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

ALCEINI

Moose / Elk

The Moose / Elk (Alces alces) is the largest species of deer and there is a long history of arguing whether there are one or two species: Alces alces The European elk from W of the Yelisei River and Alces americana the Moose from E of the Yenisei river and North America. Among most experts the consensus is that Moose should be treated as one species, but non-surprisingly G&G argue for the recognition of two species. Unfortunately I cannot access the study (Boeskorov 1999) which supplies the data for the split, so I will simply follow the description given by G&G


Skins

alces is described as having an evenly brownish overall color, with legs whitish. americana is said to have an overall color from light brown to black, with legs from light to dark brown or grey.

Skulls & Antlers

In the skull of alces the posterior end of the nasal branch of the premaxilla is spoonlike, whereas in americanca it is narrow and pointed, though in both populations the other type also exists in rare cases. Antlers of alces mostly have a single palm, whereas most americana have a double palm. Subspecies cameloides from Mongolia and N China has almost unpalmated antlers. Antler shaft averages 91.3 mm in alces and 120 mm in americana.

Additional data

G&G state that the body size index (body length as percentage of shoulder height) of alces is 80.8-84.5%, whereas this is 77.8-89.6% in americana. There also exists a difference in karyotype between alces (2n = 68) and americana (2n = 70). Boeskorov (2003) also mentions that different rutting calls and fauna of helminth worms have been described between alces and americana

Although there seem to be clear differences in karyotype, as well as differences in size, skull and antlers, the overall picture is not as clear. Genetic research (e.g. Hundertmark et al. 2002; Hundertmark & Bowyer 2004) has shown that until very recently, ~60.000 years, there was no difference between americana and alces. So all differences have originated very recently. Moose also only seem to have colonised America in the past 15.000 years. Lister et al. (2004) compared the shape of the premaxilla (part of skull) across the range of Alces alces and found that though the European type dominated west of the Yenisei River and the American type E of the Yelnsei River, there was a very broad zone, where both types commonly occured and only in the western and eastern extremes of the range did all individuals have the same premaxilla type. This indicates a broad hybrid zone.

Summarizing

Moose / Elk present an interesting case where we seem to have speciation in progress and this is a case where in an evolutionary species concept one would recognize two separate species, whereas under a biological species concept both taxa belong to the same species. I am not sure what to make of this and we cannot look into the far future to see how both taxa would develop. Unfortunately not much seems to be known of animals in the hybrid zone and it would be interesting to see whether animals with hybrid characteristics have lower reproductive success than others. Until more evidence becomes available it seems better to me to keep americana as a subspecies of Alces alces.


In addition to the question whether alces and americana constitute different species, there is also debate about the validity of the multiple subspecies described in N America. Bowyer et al. (2002) provide some evidence that gigas from Alaska is not only bigger because of a different diet, but that there is some genetic basis for this as well.

The following 8 subspecies are normally recognized, of which all but alces would belong to the putative species americana:

A.a. alces Europe, Russia W of Yenisei River
A.a. pfizenmayeri Siberia, E of Yenisei River
A.a. burturlini extreme E Russia (sometimes included in gigas)
A.a. cameloides E Mongolia, extreme N China, SE Russia
A.a. gigas Alaska
A.a. shirasi NW USA,
A.a. andersoni, most of W and C Canada and parts of N USA
A.a. americana E Canada and NE USA

A.a. alces

@alexkant , Mogilev Zoo, Belarus


@gentle lemur , Highland Wildlife Park, Kingussie, UK

A.a. gigas

@zoo_enthusiast , Alaska Wildlife Conservation Center, USA


@Pleistohorse , Alaska, USA

A.a. andersoni

@geomorph , Minnesota Zoo, Minneapolis, USA

A.a. shirasi

@savethelephant , Shiras Moose Rocky Mountain National Park, USA

A.a. americana

@Pleistohorse , Maine, USA

References

Bowyer et al. (2002): https://www.researchgate.net/profil...d_genetics/links/0deec522f8757ad798000000.pdf

Boeskorov (2003): http://natuurtijdschriften.nl/download?type=document&docid=523503

Hundertmark et al. (2002): Mitochondrial Phylogeography of Moose (Alces alces): Late Pleistocene Divergence and Population Expansion - ScienceDirect

Hundertmark & Bowyer (2004): https://www.researchgate.net/profil...ics-evolution-and-phylogeography-of-moose.pdf

Lister et al. (2004): https://www.researchgate.net/profil...graphical-variation-in-cranial-morphology.pdf

 

Thanks for the update. Presumably at Howletts they are in the 'deer park' with the Nilgai, Blackbuck etc? Any idea how many animals this involved and why?

