Content uploaded by Peter Wirtz
Author content
All content in this area was uploaded by Peter Wirtz on Sep 10, 2018
Content may be subject to copyright.
Accepted by W. Holleman: 11 Apr. 2018; published: 21 Jun. 2018
ZOOTAXA
ISSN 1175-5326 (print edition)
ISSN
1175-5334
(online edition)
Copyright © 2018 Magnolia Press
Zootaxa 4438 (2): 381
–
393
http://www.mapress.com/j/zt/
Article
381
https://doi.org/10.11646/zootaxa.4438.2.12
http://zoobank.org/urn:lsid:zoobank.org:pub:673356D1-F47F-4CB2-AEFA-3ACF3DB1A3C8
Didogobius janetarum sp. nov., a new cryptobenthic goby species from the Cape
Verde Islands (Teleostei: Gobiidae)
ULRICH K. SCHLIEWEN
1,4
, PETER WIRTZ
2
& MARCELO KOVAČIĆ
3
1
SNSB-Bavarian State Collection of Zoology (ZSM), Department of Ichthyology, Münchhausenstr. 21, D-81247 München, Germany.
E-mail: schliewen@zsm.mwn.de
2
Centro de Ciências do Mar, Universidade do Algarve, Campus de Gambelas, PT 8005-139 Faro, Portugal
3
Prirodoslovni muzej Rijeka, Lorenzov prolaz 1, HR–51000 Rijeka, Croatia
4
Corresponding author
Abstract
Didogobius janetarum sp. nov. is described from five specimens collected from small caves and rock crevices between
12 and 20 m depth off two locations of Santiago Island, Cape Verde Islands. The species differs from all currently de-
scribed congeners by the combination of the following characters: (1) 27 vertebrae, (2) second dorsal fin I + 10, (3) pos-
terior quarter of predorsal region in front of first dorsal fin origin scaled, with several rows of cycloid scales, (4) body
squamation cycloid anteriorly and ctenoid posteriorly, (5) scales in the lateral series 30–32, (6) anterior oculoscapular ca-
nal present, (7) preopercular head canal absent, (8) suborbital row 7 each a single papilla, (9) suborbital rows 2 and 4 close
to orbit, and by (10) branchiostegal membranes uniformely black below preopercle, forming a V-shaped mark. Definitions
for all used meristic counts are presented to serve as a reference for gobioid meristic studies. The genus is rediagnosed to
accommodate recently described Didogobius species.
Key words: Gobiidae, new species, eastern Atlantic, Didogobius, Chromogobius, meristics
Introduction
With almost 2200 valid species, Gobiiformes are the most species rich and yet under-explored predominantly
marine fish order worldwide, a fact that is also reflected by more than 373 new species descriptions over the course
of the last ten years (Eschmeyer & Fong 2018). Even among the comparatively well-known monophyletic Afro-
European gobies of the Gobius-lineage (sensu Agorreta et al. 2013), many species have been described in the last
decade, e.g., three out of ten species of the closely related genera Chromogobius de Buen, 1930 and Didogobius
Miller, 1966 (Schliewen & Kovačić 2008, Van Tassell & Kramer 2014). In 2010 one us (PW) collected a single
specimen of an apparently new Didogobius species at the coast of Santiago Island (Republic of Cabo Verde), which
appeared superficially similar to the recently described D. helenae Van Tassell & Kramer, 2014. Collection of four
additional specimens, including both sexes and one juvenile, in 2016 allowed assessment of variation of putatively
diagnostic characters, which were revealed to be consistent. We therefore describe the species formally. Since
neither the most recently described Didogobius species, D. helenae, nor the new species fully agree with the most
recent diagnosis of Didogobius in Schliewen & Kovačić (2008), we provide a modified diagnosis of the genus
rather than erecting another new monotypic genus for the new species presented here.
Material and methods
Specimens were captured while SCUBA diving with the use of a hand net and fixed in lab-grade 96% ethanol
directly after collection. Distance measurement methods follow detailed definitions provided in Schliewen &
Kovačić (2008) and were taken with a Mitutoyo 505-732 dial calliper (accuracy 0.01 mm) or, in order to avoid
SCHLIEWEN ET AL.
382
·
Zootaxa 4438 (2) © 2018 Magnolia Press
specimen damage, in some cases from variously enlarged digital x-rays or specimen photos with a 10.0 mm size
standard (see Discussion). Meristic counts were taken according to the following precise definitions (abbreviations
in brackets), which are given here in detail, because there is, to our knowledge, yet no compilation of recently used
meristics definitions used in goby descriptions. Scales in the lateral series (LL): counted on body from pectoral-fin
axil along lateral midline to the end of caudal peduncle with the scales on C counted separately, showed
additionally with a “+”; scales in transversal series (TR): from the anterior origin of the anal fin obliquely upwards
and rearwards to the base of D2; caudal peduncle scale count (CP) after Van Tassell & Kramer (2014), based on
Larson (2001): beginning at the first normal scale (not reduced in size or shape) on top of the caudal peduncle
immediately in front of caudal fin, and following the scale rows down and forward to the ventral edge of the
peduncle, then around and back to the original scale; predorsal scale row count (PC): scale rows counted along
mid-dorsal line counted from D1 origin to anterior end of scaled predorsal area. Fin elements terminology as
summarized in Fricke (1983): spines are unsegmented and unbranched elements (in Roman numerals), rays are
segmented elements which may or may not be branched (in Arabic numerals), and procurrent rays are short
unbranched and unsegmented short rays at the upper and lower margin of the caudal fin; first dorsal fin (D1)
spines: all; second dorsal fin (D2), anal fin (A) and pelvic fin (V) elements: all, last bifid ray of D2 and A counted
as a single ray; pectoral fin (P) rays: all elements counted at base of P; branched caudal fin (C) rays: all branched C
rays; segmented caudal fin (C) rays: all branched and unbranched segmented rays; upper and lower procurrent rays
of caudal fin (only countable on x-rays): all visible caudal elements (excluding segmented rays), often becoming
progressively shorter if counted from C (therefore the count is only a minimum, because very small elements are
hardly visible on x-rays. Number of caudal fin rays inserting in hypuralia and parhypural separately counted for
hypuralia 1 + 2, hypuralia 3 + 4, hypural 5 and parhypural; rays taking an intermediate position, if present, are
