Three new species of Australoheros from Brazil

[cichla]

Three new species of Australoheros from southeastern Brazil, with taxonomic notes on Chromys oblonga, Heros autochton and H. jenynsii (Teleostei: Labroidei: Cichlidae)

http://www.vertebrate-zoology.de/vz62-1 ... _83-96.pdf

[Juan Artigas]

Thank you so much!

[michi tobler]

It seems like the number of Australoheros taxa grows with the same speed as the guppy population in my office...
Thanks for posting!

[Rico Morgenstern]

That's interesting. Three further species added to the genus, certainly not the last! This time, also the "old" nominal species Chromys oblonga Castelnau, 1855, Heros autochthon Günther, 1862 and Heros jenynsii Steindachner, 1869 have been considered.I find it a good idea to regard C. oblonga a nomen dubium. The state of the type specimen does hardly allow any proper identification, and the type locality is doubtful. So, for the time being, we should forget the name as it is currently not applicable to any known species.

If the newly designated lectotype of H. autochthon is really part of the original type series (if not, the designation is null and void) remains to be demonstrated, as it differs considerably from the remaining three types, and was listed separately whitout locality and collector data by Regan (1905). Anyway, Ottoni's conclusion that this specimen is not an Australoheros, or, indeed, not even belonging to the tribe "Heroini" (BTW the tribe name Therapsini Allgayer, 1989 has priority!), is not substantiated. All visible remains of the color pattern agree with other Australoheros, and the one feature which is said to exclude the specimen from being an Australoheros, namely the uninterrupted trunk bars 6-7 (actually this is an over- or misinterpretation of the original generic diagnosis), is not at all discernable on the photo (note that the actual specimen was not examined).

Thus, the status of H. autochthon remains unresolved, and Ottoni's attempt to get rid of a potential senior synonym of one or (more likely) several of his new species has only added to the existing confusion.

[Willem Heijns]

Another challenging read! I'll have yet to study the paper in detail, but at first glance a few things spring to mind:

Australoheros mattosi from the Rio São Francisco is most likely a valid name. There might even be a second Australoheros species in that drainage, because Ottoni didn't take Britski's paper (1988) into account; a paper showing a differently shaped fish:



As for Australoheros montanus, its range fits pefectly within the range of many of Ottoni's other Australoheros species, the ones I consider to be identical to Australoheros autochthon.

Australoheros tavaresi is the most interesting new name. This species apparently occurs in the Rio Paraná drainage and may therefore not be part of the species complex occurring in the coastal rivers of Brazil. Makes you wonder what you would find travelling down the Paraná until you reach the range of Australoheros kaaygua and the like. Quite a stretch of river not yet researched for Australoheros.

It gets hilarious when you read the part on the status of Australoheros autochthon. In his earlier paper Ottoni didn't even mention the name, now he disregards it as not being an Australoheros and "probably not even belonging to the tribe Heroini". All he had to get to this conclusion were pictures of the syntypes of Heros autochthon (he never saw the fish themselves) and we all know how difficult it is to identify species from pictures. Also I disagree with his opinion that the three paralectotypes are in worse condition that the chosen lectotype. It's the other way around, as I will show in the second part of my article on this genus in the July issue of Cichlid News magazine. Further comments will be published there too.

[cichla]

So far it is known (from the biography) Lord Stuart did not leave Rio. Thus it is likely that autochthon is collected there. The fourth syntype (now designated as the lectotype) differs from the other three specimens (see figure below) in body shape. It might be that these specimens are not conspecific. Could it be that the fourth specimen (now labeled as the lectotype) is actually not part of the syntype series. Why should autochthon not be an Australoheros? It has >4 anal fin spines and caudal fin spot is situated as in other species of the genus. The figure (see below) shows that the fourth specimen is closer to Asutraloheros than to cichlasomatines (be aware sample size is very very small and it needs to be reproduce with a comprehensive data set).

[Rico Morgenstern]

Australoheros tavaresi is the most interesting new name. This species apparently occurs in the Rio Paraná drainage and may therefore not be part of the species complex occurring in the coastal rivers of Brazil. Makes you wonder what you would find travelling down the Paraná until you reach the range of Australoheros kaaygua and the like. Quite a stretch of river not yet researched for Australoheros.

As the headwaters of the Rio Tieté, from where the type series of A. tavaresi was taken, very closely approach the coast, it is not unlikely, that some these rivers were formerly part of a coastal drainage and have changed their course by river capture and/or tectonic activities.

Some evidence is presented here: http://www.scielo.br/pdf/ni/v4n2/v4n2a09.pdf

'Geophagus' brasiliensis (the next candidate to be split up into umpteen indindistinguishable species??? ) is also present there, although it might have been introduced by man (Stawikowski & Werner 2004, quoted from Lucena & Kullander 1992). This, in turn, could then also be true for the Australoheros species.

[cichla]

Thank you Rico for the link. There is also a connection between upper rio Parana and rio Sao Francisco (1). Closely related cichlid species distributed in the upper rio Sao Francisco and upper upper rio Parana (2). Willem, this is why I think that it is more likely that tavaresi has more 'connections' to the A. autrani species group than to A. kaaygua & Co.

