Ungulate taxonomy revisited: the evidence for the splits of G&G

[overread]

It surprises me as well, I think part of what it is is when they already know many species but the animal is not standing sideways to the viewer. Photographic guides (at least good ones) tend to show more alternate angles compared to drawn.
I can see it working for species such as birds where many times you don't get an ideal view so a photograph can be closer to what you see in reality.

I agree if its a side on picture VS photo the picture is likely going to be clearer. In the end I think complimentary is the best approach, but I generally favour the drawn plates myself.

 

[LaughingDove]

It surprises me as well,

I am interested in what kinds of experienced wildlife watchers and what circumstances/needs they have so that they prefer photographic field guides.

I think part of what it is is when they already know many species but the animal is not standing sideways to the viewer. Photographic guides (at least good ones) tend to show more alternate angles compared to drawn. I can see it working for species such as birds where many times you don't get an ideal view so a photograph can be closer to what you see in reality.

I don't think that makes sense. The whole point of having drawings rather than photographs is that you can show clear angles and emphasise key identification points. I don't see this impeding identification when a bird is not at an optimum angle or view, in fact drawings can emphasise the impression that you are likely to get of a bird much better than a photograph can.

Can you suggest a good photographic guide? I don't think I know any.

I agree if its a side on picture VS photo the picture is likely going to be clearer. In the end I think complimentary is the best approach, but I generally favour the drawn plates myself.

I should note that the main reason that I'm arguing is that I disagree with the original point that more experienced birders/wildlife watchers prefer photographic guides. If anything it's probably the other way around where inexperienced birders chose a guide with pretty colour pictures and nice backgrounds over a guide with plain drawings on a plain background.

ditto for me. I've have never heard of or met any serious birders who think photographic identification guides are better than drawn ones. (That was almost literally repeating what LaughingDove said, but there doesn't seem any other way to phrase it). Maybe overread has only met English birders who don't actually need guidebooks because they already know all the birds in England? In any case, any photographic guide which covered every species in a region would be useless as a field-guide because it would be too difficult to make comparisons between species and it would be far too large and heavy. That's the whole point of drawn field-guides - having multiple similar species on one plate for ease of identification and reduction of book size.

I think you're going to struggle to find a birder who is so experienced that they can identify every vagrant wader or warbler in England but would carry around a photographic field guide

P.s. apologies for hijacking the thread a bit

 

[lintworm]

Can you suggest a good photographic guide? I don't think I know any.

The new British Bird guide is as good as it gets for photographic guides (ignoring the multitude of errors). Another very nice photographic guide is the guide to the larger mammals of Tanzania by Foley et al., this is the best African mammal field guide I have seen so far.

 

[Pertinax]

I think you're going to struggle to find a birder who is so experienced that they can identify every vagrant wader or warbler in England but would carry around a photographic field guide

P.s. apologies for hijacking the thread a bit

Most of the guides with photographs don't carry ones of the very rarest species I don't think. I've been birdwatching for many years and always use fieldguides with artist-drawn plates as they show all the definitions you need for identification. The only thing is the colour is never exactly the same as in real life- that's where a photograph may score. I think the same for Butterflies too- line drawings are best.

 

[TeaLovingDave]

Can you suggest a good photographic guide? I don't think I know any.

This is a pretty good one, to be fair:

https://www.amazon.co.uk/Crossley-ID-Guide-Britain-Ireland/dp/0691151946

 

[overread]

Crossley is the one I'd have held up as well and seems to be setting a higher quality trend. I agree that they tend to favour more common species over rarer ones and that quality photographic are rarer.

 

[Jurek7]

I can see the merit of photographic guidebooks to mammals. Colors are not so important with mammals, but more size, shape, quality of fur, which transform better on a photograph. However, I know of no birdwatcher who uses photo guidebook as a primary guidebook.

For Ungulate taxonomy and similar books, they don't consider individual variation. Anybody knows how variable can be foxes or deer (or insert any common mammal from your area) in size and shade of color. These individual or local variantsin a less known species could easily be taken for sub-specific or specific distinction.

 

[LaughingDove]

This is a pretty good one, to be fair:

The Crossley ID Guide: Britain and Ireland (The Crossley ID Guides): Amazon.co.uk: Richard Crossley, Dominic Couzens: 9780691151946: Books

Crossley is the one I'd have held up as well and seems to be setting a higher quality trend. I agree that they tend to favour more common species over rarer ones and that quality photographic are rarer.

Hmm... looking at some sample pages it seems rather good. Certainly aimed at beginners though (as it is described on the Amazon page itself) and I don't see it being the preferred book of experts .

The new British Bird guide is as good as it gets for photographic guides (ignoring the multitude of errors). Another very nice photographic guide is the guide to the larger mammals of Tanzania by Foley et al., this is the best African mammal field guide I have seen so far.

Yeah, forgot about the Tanzania one. I had actually seen that and thought it was pretty good. There aren't really many confusion species dealt with though. I'd like to see them try a photographic guide to the small rodents and shrews of Tanzania

 

[Hix]

I've been birdwatching for many years and always use fieldguides with artist-drawn plates as they show all the definitions you need for identification. The only thing is the colour is never exactly the same as in real life- that's where a photograph may score.

I have to agree. I recently went to Borneo and bought two different bird guides, and the illustrations varied widely between the two, mainly in colour - in one book the illustrations were sombre, and the other the colours were rather garish. In the end, to identify some species of rather non-descript bulbuls I ended up using photos on the internet to confirm some suspicions.

