BULLETIN OF MARINE SCIENCE, 70(1): 33–39, 2002
TAXONOMY AND ZOOGEOGRAPHY OF MICRODESMUS CARRI GILBERT (MICRODESMIDAE)
Felipe Amaya, Arturo Acero P. and María M. Criales
ABSTRACT This contribution describes juvenile specimens of the cienaga wormfish Microdesmus carri (Microdesmidae) collected at the entrance of the estuarine lagoon Ciénaga Grande de Santa Marta on the Colombian Caribbean. This is the first report of M. carri for South American waters and the third of the species since it was originally described from the Caribbean coast of Costa Rica and later reported for Belize and the Gulf of Mexico. Meristic and morphometric data of the Colombian material are similar to those described for Costa Rica but differences from the Belize and Gulf of Mexico material were found in the fin counts. The species is considered here a southern Caribbean endemic, whose distribution may be explained by geological and oceanographic factors.
The gobioid fish family Microdesmidae is widespread in marine tropical waters and includes two subfamilies: Microdesminae (wormfishes) and Ptereleotrinae (dartfishes or firefishes) (Nelson, 1994). However, Thacker (2000) excluded the dartfishes from the Microdesmidae. Wormfishes burrow in soft muddy and sandy bottoms in shelf ecosys- tems from coral reefs to estuaries and from tidepools to about 40 m depth (Robins et al., 1986; Nelson, 1994). Seven species of wormfishes have been described for the western Atlantic, two in the genus Cerdale (C. fasciata and C. floridana) (Dawson, 1974) and five in the genus Microdesmus (M. bahianus, M. lanceolatus, M. longipinnis, M. lucus and M. carri) (Gilbert, 1966; Dawson, 1977). The only wormfish previously recorded from the Colombian Caribbean is C. floridana (Acero and Garzón, 1986). Microdesmus carri was originally described by Gilbert (1966) based on 36 specimens collected from Tortuguero, Costa Rica (10∞30'N 83∞30'W), western Atlantic, in May 1963 and August 1963 and 1964. Recently, Smith and Thacker (2000) reported on additional collections of the species from the southwestern Gulf of Mexico and Belize; however, some interesting differences between their material and that from Costa Rica and Colombia merit discus- sion. The objective of this paper is to report M. carri for the Colombian Caribbean and to discuss its zoogeography. Information is also provided about the ecology of the M. carri material captured at the Boca de la Barra, Ciénaga Grande de Santa Marta (CGSM), Colombia.
MATERIAL AND METHODS
Collections were performed at Boca de la Barra (10∞59'N, 74∞17'W), the connection of the es- tuarine CGSM to the Caribbean Sea, with a moored channel net (800 m) suspended 50 cm below the surface. The cod end of the net was set up at sunset and retrieved after dawn for two consecutive days during the full and the new moon monthly from February 1998 to February 1999 (Criales et al., 2002). Fish larvae and juveniles were sorted and identified to the lowest possible taxon. Fin elements were counted on 34 specimens following Gilbert (1966) and using an ocular micrometer in a stereomicroscope. Measurements were made on 15 specimens, which were cleared and stained using the Smith and Richardson (1979) technique (Table 1). Descriptive information about the species is included based on Colombian material and the original description.