denoted separately; total number of caudal fin rays inserting in the area of the hypuralia and parhypural: all rays
including intermediately inserting rays (if present). Number of epurals (EPU), i.e. the flattened bones bones above
the dorsal ridge of the urostyle are denoted as of Birdsong et al. (1988), however in very small specimens the
flattened and very thin epurals are sometimes difficult to recognize on x-rays. Vertebrae terminology and spinous
dorsal fin pterygiophore formula (pty) follows Birdsong et al. (1988). Precaudal (abdominal) vertebrae are all
anterior vertebrae without a closed haemal arch (Birdsong et al. 1988) and with pair of separated parapophyses
(Hilton 2011). On lateral x-rays both characters may be hard to confirm for transitional vertebra. In gobies, the last
(i.e. transitional) precaudal vertebra may have enlarged parapophyses, which, in lateral view, are intermediate in
shape between the shape of a closed haemal arch and corresponding haemal spine of the first caudal vertebra and
the shape of parapophyses of penultimate precaudal vertebra, and which have to be counted as the last precaudal
vertebra (compare with Fig. 2). Caudal vertebrae: all vertebrae with a closed haemal arch and including the urostyl;
pterygiophore formula: the initial digit of the formula indicates the count of the interneural space into which the
first pterygiophore of the D1 inserts. The following digits are separated by a dash ( or “minus”) from the initial one
and each represents one interneural space, starting with the one into which the first pterygiophore inserts and
ending with the one representing the interneural space before the insertion of the first pterygiophore of D2, whether
it is supports a D2 spine or not; each number of these digits indicates the count of pterygiophores inserting in the
respective interneural space including zero for no pterygiophores in the interneural space (if no ray or spine is
supported by a pterygiophore, its number is supplied by an asterisk); anal fin (A) pterygiophores anterior to the first
haemal spine as of Birdsong et al. (1988): count of A pterygiophores inserting before or directly over the tip of the
first haemal spine. Terminology of lateral line system follows Miller (1986) based on Sanzo (1911). All type
specimens were reversibly stained in 2% solution of Cyanine Blue in distilled water (Saruwatari et al. 1997) for
studying scales and sensory papillae rows. X-rays were prepared on an UltraFocus Digital Radiography System
(Faxitron Bioptics, Arizona, USA). Apart from the description of the sensory papillae pattern and dentition
characters, the description is based on inspection of all four type specimens of the new species. Sensory papillae
pattern is described in detail form the holotype and one paratype. Dentition characters and the shape of the last
precaudal vertebra of the holotype were inspected on a rapid Micro-CT scan at 120 kV and 110 A for 20 minutes
(2398 projections) using a phoenix x nanotom m cone-beam machine (GE Measurement and Control, Wunstorf,
Germany) of the holotype only. For the micro-CT the specimen was stabilized using polystyrene in a plastic
housing containing a small amount of 75-80% ethanol at the bottom of the vessel to avoid drying of the specimen.
All material was preserved in 96% ethanol after collection, which may cause shrinkage and shorter soft-tissue
based distance measurements as compared with respective measurements of formol-preserved comparative
material of previous studies (Buchheister & Wilson 2005, König & Borcherding 2012).
Zootaxa 4438 (2) © 2018 Magnolia Press
·
383
NEW SPECIES OF DIDOGOBIUS FROM CAPE VERDE IDS
Collections. Type specimens of the new species have been deposited at the Bavarian State Collection of
Zoology, Munich, Germany (ZSM) and the Natural History Museum, Rijeka, Croatia (PMR).
Taxonomy
Generic identification
Generic identification is mostly consistent with the diagnosis of the genus Didogobius Miller, 1966 in Schliewen &
Kovačić (2008) except for the predorsal squamation. However, the recently described D. helenae differs
significantly from other Didogobius species by the complete absence of head canals. Therefore we modified the
first character of the generic diagnosis to accommodate the new species described herein and the fourth and eight
characters to accommodate D. helenae. The genus re-diagnosis is as follows: Didogobius are Gobiinae sensu
Pezold (1993) with the following characters: (1) suborbital papillae without row a; (2) six suborbital transverse
rows present, row 3 missing, last row 7 represented by a single papilla or several papillae near pore α or if the head
canal is absent it is near replacement large papilla; (3) suborbital row 5 long, from near eye to near row d; (4) paired
rows of interorbital transversal papillae absent, if head canal absent, individual large replacement papillae are
present but not a pair of transversal rows. (5) Head naked, predorsal area naked or scaled posteriorly (scaled up to
six rows of cycloid scales in middorsal and up to approx. 1/4 of predosal area length); (6) no mental barbels; (7)
pelvic disc complete with fully developed anterior membrane (frenum); (8) head canals variably reduced from
anterior oculoscapular and preopercular canals present and posterior oculoscapular canal absent to complete
absence of head canals. Note, that we clarified previous versions of character (2) with regard to the row identity of
the one suborbital transversal row missing in Didogobius species. In previous Didogobius species descriptions
(Miller 1966, 1992; Ahnelt & Patzner 1995, Schliewen & Kovačić 2008, Van Tassel & Kramer 2014) the
remaining suborbital row had always been marked as row 3/4. However, it had also been stated in the previous
descriptions (e.g. Schliewen & Kovačić 2008) that row 4 is the missing row. The position of the remaining 3/4 row
in all Didogobius species is below mideye i.e. below the pupil center as is the row 4 in the closely related
Chromogobius species (Miller 1971, Van Tassel 2001). Therefore, we consider it as the most parsimonious
explanation that row 3 and not row 4 is the lost row, and consequently we modified character (2) in the generic
rediagnosis in that sense.
