Species of the Macrognathus aculeatus group in Myanmar with remarks on M. caudiocellatus (Teleostei: Synbranchiformes: Mastacembelidae)

Abstract

The species of the Macrognathus aculeatus group from Myanmar are reviewed and three new species are described. Macrognathus dorsiocellatus, new species, previously identified as M. aral, is distinguished by rostral tooth plates 19-23, dorsal fin spines 14-22, 7-11 small ocelli along soft dorsal fin, with ventrally open, incomplete white rim; M. obscurus, new species, is distinguished by rostral tooth plates 8-10, dorsal fin spines 20-22, ocelli along dorsal fin developed as small irregularly arranged dark spots or absent, M. pavo, new species, is distinguished by the presence of only 4-6 dorsal-fin spines, only 6-8 rostral tooth plates and details of the colour pattern. In addition, M. lineatomaculatus, new species, is described from India and Nepal and it is distinguished by rostral tooth plates 15-17, dorsal fin spines 19-22, and large black blotches along dorsal fin. The systematic position of M. caudiocel-latus is reviewed and the species is retransferred to Mastacembelus.

Full text

Verlag Dr. Friedrich Pfeil
ISSN 0936-9902
Ichthyological Exploration
of Freshwaters
Volume 20
Number 4
An international journal for fi eld-orientated ichthyology

Ichthyological Exploration of Freshwaters
An international journal for fi eld-orientated ichthyology
Volume 20 • Number 4 • December 2009
pages 289-384, 47 fi gs., 18 tabs.
Managing Editor
Maurice Kottelat, Route de la Baroche 12, Case postale 57
CH– 2952 Cornol, Switzerland
Tel. + 41 - 32 - 4 62 31 75 / Fax + 41 - 32 - 4 62 22 59 / E-mail mkottelat@dplanet.ch
Editorial board
Pier Giorgio Bianco, Dipartimento di Zoologia, Università, Napoli, Italy
Ralf Britz, Department of Zoology, The Natural History Museum, London, United Kingdom
Sven O. Kullander, Naturhistoriska Riksmuseet, Stockholm, Sweden
Helen K. Larson, Museum and Art Gallery of the Northern Territory, Darwin, Australia
Lukas Rüber, Department of Zoology, The Natural History Museum, London, United Kingdom
Ivan Sazima, Museu de Zoologia, Unicamp, Campinas, Brazil
Paul H. Skelton, South African Institute for Aquatic Biodiversity, Grahamstown, South Africa
Heok Hui Tan, Raffl es Museum of Biodiversity Research, National University of Singapore, Singapore
Ichthyological Exploration of Freshwaters is published quarterly
Subscriptions should be addressed to the Publisher:
Verlag Dr. Friedrich Pfeil, Wolfratshauser Str. 27, D– 81379 München, Germany
PERSONAL SUBSCRIPTION : EURO 100 per Year/volume - 4 issues (includes surface mail shipping)
INSTITUTIONAL SUBSCRIPTION : EURO 180 per Year/volume - 4 issues (includes surface mail shipping)
Manuscripts should be addressed to the Managing Editor:
Maurice Kottelat, Route de la Baroche 12, Case postale 57, CH –2952 Cornol, Switzerland
CIP-Titelaufnahme der Deutschen Bibliothek
Ichthyological exploration of freshwaters : an international
journal for fi eld-orientated ichthyology. – München : Pfeil.
Erscheint jährl. viermal. – Aufnahme nach Vol. 1, No. 1 (1990)
ISSN 0936-9902
Vol. 1, No. 1 (1990) –
Copyright © 2009 by Verlag Dr. Friedrich Pfeil, München, Germany
All rights reserved.
No part of this publication may be reproduced, stored in a retrieval system, or transmitted in any form or by any
means, electronic, mechanical, photocopying or otherwise, without the prior permission of the copyright owner.
Applications for such permission, with a statement of the purpose and extent of the reproduction, should be
addressed to the Publisher, Verlag Dr. Friedrich Pfeil, Wolfratshauser Str. 27, D-81379 München, Germany.
Printed by Advantage Printpool, Gilching
ISSN 0936-9902
Printed in the European Union
Verlag Dr. Friedrich Pfeil, Wolfratshauser Str. 27, D-81379 München, Germany
Tel. + 49 - (0)89 - 74 28 27 0 – Fax + 49 - (0)89 - 72 42 77 2 – E-mail: info@pfeil-verlag.de – www.pfeil-verlag.de

295
Ichthyol. Explor. Freshwaters, Vol. 20, No. 4
Ichthyol. Explor. Freshwaters, Vol. 20, No. 4, pp. 295-308, 7 figs., 3 tabs., December 2009
© 2009 by Verlag Dr. Friedrich Pfeil, München, Germany – ISSN 0936-9902
Species of the Macrognathus aculeatus group in Myanmar
with remarks on M. caudiocellatus
(Teleostei: Synbranchiformes: Mastacembelidae)
Ralf Britz*
The species of the Macrognathus aculeatus group from Myanmar are reviewed and three new species are described.
Macrognathus dorsiocellatus, new species, previously identified as M. aral, is distinguished by rostral tooth plates
19-23, dorsal fin spines 14-22, 7-11 small ocelli along soft dorsal fin, with ventrally open, incomplete white rim;
M. obscurus, new species, is distinguished by rostral tooth plates 8-10, dorsal fin spines 20-22, ocelli along dorsal
fin developed as small irregularly arranged dark spots or absent, M. pavo, new species, is distinguished by the
presence of only 4-6 dorsal-fin spines, only 6-8 rostral tooth plates and details of the colour pattern. In addition,
M. lineatomaculatus, new species, is described from India and Nepal and it is distinguished by rostral tooth plates
15-17, dorsal fin spines 19-22, and large black blotches along dorsal fin. The systematic position of M. caudiocel-
latus is reviewed and the species is retransferred to Mastacembelus.
* Department of Zoology, The Natural History Museum, London, SW75BD, United Kingdom.
E-mail: r.britz@nhm.ac.uk
Introduction
Spiny eels of the Macrognathus aculeatus species
group form a monophyletic unit characterized
uniquely among mastacembelids by a series of
paired tooth plates that support the long rostral
tentacle. Sufi (1956), who published the last revi-
sion of all Asian mastacembelids, recognized only
a single species in this group: M. aculeatus. This
species was thought to have an extremely wide
distribution (“India, Pakistan, Burma, Ceylon,
Siam, Malay Peninsula, Java, Sumatra, Borneo
and Moluccas” according to Sufi, 1956: 103) and
to be also highly variable morphologically. Sufi’s
(1956) view was commonly accepted until Roberts
(1980: 388-389) revised this group of mastacem-
belids and recognized three species, M. aral from
“all of the larger river systems of the Indian sub-
continent, including Sri Lanka and Burma”,
M. siamensis from “the Mekong basin of Thailand
and Kampuchea (and presumably in Laos), as
well as in the Chao Phraya and other rivers of
Thailand southwards to the northern end of the
Malay peninsula”, and M. aculeatus from “the
southern half of the Malay peninsula; several of
the principal rivers of Sumatra; the Kapuas River
of Borneo; and northern Java as far east as the
Brantas River”.
During three recent visits to the Swedish
Museum of Natural History in Stockholm, I had
the opportunity to study their holdings of speci-
mens of the M. aculeatus group from Myanmar
and to compare them with those from the neigh-
bouring countries India and Thailand. This was

