The additions to, name changes and deletions of questionable records of some species listed from the PNSAV that are presented here are based on a review of those previous papers and incorporation of information from two additional sources: georeferenced records of species present in the PNSAV obtained from the digital databases of four major online aggregators that contain biogeographic information on fishes in the Gulf of Mexico, and our own collections and observations in the PNSAV. We reviewed and assessed the validity of the names used and questionable records of various species listed by Del Moral-Flores et al. (2013) and similarly reviewed the list of species arising from a subsequent study by Ayala-Rodríguez et al. (2016). Tello-Musi et al. (2018) provided further information on one species.

In recent years various efforts have led to large databases on the distributions of species becoming available through online museum databases, and from online aggregators that collate and distribute data from museums and a broad range of additional science sources. We took advantage of this trend by obtaining georeferenced records for species of fishes present in the area of the PNSAV from six major aggregators:

i) the Mexican National Commission for the Use and Conservation of Biodiversity (CONABIO: http://www.conabio.gob.mx/informacion/gis/) a national aggregator that collects data from Mexican science sources, and three aggregators that obtain data from a wider range of international sources;

ii) Integrated Digitized Biocollections (iDigBio: https://portal.idigbio.org/portal/search), an NSF sponsored effort run by the University of Florida that provides digital data from US collections;

iii) the Global Biodiversity Information Facility (GBIF: https://www.gbif.org/), which draws data from 45,000+ datasets on a broad range of organisms from a wide range sources scattered in most major areas of the globe;

iv) Fishnet2 (http://www.fishnet2.net/), which aggregates data from ~75 museum databases in North America (mainly), Europe, Asia and Australia;

v) the Ocean Biogeographic Information System (OBIS: https://obis.org/), a clearing house for data that aggregates museum and local-aggregator data on various aspects of the biology of marine organisms, including their geographic distributions, is hosted in Belgium, and has 13 regional nodes scattered around the world, including the USA; and

vi) FishBase (http://www.fishbase.org), an international aggregator supervised by a consortium of nine non-USA international institutions that takes data on fishes in general from a broad range of sources.

These aggregators often recycle some data amongst themselves. To obtain data on occurrences of fishes from the PNSAV we searched each of those databases for georeferenced species records within a quadrat with latitudinal and longitudinal limits that closely bounded the PNSAV, with latitudes from 19.04° to 19.26°N, and longitudes from -95.75° to -96.18° W. Individual georeferenced records can be obtained from each aggregator. Since water depths within almost all of the PNSAV, particularly around the reefs, do not exceed 50m (Liaño-Carrera et al. 2019) we included in our results only those species known to occur at depths between 0–50 m in other parts of their geographic ranges. We excluded records of species of poeciliids, characids and cichlids as those are primarily or exclusively freshwater taxa. The records obtained from those aggregators were reviewed, to check for inconsistencies between putative occurrences and the known geographic ranges of species, which are not uncommon (e.g., see Robertson 2008), and to ensure included occurrences relate to updated taxonomic nomenclature, based on that in Eschmeyer’s Catalog of Fishes (Fricke et al. 2019). The PNSAV lies within the known geographic ranges of all species included in this update whose records in that MPA came from the aggregators.

Omar Domínguez-Domínguez (ODD) led a collecting expedition to the PNSAV in 2015 as part of a study of connectivity among reef fish populations throughout different reef areas in the Mexican tropical west Atlantic. That effort focused on both readily visible and small, cryptic fishes hiding in the reef matrix. For the latter the anesthetic clove oil was used to make collections (e.g., see Robertson et al. 2019). Voucher specimens of all small cryptic species collected by ODD were preserved in ethanol and have been deposited in the Colección de Peces de la Universidad Michoacana de San Nicolás de Hidalgo (curator MC Xavier Madrigal, xmguridi@yahoo.com).

Horacio Pérez-España (HP-E), based at the Universidad Veracruzana in Veracruz City, has spent decades studying reef fishes in the PNSAV. During a week in May 2019 D Ross Robertson (DRR), Carlos J Estapé (CJE), and Allison Morgan Estapé (AME) made scuba and snorkeling dives at a variety of inner and outer reefs in the northern and southern parts of the PNSAV. That activity led to the observations of species not on any previously published lists, or in online databases, and photographic records of various species.