 

Very interesting! It looks like moose, elk, bison and grizzly bears in America are populations which migrated from Asia during the last Ice Age, and guanaco, cougar and jaguar became almost extinct (went through a population bottleneck). Maybe humans, besides wiping out mammoths, giant sloths etc. almost wiped out these species currently alive?

 

So does Berlin keep Indian Sambar now? It'd be nice if the ex-Berlin animals could have made it back into the US population,though the US government would likely never allow it.

~Thylo

 

I have seen a mixed herd of Wapiti and Red Deer at Grant's Farm - the animals looked completely different.

 

I once saw a Red Deer stag and a Wapiti cow standing in the brush together while I was driving from Del Rio to San Antonio along Highway 90. The old nursery rhyme about Jack Sprat came to mind...

 

Recent photos of moose in the Minnesota Zoo gallery should be andersoni since they took in a bunch of orphaned calves from northern Minnesota a few years back.

In North America moose often turn up hundreds of km outside their regular range, in South Dakota, Iowa, etc. With that sort of dispersal ability and a lack of major geographic barriers it is tough to envision populations differentiating much from one another. Makes me wonder how Eurasian populations were able to split into three quite different forms. Perhaps different glacial refugia?

Interestingly, some of those species colonized America during different Pleistocene interglacials. Moose and Elk came during the last interglacial, but Grizzly Bears arrived earlier, and Bison earlier still. There is a large body of literature on population size changes of these species, with lots of debate on the relative importance of humans vs climate fluctutations in causing bottlenecks.

 

White-tailed deer

The White-tailed deer (Odocoileus virginianus) is a medium sized species occurring throughout most of North and Central America, as well as northern and northwestern South America. Traditionally up to 39 subspecies have been recognized, though genetic data indicate that this is an overestimate and genetic groupings often are not the same as the originally described subspecies. The following subspecies have been traditionally recognized (taken from IUCN red list):

Morphological and genetic research in Venezuela (Molina & Molinari 1999; Moscarella et al. 2003), as well as genetic research in Mexico (de La Rosa-Reyna et al. 2012) seem to indicate differences between northern and southern White-tailed deer and there might be a case for recognizing multiple White-tailed deer species, but for this extensive sampling throughout the range of this species is necessary.


N-American White-tailed deer:

@Giant Eland , Oklahoma City Zoo, USA


@ThylacineAlive , Beardsley's Zoo, USA

S-American White-tailed deer:

@Giant Eland , Parque Jaime Duque, Bogota, Colombia


@Giant Eland , Bioparque Los Ocarros, Villavicencio, Colombia


Mule deer

The Mule deer (Odocoileus hemionus) is a medium-sized deer subspecies from western North America, ranging from S Alaska to N Mexico. Traditionally up to ten subspecies have been recognized (taken from IUCN red list):

This view is however challenged by genetic and morphological data. There seems to be a main difference between the Mule deer from the NW part of the distribution, the Black-tailed deer and the other Mule deer. Latch et al. (2014) could identify three groups based on genetic analysis sitkensis and columbianus formed two clades within the Black-tailed deer and all the other subspecies grouped together in hemionus without much differentiation between them. This difference is explained as the groupings having different glacial refugials and since the retreat of the glaciers an extensive hybrid zone has established. These hybrids are not infertile and a large stable hybrid swarm exists (Latch et al. 2011) and there seems no evidence of reduced fertility of hybrids. This complex has been described as a species in the making, but based on current evidence the two Black-tailed deer groups are best treated as subspecies of hemionus.

O.h. hemionus

@Dibatag , Calgary Zoo, Canada


@Giant Eland , Hutchinson Zoo, USA


O.h. sitkensis

@Arizona Docent , Alaska Wildlife Conservation Center, Anchorage, USA


@Pleistohorse , Alaska Zoo, Anchorage, USA

O.h. columbianus

@ThylacineAlive , California Living Museum, Bakersfield, USA


@Blackduiker , California Living Museum, Bakersfield, USA


References

De La Rosa-Reyna et al. (2012): Genetic diversity and structure among subspecies of white-tailed deer in Mexico

Latch et al. (2014): Error - Cookies Turned Off

Latch et al. (2011): Error - Cookies Turned Off

Molina & Molinari (1999): https://www.researchgate.net/profil...lar_traits/links/57d19cb308ae0c0081e0502b.pdf

Moscarella et al. (2003): Phylogeography, Population Structure, and Implications for Conservation of White-Tailed Deer ( Odocoileus virginianus ) in Venezuela

next: the Mazama mess.