296
Britz: Macrognathus aculeatus group
complemented by additional material from other
institutions. A thorough analysis of the material
demonstrated that the species in Myanmar con-
sidered M. aral by Roberts (1980, 1986) actually
represents an undescribed species and that the
species-level diversity of this species group is
even higher. The present paper serves to describe
these new species.
Material and methods
Measurements were taken with digital callipers
to the nearest 0.1 mm. Obtaining accurate meas-
urements in mastacembelids usually proves dif-
ficult because specimens are frequently preserved
in a bent and twisted state. Standard length,
predorsal length (spinous and soft), preanal
length, head length and snout length were meas-
ured from the anterior tip of the body of the
premaxilla, thus not including the proboscis,
which is often twisted and distorted. This land-
mark can be located by bending the rostrum
ventrally and pushing the caliper towards the eye
until a firm point is found. Vertebral numbers
were counted from radiographs. Fin ray counts
for unpaired fins were also obtained from radio-
graphs, those of pectoral fins from alcohol speci-
mens under transmitted light. The count of cau-
dal-fin rays includes also the small and sometimes
tiny procurrent rays. Counting the number of
pectoral-fin rays in alcohol specimens was not
without difficulties and small fin rays at the
margins of the fin may have been overlooked.
Several specimens of the different samples that
were studied for the present paper show vertebral
elements with double or even more numerous
neural and/or haemal spines. Specimens with
this kind of teratological variation were excluded
from the analysis of meristic data. Abbreviations
used: BMNH, The Natural History Museum,
London; CAS, California Academy of Sciences,
San Francisco; NRM, Swedish Museum of Natu-
ral History, Stockholm; USNM, National Mu-
seum of Natural History, Smithsonian Institution,
Washington.
Macrognathus dorsiocellatus, new species
(Fig. 2)
Holotype. NRM 60293, 133.3 mm SL; Myanmar:
Bago Division: roadside stream about 64km on
road Taungoo–Nyaunglaybin, 18°19'5" N 96°30'
7" E; F. Fang et al., 19 Mar 1997.
Paratypes. NRM 24779, 3, 192.7-222.8 mm SL;
Myanmar: Yangon Division: Yangon River, S of
Yangon, W bank (probably c 12 miles from Yan-
gon city); 22 Jan-2 Jul 1935. – USNM 372501, 9,
90.0-125.1 mm SL; Myanmar: Yangon Division:
Thanlyin market; C. J. Ferraris, 18-26 Apr 1996.
– NRM 39701, 2, 109.1-126.2 mm SL; Myanmar:
Yangon Division: Yangon, Kyimyindine Town-
ship, Central Fish Market, 16°47'0" N 96°10'0" E;
F Fang & A Roos 14 Mar 1997. – NRM 58246, 2,
132.3-137.3 mm SL; Myanmar: Yangon Division:
Barlar Chaung, approximately 40 km (25 miles)
on the way from Yangon northeast to Bago ; Del-
ling et al., 15 Mar 2008. – NRM 58205, 8, 122.5-
162.0 mm SL; Myanmar: Yangon Division: Barlar
Chaung, approximately 40 km (25 miles) on the
way from Yangon northeast to Bago; Delling et
al., 15 Mar 2008. – NRM 39873, 134.6 mm SL;
Myanmar: Bago Division: Bago, fish market,
100°E
95°E
90°E 100°E
95°E
90°E
10°N
15°N
25°N
20°N
10°N
15°N
25°N
20°N
India
India
B
angla-
angla-
desh
desh
C
hina
hina
Laos
Laos
T
hailand
hailand
V
iet-
iet-
nam
nam
India
Bangla-
desh
China
Laos
Thailand
Viet-
nam
M. dorsiocellatus
M. obscurus
M. morehensis
M. pavo
Fig. 1. Sampling localities of species of the Macro gnathus
aculeatus group in Myanmar.

297
Ichthyol. Explor. Freshwaters, Vol. 20, No. 4
17°20'0" N 96°29'0" E; F. Fang & A. Roos, 15 Mar
1997. – NRM 39887, 146.2 mm SL; same data as
holotype. – BMNH 2009.7.3.1-2, 2, 112.5-156.0 mm
SL; Myanmar: Bago Division: Daik U township,
pool beside road next to ricefield, 17°44'06" N
96°07'22" E; 17; R. Britz et al. 19 Oct 2008. – NRM
39670, 250 mm SL; Myanmar: Bago Division:
river crossing road about 27 km from Bago to
Waw, near Waw, 17°27'43" N 96°39'9" E; F. Fang
& A. Roos, 15 Mar 1997. – USNM 372501, 5, 127.8-
174.8 mm SL; Myanmar: Bago Division: Nga Zin
Yaing creek near Taungoo; C. Ferraris et al. 25
Oct 1997. – NRM 48647,107.3 mm SL; Myanmar:
Mon State: about 20 km SE of Moulmein, Kyaik-
maraw Market near Ataran Chaung, 16°22'59" N
97°43'59" E; T. R. Roberts, 1-6 Sep 2000. – NRM
58327, 1, 157.0 mm SL; Myanmar: Mon State:
Kyaikto market; F. Fang and Thein Win, 15 Mar
2008. – USNM 378907, 4, 153.8-158.0 mm SL;
Myanmar: Mandalay Division: Mandalay mar-
kets; R. Britz & R. Roesler 20 Mar 2003. – USNM
344664, 3, 128.4-188.3 mm SL; Myanmar: Manda-
lay Division: Mandalay markets; C. J. Ferraris et
al., 9-10 Nov 1996. – NRM 12915, 2, 106.5-165.3 mm
SL; Myanmar: Mandalay Division: Mandalay;
1935. – NRM 14945, 3, 101.2-145.1 mm SL; Myan-
mar: Mandalay Division: Mandalay area, 12 dif-
ferent sites, 22°0'0" N 96°5'0" E; O. Hetzel, 1 Mar-
30 Apr 1935. – BMNH 2009.7.3.3-4, 2, 152.3-
200.7 mm SL; Myanmar: Shan State: Kalaw, Kalaw
market, reportedly from Mandalay Division:
Thazi: Minhla Lake; R. Britz & O. Crimmen 1 Nov
2008.
Diagnosis. Macrognathus dorsiocellatus is a mem-
ber of the M. aculeatus species group. It differs
from all other species of this group by the follow-
ing combination of characters: rostral tooth plates
Fig. 2. Macrognathus dorsiocellatus, NRM 60293, holotype, 133.3 mm SL; Myanmar: Bago Division, roadside stream
about 64 km on road Taungoo–Nyaunglaybin.
Fig. 3. a, Macrognathus siamensis, BMNH 1861.10.8.14, 138 mm SL; Thailand: Pechabore [Petchaburi]; b, M. aral,
NRM 40555, 169 mm SL; India: Western Bengal: Kolkata.
a
b

298
19-23, dorsal-fin spines 14-22, and 7-11 small
ocelli along soft dorsal fin, with ventrally open,
incomplete white rim.
Description. For general appearance see Figure 2.
Morphometric data are provided in Table 1 and
vertebral and dorsal-fin spine counts are pre-
sented in Tables 2 and 3.
Body elongate, oval in cross section, but
strongly laterally compressed in its caudal area.
Depth 7.1-9.3 times in SL and body width 1.9-2.8
times in its depth. Head pointed with median
fleshy rostral tentacle projecting from upper jaw;
tentacle with anterior naris at end of nasal tube
originating at its subdistal tip, rim of anterior
naris with six fimbriae; lips fleshy; jaws with
numerous small, pointed teeth; gill membrane
connected to isthmus; opercular opening large
ventrally but extending dorsally only to level of
horizontal through upper third of pectoral-fin
base. Preopercular region without any spines.
Minute scales on body, opercular area and cheek;
a narrow band extending from above eye to pos-
terior naris, and on base of rostral tentacle.
Pectoral-fin rays 20-24. Dorsal fin with 14-22
spines and 51-61 rays. Anal fin with 3 spines and
51-60 rays. Caudal-fin rays 16-20. Dorsal, caudal,
and anal fins separated. First dorsal-fin pterygi-
ophore inserted behind 13th to 20th neural spine,
pterygiophore of last dorsal fin spine inserted
behind 32nd to 36th neural spine. First anal-fin
pterygiophore inserted behind haemal spine of
32nd to 35th vertebra.
Anal-fin spines covered and concealed by
thick skin, second spine longest. Penultimate
dorsal-fin spine longest, followed by shorter last
spine, completely concealed by thick skin, difficult
to detect.
Lateral line extending from shoulder girdle
in a longitudinal line, higher on body in ab-
dominal region descending to middle of body in
caudal region and ending some distance in front
of caudal fin. Vertebrae: 32-35 + 40-46 = 72-80.
Colour in alcohol. Colour dark brown on dor-
sum, becoming lighter brown on side, whitish on
ventrum. Narrow white longitudinal line from
above eye posteriorly along body, where it marks
lateral-line canal. Area below white narrow line
commonly with darker brown band that extends
to snout.
Series of 7-11 ocelli along base of dorsal fin.
Individual ocelli with dark brown to black center
and white rim that fails to close ventrally. A few
specimens (e. g. NRM 48647, 58205) with faint
oblique anteriorly-directed bars originating from
bases of ocelli. Posterior two-thirds of soft dorsal
fin with longitudinal dark-brown striations. Soft
Table 1. Selected morphometric data for Macrognathus dorsiocellatus, M. obscurus, M. pavo and M. lineatomacula-
tus. H, Holotype; SD, standard deviation.
Britz: Macrognathus aculeatus group
M. dorsiocellatus M. obscurus
H n = 39 mean ± SD H n = 3 mean ± SD
Standard length (SL) in mm 133.3 97-250 135 99.6-135
In percent of standard length
Head length (HL) 18.4 16.6-21.5 18.7±1.3 18.1 18.1-18.9 18.4±0.4
Predorsal length of spinous dorsal fin 39.6 35.3-42.8 39.5±1.9 33.3 33.3-38.5 35.4±2.7
Predorsal length of soft dorsal fin 63.5 63.5-70.2 66.7±1.4 69.3 65.5-69.3 67.1±1.9
Preanal length 54.4 56.4-65.5 61.2±1.6 60.4 59.8-60.8 60.4±0.5
Pectoral fin length 6.4 5.9-7.9 6.9±0.6 6.7 6.4-7.0 6.7±0.3
Body depth at pectoral fin 9.6 7.2-9.8 8.3±0.7 8.8 7.2-8.8 8.1±0.8
Body depth at 1st anal-fin spine 11.9 11.2-14.1 12.3±0.7 11.9 11.9-13.7 12.9±1.0
Body width at 1st anal-fin spine 5.9 4.5-6.7 5.8±0.5 5.5 5.5-6.4 5.9±0.5
In percent of head length
Snout length 37.6 31.8-42.8 36.7±2.7 37.4 38.8-40.4 39.9±1.0
Eye diameter 11.0 9.0-14.1 11.4±1.3 12.7 12.7-14.9 13.4±1.3
Ratios
Depth/width of body 2.0 1.9-2.8 2.2±0.2 2.2 2.1-2.3 2.2±0.1
Standard length/Head lenth 5.4 4.7-6.3 5.4±0.4 5.5 5.3-5.5 5.4±0.1
Standard length/body depth at 1st anal-fin spine 8.4 7.1-9.0 8.1±0.4 8.4 7.3-8.4 7.8±0.6

