Journal archives for October 2022

October 01, 2022

A new hypothesis for the steenbok (Raphicerus campestris) in the Highveld of South Africa: it was naturally absent

It is easy to assume that the steenbok (Raphicerus campestris) was indigenous to the Highveld ( when European explorers arrived.

However, I suggest that it was actually absent.

This would help to explain why its current taxonomic status in the Highveld is so nebulous (

My argument is based on an ecological rationale, but seems consistent with the historical record.

The steenbok tends to be taken for granted as widespread and ecologically tolerant. However, it is more ecologically specialised than first it seems, in diet and habitats.

This species is part of a guild of relatively small herbivores including

All eat both grasses and dicotyledonous plants, in various combinations according to the seasons (

Because it has so many competitors, R. campestris may have been somewhat limited in occurrence in prehistoric South Africa, when the full fauna of the Holocene remained.

The Highveld contained by far the most complex fauna of ungulates of any region of treeless grassland on Earth ( and This may have precluded a niche for R. campestris under natural conditions.

I suspect that it was only when the fauna was disrupted by human settlement that R. campestris spontaneously entered the Highveld, from the west and north.

The idea is that it filled in for the species virtually exterminated by settlers, particularly the formerly abundant and migratory A. marsupialis.

This is not to claim that R. campestris became abundant in the Highveld, but merely that it became viable there because of, rather than despite, anthropogenic disturbance.

Three subspecies were thus hypothetically recruited to the Highveld, then mixing there through hybridisation. These are R. c. campestris (southwesterly origin), R. c. steinhardti (northwesterly origin), and R. c. capricornis (northeasterly origin).

Another basis for my hypothesis is habitat, particularly the natural availability of cover.

In its original state, the Highveld was treeless over extensive areas, partly owing to the intense natural herbivory. This hypothetically made the grassland too open for R. campestris.

There are two situations, in the Highveld today, with some shrubby indigenous cover, viz.

  • perennial drainage lines, and
  • scattered rocky outcrops.


The settlement of the Highveld has boosted the incidence of woody plants, for various reasons.

I consulted du Plessis S F (1969, The Past and Present Geographical Distribution of the Perissodactyla and Artiodactyla in Southern Africa), for information on the occurrence of the steenbok.

On page 100, du Plessis states for 'Orange Free State':
"Though no doubt occurring everywhere in this province in the past, practically no written records could be traced...Only Smith (Kirby, 1939) in 1835 mentions it as being common near the confluence of the Riet and Modder rivers".

This location ( is actually west of Free State province and the Highveld, being located in the current Mokala National Park (

Much of the Highveld occurs in the former Transvaal ( However, once again du Plessis (1969) largely draws a blank in the historical record.

The exceptions are as follows:

  • "Mauch (Petermann, 1870): north of Lydenburg"
  • "Holub (1881): near the Vaal River in western Transvaal"
  • "Holub (1890): between Bloemhof and Christiana"
  • "Baldwin (1894): the vicinity of Potchefstroom"
  • "Randall (1895): the open flats in the Barberton District."

The relevant locations mentioned are:

All of these are rather marginal to the Highveld, except for Potchefstroom. However, the latter is located in the valley of a major drainage line (

On page 103, du Plessis states the incidence, as of 1969, as follows:
"Orange Free State: Van Ee (1962): fairly generally distributed throughout the province with the greatest numbers along the rivers and in the mountainous districts of the east, especially in the Fouriesburg, Tweeling, Petrus Steyn, Boshoff, Theunissen, Brandfort, Hoopstad and Koffiefontein districts. Roberts (1963): 45 in the Willem Pretorius Game Reserve".

The status in Transvaal (page 104) was similarly widespread, as of 1969.

So, dear readers, please prove me wrong in my suggestion that the steenbok was naturally absent from the Highveld, including virtually the whole of what is now Free State.

One way to do so might be to consult a work to which I currently lack access. This is by Skead (, covering that part of the Highveld falling within the Eastern Cape (,

In the meantime, the following is a compendium of all the current photos of R. campestris, located in the Highveld, in iNaturalist:

Only four of the above observations show the animals clearly.

