The Cayo Arcas reef complex is situated at 20.21°N, -91.98°W, 145 km from the mainland. This complex comprises reefs fringing a cluster of three sand cays that are spread over an area of ~4 km by ~2.6 km (Figure 2). The largest sand cay, which is 1 km long, has an elongate, 4 km long, crescent-shaped reef that runs along and to the northwest of its eastern side. Between the eastern beach of this cay and the exposed reef crest, there is a narrow, very shallow (~1m deep) sandy lagoon. The smaller eastern cay has a similar, much smaller lagoon (see Figure 2). On the western side of the main cay there is a large sand-floored semi-lagoon with scattered submerged patch reefs, some with abundant macroalgal growth, that gradually slopes off westwards into deeper water over a distance of ~2 km from the main cay. Reef development apparently extends down to ~25m (see http://www.gulfbase.org/reef/view.php?rid=cac1).

Figure 2. 

Schematic representation of Cayo Arcas reef complex. Solid lines indicate edges of emergent reef. Solid fill = sand cays. Locations of dives sites for 2013 and 2016 are indicated by symbols.

Mangroves and beds of seagrasses (turtlegrass, Thalassia testudinum Koenig 1805, and manatee-grass, Syringodium filiforme Kutzing in Hohenacker, 1860, represent major ancillary habitats adjacent to many reefs in the tropical northwest Atlantic. Those habitats are used by many reef fishes, often as nursery habitats. While turtlegrass often forms large beds within which dense growths of flat, strip-like blades provide ample shelter for fish, manatee-grass has thin, thread-like blades and a sparser growth pattern that provide lower quality shelter for reef fishes. There are no mangroves at Cayo Arcas. The www.gulfbase.org marine species list for reefs in the Mexican section of the Gulf of Mexico indicates that turtlegrass does occur at Cayo Arcas (see p 27 of http://www.gulfbase.org/checklist/pdfs/marine-spp-list.pdf). That gulfbase list cites Flores (1992) as the source of this information. A copy of Flores (1992) was reviewed, and it was found that, while a table that indicates what terrestrial and marine flowering plants occur at different islands shows turtlegrass as present at Cayo Arcas (and at all the other islands treated in that book), it is not mentioned in the detailed text treatment of plants at Cayo Arcas. Hence that table record may have been an error. Flores (1992) does indicate that there were no mangroves at Cayo Arcas, but does not include any mention of manatee grass at any location. Turtlegrass was not seen by us or any other divers from UNAM-Sisal studying shallow water organisms during expeditions there in April and August 2016. In contrast, manatee grass was found at Cayo Arcas during 2016, and may be restricted to the back-reef lagoon on the eastern side of the main cay, and perhaps a similar lagoon next to the eastern cay (Figure 2). In August 2016 there were windrows of dead seagrass composed exclusively of manatee-grass threads along the entire beach on the eastern side of the main cay. As both T. testudinum and S. filiforme are restricted to shallow water it is unlikely that turtlegrass was present but missed during the 2016 expeditions. Hence two major ancillary habitats available for reef fishes throughout much of the wider Caribbean area are entirely (mangroves) or largely (seagrass) lacking at Cayo Arcas.

The earliest information available on Cayo Arcas fishes comes from an unpublished Master’s thesis by Garduño (1988). Part of this information, based on censuses of fishes made at depths between 10-42 m at Cayo Arcas during 1984-86, was published in Garduño and Chavez (2000). In 2013 HP-E censused reef fishes on twelve 25mX4m transects at one site on the Arcas reef complex on July 31, while ENL censused 24 such transects at 6-11 m depth, divided among five sites on August 6–7 (Figure 2). Most recently, during August 20-27, 2016, DRR & FPI surveyed the Cayo Arcas reefs to assess the population status and distribution of an exotic Indo-Pacific damselfish, Neopomacentrus cyanomos that has become established in much of the southwest Gulf of Mexico, including Cayo Arcas (see Robertson et al. 2016, Simoes and Robertson 2016). During that period 14 dives were made at 11 sites on that reef complex (Figure 2). Each dive, which ranged from 5–30 m in maximum depth, lasted 1-1.5 hrs, and covered an area of ~0.25 ha. In addition, three dives were made to depths of 40 m on a small petroleum-loading platform, located 1.5 km southwest of that reef cluster (see Figure 2, and Simoes and Robertson 2016). That platform is at the northern end of the complex of several hundred platforms that form Mexico’s largest offshore oilfield in the Gulf of Mexico. During those dives notes were made of the presence and abundance of different species of reef fishes. Due to the briefness of the 2016 visit, the primary objective of those dives was to gain a broad-brush picture of the abundance and distribution of both N. cyanomos and native fishes across a variety of habitats, rather than obtaining quantitative data on densities on small transects such as those used by HP-E and ENL.

