussed in greater detail below.

(c) Number of Dorsal and Anal Fin-rays. The dorsal fin-rays begin to differentiate so that their basal structures are sufficiently well developed to be counted at a total length of about 37.0mm. At this length the difference between the dorsal and anal origins is about 25 complete myomeres and throughout growth and metamorphosis the dorsal origin advances towards its final position above the middle of the pectoral at about the same rate as the advance of the vent. The anal rays do not differentiate until the body length is about 55mm, when they number about 120. There is a slow increase in the number of dorsal and anal fin-rays to their final numbers of about 180 and 140 respectively. Longer leptocephali, showing little sign of normal metamorphosis, have a greater number of dorsal fin-rays, about 205–210.

(d) Teeth (Text-fig. 5). At a length of about 5mm the larva has 2–3 very acute teeth on the upper jaw, of which the first is long, needle-like, curved slightly downwards but generally directed forwards. On the lower jaw there are also 2–3 similar teeth, the first of which is directed anterodorsally. The lower teeth always project outside the upper and are usually fewer in number during all stages of development. At 20mm total length the upper teeth number about 5, page 33
Text-fig. 5.—L. Gnathophis habenatus ?longicaudatus, growth and early metamorphosis, showing in each figure lateral views of body, snout and tip of caudal region. Fig. A—10.2mm total length, I.F.O. (Cronulla), Warreen St. 132/39, Eastern Australia, 5/5/39. Fig. B—20.3mm total length, I.F.O. (Cronulla), Warreen St. 127/39. Eastern Australia, 3/5/39. Fig. C—37.0mm total length, I.F.O. (Cronulla), Warreen St. 26/40, Eastern Australia, 20/4/40. Fig. D—85.2mm total length, I.F.O. (Cronulla), Warreen St. 138/39, Eastern Australia, 16/5/39. Fig. E—138.2mm total length, I.F.O. (Cronulla) regd. no. B137, Warreen St. 284/39, off Tasmania, 12/11/39.

Text-fig. 5.—L. Gnathophis habenatus ?longicaudatus, growth and early metamorphosis, showing in each figure lateral views of body, snout and tip of caudal region. Fig. A—10.2mm total length, I.F.O. (Cronulla), Warreen St. 132/39, Eastern Australia, 5/5/39. Fig. B—20.3mm total length, I.F.O. (Cronulla), Warreen St. 127/39. Eastern Australia, 3/5/39. Fig. C—37.0mm total length, I.F.O. (Cronulla), Warreen St. 26/40, Eastern Australia, 20/4/40. Fig. D—85.2mm total length, I.F.O. (Cronulla), Warreen St. 138/39, Eastern Australia, 16/5/39. Fig. E—138.2mm total length, I.F.O. (Cronulla) regd. no. B137, Warreen St. 284/39, off Tasmania, 12/11/39.

page 34 with the first and second teeth larger than the remainder; the lower number about 4, all of which are longer than the upper. At 40mm the teeth are distributed as follows: 1 + 1 + II + 9 over 1 + IV + 4, at 85mm they are 1 + V + 21 over 1 + VI + 12 and at 138mm they are 1 + 1 + VI + 19 over 1 + VIII + 7. In some specimens, as in Anguilla, there is a minute, needle-like tooth on the dorsal surface of the snout, directed anteriorly, immediately above the most anterior tooth. As growth proceeds, the bases of the larval teeth become overgrown by the fleshy snout and the teeth appear to be shorter in later leptocephali. As the metamorphosing leptocephalus takes on the appearance of the elver the larval teeth are progressively lost, from the anterior part of the jaws posteriorly.

(e) Nasal Organ (Text-fig. 5). Until about 40mm total length the nasal organ has a single wide opening with a slightly raised rim. The upper and lower margins later grow together at the midpoint of this opening to form the two nostrils. The anterior nostril becomes tubular but remains with a simple rim until well after metamorphosis of the elver, when the scroll-like condition develops. The posterior nostril remains widely open until the elver, when it narrows to form a slit.