Dasyatis americana Hildebrand & Schroeder, 1928 and D. sabina (Lesueur, 1824) to Hypanus americanus and H. sabinus. Both species have been reassigned to the genus Hypanus by Last et al. (2016).

Gymnura micrura (Bloch & Schneider, 1801) to G. lessae. Yokota and Carvalho 2017. Yokota and Carvalho (2017) split G. micrura into two species, and named the population from the Gulf of Mexico and Atlantic USA G. lessae, leaving G. micrura restricted to the coast of South America.

Manta birostris (Walbaum, 1792) to Mobula birostris. Manta was synonymized with Mobula by Last et al. (2016).

Antennarius striatus (Shaw, 1794) to A. scaber. (Cuvier, 1817). A striatus was thought to represent a single pantropical species. However, the west Atlantic population was recently recognized as A. scaber (see Arnold and Pietsch 2012; Smith-Vaniz and Jelks 2014).

Haemulon chrysargeum Gunther, 1859) to Brachygenys chrysargeum (Gunther, 1859). Tavera et al. (2018) reassigned this species to the newly created genus Brachygenys.

Pomadasys crocro (Cuvier, 1830) to Rhonciscus crocro (Cuvier, 1830). Tavera et al. (2018) reassigned this species to the newly created genus Rhonciscus.

Bairdiella ronchus (Cuvier, 1830) to Bairdiella veraecrucis Jordan & Dickerson, 1908. Marceniuk et al. (2019) revised the genus and resurrected B. veraecrucis for the Gulf of Mexico population. Bairdiella ronchus is restricted to South America.

Kyphosus incisor (Cuvier, 1831) to Kyphosus vaigiensis (Quoy & Gaimard, 1825). Knudsen and Clements (2013) synonymized K. incisor with K. vaigiensis.

Stegastes variabilis (Castelnau, 1855) to Stegastes xanthurus (Poey, 1860). While the name S. variabilis was long applied to both Brazilian and Greater Caribbean populations of what was thought to be a single species, S. variabilis is now considered to be a Brazilian endemic, while the Greater Caribbean population is S. xanthurus (Smith-Vaniz and Jelks 2014)

Labrisomus kalisherae (Jordan, 1904) to Gobioclinus kalisherae (Jordan, 1904) Lin and Hastings (2013) revised the genus Labrisomus and split it into three, with G. kalisherae placed in Gobioclinus.

Emblemariopsis sp. to Emblemariopsis diaphana Longley, 1927. Photographs of this species (Figure 1) show it to be E. diaphana.

Gnatholepis cauerensis (Bleeker, 1853) to Gnatholepis thompsoni Jordan, 1904. The species G cauerensis is restricted to the Indo-Pacific and St Helena in the Atlantic. Gnatholepis thompsoni, which is closely related to G cauerensis, is found on both sides of the Atlantic, including throughout the Greater Caribbean (Rocha et al. 2005; Van Tassell 2011)

Del Moral-Flores et al. (2013) listed 387 species in 206 genera and 92 families, including 21 elasmobranchs and 366 bony fishes in the PNSAV. We excluded ten species from this list that were not replaced by other names, due to likely identification errors, which would reduce the number listed by that paper to 377 species.

Narcine sp. to Narcine bancrofti (Griffith & Smith, 1834). Narcine bancrofti, which is included in the Del Moral-Flores et al. (2013) list, is the only member of this genus currently recognized from the Gulf of Mexico and Caribbean. Narcine sp. may have been used due to longstanding confusion arising from misidentification of N. bancrofti as N. brasiliensis (now known to be a Brazilian endemic, see Rosa et al. 2007) or to the fact that coloration of N. bancrofti varies considerably. We excluded this record during the update.