33 34 BULLETIN OF MARINE SCIENCE, VOL. 70, NO. 1, 2002
RESULTS AND DISCUSSION
Microdesmus carri Gilbert, 1966 (Fig. 1, Table 1)
Material Examined.—A total of 112 specimens were collected at the Boca de la Barra, of which a peak of 98 juveniles were collected in July and the remaining specimens in March, June, August and September. Specimens are kept at the fish collection of the Instituto de Investigaciones Marinas y Costeras (INVEMAR 2409), Santa Marta, Colom- bia. All specimens were of uniform size ranging between 27 and 30.5 mm standard length (SL), precluding analysis of migrations or structural development of the growth of indi- viduals. Diagnosis.—M. carri can be separated from all other Microdesmus from the western Atlantic by the following combination of counts: dorsal fin with 25–26 (modally 25) spines and 67–71 (modally 69) total elements, anal fin with 33–37 (modally 35 and 36) rays and originating between verticals from dorsal-fin elements 33/34 and 36/37 (mod- ally 35/36), total dorsal and anal fin elements 100–107 (modally 104), and 36–38 (mod- ally 37) precaudal vertebrae (Table 1). The material from Belize and the Gulf of Mexico reported as M. carri by Smith and Thacker (2000) has generally higher fin counts. Description.—The head is somewhat depressed anteriorly, with a blunt tip, 10.8–12.2 % SL; the eyes are lateral on the highest part of the head, their diameter 20.3–25.4 % of the head length, the lips are fleshy but not very large (Fig. 1, Table 2). The lower jaw has only one row of short, strong, pointed, conical, and backward oriented teeth, which de- crease in size posteriorly. Gill openings are somewhat long, beginning immediately be- fore the insertion of the first pectoral-fin ray; the opercular margin runs vertically to- wards the isthmus. The body is elongated and tubular. The dorsal fin originates after the beginning of the pectoral fin; the predorsal fin length is 17.9–20.5 % SL. The dorsal fin is not confluent with the caudal fin and has XXV or XXVI spines and 42–46 rays, with XXV, 44 as the modal count. The total number of vertebrae ranged between 66 and 70, modally 68 (Table 1). All specimens were naked. In general our specimens coincide with the original description of the species, but Smith and Thacker’s (2000) material from Belize and the Gulf of Mexico tends to differ from the southwestern Caribbean speci- mens. The main differences based on Smith and Thacker’s (2000) description of their specimens, are centered on the higher counts of the northern material, i.e., anal-fin rays 35–39, mean close to 37, and counts in one specimen of XXVI, 46 dorsal-fin elements and anal fin originating at vertical of dorsal-fin elements 38/39. The total of 110 elements in the specimen reported by Smith and Thacker (2000) is well over 107, the upper limit of the Costa Rican and Colombian material. It may seem that the northern population is more elongated, i.e., it has higher number of dorsal and anal fin elements; nevertheless, it does not seem to clearly have higher vertebral counts than the southern population. Gil- bert (1966: 527) reported that “M. carri and M. suttkusi share the following characters, the combinations of which are unique among the species of Microdesmus: total dorsal elements 67 to 71; total anal elements 33 to 37 (33 to 36 in M. carri); origin of anal fin beneath 33rd ro 37th dorsal elements.” Dawson (1977), in his key to western Atlantic species of Microdesmus, reported that M. carri has 24–27 dorsal fin spines and 34–40 AMAYA ET AL: STATUS OF MICRODESMUS CARRI 35
Table 1. Meristic data of southwestern Caribbean specimens of Microdesmus carri. Data from Costa Rica were taken from Gilbert (1966).