299
Ichthyol. Explor. Freshwaters, Vol. 20, No. 4
anal fin uniform brownish, sometimes with white
rim. Caudal fin with series of vertical striae. Pec-
toral fin uniformly whitish with no additional
darker marks.
Distribution. Known from the Ayeyarwaddy,
lower Sittang and lower Salween basins in My-
anmar. Widespread at lower altitudes in southern
Myanmar up to the Mandalay area of the Ayeyar-
waddy basin.
Etymology. The name is derived from the Latin
words dorsum, back, and ocellatus, with ocelli,
referring to the series of ocelli along the dorsal
fin. An adjective.
Remarks. Macrognathus dorsiocellatus resembles
most closely M. siamensis (Fig. 3a), but differs from
it by more dorsal-fin ocelli (7-11 vs. 3-6), absence
of ocellus on caudal-fin base (vs. presence) and
more rostral tooth plates (19-24 vs. 11-13). It is
also readily distinguished from the other ocel-
lated Indian Macrognathus, M. aral (Fig. 3b) and
M. pentophthalmos, by a higher number of dorsal-
fin ocelli (7-11 vs. 3-6), the incomplete white rim
of the ocelli, lack of the two dark longitudinal
stripes and a striped rather than marbled caudal
fin. The other Indian/Myanmar Macrognathus
with incomplete dorsal-fin ocelli, M. morehensis
(Fig. 6), has fewer rostral tooth plates (11-16 vs.
19-24). This new species has previously been
referred to as either M. aculeatus (Day, 1878, 1889;
Vinciguerra, 1889-90; Sufi, 1956) or M. aral (Rob-
erts, 1980, 1986; Talwar & Jhingran, 1992; Vidthay-
anon et al. 2005 [note that the specimen illus-
trated in this account as M. aral from Myanmar
is not M. dorsiocellatus but most likely an Indian
specimen of the true M. aral, as evidenced by the
typical colour pattern]). Kottelat & Widjanarti
(2005) resolved the confusion surrounding the
name Macro gnathus aculeatus, once considered to
be a widespread species, through designation of
a neotype from the Brantas basin in Java.
Macrognathus aral is an Indian species with
“fluvios Tranquebarenses” (“rivers of Tranque-
bar”, the earlier transcription of Tharangambadi)
as type locality, but with a more widespread
distribution in India. With the exception of one
specimen (NRM 18350) from Tamil Nadu and the
types in the Berlin museum (ZMB 1419, ZMB
1420), all specimens of M. aral used in this study
are from northern India. They, however, corre-
spond in their meristic characters closely to the
lectotype of M. aral, as designated by Paepke
(1999) and the specimens that Pethiyagoda et al.
(2008) list as topotypic material of M. aral, with
the exception of the number of abdominal verte-
brae. Pethiyagoda et al. (2008: 43) noted that
“These specimens (ZMB 1420, ZMB 1419) differ
significantly from the Sri Lankan examples of
Macrognathus examined by us: they (ZMB 1420,
ZMB 1419) have 18 dorsal spines . . . and 36-37
precaudal vertebrae . . ., both of which are consist-
ent with our series of 18 topotypes of M. aral”.
The radiographs I received of the Bloch types
(courtesy of P. Bartsch, ZMB) show 35 abdominal
vertebrae in ZMB 1419 and 34 abdominal verte-
brae in both specimens of ZMB 1420, and not
36-37 as mentioned by Pethiyagoda et al. (2008).
In addition all specimens of M. aral that were
investigated during the present study have 32-35
abdominal vertebrae. That means that the ab-
dominal vertebral count of M. aral falls within the
range given for M. pentophthalmos by Pethiya-
goda et al. (2008) and is not a distinguishing
character of the two species. The count of 18 spines
for the three ZMB specimens by Pethiyagoda et
al. (2008) and Paepke’s (1999) count of 19 spines
for ZMB 1419 are also erroneous. ZMB 1419 has
20 and the two specimens of ZMB 1420 each have
17 dorsal fin spines.
M. pavo M. lineatomaculatus
H n = 9 mean ± SD H n = 20 mean ± SD
133.3 95.6-157.5 144.6 114.5-184.3
17.0 16.6-18.3 17.5±0.6 18.7 17.9-20.4 19.3±0.7
52.0 49.5-54.0 52.3±1.2 34.2 33.3-39.9 35.9±1.9
60.3 58.4-61.1 59.8±0.8 64.6 62.3-67.8 65.0±1.1
54.4 54.4-58.7 56.8±1.3 60.4 58.7-63.0 61.0±1.2
5.9 4.8-6.5 5.6±0.6 6.9 5.9-6.9 6.4±0.3
6.9 6.4-7.1 6.8±0.2 7.3 7.3-8.5 7.6±0.4
8.4 9.6-11.2 10.3±0.6 10.7 9.6-12.3 11.0±0.6
4.4 4.7-6.4 5.6±0.6 4.1 4.0-5.5 4.8±0.4
36.0 34.1-37.9 36.1±1.9 36.5 36.5-43.9 40.5±2.2
9.3 9.3-11.6 10.3±0.7 13.3 10.5-14.3 11.8±1.1
2.1 1.7-2.2 1.9±0.2 2.6 2.1-2.7 2.3±0.2
5.9 5.5-6.0 5.7±0.2 5.3 4.9-5.6 5.2±0.2
9.7 9.0-10.4 9.7±0.5 9.3 8.5-10.4 9.1±0.5

