A working approach to subspecies distinctions in the steenbok, Raphicerus campestris

@alanhorstmann @tonyrebelo @jakob @jeremygilmore @ludwig_muller @jwidness @colin25 @geichhorn @henrydelange @koosretief @michalsloviak @alexdreyer @chewitt1 @oviscanadensis_connerties @capracornelius @tandala

The subspecies of the steenbok (Raphicerus campestris, https://www.inaturalist.org/taxa/42375-Raphicerus-campestris) were revised by Groves and Grubb (2011, Ungulate Taxonomy). However, please see https://www.zoochat.com/community/threads/ungulate-taxonomy-revisited-the-evidence-for-the-splits-of-g-g.467230/page-3.

The treatment of Groves and Grubb largely follows that of Roberts A (1951, The Mammals of South Africa).

Until now, iNaturalist has avoided the problem of subspecies, by simply distinguishing the widely disjunct East African form, neumanni, from the nominate form of southern Africa.

The current approach, in iNaturalist, does not do justice to the subspecific distinctions in R. campestris. This is because

  • the East African form does not look particularly distinctive, despite its geographical disjunction,
  • the southwesternmost form, of Western Cape, is obviously distinctive despite grading continuously with other forms to the north and to the east, and
  • the form penetrating the edges of the Namib desert is so ecologically extreme that it should be assumed to belong to an arid-adapted subspecies, unless proven otherwise.

Therefore, what seems most plausible is some compromise between the current 'lumped' approach, of only two subspecies, and the old, 'split' approach in which seven subspecies are recognised, viz.

  • campestris (southwestern parts of South Africa)
  • fulvorubescens (Eastern Cape and adjacent areas)
  • natalensis (eastern parts of South Africa, as far north as Gauteng and the high-lying parts of Mpumalanga)
  • zuluensis (Zululand and the low-lying parts of Mpumalanga)
  • capricornis (Limpopo province through Zinbabwe to the Zambezi valley)
  • steinhardti (Namibia, southern Angola, northwestern South Africa, and presumably southern Angola)
  • neumanni (Kenya and Tanzania).

On the basis of photographic evidence, the main, obvious distinction is between the dark form of the southwesternmost part of South Africa, and the rest of the species-distribution. There seems no doubt that the nominate form, i.e. subspecies campestris, deserves recognition. This is true notwithstanding the uncertainty of its northern and eastern limits, where it intergrades with adjacent forms.

It also remains fair to assume that the disjunct form of East Africa, namely neumanni, is a valid subspecies. This is notwithstanding the surprising fact that, based on photos, it looks hardly different from the form in eastern South Africa.

This brings us to the arid-adapted, western form, the main distinctive feature of which, based on photos, is the extreme enlargement of the ear pinnae, at least in some individuals.

On one hand, it is possible that the western populations are merely an ecotype, as opposed to a subspecies. This is because the colouration is similar to eastern forms, apart from the usual tendency to pallor, that is so familiar in animals living in semi-deserts.

On the other hand, we should bear in mind that the western populations are globally unique, in ecological terms.

There is no ungulate, worldwide, of body mass less than 15 kg, that penetrates the edge of desert, other than R. campestris (https://www.inaturalist.org/observations/11048622).

A remarkable fact is that R. campestris lives both at the edge of the barren Namib, in Namibia, and under a mesic, equatorial climate in west-central Kenya.

So, it seems reasonable to recognise the subspecies steinhardti, the type location of which (https://en.wikipedia.org/wiki/Fransfontein) is in northwestern Namibia.

This leaves us with the mesic regions of southern Africa, from Eastern Cape through Free State and Kwazulu-Natal to Mozambique, Mpumalanga and Limpopo, and northwards through southeastern and western Zimbabwe to northeastern Botswana and western Zambia.

According to Roberts (1951), subspecies capricornis, of western Zimbabwe, is distinctive in nearly lacking the dark, V-shaped marking on the crown, that is normal in other subspecies (https://www.inaturalist.org/observations/136478014).

In support of a subspecific distinction, it is obvious from photos that the ears are, at least in some individuals, extremely small in Kruger National Park, in Limpopo province of South Africa.

However, it is easy to show that the dark, V-shaped marking on the crown does remain, at least faintly, even in

This seems to contradict Roberts (1951).

Given the doubtful distinctions among the various eastern populations, perhaps we should choose whichever name takes chronological priority.

The choices are

  • fulvorubescens 1822, 'Caffraria', Eastern Cape
  • natalensis 1907, 'Drakensberg, Natal'
  • zuluensis 1946, Umfolozi, now in northern Kwazulu-Natal
  • capricornis 1906, Klein Letaba, now in Kruger National Park.