Tsotal dorsal fin elements Anal fin element 687 69607173343536373 C150osta Rica 1741-791- C774olombia 16- 1151341
Total dorsal and anal fin elements 1100 120 130 140 150 160 170 10 C1--10osta Rica 1474 C18265855olombia
Position of anal-fin origin in relation to dorsal-fin elements 353/34 364/3 375/3 36/3 C1695osta Rica C190olombia 24
Precaudal and caudal vertebrae 36+360 3 +361 3 +372 3 +370 3 +371 3 +382 3 +380 3 +381 3 +32 C123263-1-osta Rica C-212521-1olombia
(modally 37) anal fin rays. Those ranges are very wide, even including the northern ma- terial. We do not have an explanation for such discrepancies. Pigmentation.—Specimens of M. carri kept in alcohol after several days of being cap- tured were relatively unpigmented. The most conspicuous pigmentation was a row of small and elongated melanophores from throat to level of pectoral-fin base, diverging into a paired series to a point slightly behind half-way between pelvic-fin base and anal- fin origin. There was also a paired series of melanophores located at both sides of the base of each anal-fin ray. Medially on the caudal peduncle there is an elongated melano- phore and on the caudal-fin base one or two additional melanophores. Specimens de- scribed by Gilbert (1966) were separated into two distinct morphological types. Two large individuals (36.0 and 53.0 mm SL) were well pigmented on body and fins, with a rounded caudal fin, the dorsal and anal fins adnate to caudal. The other 34 smaller indi- viduals (28.0 to 33.5 mm SL) were almost completely devoid of pigment on body and fins with an emarginate caudal fin and the dorsal and anal fins not adnate to caudal. M. carri from Colombia corresponds to the postlarval or juvenile stage described by Gilbert (1966). The shape and development of the anal fins also correspond to Gilbert’s descrip- tion. Gilbert (1966) found that the Tortuguero specimens collected at midday were heavily pigmented in contrast to those collected during the night,which were unpigmented. Be- cause we did not collect during daylight light hours we can not add to this discussion. Common Name.—Herewith we propose for M. carri the official common name of cienaga [pronounced sjé-na-ga] wormfish, from the Spanish ciénaga meaning shallow lagoon or swamp, which is the kind of environment in Colombia and Costa Rica where the species has been found. Ecology.—Almost nothing is known about the life history, habits or behavior of these fishes. Gilbert (1966) described M. carri from the Caribbean shores of Costa Rica. Speci- 36 BULLETIN OF MARINE SCIENCE, VOL. 70, NO. 1, 2002
Figure 1. Microdesmus carri specimen from Colombia (INVEMAR 2409), 30.5 mm standard length.
mens from Costa Rica were collected about one-half mile above the mouth of Tortuguero, a shallow water lagoon with salinity ranging between 1.5 and 20 and a muddy/sandy bottom (Gilbert, 1966). Specimens of the cienaga wormfish from Colombia were found at the entrance to the Ciénaga Grande de Santa Marta, a shallow tropical lagoon that serves as a nursery ground for a great variety of species of fishes and invertebrates (Santos- Martínez and Acero, 1991). The Boca de la Barra receives fresh water from the Río Magdalena and other rivers from the mountains of the Sierra Nevada de Santa Marta and marine waters from the Caribbean Sea. The bottom of the lagoon is muddy to sandy and the salinity fluctuates between 9 and 33 (Rueda and Mancera, 1995). Fishes of the genus Microdesmus are demersal spawners and some species have been reported associated with coral reefs in the Caribbean (Limouzy-Paris et al., 1994, 1997; Sponaugle and Cowen, 1996; Smith and Thacker, 2000). Limouzy-Paris et al. (1997) reported advanced larval stages of coral reef wormfishes (M. lanceolatus and M. longipinnis) offshore of the Florida Keys between 0–40 m in depth. Smith and Thacker (2000) described the late-stage larvae of five wormfish species from Belize, including both Cerdale and Microdesmus. Gilbert (1966) found many postlarval and juvenile specimens of M. carri entering the Tortuguero Lagoon. Gilbert and Kelso (1971) proposed that M. carri spawns in the ocean. However, they also mentioned that M. carri is included in the tismiche, which they defined as “a diverse assemblage of post-larval fishes and crustaceans that appear in the lagoon at intervals not yet determined” (Gilbert and Kelso, 1971). They also thought that the tismiche is a phenomenon similar to the Puerto Rican seti or the Philippine ipon. Silva Melo and Acero (1990) studied the biology of Sycidium antillarum, the main species that forms the Santa Marta version of the tismiche, the tití. It seems that the species that contribute with their postlarvae to the tití are freshwater benthic fishes that do not go through reproduc- tive migrations as adults, but spawn in the adults’ habitats (Silva Melo and Acero, 1990). In the case of Sycidium, the larvae migrate downstream to the sea after hatching and some days later the postlarvae are collected in the mouths of the streams where the adults live, migrating upstream. Accordingly, it may be that the cienaga wormfish, which spawns demersally, does not spawn in the ocean, but like other members of the tití (or tismiche) spawns without any significant migration by the reproducing adults. The fact that the late-stage larvae reported as M. carri by Smith and Thacker (2000) from Belize were collected in a coral reef environment, far from estuarine situations, may reinforce our hypothesis that they were dealing with a different species. Zoogeography.—Gilbert (1966) proposed that M. carri is the sister species of M. suttkusi, described by him from the Pacific coast of Costa Rica. He based that opinion in their sharing of narrow ranges of dorsal and anal fin counts. If the diagnosis of M. carri is expanded to include the material from Belize and the Gulf of Mexico, it will be hard to exclude M. suttkusi from such a diagnosis, at least on meristic grounds. In such a case, M. AMAYA ET AL: STATUS OF MICRODESMUS CARRI 37
L
J
4
6
3
7
5
2
3
5
9
U
.