It seems like the number of Australoheros taxa grows with the same speed as the guppy population in my office...

Michi, just feed them good, like two times daily. Give them some flakes and mosquito larvae. Otherwise the Guppies will lost the run. There are more Australoheros on their way.

Greetings, IS

(1) 2005: A poorly known case of watershed transposition between the São Francisco and upper Paraná river basins. Neotropical Ichthyology 3(3):449-452. http://www.scielo.br/pdf/ni/v3n3/v3n3a15.pdf
(2) 2011: Zoogeografie der südamerikanischen Cichliden. DCG-Informationen 42(9): 209-216.

[Willem Heijns]

Halfway through reading the paper by Ribeiro (2006) I thought I more or less understood what he meant. Fascinating to read how old patterns of rivers were set by those "megadomes" (Río Uruguay, Rio Paraná, Rio São Francisco). Subsequent erosion might well have caused stream capture of their headwaters by developing coastal rivers like Rio Paraiba do Sul. But then Ribeiro states that recent processes (active even today) may be responsible for "sending back an aquatic biota from the coastal rivers to the upper Paraná" (page 242). And now I'm lost. Which way did Australoheros travel?

[Rico Morgenstern]

As they are seemingly absent from large parts of the Alto Paraná draingae, I find it more likely that they have spread along the coast and have entered the more interior drainages secondarily. At least, this would be a more parsimonious explanation than one which would require to take into consideration an area-wide extinction of this otherwise seemingly successful group within such a large drainage. Of course it cannot be excluded that the seeming absence of Australoheros in the Upper Paranà drainage is actually only the absence in collections, but this would be also more than unlikely in the light of the vastness of the area, which can be regarded (at least in parts) as reasonably well sampled.

[cichla]

Biogeography: Wilem, I do not think that it needs dramatic geomorphological activities and millions of years to explain the distribution and speciation for the members of the autrani group. The differences between the species of the autrani group are comparatively small. It needs only few thousands (or even few hundreds) of years. The fluctuating sea levels during the last 10000 years has(in this case) surely a high impact to the current diversity and distribution of the species of the autrani species group.

autochthon: there are many characters which confirm the membership of the lectotype to Australoheros. However, see Ottoni (2012: 93) about autochthon: ''... herein designated ... lectotype ... does not belong to the genus Australoheros, ..., probably, even not belonging to the tribe Heroini.'' Any idea why it is not a Heroini? What else could it be?

[Willem Heijns]

So, assuming that the Alto Paraná is relatively well researched and no Australoheros is present, this raises the interesting question: why is that so? Especially since the habitat is apparently suitable for these species (i.e. Australoheros tavaresi).
Be that as it may, the most likely way the latter species has reached the Paraná is from somewhere in the coastal rivers. That would also place it in close relationship with other coastal species; the ones I would still like to call Australoheros autochthon. Might tavaresi be nothing more than just another syononym of autochthon?

[cichla]

By having 12 caudal vertebrae (Ottoni 2012: 91, tab. 2) Australoheros tavaresi is a member of the autrani group (see Ottoni 2012: tab. 3). Hence, it may presumed that tavaresi has close relationship with other coastal species of the autrani group.

Might tavaresi be nothing more than just another syononym of autochthon?

If autochthon is not a Heroini (see above) then neither tavaresi nor one of the remaining members of the Australoheros autrani group can be a synonym of it.

[Willem Heijns]

We mustn't put ideas into the minds of our readers, Ingo. Unless of course you really believe Heros autochthon not to belong to the tribe Heroini. If so, I would very much like to read your "strong arguments".

As for vertebral count, here's an interesting notion: all Australoheros species of which Ottoni is the (co)author appear to have 14 abdominal vertebrae and 12 caudal vertebrae (except Australoheros acaroides, but then you were the senior author of that ). All other Australoheros species apparently have 13 abdominal vertebrae and 13 (sometimes 14) caudal vertebrae.
We can only hope a counting error is at hand here (counting one caudal vertebra as an abdominal vertebra), because otherwise all Ottoni's Australoheros species wouldn't fit the diagnosis of the genus which says all Australoheros species have 13 abdominal and 13 (14) caudal vertebrae.

[cichla]

Unless of course you really believe Heros autochthon not to belong to the tribe Heroini. If so, I would very much like to read your "strong arguments".

No, I do not have arguments that autochthon is not a member of Heroini. But only some that it belongs to Australoheros (or at least to Heroini). With the rhetorical question I just try to find out whether there are arguments which support Ottoni's decision or not.

We can only hope a counting error is at hand here (counting one caudal vertebra as an abdominal vertebra), because otherwise all Ottoni's Australoheros species wouldn't fit the diagnosis of the genus which says all Australoheros species have 13 abdominal and 13 (14) caudal vertebrae.

Oops , another strange thing.

[Lisachromis]

Colour pic - http://novataxa.blogspot.ca/2012/05/201 ... tanus.html

[DRE]

Also great! Who is the photographer here?