It seems likely that the populations in Eastern Cape represent intergradation with the nominate subspecies.

The name 'natalensis' seems to be invalid, i.e. a nomen nudum (see comment below).

Therefore, it resolves to a choice among the remaining two, and capricornis takes priority.

On this basis, I suggest that we provisionally recognise Raphicerus campestris capricornis (Thomas & Schwann, 1906).

In summary, I propose that we adopt the following subspecific names in iNaturalist:

  • campestris (Western Cape and adjacent parts of Northern Cape and Eastern Cape)
  • capricornis (northern Kwazulu-Natal, Mpumalanga, Limpopo, Gauteng, Mozambique, Zimbabwe, Zambia, easternmost Botswana, northeasternmost parts of both Namibia and Botswana)
  • steinhardti (Namibia, Angola, most of Botswana, most of Northern Cape)
  • neumanni (Kenya, Tanzania).

In all other regions, just identify to species-level, owing to the likelihood of intergradation. This includes the whole of Free State and North West (https://en.wikipedia.org/wiki/North_West_(South_African_province)), the eastern half of Eastern Cape, and eastern Botswana adjacent to Zimbabwe.

I leave readers with a few, carefully selected photos, illustrating the range of variation in the appearance of R. campestris.

The following shows how dark the nominate subspecies can be (https://www.inaturalist.org/observations/11241069). There is considerable individual variation in R. campestris, but no individual of any other subspecies is anywhere near as dark as this.

The following show the extreme variation in the size of the ear pinnae within R. campestris (https://www.shutterstock.com/image-photo/sharpes-grysbok-standing-shade-1917779507 vs http://www.rupavadodaria.com/my-first-veganniversary/).

The following show the variation in the dark marking in the rostrum, adjacent to the rhinarium. This is absent in some individuals of R. c. steinhardti (https://www.inaturalist.org/observations/6611664), whereas it reaches to between the eyes in some individuals of R. c. capricornis (https://www.alamy.com/male-steenbok-raphicerus-campestris-kruger-national-park-south-africa-image68361886.html?imageid=7C5898AA-9B45-4447-98EF-CB92AEA025CF&p=196821&pn=1&searchId=81afea0714a7b7094a750bf43216e26d&searchtype=0).

The following show how different the conspicuousness of the ear pinnae can be in R. campestris, depending on direction of illumination and whether the pale hair-curtains are open or closed (https://www.shutterstock.com/image-photo/steenbok-spotted-namibian-desert-710362444 vs https://stock.adobe.com/search?k=steenbok&asset_id=83926648).

Posted on 28 September, 2022 17:33 by milewski milewski


Posted by tonyrebelo almost 2 years ago

@tonyrebelo @jakob

Smithers (1971, The Mammals of Botswana, page 219) states:
"The populations in the Kalahari, including the Makgadigadi and its environs, are lighter in colour and yellower on the body and forehead, lacking the reddish tinge of those from the eastern, north-eastern sectors and the Okavango which are darker and redder. These eastern populations may be intermediates between R. c. steinhardti and R. c. capricornis Thomas and Schwann, from the eastern Transvaal, and in the Okavango between R. c. steinhardti and R. c. kelleni Jentinck from Angola."

Posted by milewski almost 2 years ago

The following shows that infants of Raphicerus campestris campestris are as dark as adults:


Compare the above with Raphicerus campestris capricornis:


Posted by milewski almost 2 years ago

@tonyrebelo @michalsloviak @geichhorn

The following is noteworthy, because it seems to be the northernmost record for Raphicerus campestris (capricornis? kelleni?) in Zambia:

The location is ostensibly https://en.wikipedia.org/wiki/Liuwa_Plain_National_Park.

Actually, there are two photos of R. campestris, buried because of mislabelling, here:

Also, scroll in the following for another photo of Raphicerus campestris (capricornis? kelleni?) in Liuwa Plain National Park, western Zambia: https://www.wildlifeworldwide.com/group-tours/zambias-wild-west-kafue-liuwa-plain

Posted by milewski almost 2 years ago
Posted by milewski almost 2 years ago

@tonyrebelo @beartracker @paradoxornithidae @matthewinabinett @capracornelius @geichhorn

The following (https://depositphotos.com/523443750/stock-photo-steenbok-lying-red-sand-dune.html), of Raphicerus campestris steinhardti in Kgalagadi Transfrontier Park, shows something of interest, w.r.t. the adaptation to aridity in this subspecies.

This adult male individual shows what seems to be a dewlap, a thermoregulatory feature otherwise associated only with large-bodied sympatric ungulates such as Taurotragus oryx, Equus hartmannae, and possibly Strepsiceros strepsiceros (https://www.krugerparksa.com/Kruger-national-park-kudu-cow-head-profile.jpg) and Oryx gazella (https://www.istockphoto.com/photo/a-gemsbok-standing-in-close-up-at-the-kgalagadi-national-park-in-south-africa-gm1019399344-273960315).

It seems possible that the only bambi on Earth that possesses a dewlap is R. campestris. Indeed, I cannot offhand recall any ungulate of body mass less than 100 kg that possesses a dewlap (Ammelaphus imberbis needs checking).

However, this is complicated. Raphicerus campestris possesses an intermandibular gland in both sexes (https://journals.co.za/doi/pdf/10.10520/AJA00382809_3949). This gland is usually not noticeable.

The distension of the skin in this specimen, corresponding to the location of the intermandibular gland, suggests that the same stretch of skin may possibly serve two functions, viz. a) thermoregulation, and b) production of scent for social purposes.

Posted by milewski almost 2 years ago

What I dont understand is why there is the mismatch.
Admittedly (Wiki): "Up to 24 subspecies have been described from Southern Africa" is a bit of an overkill.

However, how is the current iNat compromise reached?

The old iNat approach was only 2 subspecies, and the old, 'split' approach in which 7 subspecies are recognised.
Now iNat recognizes 3 subspecies, but "Mammal Species of the World: A Taxonomic and Geographic Reference" has 4 (campestris, neumanni, capricornis & kelleni)
G&G split campestris subspecies into: campestris, fulvorubescens, natalensis (including zuluensis), capricornis, steinhardti (including kelleni)

iNat old | iNat curr | finest scale
campestris | campestris | campestris (sw S Afr)
campestris | campestris | fulvorubescens (E Cape & adj.)
campestris | campestris | natalensis (e S Afr -to Gau & grassveld Mp)
campestris | campestris | zuluensis (Zululand & lowveld Mp)
campestris | campestris | capricornis (Lim - Zim - Zambezi valley)
campestris | steinhardti | steinhardti (Namibia, s Angola, nw S Afr)
neumanni | neumanni | neumanni (Kenya and Tanzania).

iNats External Taxonomic Authority List for Mammals (Mammalia) is currently frameworked to the 2019 ASM Mammal Diversity Database, which is no longer available online. This authority applies to all mammal taxa between class and species. (Mammal subspecies are not covered by a framework.)
Presumably then, we can motivate for subspecies based on any other reference.

So how and who is going to campaign for iNaturalist to follow the finest scale (if we follow the above and split these up)?

In the interim these "comments" on subspecies are meaningless: we cannot search or filter on it easily.
One option is to use an observation field to allow extraction of these (it wont be a community ID!)

But what is the general feeling? Should we continue posting notes, or use an observation field, or motivate to add the other subspecies?
(esp in the light of https://www.zoochat.com/community/threads/ungulate-taxonomy-revisited-the-evidence-for-the-splits-of-g-g.467230/page-3 - how do we justify this?)

Posted by tonyrebelo almost 2 years ago

In my view, the minimum distinction, at the level of subspecies within Raphicerus campestris, would be a) the nominate ssp. (Western Cape, with some occurrence in adjacent Northern province and Eastern Cape) and b) all the rest.

Yes, this does mean that I would find it more acceptable to 'lump' East African neumanni than to 'lump' the southwestern Cape (nominate) campestris.

And yes, I realise that this seems to go against the fact that neumanni has reproductive isolation, whereas nominate campestris does not.

The nominate (southwesternmost) campestris is distinctive in its dark ground-colour, minimal countershading and superciliary, and (apparently) minimal expression of the radial gland. In retaining the full extent of white of the buttock flag, it emerges as the form of Raphicerus with the strongest 'flight announcement', reflecting its relationship with predators.

This tendency to darkness parallels, in biogeographic terms, what we see in Sylvicapra grimmia. As in the case of R. campestris, I would sooner relinquish all other sspp. of S. grimmia than relinquish the distinctively-coloured ssp. of the southwestern Cape - which, like nominate R. campestris, has retained a bright-hued forehead but has lost most of its countershading.

When one considers R. c. campestris (strict sense), S. grimmia grimmia, and R. melanotis (the latter also dark, relative to R. sharpei), we seem to have a significant biogeographical pattern, spanning two clades.