.
.
.
.
.
.
.
.
.
,
1
8
3
0
1
0
5
6
4
4
1
h
2
1
t
g
n
e
l
.
t
h
2
8
3
7
5
5
3
1
9
u
t
.
.
.
.
.
.
.
.
.
.
o
15
p
8
33
01
10
02
57
6
49
40
16
e
n
20
15
s
d
y
L
d
n
o
S
b
3
9
4
7
6
3
2
3
1
,
.
.
.
.
.
.
.
.
.
.
h
14
0
36
03
10
03
56
7
45
43
26
t
D
32
10
g
B
n
,
e
l
h
t
d
g
a
n
4
9
4
7
7
2
3
1
1
e
.
.
.
.
.
.
.
e
.
.
.
13
l
h
9
39
06
13
06
57
7
40
45
27
24
17
c
i
L
v
H
l
e
,
p
h
t
e
4
0
4
7
7
8
7
5
1
r
g
.
.
.
.
.
.
.
.
.
.
12
p
n
0
38
16
18
06
52
7
44
47
27
e
30
15
l
L
V
d
r
P
a
,
d
h
2
8
3
7
5
3
1
9
8
n
.
.
.
.
.
.
.
.
.
.
t
a
11
7
38
00
18
03
58
3
48
30
17
g
t
23
17
n
s
e
l
L
l
S
a
r
.
i
o
3
8
3
7
7
0
5
3
0
r
t
.
.
.
.
.
.
.
.
.
.
r
c
0
32
07
10
05
56
7
43
41
27
a
e
31
17
c
p
e
s
r
u
P
m
s
L
4
0
3
7
8
8
4
2
9
e
.
.
.
.
.
.
.
.
.
.
P
d
9
30
17
10
03
58
7
42
45
14
P
o
25
10
,
r
h
c
t
i
g
M
n
e
f
l
1
7
3
7
0
8
1
0
7
o
.
.
.
.
.
.
.
.
.
.
l
8
38
09
15
03
59
6
43
46
16
s
a
23
17
n
n
e
a
e
m
r
i
p
c
e
L
2
7
3
7
7
3
4
2
9
p
.
.
.
.
.
.
.
.
.
.
s
A
8
30
04
10
03
50
7
47
44
14
P
20
17
d
,
e
h
n
t
i
g
a
t
n
s
e
3
8
3
6
3
0
1
9
9
l
.
.
.
.
.
.
.
.
.
.
d
9
36
04
19
06
51
7
43
36
16
l
n
25
17
a
a
s
r
d
o
e
d
r
e
a
r
e
3
9
3
6
4
2
1
0
9
l
p
.
.
.
.
.
.
.
.
.
.
c
8
30
05
16
01
53
7
47
46
16
L
20
10
n
i
D
)
P
,
m
r
e
m
t
3
9
3
0
5
0
3
1
9
.
.
.
.
.
.
.
.
.
.
e
n
9
30
01
16
11
54
8
43
40
16
i
m
20
15
(
a
i
a
t
d
a
e
d
y
4
9
4
7
5
7
4
1
9
c
e
.
.
.
.
.
.
.
.
.
.
i
r
9
35
07
14
00
56
7
40
43
16
t
D
25
10
e
E
m
,
o
h
t
h
g
p
4
8
3
7
1
3
1
9
r
n
.
