75% of all species of catfish come from South America

[Dinyar]

about 75% of all species of catfish come from South America

I didn't know that, and find it surprising. Would anyone be able to provide a breakdown of # of species by continent? Thanks.

[Silurus]

Here is a very rough breakdown obtained at a quick & dirty guessstimation from the ACSI website:

Africa (+Madagascar): 488
Asia: 483
Australia+New Guinea: 66
North America (+Mexico): 52
South America (+Central America S of Mexico): 1647

These are useful only as ball park figures.

[MatsP]

Copy Silurus figures into Excel, and we realize that my guestimations was off by some amount...


Africa (+Madagascar): 488 18%
Asia: 483 18%
Australia+New Guinea: 66 2%
North America (+Mexico):52 2%
South America (+CA): 1647 60%
Total 2736

I was actually under the impression that a larger proportion of species were from South America.

Of course, until ALL species have been identified, we won't know exactly how many they are.

My point, however, was that (especially for the aquarium trade) a very large portion of the different species of catfish originate from south america. The exact proportion was a off by some amount. Sorry about that.

--
Mats

[Owch]

Here is a very rough breakdown obtained at a quick & dirty guessstimation from the ACSI website:

Africa (+Madagascar): 488
Asia: 483
Australia+New Guinea: 66
North America (+Mexico): 52
South America (+Central America S of Mexico): 1647

These are useful only as ball park figures.

What about European Catfish?

[Silurus]

That got lumped into Asia (I should have labeled that Eurasia).

I am using the six-continent scheme.

[Dinyar]

That's quite an eye-opener. Thanks, Silurus.

Would it be a lot of trouble to put these numbers for Siluriformes into the context of all freshwater fish? E.g., Continent, # all FW species, # catfish species?

[Silurus]

Would it be a lot of trouble

Yes, because no exact (or even near enough to exact) figures exist for what you are asking of.

These are the numbers culled from the Lundberg et al. (2000) paper (So many fishes, so little time: an overview of recent ichthyological discovery in continental waters. Annals of the Missouri Botanical Garden 87: 26â??62):

N America: 1050
S & Central America: >5000
Europe: 360
Africa: 3000
Tropical Asia (this excludes temperate Asia): 3000
Australia & New Guinea: 500

[Walter]

Hi,
I guess, Dyniars indication of percentage would be +/- correct, if we include the number of (extimated) undescribed species.

[Dinyar]

Would it be a lot of trouble

Yes, because no exact (or even near enough to exact) figures exist for what you are asking of.

These are the numbers culled from the Lundberg et al. (2000) paper (So many fishes, so little time: an overview of recent ichthyological discovery in continental waters. Annals of the Missouri Botanical Garden 87: 26â??62):

N America: 1050
S & Central America: >5000
Europe: 360
Africa: 3000
Tropical Asia (this excludes temperate Asia): 3000
Australia & New Guinea: 500

Thanks. Certainly a good benchmark. Eyeballing these numbers confirmed what I had suspected, namely that catfish make up a much bigger percentage of all FW species in S & C America (~1/3) than in other continents (~1/6 in Africa and Asia).

Any views on why this is the case?

[Silurus]

That's a tough one to answer.

One possibility that immediately comes to mind is that this reflects the history of the group. The fact that the most "primitive" catfishes (Diplomystes and, depending on who you follow, loricarioids and/or cetopsids) are South American would lend credence to this line of reasoning.

[Walter]

Hi,

Any views on why this is the case?

cypriniformes as barbs and most notably loaches as competitors for ecological niches are missing in central and south America.
Most of the benthal niches occupied from loaches in Asia are occupied of catfishes in SA - so they had/have more room to spread in diversity.

[Silurus]

cypriniformes as barbs and most notably loaches as competitors for ecological niches are missing in central and south America.
Most of the benthal niches occupied from loaches in Asia are occupied of catfishes in SA - so they had/have more room to spread in diversity.