And then there are hand-drawn illustrations that are just badly drawn and don't look life-like at all. But I guess that's another story!

Umm..... we appear to have hijacked Lintworm's thread - perhaps a Mod could pull out these posts and create a new thread?



Hix

 

[lintworm]

Kob

The Kob (Kobus kob) is an antelope from west and central Africa and is the sister species to the Puku (Kobus vardoni), with which it was sometimes considered conspecific, also based on Mtdna data (Birungi 1999).

Normally three different subspecies are recognized, based on obvious differences in coat patterns and general size:

K.k. kob Senegal to CAR and DR Congo
K.k. thomasi Uganda, South Sudan west of the Nile, NE DR Congo
K.k. leucotis South Sudan east of the Nile and Gambella, Ethiopia

(source wikipedia, based on IUCN red list)

K.k. kob is a lot lighter than the 2 more eastern subspecies. Kingdon gives the weight of the males as 63 kg (58-67, n=3), whereas this is on average 96.9 kg (88.4-107.9) for K.k. thomasi. K.k. leucotis is different from the other subspecies in the fact that dominant males become almost black, but still with clear white markings on the head.

G&G split this species into 4, elevating all subspecies to species level and naming the the kob east from Lake Chad loderi, just as the ones from Sudan west of the nile. According to them thomasi is restricted to Uganda (and formerly also W Kenya & Tanzania). Interestingly in Castello's bovid field guide, the specimens from W Sudan belong to thomasi, not to loderi.

Sample size
No sample sizes are reported for the skins seen

Skulls (males/females)
kob (12-19/4)
loderi
Central Africa (29-32/1)
W Sudan (11-13/2)
thomasi (7-11/3)
leucotis (2-3/0)

Horns (males only)
kob (16-21)
loderi
Central Africa (30-33)
W Sudan (13)
thomasi (9)
leucotis (3)

Sample sizes for females are very small, sample sizes for males are quite good, except for leucotis

Skins

kob skins are orange-fulvous to tawny, with a whitish ring arounds each eye and base of the ear. Ears are fulvous on the back, with indistinct black tips. Blackish stripe down the front legs is indistinct and thin, usually interrupted by a white band above the hoofs. loderi is said to be tawny, with a black line down the front of the forelegs, thinner black line down the hind shank. White of the pasterns is separated in 2 by a thin black line. Leg lines are of varying thickness. Inside of the ears tends to be white. thomasi is said to have deep black and distinctive markings, with the white area around the eye also much larger than loderi, the back of the ears is also showing a tendency to whiteness. leucotis males are said to be deep black, with large white patches around the eyes and ears and as a continum through the muzzle, chin, upper throat, chest and inner side of the upper portions of the limbs. Females and brown colored males have more white around the eye and the leg stripe is more light brown, reaching the shoulder.

Skulls
leucotis skulls overlap significantly with thomasi skulls and often fall completely within the variation of thomasi, though on average the 2-3 samples are slightly smaller, they cannot be distinguished from thomasi. thomasi skulls are significantly larger than kob skulls, up to the point that for the 5 measurements listed, there is only real overlap in the nasal breadth between the two taxa. The loderi samples from C Africa and W Sudan are indistinguishable and in size they are close to the thomasi samples, though on average slightly smaller. For the few female skulls the same pattern is visible with the loderi and thomasi samples being practically the same, whereas the kob sample is smaller and falls completely outside the (limited) variation of loderi and thomasi.

Horns
On average leucotis has longer and wider, but thinner horns than thomasi, but variation is large within the taxa and leucotis sample size is very small, they could be well distinct though. kob has the smallest and thinnest horns on average and given the available sample sizes they might well be significantly smaller than the other taxa, when tested statistically. Overall horns of loderi and thomasi are similar, though thomasi may have slightly thicker horns (3 mm at the base...)

Additional data
There has been some genetic work on the different Kob taxa and it is quite confusing. The first study (Birungi & Arctander 2000) find evidence for 2 different lineages based on Mtdna, one predominant in northern populations, the other predominant in western and southern populations. Both lineages overlap in Semliki and Murchison Falls, Uganda. Lorenzen et al. (2007) took samples from all 3 subspecies recognized at that point (leucotis, kob, thomasi), they find the the thomasi in Murchison Falls are genetically very similar to leucotis, though phenotypically they are indistinguishable from the thomasi in nearby Queen Elizabeth NP. They propose that Kob were separated in 2 refuge areas, one in the west and one in the east of their distribution and that the Murchison Falls samples are a result of hybridisation between the western kob+thomasi and the eastern leucotis populations. The authors also suggest that based on their results gene flow between thomasi and leucotis is still going on. G&G acknowledge this study but only highlight that there is some differentation between the 3 taxa based on the microsatellites. They do not integrate the other (for them unfavorable) results in their own morphological findings, except by proposing that the leucotis phenotype is selected against in Murchison Falls (or selected in favor of the microsatellites).

Summarizing
Though at first glance the Kob represents a species with 3 very different subspecies, the western being very small and the eastern having a completely different coloration in dominant males, their evolutionary history is making things much more complicated. There is evidence of two refuges, with the eastern one corresponding to leucotis and the western to loderi+kob+thomasi, but there is also evidence for subsequent mixing between the refuges. Based on genetic and morphological work I see no reason to recognize the loderi taxa, even as a subspecies, as it is almost completely similar to thomasi. I also have my doubts about elevating any of the other taxa to species level either. Though morphologically distinct, there is hardly any genetic difference between kob and thomasi and though leucotis is also morphologicall distinct, it has shown to probably still have active gene flow with thomasi. So for now I see no reason not to continue threating all normally recognized subspecies as subspecies and retain loderi within thomasi. More genetic work might slightly change the picture and the leucotis is well worth conserving as a seperate genetic lineage, but given the current data not worthy of species status in my opinion.

