Ever since it was first discovered in aquarium exports nearly two decades ago, the Golden Fairy Basslet (Gramma dejongi) has kept an air of mystery about it. Sightings of this beautiful yellow fish in the wild have been remarkably few, and, until the recent breeding success by aquaculturist Todd Gardner, aquarium specimens were exceedingly scarce and impossible to find in the US. Today, we’re getting our best look yet at this enigmatic grammatid thanks to Bruce Carlson, but these new images are likely to raise some difficult questions about the validity of this “species”.

When it was first described in 2010, Gramma dejongi was known only from specimens collected around Trinidad, Cuba, roughly midway along the southern shores of the island. The species was named for Arie DeJong of DeJong Marinelife (a large European aquarium wholesaler), who discerned that this fish looked quite different from the common Caribbean Fairy Basslet or Royal Gramma (Gramma loreto). He sent specimens over to famed ichthyologist Dr. Jack Randall who, along with Dr. Ben Victor, described a new species based upon three small individuals.

In this description, these authors note that there are really no morphological differences to distinguish the two, nor was there any indication of a genetic distinction in the CO1 gene. This “barcode” gene is used heavily in taxonomy to identify closely related species, and the identicalness of it here was certainly suggestive that the yellow Cuban specimens might simply be an unusual variation of the typical bicolored G. loreto. As the authors themselves stated, the similarities “[raise] the possibility that the new species is a local color variant and not a separate species.”

It was noted at the time that G. dejongi might be associated with slightly deeper reefs (>20 meters) in comparison to the Royal Gramma, but this new footage, taken in May of this year at the Jardines de la Reina, shows that the two simply co-occur in the middle portions of the altiphotic zone. Observations of five specimens were made, all between 45-75 feet (13-22 meters), and, in each instance, the yellow dejongi phenotype was vastly outnumbered by its bicolored congener. Such low abundance strongly argues against a viable population being present, especially when we compare it to the commonness of related taxa.

The type specimens of G. dejongi were all quite small, with the largest being only 45mm in length (G. loreto can reach nearly twice this), but observations since then have shown the two reach an equivalent size in adulthood—a nice plump specimen can be seen in this video at the 3:24 mark. One possible explanation for the relative rarity of larger individuals is that they make for an easier prey item to predatory fishes, especially the introduced Lionfish. As related by Arie De Jong, one of these was found to have a dozen dejongi in its belly, but none of G. loreto, whose bicolored patterning likely makes for much better camouflage.

When Randall and Victor described Gramma dejongi, they did so on assumptions that no longer hold true. This fish is not ecologically restricted to deeper reefs, as seemingly every time it has been observed it has been in the close company of G. loreto. There is no size discrepancy either. And the species isn’t endemic to Cuba, as it has since been found at Little Cayman Island, well to the south. When we take into account the fact that hybrid phenotypes have regularly been spotted, that there are no morphological distinctions, and that the genetics of the two are virtually identical, the evidence quickly mounts that this yellow fish is not really a valid “species”.

But I say this with a caveat. Gramma loreto is clearly in the process of speciating. We can see evidence of this with the strong split between it and its sister species, the Brazilian Fairy Basslet (G. brasiliensis). We find more evidence in the Caribbean with the unique population that occurs in the Bahamas, which appears to be midway on the path to developing a dejongi-like phenotype. And there’s plenty of genetic evidence out there in other groups that supports the waters south of Cuba as being a region of endemism.

Gramma dejongi really does stretch the limits of what we can taxonomically allow as a valid “species”. If we accept it as being distinct, which one could certainly make an argument for, we also have to accept that the yellow phenotype is simply an uncommon variation and that most specimens are essentially identical to G. loreto. That species was originally described from Matanzas, Cuba, located on the northern shores of the island. And if we take into account what the Caribbean would have looked like during the many recent periods of lowered sea levels, it’s clear that the two Cuban coasts would have been quite isolated from one another. This is the crucible of speciation, but our understanding of its patterns in the West Atlantic is still being deciphered, and nowhere is our ignorance laid quite so bare as it is with these colorful little basslets.

• Lohr, K.E., Camp, E.F. and Manfrino, C., 2014. First record of the basslet Gramma dejongi outside of Cuba. Coral Reefs, 33(1), pp.221-221.
• Molina, W.F., da Costa, G.W.W.F., Cioffi, M.B. and Bertollo, L.A.C., 2012. Chromosomal differentiation and speciation in sister-species of Grammatidae (Perciformes) from the Western Atlantic. Helgoland Marine Research, 66(3), p.363.
• Victor, B.C. and Randall, J.E., 2010. Gramma dejongi, a New Basslet(Perciformes: Grammatidae) from Cuba, a Sympatric Sibling Species of G. loreto. Zoological Studies, 49(6), pp.865-871.