Also see

Posted on October 01, 2022 06:54 AM by milewski milewski | 6 comments | Leave a comment

October 02, 2022

Selected views of the bambi genus Madoqua, plus noteworthy photos mislabelled as dik-diks on the Web


showing anomalous whitish at knee







kirkii and



Posted on October 02, 2022 11:16 PM by milewski milewski | 26 comments | Leave a comment

October 03, 2022

An attempted photo-guide to the subspecies of the steenbok (Raphicerus campestris) is partly a self-refuting exercise

@tonyrebelo @jeremygilmore @ludwig_muller @capracornelius @simontonge @tandala @geichhorn @michalsloviak @colin25 @alanhorstmann @jakob @jwidness @dejong @davidbygott @koosretief @gigilaidler

Today, I set out to illustrate - in the spirit of a field guide-book - the various subspecies of the steenbuck (Raphicerus campestris,

I was only partly successful, and this was not because of a lack of photos on the Web

Something important to realise, which has not emerged in the literature, is the following.

The northern subspecies ( is only slightly different from the subspecies found in eastern South Africa (

This is despite a wide geographical disjunction and the associated reproductive isolation.

It is instead the southernmost subspecies, living in a temperate zone, that is recognisable in photos.

What this means is that only two subspecies can be claimed with confidence, viz.

  • nominate campestris, and
  • neumanni, in a broadened sense.

The remaining subspecies (capricornis and steinhardti) are recognised mainly because of an assumption that their own wide geographical and climatic spread, from the edge of the Namib desert to the edge of miombo woodland in Zimbabwe, must surely be reflected in subspeciation.

It remains possible that even the adaptation to aridity within this species is ( is merely a matter of ecotypes, rather than a matter of subspeciation.

Another important point is that there is enough individual variation in the steenbok to blur any subspecies-distinctions.

The result:

The distinctions among neumanni, capricornis, and steinhardti are so slight that I have found it hard to compile any 'typical' photos of them, in which the subspecific differences are self-evident.

The following is my best attempt, after sifting through thousands of photos.


Western Cape and adjacent parts of Northern Cape and Eastern Cape, South Africa

ground-colour dark

forehead rich-hued

white features (except for buttocks) minimal, particularly on face and inner upper hindleg


southern Angola, most of Botswana, Namibia except for eastern Caprivi, part of Northern Cape of South Africa

ground colour pale

size of ear pinnae maximal

radial gland noticeable


Mpumalanga, Limpopo, northeastern Kwazulu-Natal, eSwatini, southern Mozambique, Zimbabwe, eastern Caprivi of Namibia, eastern and northeastern Botswana, southwestern Zambia

forehead rich-hued

size of ear pinnae minimal

radial gland noticeable


Kenya, Tanzania

Despite the geographical isolation of the East African part of the species-distribution, I have failed to find any consistent difference between this supposed subspecies and capricornis.

Posted on October 03, 2022 04:22 PM by milewski milewski | 7 comments | Leave a comment

October 05, 2022

White on the buttocks: a previously overlooked species-difference in dikdiks of the Madoqua kirkii-Madoqua damarensis complex?

@dejong @tbutynski

In recent Posts, I have shown previously overlooked variation among the species/subspecies in two genera of bambis, namely Ourebia ( and Raphicerus (

I now turn to another genus of bambis, namely Madoqua (

The species and subspecies of Madoqua may seem too taxonomically complicated for many naturalists to disentangle.

De Jong and Butynski (2017, have greatly clarified the distinctions among various forms in the kirkii-complex and superficially similar Madoqua guentheri.

However, they seem to have overlooked what is perhaps major variation in adaptive colouration among the species and subspecies in this geographically bewildering set of taxa.

I refer to the incidence, extent, and flaring of whitish pelage on the buttocks - which in some taxa is displayed to potential predators.

What I have noticed, after perusing hundreds of photos, is that there is a range of apparency of white on the hindquarters, as follows:

  • maximal in M. guentheri, minimal in M. damarensis (restricted to southern Africa), and intermediate in the kirkii-complex of East Africa, and
  • within the kirkii-complex: maximal in nominate kirkii and hindei, vs minimal in thomasi and cavendishi.

In M. guentheri, there is a well-developed, white buttock flag ( This is normally folded out of sight, but is activated in alarm.

In M. damarensis, there no buttock flag. Furthermore, whitish pelage is absent from the buttocks, being restricted to the inner surface of the uppermost hindleg, where it is obscured from view (, even when the animal flees.