300
Macrognathus obscurus, new species
(Fig. 4)
Holotype. NRM 36330, 135 mm SL; Myanmar:
Kachin State: Hpa Lap Chaung just south of
Yuzana Myaing village (8 km left from Myitkina
Myitzon rd km 11), 25°31'57'' N 97°22'19'' E; F.
Fang & A Roos,1 April 1997.
Paratypes. USNM 372502, 2, 99.6-125.6 mm SL;
Myanmar: Kachin State: Hpa Lap stream, of Nam
Chin Chaung, of Nan Kwe Chaung, NW of My-
itkina, 25°31'36'' N 97°22'45'' E; C. Ferraris et al.,
6 Nov 1997.
Diagnosis. Macrognathus obscurus is a member
of the M. aculeatus species group. It differs from
all other species of this group by the following
combination of characters: rostral tooth plates
8-10, dorsal-fin spines 20-22, and ocelli along
dorsal fin developed as small irregularly arranged
dark spots or absent.
Description. For general appearance see Figure 4.
Morphometric data are provided in Table 1 and
vertebral and dorsal-fin spine counts are pre-
sented in Tables 2 and 3.
Body elongate, oval in cross section, but
strongly laterally compressed in its caudal area.
Depth 7.3-8.4 times in SL and body width 2.1-2.3
times in its depth. Head pointed with median
fleshy rostral tentacle projecting from upper jaw;
tentacle with anterior naris at end of nasal tube
originating at subdistal tip of rostral tentacle, rim
of anterior naris with six fimbriae; lips fleshy;
jaws with numerous small, pointed teeth; gill
membrane connected to isthmus; opercular open-
ing large ventrally but extending dorsally only
to level of horizontal through upper third of
pectoral-fin base. Preopercular region without
any spines. Minute scales on body, opercular area,
and cheek; a narrow band extending from above
eye to posterior naris, and on base of rostral
tentacle.
Pectoral-fin rays 22-24. Dorsal fin with 20-22
spines and 50-55 rays. Anal fin with 3 spines and
49-51 rays. Caudal-fin rays 16-18. Dorsal, caudal,
and anal fins separated. First dorsal-fin pterygio-
phore inserted behind 12th to13th neural spine,
pterygiophore of last dorsal fin spine inserted
behind 33rd to 35th neural spine. First anal-fin
pterygiophore inserted behind haemal spine of
32nd vertebra.
Anal-fin spines covered and concealed by
thick skin, second spine longest. Penultimate
dorsal-fin spine longest, followed by shorter last
spine, completely concealed by thick skin, difficult
to detect.
Lateral line extending from shoulder girdle
in a longitudinal line, higher on body in ab-
dominal region descending to middle of body in
caudal region and ending some distance in front
of caudal fin. Vertebrae: 32 + 40-42 = 72-74.
Colour in alcohol. Head and body with narrow
white longitudinal line from above eye posteri-
orly along body, where it marks lateral-line canal,
on body broken up by darker spots into a series
of shorter lines with round, spot-like, white marks.
Below white narrow line area on head and body
darker brown with white speckling. Colour dark
brown on dorsum, followed by lighter brown
band in dorsal third of body from above eye
extending to caudal area. Ventrum pale whitish.
A series of 8 or 9 very small, sometimes indistinct
black spots along base of dorsal fin, not expressed
as ocelli with white rim as in M. dorsiocellatus. Soft
dorsal fin whitish with brown oblique longitudi-
nal striations. Soft anal fin with brownish base,
whitish rim and a few striae on posterior fifth.
Caudal fin with a series of vertical somewhat
irregular striae. Pectoral fin uniformly whitish
with a few brownish marks at the base.
Fig. 4. Macrognathus obscurus, NRM 36330, holotype, 135 mm SL; Myanmar: Kachin State: Hpa Lap Chaung.
Britz: Macrognathus aculeatus group

301
Ichthyol. Explor. Freshwaters, Vol. 20, No. 4
Distribution. Known from Hpa Lap stream near
Myitkina in northern Myanmar, the type locality
of Danio kyathit (Fang, 1998).
Etymology. From the Latin adjective obscurus,
dark, inconspicuous, alluding to the lack of the
prominent dorsal-fin ocelli or spots, otherwise
typical of the M. aculeatus species group.
Remarks. Macrognathus obscurus resembles
M. morehensis (Fig. 6) in number of rostral plates,
but differs from it in number of dorsal-fin spines
(20-22 vs. 11-16). It also superficially resembles
M. dorsiocellatus (Fig. 2), but is easily distinguished
by the absence of the dorsal-fin ocelli and by the
narrow longitudinal white line that marks the
lateral-line canal being interrupted by darker
spots.
Macrognathus pavo, new species
(Fig. 5)
Holotype. BMNH 2009.7.3.5, 133.3 mm SL; My-
anmar: Rakhine State: Kyeintali Chaung; Han
Thet Aung, 23 Nov 2008.
Paratypes. BMNH 2009.7.3.6, 1, 122.5 mm SL;
same data as holotype. – BMNH 2009.11.3.1-7, 7,
95.6-157.5 mm SL; Myanmar: Rakhine State:
Kyeintali Chaung; Han Thet Aung, 25 Jan 2009.
– BMNH 2009.12.14.1-6, 6, 103.4-147.2 mm SL;
Myanmar: Rakhine State: Kyeintali Chaung; Han
Thet Aung, Apr 2008. – BMNH 2009.12.14.7-8, 2,
89.8-98.7 mm SL; Myanmar Rakhine State: Taung
Chaung: Kyeintali Chaung drainage, 18°00'37'' N
94°32'0'' E; Zaw Win & Han Thet Aung 2 Dec
2009.
Diagnosis. Macrognathus pavo is a member of the
M. aculeatus species group. It differs from all
other Asian mastacembelids by the presence of
only 4-6 dorsal-fin spines (vs. 11-40). It differs
further from all other species of the M. aculeatus
group by the presence of only 6-8 rostral tooth
plates (vs. from 9 in M. meklongensis to 55 in
M. tapirus) and details of the colour pattern.
Description. For general appearance see Figure 5.
Morphometric data are provided in Table 1 and
vertebral and dorsal-fin spine counts are pre-
sented in Tables 2 and 3.
Body elongate, oval in cross section, but
strongly laterally compressed in its caudal area.
Depth 9.0-10.4 times in SL and body width 1.7-2.2
times in its depth. Head pointed with median
fleshy rostral tentacle projecting from upper jaw;
tentacle with anterior naris at end of nasal tube
originating at its subdistal tip, rim of anterior
naris with six fimbriae; lips fleshy; jaws with
numerous small, pointed teeth; gill membrane
connected to isthmus; opercular opening large
ventrally but extending dorsally only to level of
horizontal through upper third of pectoral-fin
base. Preopercular region without any spines.
Minute scales on body, opercular area, and cheek;
a narrow band extending from above eye to pos-
terior naris, and on base of rostral tentacle.
Pectoral-fin rays 21-22. Dorsal fin with 4-6
spines and 53-55 rays. Anal fin with 3 spines and
47-51 rays. Caudal-fin rays 16-18. Dorsal, caudal,
and anal fins separated, but contiguous. Pterygi-
ophore of first dorsal fin spine inserted behind
25th to 27th neural spine, pterygiophore of last
dorsal-fin spine inserted behind 31th to 32th neural
spine. First anal-fin pterygiophore inserted behind
haemal spine of 31nd to 32th vertebra.
Anal-fin spines covered and concealed by
thick skin, second spine longest. Penultimate
dorsal-fin spine longest, followed by shorter last
spine, completely concealed by thick skin, difficult
to detect.
Fig. 5. Macrognathus pavo, BMNH 2009.7.3.5, holotype, 133.3 mm SL; Myanmar: Rakhine State: Kyeintali
Chaung.

302
Lateral line extending from shoulder girdle
in a longitudinal line, higher on body in ab-
dominal region descending to middle of body in
caudal region and ending some distance in front
of caudal fin. Vertebrae: 31-32 + 41-44 = 73-76.
Colour in alcohol. Head and body with black
stripe extending from in front of eye to horizon-
tal through middle of anal fin base. Bordered
dorsally by narrow, whitish to lighter brown
longitudinal line from in front of eye to anterior
third of caudal area. Upper half of side of body
with light brown longitudinal band from eye to
base of caudal fin. Middle of dorsum light brown
bordered laterally by a paired dark brown to black
line with irregular edge, extended posteriorly as
a series of small black spots along dorsal-fin base.
Ventral half of body lighter brown and set off
distinctly against whitish ventrum. A series of 16
to 20 prominent small black spots along base of
dorsal fin, and a series of 6 to 8 tiny black spots
along posterior third of anal-fin base. Soft dorsal
and anal fins whitish with brown oblique lines
of pigment following fin rays and tiny spots in
dorsal third of fin. Caudal fin also with brown
pigment lines following fin rays and peppering
of tiny black spots. Pectoral fin uniformly whitish
without any marks.
Distribution. Known from Kyeintali Chaung in
the Rakhine Yoma mountain range.
Etymology. From the Latin word “pavo”, pea-
fowl, referring to the numerous black spots along
the dorsal and anal fins but also on the body of
this species. A noun in apposition.
Remarks. Macrognathus pavo is one of the most
unusual in the M. aculeatus group because of its
colour pattern, the lowest number of rostral
toothplates, and the possession of the lowest
number of dorsal-fin spines in any Asian mast-
acembelid. This is correlated with a greater pre-
dorsal length of 49.5-54.0 % SL and a more pos-
terior insertion of the first spine-bearing dorsal-fin
pterygiophore (behind neural arch of 24-27th
vertebra vs. 12th-22nd ) than in the other Myanmar
species of the M. aculeatus group. The occurrence
of M. pavo in the Rakhine Yoma emphasizes again
the high level of endemism of the fish fauna of
this area, as pointed out previously (Britz, 2007;
Kullander & Fang 2004, 2009).
Table 2. Vertebral numbers of species of Macrognathus aculeatus group. Counts for holotypes and lectotype in
bold.
abdominal
vertebrae caudal vertebrae total vertebrae
31323334 35 37 38 39 40 41 42 43 44 45 46 69 70 71 72 73 74 75 76 77 78 79 80
M. dorsiocellatus 32417 1 2 7 1913 31 1 1 5 1812 6211
M. obscurus 32121
M. pavo 451531
251
M. morehensis 22 2
M. lineatomaculatus Jorai River India 322
21 41
Chittawan River Nepal 3 18 2 1 13 8 1 1 1210
M. aral ZMB 1420 lectotype 11 1
ZMB 1419 paralectoypes 2 2 2
Western Bengal NRM40555, 40566, 40598 4353 21453 1 233411
Tamil Nadu NRM 18350 1 1 1
Nepal SU 52965 2 2 1 2 1 2 1 1
Pakistan CAS 76437 1 1 1
M. pentophthalmos MCZ 166239 11 1
M. siamensis BMNH 1861.10.8.14 11 1
NRM 24786 5 1 2 1 1 1 2 1 1
NRM 24787 1 1 1
NRM 51047 1 1 1
M. meklongensis NRM 24780 holotype 11 1
NRM 24781 1 1 1 1 1 1
Britz: Macrognathus aculeatus group