Kobus kob kob

@KevinVar, Mole NP, Ghana

@KevinVar , Mole NP, Ghana


Kobus kob thomasi

@Hix , Murchison Falls NP, Uganda

@Giant Eland , San Diego Safari Park, USA

References

Birungi, J. 1999. Phylogenetic relationships of Reduncine antelopes (subfamily: reduncinae) and population structure of the Kob (Kobus kob). PhD thesis Makerere University, Uganda

Birungi & Arctander 2000:
Large sequence divergence of mitochondrial DNA genotypes of the control region within populations of the African antelope, kob (Kobus kob)

Lorenzen et al. 2007:
https://www.researchgate.net/profil...ene-refugia-of-the-kob-antelope-Kobus-kob.pdf

Next stop: Lechwe

 

[lintworm]

Southern lechwe

The Southern lechwe (Kobus leche) is a reduncine antelope, with a distribution in Southern Africa focused on Zambia and is closely related to the Nile lechwe from (South) Sudan and Gambella. Traditionally Southern lechwe have been treated as one species, with four subspecies, of which one, robertsi, is extinct. However recently the Southern Lechwe from Upemba in the DR Congo have been described as an additional (sub-)species.

Currently the following extant subspecies are recognized:

K.l. leche (Red lechwe) in Botswana, Namibia, Angola and Zambia
K.l. kafuensis (Kafue lechwe) in the Kafue Flats, C Zambia
K.l. smithemani (Black lechwe) in the Bangweulu Swamps and Chambeshi system, N Zambia
K.l. anselli (Upemba lechwe) in the Upemba wetland DR Congo

The first three subspecies are also kept in European zoos.

For a distribution map see: IUCN Red List maps

If you zoom in enough, subspecific distribution is also visible

Naturally G&G elevate all these taxa, including robertsi to species status. Their book Ungulate Taxonomy only covers Lechwe very briefly and uses data only from a paper by Cotterill in 2005, who described the Upemba lechwe in that paper and gave an in-depth analysis as to why the different Southern lechwe taxa are distinct. I will also purely draw from his paper, which can be found here:
The Upemba lechwe, Kobus anselli: An antelope new to science emphasizes the conservation importance of Katanga, Democratic Republic of Congo (PDF Download Available)

Sample size

As robertsi is extinct and available sample size is extremely limited, I will leave this taxa out for this post.

No sample size for skins has been reported

Skulls & Horns (males only)
anselli (13-15)
kafuensis (53-75)
leche (26-41)
smithemani (26-37)

Sample sizes are relatively large and should thus be large enough to reliably detect significant differences.

All these sample sizes and results on skulls and horns are based on Table 1 in Cotterill 2005.

Skins
leche, kafuensis & anselli all have a similar light to bright red-brown skin color. Adult male kafuensis develop dark shoulder patches at the top of their thick black leg-stripes, their white throat stripe is continuous with the white chin and the white underside, just as in leche. In leche the dark leg stripe is also broad, but only up to the elbow. The back of the ears is light red-brown in leche, but white in anselli. anselli only has a thin leg stripe and only on the lower leg and their white throat stripe is reduced or absent. smithemani adult males are different from all other taxa in that they have many dark hairs on the face, neck, shoulder and flanks. Their white throat stripe is also continuous.

Horns
kafuensis has significantly larger horns, a wider horn curve and a wider span than the other 3 taxa and basically all other taken measurements for horns were also larger in kafuensis than for the other 3. leche has a larger basal horn width and basal horn spread than smithemani and anselli, though smaller than kafuensis. smithemani and anselli are very similar on univariate measurements, except that inner horn width is slightly larger in smithemani, though with a MANOVA the basal horn width and horn length are also tested to be significantly larger in smithemani (Table 2).

Very short statistical interlude
A MANOVA is a multivariate analysis (as opposed to univariate) and tests multiple dependent variables (in this case horn measurements) at once against an independent variable (in this case the taxa), as opposed to a univariate test that can only deal with one at a time. A MANOVA can thus also test relationships between different dependent variables. Myself I normally work mostly with univariate statistics, so I am not very familiar with MANOVA, but given the results they are more likely to give significant differences as opposed to multiple ANOVA tests. Interestingly MANOVA are apparently not the best test when you dependent variables are largely correlated with each other (which they 100% are in this study), so the validity of these outcomes vs. a normal ANOVA could be questioned, especially as he does not provide r values for correlations between dependent variables... So for the skulls I will ignore the MANOVA results again as it's results might be misleading

Skulls
On average most data on skull characteristics are somewhat similar, though kafuensis and leche are generally larger as compared to anselli and smithemani and this difference is probably significant for some of the 10 measured skull variables, though the actual difference is only in millimeters and there are no huge differences, but they are still distinct.

Cotterill then continues with other Multivariate statistic techniques to plot the different taxa against a number of skull and horn characteristics combined and with a combination of 10+ characteristics he is able to differentiate them. The sample size used for this is however not sufficient to draw conclusions, as sample size of individual taxa is hardly larger than of the number of variables included in the analysis. This can result in the PCA showing differences that are not there.

