The systematic status of Heros margaritifer

Kullander lists Heros margaritifer Günther, 1862 as a valid species in the genus Amphilophus noting that the type locality is doubtful and that the species was not reported with certainty since the original description. Günther descripted this species on the basis of a single fish that was supposedly collected in Lake Peten, Guatemala. Bussing & Martin (1976) redescribed the species on the basis of the type specimen. They also indicate that the actual type locality may in fact be in the Belize River system in Belize or on the Atlantic versant of Guatemala.

In the past, fish from various localities were linked to this name. Conkel (1993) shows fish as A. margaritifer that supposedly came from Lake Peten. However, although various people looked for the according fish there, nobody was ever able to find similar fish. Stawikowski & Werner (1998) hypothesized that Conkel might have pictured a natural hybrid. Fishes collected by Wessel in Honduras were identified as A. margaritifer by R. R. Miller (Wessel, 1996). According to Stawikowski & Werner (1998), the according publication, however shows A. longimanus.

Buchhauser (2005) brings up the discussion again. He pictures not only the cichlid collected by Wessel (identified as A. margaritifer), but also a fish collected by Werner in the Rio Lancetilla, Honduras (identified as A. robertsoni) and cichlids he caught in the Quebrada de Barracon, also Honduras. Indeed, the fish of Wessel and Werner look very similar. The fish Buchhauser caught in Honduras, however, look way different and resemble indeed the fish Günther pictured in the first description. Buchhauser points out that the newly collected fish differ from A. robertsoni in having blue pearly spots, a red belly, bulging lips and a different head profile. Furthermore he argues that only 100 km more in the west, fish can be found that look like A. robertsoni from México, Belize and Guatemala. This leads to his conclusion that the fish caught by Wessel, Werner and himself is not A. robertsoni but potentially A. margaritifer.

Taking into account the redescription of Bussing & Martin (1975), further differences to A. robertsoni become evident, namely the absence of the small black spots on the interradial membranes of the spinous dorsal fin, the presence of pearllike spots on each scale of the body and presence of a red edge in the dorsalis. However, Buchhauser's fish also differs in some key characteristics of A. margaritifer given by these authors: The lateral blotch is present, the third and fourth trunk bars are not very prominent. These color patterns, however, are known to vary depending on the mood in living specimens and the coloration of preserved fish is not known from Buchhauser's samples. Furthermore, it is unclear if the brown-orange blotch on the opercle of Buchhauser's fish is the "crescentic blotch on the posteroventral margin of the opercle" mentioned by Bussing & Martin (1975).

Thus, it seems still premature to me to conclude that Buchhauser's fish is the enigmatic A. margaritifer. Further questions need to be addressed first:

• Before any conclusion, all the mentioned forms should first be compared with the type material. Comparisons of photos are not sufficient to draw any conlusions.
• Is it possible that there is a gradual change in the cichlid forms from "classical" A. robertsoni to the fish collected by Wessel, Werner and Buchhauser? Is it possible that Werner was right and the fish identified as A. margaritifer by others are in fact geographical variants of A. robertsoni? Furthermore, if the type locality of A. margaritifer indeed is in Belize, is A. robertsoni at the end a synonym to A. margaritifer?
• Given that the fish collected by Wessel, Werner and Buchhauser are not A. robertsoni, what is the actual evidence that they are identical with the type specimen of A. margaritifer?