303
Ichthyol. Explor. Freshwaters, Vol. 20, No. 4
Macrognathus morehensis
Remarks. This species was described by Arun-
kumar & Tombi Singh (2000) from specimens
collected in the Yu River in Manipur, which drains
into the Chindwin in Myanmar. The authors
provided only a very rough illustration of the
holotype, which can barely be recognized as a
mastacembelid and on which no details can be
discerned. They list the combination of the fol-
lowing characters as diagnostic: “(i) 11 to 16
dorsal fin spines, (ii) 20-25 black broad transverse
bars on the body, (iii) 8 to 11 rostral tooth-plates,
(iv) 12 to 14 black spots that are imperfect ocelli
at the base of dorsal fin rays, (v) 10 to13 distinct
black spots at the base of anal fin rays, (vi) 6 black
oval spots at the base of dorsal spines, (vii) 5 to 7
oblique striations of black dots arranged in paral-
lel longitudinal rows at the dorsal and anal fin
rays, (viii) 7 to 10 black lines of striations formed
by dots at the caudal fin, (ix) a single ocellus at
the base of caudal fin and (x) 76 vertebrae.”
A photo of the holotype of M. morehensis was
provided by W. Vishwanath and is reproduced
herein (Fig. 6a). He also informed me (e-mail, Sept
2009) that only the holotype but none of the
8 paratypes listed by Arunkumar & Tombi Singh
(2000) have actually been deposited at MUMF;
their whereabouts are unknown. The photo of
the holotype of M. morehensis shows a depig-
mented fish with a series of indistinct ocelli along
the dorsal-fin base and several faint bars along
the posterior part of the body. The caudal fin has
at least 10 more or less vertical striae and the
posterior half of the dorsal fin shows a series of
mottled striations. The anal and pectoral fins are
whitish and do not seem to have any darker pig-
ments. That means of the seven colour pattern
characters listed in the diagnosis only four can
be observed in the photograph of the holotype.
Among the mastacembelids from Myanmar
used for the present study were two specimens
(USNM 372595, NRM 59031) from the Chindwin
basin that fit the description of M. morehensis and
have most of the diagnostic characters listed by
Arunkumar & Tombi Singh (2000). The two well-
preserved specimens identified as M. morehensis
in this study have 19-20 wide brown oblique bars
along the body, a whitish belly and a narrow
white line marking the lateral line canal (Fig. 6b).
There is also a series of 9-10 incompletely enclosed
dorsal fin ocelli, the base of which is continuous
with the lateral bars. A series of up to 8 incomplete
ocelli extends along the anal-fin base. Dorsal,
Table 3. Dorsal-fin spine numbers of species of Macrognathus aculeatus group. Counts for holotypes and lectotype
in bold.
4 5 6 11 14151617181920212223
M. dorsiocellatus 112
10 178631
M. obscurus 21
M. pavo 135
M. morehensis 2
M. lineatomaculatus Jorai River India 14 1
Chittawan River Nepal 24865
M. aral ZMB 1420 lectotype 1
ZMB 1419 paralectotypes 2
NRM 40555, 40566, 40598 Western Bengal 153312
NRM 18350 Tamil Nadu 1
SU 52965 Nepal 121
CAS 76437, 24255 Pakistan 11
M. pentophthalmos MCZ 166239 1
M. siamensis BMNH 1861.10.8.14 1
NRM 24786 122
NRM 24787 1
NRM 51047 1
M. meklongensis NRM 24780 holotype 1
NRM 24781 paratypes 2

304
caudal and anal fins show a series of dark brown
striations, more or less horizontally arranged on
dorsal and anal fins and vertically on the caudal
fin. Both specimens have 11 dorsal-fin spines and
10 and 11 rostral toothplates, respectively. The
colour pattern of M. morehensis somewhat resem-
ble M. dorsiocellatus in having a series of black
spots with an incomplete white rim at the dorsal
fin base, but M. morehensis differs clearly from the
latter in having fewer dorsal-fin spines and rostral
toothplates.
Macrognathus lineatomaculatus, new species
(Fig. 7)
Holotype. BMNH 2009.7.3.7, 144.6 mm SL; India:
West Bengal: Jalpaiguri district: Jorai River a
tributary of the Sankosh at Laskarpara, Barobisha
town outskirts (26°28'52" N 89°49'30" E); M. Das,
3 April 2008.
Paratypes. BMNH 2009.7.3.8-12, 5, 132.0-
184.3 mm SL; same data as holotype. – CAS 50320,
116.8 mm SL; Nepal: Rapti River, Chitawan Val-
ley at Kasa Darvar; T. R. Roberts, 27 Apr 1975.
– CAS 50331, 3, 87.4-142.8 mm SL; Nepal, Rapti
River basin, Chitawan Valley, Dudara River,
tributary to Rapti from Churia Hills; T. R. Roberts,
25 Apr 1975. – CAS 42592, 28, 104.1-160 mm SL;
Nepal, Upper Rapti River, 4 miles north of Hi-
taura; T. R. Roberts, 02 May 1975.
Diagnosis. Macrognathus lineatomaculatus is a
member of the M. aculeatus group. It differs from
all other species of that group by the following
combination of characters: rostral tooth plates
15-17, dorsal-fin spines 19-22, and large (eye size)
black blotches along dorsal fin. It is distinguished
from its other Indian congeners of the M. aculea-
tus group as follows: from M. aral, M. pentophthal-
mos and M. morehensis by presence of dorsal fin
blotches (vs. white rimmed dorsal fin ocelli), from
M. aral and M. pentophthalmos by fewer rostral
tooth plates (15-17 vs. 19-27), and in addition from
M. pentophthalmos and M. morehensis by more
dorsal-fin spines (19-22 vs. 11-16) and from M. mo-
rehensis by more rostral tooth plates (15-17 vs. 8-11).
Description. For general appearance see Figure 7.
Morphometric data are provided in Table 1 and
vertebral and dorsal-fin spine counts are pre-
sented in Tables 2 and 3.
Body elongate, oval in cross section, but
strongly laterally compressed in its caudal area.
Depth 8.9-9.5 times in SL and body width 2.3-2.7
times in its depth. Head pointed with median
fleshy rostral tentacle projecting from upper jaw;
tentacle with anterior naris at end of nasal tube
originating at its subdistal tip, rim of anterior
naris with six fimbriae; lips fleshy; jaws with
numerous small, pointed teeth; gill membrane
connected to isthmus; opercular opening large
ventrally but extending dorsally only to level of
horizontal through upper third of pectoral fin
Fig. 6. Macrognathus morehensis. a, MUMF 203/8A, holotype, 155 mm TL; India: Manipur: Maklang (photograph
courtesy of W. Vishwanath); b, NRM 59031, 149 mm SL; Myanmar: Chin State: road Kalaymyo-Tamu.
a
b
Britz: Macrognathus aculeatus group

