ArticlePDF Available

A new species of the eleotrid genus Thalasseleotris (Teleostei: Gobioidei) from New Zealand coastal waters


Abstract and Figures

The eleotrid Thalasseleotris iota sp. nov. is described from reef habitats in New Zealand coastal marine waters at 2–34 m depth. It is distinguished from its only known congener, T. adela Hoese and Larson, from Australia, by having a narrower gill opening with lower edge extending forward to below posterior margin of eye or slightly behind eye (versus under middle of eye), less extensively scaled body, with naked area reaching to below second dorsal origin or beyond (versus near end of first dorsal fin), higher transverse scale count (usually 7–8 versus 6–7), lower vertebral count (10+16 versus 10+17), anterior nasal tube distinctly longer than posterior nasal tube (versus sub‐equal); two anal pterygiophores before first haemal spine (versus 1), and the uppermost and lowermost scale on base of caudal with enlarged peripheral cteni (versus no enlarged peripheral cteni). In addition, there are several colour differences, including a broad dark area on the basal third of the pectoral fin (versus absent) and thin wavy‐grey lines crossing dorsal, anal and caudal fins (versus absent). Maximum adult size is 32.6 mm SL (39.4 mm total length), which makes the pygmy sleeper one of New Zealand's smallest known vertebrate species.
Content may be subject to copyright.
Hoese & Roberts—New pygmy sleeper from New Zealand 417
A new species of the eleotrid genus Thalasseleotris (Teleostei:
Gobioidei) from New Zealand coastal waters
Douglass F. Hoese1 and Clive D. Roberts2
Abstract The eleotrid Thalasseleotris iota sp. nov. is described from reef habitats in
New Zealand coastal marine waters at 2–34 m depth. It is distinguished from its only
known congener, T. adela Hoese and Larson, from Australia, by having a narrower gill
opening with lower edge extending forward to below posterior margin of eye or slightly
behind eye (versus under middle of eye), less extensively scaled body, with naked area
reaching to below second dorsal origin or beyond (versus near end of first dorsal fin),
higher transverse scale count (usually 7–8 versus 6–7), lower vertebral count (10+16
versus 10+17), anterior nasal tube distinctly longer than posterior nasal tube (versus sub-
equal); two anal pterygiophores before first haemal spine (versus 1), and the uppermost
and lowermost scale on base of caudal with enlarged peripheral cteni (versus no enlarged
peripheral cteni). In addition, there are several colour differences, including a broad
dark area on the basal third of the pectoral fin (versus absent) and thin wavy-grey lines
crossing dorsal, anal and caudal fins (versus absent). Maximum adult size is 32.6 mm SL
(39.4 mm total length), which makes the pygmy sleeper one of New Zealand’s smallest
known vertebrate species.
Keywords Eleotridae; Thalasseleotris iota sp. nov.; pygmy sleeper; taxonomy; New Zealand
Fishes of the family Eleotridae, commonly known as sleepers or gudgeons, are primarily found
in freshwater. Marine eleotrids are rare, with only three genera restricted to fully marine reef
environments: the Indo-Pacific coral reef genus Calumia Smith, the temperate New Zealand
genus Grahamichthys Whitley, and the temperate Australasian genus Thalasseleotris Hoese
& Larson. Based on a number of shared derived characters, Hoese & Gill (1993) suggested a
close relationship between Grahamichthys and Thalasseleotris.
The genus Thalasseleotris resembles the gobiid genus Hetereleotris in having a similar body
form (10+17 vertebrae and first gill slit closed), but osteological study showed it belonged
to the family Eleotridae, with a number of features separating it from other gobioid fishes.
These included: no mesopterygoid; postcleithra absent; maxillo-dentary ligament attaching to
distinct process on outer face of premaxilla; urohyal without transverse shelf on ventral margin
(broad in most eleotrids); and basihyal triangular (Hoese & Larson 1987). Thalasseleotris was
erected by Hoese & Larson (1987) to accommodate the monotypic T. adela, discovered from
Tasmania in 1972 and eastern Australian coastal waters in 1974. This species, known as the
dusky marine gudgeon (Hoese 1994), was subsequently found to be widespread in southern
1Australian Museum, 6 College Street, Sydney, NSW 2010, Australia.
2Museum of New Zealand Te Papa Tongarewa, P.O. Box 467, Wellington, New Zealand.
R05006 Received 30 April 2005; accepted 4 July 2005; Online publication date 6 December 2005
Journal of the Royal Society of New Zealand
Volume 35, Number 4, December, 2005, pp 417–431
Journal of the Royal Society of New Zealand, Volume 35, 2005
Australia, inhabiting marine rocky reefs from Sydney, New South Wales, to Rottnest Island,
Western Australia.
Since 1984, a new species of marine eleotrid has been collected using rotenone ichthyocide
from subtidal reefs around New Zealand during Museum of New Zealand Te Papa Tongarewa
(Te Papa) fish surveys carried out by the junior author and colleagues. Previously, this fish
had eluded collection because it is difficult to observe and capture underwater, being small
(usually <30 mm SL) and semi-transparent. However, once its microhabitat was recognised
(sediment-covered rocky ledges on sides and floors of holes, cracks, and small caves in shallow
reefs), the sample rate improved. Scuba divers were able to collect substantial numbers of the
new sleeper (up to 32 specimens per station), particularly when using light from an underwater
torch to illuminate dark rocky recesses in the reef. Ryan & Paulin (1998) recorded it as the
pygmy sleeper, a common name in use by the Te Papa fish team at that time, and suggested
that the species is the smallest vertebrate in New Zealand.
During Te Papa fish surveys, a total of 376 specimens in 98 lots were collected, sufficient
to provide accurate data on character ranges and geographic variation. Examination of these
specimens has shown they belong to the eleotrid genus Thalasseleotris, and comparison with
the Australian species T. adela shows the New Zealand species to be distinct. The main pur-
pose of the present paper is to name and describe this new species of pygmy sleeper, which
is endemic to New Zealand coastal reefs.
In descriptions presented here, an asterisk indicates the value is that of the holotype; figures in
brackets in the text indicate the number of specimens examined with the preceding value. In
the list of material examined, the number of specimens is given in parentheses following the
registration number, and the size range in millimetre standard length is given. Station (stn.)
numbers refer to Te Papa Fish Section field surveys by scuba divers using rotenone, as follows:
B, Bare Island, Hawke Bay 1990; E, East Cape 1992, 1993, 1998; F, Fiordland 1993, 1994,
1995, 1998; H, Haast area, Westland 1999; K, Kapiti Island 1996; S, Stewart Island 1992.
Counts and measurements mostly follow the methods outlined by Hubbs & Lagler (1958).
In the species studied, the first ray of the anal fin and second dorsal fin is always a spine. The
remaining rays in these fins are segmented. The last ray of the second dorsal and anal fins,
as counted, consists of two elements in contact though distinct through the extreme base.
Segmented caudal rays include total segmented rays, and in most specimens there are eight
dorsal and eight ventral rays. Procurrent caudal rays are recorded as dorsal/ventral rays. The
first dorsal pterygiophore formula follows Birdsong et al. (1988).
Longitudinal scale counts (LS) are taken along the midline of the sides of the body, from
the central base of the pectoral fin to the end of the hypural, excluding small scales on the base
of the caudal fin. Transverse scale counts are taken from the anal spine upward and backward
to the base of the second dorsal fin (TRB), with very small scales, if present, adjacent to the
anal spine and dorsally below second dorsal fin not counted. Scale cteni terminology follows
Roberts (1993).
Vertebrae are counted from radiographs, and include the urostyle. Count of ventral caudal
peduncle spots includes distinct black spots or bars on ventral midline of caudal peduncle
behind anal insertion and before caudal insertion; a spot often present on the caudal base is
not included in the count. Caudal vertebrae are defined here as those with a closed haemal
arch and a developed haemal spine. Generally in gobioids, one or two anal pterygiophores
precede the first haemal spine (whether one or two depends on the genus, except for the current
genus). The osteology was determined from trypsin-prepared, cleared specimens, stained with
alizarin red and in some cases, counterstained with Alcian blue.