Pterois volitans (Linnaeus, 1758). This invasive Indo-Pacific lionfish was first reported on the USGS invasive species website at an oil platform near Cayo Arcas in late 2012 (http://nas.er.usgs.gov/queries/SpecimenViewer.aspx?SpecimenID=292473. Fewer than a dozen individuals of this species were seen by the group of eight divers during the August 2016 expedition. Those occurred at depths of between 8-20 m on the reef, and at 30 m on the oil platform. Lionfish were first reported at Alacran reef, 350 km northeast of Cayo Arcas, in 2010, and on coastal reefs ~450 km west of Cayo Arcas in 2012 (http://nas.er.usgs.gov/queries/collectioninfo.aspx?SpeciesID=963). At Alacran Reef lionfish seem to be concentrated at mesophotic depths (Aguilar-Perera et al. 2016). It is not known how abundant lionfish is in the Cayo Arcas area below 25 m depth.

Hypoplectrus nigricans (Poey, 1852). This species is known to vary geographically in color and shape (Aguilar-Perera 2004). At Cayo Arcas in 2016 all H. nigricans were similarly colored, with a translucent, dark blue-black head, body and fins, except the tail fin, which was translucent grey, with darker upper and lower borders.

Lutjanus apodus (Walbaum, 1792). Garduño (1988)Garduño and Chavez (2000) listed only two snappers, L. apodus and Ocyurus chrysurus (Bloch, 1791) as present at Cayo Arcas. The seven members of this family that were observed during 2016 included six species of Lutjanus, but not L. apodus. L. apodus also was not recorded in 2013. Two of those Lutjanus species, L. griseus (Linnaeus, 1758) and L. jocu (Bloch & Schneider, 1801) were common across on the Arcas reefs in 2016. L. apodus, which is easily recognizable and probably the commonest and most widely distributed shallow-living member of its genus on Caribbean reefs, occurs through much of the Gulf of Mexico (see http://www.iucnredlist.org/details/155152/0, and http://biogeodb.stri.si.edu/caribbean/en/thefishes/species/3684. If currently present in the Arcas reef complex L. apodus must be rare.

Neopomacentrus cyanomos. During a visit to Cayo Arcas in April 2016 NS found it to occur both on the reefs and the adjacent oil-loading platform, where it was superabundant (Simoes and Robertson 2016). During the August 2016 expedition we found this species in small aggregations (up to ~150 fish, but usually about several dozen, including many juveniles) at all dive sites on the reef complex. It also formed a large, dense aggregation of many tens of thousands of individuals on the oil-loading platform (see Simoes and Robertson 2016). Neither HP-E nor ENL recorded this species in 2013. However, a 2016 review by HP-E of photographs he took during his July 2013 expedition show that this species was present on the Cayo Arcas reef then (Figure 3).

Figure 3. 

Neopomacentrus cyanomos at Cayo Arcas, July 2013 (Photo HP-E); 11 N. cyanomos are visible, and can be identified by the presence of a black blotch on the shoulder, and a large white blotch on the rear edge of the dorsal fin; a single Chromis multilineata (without the shoulder- and dorsal-fin blotches) is present in the foreground immediately above the brain coral.