.
.
.
.
.
.
.
.
.
123456789 0
e
o
9
38
01
13
04
56
8
43
43
16
l
25
15
M
w
.
a
2
j
r
e
l
e
L
L
L
L
L
b
L
p
L
J
D
a
D
D
A
P
V
L
n
p
T
u
H3
U5
E8
P4
P3
P5
P3
B6
S0
S8 38 BULLETIN OF MARINE SCIENCE, VOL. 70, NO. 1, 2002 carri may well end up being the only wormfish known from both sides of the continent. However, we propose this as a species complex, including M. carri, M. suttkusi, and one undescribed species from the northern Caribbean and the Gulf of Mexico. If this is the case, M. carri is one of the fish species endemic to the southern Caribbean (Costa Rica- Venezuela). Acero (1984, 1985) discussed some of the known cases of fish endemicity in that Caribbean region. Some species are rocky reef inhabitants, such as those of the fami- lies Chaenopsidae, Haemulidae, and Serranidae. However, most of the fishes endemic to the southern Caribbean are inhabitants of brackish water/muddy bottom environments (i.e., pristigasterids, engraulids, ariids, batrachoidids, and sciaenids). Before the uplift of the Panama isthmus, the southern Caribbean and the eastern Pacific formed a biogeo- graphic province dominated by brackish and murky waters and relatively cool condi- tions. The southern Caribbean includes several endemic fish species, which are phyloge- netically more related to eastern Pacific species than to those of the northern Caribbean. Examples of these are species of the genera Acanthemblemaria, Ctenosciaena, Ekemblemaria, Paralabrax, Selenaspis, and Thalassophryne, among many others (Acero, unpubl.). If M. carri and M. suttkusi are sister species, as proposed by Gilbert (1966) and us, it would fit well in the biogeographic pattern discussed above. Ocean currents and hydrodinamic processes also play an important role in the distribution of marine organ- isms. In the southwestern Caribbean the prominent cyclonic Panama-Colombia Gyre has been revealed to be a quasi-permanent feature of the circulation interacting with the shelf waters off Colombia, Panama, Costa Rica, Nicaragua and the plume of the Magdalena River (Mooers and Maul, 1998). This mesoscale gyre may provide a feasible retention mechanism for these endemic species of the southwestern Caribbean.
ACKNOWLEDGMENTS
We thank W. Richards (National Marine Fisheries Center, Miami), and C. Paris-Limouzy (Rosenstiel School of Marine and Atmospheric Science, University of Miami) for their advice and comments; to A. López (Instituto de Investigaciones Marinas y Costeras, INVEMAR), and the fishermen of the Ciénaga Grande of Santa Marta and D. Garizábal for their assistance in the field. Funding was provided by COLCIENCIAS (Colombian Institute for Science and Technology) through a cooperative project between the Instituto de Investigaciones Marinas (INVEMAR) and the Rosenstiel School of Marine and Atmospheric Science (RSMAS), University of Miami. This is contribution 723 of INVEMAR and 189 of the Marine Biology Graduate Program of Universidad Nacional de Colombia.