So, I would not necessarily argue for more sspp. than have hitherto been recognised in iNaturalist, useful though that might be. But I would argue that if we are going to concede only two sspp., one of them should be the nominate form, in the strict sense, not in the current loose sense in which the distinctiveness of R. c. campestris is lost by lumping it with all the other forms from the upper Karoo to western Zambia.

Posted by milewski almost 2 years ago

and what name would be applicable to your non campestris campestris?

Posted by tonyrebelo almost 2 years ago

The form kelleni (Jentinck, 1899) would take priority over capricornis (1906). The specimen from Angola (Cahama, Caculovar River), was collected remarkably early, despite its remote location (https://en.wikipedia.org/wiki/Cahama).

The form neumanni was also collected early (before 1906), so I will check the precise date.

However, it looks like being between kelleni and neumanni.

Posted by milewski almost 2 years ago

The answer, for neumanni, seems to be 1894 (https://www.tandfonline.com/doi/abs/10.1080/00222934508527497).

Also see https://www.mindat.org/taxon-4262648.html.

So, if we recognised only the nominate ssp. and one other ssp., the latter would seem to be R. c. neumanni (Matschie, 1894).

Posted by milewski almost 2 years ago

So you are proposing:

Milewski | iNat old | iNat curr | finest scale
campestris | campestris | campestris | campestris (sw S Afr)
neumanni | campestris | campestris | fulvorubescens (E Cape & adj.)
neumanni | campestris | campestris | natalensis (e S Afr -to Gau & grassveld Mp)
neumanni | campestris | campestris | zuluensis (Zululand & lowveld Mp)
neumanni | campestris | campestris | capricornis (Lim - Zim - Zambezi valley)
neumanni | campestris | steinhardti | steinhardti (Namibia, s Angola, nw S Afr)
neumanni | neumanni | neumanni | neumanni (Kenya and Tanzania).

Posted by tonyrebelo almost 2 years ago

Thanks for laying it out in a technically correct way.

However, it would go too far to say that I am proposing this.

Rather, my position is that I propose, as a working system, four sspp. After all, we already have something similar in Sylvicapra grimmia (arid-adapted, western steinhardti vs mesic forms to the east and northeast), and nobody should quibble with a wide disjunction implying subspeciation in East Africa.

But, if I have to settle for just two sspp., then yes, I would prefer what you have laid out here to the alternatives.

Posted by milewski almost 2 years ago

So you are proposing as a working system:

Milewski prop | iNat old | iNat curr | finest scale
campestris | campestris | campestris | campestris (sw S Afr: Fynbos)
capricornis | campestris | campestris | fulvorubescens (Thicket & coast Grassvld - E Cape)
capricornis | campestris | campestris | natalensis (Grassveld)
capricornis | campestris | campestris | zuluensis (Lowveld s of Limpopo)
capricornis | campestris | campestris | capricornis (Tropical Lowveld)
steinhardti | campestris | steinhardti | steinhardti (Namib & Kalahari & Nama Karoo)
neumanni | neumanni | neumanni | neumanni (Kenya and Tanzania).

Posted by tonyrebelo almost 2 years ago

Yes, that is correct, except that I hesitate to recognise fulvorubescens as more than intergradational. Therefore, any specimens in Eastern Cape that do not fit the nominate would be as aptly classified under the nominate as under capricornis, not so?

Posted by milewski almost 2 years ago

I would argue that eastern Cape it should be done as either campestris or capricornis depending on visible features. But what exactly is the cutoff - intermediates would be unclassifiable.

Posted by tonyrebelo almost 2 years ago


There is an intriguing hint that the nominate ssp. may perhaps differ also in sexual dimorphism.

Haltenorth and Diller (1977), on page 52, state as follows for R. campestris:
"old males usually darkest".

They also, on page 53, state as follows for R. melanotis (including sharpei):
"old and young males darker than old and young females".

Now, I have examined many photos of male and female together, in the case of R. c. capricornis and R. c. steinhardti, without noticing any sexual difference in colouration.

So, I assume that, if there is a sexual difference in colouration, it is restricted to the nominate, which is sympatric, and perhaps in this respect convergent, with R. melanotis.

If this subtle sexual dimorphism holds up to scrutiny in R. c. campestris sensu stricto, then this would be yet another aspect setting it apart from the rest of its species, not so?

One way to test this might be to find the five darkest specimens of the nominate form on the Web, and check if they are all males.

Posted by milewski almost 2 years ago


In reply to your question re fulvorubescens:

A noteworthy aspect of the distribution of R. campestris is that it reaches its limit in the central Eastern Cape, being absent from not only the Natal Midlands but also the Eastern part of the Eastern Cape (where it is, as it were, replaced by Ourebia ourebi).