Additional data
There have been several studies looking at genetic differences between the different Southern Lechwe taxa, but both Birungi & Arctander (2001) and De Meneghi et al. (1995) find only very little difference between the different taxa, though distinct the absolute difference is very small. Cotterill as part of his PhD thesis (Cotterill 2006) also worked on genetics and his thesis is well worth a read as it also covers the geological history of the Southern lechwe distribution and much more. He shows that kafuensis is a recent offshoot from leche and has been separated 35- 240 thousand years and that the other taxa have already been separated longer, possibly since the early Pleistocene. The different taxa have probably been relatively isolated from each other ever since and the explanation for that may very well lie in the climatic conditions of the time and the dependence of the species on wetlands.

Summarizing
The 4 extant Southern lechwe taxa are somewhat distinct, though the multivariate analysis is not good enough to trust the conclusion that the taxa are indeed distinct. The taxa do however not interbreed and are spatially separated, though most probably they still can interbreed in captivity. Though they have been genetically separated for quite some time, the morphological and genetical differences are rather small, so I would favor that the Southern lechwe is retained as one subspecies, with 4 clearly defined subspecies.

This Southern lechwe is however a tricky example as they seem to hover a bit between the classical subspecies and classical species definitions and though it is sound to protect each taxa from a conservation perspective, it really could divert money to other rare taxa that are more distinct from it's sister taxa....

Kobus leche leche

@Maguari , Khwai Community Area, Botswana

Kobus leche kafuensis

@Patrick87 , Zoo Berlin, Germany

Kobus leche smithemani

@Giant Eland , Safari Madrid, Spain

No pictures of anselli have been uploaded to the gallery.

References

Birungi & Arctander 2001:
Molecular Systematics and Phylogeny of the Reduncini (Artiodactyla: Bovidae) Inferred from the Analysis of Mitochondrial Cytochrome b Gene Sequences

Cotterrill 2005: The Upemba lechwe, Kobus anselli: An antelope new to science emphasizes the conservation importance of Katanga, Democratic Republic of Congo (PDF Download Available)

Cotterrill 2006:
The evolutionary history and taxonomy of the Kobus leche species complex of South-Central Africa in the context of Palaeo-drainage dynamics. PhD Thesis, University of Stellenbosch

De Meneghi et al. 1995:
http://rcin.org.pl/Content/12357/BI002_26822_Cz-40-2_Acta-T40-nr25-303-308_o.pdf


Next stop: Waterbuck

 

[lintworm]

Now it is time for a species most of you will be familiar with, not only with the species, but also with both widely recognized subspecies, as they are present in a large number of zoos.

Waterbuck

The waterbuck (Kobus ellipsiprymnus) is among the heaviest antelope and has a wide distribution throughout tropical Africa, south of the Sahara it is only absent from Gambia, Mauritania, Equatorial guinea, Djibouti and Lesotho. Traditionally two subspecies have been recognized, allthough they have sometimes be described as 2 species:

K. defassa (Defassa waterbuck) west, north-east and central Africa, generally west of the great rift valley
K. ellipsiprymnus (Ellipsen/Common waterbuck) south-east Africa, generally east of the great rift valley.

(source: wikipedia)

Some authors also recognize other subspecies other than these two, but they are hardly followed and are also not specifically mentioned by G&G. They do however split the species by elevating both subspecies to species status.

Sample size
No sample sizes for skins are reported

Skulls & horns (males only)

defassa
W Africa, W Sudan 16-19
Ethiopia, NE Sudan 5-7
Rutshuru 8-12
Ankole 4-7
Mongalla, Karamoja 3-4
Serengeti 6-11
Laikipia, Baringo 3-5
Zambia, Angola 7-10

ellipsiprymnus
S Africa 5-6
Gorongoza 1
Malawi 1
S Tanzania 1-2
Tsavo 1
Thika 2-3

Sample sizes for defassa are relatively large, especially when pooled. Sample sizes for ellipsiprymnus are rather small, especially when treated separately.

Skins
The main distinguishing feature is that ellipsiprymnus has a white ring around the rump, which defassa lacks. Additionally the coats of defassa are generally a bit more rufous brown, where ellipsiprymnus is more grey brown.

Horns & skulls
There is some variation between different localities for each taxon, but overall there is no consistent differences and differences within measurements are also small when comparing populations. The only thing that springs out is that the ellipsiprymnus from S Africa are generally larger than the other localities (including defassa). Sample sizes are very small though for other ellipsiprymnus localities and this fits rather nicely with Bergmann's rule that states that animals away from the equator tend to be bigger than their tropical relatives. There may thus well be a cline from the tropics to S Africa.

Additional data
This is where it gets interesting. Kingswood et al. (1998) reported different karyotypes for ellipsiprymnus (2n=50-52), whereas defassa (2n=53-54), there is thus a chromosomal difference between both taxa that might limit gene flow. There are however well-known cases of hybridisation between both taxa in Kenya (Nairobi NP, Samburu & Tsavo) and a study from Lorenzen et al. (2006) showed this, despite significant mtDNA differences between both taxa. These hybridization events are probably very recent, somewhere after the last ice age, but possibly within the last few centuries. They hypothesize that the original Waterbuck population got split into two refugia, one in the west and one in the southeast, leading to the development of 2 separate taxa in a time without any gene flow between populations. When conditions where favorable again both taxa extended their distributions and met in Kenya, where they hybridised. It has been hypothesised that the differences in chromosomes might hinder gene flow and is an emerging reproductive barrier.

Summarizing
There are consistent differences between both taxa in terms of morphology, karyotype and Mtdna. There is however some hybridization going on between both taxa. Heller et al. (2013) use the last point, together with the only relatively small Mtdna difference (though significant) as reasons why G&G have split this species without good reason. But as explained in the HBW checklist by Del Hoyo & Collar, a narrow hybridisation zone can also be considered as evidence that there is resistance against hybridizing, so should be used as an extra argument for splitting, as opposed to the biological species concept. Overall I feel there is quite good evidence for splitting the waterbuck into two species, with a narrow hybridization zone in central Kenya, as both taxa clearly are 2 separate evolutionary lineages that are diagnosable outside their hybrid zone on multiple criteria. This would then lead to the recognition of the following species: Common/Ellipsen waterbuck (Kobus ellipsiprymnus, Ogilby 1833) and the Defassa waterbuck (Kobus defassa, Rueppell 1835)

next: the remaining reduncine antelope.

Kobus ellipsiprymnus

@Maguari , Moremi Game Reserve, Botswana

@vogelcommando , Burgers' Zoo, Netherlands


Kobus defassa

@Patrick87 , Berlin Zoo, Germany

References
Heller et al. (2013):
Is Diagnosability an Indicator of Speciation? Response to “Why One Century of Phenetics Is Enough” | Systematic Biology | Oxford Academic

Lorenzen et al. (2006):
Hybridization between subspecies of waterbuck (Kobus ellipsiprymnus) in zones of overlap with limited introgression

Kingswood et al. (1998):
https://watermark.silverchair.com/a...P4BdY8NqPkDwPWfIc1aKBXEk0k0cG9NR2zVkIohFwFqMY

 

[lintworm]

Before moving on to the next tribe, there are a few Reduncine antelopes that have not been mentioned yet

Grey rhebok
The Grey rhebok (Pelea capreolus) is the most primitive of this tribe and occurs only in grasslands in S Africa, Swaziland and Lesotho. No subspecific variation has been described and thus remains a monotypic species (and genus) within G&G's proposed taxonomy.

@Giant Eland , Larry Johnson's Place, USA

@Giant Eland , Larry Johnson's Place, USA

Nile lechwe
The Nile lechwe (Kobus megaceros) is a monotypic species, sister to the Southern lechwe that occurs in Sudan and Gambella, Ethiopia. No subspecific variation has been described and this taxon is treated as a monospecific species in all taxonomic classifications.

@vogelcommando , Safaripark Beekse Bergen, Netherlands

@vogelcommando , Safaripark Beekse Bergen, Netherlands

Puku
The Puku (Kobus vardoni) is most closely related to the Kob and has sometimes been merged into one superspecies. Based on morphology and Mtdna Puku are now widely accepted as a species in it's own right. Currently this species from south-central Africa (from Tanzania to northern Botswana) is mostly treated as monospecific, though it is mentioned in Mammals of Africa that research into geographical variation within Puku has not been investigated.

G&G analyse skulls and horns from 4 different regions (Botswana, Tanzania, W+C Zambia, Luangwa Valley + Malawi). Overall sample sizes are very small, with only for females from W+C Zambia a sample size larger than 3. So any differences might well be due to the extremely small sample size. The only difference they mention is the broad skull with small nasals from Tanzania. But the sample for the female falls completely within the variation of the Luangwa Valley + Malawi sample and for the males only 1 sample per region was taken, except for 2 horns from Botswana. So any comparison there is meaningless...

In Castello's Bovids of the World 2 subspecies are recognized: vardoni from Tanzania, Malawi and N Zambia and senganus from Namibia, Botswana, Angola and S Zambia. senganus is said to be smaller and darker colored, especially on the head. But the average sizes given do not support this and the pictures also do not show any difference, except that the pictures for senganus are taken in bright sunlight, so look more reddish...

Overall I do not see any evidence for any subspecific distinction in Puku and it is well treated as a separate species, though it has been argued relatively recently by Arctander & Birungi that genetic data do support including Puku into the Kob...

No pictures of Puku have been uploaded to the gallery yet.

Next will be the Oribi.

 

[lintworm]

Though there has been a pause, I will soon post the next part. For now there is something else to read, not about ungulates but primates, the other mammal group that is dominated by splitters. Ian Tattersall, the godfather of lemurs, wrote a paper 10 years ago in which he examines the evidence for all the splits that have taken place in Madagascar's lemurs. Even though the paper is now 10 years old it is still very readable and kind of similar, with what I am trying to do on ungulates in this thread, but then done by a real expert in that field:

http://www.globaldiv.eu/SummerSchool/docs/Groenweld/tattersall_2007.pdf

 

[LaughingDove]

Ian Tattersall, the godfather of lemurs

That would make a great email signature!

 

[lintworm]

Oribi

The Oribi (Ourebia oerebi) is the only species in the Oribini tribe, which in turn is sister to the Reduncine antelope discussed before. The oribi is a smaller species of antelope that occurs widely throughout Africa and many subspecies have been described. The validity of many of them is debatable and have not been tested with molecular work

In Kingdon's Mammals of Africa the following 12 subspecies are described, of which one is extinct:

O. o. ourebi South Africa & S Mozambique
O. o. hastata Malawi, Zimbabwe, N Mozambique, E Zambia, SE Tanzania
O. o. rutila Angola Botswana, Namibia, W Zambia and perhaps S DR Congo
O. o. cottoni SW Kenya to C Tanzania
O. o. masakensis S Uganda, Rwanda, NW Tanzania
O. o. aequatoria N Uganda, South Sudan
O. o. kenyae Lower slopes of Mt. Kenya, now extinct
O. o. haggardi coastal Kenya, S Somalia
O. o. gallarum C Ethiopia
O. o. montana W Ethiopia, SE Sudan
O. o. goslingi N DR Congo
O. o. quadriscopa Senegal to Nigeria

Maybe somewhat surprisingly G&G do not recognize 12 species, but "only" 4 species of Oribi.

O. ourebi, all Oribi south of the Zambezi river, so including ourebi and parts the range of hastata
O. hastata, including haggardi, kenyae, cottoni, rutilus, masakensis
O. montana, including goslingi, gallarum, aequatoria
O. quadriscopa

G&G note that even within their proposedspecies the pelage is rather variable, just as other characters, so they clearly state that this is just provisional.

Sample size

Skins
Sample size for skins seen is not reported

Skulls (male/female)
quadriscopa (4-6/8-9)

goslingi (6-11/5-13)
aequatoria (8-20/7-12)
montana (25-35/9-15)
gallarum (10-19/6-7)

cottoni (10-22/2-6)
pitmani (17-20/9-10)
kenyae (6-7/2)
haggardi (5/1)
hastata (6/2)
W Zambia (8-16/3-5)
rutila (3-6/3)

ourebi (2-3/2-3)

Horns (males only)

quadriscopa (1)

goslingi (9)
aequatoria (20)
montana (36)
gallarum (18)

cottoni (21)
pitmani (16)
kenyae (7)
haggardi (5)
hastata (5
W Zambia (17)
rutila (7)

ourebi (3)

Sample sizes are good for hastata and montana, but not for ourebi and quadriscopa which have a small sample size.

Skins
quadriscopa is said to be quite distinct, though pelage colour varies, what actually makes this species distinct is not mentioned by G&G. In Castello's Bovid field guide it is mentioned that the pelage is speckled gray fawn to deep orange and the tail is black. montana does usually not have a black tail and normally also not a a darker one, except for the Ankole population (Uganda) pelage color varies again, according to Castello form dark gray to orange buff. hastata has a black tail even on the underside and is distinguished by the much heavier banding of richer more yellow hairs. ourebi has a more rufous colour and also a black tail. The pictures in Castello's book here look misleading though, due to different light conditions (and maybe even photoshop).

Skulls
The skull of quadriscopa is said to be quite distinct and Gt 1 is indeed on average smaller than in other taxa, sample size is small though and data fall almost completely within the variation of other taxa. hastata and montana are almost identical with regards to their skull features and can't be distinguished. Sample size for ourebi is very small and falls within the variation of hastata. ourebi is said to be the only taxa in which the males are larger than the females, but this is not apparent from their small dataset on ourebi and is only visible for Gt 1, where it seems that the female dataset has 1 outlier, making it seem smaller on average...

Horns
Horns of quadriscopa and hastata are indistinguishable. Horns of ourebi and montana are also indistinguishable, though their horns are on average slightly larger than those from quadriscopa and hastata, the variation is however huge and no distinctions can be made.

Additional data
I have been unable to find any genetic studies looking at Oribi taxa and data on this species taxonomy seems to be lacking at all.

Summarizing
Though G&G have reduced the Oribi from 12 to 4 taxa there does not seem to be any evidence on which their proposed species can really be distinguished. There are some small differences, mainly between quadriscopa being very slightly smaller on average and hastata often missing the black on the tail. But there is no reason to recognize more than one species of Oribi. Even recognition of subspecies is difficult based on the provided data, given the huge variation within populations when it comes to pelage color and size. For now recognizing the proposed 4 species as subspecies could be used as a working hypothesis until further testing is done, but even that feels premature, when the only real difference is whether the tail is black or not and whether that is the case on both sides of the tail...

Oerebia oerebi hastata

@Hix , Murchison Falls NP, Uganda

Oerebia oerebi montana

@lintworm , Senkele Wildlife Sanctuary, Ethiopia

Next the Antilopini tribe, where I will start with the Dorcas gazelle and thus also venture out of Africa.

 

[lintworm]

Dorcas gazelle

The Dorcas gazelle (Gazella dorcas) is a widely distributed gazelle occuring in throughout the Sahara and Sahel regions and into the Arabian peninsula and the Horn of Africa. the extinct G. saudiya was considered a subspecies of the Dorcas gazelle, but is now considered a separate species.

In Mammals of Africa the following subspecies are recognized:
G.d. pelzelni Somaliland, Djibouti Ethiopia and Eritrea
G.d. isabella Sinai, Israel, Egypt, Sudan to N Eritrea and N Ethiopia
G.d. becarrii Highlands of the upper Anseba river in Eritrea
G.d. osiris Sudan west to Mauritania, Western Sahara, Morocco and Algeria
G.d. massaesyla Algeria and Northern Morocco
G.d. dorcas Egypt, Sudan, Chad, Libya & Tunisia

G&G retain all of these subspecies, but elevate G.d. pelzelni to species status.

Sample sizes
No sample sizes for horns, skulls or skins are reported.

Skins
pelzelni is overall relatively similar in colouration and skin markings, the colour is said to be a bright sandy ochre, which lighter than in isabella, the neighbouring taxon

Skull & horns
Horns are reported to be longer than in pelzelni in other Dorcas gazelles and contrary to other Dorcas gazelles the tips do hardly/not point inwards at the tips. pelzelni is said to be larger and longer legged than the other Dorcas gazelles.

Additional information

In the same year as G&G published their book Ungulate Taxonomy, Lerp et al. published a study looking at Mtdna of Dorcas gazelles across their range. Their results indicate very little intraspecific genetic structure and conclude that all Dorcas gazelles can be treated as one evolutionary significant unit. pelzelni was placed within the other Dorcas gazelles and based on Mtdna does not warrant recognition as a separate species. They mention that the differences between pelzelni and the other Dorcas gazelles may be due to adaptations to habitats with higher humidity in the former. Lerp et al. could also not find any evidence for the recognition of separate subspecies within the Dorcas gazelle.