Alosa sapidissima (Wilson, 1811) is a temperate species with a native range in eastern North America from Canada to the central east coast of Florida (Natureserve and Daniels 2019). There are only two members of the genus with established populations in the GoMx, both of which are endemic to the northern Gulf. Alosa alabamae Jordan & Evermann, 1896, is restricted to the northeast section of the gulf (Natureserve 2010). Alosa chrysochloris (Rafinesque, 1820) ranges more widely, as far south as the Texas/ México border (Robertson and Caruso 2018), and is the most likely candidate for any Alosa found in in the southwest GoMx. Adults of Alosa spp. are marine, but spawn in rivers, and juveniles can be found in estuaries (Natureserve and Daniels 2019; Limburg 1996, O’Connell et al. 2004). Castro-Aguirre et al. (1999) did not record any members of this genus in estuaries on lagoons of México, which would be expected if they lived in Mexico and spawned there in rivers. The only aggregator records of any Alosa species in México are a few in GBIF (and Fishnet2 and FishBase) of A. pseudoharengus (Wilson, 1811) in Campeche, southern Veracruz state (not the PNSAV) and off the northeast tip of the Yucatan peninsula. While those Yucatan records represent misidentified Harengula jaguana Poey, 1865, the other records are of an Alosa species of uncertain identity, possibly A. chrysochloris, but not A. pseudoharengus (Hector Espinoza Pérez pers. comm. September 2019). Given that Harengula jaguana has been confused with Alosa elsewhere and is listed as present in the PNSAV by Del Moral-Flores et al. (2013), we suggest that the record of Alosa sapidissima at the PNSAV should be viewed as incertae sedis. We excluded it during the update.

Hypoplectrus puella (Cuvier, 1828) to H. floridae Victor, 2012. González-Gándara et al. (2012, 2013) recorded H. nigricans (Poey, 1830) (a look-alike congener of the Veracruz endemic H. atlahua Tavera & Acero P., 2013), H. puella (a a look-alike congener of the species described from Florida, H. floridae), and H. unicolor (Walbaum, 1792) (a look-alike congener of the Veracruz endemic H. castroaguirre Del Moral-Flores et al. 2012) from reefs around Tuxpan, 250 km north of the PNSAV. Among those species Del Moral-Flores et al. (2013) listed only H. atlahua, H. castroaguirre and H. puella from the PNSAV. Subsequently H. floridae was noted from one of the Tuxpan reefs by González-Gándara (2014) and in the PNSAV by Tello-Musi et al. (2018). Given that their look-alike congeners are present in Veracruz it seems unlikely that any H. nigricans, H. puella, and H. unicolor also are present. Hence, just as the records of Avalos et al. (2008) of H. nigricans and H. unicolor in the PNSAV were replaced by H. atlahua and H. castroaguirre, respectively, in Del Moral-Flores et al. (2013) the H. puella record of Del Moral-Flores et al. (2013) from the PNSAV most likely refers to H. floridae. For comparison, images of H. floridae from the PNSAV and Florida, and of H. puella from the Caribbean are presented in Figure 2.

Cynoscion jamaicensis (Vaillant & Bocourt, 1883). This species is largely restricted to South America and extends no further north than Honduras on the continental shoreline (Fredou and Villwock de Miranda 2015). This record most likely represents a misidentification of one of the three species of Cynoscion that have been found in the PNSAV, but were not included in the Del Moral-Flores et al. (2013) list (see Table 1). Castro-Aguirre et al. (1999) did not record it from México. This record was excluded during the update.

Stegastes fuscus (Cuvier, 1830) to Stegastes adustus (Troschel, 1865). While the specific name fuscus was long applied to the Caribbean dusky damselfish as Pomacentrus fuscus under the assumption that there is a single west Atlantic species, S. fuscus is a Brazilian species not known to be present in the Greater Caribbean (Carter and Kaufman 2003). As S. adustus is in the list of Del Moral-Flores et al. (2013) we excluded this record during the update.

Stegastes pictus (Castelnau, 1855) to Stegastes partitus (Poey,1868). Stegastes pictus is a Brazilian species (Carter and Kaufman 2003), juveniles of which resemble some individuals of the variably colored S. partitus. No other Greater Caribbean Stegastes species has a color pattern resembling that of S. pictus. The only records of S. pictus in the Greater Caribbean are of a few vagrants on the lesser Antilles in the southeast corner of the Caribbean, where vagrants of other species of Brazilian endemics also are known to occur. We note that S. partitus is in the Del Moral-Flores et al. (2013) list, and we excluded the S. pictus record from the update.