In thomasi and cavendishi, the white pelage on the inner upper hindleg hardly reaches the inner surface of the buttock (cavendishi: ; cavendishi/thomasi:

Instead, there is a darkened effect, owing to the grey, grizzled pelage on the inner buttock being viewed from a certain angle (also see and

In nominate kirkii and (to a lesser extent?) hindei, considerable whitish is normally apparent on the buttocks. It remains unknown whether this can be activated by piloerection.

What emerges is that, w.r.t. the colouration of the buttocks, the southern and western forms of the kirkii-complex in East Africa may be more similar to the widely disjunct damarensis of southern Africa than they are to nearby kirkii, the nominate, northeasternmost form.

This raises the possibility that there are two species, one of them extremely disjunct - in a biogeographic parallel to sympatric Raphicerus campestris, which also shows wide disjunction between southern and East Africa.

These would be

If so, the basic geographical feature separating the two spp. would be the eastern arm of the Great African Rift (, from Lake Turkana ( in the north to Lake Manyara ( in the south.

Such geographical separation would be analogous with that previously recognised between

  • Connochaetes albojubatus and Connochaetes mearnsi, and
  • Kobus ellipsiprymnus and Kobus defassa.

The following show the buttock flag in Madoqua guentheri:

The following shows that there is no analogous flag in cavendishi/thomasi:


Many photos of M. guentheri show a whitish horizontal mark on the hindquarters (just below the small tail), which I have never seen in any other species in the genus:

This indicates a fur-fold which opens out into a broad, white, buttock flag.

This resembles the buttocks in Raphicerus campestris ( However, it is more pronounced, with a more complex, origami-like unfolding - both vertically and horizontally - of panels of pelage.

KIRKII (nominate)

The following show that there is more extensive white on the hindquarters than in damarensis.

This white pelage is visible without any particular activation, partly because the panel of grizzled, grey pelage of the outer buttock is shorter than in damarensis, thomasi, and cavendishi.

The adaptive colouration thus seems to confirm that damarensis is a different species from nominate kirkii, and not merely a subspecies.

second photo in

The following shows that the whitish on the buttocks can be seen at some distance:

The following is the only photo of nominate kirkii, in posteriolateral view, that could be confused with damarensis on the basis of the colouration of the hindquarters:


The following show that considerably more white is visible on the hindquarters than in damarensis. This seems, at least partly, because the pelage of the outer buttock is shorter than that of damarensis.

Whether this qualifies as a buttock flag remains unclear, because this would depend on

The following, at the lower Tana River in eastern Kenya, do not show any white on the buttocks:


The following show that, if white is more evident on the hindquarters of cavendishi than on those of damarensis, the difference is slight.

The following individual shows the maximum extent of whitish on the buttocks:


The following does show some white on the buttocks:

INTERGRADATIONAL between cavendishi and thomasi, in Tarangire and Lake Manyara National Parks, Tanzania:

The following, in Arusha National Park, look like thomasi rather than hindei:


The following show that the occurrence of whitish on the hindquarters is limited to the inner surface of the upper hindleg, with hardly any sign of any extension on to the buttocks.

Second photo in

tail raised, revealing the black bare skin around anus:

The following does show some white:

Posted on October 05, 2022 01:53 AM by milewski milewski | 9 comments | Leave a comment

Summary of my recent observations: diagnostically pale features on the hindquarters of bovid bambis (Ourebia, Raphicerus, and Madoqua)

@joni_overbosch @beartracker @davidbygott @tonyrebelo @ludwig_muller @jeremygilmore @tandala @dejong @michalsloviak @oviscanadensis_connerties @capracornelius @tbutynski @zarek @simontonge @jwidness @johnnybirder @calebcam @jakob @henrydelange @wasinitourguide @geichhorn @paradoxornithidae @jacqueline_llerena @koosretief @justinhawthorne @happyasacupcake @marcelo_aranda @enricotosto96 @diegoalmendras @michaelweymann @fionahellmann


The following aspects of adaptive colouration have previously been overlooked. They may additionally be diagnostic, taxonomically.


Ourebia ourebi differs from Ourebia montana in possessing a bleeze on the hindquarters.