What about Africa (where there are no loaches), then?
And there are plenty of bottom-dwelling non-siluriforms that could potentially compete with catfishes in South America (the gymnotiforms come to mind).

[Walter]

Hi HH,
Gymnotiformes are bottom dwellers? I don´t think so. Take a look at the swimm bladder, take a look at body shape.

All I tell is a guess, don´t forget...

In Africa there´s a different situation - Africa has a much smaller species diversity than South America (I think, you know the "Refugial Theories" of Haffner - and it´s adaptions - and the hypothesis why there´s such a high diversity in tropical SAmerica).
And if you subtract the number of Cichlid species of Lakes Malawi, Tanganyika and Viktoria, which percentage of all remaining african fish species would catfishes be then?
Would be interesting to count.

As interesting, as adding the number of loaches and catfish species in Asia and compare this number of (mostly) "bottom dwelling" species to the number of all Asian fishes.

[Silurus]

If Gymnotiformes are not bottom dwellers, why do they form such a high percentage of the benthic fish fauna in South America?

[Shane]

Another set of contributing factors to keep in mind are the radical geological changes that South America has undergone in the last few million years. These include the rising of the Andes, the reversal of the Amazon's flow, the seperation of the Amazon basin from the Orinoco (and subsequent creation of the Magdalena and Maracaibo basins). Not only have these events seperated species, such as the Orinoco Panaque nigrolineatus from its Amazonian relatives (L 27, 191, 330 etc), but created the Andean based whitewater environments (Rios Meta, Apure, Guarico, etc) and Guayana/Brazilian shield blackwater/clear water environments (Tapajos, Xingu, Orinoco, etc).
-Shane

[Walter]

Hi HH,

If Gymnotiformes are not bottom dwellers, why do they form such a high percentage of the benthic fish fauna in South America?

sorry, I´m no native speaker, but for me a "bottom dweller" is a fish, that contacts the surface of the substrat most of his life time.
Benthos ist more than "the bottom" - it´s the bottom and organism living in, on and above the bottom.´

And if you read the Thome-Souza and Chao paper carefully, you will recognize, that they only observed "deep channels of neotropical rivers" (p. 128: Study area. Research was conducted in two main channels of the Rio Negro basin, one sampling area was located on the lower Rio Branco...)

Deep channels are not the places of the most diversity in species.

Then: They used a otter drawl (p. 129: Samples were collected using a 5.28 m otter trawl... 10 minutes...)
That´s not the method to catch the (in my sight) "real bottom dwellers" like loricariids.
Take a look on Table 1 - p. 133 and 134 - only few specimen of loricariidae have been caught, but many pimelodids and doradids.

In my opinion you cant compare this work of Thome-Souza and Chao to all rivers of amazonia - it´s only a insight in the fish communities of deep river channels (and, as they annotate, communities change during the season because of migratory fish species).

[Walter]

Hi Shane,

Another set of contributing factors to keep in mind are the radical geological changes that South America has undergone in the last few million years. These include the rising of the Andes, the reversal of the Amazon's flow, the seperation of the Amazon basin from the Orinoco (and subsequent creation of the Magdalena and Maracaibo basins). Not only have these events seperated species, such as the Orinoco Panaque nigrolineatus from its Amazonian relatives (L 27, 191, 330 etc), but created the Andean based whitewater environments (Rios Meta, Apure, Guarico, etc) and Guayana/Brazilian shield blackwater/clear water environments (Tapajos, Xingu, Orinoco, etc).
-Shane

that´s in the extended "Refugial Theory" of Haffner.
But the most important factors probably were the frequent changes of climatic conditions during the ice-ages (is "glacean" correct in english?).

[sidguppy]

And if you subtract the number of c*****d species of Lakes Malawi, Tanganyika and Viktoria, which percentage of all remaining african fish species would catfishes be then

That's a non-argument, because if you would be consequent you'd have to substract the three groups of catfish that have shown adaptive radiation as well; Synodontis and Chrysichthys in Lake Tanganyika and Bathyclarias in Lake Malawi.

for some reason cichlids evolve into new species at a much faster rate than catfishes, perhaps because many catfishes are perfectly suited for oppertunistic lifestyles where there's no need to evolve into a new species.

the ease from wich the genuine Lakedweller S petricola can be kept in a wide arrange of tanks, springs to mind. Synodontis are often extremely "flexible" when it comes to changing environment.
still they DO evolve, and with the "petricola-polli-dhonti" clade (all have the characteristic enlarged upper lip, short whiskers and small eyes), this process is happening as we speak: where once it was decided only 8 or so species dwell in Tanganyika, we now find new species, subspecies and local varieties every year.....

If you substract all these as well (they evolved with the cichlids using the same oppertunities of a new Lake), you miss out a fairly large batch of African cats as well.
Apart from Clarias, Synodontis and perhaps Amphilius, most African genera don't have that many species. it's an old continent with a crowd of very adaptable and oppertunistic catfish-species, some of wich have spread through a huge area with very different environments. Synodontis schall, Auchenoglanis occidentalis, Bagrus docmac, Clarias gariepinus etc.

[Walter]

Hi Sidguppy,

That's a non-argument, because if you would be consequent you'd have to substract the three groups of catfish that have shown adaptive radiation as well; Synodontis and Chrysichthys in Lake Tanganyika and Bathyclarias in Lake Malawi.

.

that was no argument, that was a question on my part.
And of course you have to substract all species of fish of the big lakes.

Africas aquatic habitat consist of a much higher percentage of huge lakes than South American water habitats do.
And in these habitats cichlids are the by far dominating species (e.g. about 3/5 of all fish species of Lake Tanganyika).

In running water systems they are not that dominating.

Neotropical water habitats are dominated by running water systems.

So I can rephrase my question:
It could/would be interesting to compare African and South American diversity/numbers/percentages of catfish species to the numbers of all fish species in running water systems of these continents.

[sidguppy]

I think then SA would win hands down.

Africa's tropical rainforest area is much smaller than SA; hence less species.

[Walter]

Hi Sidguppy,

but the interesting point is not the total number of species, but the percentage of catfish species in comparison to all species of fish in these running water habitats.

I wonder, if this percentage would be similar.

[sidguppy]

Africa has barbs, many many many barbs, next to it's own varieties of characins.

also many more killi-fish, especially the seasonal kind (Nothobranchius, Aphyosemion etc); Polypterids, loads and loads of Mormyrids, herrings, sardines, Anabantoids, more puffers etc.

the overall non-catfish fauna in Africa is much larger than in SA, so yes I still think that even without counting the Riftlakes, the catfishfauna doesn't have the same percentage in comparison to SA.

[Shane]

Africa's tropical rainforest area is much smaller than SA; hence less species.

Macahado Allison's "Los Peces de Los Llanos de Venezuela: Un Ensayo Sobre su Historia Natural" (Fishes of the Venezuelans plains, a study of their natural history) puts forth the following figures:
Catfishes: 30% of all spp.
Tetras: 35%
Gymnotiformes: 15%
Cichlids: 10%
The remaining 10% are divided amongst killies, eels, flounders, rays, clupeiforms, etc.

Note that this is not in the rainforest. In fact, it is outside the South American rain forests (such as in Andean streams) where catfishes are most likely to be the dominant, if not only (esp. in the case of astroblepids), fish present. Collecting in the Rio Aragua, an Andean stream, for example turns up one Farlowella, two Chaetostoma spp., and a pimelodid. Non-catfishes are one sp. of pike cichlid, one Creagrutus tetra sp, and the guppy.

One additional factor that must be taken into consideration is that the description of a species as new/distinct is a human decision that may or may not reflect the actual evolutionary history (although that is certainly the goal). Future taxonomists may decide that there are really only 6-8 genetically distinct cichlid spp on Lake Malawi just as future taxonomists may decide that all long nosed and short nosed Corydoras represent distinct spp (this has already been done to some extent) and double the number of cory spp. overnight.
-Shane