Species identification
Apart from the new species, Didogobius Miller, 1966 contains seven valid species, which are all easily
differentiated from D. janetarum sp. nov. by the following characters (for a comprehensive comparison of
character states see compilations in Schliewen & Kovačić (2008; Table 2) and Van Tassell & Kramer (2014; Table
2). It differs from the type species of Didogobius D. bentuvii Miller, 1966 by having 27 vs. 28 vertebrae and
rounded vs. lanceolate caudal fin; from D. kochi Van Tassell, 1988, D. schlieweni Miller, 1993, D. amicuscaridis
Schliewen & Kovačić, 2008 and D. wirtzi Schliewen & Kovačić, 2008 by having only 10 vs. 11–14 D2 soft rays;
from all species except D. kochi Van Tassell, 1988 (specimens from the Canary islands) by the posterior quarter of
predorsal region in front of first dorsal fin origin with several rows of cycloid scales vs. predorsal midline
completely naked (naked in also in specimens from Cape Verde Islands (pers. obs. — see Comparative material
examined); from D. bentuvii, D. kochi and D. schlieweni by a posterior ctenoid body squamation vs. posterior
cycloid body squamation; from D. amicuscaridis, D. bentuvii, D. kochi, D. schlieweni and D. wirtzi by having only
30–32 scales in the lateral series vs. having between 33 and 70 scales in the lateral line series; from D. helenae by
having anterior oculoscapular head canal present vs. absent; from D. amicuscaridis, D. kochi, D. schlieweni, D.
splechtnai, D. wirtzi by preopercular canal absent vs. present; from D. amicuscaridis, D. wirtzi by suborbital row 7
being each a single papilla vs. several papillae, and by having suborbital rows 2 and 4 close to orbit, vs. distant
from orbit; and finally from all other Didogobius species by branchiostegal membranes being uniformely black
below preopercle, forming a V mark vs. no uniformely black V mark in that position.
SCHLIEWEN ET AL.
384
·
Zootaxa 4438 (2) © 2018 Magnolia Press
Didogobius janetarum, spec. nov.
(Figs. 1a, b, c; 2, 3, 4)
Holotype. ZSM 45303, male, 27.3 + 6.2 mm, Republic of Cabo Verde, Santiago Island, King Bay at Tarrafal
(15.275° - 23.758°), from a horizontal rock crevice in approx. 12 m depth, May 2016, collected with a handnet by
P. Wir t z.
Paratypes. ZSM 45302, juvenile, 18.2 + 4.9 mm, Republic of Cabo Verde, Santiago Island, cliff approx. 2.5
km west of Cidade Velha from a small cave in 14 m depth rock crevice (14.916° - 23.626°), May 2016, collected
with a handnet P. Wirtz; ZSM 40136, female, 24.7 + 5.7 mm, Republic of Cabo Verde, Santiago Island, SSW of
Tarrafal (14.916081° - 23.626626°), dive site “Danger” (15.255° - 23.754°) in approx. 20 m depth, March 19th
2010, collected with a handnet by P. Wirtz; PMR VP4134 (ex ZSM-PIS-GO1804); female, 22.8 + 5.5 mm, same
collection data as holotype ZSM 45303.
Additional material. ZSM-PIS-GO 1803, juvenile, 12.2 + 2.3 mm, same collection data as paratype ZSM 45302.
Comparative material examined. Didogobius kochi Van Tassell, 1988: male, 29.1 + 7 mm (Sao Vicente
Island, Republic of Cabo Verde); ZSM 43050, female, 36.6 + 8 mm and male, 25.6 + 6 mm, ZSM 35462 (Sal
Island, Republic of Cabo Verde).
Diagnosis. Didogobius janetarum spec. nov. differs from all currently described congeners by the combination
of the following characters: (1) 27 vertebrae, (2) D2 I + 10, (3) posterior quarter of predorsal region in front of D1
origin scaled, with several rows of cycloid scales, (4) body squamation cycloid anteriorly and ctenoid posteriorly,
(5) scales in the lateral series 30–32, (6) anterior oculoscapular canal present, (7) preopercular head canal absent,
(8) suborbital row 7 each a single papilla, (9) suborbital rows 2 and 4 close to orbit, (10) branchiostegal membranes
uniformely black below preopercle, forming V-mark.
Description. General morphology. Body proportions and meristics of the holotype and four paratypes are
given in Table I. For a general view see Figs 1a, 1b and 1c. Body moderately elongate and laterally compressed,
head depressed; interorbital space narrow (approx. 20.0–23.5% of eye diameter), dorsolaterally positioned eyes.
Mouth oblique (~35
°
from horizontal), lower jaw slightly projecting, posterior angle of jaws below center of pupil.
Snout shorter than eye and rounded in dorsal view. Anterior nostril tubular (longer than diameter), without process
from rim, barely reaching but not overlapping the upper lip; posterior nostril with slightly erected rim, but not
tubular. Upper lips anteriorly slightly thinner than laterally. Branchiostegal membrane attached along entire lateral
margin; posterior margin of operculum approximately reaching frontal edge of pectoral fin base. Cranial roof
covered by dorsal axial musculature. Pectoral girdle without dermal flaps on anterior edge.
Fins. D1 VI; D2 I/10 (last bifid); A I/9; P 16–17 (counted only on left side); V (left/right) I/5 + 5/I; C
(branched/segmented rays) 14–15/17. Fin lengths and proportions are given in Table I. First D1 slightly longer
(males) or almost as long (females) as second spine, third to sixth spines becoming progressively shorter;
interdorsal space distinct and without fin membrane connection between D1 and D2; longest D2 rays reaching base
of uppermost caudal fin rays. A originates slightly posterior of vertical through D2 origin; C rounded, shorter than
head length; uppermost rays of P not free of membrane, P reaching D2; V complete and rounded with ray 4 as long
as ray 5, and a well-developed anterior pelvic membrane (frenum), its height in midline approx. 1/3 of V spine
length and at its lateral margins approx. 2/3 of V spine length.
Scales. Body covered anteriorly with cycloid and posteriorly with ctenoid scales; ctenoid scales commence
midlaterally from the vertical below base of the fourth spine of D1, dorsally and ventrally further caudally from
approximately the vertical below the base of 2nd ray of D2 and third ray of A. Head area naked, predorsal area with
up to six rows (holotype) or fewer (paratypes) cycloid scales in front of D1 origin, covering at maximum the
posterior quarter of the predorsal region and extending more anteriorly along the middorsal line than laterally, i.e.
with more middorsal predorsal scale rows than dorsolateral ones (see Fig. 1c), prepectoral area naked, breast with
few cycloid scales posteriorly; belly completely scaled with cycloid scales. One row of slightly ctenoid scales on
caudal fin base. LL 30–32, TR 12, CP 12–14.