Halichoeres pictus (Poey, 1860) and Halichoeres socialis Randall & Lobel, 2003 to Halichoeres burekae Weaver & Rocha, 2007. The terminal phase male of H. burekae resembles that phase of both H. pictus and H. socialis, and the initial phases of H. socialis and H. burekae also are very similarly colored. Those three species form a clade within the new world Halichoeres species, in which H. burekae and H. socialis are sisters (Wainwright et al. 2018). Halichoeres pictus is a conspicuous species widely distributed on reefs throughout most of the Greater Caribbean, while H. socialis is a Belize endemic. Halichoeres burekae is abundant on reefs throughout the southwest Gulf of Mexico (Aguilar-Perera and Tuz-Sulub 2009; Robertson et al. 2016a, b). The southwest gulf records of H. pictus and H. socialis predate the description date for H. burekae and most likely refer to that species, as there are no verified recent records of either of these two species in the southwest Gulf of Mexico since H. burekae was described. Halichoeres burekae is a common inhabitant of PNSAV reefs (our observations) that is in the Del Moral-Flores et al. (2013) list. We excluded these two records during the update.

Ophioblennius atlanticus (Valenciennes, 1836). The name O. atlanticus was originally applied to the populations in both the west and east Atlantic. However, the Greater Caribbean population is now recognized as O. macclurei (Silvester, 1915), and O. atlanticus refers to the east Atlantic population only (Collette et al. 2003). As the Del Moral-Flores et al. (2013) list includes O. macclurei as well as O. atlanticus the record of O. atlanticus was excluded during the update.

Eleotris pisonis (Gmelin, 1789) to Eleotris amblyopsis (Cope, 1871). Pezold and Cage (2002) revised the genus and found that E. pisonis is restricted to eastern South America. Eleotris amblyopsis has been collected in the study area (see Table 1). We excluded this record when constructing the update. It should also be noted that Eleotris perniger (Cope, 1871) which ranges from Veracruz south to Brazil (Pezold et al. 2015) also has aggregator records very near the PNSAV and probably occurs within it.

Elacatinus evelynae (Böhlke & Robins, 1968) to Elacatinus prochilos (Böhlke & Robins, 1968), which is on the Del Moral-Flores et al. (2013) list. Elacatinus evelynae, which has a color pattern very similar to that of E. prochilos, is restricted to the Bahamas, Antilles and central Caribbean. It is not known from the northwest Caribbean. Elacatinus prochilos does occur along the coast of the northwest Caribbean from Honduras to northeast Yucatan and hence is the more likely of the two species to be present at Veracruz. There are no records of either species from the reefs of Campeche bank. We excluded this record from the update

Tigrigobius dilepis (Robins & Böhlke, 1964) and Tigrigobius saucrus (Robins, 1960) to Tigrigobius redimiculus (Taylor & Akins, 2007). Records of T. dilepis and T. saucrus in the PNSAV precede the date of the relatively recent description of T. redimiculus, which was based on specimens from the PNSAV. These three species have similarly structured color patterns, with the dark marks on the head and body ranging from brown in T. saucrus to red in T. dilepis to a brown body with a red head in T. redimiculus (Figure 3). Tigrigobius redimiculus is endemic to the southwest Gulf of Mexico, where it ranges from reefs of Veracruz state to Alacranes reef on the central Campeche Bank. The older Veracruz record is the only one for T. dilepis anywhere in the GoMx, while T. saucrus has confirmed records in the GoMx only at the Florida Keys and northern Cuba. No other species of goby in the wider Caribbean as similar to T. redimiculus as are T. dilepis or T. saucrus is known from the Gulf of Mexico. Tigrigobius redimiculus was common on massive coral heads in very shallow water on all reefs visited, but no T. dilepis or T. saucrus (Figure 3) were observed, despite searches for them by DRR, CJE and AME in May 2019. We excluded these two records from the update.

The study of fishes in the PNSAV by Ayala-Rodriguez et al. (2016) was focused primarily on larval fishes. However, they also added 16 species, based on records of adults, that were not included by Del Moral-Flores et al. (2013), including two deep-water species (Bregmaceros cantori (Milliken & Houde, 1984) and Tetragonurus atlanticus (Lowe, 1839)) we do not include here, and three questionable records that we discuss below.