305
Ichthyol. Explor. Freshwaters, Vol. 20, No. 4
Fig. 7. Macrognathus lineatomaculatus. a, BMNH 2009.7.3.7, holotype, 144.6 mm SL; India: West Bengal: Sankosh
River drainage; b, CAS 42592, 143.7 mm SL; Nepal: Upper Rapti River.
a
b
base. Preopercular region without any spines.
Minute scales on body, opercular area, cheek,
and; a narrow band extending from above eye to
posterior naris, and on base of rostral tentacle.
Pectoral-fin rays 22-23. Dorsal fin with 19-22
spines and 50-57 rays. Anal fin with 3 spines and
47-54 rays. Caudal-fin rays 16-17. Dorsal, caudal,
and anal fins separated. First dorsal-fin pterygi-
ophore inserted behind 12th to 14th neural spine,
pterygiophore of last dorsal fin spine inserted
behind 34th to 35th neural spine. First anal-fin
pterygiophore inserted behind haemal spine of
33rd to 34th vertebra.
Anal-fin spines covered and concealed by
thick skin, second spine longest. Penultimate
dorsal-fin spine longest, followed by shorter last
spine, completely concealed by thick skin, difficult
to detect.
Lateral line extending from shoulder girdle
in longitudinal line, higher on body in abdominal
region descending to middle of body in caudal
region and ending at some distance in front of
caudal fin. Vertebrae: 33-34 + 40-43 = 74-76.
Colour in alcohol. Head and body with a light,
whitish-beige band extending from posterior naris
in front of eye to caudal fin. Bordered dorsally
by dark brown longitudinal line from back of
head to caudal fin. This line frequently broken
up into eye size blotches beneath spinous dorsal
fin. Ventral to whitish-beige band with a dark
brown stripe from snout to caudal fin, sometimes
expressed as series of dark blotches slightly be-
neath lateral line canal in abdominal area and at
level of lateral line canal in caudal area. More
ventrally side brown and more or less well sepa-
rated from light brown to whitish ventrum.
Series of up to 10 circular black blotches along
base of soft dorsal fin, sometimes only expressed
in posterior part of fin. Proximal area of soft
dorsal fin whitish, with series of darker small
circular or irregular marks, sometimes forming
second series of smaller blotches above basal
series of blotches. Rim of dorsal and anal fins
whitish, but fins with dusky subdistal band.
Caudal fin dusky sometimes with a few darker
roundish marks. Pectoral fin whitish with a
darker brown mark at base and some brown
chromatophores along fin rays.
Distribution. Macrognathus lineatomaculatus is
known from the Jorai River in Western Bengal,
India, a tributary of the Sankosh, and the Rapti
River basin in Nepal.
Etymology. A combination of the Latin words
lineatus, striped, and maculatus, with blotches,
referring to its row of blotches along the dorsal-fin
base and sometimes along the middle of the body.
An adjective.
Remarks. Macrognathus lineatomaculatus has been
figured as M. aral by Roberts (1980: fig. 2b). It
differs from M. aral, however, in colour pattern
as well as number of rostral tooth plates (15-17
vs. 20-24).

306
Discussion
Among mastacembelids, species of the Macro-
gnathus aculeatus group are easily recognized by
their comparatively longer rostral tentacle that is
supported by a varying number of pairs of tooth
plates most likely representing a fragmented
premaxilla (Sufi, 1956). The first mention of this
unique character was probably by Schneider
(1801: 478) when he described Rhynchobdella ori-
entalis as having “labium superius productum in
rostrum infra excavatum, plicis transversis
plurimis, carina longitudinalis divisis” (upper lip
produced into a rostrum, hollowed out below,
with numerous transverse folds, divided by a
longitudinal keel). Sufi (1956) studied the details
of the structure of this unusual rostrum and
clarified that it includes a series of tooth plates
along the ventral face of the proboscis in front of
the maxilla. He concluded that this feature dis-
tinguishes Macrognathus from Mastacembelus.
Day’s (1878) and Sufi’s (1956) conclusion of a
monotypic Macrognathus was followed until
Roberts (1980) revised the genus and recognized
three species, M. aculeatus, M. aral, and M. siamen-
sis. Roberts (1986) later described a fourth species
from this species group, M. meklongensis from the
Mae Khlong River basin in Kanchanaburi, Thai-
land. A fifth species, M. morehensis (Arunkumar
& Tombi Singh, 2000) was added from the Yu
River system in Manipur, a tributary of the Chind-
win in Myanmar. Kottelat & Widjanarti (2005)
demonstrated that Fowler’s Mastacembelus pauci-
spinis from Trang in Thailand is actually a sixth
valid species of the Macrognathus aculeatus group,
but because the name was preoccupied they cre-
ated the replacement name M. tapirus. And fi-
nally, Pethiyagoda et al. (2008) concluded re-
cently that the spiny eel in Sri Lanka formerly
considered M. aral, is a separate species, M. pent-
ophthalmos, which is now apparently extinct. There
have thus been seven recognized species in the
M. aculeatus group to date. The present paper
brings this figure up to 11.
Travers’ concept of Macrognathus. Cuvier &
Valenciennes (1832) distinguished two genera
among the “Rhynchobdelles”, Rhynchobdella « qui
ont le museau concave et strié en dessous [the
pairs of rostral tooth plates described in Sufi,
1956], et les trois nageoires verticales séparées »
and Mastacembelus « dont le museau est simple-
ment conique, sans stries ni concavité, et dont les
nageoires verticales sont plus ou moins complète-
ment réunies ». This distinction has been adopted
by various authors until Travers (1984a-b), how-
ever, with Macrognathus La Cepède, 1800 replac-
ing Rhynchobdella Bloch, in Schneider, 1801 as the
valid name for this genus. Travers (1984b) hy-
pothesized that eight species included until then
in the genus Mastacembelus are actually more
closely related to the species of the genus Macro-
gnathus than to other Mastacembelus species. Ac-
cordingly, he chose to add these eight species to
Macrognathus thus expanding the genus to the
effect that the series of tooth plates was not diag-
nostic of the genus anymore but only of the spe-
cies of the M. aculeatus group.
Travers (1984b) listed eight synapomorphies
(61-68) for his expanded Macrognathus, in which
he included also M. caudiocellatus. As material
studied for the species M. caudiocellatus, Travers
(1984a: 7) listed only the three syntypes and
noted that he used radiographs to study them.
No specimens of M. caudiocellatus were thus avail-
able to him for dissection and, of the eight synapo-
morphies for his expanded genus Macrognathus,
only one or two are potentially observable in
radiographs of M. caudiocellatus (64. Dorsal surface
of frontal slopes ventrolaterally; 65. Elongate
narrow neural and haemal spines, associated with
a deep body) and a third (68. 6 slender digitiform
fimbriae around rim of each anterior nostril as-
sociated with relatively long rostral appendage)
is potentially observable in alcohol specimens. I
checked the three type specimens of M. caudiocel-
latus (BMNH 18) and found the snout in all of
them bent or distorted and not preserved well
enough to be able to evaluate the structure of the
anterior nares. Newly collected alcohol specimens
of M. caudiocellatus (NRM 28481, 58401), however,
fail to confirm the presence of 6 fimbriae (finger-
like skin projections) around the rim of the ante-
rior nostril as suggested by Travers (1984b), but
rather the typical condition of most Mastacembe-
lus, with just two finger-like fimbriae and two
larger fimbrules (= skin flaps). The inclusion of
M. caudiocellatus in Macrognathus is thus not sup-
ported even in the wider sense of Travers (1984b)
and the species is placed in Mastacembelus.
Comparative material. Macrognathus aral: 25 speci-
mens, 113-193 mm SL. ZMB 1419, lectotype; India:
fluvios Tranquebarenses. – ZMB 1420, 2, paralectotypes
according to Paepke (1999). – NRM 40598, 6; India:
Western Bengal, Kolkata, New Market, Jadu Babris Fish
Britz: Macrognathus aculeatus group