Teet h. Teeth in lower jaw in two rows. Outer row with five large-sized teeth frontally on each side, caniniform,
pointing slightly backwards. Inner frontal teeth small and conical, numerous, more or less in one row in frontal
position. Inner posterior lateral teeth few and large, caniniform. Teeth in upper jaw in two rows. Outer row with 12
(left side) or 11 (right side) teeth, the frontal three ones on each side large and caniniform, the posteriorlateral ones
medium to small, decreasing in size posteriorly; inner row with one (right side) or three (left side) large frontal
teeth, caniniform.
Zootaxa 4438 (2) © 2018 Magnolia Press
·
385
NEW SPECIES OF DIDOGOBIUS FROM CAPE VERDE IDS
TABLE 1. Morphometric and meristic characters of D. janetorum sp. nov. holotype and all three paratypes (left column for each specimen: absolute values, right column as proportional
measurements in %).
D. janetarum D. janetarum D. janetarum D. janetarum
Holotype Paratype Paratype Paratype
ZSM 45303 ZSM 45302 ZSM 40136 PIS-GO-1804
Distance measurements % % % %
Tl, total length 32.5 21.7 28.7 27.1
Sl, standard length 26.4 17.5 23.3 21.8
%Sl A1 I, 1st anal spine length 1.0 3.8 0.8 4.6 1.7 7.3 6.4 29.4
Ab, anal fin base 5.4 20.3 3.6 20.6 4.5 19.4 5.5 25.3
Ad, body depth at anal fin origin 4.2 15.7 2.8 16.0 3.7 15.8 3.4 15.5
Aw, body width at anal fin origin 2.8 10.7 1.7 9.7 2.5 10.7 2.1 9.7
Bd, body depth 4.0 15.2 3.3 19.2 4.6 19.9 3.6 16.3
Cl, caudal fin length 6.0 22.8 4.3 24.5 5.4 23.0 5.3 24.4
CP, caudal peduncle length 5.9 22.3 5.4 30.9 4.7 20.0 5.4 24.8
CPd, caudal peduncle depth 3.1 11.9 2.1 12.0 2.8 11.9 2.5 11.3
D1 I, 1st dorsal spine length 4.5 16.9 3.4 19.7 4.3 18.3 3.4 15.7
D1 II, 2nd dorsal spine length 2.7 10.0 2.0 11.3 3.1 13.5 3.1 14.1
D1 III, 3rd dorsal spine length 2.1 7.9 1.8 10.3 2.7 11.5 2.8 12.8
D1b, first dorsal fin base 3.4 12.7 2.2 12.6 3.0 12.9 2.8 12.7
D2 I, 1st dorsal spine length 2.6 9.7 2.1 12.2 3.0 12.8 2.0 9.3
D2b, second dorsal fin base 6.7 25.4 4.5 25.8 6.2 26.7 5.0 22.9
Hl, head length 7.3 27.6 5.6 32.1 7.3 31.3 6.1 28.0
Hw, head width 5.3 20.2 3.3 19.1 4.6 19.8 4.1 18.7
IDs, interdorsal space 2.2 8.3 1.5 8.6 2.3 9.7 2.6 12.0
Pl, pectoral fin length 4.8 18.3 4.1 23.4 3.8 16.3 4.0 18.5
SN/A, snout to A 15.6 59.2 10.5 59.9 14.3 61.4 13.4 61.8
SN/AN, snout to anus 14.9 56.4 10.0 57.2 13.2 56.8 12.7 58.5
SN/D1, snout to D1 10.1 38.1 6.8 38.7 9.4 40.4 8.5 39.0
......continued on the next page
SCHLIEWEN ET AL.
386
·
Zootaxa 4438 (2) © 2018 Magnolia Press
TABLE 1. (Continued)
D. janetarum D. janetarum D. janetarum D. janetarum
Holotype Paratype Paratype Paratype
ZSM 45303 ZSM 45302 ZSM 40136 PIS-GO-1804
SN/D2, snout to D2 15.3 58.0 10.2 58.4 14.2 61.1 12.9 59.4
SN/V, snout to V 7.9 30.0 5.2 29.7 7.4 31.9 6.8 31.3
V/AN, pelvic to anus 6.3 24.0 4.9 27.9 6.0 25.8 12.7 58.6
Vd, body depth at pelvic fin origin 3.9 14.8 2.7 15.6 4.0 17.0 3.2 14.7
Vl, pelvic fin length 5.8 21.8 3.3 18.9 5.0 21.3 5.8 26.8
VSI, pelvic spine length 2.2 8.3 1.2 6.9 1.2 5.1 1.6 7.3
Vw, body width at pelvic fin origin 3.6 13.5 2.8 16.3 3.2 13.9 3.0 13.7
%CP CPd, caudal peduncle depth 3.1 53.1 2.1 38.7 2.8 59.7 2.5 45.7
%Hl AULw, anterior upper lip width 0.3 4.1 0.3 5.3 0.5 6.8 0.3 4.9
CHd, cheek depth 1.1 15.1 0.9 16.0 1.1 15.1 0.9 14.8
E, eye diameter 2.2 30.2 1.7 30.3 2.2 30.1 2.0 32.8
Hd, head depth 2.9 39.6 2.1 37.3 3.2 43.7 3.0 48.6
Hw, head width 5.3 73.3 3.3 59.4 4.6 63.1 4.1 66.7
LPd, lateral preorbital depth 0.6 8.2 0.5 8.9 0.5 6.8 0.5 8.2
MULw, maximum lip width 0.4 5.5 0.4 7.1 0.4 5.5 0.4 6.6
PO, postorbital length 4.2 57.6 2.9 51.7 4.9 67.1 4.3 70.5
SN, snout length 1.5 20.6 1.1 19.6 1.3 17.8 1.1 18.0
ULl, upper lip length 3.0 41.1 1.9 33.9 2.6 35.6 1.8 29.5
%E I, interorbital width 0.5 22.7 0.4 23.5 0.5 22.7 0.4 20.0
Meristics
Vertebrae (precaudal + caudal) 10+17 10+17 10+17 10+17
dorsal pterygiophore insertion pattern 3-22110 3-22110 3-22110 3-22110
......continued on the next page
Zootaxa 4438 (2) © 2018 Magnolia Press
·
387
NEW SPECIES OF DIDOGOBIUS FROM CAPE VERDE IDS
TABLE 1. (Continued)
D. janetarum D. janetarum D. janetarum D. janetarum
Holotype Paratype Paratype Paratype
ZSM 45303 ZSM 45302 ZSM 40136 PIS-GO-1804
anal pterygiophores anterior to the first haemal spine 2 2 2 2
scales in lateral series 32+1 32+1 32+1 30+1
scales in transverse series 12 12 12 12
scales - circumpeduncle 12 14 12 12
first dorsal fin spines VI VI VI VI
second dorsal fin elements I,10 (last bifid) I,10 (last bifid) I,10 (last bifid) I,10.5 (last bifid)
anal fin elements I.9 I.9 I.9 I.9
pectoral fin rays 17 17 16
pelvic fin elements I,5 I,5 I,5 I,5 I,5
caudal fin rays hypurals 1+2 5 5 5 5
caudal fin rays hypurals 3+4 6 6 6 6
caudal fin rays hypurals 5 2 2 2 2
caudal fin rays parhypural 1 1 1 1
caudal fin rays of all hyporalia and parhypural 14 14 14 14
caudal fin branched rays 15 15 15 14
caudal fin segmented rays 17 17 17 17
SCHLIEWEN ET AL.
388
·
Zootaxa 4438 (2) © 2018 Magnolia Press
FIGURE 1. Didogobius janetorum sp. nov. (ZSM 45303), male, 27.3 SL, Republic of Cabo Verde, Santiago Island, King Bay
at Tarrafal, preserved holotype, A: lateral left view, B: ventral, C: dorsal and D: X-ray of lateral left view. Photos by U.
Schliewen.
Zootaxa 4438 (2) © 2018 Magnolia Press
·
389
NEW SPECIES OF DIDOGOBIUS FROM CAPE VERDE IDS
Osteology. Vertebral column and pterygiophore insertion pattern (pty) (Fig. 1d). 9 precaudal and 18 caudal
vertebrae (including urostyle); total count: 27. Pty 3-22110; two pterygiophores anterior to the first haemal spine.
One epural. Number of C rays inserting in the hypural 5: 2, 3 + 4 (fused): 6, hypural 1 + 2 (fused): 5 and
parhypural: 1, total number of C rays inserting in hypurals, and parhypural: 14; fused hypural 1 + 2 and 3 + 4
separated by a large gap, which is not inserted by a branched caudal ray.
Lateral line system (Fig. 2). Head with anterior oculoscapular canal with pores σ, λ, κ, ω, α, β, ρ, paratype ZSM
45302 with open furrow between left σ and λ. Posterior oculoscapular canal and preopercular canal absent. Rows
and number of sensory papillae as follows, counted on left side of holotype ZSM 45303 (first value) and of
paratype PMR VP4134 (second value): (I) preorbital: snout with four rows in median preorbital series. Row r
(4, 4)
median to pore σ. Upper row s
1
(3, 3) transversal near posterior nostril, lower s
2
(1, 1) near anterior nostril, and s
3
longitudinal above upper lip (2, 3). Lateral series c in four parts: superior c
2
between posterior and anterior nostrils
(1 + 5, 2 + 4) as two rows; middle transversal c
1
(4, 3) below anterior nostril; inferior upper c
2
(3, 2) and lower c
1
(2,
2) as two rows between lips and row 1. (II) suborbital: six transverse and two longitudinal rows of sensory papillae
on cheek. Rows 1-5 before longitudinal row b; row 6 divided by b in superior (6s) and inferior sections (6i); row 7
near pore α (1: 8, 6, 2: 7, 5, 4: 7, 6, 5: 10, 6, 6s: 6, 3, 6i: 9, 10, 7: 1, 1). Row 1 slightly in front of eye orbit. Rows 2
and 4 close to orbit. Row 6i ending ventrally below level of row d. Longitudinal row b (8, 6) starting anteriorly
behind row 5 (holotype ZSM 45303) or at row 6 (paratype PMR VP4134), ending posteriorly far behind rear
border of eye. Longitudinal row d (18, 15) continuous. (III) preoperculo-mandibular: external row e and internal
row i divided into anterior (e: 21, 18, i: 7, 8), and posterior sections (e: 20, 16, i: 7, 7); row f longitudinal (4, 6). (IV)
oculoscapular: anterior longitudinal row x
1
(9, 7) from behind pore β to behind pore ρ; posterior longitudinal row x
2
(2, 2) above transversal row y (1, 1); transversal row z (7, 7) below pore ρ; transversal row q (4, 5) behind pore ρ;
row u as two large longitudinally arranged papillae behind row q on the place of absent posterior oculoscapular
canal; transversal axillary rows as
1
(2, /), as
2
(3, 4), as
3
(6, /) present and longitudinal rows la
1
(1, 1) and la
2
(2, 1)
above as
2
and as
3
; as
1
and as
3
not visible in paratype PMR VP4134. (V) opercular: transverse row ot (21, 16);
superior longitudinal row os (8, 7); and inferior longitudinal row oi (6, 3); two large papillae present on the place of
absent preopercular canal. (VI) anterior dorsal: transversal row n behind pore ω (9, 8); transversal rows o (/, 2)
divided from each other; longitudinal row g (/, 4) ends posteriorly to front row o, longitudinal row m (/, 2) behind
and below of row g; longitudinal row h continuous (/, 6); rows o, g, m, h not visible in holotype ZSM 45303.
FIGURE 2. Didogobius janetorum sp. nov., head LL sensory papillae and canal pores (ZSM 45303), male, 27.5 mm SL,
Republic of Cabo Verde, Santiago Island, King Bay at Tarrafal, holotype. Drawing by M. Kovačić.
SCHLIEWEN ET AL.
390
·
Zootaxa 4438 (2) © 2018 Magnolia Press
FIGURE 3. Didogobius janetorum sp. nov. (ZSM 45303, live holotype), male, 27.3 SL, Republic of Cabo Verde, Santiago
Island, King Bay at Tarrafal. Photo taken directly after capture by P. Wirtz.
Coloration. In life (based on photograph of holotype (Fig. 3; right side) shortly after capture. Seven dark
orange-brown vertical bands extending on flanks from dorsal midline: the most posterior stripe on caudal peduncle
and two anterior stripes below D1 and predorsal area strait vertical, the middle four stripes broader and connected
ventrally forming a zig-zag pattern, i.e. the seven stripes form a pattern similar to the sequence of letters IAAAAII
from caudal peduncle to head: the anteriormost originates from the predorsal area anterior to the origin of D1 and
splits ventrally with the posterior branch extending behind pectoral fin base, and the anterior extending obliquely to
meet the posterior border of the opercle; the second from the below anterior D1 base from approx. third spine to
approx. fourth spine; the third extending from posterior base of the D1, from approx. fifth to sixth spine, the fourth
and fifth from anterior and central base of D2, from approx. second to third and sixth to seventh ray, respectively,
the sixth from the posterior base of D2 (from approx. tenth ray) and the anterior caudal peduncle, and the seventh
extending at middle caudal peduncle. Posterior end of caudal peduncle and caudal fin base brown. White areas of
the IAAAAII pattern are two poorly visible whitish vertical stripes on caudal peduncle and two clear stripes
anteriorly below D1 and predorsal area; in between three white dorsal saddles are present below D2, seven white
triangles along the lower flanks (the three posteriormost on caudal peduncle poorly visible), and several small
whitish marks inside orange-brown vertical bands. Nape anterior to first orange-brown band with a narrow white
band followed anteriorly by an orange-brown area with embedded whitish marks and with two pairs of orange-
brown lateral “legs” extending from that area laterally, followed behind eye by another narrow white band. One
orange-brown oblique stripe extends ventrally from orbit slightly backwards ending at about center of the cheek,
another one from orbit to mouth; below those two marks are present, one on posterior angle of mouth and one on
lower cheek behind the first one. Preopercle and operculum with three irregular oblique orange bands (forming a “/
I/”-pattern) connected dorsally to form a distinctive orange brown pattern. Lower part of anterior pectoral fin base
with a prominent white area followed anteriorly by an orange brown band or blotch; anterior parts of
branchiostegal membranes, i.e. ventrally below preopercle blackish-grey, forming an intensively pigmented black
spot at each of the upper lateral sides. Ventral body coloration varies from white to grey. Pectoral and ventral fins
with white rays and clear membranes, anal fin rays white with a membranes having a dusky hue; D1 whitish with a
white posterior margin, and four to three brown oblique bands, the anteriormost being darker and broader than the
posterior ones; first D1 spin with contrasting dark-brown and white bands partially corresponding with the D1
bands. D2 white with seven to six oblique orange brown series of brown dots on fin rays, the anteriormost being
darker and broader than the posterior ones; C base with a narrow brown bar embedded in a white field, the
remaining C area whitish to clear with eight or fewer pale rows of brown dots on C rays. Variation (based on color
photographs of three additional specimens (not shown)): smaller individual may have less ill-defined and more
contrasting lateral bands and orange head coloration pattern may be slightly different in size and distribution of
colored elements.
Preserved color. Body bars and blotches on flanks and operculum the same as in life, but orange-brown live
colors have changed after preservation to dark brown, and previously white areas have become whitish-beige, the
pattern being less distinctive than in living specimens; head, nape, cheek and snout bands and blotches also less
distinct; branchiostegal membranes below preopercle uniformely black (without a clear dark spot on the upper
part) forming V mark from ventral view, otherwise ventral whitish-grey; narrow bar on caudal fin base black but ill
Zootaxa 4438 (2) © 2018 Magnolia Press
·
391
NEW SPECIES OF DIDOGOBIUS FROM CAPE VERDE IDS
defined, white area on C fin base prominent and larger than in living specimens; D1 and D2 whitish-translucent
with blackish-brown bands; ventrals and pectorals dusky; anal fin with blackish-brown rays and whitish-
translucent membranes. Variation (based on all five specimens): smaller individual may have less ill-defined and
more contrasting lateral bands and head coloration pattern may be slightly different in size and distribution of
elements – analogous to variation in life coloration (see above).
Etymology. The species name is dedicated to Mrs. Janet Camp and Ms. Janet Eyre, who generously supported
our goby research. A noun in feminine genitive (plural).
Distribution and habitat. Didogobius janetarum sp. nov. is only known from two locations of Santiago
Island, Republic of Cabo Verde. It was collected exclusively from small caves and rock crevices between 12 and 20
m depth. One specimen was observed lurking out of a small cave before it was collected, whereas the other four
known specimens were retrieved after spraying a fish anaesthetic (clove oil) into rock crevices on vertical walls
covered with the coral Tubastrea caboverdiana.
Discussion
Genus level classification and putative relationships. Didogobius, Chromogobius and Gammogobius appear
closely related according to morphological similarity (Schliewen & Kovačić 2008, Van Tassell & Kramer 2014)
and preliminary genetic data (Agorreta et al 2013), but a comprehensive phylogenetic analysis including all taxa is
still missing. The current generic classification of members of these three genera is based on differences of two
selected character states of the head lateral line system, i.e. (i) six vs. seven suborbital rows present and (ii)
suborbital row 5 being long vs. being short (i.e. characters (6) and (7) of the revised Didogobius genus diagnosis
provided in this paper). These two character states have been weighted apparently stronger than differences of
other characters among the twelve members currently placed in these three genera, e.g., presence of a long row 7
vs. row 7 as single papilla, which is diagnostic for the closely related species pair D. amicuscaridis and D. wirtzi
within Didogobius, or vertebrae number, caudal fin shape and eye size, as only the type species of Didogobius, D.
bentuvii has 28 vertebrae, a lanceolate caudal fin and strongly reduced eyes as compared to the single
Gammogobius species, all Chromogobius and all other Didogobius, which have 27 vertebrae, a rounded caudal fin
and large eyes (Miller 1971, Scsepka & Ahnelt 1999, Schliewen & Kovačić 2008, Van Tassell & Kramer 2014).