Menidia menidia (Linnaeus, 1766). The generally recognized geographic range of this species is limited to the east coast of North America, from central Florida to Newfoundland (Carpenter and Munroe 2015). This record likely relates to a congener, e.g., M. peninsulae, which was not recorded in the PNSAV by either Ayala-Rodríguez et al. (2016) or Del Moral-Flores et al. (2013), and the known range of which extends along the northern coast of the GoMx and south along the western coast to at least Tamiahua, 275 km from Veracruz city in the northern part of Veracruz state (Castro-Aguirre et al. 1999; Chao et al. 2015 a; Raz-Guzmán et al. 2018). The update does not include this record.

Cynoscion regalis (Bloch & Schneider, 1801). The generally recognized geographic range of this species is the east coast of North America from Nova Scotia to southeast Florida, with occasional individuals on the southwest coast of Florida (Chao, 2003). Cynoscion nebulosus (Cuvier, 1830), which Ayala-Rodriguez et al. (2016) also recorded in the PNSAV, is a look-alike sister species that is sometimes misidentified as C. regalis (Chao, 2003). The known range of C. nebulosus extends from New York south throughout the Gulf of Mexico (except Cuba) (Chao et al. 2015b). Raz-Guzmán et al. (2018) recorded this species, but not C. regalis, in northern Veracruz state. Similarly, Castro-Aguirre et al. (1999) recorded C. nebulosus but not C. regalis from México. This record was not included in the update.

Membras vagrans (Goode & Bean, 1879); type locality Pensacola, Florida. Ayala-Rodriguez et al. (2016) and Raz-Guzmán et al. (2018) listed both M. martinica (Valenciennes, 1835) and M. vagrans at the PNSAV and at Tamiahua lagoon, 275 km north of the PNSAV, respectively. Castro-Aguirre et al. (1999, p. 191) treated M. vagrans as valid and provided a dichotomous key that separated M vagrans and M. martinica on the basis of non-overlapping numbers of anal fin rays: 14–18 for M. vagrans and 19–22 for M martinica. However, the geographic range of M. vagrans is overlapped completely by that of M. martinica, Miller (2006, p. 201) listed M. vagrans as a synonym of M. martinica, both McEachran and Fechhelm (1998, p. 886) and Robins et al. (2018, p. 185) did not include M. vagrans and gave anal fin ray counts for M. martinica of 14–21, completely overlapping the range given by Castro-Aguirre et al. (1999) for M. vagrans. In addition, Chernoff (1986) did not include M. vagrans in his revision of the Menidine silversides, and Chernoff (2003) did not include it in the FAO guide to the fishes of the northwest Atlantic. Hence it seems best at present to regard M. vagrans as a synonym of M. martinica. We did not include this record in the update.

We found records of 95 additional species not listed by Del Moral-Flores et al. (2013) that are known to occur in depths shallower than 50 m elsewhere in their geographic ranges. Those, which include two elasmobranchs, are from 73 genera and 41 families (Table 1), with eight of those families and 42 of those genera not recorded by Del Moral-Flores et al. (2013). Seventy-one (74.7%) of the additional records came from the six aggregators. While those aggregators produced the great majority of additional records only seven species (9.9% of those in aggregator databases) were recorded in all six aggregator databases. In addition, 10 (14.1%) of those 71 species were recorded from only one aggregator, eight from GBIF and one each from FishBase and OBIS. GBIF provided the greatest number of additional aggregator records, 61 species, but missed 14.1% of species recorded by one or more of the other aggregators. CONABIO recorded 18 additional species, iDigBio 49 species, Fishnet2 42 species, FishBase 15 species, and OBIS 18 species. Given this degree of variability in numbers and identity of species recorded by different aggregators it is evident that records need to be obtained from multiple aggregators to assemble comprehensive checklists. Further, two aggregators that draw data from the same sources do not necessarily provide the same set of georeferenced records for the same species: that table shows concurrence of additional species records among those extracted from iDigBio and Fishnet2 in only 37 (69.8%) of 53 cases in which either source provided a record, with five cases of species for which records extracted directly from Fishnet2 were not present in iDigBio. In contrast, GBIF, which also receives Fishnet2 data, did record all species recorded by Fishnet2.

