Some aspects of Dendrobatids in Venezuela:
Declines and nomenclature

César Luis Barrio-Amoros,
Executive Director Fundación Andigen A,
Apartado Postal 210,
5101-A Mérida, Venezuela.
E-mail: cesarlba@yahoo.com

According to current data, Venezuela is the sixth most amphibian-diverse country in the world with over 300 species known and still many more to discover and report (Barrio-Amoros, 1998; unpublished data). Of the known species, Rodriguez and Rojas Suarez (1995-1999) in the Red Book of Venezuelan Fauna identify three endangered categories containing eleven species of amphibians. The Dendrobatids, Dendrobates leucomelas and Minyobates steyermarki are placed in the lower category of 'Lower risk, least concern.'

Dendrobates leucomelas is an attractive pet animal, hence the current concern about its extinction in Venezuela. However, the selling price for this animal internationally has decreased, as it is easy to rear in captivity. Our own observations show that D. leucomelas is abundant in all places sampled (Barrio Amoros y Fuentes, 1998).

Minyobates steyermarki should be considered under a threatened species category since its restricted habitat (the top of Cerro Yapacana, an isolated 'tepui' in Amazonas State) makes it vulnerable to any natural or artificial catastrophe.

As is now known, an illegal expedition extracted some 150 animals for a German breeding centre. However the main threat comes from open-air gold mining, mostly by Brazilian 'garimpeiros' which causes widespread deforestation, soil erosion, general predation and mercury poisoning (Barrio-Amoros y Fuentes, 1999).

Other Venezuelan Dendrobatids mentioned in Vial and Saylor's (1993) list are the following: Colostethus alboguttatus C. collaris, C. duranti, C. haydeeae, C. humilis, C. leopardalis, C. mandelorum, C. mayorgai, C. weridensis, C. orostoma, C. saltuensis and C. serranus (for comments on nomenclature, see below). Little is known about their population status although it is likely that some are undergoing local declines (some inhabit minute areas isolated by human activity). C. humilis is not an uncommon species in the Andean foothills (Walter Schargel, pers. com.). C. leopardalis, once abundant in the Paramo de Mucabaji is currently extremely difficult to locate. Mannophryne collaris has undergone a drastic reduction in numbers in some areas such as the City of Mérida and its surroundings in the Venezuelan Andes (La Marca, 1995) due to human expansion, but it is still abundant in areas of the Barinas foothills (Barrio Amoros y Fuentes, 1999). The other species lack any kind of monitoring program so as to affirm their inclusion in the list.

Aromobates nocturnus is an apparently unique Dendrobatid by its peculiarities (Myers et al. 1991). This species is found at 2200m elevation at its type locality in Trujillo State in an Andean cloud forest. It has recently been the subject of an intense search by teams from the Universidad de los Andes without success.

Venezuela possesses a very high rate of dendrobatid endemism, particularly as regards mountain ranges (Andes, Coastal Range, and Tepuis or highlands of the Guianan shield). Many of these endemic species show limited population distributions and are in danger because of human activity (Andes) or because they are susceptible to natural disasters (drought, floods and bacterial, viral or fungal infections) and to over collection. Besides the already named species, the following taxa should be taken into consideration when listing species likely to become extinct in the near future, though this is not the result of monitoring or specific studies: Colostethus lamarcai, C riveroi.

Guianan Shield Species
(Colostethus ayarzaguenai, C. brunneus, C. guanayensis, C. murisipanensis, C. parimae, C. parkerae, C. praderioi, C. roraima, C. sanmartini, C. shrevei, C. tamacuarensis, C. tepuyensis, C. undulatus), though some are only known from one or a few specimens, are not suspected of being in danger, as they inhabit remote and difficult to explore areas. In my experience, some of these species (plus two new ones under study) are still abundant (also see Myers et al. 2001 for C. undulatus).

Colostethus bromelicola is a very difficult species to find (as it lives on high bromeliads) and since its description in 1956 nothing is known.