For now there is no reason to recognize the Pelzeln's gazelle as a separate species and even all the originally recognized subspecies might not warrant subspecies status, even though they differ slightly morphologically.

Gazella dorcas "osiris"

@carlos55 , Zoo Barcelona, Spain

@TeaLovingDave , Wilhelma Stuttgart, Germany

Gazella dorcas "dorcas"

@devilfish , Giza Zoo, Egypt

Gazella dorcas "massaesyla"

@Zoo Tycooner FR , La Vallée des Oiseaux, Morocco

Gazella dorcas "neglecta" (generally included in osiris)

@Giant Eland , San Diego Safari Park, USA

Gazella dorcas "isabella"

@Giant Eland , Al Ain Zoo, UAE

Gazella dorcas unknown origin

@gust1 , Kuwait Zoo, Kuwait

next: Thomson's gazelle

Lerp et al. 2011: https://s3.amazonaws.com/academia.edu.documents/45066765/A_phylogeographic_framework_for_the_cons20160425-4264-1v5bzvi.pdf?AWSAccessKeyId=AKIAIWOWYYGZ2Y53UL3A&Expires=1512235823&Signature=r6PdlLMovq+pQwfH+TOrkloWfjY=&response-content-disposition=inline; filename=A_phylogeographic_framework_for_the_cons.pdf

 

[lintworm]

Because I will have to print some documents before I can comfortably write about Thomson gazelle splitting, I will continue with the Gerenuk.

As a sidenote: I won't cover anything on the treating of Mongalla & Eritrean gazelle as distinct species, as this is also recognized by the IUCN red list and Kingdon's Mammals of Africa. Additionally I will also not cover the split of the Grant's gazelle in three separate species, the reasons for this split are clearly shown in the following article:
https://www.researchgate.net/profile/Eline_Lorenzen/publication/225617816_Three_reciprocally_monophyletic_mtDNA_lineages_elucidate_the_taxonomic_status_of_Grant's_gazelles/links/5553c32708ae6fd2d81f232c.pdf

Gerenuk

The Gerenuk (Litocranius walleri) is an interesting looking gazelle which extremely long legs and neck, which it uses to reach browse that is out of reach for many other browsers in it's range. Gerenuks are confined to arid and semi-arid areas of Northeastern Africa, ranging from northern Tanzania to Djibouti and Ethiopia in the north.

Traditionally two subspecies have been recognized, though there has been and still is some discussion on the exact ranges of both.

Kingdon's Mammals of Africa mentions the two subspecies and their distributions based on (Grubb 2002) as follows:

L.w. walleri NE Tanzania, Kenya, S Somalia and S Ethiopia
L.w. sclateri N Somalia, Djibouti and adjacent parts of Ethiopia

G&G recognize both subspecies as distinct species, but mention different distributions, leaving sclateri confined to N Somalia and Djibouti and a minimal part of Ethiopia (Though in Castello's field guide this is extended again, but not in HMW).

Sample sizes
G&G do only refer to measurements already published in Grubb (2002), which are as following:

Skins

walleri 7
sclateri 12

Skulls (males/female)
walleri (22-27/6-7)
sclateri (9-12/4-5)

Sample sizes are thus not large, but for male skulls especially big enough to derive any conclusions from.

Skins

Contrary to what has been published earlier, Grubb (2002) did not find any colour differences between the two taxa. There was however a consistent difference between walleri and sclateri, walleri has a reversed band of hair on the dorsal midline of the neck, whereas sclateri has none. This difference was consistent on all skins.

Skulls

Both in males and females skulls of sclateri are distinctly larger than in walleri. Greatest skull length and preorbital length do almost completely not overlap (94-96%) for both male and female, in other skull characteristics this difference is also visible, but the percentage of overlap is smaller. There were no differences in skull measurements between different localities of walleri.

Additional data

Unfortunately there have been no studies done on the genetics of Gerenuk and there is no additional data available on separation on both Gerenuk taxa.

Summarizing

There are some interesting differences between both sclateri and walleri and these differences are consistent, even though they are practically impossible to distinguish in the field. sclateri is significantly larger (though we are talking about 1.5 cm (about 7.5%) difference in skull length) and also misses the characteristic reversed band of hairs on the neck that is present in walleri. Grubb (2002) notes that there is no apparent cline visible in the data, though there were no samples from the Ogaden region (SE Ethiopia). It would be interesting and of high conservation value to look at whether there is any genetic difference between the two Gerenuk taxa, as there might be more cryptic diversity than is apparent at first sight and genetics could help to answer the question whether we should treat sclateri and walleri as subspecies or as distinct species.

So what about the actual distribution of sclateri?

There is some unclarity about the actual distribution of both sclateri and walleri and there is no definitive answer to it. Grubb (2002) suggest that the Eastern highlands of Ethiopia (Chercher mountains) and the Golis mountains of Somaliland are/were the barrier between both taxa given the samples he analysed. In both Antelopes of Africa and Castello's Bovid field guide the range is extended well into Puntland (NE Somalia). In HMW and G&G's in Ungulate Taxonomy is contracted to only cover Somaliland, Djibouti and only just over the border in Ethiopia.

Based on the sample locations from Grubb (2002), all these other maps do seem false as he has sclateri samples from well into Ethiopia and a walleri sample origininated from close to Puntland (Galkayo).

