INTRODUCTIONTop

Fish larvae often go through very complex processes of metamorphosis during growth. A wide variety of developmental patterns, characterized by differential growth of different regions of the body at different sizes, are described for marine species and justified via ecomorphological hypotheses about the allocation of energy during growth (Russo et al. 2007Russo T., Costa C., Cataudella S. 2007. Correspondence between shape and feeding habit changes throughout ontogeny of gilthead sea bream Sparus aurata L., 1758. J. Fish. Biol. 71: 629-656.). Certain species undergo gradual changes, whereas other species experience threshold effects in their ecomorphological relationships during ontogeny (Kouttouki et al. 2006Kouttouki S., Georgakopoulou E., Kaspiris P., et al. 2006. Shape ontogeny and variation in the sharpsnout seabream, Diplodus puntazzo (Cetti 1777). Aqua. Res. 37: 655-663. , Russo et al. 2009Russo T., Pulcini D., Bruner E., et al. 2009. Shape and size variation: growth and development of the dusky grouper (Epinephelus marginatus Lowe, 1834). J. Morph. 270: 83-, Frédérich et al. 2008Frédérich B., Adriaens D., Vandewalle P. 2008. Ontogenetic shape changes in Pomacentridae (Teleostei, Perciformes) and their relationships with feeding strategies: a geometric morphometric approach. Biol. J. Linn. Soc. 95: 92-105., 2012Frédérich B., Colleye O., Lepoint G., et al. 2012. Mismatch between shape changes and ecological shifts during the post-settlement growth of the surgeonfish, Acanthurus triostegus. Front. Zool. 9: 8.).

In fish species with pelagic and filiform larval stages and an elongate adult body form, the ontogenetic growth intensities are not distributed uniformly across the body but in a regular pattern, constituting a smooth, continuous gradient. The gradient is characterized by a period with terminal growth centres (high intensities), corresponding to head and tail, and a continuous, U-shaped gradient between them (Fuiman 1983Fuiman L. 1983. Growth gradients in fish larvae. J. Fish Biol. 23: 117-123. ). In these species, the completion of the head is crucial for feeding and respiratory functions, whereas the completion of the tail is pivotal for cruising and escape reactions (Russo et al. 2009Russo T., Pulcini D., Bruner E., et al. 2009. Shape and size variation: growth and development of the dusky grouper (Epinephelus marginatus Lowe, 1834). J. Morph. 270: 83-).

The suborder Blennoidei includes 6 families and over 880 benthic species (Hastings and Springer 2009Hastings P., Springer V. 2009. Systematics of the Blennioidei and the included families Chaenopsidae, Clinidae, Labrisomidae and Dactyloscopidae. In: Patzner R., Goncalves E., Hastings P., et al. (eds), The Biology of Blennies. Enfield, NH. Science Publishers, pp. 3-30.), which inhabit shallow waters of tropical and temperate seas. The family Clinidae consists of relatively small fishes with a standard length (SL) of less than 300 mm placed in three tribes: the matritrophic (ovoviviparous) Clinini and Ophiclinini, and the oviparous Myxodini (George and Springer 1980George A., Springer V. 1980. Revision of the clinid fish tribe Ophiclinini, including five new species, and definition the family Clinidae. Smithson. Contrib. Zool. 307: 1-30., Stepien 1992Stepien C. 1992. Evolution and biogeography of the Clinidae (Teleostei: Blennioidei). Copeia 1992(2): 375-392.). The sole representative of the family Clinidae in Southeastern Pacific coasts is the genus Myxodes, with three described species: Myxodes viridis Valenciennes, 1836, M. cristatus Valenciennes, 1836, and M. ornatus (Stephens and Springer, 1973). Myxodes viridis ranges from Independencia Bay, Peru, to southern Chile (Pequeño et al. 1995Pequeño G., Lamilla J., Lloris D., et al. 1995. Comparación entre las ictiofaunas intermareales de los extremos austral y boreal de los canales patagónicos. Rev. Biol. Mar. 30(2): 155-177.). The adults feed mostly on crabs, whereas juveniles (<75 mm SL) eat primarily harpacticoid copepods, amphipods, gastropod snails and isopods (Stepien 1990Stepien C. 1990. Population structure, diets and biogeographic relationships of a rocky intertidal fish assemblage in central Chile: high levels of herbivory in a temperate system. Bull. Mar. Sci. 47: 598-612.). The pelagic larvae change from omnivory to carnivory as they develop, preying mainly on nauplii and calanoid copepodites (Ochoa-Muñoz et al. 2013Ochoa-Muñoz M., Valenzuela C., Toledo S., et al. 2013. Feeding of a larval clinid fish in a microtidal estuary from southern Chile. Rev. Biol. Mar. Oceanogr. 48: 45-57.). Although there is some knowledge of the early life history of other blennoids of the region, e.g. blennoids, dactyloscopids and tripterigiids (Ciechomski 1975Ciechomski J. 1975. Características y distribución de post-larvas del acorazado Agonopsis chiloensis (Jenyns, 1842) Jordan y Evermann, 1898, y de Tripterygion cunninghami Smitt, 1898 en aguas del Atlántico frente a la Argentina (Pisces). Physis A 84: 309-317., Balbontín and Pérez 1979Balbontín F., Pérez R. 1979. Modalidad de postura, huevos y estados larvales de Hypsoblennius sordidus (Bennett) en la Bahía de Valparaíso (Blenniidae: Perciformes). Rev. Biol. Mar. 16: 311-318., Herrera et al. 2007Herrera G., Llanos-Rivera A., Landaeta M.F. 2007. Larvae of the sand stargazer Sindoscopus australis and notes on the development of Dactyloscopidae (Percifomes: Blennioidei). Zootaxa 1401: 63-68.), the larvae of M. viridis have not been described yet.

The classical distance-based morphometry (DBM) used in the first description of a larval fish development consists of a survey of measurements along the body (biometries), which are selected a priori by making assumptions about form and function. A way to understand the patterns of allometric growth and shape change is to utilize geometric morphometry techniques. Geometric morphometry deals directly with coordinates of anatomical landmarks, either in two or three dimensions, rather than with traditional distance or angle measurements (Zeldrich et al. 2004Zeldrich M., Swiderki D., Sheets H., et al. 2004. Geometric Morphometrics for Biologists: A Primer. San Diego, CA, Academic Press.). Landmark points have been defined by Bookstein (1991)Bookstein F. 1991. Morphometric tools for landmark data. Geometry and Biology. Cambridge Univ. Press, 455 p. as loci that have names as well as Cartesian coordinates. Ontogenetic changes associated with growth and development, where there is a clear directionality from young to older organisms, are examples of shape changes (Klingenberg 2013Klingenberg C. 2013. Visualizations in geometric morphometrics: how to read and how to make graphs showing shape changes. Hystrix 24: 15-24. ), understood as the movement from a referential point towards another representing the target shape. This means that it is a vector that has a direction and a magnitude (or length; Klingenberg and Monteiro 2005Klingenberg C., Monteiro L. 2005. Distances and directions in multidimensional shape spaces: implications for morphometric applications. Syst. Biol. 54: 678-688.).

Therefore, utilizing the larvae of the kelpfish Myxodes viridis as a model species, the ontogenetic changes of early stages of a marine fish with pelagic larval phase and benthic adult phase is described using a classical (DBM) approach and geometric morphometry techniques to quantify the shape changes. From an ecomorphological perspective, the development and shape change during early ontogeny of M. viridis will reflect the requirements for improving survival during the pelagic stage. Then, two plausible predictions arise: one, the head will show the largest shape change in order to increase feeding success, or two, the tail will show the greatest shape change in order to increase predator avoidance. The objective of this work is to describe the larval development and shape changes during the early ontogeny of the kelpfish Myxodes viridis using two complementary methodologies, in order to increase the taxonomic knowledge of this family in South America.

MATERIALS AND METHODSTop

Fieldwork

During the late austral winter and spring of 2010-2012, 13 dusk and nocturnal coastal surveys (1930 to 2300 h) were conducted in El Quisco Bay (33°24’S, 71°43’W) onboard an artisanal vessel. Plankton samples were obtained in oblique tows with a Bongo net (60 cm diameter, 300 μm mesh and two TSK flowmeters) performed in nearshore waters (<1 km distance from shore) for 15-20 min from a depth of 20 m. Filtered seawater ranged from 13.1 to 437.4 m³. The plankton samples (166) were preserved in 5% buffered (sodium borate) formalin and transferred to 96% ethanol after 12 h.

Laboratory work

In the laboratory, all fish larvae from the plankton samples were sorted, counted and identified to the lowest taxonomic level. Larval Myxodes viridis were identified by the series method (Ahlstrom and Ball 1954Ahlstrom E.H., Ball O.P. 1954. Description of eggs and larvae of jack mackerel (Trachurus symmetricus) and distribution and abundance of larvae in 1950 and 1951. Fish. Bull. U.S. 52: 209-245., Ahlstrom et al. 1976Ahlstrom E.H., Moser H.G., O’Toole M.J. 1976. Development and distribution of larvae and early juveniles of the commercial lanternfish, Lampanyctodes hectoris (Günther) off the west coast of southern Africa with a discussion of phylogenetic relationships of the genus. Bull. South. Cal. Acad. Sci. 75: 138-152., Neira et al. 1998Neira F., Miskiewicz A., Trnski T. 1998. Larvae of Temperate Australian Fishes. Laboratory guide for larval fish identification. Univ. Western Australia Press, Nedlands, 474 pp.). Measurements were made to the nearest 0.01 mm under an Olympus SZ-61 stereomicroscope using a Moticam 2500 (5.0 Mpixel) video camera connected to a PC with Moticam Image Plus 2.0 software. Measurements used here are defined by Moser (1996)Moser H. 1996. Introduction. In: Moser H (ed.), The early stages of fishes in the California Current region, pp. 1-72. Kansas City, KS: Allen Press., and correspond to SL, head length (HL), preanal length (PAL), snout length (SnL), eye diameter (ED), and body depth (BD) (Fig. 1). Pigments refer solely to melanophores.

The illustrations were done using a stereomicroscope with a camera lucida. Drawings were done by hand and scanned for image edition with GIMP 2.6.12-1ubuntu 1.3 software.

Cartilage and bone staining

To set meristic characters and observe the ossification process during the development, a double staining technique modified from Menegola et al. (2001)Menegola E., Broccia M., Giavini E. 2001. Atlas of rat fetal skeleton double stained for bone and cartilage. Teratology 64: 125-133. and Gillis et al. (2009)Gillis G., Randall D., Shubin N. 2009. Chondrogenesis and homology of the visceral skeleton in the little skate, Leucoraja erinacea (Chondrichthyes: Batoidea). J. Morph. 270: 628-643. was applied to eight larvae ranging from 4.69 mm to 23.09 mm SL. After re hydration in 70% ethanol, larvae were placed in the Alcian blue-Alizarin red S acid staining solution, then washed three times in 95% ethanol and placed in the Alizarin red S basic staining solution (Menegola et al. 2001Menegola E., Broccia M., Giavini E. 2001. Atlas of rat fetal skeleton double stained for bone and cartilage. Teratology 64: 125-133.). Then, larvae were treated with graded KOH/glycerol series (Gillis et al. 2009Gillis G., Randall D., Shubin N. 2009. Chondrogenesis and homology of the visceral skeleton in the little skate, Leucoraja erinacea (Chondrichthyes: Batoidea). J. Morph. 270: 628-643.) and finally stored in 100% glycerol at room temperature.

Geometric morphometrics

Selected individuals were photographed on the left side with an Olympus SZ61 stereomicroscope attached to a camera (Motic Moticam 2500, resolution 5.0 MPixel) using Motic Image Plus 2.0 software. To carry out the geometric morphometrics, photographed larvae were divided into four groups (preflexion, flexion, postflexion and transformation). A configuration of 13 landmarks was applied to study 67 larval kelpfish (Fig. 1), ranging from 4.00 to 22.22 mm BL. In the use of landmarks it is necessary to avoid bias produced by fixative procedures, so all specimens must keep the same anatomical position. Therefore, only well-preserved individuals with the same methodology were used for geometric morphometry. Subsequently, a database was created using the tpsUtil program (version 1.58, Rohlf 2006Rohlf F. 2006. TpsDig Ver. 2.10, Digitalized Landmarks and Outlines. Department of Ecology and Evolution, State Univ. New York: Stony Book.), and the landmarks were digitized with the tpsDig program (version 2.17, Rohlf 2006Rohlf F. 2006. TpsDig Ver. 2.10, Digitalized Landmarks and Outlines. Department of Ecology and Evolution, State Univ. New York: Stony Book.). The coordinates obtained for each individual were centred, scaled and rotated to obtain a consensus figure using the Generalized Procrustes Analysis (MorphoJ, version 1.05f, Klingenberg 2011Klingenberg C. 2011. MorphoJ: an integrated software package for geometric morphometrics. Mol. Ecol. Resour. 11: 353-357. ). These new coordinates were used for further statistical comparisons. Principal component analysis (PCA) was performed to identify the main axes of shape change and the specific changes in the larval body (reflected as movements of landmarks) that explain the variability of the data. To visualilze the shape changes, warping transformation grids were applied to the drawing only, without the grid, following Klingenberg (2013)Klingenberg C. 2013. Visualizations in geometric morphometrics: how to read and how to make graphs showing shape changes. Hystrix 24: 15-24. . Procrustes ANOVA was carried out to assess the relative amounts of variation among individuals and of measurement error using MorphoJ. In general, allometric changes between somatic parameters are expressed by a potential equation (see Lleonart et al. 2000Lleonart J., Salat J., Torres G.J. 2000. Removing allometric effects of body size in morphometrical analysis. J. Theor. Biol. 205: 85-93.); however, in geometric morphometry, allometry is widely characterized by multivariate regression of shape on size (usually centroid size or log-transformed centroid size); such regressions often fit the data well and the allometric shape changes tend to affect the entire structures under study (Loy et al. 1998Loy A., Mariani L., Bertelletti M., et al. 1998. Visualizing allometry: geometric morphometrics in the study of shape changes in the early stages of the two-banded sea bream, Diplodus vulgaris (Perciformes, Sparidae). J. Morphol. 237: 137-146., Mitteroecker et al. 2004Mitteroecker P., Gunz P., Bernhard M., et al. 2004. Comparison of cranial ontogenetic trajectories among great apes and humans. J. Hum. Evol. 46: 679-698., Klingenberg and Marugán-Lobón 2013Klingenberg C., Marugán-Lobón J. 2013. Evolutionary covariation in geometric morphometric data: Analyzing integration, modularity, and allometry in a phylogenetic context. Syst. Biol. 62: 591-610.). Therefore, a regression between a regression score of each independent variable (Procrustes coordinates) and centroid size was carried out with MorphoJ (Sidlaukas et al. 2011Sidlaukas B.L., Mol J.H., Vari R.P. 2011. Dealing with allometry in linear and geometric morphometrics: a taxonomic case study in the Leporinus cylindriformis group (Characiformes: Anostomidae) with description of a new species from Suriname. Zool. J. Linn. Soc. 162: 103-130.). The allometry corresponds to the proportion of variation for which the regression accounts as a percentage of the total variation.

RESULTSTop

A total of 214 larvae were used for the dynamic description, ranging from 3.51 to 23.09 mm SL (mean ± standard deviation; 6.91±2.85 mm). Late postflexion larvae of M. viridis can be distinguished from those of other co-occurring Clinidae and Labrisomidae by the number of spines and soft rays in the dorsal fin (D XXIV-XXVI, 6-7), longer preanal distance, and persistent melanophores over the gut throughout the larval development.

Dynamic description of larval development

Body

The body shape of Myxodes viridis is elongated throughout the larval development (BD 10-20% SL) (Table 1). Body Depth is 10.33% SL (±1.35) in preflexion larvae, and slightly larger in flexion and postflexion (BD 10.5±0.82% SL and BD 11.83±1.51% SL, respectively). Notochord flexion occurs between 6.1 mm and 7.6 mm SL. Preanal length (PAL) is moderate (between 30-50% SL) to long (between 50-70% SL) (Table 1). Preflexion larvae have a PAL of 49.62% of SL (±3.47), and slightly lower in flexion and postflexion larvae (PAL 48.56±2.9% SL and PAL 47.5±2.22% SL, respectively).

Internal organs

The foregut is coiled anteriorly. The midgut has striations from the end of the coil to the constriction prior to the rectum. The gas bladder is located over the midgut; it was present in all individuals and there were no signs of change in its size or position throughout the development. In juveniles, it is difficult to observe because it is covered by myomeres (Fig. 2).

Head

In preflexion larvae HL is 17.02% SL (±1.87%), qualifying as small. In flexion larvae HL is smaller, 16.84 (±1.69% SL), but increases considerably in postflexion (HL 19.74±2.19% SL), and over 20% SL in juveniles. SnL, as a proportion of HL, is very small in preflexion larvae (SnL 14.87±4.79% HL) but increases considerably in flexion (SnL 19.38±5.12% HL) and postflexion (SnL 23.5±3.16% HL). This is due to a change in the shape of the head, from rounded with a low mouth in yolk sac, preflexion and flexion larvae to pointed with a mouth placed higher and in a terminal position.

ED is 37.41% (±5.16%) of HL in preflexion larvae, 33.44 (±4.66%) in flexion and 26.88 (±3.00%) in postflexion. All studied larvae had pigmented eyes. Nostrils are formed in individuals larger than 18 mm SL. Single supraocular cirri were observed only in juveniles (Fig. 2F).

Fin formation sequence

Pectoral fins can be distinguished in larvae from 5 mm SL (Fig. 2B) as a simple bud and a lamina. Formation of pectoral fin rays occurs late in the development, as they were observed only in larvae larger than 14 mm SL. Notochord flexion and caudal fin formation were observed in individuals larger than 6.1 mm SL (Fig. 2C). The caudal fin is completely formed after 7.6 mm SL.

Dorsal and anal fins develop almost simultaneously in advanced postflexion stages, after 8.5 mm SL. The first pterygiophores form at the base of the posterior part of both fins. All rays (6-7 in the posterior part of the dorsal fin and 24-25 in the anal fin) and spines (XXXIV-XXXVI dorsal, II anal) appear in individuals larger than 13 mm SL. No signs of pelvic fin formation were apparent in larvae smaller than 16 mm SL.

Pigmentation

The larvae of M. viridis show little pigmentation throughout the development. Early preflexion larvae usually have few small ventral melanophores, one or two (rarely three or none) in the midgut, one larger and round under the foregut (anterior to the coil), and one small and branched in the constriction anterior to the rectum. Under the ventral margin of the tail, between the anus and the caudal peduncle, there is a series of four to six small round melanophores evenly spaced at five to six myomeres; these increase in number during development and reach one melanophore at the base of each pterygiophore in juveniles. In the middle of the tail of early larvae, even in yolk sac stages, a distinctive, large and branched melanophore forms on the ventral margin, with branches that extend over the sides of the body and in the fin fold. This melanophore decreases in size during development, and in larvae larger than 10 mm SL it becomes indistinguishable from the rest of the ventral series (Fig. 2F). This melanophore is similar to the one observed in the development of the labrisomid Labrisomus xanthi (Watson 1996Watson W. 1996. Clinidae. In: Moser H (ed.) The Early Stages of Fishes in the California Current Region. Atlas N°33. California Cooperative Oceanic Fisheries Investigations, Allen Press Inc., Kansas, USA, pp. 1164-1169.).

A large dendritic melanophore that develops dorsally on the gas bladder can be observed in larvae smaller than 9 mm SL as it becomes embedded during development in larger larvae. A single, usually enlongate melanophore develops on each side of the isthmus during early preflexion.

Chondrogenesis and ossification of larval kelpfish

The double staining technique showed positive results in seven stained larvae. In preflexion larvae, only the six pairs of branchial arches and part of the esophagus were stained in a light blue colour. In flexion and postflexion larvae, these same structures, along with the lower jaw, showed a darker blue colour (Fig. 3A-D). Flexion larvae showed the first signs of cartilaginous matrix in the forming hypural plates of the caudal fin. The sequence of formation started with hypural plates 1, 2 and 3 (Fig. 3E). In late flexion and postflexion larvae, it was possible to observe all five hypurals completely formed and stained dark blue, showing that these structures are solely formed by cartilage (Fig. 3F, G) Parahypurals and epurals were not stained and not distinguishable. In the juvenile individual, hypurals were not stained, and the urostyle is the only observable caudal structure in a strong pink tone (just like the last six spines of the dorsal fin), which indicates formed bony structures (Fig. 3H).

First signs of chondrogenesis in the vertebral column were the haemal arches from the 17th vertebra (Fig. 4B). In postflexion larva (Fig. 4C) it is possible to observe all neural and vertebral arches, formed by cartilaginous tissue. There were no observable signs of chondrogenesis in vertebrae of preflexion, flexion or postflexion larvae (Fig. 4A-C). Only the juvenile individual showed completely ossified vertebral centra and neural and haemal arches, whilst neural and haemal spines were halfway formed and ventral ribs were barely visible (Fig. 4D).

Coracoscapular cartilage is visible in late flexion and postflexion larvae, at the base of the pectoral fin (Fig. 3C-D). In the juvenile individual, this same structure has ossified and divided into scapula and coracoid bones, which form a base for the proximal radial bones (also completely ossified). Distal radial bones closer to the ventral side of the body show some degree of ossification. In this same individual, the basipterygium bones, that give base to the pelvic fin, can be observed (Fig. 4D).

Geometric morphometrics

Variability of the Principal Component 1 (PC=41.74%, Fig. 5) is explained mainly by early ontogenetic changes in the vertical axis of the body, caused by a vertical compression in the abdominal area and a bending downwards of both extremes of the body (tip of the snout and posterior extreme of the tail). This main change in shape may be attributable to ontogenetic enlargement of shape as well as the effects of fixatives (formaldehyde, ethanol). Variability in PC2 (17.57%) is explained by changes in the longitudinal axis of the body, triggered mainly by the elongation of the gut, and shortening of the head produced by a decrease in the distance between the tip of the snout and the opercle (Fig. 5). Finally, PC3 (8.36% of the variation of shape) accounts for an increase in the head height, a shortening of the mandible and a deepening of the body height at the level of the anus (Fig. 5).

The variance unexplained by individual effect in the Procrustes ANOVA is the measurement error, corresponding to 1.96% (Table 2). The regression of Procrustes coordinates and centroid size (Fig. 6) estimates an ontogenetic allometry of 14.69%. The permutation test against the null hypothesis of independence gives a P value of 0.0001.

According to these results, body shape changes smoothly, in a continuous gradient throughout the larval development of Myxodes viridis. Major changes were expressed in the head and snout and, to a lesser extent, in the body height and shortening of the gut.

DISCUSSIONTop

Preflexion larval kelpfish Myxodes viridis are likely to be confused with other co-occurring fish larvae in the coastal area of the southeastern Pacific Ocean, such as gobiid Ophiogobius jenynsi, triplefin Helcogrammoides chilensis and H. cunninghami, sand stargazer Sindoscopus australis, and several labrisomid species of the genera Auchenionchus, Calliclinus and Labrisomus (Ciechomski 1975Ciechomski J. 1975. Características y distribución de post-larvas del acorazado Agonopsis chiloensis (Jenyns, 1842) Jordan y Evermann, 1898, y de Tripterygion cunninghami Smitt, 1898 en aguas del Atlántico frente a la Argentina (Pisces). Physis A 84: 309-317., Hastings and Springer 2009Hastings P., Springer V. 2009. Systematics of the Blennioidei and the included families Chaenopsidae, Clinidae, Labrisomidae and Dactyloscopidae. In: Patzner R., Goncalves E., Hastings P., et al. (eds), The Biology of Blennies. Enfield, NH. Science Publishers, pp. 3-30., Herrera et al. 2007Herrera G., Llanos-Rivera A., Landaeta M.F. 2007. Larvae of the sand stargazer Sindoscopus australis and notes on the development of Dactyloscopidae (Percifomes: Blennioidei). Zootaxa 1401: 63-68.) (Table 3). The presence of a large ventral branched melanophore in the middle of the tail is a good character to distinguish preflexion and flexion larvae of M. viridis from the larvae of most of the other species. The larvae of two species also have a rather similar feature, the gobiid O. jenynsi and the labrisomid A. crinitus. The larvae of O. jenynsi have conspicuous melanophores along the ventral margin of the tail (Herrera 1984Herrera G. 1984. Descripción de estados post-embrionales de Ophiogobius jenynsi Hoese 1976 (Gobiidae, Blenniioidei). Rev. Biol. Mar. 20: 159-168.); however, these differ in the extension and branching pattern from those of M. viridis. Additionally, the larvae of O. jenynsi have a longer PAL (>55%). The larvae of A. crinitus can be distinguished from those of M. viridis by having a smaller PAL, lower myomere numbers (13+29-30 vs. 16-18+31-34), more pectoral rays numbers (13 vs. 11), and two prominent stellate melanophores over the head that are absent in larval M. viridis (Landaeta, unpublished data). The postflexion larvae of other co-occurring Blenniioid species seem rather similar to M. viridis, but they can be easily identified based on a shorter PAL, 27-31% SL in Sindoscopus australis (Herrera et al. 2007Herrera G., Llanos-Rivera A., Landaeta M.F. 2007. Larvae of the sand stargazer Sindoscopus australis and notes on the development of Dactyloscopidae (Percifomes: Blennioidei). Zootaxa 1401: 63-68.), and 30-38% SL in Helcogrammoides chilensis (Pérez 1979Pérez R. 1979. Desarrollo postembrionario de Tripterygion chilensis Cancino 1955, en la Bahía de Valparaíso (Tripterygiidae: Perciformes). Rev. Biol. Mar. 16: 319-329.).

Within the suborder Blennioidei, larvae are unspecialized and show few adaptations, morphology or pigmentation for pelagic life (Herrera and Lavenberg 1999Herrera G., Lavenberg R. 1999. Larval Labrisomidae (Pisces: Blennioidei) from the Galápagos Islands. Contrib. Sci. 48: 1-14), except for the presence of a gas bladder, absent in adult stages. The family Clinidae has been recognized as a natural group (Stepien et al. 1997Stepien C., Dillon A., Brooks M., et al. 1997. The Evolution of Blennioid Fishes Based on an Analysis of Mitochondrial 12S rDNA In: Kocher T., Stepien C. (ed.) Molecular Systematics of Fishes, Academic Press, London, pp. 245-270., Lin 2009Lin H. 2009. Evolution of the suborder Blennioidei: phylogeny and phylogeography of a shallow water fish clade. Ph.D. thesis, Univ. California, San Diego, 150 pp.) that is related to Labrisomidae, Chaenopsidae and Dactyloscopidae. Relationships among the latter two are not fully resolved (Lin 2009Lin H. 2009. Evolution of the suborder Blennioidei: phylogeny and phylogeography of a shallow water fish clade. Ph.D. thesis, Univ. California, San Diego, 150 pp., Betancur et al. 2013Betancur-R.R., Broughton R.E., Wiley E.O., et al. 2013. The Tree of Life and a new classification of bony fishes. PLoS Curr. 18: 5.). The larvae of Clinidae, Labrisomidae, Chaenopsidae and Dactyloscopidae have elongated and relatively compressed bodies throughout the early stages, a rounded and comparatively small head, and short snouts. A common feature observed in the known larvae of the families is the presence of a pair of melanophores that develop below the pectoral fins in yolk sac larvae, at the level of the liver. The fate of these varies, as they become elongate in chaenopsids and fuse in the ventral midline under the basipterygium (e.g. Okiyama 1988Okiyama M. 1988. An atlas of the early stage fishes in Japan. Tokai Univ. Press, Tokio. 1154 pp., Brogan 1992Brogan M. 1992. Ecology of larval fishes around reefs in the Gulf of California, México. PhD. thesis, Univ. Arizona, 161 p., Herrera and Lavenberg 2002Herrera G.A., Lavenberg R.J. 2002. Larval blennies from the Galapagos and Cocos Islands: families Tripterygiidae, Dactyloscopidae, and Chaenopsidae (Perciformes, Blennioidei). Contrib. Sci. 488: 1-15. ). Adittionally, clinids can be distinguished by a longer preanal length (clinids, >45% SL; labrisomids, 35-44% SL) (Watson 1996Watson W. 1996. Clinidae. In: Moser H (ed.) The Early Stages of Fishes in the California Current Region. Atlas N°33. California Cooperative Oceanic Fisheries Investigations, Allen Press Inc., Kansas, USA, pp. 1164-1169.).

Among the Blennioidei, larvae are unspecialized and show few adaptations, morphology or pigmentation for pelagic life (Herrera and Lavenberg 1999Herrera G., Lavenberg R. 1999. Larval Labrisomidae (Pisces: Blennioidei) from the Galápagos Islands. Contrib. Sci. 48: 1-14). A common feature of the larvae of the three above-mentioned families is the presence of a midlateral melanophore under the pectoral fin base, which moves ventrad and cephalad to near the ventral midline during development.

Both clinid and labrisomid larvae, throughout the early stages, have elongated and relatively laterally compressed bodies, with rounded and comparatively small head and snout length, although clinids can be distinguished for a longer preanal length (clinids: >45%SL; labrisomids: 35-44%SL) (Watson 1996Watson W. 1996. Clinidae. In: Moser H (ed.) The Early Stages of Fishes in the California Current Region. Atlas N°33. California Cooperative Oceanic Fisheries Investigations, Allen Press Inc., Kansas, USA, pp. 1164-1169.).

The larvae of M. viridis show unusual developmental features within the Clinidae, and the Blennioidei. The pelvic fin develops at a comparatively larger size, as they were not observed in larvae (<16 mm) but only in juveniles (>20 mm). Furthermore, the dorsal and anal fin rays appear at larger sizes. In M. viridis, they form after 9.5 mm SL, whereas in other Blennoidei they are usually formed at smaller sizes (e.g. Watson 1996Watson W. 1996. Clinidae. In: Moser H (ed.) The Early Stages of Fishes in the California Current Region. Atlas N°33. California Cooperative Oceanic Fisheries Investigations, Allen Press Inc., Kansas, USA, pp. 1164-1169., Herrera and Lavenberg 2002Herrera G.A., Lavenberg R.J. 2002. Larval blennies from the Galapagos and Cocos Islands: families Tripterygiidae, Dactyloscopidae, and Chaenopsidae (Perciformes, Blennioidei). Contrib. Sci. 488: 1-15. ).

It is assumed that larvae of M. viridis start their ossification process after hatching, since the smallest observed larva (4.69 mm SL) only showed faint signs of cartilage formation on the gill arches. Skull structures in general are the first to develop, since they sustain structures that are critical for feeding, respiratory and neurological functions (Koumoundouros et al. 2001Koumoundouros G., Divanach P., Kentouri M. 2001. Osteological development of Dentex dentex (Osteichthyes, Sparidae): dorsal, anal, paired fins and squamation. Mar. Biol. 138: 399-406.). The early formation of cartilage in the coracoscapular bone may suggest that the pectoral fin fold provides stability or some degree of directionality in the trajectory of the larvae in the water column (Ott et al. 2009Ott A., Löffler J., Ahnelt H., et al. 2009. Early development of the postcranial skeleton of the Pikeperch Sander lucioperca (Teleostei: Percidae) relating to developmental stages and growth. J. Morph. 273: 894-908).

Larval duration for Myxodes viridis was recently estimated based on transition marks in the otolith, and ranged from 69 to 118 days (Mansur et al. 2014Mansur L., Plaza G., Landaeta M., et al. 2014. Planktonic duration in fourteen species of intertidal rocky fishes from the south-eastern Pacific Ocean. Mar. Freshw. Res. 65: 901-909.). The giant kelpfish Heterostichus rostratus has a pelagic larval duration (PLD) of around two months (Stepien 1986Stepien C. 1986. Life history and larval development of the giant kelpfish, Heterostichus rostratus Girard, 1854. Fish. Bull. 84(4): 809-826.). Other species from temperate rocky reefs of central Chile have a PLD of two to four months (Plaza et al. 2013Plaza G., Landaeta M., Espinoza C., et al. 2013. Daily growth patterns of six species of young-of-the-year of Chilean intertidal fishes. J. Mar. Biol. Ass. UK 93: 389-395.).

The effects induced on fishes by fixation and preservation vary among species (Tucker and Chester 1984Tucker J., Chester A.J. 1984. Effects of salinity, formalin concentration and buffer on quality of preservation of southern flounder (Paralichthys lethostigma) larvae. Copeia 1984: 981-988., Sagnes 1997Sagnes P. 1997. Potential artefacts in morphometric analyses of fish: effects of formalin preservation on 0+ grayling. J. Fish. Biol. 50: 910-914.) and influence the results of geometric morphometric analysis of adult (Martínez et al. 2013Martínez P., Berbel-Filho W., Jacobina U. 2013. Is formalin fixation and ethanol preservation able to influence in geometric morphometric analysis? Fishes as a case of study. Zoomorphology 132: 87-93.) as well as larval stages (Nikolakakis et al. 2014Nikolakakis S., Bossier P., Kanlis G., et al. 2014. Protocol for quantitative shape analysis of deformities in early larval European seabass Dicentrarchus labrax. J. Fish. Biol. 84: 206-224.). The samples utilized for geometric morphometrics in this study were initially fixed in formaldehyde and transferred to 96% ethanol after 12 h. The major changes in the shape of M. viridis during the larval development were focused mainly in the head, in addition to a shortening in the preanal length. These changes, particularly the bending of the extremes of the body, may be due to formalin-ethanol fixatives (C.P. Klingenberg, pers. comm.). Although this methodology has few effect on fish larvae shape (Nikolakakis et al. 2014Nikolakakis S., Bossier P., Kanlis G., et al. 2014. Protocol for quantitative shape analysis of deformities in early larval European seabass Dicentrarchus labrax. J. Fish. Biol. 84: 206-224.), it is important to carry out new studies about the effects of fixatives in larval fish shape, and/or to study the shape change with anaesthetized larvae and juveniles.

Also, the shape changes coincided with variations in the diet of larvae from preflexion to postflexion (change from omnivory to exclusive carnivory, Ochoa-Muñoz et al. 2013Ochoa-Muñoz M., Valenzuela C., Toledo S., et al. 2013. Feeding of a larval clinid fish in a microtidal estuary from southern Chile. Rev. Biol. Mar. Oceanogr. 48: 45-57.), and may be considered as a preparation for adult feeding tactics in a benthic habitat. Adults use their pointed head to search and capture prey under rocks and boulders (Landaeta, pers. obs.). Also, the subtle changes in the body shape throughout larval development, with a slow development of the tail and caudal rays, suggest that the ability to swim early in life is not well developed for Myxodes (Fuiman 1983Fuiman L. 1983. Growth gradients in fish larvae. J. Fish Biol. 23: 117-123. ). The Fuiman method for estimating growth gradients in fish larvae uses the eye diameter as standard, so it does not consider the whole shape of the larvae in order to calculate the allometric growth. Although the allometry is classically estimated from a power function (Huxley 1932Huxley J.S. 1932. Problems of Relative Growth. London: Methuen and Co. 312 p.), in geometric morphometry the allometry is better explained by a regression between Procrustes coordinates and centroid size (Klingenberg and Marugán-Lobón 2013Klingenberg C., Marugán-Lobón J. 2013. Evolutionary covariation in geometric morphometric data: Analyzing integration, modularity, and allometry in a phylogenetic context. Syst. Biol. 62: 591-610.).

Similarly, the blenniid Ecsenius strictus show slow critical swimming speeds (U-critic, 5.5±2.2 cm s^–1) compared with those of other coral reef larvae (>30 cm s^–1, Fisher et al. 2005Fisher R., Leis J.M., Clark D.L., et al. 2005. Critical swimming speeds of late-stage coral reef fish larvae: variation within species, among species and between locations. Mar. Biol. 147: 1201-1212.). Additionally, allometric growth, estimated by the regression of the Procrustes coordinates and centroid size, accounted for a small portion of the shape change (less than 15% of variation). This finding suggests that the final shape (juvenile) is reached early in the development, and most changes occur during the preflexion and flexion stages. Therefore, we conclude that early shape changes during the development of the kelpfish Myxodes viridis are aimed at an increase in feeding success (because of the modifications in the head shape) and probably an increase in the digestion capabilities (due to a shortening of the gut when larvae are mostly carnivorous), rather than an increase in predator avoidance (through the development of the tail and caudal bones).

ACKNOWLEDGEMENTSTop

We appreciate the help and comments of Prof. Fernando Balbontín (Universidad de Valparaíso). Also, we thank Dr. C.P. Klingenberg (University of Manchester) for comments and suggestions, during the course “Integration and Modularity with Geometric Morphometrics”, organized by Transmitting Science. Comment of two reviewers improved an early version of the manuscript. This study was partially funded by the grants of the projects Fondecyt 1100424 to Dr. F. Patricio Ojeda (Pontificia Universidad Católica de Chile), Fondecyt 1120868 to Dr. Gabriela Muñoz (Universidad de Valparaíso) and INACH RT_04-13 to MFL.

REFERENCESTop

Ahlstrom E.H., Ball O.P. 1954. Description of eggs and larvae of jack mackerel (Trachurus symmetricus) and distribution and abundance of larvae in 1950 and 1951. Fish. Bull. U.S. 52: 209-245.

Ahlstrom E.H., Moser H.G., O’Toole M.J. 1976. Development and distribution of larvae and early juveniles of the commercial lanternfish, Lampanyctodes hectoris (Günther) off the west coast of southern Africa with a discussion of phylogenetic relationships of the genus. Bull. South. Cal. Acad. Sci. 75: 138-152.

Balbontín F., Pérez R. 1979. Modalidad de postura, huevos y estados larvales de Hypsoblennius sordidus (Bennett) en la Bahía de Valparaíso (Blenniidae: Perciformes). Rev. Biol. Mar. 16: 311-318.

Betancur-R.R., Broughton R.E., Wiley E.O., et al. 2013. The Tree of Life and a new classification of bony fishes. PLoS Curr. 18: 5.
http://dx.doi.org/10.1371/currents.tol.53ba26640df0ccaee75bb165c8c26288

Bookstein F. 1991. Morphometric tools for landmark data. Geometry and Biology. Cambridge Univ. Press, 455 p.

Brogan M. 1992. Ecology of larval fishes around reefs in the Gulf of California, México. PhD. thesis, Univ. Arizona, 161 p.

Castillo R.M., Pequeño G. 1998. Sinopsis de Tripterygiidae de Chile (Osteichthyes: Perciformes). Gayana Zool. 62: 109-133.

Ciechomski J. 1975. Características y distribución de post-larvas del acorazado Agonopsis chiloensis (Jenyns, 1842) Jordan y Evermann, 1898, y de Tripterygion cunninghami Smitt, 1898 en aguas del Atlántico frente a la Argentina (Pisces). Physis A 84: 309-317.

Fisher R., Leis J.M., Clark D.L., et al. 2005. Critical swimming speeds of late-stage coral reef fish larvae: variation within species, among species and between locations. Mar. Biol. 147: 1201-1212.
http://dx.doi.org/10.1007/s00227-005-0001-x

Frédérich B., Adriaens D., Vandewalle P. 2008. Ontogenetic shape changes in Pomacentridae (Teleostei, Perciformes) and their relationships with feeding strategies: a geometric morphometric approach. Biol. J. Linn. Soc. 95: 92-105.
http://dx.doi.org/10.1111/j.1095-8312.2008.01003.x

Frédérich B., Colleye O., Lepoint G., et al. 2012. Mismatch between shape changes and ecological shifts during the post-settlement growth of the surgeonfish, Acanthurus triostegus. Front. Zool. 9: 8.
http://dx.doi.org/10.1186/1742-9994-9-8

Fuiman L. 1983. Growth gradients in fish larvae. J. Fish Biol. 23: 117-123.
http://dx.doi.org/10.1111/j.1095-8649.1983.tb02886.x

George A., Springer V. 1980. Revision of the clinid fish tribe Ophiclinini, including five new species, and definition the family Clinidae. Smithson. Contrib. Zool. 307: 1-30.
http://dx.doi.org/10.5479/si.00810282.307

Gillis G., Randall D., Shubin N. 2009. Chondrogenesis and homology of the visceral skeleton in the little skate, Leucoraja erinacea (Chondrichthyes: Batoidea). J. Morph. 270: 628-643.
http://dx.doi.org/10.1002/jmor.10710

Hastings P., Springer V. 2009. Systematics of the Blennioidei and the included families Chaenopsidae, Clinidae, Labrisomidae and Dactyloscopidae. In: Patzner R., Goncalves E., Hastings P., et al. (eds), The Biology of Blennies. Enfield, NH. Science Publishers, pp. 3-30.
http://dx.doi.org/10.1201/b10301-3

Herrera G. 1984. Descripción de estados post-embrionales de Ophiogobius jenynsi Hoese 1976 (Gobiidae, Blenniioidei). Rev. Biol. Mar. 20: 159-168.

Herrera G.A., Lavenberg R.J. 1999. Larval Labrisomidae (Pisces: Blennioidei) from the Galápagos Islands. Contrib. Sci. 478: 1-14.
http://www.nhm.org/site/sites/default/files/pdf/contrib_science/CS478.pdf

Herrera G.A., Lavenberg R.J. 2002. Larval blennies from the Galapagos and Cocos Islands: families Tripterygiidae, Dactyloscopidae, and Chaenopsidae (Perciformes, Blennioidei). Contrib. Sci. 488: 1-15.
http://www.nhm.org/site/sites/default/files/pdf/contrib_science/CS488.pdf

Herrera G., Llanos-Rivera A., Landaeta M.F. 2007. Larvae of the sand stargazer Sindoscopus australis and notes on the development of Dactyloscopidae (Percifomes: Blennioidei). Zootaxa 1401: 63-68.

Huxley J.S. 1932. Problems of Relative Growth. London: Methuen and Co. 312 p.

Klingenberg C. 2011. MorphoJ: an integrated software package for geometric morphometrics. Mol. Ecol. Resour. 11: 353-357.
http://dx.doi.org/10.1111/j.1755-0998.2010.02924.x

Klingenberg C. 2013. Visualizations in geometric morphometrics: how to read and how to make graphs showing shape changes. Hystrix 24: 15-24.

Klingenberg C., Monteiro L. 2005. Distances and directions in multidimensional shape spaces: implications for morphometric applications. Syst. Biol. 54: 678-688.
http://dx.doi.org/10.1080/10635150590947258

Klingenberg C., Marugán-Lobón J. 2013. Evolutionary covariation in geometric morphometric data: Analyzing integration, modularity, and allometry in a phylogenetic context. Syst. Biol. 62: 591-610.
http://dx.doi.org/10.1093/sysbio/syt025

Koumoundouros G., Divanach P., Kentouri M. 2001. Osteological development of Dentex dentex (Osteichthyes, Sparidae): dorsal, anal, paired fins and squamation. Mar. Biol. 138: 399-406.
http://dx.doi.org/10.1007/s002270000460

Kouttouki S., Georgakopoulou E., Kaspiris P., et al. 2006. Shape ontogeny and variation in the sharpsnout seabream, Diplodus puntazzo (Cetti 1777). Aqua. Res. 37: 655-663.
http://dx.doi.org/10.1111/j.1365-2109.2006.01475.x

Lin H. 2009. Evolution of the suborder Blennioidei: phylogeny and phylogeography of a shallow water fish clade. Ph.D. thesis, Univ. California, San Diego, 150 pp.

Lleonart J., Salat J., Torres G.J. 2000. Removing allometric effects of body size in morphometrical analysis. J. Theor. Biol. 205: 85-93.
http://dx.doi.org/10.1006/jtbi.2000.2043

Loy A., Mariani L., Bertelletti M., et al. 1998. Visualizing allometry: geometric morphometrics in the study of shape changes in the early stages of the two-banded sea bream, Diplodus vulgaris (Perciformes, Sparidae). J. Morphol. 237: 137-146.
http://dx.doi.org/10.1002/(SICI)1097-4687(199808)237:2<137::AID-JMOR5>3.0.CO;2-Z

Mansur L., Plaza G., Landaeta M., et al. 2014. Planktonic duration in fourteen species of intertidal rocky fishes from the south-eastern Pacific Ocean. Mar. Freshw. Res. 65: 901-909.
http://dx.doi.org/10.1071/MF13064

Martínez P., Berbel-Filho W., Jacobina U. 2013. Is formalin fixation and ethanol preservation able to influence in geometric morphometric analysis? Fishes as a case of study. Zoomorphology 132: 87-93.
http://dx.doi.org/10.1007/s00435-012-0176-x

Menegola E., Broccia M., Giavini E. 2001. Atlas of rat fetal skeleton double stained for bone and cartilage. Teratology 64: 125-133.
http://dx.doi.org/10.1002/tera.1055

Mitteroecker P., Gunz P., Bernhard M., et al. 2004. Comparison of cranial ontogenetic trajectories among great apes and humans. J. Hum. Evol. 46: 679-698.
http://dx.doi.org/10.1016/j.jhevol.2004.03.006

Moser H. 1996. Introduction. In: Moser H (ed.), The early stages of fishes in the California Current region, pp. 1-72. Kansas City, KS: Allen Press.

Neira F., Miskiewicz A., Trnski T. 1998. Larvae of Temperate Australian Fishes. Laboratory guide for larval fish identification. Univ. Western Australia Press, Nedlands, 474 pp.

Nikolakakis S., Bossier P., Kanlis G., et al. 2014. Protocol for quantitative shape analysis of deformities in early larval European seabass Dicentrarchus labrax. J. Fish. Biol. 84: 206-224.
http://dx.doi.org/10.1111/jfb.12284

Ochoa-Muñoz M., Valenzuela C., Toledo S., et al. 2013. Feeding of a larval clinid fish in a microtidal estuary from southern Chile. Rev. Biol. Mar. Oceanogr. 48: 45-57.
http://dx.doi.org/10.4067/S0718-19572013000100005

Okiyama M. 1988. An atlas of the early stage fishes in Japan. Tokai Univ. Press, Tokio. 1154 pp.

Ott A., Löffler J., Ahnelt H., et al. 2009. Early development of the postcranial skeleton of the Pikeperch Sander lucioperca (Teleostei: Percidae) relating to developmental stages and growth. J. Morph. 273: 894-908
http://dx.doi.org/10.1002/jmor.20029

Pequeño G., Lamilla J., Lloris D., et al. 1995. Comparación entre las ictiofaunas intermareales de los extremos austral y boreal de los canales patagónicos. Rev. Biol. Mar. 30(2): 155-177.

Pérez R. 1979. Desarrollo postembrionario de Tripterygion chilensis Cancino 1955, en la Bahía de Valparaíso (Tripterygiidae: Perciformes). Rev. Biol. Mar. 16: 319-329.

Plaza G., Landaeta M., Espinoza C., et al. 2013. Daily growth patterns of six species of young-of-the-year of Chilean intertidal fishes. J. Mar. Biol. Ass. UK 93: 389-395.
http://dx.doi.org/10.1017/S0025315412000859

Rohlf F. 2006. TpsDig Ver. 2.10, Digitalized Landmarks and Outlines. Department of Ecology and Evolution, State Univ. New York: Stony Book.

Russo T., Costa C., Cataudella S. 2007. Correspondence between shape and feeding habit changes throughout ontogeny of gilthead sea bream Sparus aurata L., 1758. J. Fish. Biol. 71: 629-656.
http://dx.doi.org/10.1111/j.1095-8649.2007.01528.x

Russo T., Pulcini D., Bruner E., et al. 2009. Shape and size variation: growth and development of the dusky grouper (Epinephelus marginatus Lowe, 1834). J. Morph. 270: 83-96.
http://dx.doi.org/10.1002/jmor.10674

Sagnes P. 1997. Potential artefacts in morphometric analyses of fish: effects of formalin preservation on 0+ grayling. J. Fish. Biol. 50: 910-914.
http://dx.doi.org/10.1111/j.1095-8649.1997.tb01986.x

Sidlaukas B.L., Mol J.H., Vari R.P. 2011. Dealing with allometry in linear and geometric morphometrics: a taxonomic case study in the Leporinus cylindriformis group (Characiformes: Anostomidae) with description of a new species from Suriname. Zool. J. Linn. Soc. 162: 103-130.
http://dx.doi.org/10.1111/j.1096-3642.2010.00677.x

Stephens J., Springer V. 1973. Clinid fishes of Chile and Peru, with description of a new species, Myxodes ornatus from Chile. Smithson. Contrib. Zool. 159: 1-24.
http://dx.doi.org/10.5479/si.00810282.159

Stepien C. 1986. Life history and larval development of the giant kelpfish, Heterostichus rostratus Girard, 1854. Fish. Bull. 84(4): 809-826.

Stepien C. 1990. Population structure, diets and biogeographic relationships of a rocky intertidal fish assemblage in central Chile: high levels of herbivory in a temperate system. Bull. Mar. Sci. 47: 598-612.

Stepien C. 1992. Evolution and biogeography of the Clinidae (Teleostei: Blennioidei). Copeia 1992(2): 375-392.
http://dx.doi.org/10.2307/1446198

Stepien C., Dillon A., Brooks M., et al. 1997. The Evolution of Blennioid Fishes Based on an Analysis of Mitochondrial 12S rDNA In: Kocher T., Stepien C. (ed.) Molecular Systematics of Fishes, Academic Press, London, pp. 245-270.
http://dx.doi.org/10.1016/B978-012417540-2/50016-6

Tucker J., Chester A.J. 1984. Effects of salinity, formalin concentration and buffer on quality of preservation of southern flounder (Paralichthys lethostigma) larvae. Copeia 1984: 981-988.
http://dx.doi.org/10.2307/1445343

Watson W. 1996. Clinidae. In: Moser H (ed.) The Early Stages of Fishes in the California Current Region. Atlas N°33. California Cooperative Oceanic Fisheries Investigations, Allen Press Inc., Kansas, USA, pp. 1164-1169.

Zeldrich M., Swiderki D., Sheets H., et al. 2004. Geometric Morphometrics for Biologists: A Primer. San Diego, CA, Academic Press.