Dendrobatid frogs do not seem to be affected by the introduction of two foreign anuran species in Venezuela: Eleutherodactylus johnstonei (Antillean coqui) and Rana catesbeiana (North American bullfrog), due to the fact that none occupy the same habitat. Very little is known of the decline of Dendrobatids in Venezuela and this is usually based on conjecture or unproven hypotheses (which should rely on proper methodology such as systematic long-term monitoring). I recommend immediate monitoring of species highly susceptible to extinction, such as Aromobates nocturnus.

Nomenclature
Another interesting aspect to discuss is the nomenclature of Venezuelan Dendrobatids. It is known that the evolution and speciation of dendrobatid frogs occurred in North-western South America. The main centre is said to be in Colombia (with 71 species; 46 of which are Colostethus, A. Acosta pers. com.), Venezuela (with 47 species: 41 Colostethus sensulato, see below), Ecuador (with 44 species: 30 Colostethus; (Coloma, pers. com.) and perhaps also in Perú (with 43 known species; 14 Colostethus, V. Morales, pers. com.). These numbers are subject to changes, especially due to the description of new species. Currently, I have at least 6 new Colostethus species to describe. In the short time since Barrio & Fuentes' 1999 synopsis the whole panorama has suffered severe changes. According to the most recent genetically cued reference (Vences et al. 2000), only two genera are mophyletic: Dendrobates and Phyllobates. Myers (1987) transferred several Dendrobates into his new genus Minyobates, and some Phyllobates and Colostethus into Epipedobates. Although none are mophyletic, they are adequate to accommodate a stable position in taxonomy. However, Vences et al. (2000), analysing mitochondrial ribosomal sub-unit gene sequences, recently placed Minyobates into the synonymy of Dendrobates. Likewise, the controversial genus Allobates (Zimmermann & Zimmermann, 1988) is currently accepted in basis of genetic consistency to include A. femoralis and A. zaparo. Finally, the same authors definitively placed in the synonymy of Epipedobates the long-discussed genus Phobobates (Zimmermann & Zimmermann, 1988).
It is also outstanding that Venezuela harbours more genera than any other country, principally due to the effort given to some weak characteristics to determine Colostethus groups. Myers et al. (1991) erected Aromobates in light of several strange features for Dendrobatids, large size (up to 62mm), nocturnal activity, aquatic habits, extensive feet webbing, and a malodorous smell. La Marca (1996) placed doubt on the validity of the genus, because some of these characters are also shared by other species, like size (Colostethus oblitteratus, C. riveroi), nocturnal activity (C. riveroi, which is not true by personal observation) and odour (C. leopardalis). However, I consider there to be enough evidence to maintain this genus valid. In Venezuela, two of Rivero's groups (1988), the 'trinitatis (=collaris)' group and the 'alboguttatus' group were respectively raised to genera by La Marca (1992, 1991 '1994'). Mannophryne was created by three 'pretended' synapomorphies: black collar, yellow throats in females and a developed aggressive mating behaviour. In fact, this behaviour is only documented in three of the 12 species considered in this genus. Collars are present in several other species in Colombia and Ecuador, yet none are considered as Mannophryne. These characters, based on colouration and behaviour, are undoubtedly too weak to maintain a generic status. I do not doubt that a monophyly maybe possible in this group, but a further generic statement is required to maintain the genus properly. Nephelobates was erected to include Dendrobatids with fang-like teeth and a dermal anal flap, two other non-unique characters (Lynch, 1982; Myers et al. 1991). The genus contains eight species. I completely reject the argument of a monophyly because these subjective and weak arguments are insufficient to defend and accommodate a dendrobatid genus. I will show my developed arguments in a future paper (in progress with Hinrich Kaiser). Furthermore, Kaiser et al. (1994), Coloma (1995), and Grant et al. (1997) provide further doubts about the generic recognition of these.

In both cases, the publication of the descriptions of these two genera appeared in obscure local and irregular university journals (La Marca 1991 '1994'; 1992). I consider that the description of a new genus of Dendrobatid is a very important question and deserves its matter appearing to light in the best herpetological journals.

César Luis Barrio-Amoros,
Executive Director Fundación Andigen A,
Apartado Postal 210,
5101-A Mérida, Venezuela.
E-mail: cesarlba@yahoo.com

We wish to thank the author for this fascinating paper. A full list of references was supplied but due to constraints on space we have been unable to include these.