The additional species records also include 25 species not in the aggregator databases: 12 of those recorded by Ayala-Rodríguez et al. (2016), one by Avalos et al. (2008), four collected by ODD and students in 2015 (in addition to 81 species they collected that are on the Del Moral-Flores et al. (2013) list), and seven species observed (plus one previously unnamed species on the Del Moral-Flores et al. (2013) list subsequently identified), and in three cases photographed, by DRR, CJE and AME during one week of diving and snorkeling in May 2019. Additional records also include two invasive Indo-Pacific species: Pterois volitans (Linnaeus, 1758), known from the PNSAV since the beginning of 2012 (Santander-Monsalvo et al. 2012), and Neopomacentrus cyanomos (Bleeker,1856) (see Figure 4), which was first recorded in the PNSAV by Horacio Pérez-España (HP-E) in early 2014 (see Robertson et al. 2016b). In addition, one species recorded by Tello-Musi et al. (2018) (Hypoplectrus floridae) effectively replaced one of the species (H. puella) on Del Moral-Flores (2016) list.

The additional species added since Del Moral-Flores et al. (2013) and discussed here include species with a range of biogeographic distributions, 32 Greater Caribbean endemics (including four GoMx endemics), 13 Northwest Atlantic endemics (found in and to the north of the Greater Caribbean), 33 West Atlantic endemics found in both the Greater Caribbean and Brazil, four transatlantic species, seven circumtropical species, and two aliens from the Indo-Pacific.

Mobula aff. birostris (the Caribbean manta; see Stevens et al. 2018). A large individual of this unnamed species, which has a distinctively different color pattern to that of M. birostris (Walbaum, 1792) (see Stevens et al. 2018), the only other morphologically similar species in the wider Caribbean, was closely observed by CJE, AME and DRR as it circled overhead during one dive; unfortunately poor visibility then did not allow for an adequate photograph. Haemulon boschmae (Metzelaar, 1919) was photographed by the wreck Riva Palacio (Figure 5), Calamus nodusus Randall & Caldwell, 1966 was photographed on De Enmedio reef (Figure 6); DRR, CJE and AME observed, and CJE photographed Canthigaster jamestyleri Moura & Castro, 2002 on Anegada reef (Figure 7), including one aggregation of 5 adults, in relatively shallow water for this species (14–20 m depth). H P-E had noticed this species previously on PNSAV reefs, present in some years, not in others. We repeatedly observed schools of Kyphosus spp. containing young adults of Kyphosus cinerascens (Forsskål, 1775) on several reefs which, due to its distinctly elevated dorsal and anal fins (see Knudsen and Clements 2013), is easy to distinguish from other members of the genus. DRR observed Entomacrodus nigricans Gill, 1859 living in barnacles in 0.5 m depth water, its typical habitat, at the base of a lighthouse on each of two emergent reefs. CJE photographed Emblemariopsis diaphana Longley, 1927 (Figure 1) at Isla Verde, and Blanca reefs, Emblemaria pandionis Evermann & Marsh, 1900 (Figure 8) on Enmedio reef, and Coryphopterus punctipectophorus Springer, 1960 (Figure 9) on Anegada reef. DRR observed several pairs of Nes longus (Nichols, 1914), perched at the mouths of snapping-shrimp burrows in which they live, on a sand bottom with abundant live Strombus pugilis Linnaeus, 1758, ca. 25 m away from the base of Enmedio reef at 15 m depth. Elacatinus jarocho Taylor & Akins, 2007 (Figure 10), and Halichoeres burekae (Figure 11) both endemic to the southwest GoMx and on the Del Moral-Flores et al. (2013) list, were common and present on all reefs visited.

Two species of Hypoplectrus that are endemic to the southwest GoMx were recently described, both of which are present in the PNSAV. The descriptions were based on few specimens and did not adequately cover the range of variation in live coloration we have observed, and photographed, in both species at the PNSAV. As color patterns are important taxonomic aids for identifying Hypoplectrus species and often vary within as well as between species we present additional information on variation in both species.