LITERATURE CITED
Acero P., A. 1984. The chaenopsine blennies of the southwestern Caribbean (Pisces: Clinidae: Chaenopsinae) I. Systematic analysis and zoogeography. An. Inst. Inv. Mar. Punta Betín 14: 29–46. ______. 1985. Zoogeographical implications of the distribution of selected families of Carib- bean coral reef fishes. Proc. Fifth Int’l. Coral Reef Congress, Tahiti: 433–438. ______and J. Garzón F. 1986. Peces de las islas del Rosario y de San Bernardo (Colombia). II. Tres nuevos registros para el Caribe sur y 16 m·s para la costa norte continental colombiana. An. Inst. Inv. Mar. Punta Betín 15/16: 3–29. AMAYA ET AL: STATUS OF MICRODESMUS CARRI 39
Criales, M. M., F. Amaya, C. Yeung, A. López, D. Jones and W. Richards. 2002. Larval supply of fishes, shrimps and crabs into the nursery ground of the Ciénaga Grande of Santa Marta, Co- lombian Caribbean. Carib. J. Sci. 38 (1–2) (in press). Dawson, C. E. 1974. A review of the Microdesmidae (Pisces: Gobioidea) l. Cerdale and Clarkichthys with descriptions of three new species. Copeia 1974: 409–448. ______. 1977. A new western Atlantic wormfish (Pisces: Microdesmidae). Copeia 1977 (1): 7–10. Gilbert, C. R. 1966. Two new wormfishes (Family Microdesmidae) from Costa Rica. Copeia 1966 (2): 325–332. ______and D. P. Kelso. 1971. Fishes of the Tortuguero area, Caribbean Costa Rica. Bull. Fla. State Mus. Biol. Sci. 16 (1): 1–54. Limouzy-Paris, C. B., M. F. McGowan, W. J. Richards, J. P. Umarán and S. S. Cha. 1994. Diversity of fish larvae in the Florida Keys: results from SEFCAR. Bull. Mar. Sci. 54: 857–870. ______, H. C. Graber, D.L. Jones, A. Röpke and W. J. Richards. 1997. Translocation of larval coral reef fishes via sub-mesoscale spin-off eddies from the Florida Current. Bull. Mar. Sci. 60: 966–983. Mooers, C. N. K. and G. A. Maul. 1998. Intra-Americas sea circulation. Pages 183–208 in A. R. Robinson and K. H. Brink, eds. The sea, vol. 11. John Wiley & Sons, New York. Nelson, J. S. 1994. Fishes of the world, 3rd ed. John Wiley & Sons, New York. 600 p. Robins, C. R., G. C. Ray and J. Douglass. 1986. A field guide to Atlantic coast fishes of North America. Houghton Mifflin Co., Boston. 354 p. Rueda, M. and E. Mancera. 1995. Alteraciones fisicoquímicas de la columna de agua, generadas por el uso del boliche (método de pesca artesanal) en la Ciénaga Grande de Santa Marta, Caribe colombiano. An. Inst. Inv. Mar. Punta Betín 24: 23–37. Santos-Martínez, A. and A. Acero P. 1991.Fish community of the Ciénaga Grande de Santa Marta (Colombia): composition and zoogeography. Ichthyol. Explor. Freshwaters 2(3): 247–263. Silva Melo, L. and A. Acero P. 1990. Sistemática, biología y ecología del tití Sycidium antillarum Grant (Pisces: Gobiidae) en la región de Santa Marta, Colombia. An. Inst. Inv. Mar. Punta Betín 19–20: 153–172. Smith, D. G. and C. E. Thacker. 2000. Late-stage larvae of the family Microdesmidae (Teleostei: Gobioidei) from Belize, with notes on systematics and biogeography in the western Atlantic. Bull. Mar. Sci. 67: 997–1012. Smith, P. E. and S. L. Richardson. 1979. Técnicas modelo para prospecciones de huevos y larvas de peces pelágicos. FAO. Documento de Pesca (175). 107 p. Sponaugle, S. and R. Cowen. 1996. Nearshore patterns of coral reef fish larval supply to Barbados, West Indies. Mar. Ecol. Prog. Ser. 133: 13–28. Thacker, C. 2000. Phylogeny of the wormfishes wormfishes (Teleostei: Gobioidei: Microdesmidae). Copeia 2000(4): 940–957.
DATE SUBMITTED: March 6, 2000. DATE ACCEPTED: September 12, 2001.
ADDRESSES: (F.A.) Instituto de Investigaciones Marinas (INVEMAR), Apartado 1016, Santa Marta, Colombia. (A.A.P.) Universidad Nacional de Colombia, Instituto de Ciencias Naturales, Apartado 1016 (INVEMAR), Santa Marta, Colombia. (M.M.C.) Rosenstiel School of Marine and Atmospheric Science, University of Miami, 4600 Rickenbacker Causeway, Miami, Florida 33149. E-mail: