Early life history of two Neotropical Triportheidae fish (Characiformes)

Ruineris Almada Cajado^1,2 , Fabíola Katrine Souza Silva^1,3, Lucas Silva Oliveira^1,4, Zaqueu dos Santos^1,3 Andréa Bialetzki⁵ and Diego Maia Zacardi^1,3

PDF: EN    XML: EN | Cite this article

Abstract​

---

Introduction​

---

The Triportheus Cope, 1872, popularly known as narrow hatchet fish and elongate hatchet fish, comprises small teleost fish exclusive to the Neotropical region (Malabarba, 2004). Species of this genus can be easily distinguished by the presence of long pectoral fins, expanded coracoid bones, and lateral line with a pronounced postventral slope close to the pectoral fin (Van der Sleen, Zanata, 2018). They are pelagic fish, perform medium-distance migrations for reproductive purposes, usually between September and October, and have omnivorous feeding habits (Doria, Queiroz, 2008; García-Dávila et al., 2018; Silvano et al., 2020).

In the Amazon River and its tributaries are recorded eight species of this genus– Triportheus albus Cope, 1872, T. angulatus (Spix & Agassiz, 1829), T. auritus (Valenciennes, 1850), T. brachipomus (Valenciennes, 1850), T. culter (Cope, 1872), T. curtus (Garman, 1890), T. pictus (Garman, 1890), and T. rotundatus (Jardine, 1841) (Dagosta, Pinna, 2019). Among these species, T. auritus, T. albus,and T. angulatus stand out for their high abundance, being frequently recorded in several biotopes in the white, clear, and black waters of the Amazon basin (Araújo et al., 2017; Imbiriba et al., 2020; Silvano et al., 2020). These species are widely used as a source of protein by riverine populations in subsistence fishing along the Solimões-Amazonas system channel and its tributaries (Batista et al., 2012; Isaac et al., 2016; Zacardi, 2020). Additionally, they are becoming increasingly common in regional markets as an alternative to meet the demand for fish despite the accelerated decline of other widely exploited species (Batista et al., 2012; Ferraz, Barthem, 2016; Isaac et al., 2016).

Despite the ecological relevance and economic potential of these species, information about their phenotype during early ontogeny is incipient, with just over 22% of the 18 species of Triportheus having their embryonic and/or larval stages described (Reynalte-Tataje et al., 2020; Fricke et al., 2022). In the Amazon basin, only T. auritus had its initial development characterized (Cajado et al., 2021), while Oldani (1979) and Garcia et al. (2016) presented a succinct and incomplete description of the early life stages of T. paranensis (both currently = T. nematurus) and T. angulatus (T. nematurus/T. signatus sensu Lopes, 2020) in the Paraná River basin.

Integrative approaches that describe the morphological, meristic, and morphometric characters of the different ontogenetic phases are fundamental for the accurate identification of the initial stages of fish development and an important step for studies on ichthyoplankton ecology and fisheries biology (Reynalte-Tataje et al., 2020; Zacardi et al., 2020a,b). The difficulty in identifying the material collected in nature is one of the larger obstacles encountered in ichthyoplankton studies, due to the intense interspecific similarity of fish during the initial life cycle (Cajado et al., 2021; Lima et al., 2021; Silva et al., 2021; Oliveira et al., 2022).

Here, we intend to expand the basic knowledge of the early ontogeny of Neotropical fish, through the detailed description of larvae and juveniles of T. angulatus and T. albus that occur in the Amazon basin. This study was based on morphology, meristic counts and morphometric measurements, and estimated the growth patterns throughout the early development of the two species. Furthermore, we propose an identification key for these two species and other congeners that occur in sympatry approaching the early ontogeny.

Material and methods

---

Collection of biological material. The larvae and juveniles of fish used in this study came from sampling carried out in the middle stretch of the Solimões River and lower Japurá River, around the Mamirauá Sustainable Development Reserve (Reserva de Desenvolvimento Sustentável de Mamirauá) – MSDR (03º08’S 64º45’W and 02º36’S 67º13’W) in the state of Amazonas. Also, in the lower Amazon River in the state of Pará: (1) open water limnetic zone (02º28’42”S 54º38’04”W); (2) under macrophyte stands in lakes (02º26’44”S 54º16’53”W) and (3) lake channels in floodplain areas (02º12’41”S 54º45’42”W and 02º18’52”S 54º43’11”W). In addition to collections along the lower and middle stretch of the Tapajós River, Pará, Brazil (02º28’46”S 55º 04’34”W and 05º32’47”S 57º05’35”W) and in different periods (Rise, flood, drought e fall) (Fig. 1).

FIGURE 1| Map of the Eastern Amazon, showing the capture sites of the Triportheus used in the study. MSDR: Mamirauá Sustainable Development Reserve.

The larvae and juveniles were collected for 12 years, between 2010 and 2022, by horizontal trawling in the subsurface of the water column (open water) using a conical-cylindrical plankton net (0,3 mm mesh). For sampling in macrophyte stands floating was used sieve fish net (0,5 mm mesh) of 1.0 x 1.5 x 1.0 m, while for sampling near sandbanks and marginal areas of rivers and lakes was used a seine net (1 mm mesh) of 5 m x 1.5 m.

Analysis of biological material. After capture, the specimens were euthanized with benzocaine (250 mg/L) and fixed in a 10% formalin solution buffered with calcium carbonate. In the laboratory, Triportheus larvae and juveniles were sorted, separating them from plant material and plankton, then identified at the species level, using the regressive developmental sequence technique from the known juveniles to the smallest individuals, as suggested by Nakatani et al. (2001). The juveniles were identified using specific literature (Malabarba, 2004). After identification, the specimens were classified according to their degree of development as proposed by Ahlstrom et al. (1976) and modified by Nakatani et al. (2001) in larval (preflexion, flexion, and postflexion stages) and juvenile periods.

The description of the larval and juvenile periods was based on the observation of the main morphological events and the degree of initial development, in addition to meristic and morphometric characters, and the individuals that best represented the characteristics of the species were photographed and illustrated. It is noteworthy that the choice of the best individuals was based on the integrity of the specimen, considering the state of the fins, body shape, and the presence or absence of pigmentation (in this case, we chose the individuals that best represented the patterns observed for the species at a certain stage of development). Moreover, voucher specimens were deposited in the Coleção de Referência de Ovos e Larvas de Peixes, Laboratório de Ecologia do Ictioplâncton e Pesca em Águas Interiores (CROLP LEIPAI), Universidade Federal do Oeste do Pará (https://specieslink.net/col/CROLP-LEIPAI/), LEIPAI 397 to LEIPAI 431 (T. angulatus), and LEIPAI 432 to LEIPAI 459 (T. albus).

Data analysis. Morphometric measurements (expressed in mm) were performed using a binocular stereomicroscope (Leica S9i) coupled to an integrated digital color camera for image capture and analysis – software (Leica LAS EZ). Eleven morphometric characters were measured (Ahlstrom et al., 1976): head depth (HD), body depth (BD), head length (HL), snout length (SNL), standard length (SL), eye diameter (ED) and the length from the snout to the origins of anal (SNA), dorsal (SND), pectoral (SNP), and pelvic (SNV) fins. Additionally, the depth of the body toward the anus (BDA) was measured. Notably for SND and SNA in the preflexion stage, the distance from the snout to the beginning of the embryonic dorsal fin and the beginning of this membrane after the anus were considered, respectively. For meristic characterization, we counted the preanal, postanal and total number of myomeres, and the number of unbranched rays and/or branched rays present in anal (A), dorsal (D), pectoral (P), and pelvic (V) fins.

For the analysis of morphometric relationships of larvae and juveniles (expressed as a percentage), the variables HD, SNL, and ED were related to HL, while BD, BDA, HL, SNA, SND, SNP, and SNV were related to SL. Body relationships for BD (BD/SL), HL (HL/SL), and ED (ED/HL) were established using the criteria suggested by Leis, Trnski (1989). Additionally, the proportions of HD in relation to BDA and BD were explored to better understand the morphometric relationships throughout development.

To evaluate body growth patterns, regression models were used in which the morphometric variables (dependent), except BDA, were plotted in relation to SL and HL (independent variables). These relationships were described by different growth models, which may indicate relevant biological processes linked to early ontogeny (Kováč et al., 1999). The hypothesis of continuous isometric growth was tested using a simple linear regression model. Two alternative developmental hypotheses were also tested: gradual allometric growth (quadratic regression) and discontinuous isometric growth (piecewise linear regression – characterized by breakpoints that highlight divergent growth rates). The selection of the best model for each analyzed relationship was based on the F test, with a significance level of p < 0.05.

The statistical significance of differences between morphometric variables was assessed using an Analysis of Covariance (ANCOVA), were HD, SNL, ED, BD, HL, SND, SNA, SNV, and SNP were response variables and SL, and HL were covariates. Before performing the ANCOVA, the data were transformed into log and a significance level of p < 0.05 was adopted. Regression analyses were performed using StatisticaTM 7.0 software StartSoft and the ANCOVA was performed using software R version 4.1.1.

Results​

---

A total of 248 specimens were analyzed 104 of Triportheus albus (25 preflexion, 18 flexion, 45 postflexion, and 16 juveniles) and 143 of T. angulatus (46 preflexion, 54 flexion, 13 postflexion, and 30 juveniles). Yolk sac larvae of both species were not found during sampling. Individuals in the early stages, such as preflexion and flexion, were captured mainly in open water. In contrast, the more developed ones (e.g., postflexion and juveniles) were caught under macrophyte stands and in shallow and marginal areas of rivers and lakes.

Triportheus albus

Larval period. Preflexion (Figs. 2A-B):The standard length ranges from 3.29 to 6.13 mm (mean ± SD = 4.74 mm ± 0.77). The notochord is rectilinear and visible through transparency. There are no traces of the yolk sac. The body is elongated in a fusiform shape, the dorsal profile concave, and the head less deep than the trunk (HD/BD – 72.79 to 99.26%, mean ± SD = 89.11 ± 6.70%). The snout is rounded, but at the end of the stage, it becomes pointed; the mouth is superior and the inferior jaw long, with the entire dentary exposed when in dorsal view. The nostrils are simple, and the operculum is formed. The eyes are spherical and completely pigmented. The swim bladder is inflated and takes up a great space in the abdomen. There are pigments involving the swim bladder upper portion and, rarely, the lateral region of the digestive tube. Chromatophores are observed in the ventral region, located at the anterior portion of the digestive tube, sometimes in the median region, little conspicuous under it and abundant posteriorly the anus, but without reaching the caudal peduncle. In individuals larger than 4.50 mm SL, scarce punctate pigments appear in the dorsal fin region and parallel to the notochord. The finfold is hyaline, involves the body dorsoventrally from the second half of the intestine to the midline of the stomach (SND/SL – 41.43 to 51.37%, mean ± SD = 46.72 ± 2.52). Only the pectoral fin button is present. The total myomeres number ranges from 38 to 39 (18–19 preanal and 19–21 postanal).

Flexion (Figs. 2C–D): The standard length ranges from 6.15 to 9.13 mm (mean ± SD = 7.73 mm ± 0.95). The notochord tip is flexed by the appearance of the hypural plate. The body remains in the fusiform shape, whose trunk is the highest part, but without a sharp angle between the anterior and posterior regions. Snout, mouth, maxilla, nostrils, eyes, swim bladder, and pectoral fin did not show any changes. The anus is located posteriorly to the middle of the body. The color pattern is similar to the previous stage, however, the pigmentation parallel to the notochord, in the ventral and cephalo-dorsal regions, becomes more conspicuous. This dorsal pigmentation extends in a band of sparse chromatophores, dendritic on the cephalic plate and punctate on the dorsum, to the place of origin of the adipose fin, but never over the urostyle. Few punctate pigments (rarely dendritic) are observed in the snout, premaxilla, dentary, and in more developed individuals, at the base of the caudal rays. Internal melanophores also appear along the cleithrum at the base of the pectoral fin. In this stage, there is the odd fins delineation and the formation of their rays, which at the end of the stage are, for the most part, developed. Similarly, the finfold is observed only in the region of the pelvic fin origin. The adipose fin is in development. The caudal rays begin to segment, and the shape of the caudal fin is modified, initiating the division into two lobes. The total myomeres number ranges from 38 to 39 (18–22 preanal and 17–20 postanal).

Postflexion (Figs. 2E–F): The standard length ranges from 10.22 to 18.91 mm (mean ± SD = 13.82 mm ± 2.63). The notochord and swim bladder are no longer visible through transparent because the muscle tissue. At the beginning of the stage, the shape of the body resembles flexion, but during development, the appearance of the ventral keel is observed, and the body acquires a deep and compressed shape. The nostrils become double. Initially, the pigmentation is conspicuous, enhancing the pattern of the previous stage, but punctiform and dendritic chromatophores appear through the maxilla, operculum, pre-operculum, around the eyes, on the inner part of the lateral surface of the digestive tract, and in rays of the odd fins, except the adipose fin. In individuals larger than 14.00 mm SL, pigments are observed in the pectoral unbranched ray. At the end of the stage, bands are formed in the dorsal and mediolateral regions of the body, which extend from the snout and pre-operculum, respectively, to the caudal peduncle. Additionally, caudal fin pigmentation becomes more concentrated at the ends of the rays. The midline under region of the body is almost hyaline and when appear pigments, these are concentrated at the caudal fin base. In this stage, the appearance of the scales is observed, which have pigmented distal edge. The odd fins are in the final stage of branching and segmentation of the flexible rays and the adipose fin is already formed. There are remnants of the finfold in the ventral region. At 10.22 mm SL, the first pectoral fin rays are observed. At 12.00 mm SL the pelvic fin rays appear, but still do not have all the formed and segmented elements. The myomeres total number range from 38 to 39 (17–18 preanal and 21–22 postanal), while unbranched rays and branched range from ii,8–9 dorsal; iii,22–27 anal; i,5–6 pelvic and, i,4–13 pectoral (Tab. 1).

TABLE 1 | Variables analyzed (mm), minimum values (Min), maximum values (Max), mean (Mean), standard deviation (SD), and morphometric relationships (%) found for larvae and juveniles of Triportheus albus. Abbreviations: AF, absent fin; BD, body depth; BDA, depth body towards the anus; ED, eye diameter; HD, head depth; HL, head length; N, number of analyzed individuals; n, number of individuals with the mode of myomeres and rays; NV, not visible; SL, standard length; SNA, snout distance to the anal fin; SND, snout distance to the dorsal fin; SNL, snout length; SNP, snout distance to the pectoral fin; SNV, snout distance to the pelvic fin.

Juvenile period (Fig. 2G): In this period, the standard length ranges from 22.33 to 49.39 mm (mean ± SD = 29.56 mm ± 6.47). The body is compressed laterally, the eyes are large, the mouth is superior, the nostrils are double, and the anus is located posteriorly to the middle of the body. There is one large series of scales on the ventral keel and 32 to 35 scales arranged along the lateral line. The pectoral fins are long and only reach the origin of the pelvic fin. In preserved specimens, a dark dorsolateral band is observed extending from the snout to the caudal peduncle. In addition, dendritic and punctate chromatophores are present at the base of the inferior jaw, on the dentary, in the maxilla, around the eyes, and opercula. Pigmentation is distributed over the pectoral fin unbranched ray and between the dorsal and anal fin rays. The pelvic and adipose fins are hyaline, and the caudal fin color pattern is well defined, where the pigments are concentrated at the base and at the tip of the rays, however, the upper lobe is densely pigmented. Under of the body midline, pigmentation is scarce, when present, it is limited to the base of the anal fin and delineating the scales margin. The complete formation of the fins (branching and segmentation of the rays) occurs in this period, following the sequence: caudal, anal (iii, 26–28), dorsal (ii,8–9), pectoral (i,11–13), and pelvic (i,5–6).

FIGURE 2| Early development of Triportheus albus: (A) preflexion early (3.91 mm SL); (B) preflexion late (6.13 mm SL); (C) flexion early (7.01 mm SL), (D) flexion late (8.52 mm SL), (E) postflexion early (11.17 mm SL), (F) postlexion late (17.17 mm SL), and (G) juvenile (27.58 mm SL). Scale bar = 1 mm.

In the preflexion stage, the body is long and low (14.34 to 19.41% of SL), becoming moderate in flexion (13.06 to 21.13% of SL), postflexion (19.64 to 27.30% of SL) and juvenile (28.20 to 30.86% of SL). The head is small in preflexion (18.36 to 24.12% of SL) and flexion (19.24 to 26.59% of SL), becoming moderate in postflexion (25.14 to 31.59% of SL) and juvenile (24.61 to 30.25% of SL). Eyes range from moderate to large throughout early development (28.81 to 40.17% of HL). Snout length (SNL/HL), head depth (HD/HL), and distance of snout-dorsal (SND/SL), anal (SNA/SL), and pectoral (SNP/SL) fins, increased along the initial ontogeny, only the distance of snout-pelvic (SNV/SL) fins maintained its proportions (Tab. 1).

On growth patterns, the eye diameter exhibited continuous isometric development (linear regression) (Fig. 3B). The snout length had positive allometric growth (quadratic regression) (Fig. 3A). All other variables related to head length and standard length showed discontinuous isometric growth, therefore, they were better represented by the piecewise linear regression model. These variables showed an abrupt change in development after the breakpoint observed in the postflexion stage. For head length, the distance of snout-anal, dorsal, and pectoral fins the growth rate decreased after the breakpoint, while head deep, body deep, and distance of snout-pelvic fins increased growth velocity (Tab. 2; Fig. 3).

TABLE 2 | Values of Linear (L), quadratic (Q) and piecewise (S) regression analyzes of morphometric variables in relation to head length (HL) and standard length (SL) of Triportheus albus larvae and juveniles. R² = coefficient of determination. BM = best model, BP = breaking point (mm), a and b = regression parameters and N = number of individuals analyzed. Values in bold represent a significant difference (p < 0.05).

FIGURE 3| Body ratios (mm) between head length and snout length (A), eye diameter (B), and head depth (C), and standard length and body depth (D), head length (E), distance from snout to anal fin (F), distance from snout to dorsal fin (G), distance from snout to pectoral fin (H) and distance from snout to pelvic fin (I) during the early development of Triportheus albus.

Triportheus angulatus

Larval period. Preflexion (Figs. 4A–B): The standard length ranges from 4.09 to 6.02 mm (mean ± SD = 4.69 mm ± 0.43). The rectilinear notochord is visible through transparency, and remnants of the yolk sac are still present. The body is elongated in a fusiform shape, convex and the head depth is generally greater than the body depth (HD/BD – 94.86 to 111.82%, mean ± SD = 102.98 ± 3.97%). The snout is rounded, and the mouth is terminal, but at the end of the stage (~5.00 mm SL) it becomes superior. The inferior jaw is short and barely visible in the dorsal view. The nostrils are simple, and the opercula are formed. The eyes are spherical and completely pigmented. The swim bladder is inflated and occupies a great space in the abdomen. The intestine is functional, straight and elongated, with the anus located in the body middle region. Individuals smaller than 4.60 mm SL have a few chromatophores in the occipital region (from one to three pigments); a horizontal line of pigments is noted to extend between the posterior region of the eye and the end of the operculum, analogous to a mask. Pigments are concentrated on the upper of the swim bladder and anteroventral portion of the digestive tract and spaced along the ventral region after the anus. In larger individuals (~5.00 mm SL) the coloration intensifies, and various dendritic pigments are distributed from the cephalic region to approximately the origin of the embryonic fin. Scarce dendritic pigments appear parallel to the notochord, and in the lower part of the intestine. The finfold is hyaline and can be seen enveloping the body dorsoventrally, from the first half of the intestine, posterior to digestive tract, to the origin of the pectoral fin button (SND/SL – 34.16 to 41.00%, mean ± SD = 37.07 ± 2.21). Only the pectoral fin button is evident. The total number of myomeres ranges from 37 to 39 (17–19 preanal and 19–22 postanal).

Flexion (Figs. 4C–D): The standard length ranges from 6.12 to 9.61 mm (mean ± SD = 7.41 mm ± 0.68). The notochord tip is flexed by the appearance of the hypural plate and there are no more remnants of the yolk. The form of the body is initially similar to the preflexion stage and, as it develops, it acquires an angular shape, with the anterior region clearly deeper than the posterior. The snout, mouth, maxilla, nostrils, eyes, swim bladder, and pectoral fin showed no changes with the preflexion stage. The anus is in the region posterior to the middle of the body. The color pattern is similar to the previous stage, however, there is the appearance of dendritic pigments throughout the dorsal region arranged in a series that intensifies with development. Along the intestine, the pigments are conspicuous and almost continuous. Midline body pigments become more visible, reaching the urostyle and sometimes around it. On flanks with approximately 9.00 mm SL, a row of internal melanophores parallel to the notochord emerges over the epidermis as continuous filiform chromatophores, extending from the caudal peduncle to the beginning of the operculum. Dendritic chromatophores are distributed between the caudal fin rays, but in greater concentrations at the median rays base. A continuous filiform band traced the base of the anal fin. Pigments can be observed on the snout, surrounding the premaxillary, under the dentary, base of the mandible, maxilla, nostrils, around the eyes, inner part of the operculum, arches, and gill filaments, on the lateral surface of the digestive tract and intestine. Also, internal chromatophores appear along the cleithrum, which, in larger individuals (9.00 mm SL), surround the pectoral fin base, connect to the swim bladder melanophores. Still, in flexion, there is the delineation of the odd fins and the formation of their rays, which at the end of the stage are mostly developed. Likewise, the embryonic membrane is observed only in the region of origin of the pelvic fin. The adipose fin is in development. The caudal rays begin to segment, and the shape of the caudal fin is modified, initiating the division into two lobes. The myomeres total number ranges from 37 to 39 (18–22 preanal and 17–20 postanal).

FIGURE 4| Early development of Triportheus angulatus: (A) preflexion early (4.53 mm SL); (B) preflexion late (5.00 mm SL); (C) flexion early (7.50 mm SL), (D) flexion late (9.61 mm SL), (E) postflexion early (11.59 mm SL), (F) postflexion late (13.73 mm SL), and (G) juvenile (21.69 mm SL). Scale bar = 1 mm.

Postflexion (Figs. 4E–F): The standard length ranges from 9.83 to 15.40 (mean ± SD = 11.94 mm ± 1.43). The notochord and swim bladder are no longer visible through transparency due to dense muscle tissue. The body profile remained the same. At the end of the phase, the nostrils are double. The pigmentation pattern is more intense in the previous stage. Numerous dark dendritic chromatophores appear on the flanks that outline and enhance the myomeres’ pattern. Additionally, pigments are observed on the rays of the odd fins, including the adipose fin. The caudal fin stands out, where a dense pigmentation at the base of the rays is distributed along the median segments and is connected to a vertical strip arranged at the end of the caudal rays. The dorsal fin pigmentation is composed of filiform chromatophores in the basal part and circular dendritic in its distal portion. The dorsal, anal, and caudal fins are in the final stage of branching and segmentation of the flexible rays. The adipose fin is formed and embryonic membrane remains only in the ventral section, in contrast to the appearance of the pelvic fin that remains throughout the postflexion stage. At approximately 11.08 mm SL and 13.73 mm SL, the first pectoral fin rays and the pelvic fin rays appear, respectively, however, despite having developed rays, still do not have all the elements formed and segmented at the end of this stage. The myomeres total number varies from 38 to 39 (19–21 preanal and 17–19 postanal) and of unbranched and branched rays range from ii,9 dorsal, iii,27–30 anal, 5 pelvic, and i,6–8 pectoral (Tab. 3).

TABLE 3 | Variables analyzed (mm), minimum values (Min), maximum values (Max), mean (Mean), standard deviation (SD), and morphometric relationships (%) found for larvae and juveniles of Triportheus angulatus. Abbreviations: AF, absent fin; BD, body depth; BDA, depth body towards the anus; ED, eye diameter; HD, head depth; HL, head length; N, number of analyzed individuals; n, number of individuals with the mode of myomeres and rays; NV, not visible; SL, standard length; SNA, snout distance to the anal fin; SND, snout distance to the dorsal fin; SNLs, snout length; SNP, snout distance to the pectoral fin; SNV, snout distance to the pelvic fin.

Juvenile period (Fig. 4G): Among the individuals, the standard length ranged from 18.08 to 53.71 mm (mean ± SD = 25.65 mm ± 6.70). The body is deep and compressed laterally, the eyes are large, the mouth is superior, the nostrils are double, and the anus is located posteriorly to the middle of the body. They have two series of large scales on the ventral keel and 34 to 37 scales along the lateral line. The pectoral fins are long, extending beyond the origin of the pelvic fin. Punctiform and dendritic pigmentation are distributed throughout the body. On the flanks, pigmentation outlines the edge of the scales. Chromatophores are distributed over the adipose fin and rays of all fins, including the pectoral and pelvic fins, which were previously hyaline. The color pattern of the caudal fin is notable, provides a band of chromatophores that are distributed from the base to the end of the median rays and is linked to a dark vertical stripe arranged in the distal region of the rays, acquiring the Y- shape. In this period of development, all fins are formed (branched and segmented) with the following formation sequence: caudal, anal (iii,27–31), dorsal (ii,9), pectoral (i,11–13), and pelvic (i,6).

The body is long and low in the preflexion stage (12.10 to 17.20% of SL) and varies for moderate in flexion (14.67 to 20.45% of SL), postflexion (18.46 to 25.04% of SL) and juvenile (28.02 to 35.16% of SL). The head length varies from small to moderate (17.58 to 32.78% of SL) and the eyes diameter varies from moderate to large (32.54 to 43.13% of HL) along the ontogeny. Head depth (HD/HL), snout length (SNL/HL), and distance of snout-fin pelvic (SNV/SL) maintained their proportions. The distance snout-fin dorsal (SND/SL), anal (SNA/SL), and pectoral (SNP/SL) distances increased during the ontogeny (Tab. 3).

Regarding growth pattern, the snout length, and the distance of snout-anal and ventral fins exhibited continuous isometric development (linear regression) (Figs. 5A, F, I). All other variables related to head length and standard length showed discontinuous isometric growth, therefore, they were better represented by the piecewise linear regression model. These variables underwent an abrupt change in development after the breakpoint, observed in the postflexion stage. For eye diameter, head length, distance snout-dorsal, and pectoral fins the growth rate decreased after the breakpoint, whereas head depth and body depth increased in growth velocity (Tab. 4; Figs. 5B, E, G and H, C, D).

TABLE 4 | Values of Linear (L), quadratic (Q) and piecewise (S) regression analyzes of morphometric variables in relation to head length (HL) and standard length (SL) of Triportheus angulatus larvae and juveniles. R² = coefficient of determination. BM = best model, BP = breaking point (mm), a and b = regression parameters and N = number of individuals analyzed. Values in bold represent a significant difference (p < 0.05).

FIGURE 5| Body ratios (mm) between snout length (A), head length and eye diameter (B), head depth (C), and standard length ratio (mm) between body depth (D), head length (E), distance from snout to anal fin (F), distance from snout to dorsal fin (G), distance from snout to pectoral fin (H) and distance from snout to pelvic fin (I) during the early development of Triportheus angulatus.

Identification key for some Triportheus species from the Amazon basin during early ontogeny (Fig. 6)

1a.. Less than 40 myomeres total number……………….. 2

1b.  More than 40 myomeres total number (Fig. 6D)……………….. Triportheus auritus (see Cajado et al., 2021)

2a.  Convex dorsal profile in early stage larvae (Fig. 6A); inner mediolateral band emerging in the epidermis in mostly dashed and continuous pigments (Fig. 6B); presence of evident angle between the anterior and posterior regions of the body in more developed larvae; conspicuous pigmentation on the median rays connecting to a vertical band at the tip of the caudal-fin rays (Fig.6C)……………….. Triportheus angulatus

2b.  Concave dorsal profile in early stage larvae (Fig. 6A’); mediolateral band only in the epidermis, mostly in dotted pigments (Fig. 6B’); fusiform body without evident angle in more developed larvae; conspicuous pigmentation at the tip of the upper lobe of the caudal fin (Fig. 6C’)……………….. Triportheus albus

FIGURE 6| Illustrated summary of the identification key for Amazonian Triportheus larvae. Letters explain characters mentioned in the identification key. Scale bars = 1 mm.

Morphometric comparisons. All variables related to head length, except for snout length, were significantly different between the two species (ANCOVA, p < 0.05), with eye diameter and head depth being higher in T. angulatus (Tab. 5). The variables head length and distance of snout-dorsal, anal, and pelvic fins differ among species. The distances of snout-anal and pelvic fins were greater in T. albus throughout the early ontogeny. The distance of snout-dorsal fin and head length is initially longer in T. albus, but these changes in the juvenile period, becoming longer in T. angulatus (Tabs. 1, 3).

TABLE 5 | Results of the analysis of covariance (ANCOVA) for the variables obtained in the individuals Triportheus albus and Triportheus angulatus in relation to standard and head length.