Other character states appear to have a mosaic-like distribution across all genera: the presence and extent of
predorsal squamation (present only in D. kochi and D. janetorum spec. nov. and C. zebratus (Miller 1971, Van
Tassell 1988, Schliewen & Kovačić 2008, Van Tassell & Kramer 2014); the different extent or loss of the anterior
oculoscapular and preopercular cephalic canals, which varies from complete loss of all canals in D. helenae (Van
Tassell & Kramer 2014) to loss of only the posterior oculoscapular canal in all Chromogobius and most Didogobius
species; the scales in the lateral series are highly variable among species in both multispecies genera as well as
head depression (Table 2, Schliewen & Kovačić 2008). Indeed, preliminary phylogenetic data based on COI-
(“genetic barcoding”)-data and a yet incomplete taxon sampling suggest that Didogobius is paraphyletic with
respect to Chromogobius (Schliewen et al., unpubl.). We conclude, that the final classification and assessment of
relationships of species currently placed in Didogobius, Chromogobius and Gammogobius has still to be settled.
Such an endeavour necessitates the critical review of the status of the type species of Didogobius, D. bentuvii,
which is currently known only from the holotype.
A close relationship of D. janetarum and D. helenae with D. splechtnai would be supported by several
character states, which have already been partially stated as being highly similar between D. helenae and D.
splechtnai by Van Tassell & Kramer (2014): (1) a similar flank coloration with brown or orange bands and narrow
white interspaces, (2) first D1 spine of males longest, others becoming progressively shorter, and (3) a lower lateral
line scale count of only 28-32 vs. 33-70 in all other species. Such a relationship is tentatively supported by
preliminary COI-barcoding data (Schliewen et al., unpubl.).
Precision of distance measurements in small goby species. Several distance measurements in the head
region of larger specimens and from other regions of the smallest paratype were taken from substantially enlarged
X-rays and/or photos (taken with a scale), i.e. in a two-dimensional plane and on a much larger scale, rather than
directly point-to-point in a three dimensions with callipers. The resulting inconsistency in data acquisition may
cause some discrepancy in results among differently taken measures. In addition, the X-rays and/or photos
SCHLIEWEN ET AL.
392
·
Zootaxa 4438 (2) © 2018 Magnolia Press
projection measures have the error of the cosine relationship of the original and projected lengths. The error,
however, should be negligible in small specimens, if the planes of photographic projection and point-to-point
length on the three dimensional specimen are narrowly angled; and they should especially negligible with regard to
errors introduced by several other factors known to increase measurement error. One source of error is avoided by
this method of two-dimensonal measurements on enlarged photos or X-rays, i.e. (i) the low sophistication of the
measuring device,introduced, e.g. by the use of comparatively crude dial callipers with high precision but low
applicability to small and partially soft structures. Other causes of error remain present even with this method,
rarely accounted for in the ichthyological literature: (ii) imprecision of distance measures definitions or complete
lack of them in compared literature data when comparing literature data with own data; (iii) interobserver variation,
playing an important role when comparing literature data with own data; (iv) effect of different preservatives on
specimen condition, i.e. important when comparing ethanol preserved specimens of various ethanol concentrations
and formol preserved specimens (Buchheister & Wilson 2005, König & Borcherding 2012), and, (v) probably most
important, small size of the measured character (Yezerinac et al. 1992, Muñoz-Muñoz & Perpiñan 2010). Most
ichthyologists (including ourselves) provide in the Material and Methods section data on the accuracy of their
measuring device (i.e. 0.01 mm SE), but an appreciation of the true measurement error is rarely given (Petrýl et al.
2014). Especially for the measurement of miniaturized fishes, a comparative investigation of measurement error
introduced by direct measurements with callipers in comparison with measurement from two-dimensional images
(photographs, X-rays) or even three dimensional images (MicroCT) would be desirable. Two-dimensional image-
based fish morphometrics have been shown to be superior over traditional calliper-based morphometrics with
regard to replicability and measurement error (Petrýl et al. 2014, Takács et al. 2016). It remains unclear, however,
to which extent of the angle between projected and real life lengths the measurement error introduced by taking
two-dimensional measurements from very small three-dimensional structures is smaller than measurement error
from using callipers to measure point-to-point three-dimensional fish structures in very small specimens, e.g., for a
30° angle projection the systematic angle error in image-based fish morphometrics would be comparatively large,
about 13.4%. On the other hand though, most ichthyologists with experience in measuring very small distances
would not be surprised with such a large point-to-point measurement error when considering very small distances
and the bulkiness of the calliper tips compared to that distance. This question is of high practical value for the study
of small gobies or miniaturized fishes in general, because at present the error of extensive distance measurements
that are often used to differentiate species of miniaturized fishes is unknown, but may nevertheless be substantial.
Acknowledgements
We would like to thank A. Cerwenka and B. Ruthensteiner (both ZSM) for providing us with a rapid MicroCT scan
of the holotype of the new species. T. Moritz (DMM) is acknowledged for a helpful discussion about definitions
about osteology, and D. Neumann (ZSM) for technical collection support. Travel costs of PW were supported by
Centro de Ciencias do Mar, Faro (CCMAR); this study received Portuguese national funds through FCT -
Foundation for Science and Technology - through project UID/Multi/04326/2013. UKS´s work has been supported
in part by SEA LIFE Deutschland GmbH grant, and MK´s work has been supported in part by Croatian Science
Foundation under the project IP-2016-06-5251.
References
Agorreta, A., San Mauro, D., Schliewen, U., Van Tassell, J.L., Kovačić, M., Zardoya, R. & Ruber, L. (2013) Molecular
phylogenetics of Gobioidei and phylogenetic placement of European gobies. Molecular Phylogenetics and Evolution, 69,
619–633.
https://doi.org/10.1016/j.ympev.2013.07.017
Ahnelt, H. & Patzner, R.A. (1995) A new species of Didogobius (Teleostei: Gobiidae) from the western Mediterranean.