(f) Pigmentation (Text-fig. 5). Changes in pigmentation are the most conspicuous changes in the development of the larva. The early leptocephalus at 5mm total length shows pigment in the characteristic somatic ventral series as about six rather irregularly-spaced, rounded spots beginning at the level of the 20th segment. There are one or two minute black spots on the ventral surface of the lower jaw and a few minute specks of black pigment above the caudal tip. A larva of 10mm shows an increase in the number of ventral spots to about 12; there are two or three spots on the lower jaw and there has been an increase in the caudal pigment to a short dorsal line of spots and a few ventral flecks. At 20mm there are about 18 ventral spots, 4–5 spots on the lower jaw and a wide scattering of pigment on the caudal fin. At about the time when the posterior movement of the vent has ceased the pigment on the lower jaw and caudal region has disappeared and there remain about 30–40 spots in the ventral series. At this time, however, a small crescentic patch of pigment appears below the iris. This is retained throughout the metamorphosis to the juvenile eel. At about 70mm-80mm in length the ventral series consists of about 45 spots beginning at about the 10th segment; there are 1–3 round spots on the throat on each side of the heart; pigment has appeared on the bases of the anal rays, deep on the ends of the hypurals, on the posterior dorsal rays and on the posterior tip of the vertebral column. Metamorphosis then rapidly takes place, and although the throat pigment is lost the ventral series is retained until the end of the metamorphosis as are the spots on the bases of the anal rays. In some of the larger leptocephali of 120mm or more a second row of spots develops internal to the first row. Just after the beginning of metamorphosis the silver iris pigment, which is so characteristic of the adult eel, develops but the black pigment above and below the iris is retained. Late metamorphosing leptocephali or glass-eels show a short line of very closely-packed spots on each side of the vent, which is perhaps due to the telescoping of the posterior part of the ventral series as the vent regresses (Text-fig. 2, F).

(g) Other Changes. At the end of the period during which there is an active movement of the vent posteriorly the caudal fin begins to differentiate from the tips of the dorsal and anal fins to form a distinct caudal fan. This is carried to the extreme in the elver, which, in common with the elvers of Anguilla, have a broad caudal fin (Text-fig. 2, D). There is no observable change in the relative page 35 position of the gall-bladder during development. In all specimens where this structure was seen its anterior margin stood within about two segments of the 37th myomere.

Geographical Range and Location of the Spawning Areas. The 155 larvae have a wide geographical range. A total of 90 specimens were collected from the eastern Australian coast between latitudes of about 24° S. and 43° S.—i.e., from the middle of the Queensland coast to Maria Island, Tasmania, and off Port Fairy, Victoria; 92 specimens were collected from Western Australia from 29°S. to 32° S.; 5 specimens from the New Caledonia region between the South Bellona Reef and Nouméa; 5 specimens from New Zealand consisting of one from the west coast of the South Island and the remainder from the Cook Strait region. Leptocephali of Gnathophis habenatus therefore appear to be more numerous in the areas relatively close to the western and eastern coasts of Australia in latitudes between 24° S. and 43° S. This does not reflect more frequent trawling in these areas since the waters of the New Caledonia region have been examined extensively in the years 1957–62 by the Institut Français d'Océanie, and although large collections of other leptocephali have been made, there have only been a few specimens of G. habenatus in these collections. The New Zealand region has been by no means ignored by institutions working this area, but the only specimens of this species collected have been occasional beachcast individuals.

In the eastern Australia-New Caledonia-New Zealand area the smallest leptocephali were collected from a locality some 20 miles from the New South Wales coast over the continental slope near the latitude of Sydney. The largest leptocephali were collected on the periphery of the area of distribution, but in general no simple pattern of increasing size towards the periphery is shown in these collections. Text-fig. 5 shows in greater detail the distribution of this group of 100 specimens. The indications are that the spawning area for eastern Australian G. habenatus is very likely to be off the New South Wales coast. The large variation in the total lengths of leptocephali collected at a number of stations would suggest spawnings at different times and that a few larvae may remain in the same area for some time, possibly as a result of countercurrents. The northward range of leptocephali along the Queensland coast and out to New Caledonia suggests that the leptocephali are not completely dependent upon the East Australian Current for their movement, and it is probable that countercurrents are important in the dispersal of the larvae.

As shown in Text-fig. 4, metamorphosis begins in most specimens at about 70mm-90mm total length. However, a few specimens in the collection are much larger than this, reaching 138.2mm. This greater length occurs only in specimens which were collected far from the indicated spawning area. For example, the five Cronulla specimens of greater length than 90mm were all collected from the coast of Victoria to Maria Island, Tasmania; one other large specimen was taken from near Nouméa, and the five New Zealand specimens are all larger than most specimens beginning metamorphosis on the Australian coast. On the other hand, four of the seven Australian Museum specimens beachcast near Sydney—that is, close to the spawning area, are all at stages in metamorphosis at a size which is less than all other metamorphosing specimens.

The geographical range of the material from Western Australia is more restricted than that from eastern Australia. The smallest specimens amongst these 92 larvae is an egg with its developing prelava taken from a position located 37 miles west of Rottnest Island in 37 metres; the largest specimen measures 110.5mm and is undergoing metamorphosis. There are two glass-eels in this collection, one page 36
Text-fig. 6.—Gnathophis habenatus, the distribution of 100 larvae in the south-west Pacific; each small circle represents at least one specimen; large circle represents 37 larvae.

Text-fig. 6.—Gnathophis habenatus, the distribution of 100 larvae in the south-west Pacific; each small circle represents at least one specimen; large circle represents 37 larvae.

of which has almost completed metamorphosis. The presence of very young leptocephali in this collection indicates that G. habenatus is also spawning off the coast of Western Australia as well as off the east coast, probably in the area west of Rottnest Island over the continental slope.

The eastern Australian collection contains only a few metamorphosing individuals and was made over a considerable area. The western Australian collection, made over a much smaller area, contains nearly one-half as metamorphosing larvae. This latter collection includes specimens which cover essentially the same size-range as the eastern Australian material excepting the few larger larvae from the furthest limits of the distribution of the latter. In both collections metamorphosis appears at a length of some 80mm, which accordingly can be regarded as the usual or normal size for metamorphosis indicating perhaps that the process is physiological and endogenous. In contrast, there are the larger metamorphosing larvae in the eastern Australian collections, of extreme size in the New Zealand collections and rare in the western Australian collections.

page 37

If the stimulus to metamorphosis is entirely endogenous, the few largest larvae may be consequent from individual irregularities in the time of onset of the process. Such individuals, because of their relatively low numbers associated with a "faulty" mechanism for metamorphosis, would not ordinarily be regarded as providing or maintaining a population of adults. The absence as yet of adults from New Caledonia and northern Queensland, supports the idea of a failure to complete normal metamorphosis.

The relative abundance of the species in New Zealand with knowledge only of large metamorphosing larvae on our coasts, suggests that a postponement of metamorphosis to sizes well above 90mm to 100mm may well be normal. Setting aside the question of a subspecific distinction in this matter (relating to the difference in preanal lengths in the adults of G. habenatus habenatus and G. h. longicaudatus, there remains the probability that the mechanism of metamorphosis is endogenous but subject to retardation in operation, a retardation dependent on an environmental control. The demonstration of adults at the extremes of the larval distribution to the north in Australia and New Caledonia and large collections of larvae for the New Zealand area are required before further analysis of the possible mechanism can be undertaken.

It has been shown that New Zealand specimens of G. habenatus differ from Australian specimens in having a slightly greater preanal length in the adult as well as slightly fewer vertebrae. If metamorphosis has been delayed by a prolonged larval life to produce a longer larva it is possible that this would have an effect on the morphological characters of the adult. This has already been shown for Conger myriaster by Takai (1959, p. 548). In order that a larger than normal leptocephalus metamorphose into a young eel with the characteristically short preanal length of G. habenatus longicaudatus then the amount of regression of the vent would be relatively greater. The vent may only be capable of a certain degree of regression for morphological reasons (although I can suggest no such specific reason) and the adult would have a relatively long preanal length, as in the New Zealand G. habenatus habenatus.

Bathymetric Distribution. Many of the hauls in which leptocephali of G. habenatus were collected were oblique or vertical and a precise depth range is therefore not available. The hauls ranged from the surface to 730m, but the majority of specimens were taken in depths ranging from the surface to 200m. The species appears to be most abundant in about 50m-100m in common with the larvae of many species of eels.