Hoese & Roberts—New pygmy sleeper from New Zealand 419
Head papillae were studied from preserved specimens (Hoese 1983). The papillae line
terminology follows Hoese (1983): LT refers to a more or less horizontal line, with each
papilla axis at right angles to the axis of the line; LL refers to a more or less horizontal line,
with each papilla axis along the axis of the line; VT refers to a more or less vertical line with
each papilla axis at right angles to the axis of the line; OT is used specifically for the oblique
lines on the operculum, with each papilla axis at a right angle to the axis of the line.
The following international standard abbreviations are used: AMS, Australian Museum,
Sydney; NMNZ, Museum of New Zealand Te Papa Tongarewa, Wellington (formerly the
National Museum of New Zealand); USNM, National Museum of Natural History, Smithsonian
Institution, Washington DC, USA; HL, head length; SL, standard length; TL, total length.
Suborder Gobioidei: Goby-like fishes.
Family Eleotridae: Sleepers, sensu Nelson (1994) and Eschmeyer (1997).
Thalasseleotris iota sp. nov.
New Zealand pygmy sleeper (Fig. 1–3, Tables 1–3)
SYNONYMY: Thalasseleotris sp.: Roberts 1993, p. 111, scale type described.
Thalasseleotris n. sp.: Ryan & Paulin 1998, p. 131, listed from NZ fiords; Roberts et al. 2005,
pp. 19, 70, listed from Milford Sound and Jackson Bay, habitat described.
MATERIAL EXAMINED: (n = 98 lots and 376 specimens).
HOLOTYPE: NMNZ P.38085, 27.2 mm SL, female (Fig. 1; Tables 1–2), “167 degree point”, Van-
couver Arm, Breaksea Sound, Fiordland, Te Papa field trip stn. F07 (45°30.65′S, 167°00.10′E),
3–15 m depth (low salinity layer 0–3 m), scuba and rotenone, 20 March 1993, M.V. Renown,
collectors C. D. Roberts, A. L. Stewart, C. D. Paulin, E. Harvey, and T. Willis.
PARATYPES (n = 22 lots and 146 specimens): AMS I.41344–001 (5), 21.2–28.7 mm (ex NMNZ
P.35061), Milford Sound; AMS I.41345–001 (2), 23.2–25.3 mm (ex NMNZ P.30228), Breaksea
Sound, same data as holotype; AMS I.41346–001 (5), 15.7–25.5 mm (ex NMNZ P.35090),
Milford Sound; LACM 56271–1 (4), 25–28 mm (ex NMNZ P.35100), Milford Sound; NMNZ
P.30228 (3), 14–30 mm, taken with holotype, “167 degree Point”, Vancouver Arm, Breaksea
Sound, Fiordland (45°30.65′S, 167°00.10′E), stn. F07, 0–15 m, 20 March 1993; NMNZ
P.32104 (4), 18–24 mm, Cove north of Evening Point, Bligh Sound, Fiordland (44°51.80′S,
167°31.40′E), stn. F28, 6–20 m, 21 March 1995; NMNZ P.32114 (5), 20–24 mm, 200 m inside
Turn Point, Bligh Sound, Fiordland (44°48.21′S, 167°32.21′E), stn. F29, 6–18 m, 22 March
1995; NMNZ P.32123 (6), 20–24 mm, East face of Alice Island, George Sound, Fiordland
(44°58.55′S, 167°25.60′E), stn. F30, 6–16 m, 22 March 1995; NMNZ P.32152 (3), 22–25 mm,
headland opposite Mount Alexander, Caswell Sound, Fiordland (45°02.61′S, 167°15.26′E),
stn. F33, 8–20 m, 24 March 1995; NMNZ P.32189 (7), 13–29 mm, “60 m Point”, north side
of Charles Sound (45°04.51′S, 167°06.63′E), stn. F37, 6–20 m, 26 March 1995; NMNZ
P.32203 (4), 25–32 mm, 300 m south of Old Point, Gold Arm, Charles Sound (45°06.33′S,
167°07.76′E), stn. F36, 5–20 m, 25 March 1995; NMNZ P.32220 (4), 20–25 mm, 0.5 nautical
miles from Toe Cove, Nancy Sound, Fiordland (45°10.36′S, 167°05.91′E), stn. F38, 5–18 m,
26 March 1995; NMNZ P.32262 (8), 13–26 mm, 1 nautical mile south of Deas Cove, south
side of Thompson Sound, Fiordland (45°12.78′S, 166°57.31′E), stn. F41, 8–24 m, 28 March
1995; NMNZ P.32288 (5), 12–22 mm, North Bay, Bauza Island, Doubtful Sound (45°17.80′S,
Journal of the Royal Society of New Zealand, Volume 35, 2005
166°54.23′E), stn. F43, 8–22 m, 29 March 1995; NMNZ P.32301 (4), 15–23 mm, Turn Point,
Crooked Arm, Doubtful Sound (45°24.38′S, 167°00.98′E), stn. F44, 6–18 m, 29 March 1995;
NMNZ P.32309 (5), 22–24 mm, Ranson Head, Malaspina Reach, Doubtful Sound (45°20.85′S,
167°01.36′E), stn. F45, 3–16 m, 30 March 1995; NMNZ P.35000 (21), 13–28 mm, east side
of Cavern Head, Preservation Inlet, Fiordland (46°04.83′S, 166°38.93′E), stn. F49, 7–10 m, 7
May 1997; NMNZ P.35061 (16), 20–31 mm, opposite Pater Point, Milford Sound, Fiordland
(44°37.43′S, 167°53.35′E), stn. F58, 8–16 m, 3 April 1998; NMNZ P.35071 (8), 21–25 mm,
Kettle Point, Milford Sound, Fiordland (44°37.33′S, 167°51.58′E), stn. F59, 10–18 m, 4 April
1998; NMNZ P.35090 (17), 16–33 mm, Bridget Point, Milford Sound, Fiordland (44°38.61′S,
167°54.86′E), stn. F61, 6–14 m, 5 April 1998; USNM 368996 (5), 23–28 mm (ex NMNZ
P.35090); USNM 368997 (5), 22.5–29.7 mm (ex NMNZ P.35061).
NON-TYPE MATERIAL: see Appendix 1.
COMPARATIVE MATERIAL EXAMINED: Thalasseleotris adela Hoese & Larson. AMS I.18241–017
(19 paratypes); AMS I.18241–035, (7 cleared and stained paratypes) and NMNZ P.30167 (ex
AMS I.18241–017, paratype), Port Hacking, Oakes Bay, NSW, Australia, 0–6 m, rotenone,
14 November 1974.
DIAGNOSIS: Thalasseleotris iota sp. nov. differs from the only other known species in the genus,
T. adela Hoese & Larson, 1987, in having a narrower gill opening, extending forward to below
posterior margin of eye or slightly behind eye (versus under middle of eye), less extensively
scaled body, with naked area reaching to below second dorsal origin or beyond (versus near
end of first dorsal fin), higher transverse scale count (usually 7–8 versus 6–7), modally lower
vertebral count (10+16 versus 10+17), anterior nasal tube distinctly longer than posterior
nasal tube (versus subequal) (Fig. 1); two anal pterygiophores before first haemal spine (ver-
sus 1), uppermost scale and lowermost scale on base of caudal fin with enlarged peripheral
cteni (versus no enlarged peripheral cteni). In addition, there are several colour differences,
including: lacking a black spot posteriorly on the first dorsal fin; usually 5–6 small black
more or less rounded spots on the ventral midline of the caudal peduncle (versus usually 4
horizontally elongate black spots); a broad black or dark brown area on basal third of pectoral
fin (versus no distinct black area or a small black spot dorsally on proximal upper 2–3 rays);
dorsal, anal, and caudal fins with thin wavy grey lines (versus uniformly pigmented); and no
dark dorsomedian line on band before first dorsal fin.
Fig. 1 Holotype of Thalasseleotris iota sp. nov., NMNZ P.38085, 27.2 mm SL, female, Vancouver
Arm, Breaksea Sound, Fiordland. Drawn by Michelle Freeborn. Size of papillae exaggerated.
Hoese & Roberts—New pygmy sleeper from New Zealand 421
DESCRIPTION: First dorsal pterygiophore formula 3(22110)* (30); first dorsal VI*(56), VII (3);
second dorsal I,8(5), I,9*(58), I,10(2); anal I,7(7), I,8*(57), I,9(1); pectoral rays 17–20, 19 in
holotype (Table 1), all rays branched except sometimes lowermost ray; pelvic rays I,5*(25);
segmented caudal rays 14(1), 15(2), 16*(40), 17(7); branched caudal rays 11(1), 12(17),
13*(21), 14(5); procurrent caudal rays 9/8(2), 9/7(1), 9/6(1), 8/7(1), 8/6(1), 7/6(4), 6/6(1),
6/5(1); vertebrae 10+15(1), 10+16(84), 10+17(1); longitudinal scale series 21–25 (Table 1);
transverse scale series (TRB) 6(1), 7(33), 8(9); ventral caudal peduncle spots 3(1), 4(57), 5(64),
5(6). Body measurements and proportions of holotype (27.2 mm SL, female) and a paratype
(28.4 mm SL, male) are given in Table 2.
Body scales peripheral ctenoid (Roberts 1993), extending forward to line from upper pec-
toral insertion to below second dorsal origin, sometimes slightly in front of origin. Ventrally
scales end in a line from or just behind pelvic base to just behind pectoral base. Belly scales
cycloid ventrally. Uppermost scale and lowermost scale on caudal base with slightly enlarged
peripheral cteni. Head, pectoral base, breast, and prepelvic area naked.
Table 1 Summary of meristic characters of Thalasseleotris iota sp. nov. and T. adela.
& non typesCharacters Holotype (n)T. adela*
First dorsal fin VI VI–VII 59 V–VII
Second dorsal fin I,9 I,8–10 65 I,9–10
Anal fin I,8 I,7–9 65 I,8–9
Pectoral fin rays 19 17–20 154 17–21
Pelvic fin I,5 I,5 25 I,5
Caudal fin branched rays 13 11–14 44 13–16
Caudal fin segmented rays 16 14–18 50 15–17
Procurrent caudal rays 8/6 6–9/5–8 12 7–8/6
Scales in longitudinal series 24 21–25 133 22–27
Scales in transverse series 7 6–8 43 6–7
Vertebrae (pre caudal +caudal) 10+16 10+15–17 86 10+17
Vertebrae (total) 26 25–27 86 27
*Data from Hoese & Larson (1987).
Table 2 Measurements and proportions of holotype and one paratype of Thalasseleotris iota sp. nov.
NMNZ P.38085
27.2 mm SL, ♀
NMNZ P.35052
28.4 mm SL, ♂
(mm) (%) (mm) (%)
Head length 8.0 29.4 8.9 31.3
Head width at posterior preopercular margin 4.9 18.0 6.5 22.9
Head depth at posterior preopercular margin 4.1 15.1 4.6 16.2
Upper jaw length 3.1 11.4 3.4 12.0
Snout length 1.9 7.0 2.2 7.7
Eye length 2.0 7.4 2.4 8.5
Body depth at pelvic origin 4.8 17.6 5.9 20.8
Body depth at anal origin 4.7 17.3 4.2 14.8
Pectoral fin length 6.5 23.9 7.9 27.8
Pelvic fin length 7.6 27.9 7.2 25.4
Caudal fin length 6.5 23.9 6.4 22.5
Journal of the Royal Society of New Zealand, Volume 35, 2005
Snout short, rounded in dorsal view, less than eye length in young, subequal to eye in large
adults; prominent bump on snout formed by protrusion of ascending processes of premaxil-
lae; both nostrils at end of short tube, anterior nostril approximately twice length of posterior
nostril; posterior nostril about one nostril diameter in front of eye, anterior nostril slightly less
than one nostril diameter from upper lip (almost in contact with lip in large adults, anterior and
posterior nostril separated by two nostril diameters; eyes elevated, shallow groove between
and behind eyes; interorbital very narrow, eyes almost touching; mouth oblique, angled about
40° with body axis; lips thick, jaws reaching posteriorly to below anterior margin to anterior
quarter of pupil, rarely almost to middle of eye; teeth minute, conical, with sharp points;
teeth in upper jaw in 4–5 rows anteriorly, tooth patch tapering laterally to 2–3 rows, teeth in
outermost row thicker and much longer than teeth in middle rows, teeth in innermost row
depressible, anteriorly subequal in length to or slightly shorter than teeth in outer row; teeth in
lower jaw in 4–5 rows anteriorly, tooth patch tapering laterally to 2 rows, teeth in outermost
row longest confined to anterior half of jaw, much larger than teeth in middle rows, innermost
row slightly shorter in length than teeth in outer row, teeth becoming smaller posteriorly; gill
opening very broad, extending forward to below posterior margin of eye or slightly behind eye,
gill membranes form a thin and short fold across isthmus; lower limb of first gill arch joined
to inner gill cover by membrane connecting lower 80–90% of arch; no spines on operculum;
chin with small rounded mental frenum, followed by two sensory papillae and with a papilla
at each side of frenum base.
First dorsal fin with broadly rounded margin, third to fifth dorsal spines longest and subequal,
fourth or fifth spine often slightly longer than third spine; second dorsal fin slightly shorter in
height to first dorsal fin; anal fin much shorter in height than first dorsal fin; base of second
dorsal fin less than distance from end of second dorsal fin to base of caudal fin; pectoral fins
with pointed margin, rays 10–13 (counted from top) longest, fin reaching to above anus to
first segmented ray of anal fin; caudal fin with rounded posterior margin, fin about two-thirds
length of head; pelvic fins widely separate, with distance between innermost rays of two fins
slightly less than half the base of each fin; pelvic fins long, reaching to before anus in young
to slightly beyond urogenital papilla in adult; pelvic rays branched, with membranes con-
necting tips of rays, but with shallow notch between; fourth pelvic ray with 3 terminal tip in
young and usually 4–5 in adult over 20 mm SL, fifth ray highly branched with 4–6 terminal
in adults over 20 mm SL, fourth and fifth rays longest, fifth ray subequal to or slightly shorter
than fourth ray and longer than third ray.
SEXUAL DIMORPHISM: Very little apparent difference between the sexes (Fig. 2; Table 2). Uro-
genital papilla similar in both sexes, broad at base and tapering slightly distally, with small
lobes at tip. Males with papilla flattened, sometimes with sparse concentration of minute black
dots covering base of papilla; lateral distal tip with 2–3 thickened lobes, contained in margin
of papilla. Females with little or no pigment on urogenital papilla; distal end with short lobes,
giving papilla a rough appearance, lengths about one-fifth length of papilla; papilla less flat-
tened and more rounded in cross section than that of male.
SENSORY PAPILLAE: Lateralis-system papillae prominent on head as short flaps, height of papilla
about 1.5–2 times width of papilla at base; no open head pores present. The pattern shows no
major difference from that figured for Thalasseleotris adela by Hoese & Larson (1987).
Usually three oval papillae along outer base of bulbous nasal sac; usually two along inner
base of nasal sac. An oblique LT line of papillae above posterior two-thirds of upper jaw. Ir-
regular LL line from above middle of jaw to below and near posterior end of pupil, followed
posteriorly by short VT line, followed by LT line of 3–4 papillae. Short LT line of 3–5 papillae
below upper LT line on cheek. Preopercular mandibular series with inner LL line composed of
Hoese & Roberts—New pygmy sleeper from New Zealand 423
Fig. 2 New Zealand pygmy sleeper, Thalasseleotris iota sp. nov.; two specimens, c. 30–35 mm TL, in
a cave at 2–3 m depth, Sentinel Cove, Mokohinau Islands, Northland, 5 November 2002. A, probable
male. B, gravid female. Underwater photos by Kendall Clements.
Journal of the Royal Society of New Zealand, Volume 35, 2005
single row of papillae and outer LT line, also composed of single row of papillae, with short
gap below and just behind posterior end of jaws in the LT line. The preopercular mandibular
series terminates anteriorly at side of chin, posteriorly terminating ventrally at preopercular
angle, but with 2–3 papillae on middle of posterior preopercular margin. Pair of LT papillae at
anterior tip of chin, behind mental frenum. Opercular series with anterior VT line and upper
oblique (OT) and separate lower oblique line (OT). LL line from posterior end of eye on nape
to just behind operculum interrupted by short sections of LT lines, composed of 1–3 papillae,
with the LT lines displaced slightly dorsally.
COLORATION IN ALCOHOL: Head and body light to dark brown. Head densely covered with minute,
evenly spaced, melanophores, denser on top of head. Oblique, dark brown bar, narrower than
pupil diameter, from anteroventral margin of eye crossing upper lip, ending dorsally on lower
lip; second dark brown, vertical bar from lower margin of eye to posterior end of jaws, darkest
posterior to jaws, often with upper portion obscure or broken into two short dark bars; two,
dark brown, oblique bars from posteroventral margin of eye, often merging with oval dark
blotches on cheek; bars often obscured by dark head coloration. A very small, dark brown,
oval, subcutaneous spot in contact with posterior margin of eye. Dorsal area behind eye dark
brown, followed by three brown cross bands, first above anterior quarter of operculum, second
above pectoral base and third before first dorsal origin, often with very thin, short, median,
longitudinal, black line on band above front half of operculum; second black, median line
on band above pectoral base, no dark line on dorsal midline on band before first dorsal fin. A
small black subcutaneous spot to either side of bump on snout. Papillae on top of head and
nape without black tips.
Upper pectoral base with scattered melanophores, not forming a distinct spot; ventrally an
elongate, oblique brown bar sloping ventrally posteriorly. Anteriorly on body a series of 4–6
brown vertical irregularly shaped brown bars from before first dorsal origin to end of first
dorsal fin, anterior bars extending onto sides of belly, posteriorly ending just below first dorsal
fin. In specimens smaller than 26 mm, a series of light brown blotches dorsally, first below
middle of first dorsal fin, second near end of second dorsal fin, on at second dorsal origin and
two below second dorsal fin; subcutaneous black spots, often paired between light blotches.
Two or three subcutaneous rectangular dark spots along base of anal fin. A series of 4–6 (usu-
ally 5 or 6) median dark brown spots ventrally on caudal peduncle from end of anal fin to
before base of caudal fin, anterior 2–3 spots sometimes fused to form a single dark brown line
anteriorly; sometimes a faint spot where caudal fin joins caudal peduncle. Posterior edges of
scale pockets much darker than centres of scales resulting in diamond shaped pattern on body.
Midline of body sometimes with small, elongate, black, subcutaneous lines, most distinctive
on caudal peduncle where a single line may extend over whole midside of caudal peduncle.
In light-coloured specimens less than 25 mm SL, midside usually with a series of 8–10 short,
vertical, dark brown bars. A dark brown vertical bar at base of caudal fin sometimes present,
often obscure. Anus usually surrounded by dark brown pigment laterally and anteriorly; in
large darker specimens the dark brown pigment extends forward as a midventral stripe onto
the belly.
Fins varying from light dusky grey to almost black. Distal margins of median and pelvic fins
lighter than rest of fin, often without pigment. Dorsal fins often with small black spots forming
thin lines in some specimens; fins often mottled grey to black with scattered light areas. Pec-
toral fin with a large, curved, black to dark brown bar near but not onto base of fin, extending
away from base of fin ventrally. Pelvic fin dusky grey to brown, with outer and distal margins
white to clear. Caudal fin with small back spots forming 6–8 wavy lines.
Hoese & Roberts—New pygmy sleeper from New Zealand 425
COLORATION OF FRESHLY COLLECTED SPECIMEN: (based on colour transparency by Andrew Stewart
of fresh specimen, NMNZ P.24653, 27 mm SL, from D’Urville Island). Head and body light
tan to dark brown; semi-transparent, clouding to opaque on death. Head spots and bands a
brownish orange. Vertical bands below first dorsal fin dark brown. Body a translucent light
brown. Posterior edges of scale pockets dark yellowish brown. Spots on dorsal and caudal
fins reddish brown.
COLORATION OF LIVE SPECIMENS: (based on underwater photos by Kendall Clements, University
of Auckland, of individuals from the Mokohinau Islands) (Fig. 2). Head and body a translu-
cent light brown to pale white. Head and body with mottling and banding formed by minute
rusty-dark brown spots on a yellowish orange background. Areas between mottling whitish to
light brown. Snout with fine brown dotting and small brown spot medially (smaller than pupil
diameter) just behind upper lip. Eye brownish orange, with light brown mottling dorsally; a
distinct, thin orange ring around pupil. Pupil emerald green. Head with an oblique thin dark
brown band from anteroventral margin of eye crossing jaws posteriorly. Four thin, brown
lines with irregular margins, radiating from eye ventrally (3) and posteriorly (1), lines often
incomplete, with gaps of light brown to grey. Elongate dark brown spot behind dorsoposterior
margin of eye. Sometimes (possibly male) with a large pearly white spot on cheek below mid-
dle of eye; behind this spot an elongate pearly white mark on preoperculum; an oval pearly
white mark (with brown near middle) covering upper half of anterior part of operculum. A
transverse brown irregular band with irregular margins dorsally above operculum, followed
by similar band above pectoral base. Posterior operculum often appearing reddish from blood
in gill arches below. Pectoral base with oblique thin brown bar ventrally. A pearly white area
above bar on middle of pectoral base extending onto pectoral fin meeting oval pearly white
area on basal quarter of middle rays. Dorsally on pectoral base a dark brown more or less
rounded blotch. Breast and isthmus white. Body with scales edged in dark brown forming a
reticulated pattern. Body with irregular mottling dorsally, forming saddles, with light central
areas. A large patch of dark brown mottling below first dorsal fin, extending ventrally onto
sides of belly, splitting into two ventral and vertical dark brown lines; followed posteriorly,
below midside, by an irregular shaped vertical bar below fifth to sixth dorsal spine, followed
posteriorly and below midside by an irregular shaped dark brown blotch slightly smaller than
eye diameter. Below anteriormost brown band on sides of belly, a subcutaneous black dash
along upper part of vertebral column; a second subcutaneous black dash below anterior part
of first dorsal fin, followed posteriorly by a similar dash below posterior end of second dorsal
fin and a similar dash just behind middle of caudal peduncle; brown dashes interspersed with
white dashes. End of caudal peduncle with a triangular vertically elongate, dark brown mark
extending onto base of caudal fin. Above caudal peduncle triangular mark a vertically elongate
dark brown spot at upper base of caudal fin. Subcutaneous black spots along anal base and
ventrally on caudal peduncle as described for preserved specimens. Fins translucent. First
dorsal fin with 2–3 curved rows of small dark brown spots. Second dorsal fin with similar
rows of spots on basal half of fin and often with white spots above brown spots, particularly
anteriorly on fin. In some individuals, dorsal fins with light greenish-brown margins and darker
greenish-brown bands between dark spots. Caudal fin with larger grey to light brown spots
forming 6 wavy bands. Pelvic and anal fins without brown pigment; pelvic fin and distal half
of pectoral fin with some scattered small white spots; pectoral fin with large pearly-white spot
on basal quarter to third of fin.
Underwater photographs suggest that the species can probably change the intensity of the
mottling and white areas to match the substrate. In situ it is very well camouflaged (Fig. 2).
Journal of the Royal Society of New Zealand, Volume 35, 2005
ETYMOLOGY: From the Greek iota = anything very small, in reference to the size of the
REMARKS: Sexual dimorphism is not obvious in preserved material examined. Overall coloration
varies from light to dark brown. The darkest samples are from Fiordland. Some other samples
appear faded and it would be difficult to document coloration differences. Small individuals
may be very pale. Since the fish apparently can lighten and darken in life, some variation is
undoubtedly related to their habitat. Because of the slight geographical variation in coloration
and pectoral ray counts, only material from Stewart Island and Fiordland in the South Island
are included in the type series.
In specimens with eight upper segmented caudal rays, the base of the first unsegmented
ray is often thickened, but the tip is slender, with the ray appearing abnormal.
Thalasseleotris iota is generally similar to the Australian Thalasseleotris adela, but differs in
a number of features. The gill opening is more extensive than T. adela, reaching to behind the
eye (versus under the eye); its anterior nostril is much longer than the posterior nostril (versus
subequal to slightly longer in T. adela). The banding under the eye is similar in the two species,
but the bands posteriorly on the head and below the first dorsal fin are more distinctive in the
New Zealand species. Similarly, the oblique bar at the base of the pectoral fin is present in
both species, but the crescent shaped black mark dorsally on the fin is absent in T. adela. The
New Zealand species is rarely uniform in colour, as is common in large male T. adela. Also,
there is a distinct black spot posteriorly on the first dorsal fin in T. adela that is absent in T.
iota. Ventrally on the caudal peduncle between the end of the anal fin and the caudal base of
the New Zealand species there are 4–6 black spots, while T. adela usually has 3–4 spots. In
larger specimens of T. iota the anterior spots may be merged into an elongate bar; juveniles
usually have 5 spots. The two species also differ in vertebral counts. Other meristic features
are similar, although the average longitudinal scale count is lower, but the transverse count
is higher, differing in both cases by an average of one scale (P < 0.001). Thalasseleotris iota
does not show the same degree of sexual dimorphism as T. adela.
The Australian species Thalasseleotris adela shows considerable geographic variation in
pectoral ray counts and longitudinal scale series counts, which were attributed to differences
in water temperature, the lowest counts occurring in warmer areas north of 35°S (Hoese &
Larson 1987, table 1). These counts were examined in T. iota in New Zealand (Table 3), but
no significant differences were found between the mean values of fishes collected from dif-
ferent areas.
Table 3 Frequency distributions of two meristic characters of Thalasseleotris iota sp. nov.
around New Zealand.
Pectoral ray counts 17 18 19 20 Mean
Northern North Island 1 15 12 1 18.4
Central New Zealand 1 7 14 6 18.9
Southern South Island 10 22 49 8 18.6
Stewart Island 0 4 4 0 18.5
Longitudinal scale counts 22 23 24 25 Mean
Northern North Island 3 8 5 4 23.5
Central New Zealand 3 13 6 3 23.4
Southern South Island 19 31 26 3 23.2
Stewart Island 2 2 3 2 23.6
Hoese & Roberts—New pygmy sleeper from New Zealand 427
Fig. 3 Distribution of Thalasseleotris iota sp. nov., based on voucher specimens in museum
Journal of the Royal Society of New Zealand, Volume 35, 2005
MAXIMUM SIZE: Measurement of 376 specimens shows that maximum adult size attained after
preservation is 32.6 mm SL (39.4 mm total length, NMNZ P.16860, largest of 3 specimens),
which makes the pygmy sleeper one of New Zealand’s smallest vertebrate species. Based
on available data, Thalasseleotris iota is the smallest fish species known from New Zealand
coastal reefs.
HABITAT: The pygmy sleeper is found at 2–34 m depth, normally inhabiting broken rocky reefs
with holes, crevices, and small caves (Roberts et al. 2005), where it perches on rocky ledges
partly overlain by silty sand or shell debris (Fig. 2). It is frequently collected underwater sharing
caves with much larger black gobies Gobiopsis atrata (Griffin, 1933). Small size, cryptic habit,
semi-transparency, and camouflage coloration make this fish very difficult to locate underwater.
However it is readily collected with rotenone ichthyocide while scuba diving. The habitat of
T. iota in New Zealand is very similar to that of T. adela in Australia, which is restricted to
rocky reefs, often in silty turbid waters, at 1–24 m depth (Hoese & Larson 1987, p. 49).
NOTES ON COLLECTIONS AND DEPTH OF CAPTURE: Over 90 collections of Thalasseleotris iota were
made by scuba divers. One specimen was sucked up alive by slurp gun (NMNZ P.18411,
23 mm SL) at 21 m depth, the rest were hand-collected following controlled use of rotenone
ichthyocide. In general, pygmy sleepers were not found in the shallowest parts of some sample
stations (C. D. Roberts and A. L. Stewart pers. obs.). The shallowest depth collected was 2 m,
even though some specimens were taken from stations with shallower upper limits (e.g., 0–15 m
depth, the type locality, stn. F07). The greatest depth recorded was 34 m (NMNZ P.21816, one
specimen, 13 mm SL). Therefore, the total depth range currently known is 2–34 m. However,
34 m is near the maximum safe working depth for compressed air scuba and may not represent
the greatest depth inhabited by this species.
ABUNDANCE: Underwater observations on scuba suggest that the pygmy sleeper is a solitary,
possibly territorial species. Usually only single individuals are sighted alive at the mouth of a
cave or on a ledge. However, when collecting using rotenone it was common for two or three
individuals to be found on the floor of a cave, while others were seen falling down the sides,
prior to capture. Hence, where habitat is suitable, several pygmy sleepers can be present living
in a small cave or large fracture in a reef. Divers at most sample stations (over 65%) collected
two or more individuals. Out of the total number of specimens collected (see Material Examined
and Appendix 1), only about 8% were taken as a single specimen per station, while over 50%
were captured at a rate of 2–8 specimens per station, and over 40% at 11–27 specimens per
station, demonstrating local abundance. At Milford Sound, frequency of capture and abundance
of the pygmy sleeper (found at 10 out of 12 stations sampled, and 134 individuals collected
in total) helped characterise a distinct fiord fish fauna (Roberts et al. 2005).
DISTRIBUTION: Thalasseleotris iota has a wide distribution in New Zealand coastal waters, where
it is endemic, extending from the Bay of Islands, Northland, to Stewart Island in the south
(Fig. 3). The three geographical groupings seen along the north-east coast of the North Island,
around Cook Strait, and at Fiordland and Stewart Island, probably reflect sampling intensity,
and the species is likely to also occur along the east coasts of both islands where habitat is
favourable. However, its apparent absence from the west coasts, other than Fiordland, may be
real and reflect the lack of its preferred habitat of sheltered rocky reefs (Roberts et al. 2005),
although collecting effort has been low in some west coast areas. The pygmy sleeper has not
been collected at the Chatham Islands, despite sampling at 17 rotenone stations at 0–21 m
depth (Roberts et al. 1991). This is one of many endemic fishes with a widespread distribution
in New Zealand coastal waters (e.g., Paulin & Roberts 1993; Francis 1996). Its sister species
T. adela in Australia also has a widespread temperate distribution, extending from New South
Wales to Western Australia, including Tasmania.
Hoese & Roberts—New pygmy sleeper from New Zealand 429
Help with preserved specimens, fish database, and radiography was given by Andrew Stewart and Robin
McPhee (Te Papa). Figure 1 was drawn by Michelle Freeborn (Te Papa); colour photographs in Fig. 2
were kindly supplied by Kendall Clements (University of Auckland). We thank the many divers who have
assisted museum fish surveys and collected fresh specimens of this new species. Helpful comments on
the manuscript were provided by Andrew Stewart (Te Papa) and John McCosker (CAS). This work was
funded by the New Zealand Foundation for Research, Science and Technology (contract MNZX0203)
‘Biosystematics of NZ EEZ Fishes” programme (C. D. Roberts, Project Leader); fieldwork was carried
out with funding assistance from the New Zealand Lotteries Board and Department of Conservation
(S&R contracts 2282 & 2360). All are gratefully acknowledged.
Birdsong R, Murdy EO, Pezold FL 1988. A study of the vertebral column and median fin osteology in gobi-
oid fishes with comments on gobioid relationships. Bulletin of Marine Science 42(2): 174–214.
Eschmeyer WN 1997. Catalog of fishes. San Francisco, California Academy of Sciences. Vol. 1–3.
2905 p.
Francis MP 1996. Geographic distribution of marine reef fishes in the New Zealand region. New Zealand
Journal of Marine and Freshwater Research 30: 35–55.
Hoese DF 1983. Sensory papilla patterns of the cheek lateralis system in the gobiid fishes Acentrogo-
bius and Glossogobius, and their significance for the classification of gobioid fishes. Records of the
Australian Museum 35: 195–222.
Hoese DF 1994. Family Eleotrididae. In: Gomon MF, Glover JCM, Kuiter RH ed. The fishes of Australia’s
South Coast. Adelaide, State Print. Pp. 810–813.
Hoese DF, Gill A 1993. Phylogenetic relationships of eleotridid fishes (Perciformes: Gobioidei). Bulletin
of Marine Science 52(1): 415–440.
Hoese DF, Larson HK 1987. New Australian fishes. Part 11. A new genus and species of eleotridid
(Gobioidei) from southern Australia with a discussion of relationships. Memoirs of the Museum of
Victoria 48(1): 35–50.
Hubbs CL, Lagler KF 1958. Fishes of the Great Lake Region. Bloomfield Hills, Michigan, Cranbrook
Institute of Science. 213 p.
Nelson JS 1994. Fishes of the world. 3rd ed. New York, John Wiley & Sons. 600 p.
Paulin CD, Roberts CD 1993. Biogeography of New Zealand rockpool fishes. In: Battershill CN, Schiel
DR, Jones GP, Crease RG, MacDiarmid AB ed. Proceedings of the Second International Temperate
Reef Symposium. Wellington, NIWA Marine. Pp. 191–199.
Roberts CD 1993. Comparative morphology of spined scales and their phylogenetic significance in the
Teleostei. Bulletin of Marine Science 51(1): 60–113.
Roberts CD, Paulin CD, Stewart AL 1991. Coast and lagoon fishes of Chatham Island: National Museum
1991 survey. Unpublished report, National Museum of New Zealand, Wellington. Held at Hector
Library. 44 p.
Roberts CD, Stewart AL, Paulin CD, Neale D 2005. Regional diversity and biogeography of coastal
fishes on the West Coast South Island of New Zealand. Science for Conservation 250. 70 p.
Ryan P, Paulin CD 1998. Fiordland underwater, New Zealand’s hidden wilderness. Auckland, Exisle
Publishing. 192 p.
Appendix over page
Journal of the Royal Society of New Zealand, Volume 35, 2005
Appendix 1 Non-type material of Thalasseleotris iota sp. nov. examined during the present study
(n = 75 lots and 229 specimens).
AMS I.41347–001 (5), 18.8–29.8 mm SL (ex NMNZ P.35077); NMNZ P.15194 (1), 18 mm, Channel
between Henry Island and Cape Home, Oakura Bay (35°23.00′S, 174°22.20′E), 14–17 m, 23 February
1984; NMNZ P.15206 (1), 19 mm, Open side of Henry Island, Oakura Bay, Northland (35°23.40′S,
174°22.70′E), 24 m, 21 February 1984; NMNZ P.16860 (3), 27–32.6 mm, Shelter Island, Fiordland,
0–5 m, 7 March 1985; NMNZ P.16913 (1), 19 mm, Entrance to Nancy Sound, Fiordland (45°07.10′S,
167°01.70′E), stn. F25, 0–18 m, 8 March 1985; NMNZ P.16952 (2), 14–18 mm, off Seymour Island,
Doubtful Sound, 28 February 1985; NMNZ P.17002 (1), 24 mm, Baxters Reef, South Bay, Kaikoura
(42°26.00′S, 173°40.80′E), 14 m, 11 March 1985; NMNZ P.17453 (1), 26 mm, Te Reakaihau Point,
Wellington (41°21.00′S, 174°47.00′E), 6 m, 9 May 1985; NMNZ P.18122 (4), 18–22 mm, Island Bay,
Wellington, 15 m, 22 January 1986; NMNZ P.18411 (1), 23 mm, Fanal Island, Mokohinau Islands
(35°57.00′S, 175°08.00′E), 21 m, 19 February 1986; NMNZ P.19850 (2), 18–24 mm, mouth of Cunaris
Sound, Fiordland, 8–9 m, 8 May 1986; NMNZ P.19879 (1), 13 mm, NE of Chalky Island, Chalky Inlet,
Fiordland, 2–4 m, 10 May 1986; NMNZ P.21235 (1), 28 mm, Goose Bay, Kaikoura (42°29.40′S,
173°31.30′E), 0–3 m, 22 February 1987; NMNZ P.21816 (1), 13.2 mm, east of Ruamahua-nui, Alderman
Islands (36°58.00′S, 176°05.00′E), 33–34 m, 6 December 1987; NMNZ P.23426 (1), 21 mm, Matakaoa,
East Cape (37°34.00′S, 178°20.00′E), 17 m, 28 June 1988; NMNZ P.23436 (1), 19 mm, Matakaoa, East
Cape (37°34.00′S, 178°20.00′E), 15 m, 26 June 1988; NMNZ P.23986 (1), 23 mm, Islet Cove, Port
Pegasus, Stewart Island, 6 m, 24 January 1989; NMNZ P.24020 (3), 19–24 mm, Port Pegasus, Stewart
Island, 11 m, 6 February 1989; NMNZ P.24596 (3), 14–21 mm, Te Horo Island, Manuhakapakapa,
D’Urville Island, Marlborough Sounds (40°53.80′S, 173°46.80′E), 10 m, 2 March 1990; NMNZ P.24642
(5), 15–25 mm, off Seal Rocks, D’Urville Island, Marlborough Sounds (40°48.40′S, 173°48.10′E),
13–15 m, 28 February 1990; NMNZ P.24653 (6), 24–27 mm, inner Port Hardy, D’Urville Island,
Marlborough Sounds (40°48.00′S, 173°54.40′E), 10 m, 23 February 1990; NMNZ P.24670 (6), 14–
22 mm, Kaitangata, Port Gore, Marlborough Sounds (41°02.50′S, 174°15.00′E), 12 m, 3 March 1990;
NMNZ P.24688 (2), 22–25 mm, The Nugget, Patuki, D’Urville Island, Marlborough Sounds (40°44.10′S,
173°57.40′E), 6 m, 24 February 1990; NMNZ P.25662 (2), 19 mm, Tawaroa Point, Titirangi Bay,
Marlborough Sounds (41°00.00′S, 174°06.00′E), 6 m, 19 March 1990; NMNZ P.25677 (2), 21–26 mm,
Worser Bay, Wellington (41°19.00′S, 174°50.00′E), 4 m, 1 December 1989; NMNZ P.25691 (2), 21–
24 mm, east entry point, Forsyth Bay, Marlborough Sounds (40°59.00′S, 174°02.00′E), 12 m, 21 March
1990; NMNZ P.26426 (2), 19–20 mm, “Passage Rock” between southern Bare Island and mainland,
Hawke Bay (39°50.30′S, 177°00.90′E), stn. B01, 15 m, 13 December 1990; NMNZ P.27691 (1), 24 mm,
Tremulous Island, Steamer Pass, Port Pegasus, Stewart Island (47°11.70′S, 167°40.20′E), stn. S12, 0–
10 m, 11 March 1992; NMNZ P.27708 (2), 19–22 mm, Pinchgut Bay, Native Island, Paterson Inlet,
Stewart Island (46°55.10′S, 168°09.60′E), stn. S06, 2–9 m, 6 March 1992; NMNZ P.27721 (1), 24 mm,
Curious Rock Reef, Port Adventure, Stewart Island (47°04.60′S, 168°10.50′E), stn. S08, 0–8 m, 8 March
1992; NMNZ P.27762 (4), 18–23 mm, “Point 8” Point, North Arm, Port Pegasus, Stewart Island
(47°10.20′S, 167°41.10′E), stn. S11, 13–18 m, 11 March 1992; NMNZ P.28050 (11), 17–24 mm,
Matatuahu Point, Tawharanui Peninsula, Hauraki Gulf (36°23.00′S, 174°49.00′E), 0–5 m, 8 April 1992;
NMNZ P.28140 (3), 17–21 mm, “Boat Cove”, Upokongaruru, west of Lottin Point, East Cape
(37°33.00′S, 178°08.00′E), stn. E07, 0–2 m, 1 May 1992; NMNZ P.28228 (2), 15–23 mm, west side of
Orete Point, Te Kaha (37°36.00′S, 177°54.00′E), stn. E10, 11–12 m, 3 May 1992; NMNZ P.28274 (1),
19 mm, bay west of Lottin Point, Waiaka Bay, East Cape (37°32.90′S, 178°08.70′E), stn. E05, 15–20 m,
1 May 1992; NMNZ P.28333 (17), 14–26 mm, south-east of Cape Runaway, East Cape (37°32.90′S,
178°00.30′E), stn. E01, 8–12 m, 24 April 1992; NMNZ P.28392 (1), 15 mm, south of Tohora Pirau,
Lottin Point (37°32.70′S, 178°10.00′E), stn. E09, 7–10 m, 2 May 1992; NMNZ P.28505 (1), 27 mm,
“Boat Cove”, Upokongaruru, west of Lottin Point, East Cape (37°33.00′S, 178°08.00′E), stn. E07, 0–2 m,
1 May 1992; NMNZ P.29812 (7), 14–25 mm, Waihau Bay, Bay of Plenty (37°36.80′S, 177°54.60′E),
stn. E31, 4–6 m, 27 January 1993; NMNZ P.29977 (2), 15–18 mm, Waipiro Bay, East Cape (38°00.00′S,
178°23.10′E), stn. E27, 21 m, 25 January 1993; NMNZ P.30055 (2), 17 mm, The Gables, East Cape
(38°31.80′S, 178°17.90′E), stn. E22, 14–17 m, 22 January 1993; NMNZ P.30073 (1), 19 mm, Tatapouri,
Gisborne (38°39.20′S, 178°09.90′E), stn. E23, 6–8 m, 22 January 1993; NMNZ P.30125 (6), 13–28 mm,
eastern Whanarua Bay, Bay of Plenty (37°40.50′S, 177°47.40′E), stn. E33, 8 m, 28 January 1993; NMNZ
P.30247 (1), 23 mm, Astronomers Point, Pickersgill Cove, Dusky Sound, Fiordland (45°47.78′S,
166°35.10′E), stn. F22, 2–18 m, 27 March 1993; NMNZ P.30258 (1), 25 mm, peninsula opposite Oak
Island, Wet Jacket Arm, Fiordland (45°39.00′S, 166°51.90′E), stn. F25, 0–33 m, 28 March 1993; NMNZ
P.30285 (2), 24–28 mm, “Little Cove”, Acheron Passage, Fiordland (45°39.88′S, 166°44.78′E), stn. F10,
0–10 m, 21 March 1993; NMNZ P.30316 (1), 20 mm, “38 Fathom Bay”, Isthmus Sound, Preservation
Inlet, Fiordland (46°02.95′S, 166°43.80′E), stn. F17, 2–18 m, 25 March 1993; NMNZ P.30347 (2),
21–23 mm, “37 Fathom Point”, opposite Otago Point, Dagg Sound, Fiordland (45°24.95′S, 166°51.30′E),
Hoese & Roberts—New pygmy sleeper from New Zealand 431
stn. F26, 0–25 m, 29 March 1993; NMNZ P.30359 (2), 23 mm, “39 Fathom Point”, Breaksea Sound,
Fiordland (45°33.75′S, 166°52.50′E), stn. F6, 6–15 m, 20 March 1993; NMNZ P.30386 (1), 32 mm,
Sunday Cove, Vancouver Arm, Fiordland (45°35.95′S, 166°45.15′E), stn. F05, 3–14 m, 19 March 1993;
NMNZ P.30434 (1), 13 mm, “55 Fathom Point”, Cunaris Sound, Chalky Inlet, Fiordland (45°58.40′S,
166°42.53′E), stn. F15, 2–31 m, 24 March 1993; NMNZ P.30449 (3), 18–27 mm, “23 Fathom Point”,
west of Fanny Bay, Dusky Sound, Fiordland (45°45.80′S, 166°49.85′E), stn. F13, 1–28 m, 23 March
1993; NMNZ P.30495 (2), 24–25 mm, “92 Fathom Point”, Bowen Channel, Dusky Sound, Fiordland
(45°43.73′S, 166°56.65′E), stn. F12, 2–20 m, 22 March 1993; NMNZ P.32179 (1), 22 mm, Paget Point,
Emelius Arm, Charles Sound, Fiordland (45°05.63′S, 167°08.45′E), stn. F35, 7–20 m, 25 March 1995;
NMNZ P.32232 (1), 18 mm, 3 nautical miles from entrance to Nancy Sound, Fiordland (45°08.63′S,
167°03.93′E), stn. F39, 8–23 m, 27 March 1995; NMNZ P.32271 (1), 25 mm, 200 m south-east of
Gardner Head, Bradshaw Sound, Fiordland (45°16.70′S, 167°08.61′E), stn. F42, 3–15 m, 28 March
1995; NMNZ P.32328 (2), 22–27 mm, Milford Sound (44°37.36′S, 167°52.00′E), stn. F12, 15–20 m,
14 November 1994; NMNZ P.32332 (2), 25–27 mm, opposite Stirling Falls, Milford Sound (44°37.36′S,
167°52.00′E), stn. F08, 10–15 m, 19 November 1994; NMNZ P.33292 (1), 14 mm, bay north of
Tahiririmongo Point, Kapiti Island (40°53.33′S, 174°52.31′E), stn. K05, 3–10 m, 7 March 1996; NMNZ
P.34140 (6), 18–23 mm, Edwardson Sound (45°54.91′S, 166°40.26′E), stn. F47, 15 m, 23 November
1996; NMNZ P.35013 (1), 24 mm, point west of Sinbad Bay, Milford Sound, Fiordland (44°39.38′S,
167°53.91′E), stn F51, 12–18 m, 30 March 1998; NMNZ P.35020 (6), 22–27 mm, Cabbage Tree Point,
Milford Sound, Fiordland (44°36.38′S, 167°49.98′E), stn. F52, 11–20 m, 31 March 1998; NMNZ P.35044
(1), 24 mm, west face of Greenstone Point, Milford Sound, Fiordland (44°35.35′S, 167°47.98′E), stn.
F56, 12–23 m, 2 April 1998; NMNZ P.35052 (3), 21–28 mm, Pater Point, Milford Sound, Fiordland
(44°38.16′S, 167°52.61′E), stn. F57, 10–15 m, 3 April 1998; NMNZ P.35077 (17), 19–32 mm, Punga
Cove, Milford Sound, Fiordland (44°36.28′S, 167°51.26′E), stn. F60, 7–14 m, 4 April 1998; NMNZ
P.35100 (10), 20–28 mm, south of Stirling Point, Milford Sound, Fiordland (44°37.45′S, 167°52.35′E),
stn. F62, 4–17 m, 5 April 1998; NMNZ P.35546 (1), 20 mm, east corner Whale Island, Bay of Plenty
(37°51.43′S, 176°59.38′E), stn. E35, 12–17 m, 1 June 1998; NMNZ P.35618 (1), 28 mm, Volkner Rocks,
Bay of Plenty (37°28.51′S, 177°08.11′E), stn. E38, 7–13 m, 4 June 1998; NMNZ P.35637 (7), 19–25 mm,
east side White Island, Bay of Plenty (37°31.15′S, 177°11.66′E), stn. E39, 7–13 m, 4 June 1998; NMNZ
P.35658 (13), 19–28 mm, Northeast Bay, White Island, Bay of Plenty (37°38.58′S, 177°11.48′E), stn.
E40, 7–13 m, 5 June 1998; NMNZ P.36262 (1), 22 mm, north side of Smoothwater Bay, South Westland
(43°57.98′S, 168°35.46′E), stn. H03, 4–6 m, 8 February 1999; NMNZ P.36486 (2), 25–28 mm, outer
Stafford Rock, South Westland (43°59.41′S, 168°31.78′E), stn. H11, 11–15 m, 12 February 1999; NMNZ
P.36657 (3), 17–20 mm, Nursery Cove, White Island, Bay of Plenty (37°31.41′S, 177°10.36′E), stn.
E43, 5–16 m, 23 April 99; NMNZ P.36676 (5), 15–21 mm, off Homestead Point, White Island, Bay of
Plenty (37°31.76′S, 177°10.68′E), stn. E44, 8–14 m, 23 April 1999; NMNZ P.36689 (5), 17–25 mm,
south-west corner, White Island, Bay of Plenty (37°31.25′S, 177°10.18′E), stn. E45, 11–18 m, 24 April
... Pygmy sleeper Thalasseleotris iota (family Eleotridae) Widespread at 20 stations, both coastal and offshore (Appendices 1-3). These specimens have contributed to a formal description of this new species (Hoese & Roberts 2005). ...