Stegastes adustus (Troschel in Muller, 1865) and S. diencaeus (Jordan & Rutter, 1897). In 2016 S. adustus, a benthic-feeding, omnivorous damselfish, was common on all hard reef substrata not covered by live corals, at depths of 0.5-7 m. HP-E recorded S. adustus but not S. diencaeus, while ENL recorded both species. In their list Garduño and Chavez 2000 did not include either species by name, but did record Eupomacentrus mellis Emery & Burgess, 1974 which is the juvenile of Stegastes diencaeus (see Robertson and Allen 1981). Adults of both these Stegastes species are dull brown to blackish fishes with similar shapes and body sizes that are often confused, even in guide books by experts: e.g. figure 287 of Randall (1996) is S. diencaeus, not S. adustus; the lower photo on p. 171 of Allen (1991) is S. diencaeus, not S. adustus; and the image on the bottom of p. 113 of Humann and DeLoach (1994) is S. diencaeus, not S. adustus (an error that was corrected in later editions of that book). In addition, juvenile S. diencaeus displaying the “mellis” coloration look quite similar to juveniles of two other Stegastes species commonly found at Cayo Arcas: S. leucostictus (Muller and Troschel in Schomburgk 1848) and S. xanthurus (Poey, 1860; recorded as S. variabilis (Castelnau, 1855) by Garduño and Chavez (2000); see http://biogeodb.stri.si.edu/caribbean/en/gallery/specie/3879, and http://biogeodb.stri.si.edu/caribbean/en/gallery/specie/3884). Given the abundance of S. adustus at Cayo Arcas in August 2016 and the apparent absence of S. diencaeus then, and the similarities between adults and juveniles of S. diencaeus and its congeners, differences in those Stegastes spp recorded during and prior to 2016 might have been due to misidentifications. In contrast, both S. adustus and S. diencaeus were recorded at Alacran Reef by González-Gándara and Arias González (2001, 2004) and both were commonly observed by DRR during two weeks of diving there in May 2016.

Halichoeres burekae Weaver & Rocha, 2007. This recently described species is endemic to the southwest and northwest Gulf of Mexico. Its known range extends from Alacran Reef on Campeche Bank to the Flower Garden Banks at the edge of the continental shelf off Texas (see http://biogeodb.stri.si.edu/caribbean/en/thefishes/species/4707; and http://maps.iucnredlist.org/map.html?id=187608). At Alacran Reef, 350 km northeast of Cayo Arcas, Aguilar-Perera and Tuz-Sulub (2009) described this species as occurring in small aggregations of a few dozen fish to as many as 200 fish, and that it is widely distributed there. Observations of the abundance and distribution of H. burekae at Alacran Reef made by DRR during May 2016 are consistent with those of Aguilar-Perera and Tuz-Sulub (2009). This species is relatively uncommon at the Flower Gardens area (http://www.iucnredlist.org/details/187608/0). During August 2016 H. burekae was recorded at all dive sites on the Cayo Arcas reefs (but not the oil-loading platform) and was one of the most abundant labrids on that reef complex. It was at least an order of magnitude more abundant than observed by DRR at Alacran Reef during June 2016. Multiple aggregations of scores to hundreds of individuals moving about in the water column 1-3m above the bottom were seen at each dive site. Those aggregations were mainly seen in areas of low relief, over both sandy and coralline-rock bottoms. A large aggregation containing thousands of individuals was seen mass spawning in midwater over a substratum of dense, live gorgonian trees on the fore-reef off the center of the main-cay reef at ~10m depth during one afternoon dive.

Scarus guacamaia Cuvier, 1829. At Cayo Arcas one or two adults of this species were seen on each of eight of the 14 dives on the reef complex in 2016. S. guacamaia also was recorded by Garduño 1988, and by HP-E in 2013. This species has been thought to be dependent on mangroves as habitat for juveniles (http://www.iucnredlist.org/details/19950/0). However Cayo Arcas lacks mangroves, and the nearest mangroves are located on the coast, 145 km away. Although <20 S. guacamaia were seen at Cayo Arcas in 2016, a more comprehensive survey across a broad range of habitats would be needed to provide a reliable estimate its total population size there.

Enneanects boehlkei Rosenblatt, 1960 and Starksia ocellata (Steindachner, 1876). One individual of each of these two cryptic species was collected as bycatch while using clove oil to anaesthetize Neopomacentrus cyanomos for collection.

Coryphopterus hyalinus Bohlke & Robins, 1962, and C. personatus (Jordan & Thompson, 1905). This pair of sister species (see Baldwinet al. 2009) apparently have essentially the same external appearance, and are differentiated by the patterns of pores on the top of the head. They also have very similar geographic, habitat and depth ranges (see http://biogeodb.stri.si.edu/caribbean/en/thefishes/species/4119 and http://biogeodb.stri.si.edu/caribbean/en/thefishes/species/4121). We include them as a single unit as, while diving, we were unable to determine whether one or both occurs at Cayo Arcas.