Hypoplectrus atlahua. The type locality of this species is offshore from Tuxpan, 250 km north along the coast from the PNSAV. The photographs presented here represent the first published of the live coloration of this species, as the original description included only photos of freshly killed specimens. Here we present a selection to show variation in the coloration of adults and describe some of that variation. We also present images and describe the juvenile color pattern, which is quite different to that of adults. We observed a full range of color patterns from that of small juveniles grading to that of the largest adults. Large adults of H. atlahua have uniform dark brownish black head, body and fins, the head usually being paler than the body (Figure 12G, I). The eyes are brown, and there is a prominent blue spot at the upper corner of the operculum, varying amounts of blue lines on the face (sometimes virtually absent: Figure 13), and a prominent blue front margin to the pelvic fins (Figure 12, and see Tavera and Acero 2013). Individuals of many other species of Hypoplectrus often have a blue spot at the upper corner of the operculum but smaller and more weakly colored than in H. atlahua. There is often an indistinct darker triangular bar extending down and back from the eye to the lower rear corner of the operculum and the body can have indistinct dark bars (Figure 12F). The body sometimes has 15–20 faint vertical blue lines extending between the dorsal and ventral body profiles (Figure 12H). Adults can change color between uniform blackish brown to mid-brown with indistinct dark blotches on the rear of the body (see Figure 12A-C, all of one fish), or they may change between a dark, indistinct barred pattern and more uniform dark pattern (see Figure 12D, E, both of another single fish).

Hypoplectrus atlahua juveniles (Figure 13) are differently colored: juveniles sometimes have pale bodies with five dark bars on the upper body, the anterior two brown, the rear three blackish, the third bar broken into two blotches, the last bar on the end of the caudal peduncle with two black spots adhering to its rear border, each of those spots with a bright white spot above it. Alternatively they sometimes have a grey-brown body, with a darker area along the side of the head and mid-flank, and a series of black blotches at the rear of the body, a vertical pair under the anterior soft dorsal, a single blotch under the rear of the soft dorsal, a large blotch before a pair of small round spots on the end of the caudal peduncle and base of the caudal fin, with whitish areas before and behind the top of the large caudal-base blotch. The fins are translucent. As fish grow, they get a progressively darker body and fins and the rear black blotches become less distinct.

Hypoplectrus castroaguirrei (Figure 14) Del Moral-Flores et al. (2011) described this species as being pale yellow, with fine blue lines on the head and chest, and blue spots on the top of the head; indistinct brown bars on the body, an oblique black bar from the top of eye down to the lower edge of the preopercle, a black blotch before the eye, both of those black marks finely edged with blue; a black blotch on the caudal peduncle; caudal, anal and pelvic fins yellow, the anal and pelvic fins with a thin blue border; the dorsal fin yellow with oblique blue lines. The type locality of this species is the PNSAV. There are very few photographs of live fish in the field available for this species (see Del Moral-Flores et al. 2011). Here we present and describe a selection taken on the reefs of the PNSAV, to provide an indication of the greater variation in this species coloration than was indicated in the original description. The ground color of the body of adults varies from pale yellowish white through mid-yellow to yellow with a brown tone over the upper body, to pale yellowish with indistinct brown bars on the upper body. The fins are yellow, and all except the caudal fin have a thin blue border. The dorsal fin, especially the soft part, is covered with many fine blue spots arranged in oblique lines, which sometimes coalesce into short, thin continuous stripes. The caudal peduncle bears a black blotch that varies considerably in size and shape, ranging from a small black blotch on the center of the upper caudal peduncle to a large, irregularly shaped blotch that covers most of the peduncle and extends forward on the rear of the body and onto the rear base of the soft dorsal fin, and sometimes is split into two separate blotches. The eye is black, surrounded by up to three black marks, including a triangular bar one angled back and down below the eye that is invariably present but varies in its length, a rounded blotch before the eye (present or absent), and a small rounded blotch above the top rear corner of the eye (present or absent). Those blotches are finely outlined with blue, there are varying amounts of blue lines on the snout, cheeks, operculum, nape, and breast, and varying arrangements of blue spots on the top of the head. The entire body of some individuals is covered with a series of ~15–20 thin vertical blue lines extending between the top and bottom profiles (Figure 14B). We have no photographs of small juveniles of this species.