307
Ichthyol. Explor. Freshwaters, Vol. 20, No. 4
Market. – NRM 40566, 5; India: Assam: Brahmaputra
River drainage: Dibrugarh Market. – NRM 40555, 4;
India: Western Bengal: Kolkata: New Market, Jadu
Babris Fish Market. – NRM 18350, 1; India: Tamil Nadu:
Vagai River, Madurai. – SU 52965, 4; Nepal: Biratnagar
and vicinity; purchased at bazaar. – CAS 24255, 1; Pa-
kistan: Sindh province: Indus River basin: Indus River,
325 miles north of Karachi (i. e. 5 miles north of Sukkur).
– CAS 76437, 1; Pakistan.
M. meklongensis: 3 specimens: 93.5-175.7 mm SL.
NRM 24780, 1, holotype; NRM 24781, 2, paratypes;
Thailand: Khanchanaburi: Meklong River basin: Kwae
Noi River, Kao Lam Dam area, Tong Pha Phum.
M. pentophthalmos: MCZ 166239, 1, 195 mm SL;
Ceylon.
M. siamensis: 8 specimens, 88.5-193 mm SL. BMNH
1861.10.8.14, holotype; Thailand: Pachebore. – NRM
24786, 5; Mekong River drainage: Ubon Ratchatani
market (presumably caught in Mun River near Ubon
Ratchatani) – NRM 24787, 1; Cambodia: Song Vam Co
Tay drainage: about 100 km SE of Phnom Phenh, Svay
Rieng. – NRM 51047, 1; Viet Nam: Thin Can Tho prov-
ince, Can Tho City.
Acknowledgements
I am grateful to S. Kullander and E. Ahlander (NRM),
J. Williams (USNM) and D. Catania, (CAS) for access
to material under their care. P. Bartsch (ZMB), pro-
vided photos and radiographs of the Bloch specimens
of Macrognathus. I also thank S. Kullander, E. Ahlander,
Fang Fang, G. Fridriksson and the rest of the fish group
at the NRM for making my stays a very pleasant expe-
rience. I am deeply indebted to B. Kajrup (NRM) for
taking most of the radiographs and photographs of the
specimens, additional photographs were provided by
P. Hurst, Photo Unit, The Natural History Museum.
The collecting trips to Myanmar, during which speci-
mens of the new species were obtained, were finan-
cially sponsored by the Swedish Natural Science Re-
search Council (R-RA 04568-316) in 1998 and the col-
lection enhancement grant scheme of The Natural
History Museum, London, in 2007 and 2008. I am grate-
ful to U Khin Ko Lay, Director General, Department of
Fisheries, Ministry of Livestock Breeding & Fisheries,
Yangon, for permits. I thank U Tin Win, Hein Aquarium,
Yangon, for the gift of specimens of various Myanmar
species among which was one of the new Macrognathus.
Andrew Rao, Malabar Tropicals, Calcutta, and Keith
Lambert, Wildwoods, Crews Hill, made the specimens
of M. lineatomaculatus available. I am also grateful to
S. Kullander, NRM, for help and company in the field.
Most of the material for this manuscript was studied
and the first drafts written during visits to the NRM in
Sept. 2006, Jan. 2007, and Dec. 2008 funded by the Eu-
ropean Union through the SYNTHESYS program (grant
SE-TAF 1950 and SE-TAF 4895).
Literature cited
Arunkumar, L. & H. Tombi Singh. 2000. Spiny eels of
the genus Macrognathus from Manipur, with descrip-
tion of a new species. Journal of the Bombay Natu-
ral History Society, 97: 117-122.
Bloch, M. E. 1786. Naturgeschichte der ausländischen
Fische. Zweiter Theil. Berlin, vi + 160 pp., pls. 145-
180.
Britz, R. 2007. Channa ornatipinnis and C. pulchra, two
new species of dwarf snakeheads from Myanmar
(Teleostei: Channidae). Ichthyological Exploration
of Freshwaters, 18: 225-344.
Cuvier, G. & A. Valenciennes. 1832. Histoire naturelle
des poissons. Tome huitième. Levrault, Paris, (1831
[1832]), xix + 509 pp., pls. 209-245.
Day, F. 1876. The fishes of India; being a natural his-
tory of the fishes known to inhabit the seas and
fresh waters of India, Burma, and Ceylon. Part 2.
Bernard Quaritch, London, pp. 169-368, pls. 41-
78.
— 1889. Fishes. Vol. II. The fauna of British India, in-
cluding Ceylon and Burma. Taylor & Francis,
London, xiv + 509 pp.
Kottelat M. & E. Widjanarti. 2005. The fishes of the
Danau Sentarum National Park and the Kapuas
Lakes area, Kalimantan Barat, Indonesia. Raffles
Bulletin of Zoology, Supplement, 13: 139-173.
Kullander, S. O. & F. Fang. 2004. Seven new species of
Garra (Cyprinidae: Cyprininae) from the Rakhine
Yoma, southern Myanmar. Ichthyological Explora-
tion of Freshwaters, 15: 257-278.
— 2009. Danio aesculapii, a new species of danio from
south-western Myanmar (Teleostei: Cyprinidae).
Zootaxa, 2164: 41-48.
Paepke, H.-J. 1999. Bloch’s fish collection in the Mu-
seum für Naturkunde der Humboldt Universität
zu Berlin – An illustrated catalog and historical
account. Gantner, Ruggell, 216 pp., 32 pls.
Pethiyagoda, R., A. Silva, K. Maduwage & l. Kariya-
wasam. 2008. The Sri Lankan spiny eel, Macro-
gnathus pentophthalmos (Teleostei: Mastacembeli-
dae), and its enigmatic decline. Zootaxa, 1931: 37-48.
Roberts, T. R. 1980. A revision of the Asian mastacem-
belid fish genus Macrognathus. Copeia, 1980: 385-
391.
— 1986. Systematic review of the Mastacembelidae or
spiny eels of Burma and Thailand, with description
of two new species of Macrognathus. Japanese Jour-
nal of Ichthyology, 33: 95-109.
Schneider, J. G. 1801. M. E. Blochii Systema Ichthyolo-
giae iconibus cx illustratum, post obitum auctoris
opus inchoatum absoluit, correxit, interpolavit Jo.
Gottlob Schneider. Sandriano, Berolini, ix + 584 pp,
110 pls.
Sufi, S. M. 1956. Revision of the Oriental fishes of the
family Mastacembelidae. Bulletin of the Raffles
Museum, 27: 93-146.

308
Talwar, P. K. & A. G. Jhingran. 1992. Inland fishes of
India and adjacent countries, Vol. 2. Balkema, Rot-
terdam, xix + pp. 543-1158.
Travers, R. A. 1984a. A review of the Mastacembeloidei,
a suborder of synbranchiform teleost fishes. Part I:
Anatomical descriptions. Bulletin of the British
Museum (Natural History), Zoology Series, 46:
1-133.
— 1984b. A review of the Mastacembeloidei, a subor-
der of synbranchiform teleost fishes. Part II: Phylo-
genetic analysis. Bulletin of the British Museum
(Natural History), Zoology Series, 47: 83-150.
Vidthayanon, C., A. Termvidchakorn & Myint Pe. 2005.
Inland fishes of Myanmar. Southeast Asian Fisher-
ies Development Center (SEAFDEC), Bangkok,
160 pp.
Vinciguerra, D. 1890. Viaggio di Leonardo Fea in Bir-
mania e regioni vicine. XXIV Pesci. Annali del
Museo Civico di Storia Naturale di Genova, Ser. 2,
9: 129-360.
Received 31 July 2009
Revised 3 November 2009
Accepted 12 November 2009
Britz: Macrognathus aculeatus group