Based on the best evidence currently available, which are Grubb's samples, sclateri thus ranges in Somaliland north of the Golis mountains, Djibouti and large parts of Ethiopia north of the Chercher mountains and east of the northern highlands, thus including Afar region (where it is absent from Awash NP, though it is sometimes listed as if it occurs there) and the most northern part of the Ethiopian Somali region. This is not a conclusive answer though and the final word is yet out, but finding an answer will be difficult as this region is currently relatively unstable...

Litocranius walleri walleri

@Patrick87, Tierpark Berlin, Germany

@Kakapo , San Diego Zoo, USA


Next will be the Thomson's gazelle or the springbok

Reference
Grubb 2002: Types, type locality and subspecies of the gerenuk Litocranius walleri (Artiodactyla: Bovidae) - Grubb - 2006 - Journal of Zoology - Wiley Online Library

 

[lintworm]

Thomson's gazelle

The Thomson's gazelle (Eudorcas thomsoni) is one of Africa's most famous antelope and has a surprisingly small distribution in Kenya and Northern Tanzania, it is however a common species within it's range.

(source: Wikipedia commons)

Mongalla gazelle and Eritrean gazelle have been considered conspecific with the Thomson's gazelle, but are generally accepted as distinct species. Interestingly Groves once treated Thomson's gazelle as a subspecies of the Cuvier's gazelle of Northern Africa, but that was not followed widely.

Kingdon recognizes two subspecies
E.t. thomsoni Rift valley and east and south of a line from the Lake Eyasi escarpment to Speke's bay (Lake Victoria)
E.t. nasalis west of the rift valley and north of the line described above, so in practice confined to the Serengeti/Ngorongori/Mara ecosystem

G&G recognize both subspecies as distinct species.

Sample sizes

G&G do not mention where their data comes from and what samples sizes were used to identify the mentioned differences in skulls, horns and skins for both taxa. The only data I could find in which differences between nasalis and thomsoni are described is a publication by Groves from 1969 (reference given below). I assume that these are the data on which he based his descriptions in Ungulate Taxonomy and as these are the only data available I will base my conclusions on them. It is annoying that G&G do not refer to this paper at all when describing nasalis and thomsoni so you have to find out yourself where their proposed differences are based on... There is supposedly also a good paper by Brooks (1961), but this one is not available at all online, if anyone has access to this paper called: study of the Thomson's gazelle (Gazella thomsonii Gunther) in Tanganyika [1961].

Skins
No sample sizes reported

skulls (males/females)
thomsoni (18-20/9-10)
nasalis (6-69/ no sample size given)

Horns
thomsoni (21/12)
nasalis (68/presumably 1)

So especially for female nasalis the known sample size is extremely small, for the others it seems relatively ok.

Skins

Several differences in pelage are described between nasalis and thomsoni. nasalis is said to have darker face stripes, a more prominent nose spot, a more extensive white crown and broader pygal stripes (stripes next to the white buttocks), all these descriptions are based on Brooks (1961). Brooks (1961) also said that nasalis is redder in colour but G&G did not find any evidence for that.

Horns

Male horns are of similar length (307 vs 308.3 on average for thomsoni and nasalis respectively), but female horns are said to be much shorter relatively in nasalis as compared to thomsoni. There seems however to be only 1 measurement for nasalis (99.6 mm) and for the 12 thomsoni samples only the mean (120.0) is given and not the standard deviation, which is often large when it comes to horn sizes. nasalis horns are also said to be less widely apart on the top but there is no data available to back up that claim.

Skulls

nasalis is said to be much smaller in both sexes but only in skull length in both sexes and the preorbital length in males this result will be significant, overall differences are relatively small though.

Additional data

There is no data on possible genetic differences between the two taxa. Though Castello's field guide report a body length for female nasalis of 70-90 cm whereas this is 89-107 for thomsoni. Shoulder height and weight are extremely similar though for both taxa, so I have some doubts about the validity of those measurements. Total length and shoulder length as given in Kingdon's Mammals of Africa (with a sample size of 58 (40 males, 18 females)) show longer total length and lower shoulder heights on average with different margins than Castello, so it would be interesting to see where Castello got his data from. Kingdon also notes in his East African Mammals (1982) a north-south decline in average measurements (though I do not have access to that book myself)

Summarizing

Though there are some slight differences in pelage, the rest of the current data supporting a split is pretty weak, there is no compelling evidence that indicates different horn sizes in females or different overall size or different shape of the horns, though skulls of nasalis, from the Serengeti are slightly smaller. Interestingly more Serengeti taxa show this, Topi and Wildebeest from the Serengeti are also slightly smaller than their conspecifics nearby. I would have loved to see more data, but based on the current evidence I see absolutely no reason to split the Thomson's gazelle into two separate species, retaining both taxa as subspecies is fine to show the slight differences between them.

Eudorcas thomsoni thomsoni

@lintworm , Lake Nakuru NP, Kenya

Eudorcas thomsoni nasalis

@Hix , Serengeti NP, Tanzania (juvenile)


next: the Springbok.

Reference

Groves (1969): http://www.zobodat.at/pdf/Zeitschrift-Saeugetierkunde_34_0038-0060.pdf

 

[Chlidonias]

There is supposedly also a good paper by Brooks (1961), but this one is not available at all online, if anyone has access to this paper called: study of the Thomson's gazelle (Gazella thomsonii Gunther) in Tanganyika [1961].

I just googled it and it seems to actually be a book (147 pages), number 25 in the Colonial Research Publications series.