Please note that this is different from Sylvicapra grimmia, which does continue eastwards ('ssp. burchellii').

Whereas it seems natural that intergradational forms occur in S. grimmia in Eastern Cape, their occurrence in R. campestris would be puzzling.

This is because the question arises, 'intergrading with what'?

For intergradation to occur in Eastern Cape, there must have been a deep southerly penetration of either ssp. capricornis or ssp. steinhardti, south of Free State. This seems rather far-fetched, for either of these sspp.

So, what seems more likely is that, to all intents and purposes, 'fulvorubescens' is just a summer-rainfall ecotype of the nominate, and that we can classify all specimens in Eastern Cape as simply the nominate.

Does this make sense?

If so, I would lump fulvorubescens with campestris, not with capricornis.

Posted by milewski almost 2 years ago

@koosretief @henrydelange @gigilaidler

Would you say, based on your experience with the steenbok in the Western Cape, that mature males are darker-coloured than adult females?

Posted by milewski almost 2 years ago

Milewski prop | iNat old | iNat curr | finest scale

campestris | campestris | campestris | campestris (sw S Afr: Fynbos)
campestris | campestris | campestris | fulvorubescens (Thicket - E Cape)

capricornis | campestris | campestris | natalensis (Grassveld- highveld)
capricornis | campestris | campestris | zuluensis (Lowveld s of Limpopo)
capricornis | campestris | campestris | capricornis (Tropical Lowveld)

steinhardti | campestris | steinhardti | steinhardti (Namib & Kalahari & Nama Karoo)

neumanni | neumanni | neumanni | neumanni (Kenya and Tanzania).

Posted by tonyrebelo almost 2 years ago


Yes, well done, I agree.

Posted by milewski almost 2 years ago

There is enough to quibble with in the following interpretation of the subspecies of Raphicerus campestris:


The problems seem to centre on the Highveld (https://en.wikipedia.org/wiki/Highveld). In and near this area, which spans three countries and seven South African provinces (https://www.intergate-immigration.com/blog/south-african-provinces/), it is currently impossible to identify any specimen of R. campestris to subspecies level.

Posted by milewski almost 2 years ago

That map just seems far too political to be true. Whomsoever drew it knows little about ecosystems, biomes or biogeographical boundaries in the region.

Posted by tonyrebelo almost 2 years ago


Another possible distinction between nominate campestris and other sspp. may be in the length of the hooves.

I refer to Roberts (1951).

campestris "As a subspecies it may be distinguished by its long hoofs and rather long ears, besides its dark colour"

fulvorubescens "hoofs shorter than in the typical western subspecies'

natalensis (information confused beyond redemption)

zuluensis "The hoofs and ears are short"

capricornis -

steinhardti "the hoofs are short"

The data are presented in Table 64, on page 590.

The numbers of individual specimens sampled are minimal, but the hooves were measured in various ways, both fore and hind. There is no statistical analysis.

I have examined the figures, and my conclusion is that there is no substance in this.

Roberts based his claim, in the nominate ssp., on just one adult specimen of each sex, from Lamberts Bay.

By the way, a careful reading of Roberts (1951) on ssp. natalensis leads me to discard this entity. The name and the information associated with it seem worthless, for several reasons.

Bottom line: everyone working on the taxonomy of R. campestris so far seems to have been incompetent. (And Roberts himself wrote that he suspected kelleni to be a misindentified oribi.)

As a result, careful examination of photos is as good an approach as any, and where the 'literature' disagrees, we should respect what we can actually see.

Posted by milewski almost 2 years ago


The subspecies name 'natalensis' is often cited for the steenbok, with an implication that the species was originally recorded in or adjacent to the Highveld (explicitly 'Drakensberg and highveld of Southern Transvaal', bottom of page 341, Roberts, 1951).

However, having read the details in Roberts (1951), I question this name.

The location given by Rothschild (1907) for the specimen of natalensis was 'Drakensberg, Natal'.

The description indicates extreme darkness, exceeding even that typical of the nominate form of the southwestern Cape. This again sounds most unlikely for the location claimed.

This specimen was probably collected near Cape Town, and then irretrievably mislabelled and thus mislocated.

Roberts (1951, page 341) himself comments "this is an inadequate and misleading description...the matter must remain in abeyance until more material is compared, as the distribution seems extraordinary".

Posted by milewski almost 2 years ago

But if we go by features, the we can test this by looking for dark Steenbok in the Drakensberg from Lesotho to Ermelo.
We should find none.

Posted by tonyrebelo almost 2 years ago

Add a Comment

Sign In or Sign Up to add comments