Cybium, 19 (1), 95–102.
Birdsong, R.S., Murdy E.O. & Pezold, F. (1988) A study of the vertebral column and median fin osteology in gobioid fishes
with comments on gobioid relationships. Bulletin of Marine Science, 42 (2), 174–214.
Buchheister, A. & Wilson, M. (2005) Shrinkage correction and length conversion equations for Theragra chalcogramma,
Mallotus villosus and Thaleichthys pacificus. Journal of Fish Biology, 67, 541–548.
Zootaxa 4438 (2) © 2018 Magnolia Press
·
393
NEW SPECIES OF DIDOGOBIUS FROM CAPE VERDE IDS
https://doi.org/10.1111/j.0022-1112.2005.00741.x
Fricke, R. (1983) A method for counting caudal fin rays of Actinopterygian fishes. Braunschweiger Naturkundliche Schriften,
1 (4), 729–733.
Hilton, E.J. (2002) Osteology of the Extant North American Fishes of the Genus Hiodon Lesueur, 1818 (Teleostei:
Osteoglossomorpha: Hiodontiformes). Fieldiana Zoology, 100, 1–142.
https://doi.org/10.5962/bhl.title.2666
Hilton, E.J. (2011) Bony fish skeleton. In: Farrell, A.P. (Ed.), Encyclopedia of Fish Physiology: From Genome to Environment.
Vol. 1. Academic Press, San Diego, pp. 434–448.
https://doi.org/10.1016/B978-0-12-374553-8.00240-9
König, U. & Borcherding, J. (2012) Preserving young of the year Perca fluviatilis in ethanol, formalin, or in a frozen state and
the consequences for measuring morphometrics. Journal of Applied Ichthyology, 28, 740–744.
https://doi.org/10.1111/j.1439-0426.2012.01958.x
Larson, H.K. (2001) A revision of the gobiid fish genus Mugilogobius (Teleostei: Gobioidei), and its systematic placement.
Records of the Western Australian Museum, 62 (Supplement), 1–233.
https://doi.org/10.18195/issn.0313-122x.62.2001.001-233
Miller, P.J. (1966) A new genus and species of gobiid fish from the eastern Mediterranean. Annals and Magazine of Natural
History, Series 13, 8 (87/88), 161–172.
Miller, P.J. (1971) A revision of the Mediterranean gobiid genus Chromogobius (Teleostei-Perciformes). Journal of Zoology
(London), 164, 305–334.
https://doi.org/10.1111/j.1469-7998.1971.tb01321.x
Miller, P.J. (1986) Gobiidae. In: Whitehead, P.J.P., Bauchot, M.-L., Hureau, J.-C., Nielsen, J. & Tortonese, E. (Eds.), Fishes of
the North-eastern Atlantic and the Mediterranean. Vol. 3. UNESCO, Paris, pp. 1019–1085.
Miller, P.J. (1993) A new species of Didogobius (Teleostei: Gobiidae) from the Adriatic Sea. Journal of Natural History, 26 (6),
1413–1419.
https://doi.org/10.1080/00222939200770791
Munoz-Munoz, F. & Perpinan, D. (2010) Measurement error in morphometric studies: comparison between manual and
computerized methods. Annales Zoologici Fennici, 47, 46–46.
https://doi.org/10.5735/086.047.0105
Petrýl, M., Kalous, L. & Memis, D. (2014) Comparison of manual measurements and computer-assistet image analysis in fish
morphometry. Turkish Journal of Veterinary and Animal Sciences, 38, 88–94.
https://doi.org/10.3906/vet-1209-9
Pezold, F. (1993) Evidence for monophyletic Gobiinae. Copeia, 1993 (3), 634–643.
Takacs, P., Vitál, Z., Ferincz, Á. & Staszny, À. (2016) Repeatability, reproducability, separative power and subjectivity of
different fish morphometric analysis methods. PLoS ONE, 11 (6), ew0157890.
https://doi.org/10.1371/journal.pone.0157890
Sanzo, L. (1911) Distribuzione delle papille cutanee (organi ciatiformi) e suo valore sistematico nei Gobi. Mittheilungen aus
der Zoologischen Station zu Neapel, 20, 251–328.
Saruwatari, T., Andrés López, J. & Pietsch, T.W. (1997) Cyanine blue: a versatile and harmless stain for specimen observation.
Copeia, 1997 (4), 840–841.
https://doi.org/10.2307/1447302
Scsepka, S. & Ahnelt H. (1999). Wiederbeschreibung von Gammogobius steinitzi Bath 1971 sowie ein Erstnachweis von
Corcyrogobius liechtensteini (Kolombatovic 1891) für Frankreich (Pisces, Gobiidae). Senckenbergiana biologica, 79 (1),
71–81.
Schliewen, U.K. & Kovačić, M. (2008) Didogobius amicuscaridis spec. nov. and D. wirtzi spec. nov., two new species of
symbiotic gobiid fish from São Tomé and Cape Verde islands. Spixiana, 31, 247–261.
Van Tassell, J.L. (1988) A new species of Didogobius (Perciformes: Gobiidae) from the Canary Islands. American Museum
Novitates, 1988 (2906), 1–8.
Van Tassell, J.L. (2001) Chromogobius (Teleostei: Gobiidae), a new species from the eastern Atlantic. Copeia, 2001 (4), 1073–
1080.
https://doi.org/10.1643/0045-8511(2001)001[1073:CTGANS]2.0.CO;2
Van Tassell, J.L. & Kramer, A. (2014) A new species of Didogobius (Teleostei: Gobiidae) from the Canary Islands. Zootaxa,
3793 (4), 453–464.
https://doi.org/10.11646/zootaxa.3793.4.4
Yezerinac, S.M., Lougheed, S.C. & Handford, P. (1992) Measurement error and morphometric studies: Statistical power and
observer experience. Systematic Biology, 41 (4), 471–482.
https://doi.org/10.1093/sysbio/41.4.471