Full-text available
A fish survey was carried out along 300 km of coast in the East Cape Region (ECR), North Island, between Whakatane (Bay of Plenty) and Gisborne (Poverty Bay) New Zealand during 1992-1999. Sampling, using mainly rotenone ichthyocide and handspear, was carried out at 0-21 m depth by teams of 4-7 divers who spent c. 200 hrs underwater collecting fish specimens and recorded identification and habitat data. A total of 148 fish species (114 genera; 65 families) were recorded in inventory lists supported by voucher specimens. Of these, 107 shallow reef fish species were analysed to compare diversity and composition in four ECR subareas: East Coast, coastal Bay of Plenty (BOP), inshore BOP islands, and offshore BOP islands. No significant difference was found between the East Coast and the coastal BOP stations, and between the BOP inshore and offshore island stations; but there was a significant difference between all coastal stations (East Coast and BOP) and all island stations (inshore and offshore) using ANOVA and Tukey post boc comparisons. Complementary results were obtained with Jaccard's, pair-wise comparison between composition of reef fish faunas, which indicated that offshore BOP islands were the least similar to coastal BOP and the East Coast (46% species shared respectively); and the East Coast and coastal BOP were the most similar (79% species shared). Most reef fishes in the ECR (78%) were widespread New Zealand species; and there was a moderate northern component (20%) and small southern component (2%). There were progressive changes in composition moving from west to east and from offshore to inshore, with a reciprocal decrease in northern species and increase in widespread, species, and a small increase in southern species. Northern species reached their limit of distribution variously at White Island, coastal BOP, East Cape, and East Coast; southern species reached their limits at East Cape. All these results provide strong support for a wide biogeographic transition throughout the ECR, and refute a narrow biogeographic boundary.
In contrast to the tropics, relatively few marine gobioid fishes live in the temperate regions of Australia and New Zealand and neighbouring islands. The marine fauna of this region is mostly confined within the Pacific Central-South Subtropical water mass, with New Zealand also being influenced by the southern Subtropical Convergence (Commonwealth of Australia 2006). In Australia, subtropical and temperate waters influence about half the continent’s fauna, with considerable overlap and intermixing of tropical and temperate faunas on the west and east coasts (Wilson and Allen, 1987). For example, the seasonal southward-flowing Leeuwin Current of Western Australia brings tropical taxa to the southern coast of Western Australia, as exemplified by the coral reefs and associated gobioid fish fauna of the Houtman Abrolhos Archipelago at 29°S (Hutchins, 1994, 1997).
Full-text available
Thalasseleotrididae n. fam. is erected to include two marine genera, Thalasseleotris Hoese & Larson from temperate Australia and New Zealand, and Grahamichthys Whitley from New Zealand. Both had been previously classified in the family Eleotrididae. The Thalasseleotrididae is demonstrably monophyletic on the basis of a single synapomorphy: membrane connecting the hyoid arch to ceratobranchial 1 broad, extending most of the length of ceratobranchial 1 (= first gill slit restricted or closed). The family represents the sister group of a newly diagnosed Gobiidae on the basis of five synapomorphies: interhyal with cup-shaped lateral structure for articulation with preopercle; laterally directed posterior process on the posterior ceratohyal supporting the interhyal; pharyngobranchial 4 absent; dorsal postcleithrum absent; urohyal without ventral shelf. The Gobiidae is defined by three synapomorphies: five branchiostegal rays; expanded and mediallyplaced ventral process on ceratobranchial 5; dorsal hemitrich of pelvic-fin rays with complex proximal head. This study represents a contribution to our ongoing clarification of the family Eleotrididae, which has served historically as a repository for genera not classified among the more derived gobioids (= Gobiidae as defined here).
Full-text available
The organization and morphology of spined scales are described from a broad-based survey of body scales of teleost fishes using scanning electron microscopy and light microscopy. Three general types of spined scale are recognized (1) crenate: simple marginal indentations and projections, (2) spinoid: spines continuous with the main body of the scale, and (3) ctenoid: spines separate from the main body of the scale, in two common configurations of transforming or peripheral ctenoid and a rare configuration of whole ctenoid. Crenate scales occur widely in the Elopocephala; spinoid scales occur widely in the Euteleostei; peripheral ctenoid scales have a restricted distribution in the Euteleostei, occurring probably independently in the Ostariophysi, Paracanthopterygii, and Percomorpha; transforming ctenoid scales are a unique specialized form of spined scale, and are a synapomorphic character diagnosing the Percomorpha; whole ctenoid scales are known from only two percomorph genera. The greatest diversity of spined scales is found in the Ostariophysi and the Percomorpha. Spined scales show great evolutionary plasticity, and it is suggested that changes in ontogenetic trajectory, as well as the evolution of new characters, contribute to the diversity of spinal structures observed.
Full-text available
Several features that have been used to classify gobioid fishes are discussed, Polarity information is provided, when evidence is available, to provide better information for further studies on gobioid relationships, Odontobutidae new family is described to include three genera (Micropercops, Odontobutis and Perccottus) that had been previously assigned to the Eleotrididae. Four synapomorphies associated with the pectoral girdle, caudal skeleton, dorsal fin, and scale morphology corroborate monophyly of a group consisting of all gobioids except odontobutids and rhyacichthyids. The remaining eleotridids are assigned to the gobiid subfamilies Butinae and Eleotridinae. Specializations in the jaw musculature and caudal skeleton are presented to define the subfamily Eleotridinae. Patterns of papilla and pterygiophore insertion are congruent within the odontobutids and butines, but show little congruence within the Eleotridinae, and we conclude that while these characters may be of some value in investigating relationships within families and subfamilies, they are uninformative of interfamilial relationships within the Gobioidei.
Full-text available
Several osteological characters of the axial skeleton are surveyed in a broad assemblage of gobioid fishes comprising approximately 200 valid genera and over 500 valid species. The characters include: pattern of interdigitation of the spinous dorsal-fin pterygiophores with the neural spines of the vertebrae; vertebral number and distribution between precaudal and caudal vertebrae; number of epural bones; and number of anal fin pterygiophores anterior to the first haemal spine. All of these features show considerable stability at the generic level and appear useful in the characterization of groups. We have been able to place the large majority (about 95%) of the species examined by us into one of 32 groups on the basis of these characters. The groups are uneven in both size and position in the taxonomic hierarchy ranging from some comprising a single genus to one containing over 50 currently recognized genera. The nature of several of the characters renders their phylogenetic polarization impossible to defend, consequently 16 of the 32 groups are only phenetically united. All of the groups are proposed without formal designation as working hypotheses in need of corroboration or refutation. The geographic distribution of the groups and their constituent genera has been determined in 10 major marine faunal regions and these data are presented.
A fish survey was carried out along 500 km of the West Coast South Island (WCSI) between Milford Sound (Fiordland) and Gentle Annie Point (Buller) during 1998-2000. Sampling at depths of 0-25 m, using mainly rotenone ichthyocide and spear, was conducted by teams of 4-6 divers who spent c. 150 hours under water collecting specimens and recorded habitat data. Intertidal rockpools and estuarine/freshwater habitats were sampled for c. 70 hours. A total of 101 marine, estuarine, and freshwater fish species in 72 genera representing 45 families were recorded, with station inventory lists supported by c. 3000 voucher specimens. Greatest fish species diversity was on subtidal rocky reefs. Species composition differed with major changes in habitat: sheltered deep reefs in the fiord; semi-exposed deep reefs at Jackson Head; exposed shallow reefs along Westland and Buller coasts. The WCSI reef fish fauna is a typical assemblage of New Zealand widespread species, but contains some distinctive elements uncommon elsewhere. Three species new to science and four rare species were collected: (1) Milford Sound (Fiordland): orange rockfish Acanthoclinus ?n.sp., pygmy sleeper Thalasseleotris n.sp. (also at Jackson Head), fiord brotula Fiordichthys slartibartfasti Paulin and eyespot clingfish Modicus tangaroa Hardy; (2) Jackson Head and Murphy's Beach (South Westland): obscure triplefin Gilloblennius abditus Hardy; (3) Mokonui and Fourteen Mile (Westland): clingfish Gastroscyphus n.sp.; and (4) Seal Island, Punakaiki (Buller): marbled brotula Bidenichthys consobrinus (Hector). On the basis of physical and biological characteristics (including fishes, invertebrates and algae), three marine regions are recognised: Fiord; Fiordland open coast-South Westland; Westland-Buller. Results support the concept of a biogeographic transition zone between the Fiordland open coast and the Westland-Buller regions.