Raphicerus campestris campestris exceeds all other subspecies in the degree of development of a buttock flag.


Madoqua damarensis differs from Madoqua kirkii in lacking white on the buttocks.

Madoqua kirkii/hindei differs from Madoqua thomasi/cavendishi in possessing white on the buttocks.


I have selected the following to illustrate the diagnostic features mentioned above.


Presence of bleeze on hindquarters in Ourebia ourebi:

Absence of bleeze on hindquarters in Ourebia montana:


Well-developed buttock flag in Raphicerus campestris campestris: and and

Relatively poorly-developed buttock flag in other sspp. of Raphicerus campestris: and and

Fibular flag in:

Raphicerus melanotis

Raphicerus sharpei


Absence of white on buttocks in:



Presence of white on buttocks in:

kirkii/hindei and and and

Posted on October 05, 2022 05:06 PM by milewski milewski | 33 comments | Leave a comment

October 06, 2022

Compendium of photos of Raphicerus sharpei on the Web, part 1

Raphicerus sharpei is among the most obscure species of antelopes in the national parks of southern Africa, being scarce, small, cryptic in colouration, timid, non-gregarious, and nocturnal.

Therefore, nobody would be surprised to find that few photos of this species have been posted on the Web.

The reality is surprising: there are so many photos available that it is hard to Post them all.

The following is a compendium of all the photos of R. sharpei that I can currently find, excluding those on iNaturalist.

Collectively, they provide excellent documentation of the appearance of this species.

Posted on October 06, 2022 12:54 AM by milewski milewski | 0 comments | Leave a comment

Compendium of photos of Raphicerus sharpei on the Web, part 2



Scroll in

Scroll in|2

Scroll in

Posted on October 06, 2022 01:25 AM by milewski milewski | 1 comment | Leave a comment

October 07, 2022

Introducing the ocular flag, a feature of adaptive colouration epitomised by dik-diks in the Madoqua kirkii/damarensis complex

@tonyrebelo @ludwig_muller @jeremygilmore @tandala @capracornelius @oviscanadensis_connerties @geichhorn @paradoxornithidae @koosretief @botswanabugs @jacqueline_llerena @zarek @dejong @tbutynski @davidbygott @diegoalmendras @michaelweymann @grinnin @beartracker

It is unsurprising that dik-diks, in the Madoqua kirkii/damarensis complex (, have inconspicuous colouration, overall.

However, what is puzzling is that the eyes do not conform to this pattern.

The dark of the eyeball, eyelids, and antorbital gland are accentuated by a ring of pale pelage, making the eyes conspicuous.


How can this anomaly be explained?

I hypothesise that the conspicuous pattern of dark/pale contrast at the eye qualifies as an OCULAR FLAG.

This would function as:

  • an announcement to potential predators that they have been spotted, and
  • a means of intraspecific communication, allowing mates and offspring to maintain visual contact in dim or dappled light.

This flag is hypothetically activated by

  • the utterance of an alarm-whistle, and
  • motion of the head, including while walking routinely during foraging.


The Madoqua kirkii/damarensis complex is comparable with Raphicerus sharpei, but the two seem to have functionally opposite patterns of colouration at the eyes.

The distribution of Raphicerus sharpei ( fills in for the disjunct distribution between M. damarensis damarensis (Namibia and Angola) and the other members of the complex (East Africa,

The similarities include:

  • habitat consisting of 'leafless thickets', tending to be associated with stony ground at the base of rocky outcrops,
  • average body mass falling within the range 5-7.5 kg,
  • colouration that is cryptic overall, partly by virtue of a grizzled effect, and
  • hindquarters that lack any buttock flag or conspicuous tail.

However, a difference is that the eye is accentuated in Madoqua, whereas it is masked in R. sharpei.

The following show the darkness around the eyes in R. sharpei:

Please note that both Kingdon ( and Dandelot ( have misportrayed this aspect of R. sharpei. They fail to show the typical masking of the eyes in this species.

My explanation of this difference between the Madoqua kirkii/damarensis complex and R. sharpei:

There is greater sociality in the former than in the latter.

In both cases, there is a monogamous pair-bond. However, R. sharpei is one of the most solitary of all antelopes in savanna biomes, whereas dik-diks retain obvious social behavior.