Warning
Prospective authors should read carefully the following instructions and
follow them when submitting a manuscript. Doing so signifi cantly hasten
publication and save money and efforts. Manuscript which do not satisfy
the instructions below may be rejected at the Editor’s discretion and will
not be returned.
Types of papers
1) Major Articles and Notes. These manuscripts are submitted to two
referees for evaluation.
2) Rapid communications. These articles will be repor ts on exciting new
results or discoveries within the scope of the journal. They are evaluated by
the Editor and the Editorial Board and are either accepted or rejected.
Submission of manuscripts
The original manuscript and two copies should be sent to the Editor,
Maurice Kottelat, Route de la Baroche 12, Case postale 57, CH- 2952
Cornol, Switzerland. Do not send a disk or original of fi gures at this stage.
A letter of transmittal is requested, giving:
1) the name, postal and email addresses and telephone of the cor-
responding author;
2) the names, postal and email adresses of up to four persons outside
the authors’ institutions who are qualifi ed to review the paper;
3) a statement that the material has not been published and is not consid-
ered for publication elsewhere and that it will not be submitted elsewhere
unless it is rejected or withdrawn. In submitting a manuscript, the author(s)
accept transfer of the copyright to the Publisher.
Co-authors, corresponding author
Authors are those who have played a signifi cant role in designing and
conducting the research and in writing the manuscript. Individuals who
have only collected data, provided material or fi nancial support, or re-
viewed the manuscript should be listed in acknowledgments. Honorary
authorship is not accepted.
Co-authors should designate a single corresponding author to whom
correspondence and proofs will be sent. All correspondence regarding
the paper should go through the corresponding author. Correspondence
will not be sent to other co-authors and correspondence from other co-
authors regarding the manuscript will neither be answered nor taken into
consideration.
Format
Text. All manuscripts are subject to editorial revision before fi nal accept-
ance for publication. The manuscript should be printed, double spaced
and with at least 2.5 cm right, left and bottom margins. Pages must be
numbered. Nothing in the manuscript should be underlined. Titles with
numerical series designations are not permitted. Titles should be brief,
fewer than 20 words and should indicate clearly the fi eld of study and
the group of fi shes investigated. All abbreviations should be explained in
the Method section (or fi gure caption when appropriate) or a reference to
published explanations should be provided; exceptions are very common
abbreviations, such as mm, km, kg, sec, min, yr, vs, SL. Footnotes are not
permitted. Do not end a line of text with a hyphen. All measurements must
be in metric units. The fi rst page should include: title of the paper, author(s),
addresses and abstract, all left justifi ed. The text should be followed by
Material Examined (if appropriate), Acknowledgments (if any), Appendix
(if any) and Literature Cited, in that order. Keys are desirable in taxonomic
papers. They should be dichotomous and not serially indented.
Nomenclature. Names of living organisms should follow the appropriate
and current International Codes of Nomenclature. Formal names of genera
and species should be written in italics. Names of authors and publica-
tion dates of scientifi c names should appear only when nomenclatural
problems are involved.
Language. Manuscripts should be written in English, French or Ger man,
but English is strongly encouraged. All papers must have a concise but
informative abstract in English. Manuscripts in French or German must
have an extensive summary in English. In taxonomic papers, the abstract
must include at least clear diagnosis of the new taxa. A second abstract,
provided by the author(s), in the language of the country or area concerned
by the text is acceptable. A maximum of two abstracts is permitted.
Acknowledgments. Identify individuals by name(s) and surname. Do not
list titles, position or institution. Acknowledge individuals, not positions.
Idiosyncrasy and private jokes are not permitted.
Literature Cited. Format for Literature Cited is that of the most recent
issue. Abbreviate names of well known journals, but provide in full the
names of lesser known journals. For books, give full name of publishing
company or institution, city and country. Manuscripts in preparation or
submitted, abstracts, in-house reports and other literature not obtainable
through normal library channels cannot be cited.
Tables. Each table should be on a separate sheet, numbered sequentially
with Arabic numerals; they should have concise but self-explanatory head-
ings. Do not insert frames, vertical rules, dotted lines or footnotes. The
location of fi rst citation of each table should be clearly indicated.
Figures. Do not submit originals for the review; send originals with the
revised manuscript. All maps, graphs, char ts, drawings and photographs
are regarded as fi gures and are to be numbered consecutively and in the
sequence of their fi rst citation in the text. When several charts or photographs
are grouped as one fi gure, they must be trimmed and spaced as intended
for fi nal reproduction. Each part of such a group fi gure should be lettered
with a lower case block letter in the lower left corner. Scale should be
indicated on the fi gure by a scale bar. Figures should never be mounted
on hard cardboard. Do not mark or write on the back of the fi gures.
All illustrations should be designed to fi t a width of 68 or 140 mm and
a depth no greater than 200 mm. Prints should be of high contrast and
glossy. Lettering should be large enough to be easily seen when reduced
onto a journal column (68 mm). Legends for fi gures must be typewritten on
a separate sheet. Identify each illustration with the author’s name, fi gure
number and direction of top margin.
Colour illustrations should preferably be slides. Slides should be framed
and properly labelled. For the review, they should be submitted as three
sets of prints, pasted on sheets of paper, or as a series of color photo-
copies. Prints to be used for publications should never be mounted and
never labelled on the back.
The decision to print in colour or in black and white any fi gure originally
submitted in colour remains with the editor and publisher. This decision
will be based on scientifi c justifi cation, quality of the original, layout and
other editorial, fi nancial and production contraints. By submitting colour
originals, the authors know and accept that they may be published in black
and white.
Even if photographs or line drawings are processed with graphics pro-
grams, original slides, negatives or drawings must always be submitted.
Software texts. The whole text should be left justifi ed. Do not right-justify.
Do not underline. Never use “L” or “I” for the numeral 1 or “O” / “o” for “0”.
Do not hyphenate any word. Do not format the text. Text should be on 3.5"
disks, readable on IBM-compatibles with MS-DOS. Do not send a disk,
when submitting the manuscript. Revised manuscript must be accompanied
by a disk satisfying above criteria, or will not be accepted.
Review
Each manuscript will be sent to two reviewers for confi dential evaluation.
When justifi ed, the reviewer’s comments will be forwarded to the author.
When submitting a revised manuscript, authors should briefl y indicate
which comments have been incorporated and the reasons for disregarding
any suggestion regarded as unacceptable. Remember that if a reviewer
had questions or did not understand you, other readers may make the
same experience and the answers should be in the manuscript and not
in a letter to the editor. Changes in style, format and layout requested by
the Editor are non-negotiable and non-observance will usually result in
rejection of the manuscript.
Revised manuscripts received more than 6 months after the reviewers’
comments had been sent will not be considered or will be treated as new
submissions.
Proofs, Reprints and Page Charges
One set of proofs will be sent to the corresponding author; they should be
checked and returned to the Editor within one week. Proofs not received
within this delay may be corrected by the Editor, at the author’s risks.
Authors may be charged for any changes other than printer’s error. Reprint
orders must be forwarded with the corrected proofs on the form supplied
by the editor. The corresponding author is responsible for contacting the
co-authors and forwarding their reprint orders.
The authors will obtain 20 reprints free of charge, additional reprints
may be ordered at cost. There will be no page charges and no charges
for justifi ed colour illustrations.
INSTRUCTIONS TO CONTRIBUTORS
Ichthyological Exploration of Freshwaters
An international journal for fi eld-orientated ichthyology

Articles appearing in this journal are indexed in:
AQUATIC SCIENCES and FISHERIES ABSTRACTS
BIOLIS - BIOLOGISCHE LITERATUR INFORMATION SENCKENBERG
CAMBRIDGE SCIENTIFIC ABSTRACTS
CURRENT CONTENTS/AGRICULTURE, BIOLOGY & ENVIRONMENTAL SCIENCES and SCIE
FISHLIT
ZOOLOGICAL RECORD
C O N T E N T S
Moritz, Timo: Nannocharax signifer, a new species of fish (Characiformes: Distichodontidae)
from the Ouémé River basin, Benin......................................................................................... 289
Britz, Ralf: Species of the Macrognathus aculeatus group in Myanmar with remarks on
M. caudiocellatus (Teleostei: Synbranchiformes: Mastacembelidae) .................................... 295
Freyhof, Jörg and Müfit Özulug: Pseudophoxinus evliyae, a new species of spring minnow
from Western Anatolia with remarks on the distribution of P. ninae and the systematic
position of P. fahirae (Teleostei: Cyprinidae) .......................................................................... 309
Bohlen, Jörg and Vendula Šlechtová: Schistura udomritthiruji, a new loach from southern
Thailand (Cypriniformes: Nemacheilidae) ............................................................................. 319
Freyhof, Jörg and Müfit Özulug: Pseudophoxinus fahrettini, a new species of spring minnow
from Central Anatolia (Teleostei: Cyprinidae) ...................................................................... 325
Turan, Davut, Maurice Kottelat and Semih Engin: Two new species of trouts, resident and
migratory, sympatric in streams of northern Anatolia (Salmoniformes: Salmonidae) ... 333
Costa, Wilson J. E. M.: Rivulus megaroni, a new killifish from the Xingu River drainage,
southern Brazilian Amazon (Cyprinodontiformes: Rivulidae) ........................................... 365
Britz, Ralf and Maurice Kottelat: Pangio longimanus, a miniature species of eel-loach from
Central Laos (Teleostei: Cypriniformes: Cobitidae) .............................................................. 371
Harold, Antony S. and Norma J. Salcedo: Creagrutus yanatili, a new species from the Río
Urubamba drainage, southeastern Perú (Teleostei: Characidae) ........................................ 377
Ichthyological Exploration of Freshwaters
An international journal for fi eld-orientated ichthyology
Volume 20 • Number 4 • December 2009
Cover photograph:
Macrognathus dorsiocellatus (photograph by Ralf Britz)
Ralf Britz
(this volume pp. 295-308)