Of the 120 observations of R. sharpei currently posted in iNaturalist, none shows more than one individual. By contrast, in the case of Madoqua damarensis, the corresponding value is 7% (13 of the 187 observations).


Dik-diks of the Madoqua kirkii/damarensis complex react to the detection of stalking predators in ways consistent with self-advertisement by means of an ocular flag.

Estes (1991) states: "Dik-diks freeze at the slightest disturbance, often with 1 leg raised. The male moves his head cautiously, trying to identify the danger, while the female remains immobile except for the questing tip of her nose...Once a predator is detected and identified, dik-diks behave in 1 of 2 ways, depending on whether it is a stalker or a courser. If it is a cat, they flee just far enough to be safe, emitting loud, explosive whistles at the first (stotting) bounds. They then proceed to keep watch and sound the alarm - possibly in a duet...The male is the principal caller...When the predator is a hyena, wild dog, or other courser, dik-diks immediately sink to the ground and lie still until the enemy passes. If discovered, they flee without whistling".

This contrasts with R. sharpei.

Skinner and Chimimba (2005) state: "they are difficult to observe and are inclined to lie up very tightly and unless disturbed by close approach remain hidden in the undergrowth. When they do run off they do so crouching low to the ground as they run through the thick underbrush, making them difficult to see. Quite often the disturbance of the undergrowth and a glimpse of a reddish body is all that can be seen of them".

The above differences in anti-predator behaviour between R. sharpei and Madoqua kirkii/damarensis are categorical. This is because there are no circumstances in which the former visually advertises itself to potential predators.

Even in the case of audial advertisement, R. sharpei lacks any snort or whistle of alarm, and its only announcement is the sound of stamping of the feet as it initially flees.

What this means:

Although similar in other ways, R. sharpei and the Madoqua kirkii/damarensis complex are extremely different in their reactions to stalking predators. The former simply hides and then, at the last moment, flees. By contrast, the latter are remarkably demonstrative for such small animals, in

Indeed, I suspect that, of all the many genera and species of bambis on Earth, dik-diks are the most demonstrative towards potential predators in the interval between initial alarm and running.

With respect to social behaviour:

For Madoqua damarensis, Skinner and Chimimba (2005) state: "The pair may be accompanied by up to two offspring... approximately 10% of groups are polygynous, containing two unrelated females...may form temporary aggregations of up to 10 individuals at sites of food of the few antelopes in which female territoriality has been documented".


For Raphicerus sharpei, the same authors state: "One usually encounters solitary adults, pairs, or a female with her single offspring...They are shy and secretive and can be overlooked in areas where in reality they are reasonably common".

What emerges is a clear role for an ocular flag in dik-diks, in a syndrome including both anti-predator and social behaviours.


If we accept that the pattern of colouration at the eye is the only adaptively conspicuous feature in certain species of dik-dik, then this creates a scientific search-image for ocular flags in other ungulates.

It is well-known that various clades of bovids and cervids contain spp. with pale superciliary ( or supraorbital ( markings.

However, the adaptive function of such markings depends on

  • the relative size of the eye, which depends mainly on body size,
  • the extension of the pale marking to encircle the whole eye (plus antorbital gland, if present), and
  • the presence of other pale markings on the figure, which would distract from the eye.

Body size is important. This is because the larger the body, the proportionately smaller the eye tends to be.

For example, Litocranius walleri resembles dik-diks in the pattern of colouration at the eye ( and

However, this is too small, relative to the whole figure, to qualify as a flag ( Instead, in this case the pattern is more likely to function disruptively, i.e. in aid of camouflage.

In summary, the argument for an ocular flag in dik-diks is based on the following:

  • body size is extremely small (5 kg) for an ungulate, making the eye proportionately large,
  • the eye is relatively large even among the various ungulates of similar, small body size,
  • the rest of the colouration is plain, with even the erectile crest, on the crown of the head, lacking dark/pale contrast,
  • the tail is too small to function as a flag, even if it were dark/pale,
  • the reaction to non-cursorial predators - presumably even by night - is one of self-advertisement, and
  • there is more sociality than expected for a small-bodied, mainly monogamous ruminant.

So, in the spirit of scientific hypothesis-testing, I can make the following prediction.

Among all the ungulates of the world, the only species qualifying for an ocular flag belong to the genus Madoqua.

Which leads to a challenge to readers:

Can anyone find any ungulate, other than dik-diks, that possesses an ocular flag?

Posted on October 07, 2022 02:43 AM by milewski milewski | 8 comments | Leave a comment

October 08, 2022

Pampas deer (Ozotoceros bezoarticus), a cervid counterpart to a combination of reedbuck (Redunca) and oribi (Ourebia)

Ozotoceros bezoarticus ( and





Posted on October 08, 2022 09:33 PM by milewski milewski | 33 comments | Leave a comment

October 09, 2022

Structure and function of the tail show that the red hartebeest (Alcelaphus caama) is a different species from Coke's hartebeest (Alcelaphus cokii)

Many naturalists may assume that Coke's hartebeest (Alcelaphus cokii, and the red hartebeest (Alcelaphus caama, are mere subspecies of a single, widespread species (

And that Groves and Grubb (2011,, in elevating them to the status of full species, were overenthusiastic as taxonomic 'splitters'.

The 'lumping' view taken by iNaturalist may seem reasonable, given that many bovids vary subspecifically in details of colouration, and the shapes of their horns.

However, the distinction between cokii and caama runs deeper than it may at first seem. This is because

  • their tails differ in structure and function, and
  • the 'flight announcements' of ruminants - being basic to their anti-predator adaptations - tend to be more evolutionarily conservative than other aspects of their appearance and behaviour.

In order to understand the relevance of the tail, we first have to appreciate the part the tail plays in the broader context of adaptive colouration.

Please see


Both Alcelaphus cokii and Alcelaphus caama are candidates for the possession of bleezes on their hindquarters. These consist of large, pale, sheeny patches on the buttocks and haunches, contrasting with the dark of the tail-tassel.

By definition, bleezes tend to highlight the figures even when stationary, making them conspicuous to scanning predators even at some distance ( and

However, on closer scrutiny, the patterns are too different for cokii and caama to be regarded as mere subspecies.

This is because

These differences are so great that caama unambivalently possesses a bleeze, but the same cannot be said for cokii.

Of thousands of photos of cokii - with various illuminations - on the Web, the following makes the strongest possible case for conspicuous colouration in this species:

The following views of cokii also indicate a bleeze ( and and

However, this pattern results from not only depigmentation of the pelage, but also poorly-understood sheen/antisheen effects - which depend on the angle of the light. Furthermore, it tends to achieve conspicuousness only in posterior/posteriolateral view (

Another basic difference is that, in cokii, the pale extends to the legs, dissipating any pale/dark contrast within the figure as a whole ( and and

By contrast, in caama,

The following ( and and show that, in full profile, cokii can look so plain that it seems to exemplify cryptic colouration. This can hardly be said for adults of caama, in any illumination.


This dichotomy between cokii and caama is compounded by an obvious difference in caudal flags ( and This may apply particularly during stotting.

In cokii, the tail is not held raised during running. Instead, it tends to be tucked in.

Alcelaphus cokii, RUNNING:
Second photo in

By contrast, in caama the tail is raised to about a horizontal position. This makes for a conspicuous display, given that the dark tail-tassel is so large.

Alcelaphus caama, RUNNING:
Scroll to two photos in
Scroll in

My interpretation is that caama, but not cokii, qualifies as possessing a caudal flag, in an anti-predator context.


Both cokii and caama stot.

However, there is a previously overlooked, categorical difference in the function of the tail during stotting.

In cokii, the tail is not displayed, instead being hidden close to the buttocks. By contrast, in caama the tail is raised demonstratively to or above the horizontal position. This means that a caudal flag is activated, during stotting, only in caama.

Alcelaphus cokii, STOTTING:

Alcelaphus caama, STOTTING:


The tail differs so much, between caama and cokii, that it would be far-fetched to include them in a single species. This seems to have been overlooked by all previous authors.

(Groves and Grubb (2011) made no attempt to analyse adaptive colouration. Therefore, their reasons, for recognising cokii and caama as different species, differ from mine.

As I have shown in this Post, conspicuous colouration - at a scale relevant to anti-predator adaptations - is

Furthermore, this applies in terms of both




Posted on October 09, 2022 10:52 PM by milewski milewski | 39 comments | Leave a comment