 

And a short interlude on Serow and Goral.

Recently a paper, Mori et al. (2019), was published on the taxonomy of Serow and Goral, based on the complete mitochondrial genome of most taxa and they lump quite a lot.

Error - Cookies Turned Off

It should be noted that they did not include evansi goral from SE Asia, which in Hassanin et al. (2012) are quite distinct from the Chinese goral and are also morphologically distinct.

Mori et al. propose to lump the Sumatran serow, Chinese serow, Himalayan serow and Indochinese serow back into one species the Sumatran serow, with divergence levels of 1% and less. Chinese serow is polyphyletic in this analysis. The Red serow (Capricornis rubidus) is distinct and sister to Formosan serow. Interestingly they used an animal from Assam as rubidus, which according to G&G is not rubidus, but an undescribed species, though it is morphologically quite similar to the rubidus from nearby Myanmar and Bangladesh. Based on morphology this systematic revision makes sense as rubidus is very distinct, whereas the other taxa are more similar.

Mori et al. also propose to lump Himalayan goral and Chinese goral again (and thus also all subspecies described, even though they did not analyse where evansi should be placed). Long-tailed goral is distinct and sister species to the clade of Himalayan + Chinese goral. Divergence between Himalayan & Chinese goral is found to be only 1.8%, which is indeed rather low and does support their conclusion. They also state that the ranges of Chinese and Himalayan goral are adjacent without a visible barrier.

All in all this is a welcome addition to the discussion on taxonomy of Serow and Goral and the conclusions based on the taxa make sense, it will be interesting to see how nuclear dna results look and when more samples become available. The last word won't have been said yet, but it is an indication of where debate might be heading to. For future research it will be critical to include evansi and also to identify whether Red serow from Assam and Myanmar are indeed the same or whether they differ.

 

I know there was talk a few years ago about lumping these two species. I assume you think this shouldn't be done?

 

I guess that would be because hybrids between these species have been described in several locations in the US? As the majority of these hybrids is sterile it is a strong indication that they should not be lumped, even ignoring obvious differences between the two species.

 

Now it gets really confusing as we start discussing the taxonomy of Brocket deer (Mazama s.l.), this is a group of small deer from the neotropics with relatively little external differences, but an interesting evolutionary history. To start with it is now clear that Mazama is not a monotypic genus and the Red brocket deer (Mazama americana) and closely related species group together with White-tailed deer (Odocoileus) and the Gray brocket deer (Mazama gouazoubira) and related species group together with Pampas deer (Ozotoceros bezoarcticus), Marsh deer (Blastoceros dichotomus) and Huemul (Hippocamelus). This is just one of multiple problems with higher level taxonomy in Neotropic deer species, and I won't go into much detail in to which genus each species should be placed. I will focus more on which species can be recognized starting with the Red brocket deer group (Mazama americana).

IUCN red list recognizes six species of Red brocket:

Mazama americana (Red brocket)
Mazama temama (Central American red brocket)
Mazama bororo (Small red brocket)
Mazama bricenii (Merida brocket)
Mazama rufina (Dwarf red brocket)
Mazama nana (Brazilian dwarf brocket.

G&G do add surprisingly little to the debate about validity of these species and whether there are more species than currently recognized. G&G name 5 extra species of Red brocket, all based on very limited data and of questionable validity. These putative species are not backed up with any genetic evidence, which has been shown in research of Red brocket to be of utmost importance, as the morphological differences are often very limited. Based on the very low sample size (e.g. jucunda is elevated to species status based on 1 specimen, zamora is split on the basis of 1 small antler, where skulls are similar to other Brocket deer), I am not going into much detail about their analysis and I will stick to a growing body of literature on Red brocket systematics. The only strong point about their analysis is that they have included some samples from Peru, Ecuador, Colombia and Trinidad, areas which are currently not covered with genetic data and would be a priority for the future.

First the lumping. bricenii and rufina are currently recognized as separate species, but this is not based on much evidence. Even G&G admit that their morphologic analysis did not reveal any clear differences. Gutierrez et al. (2015) analysed morphological data and genetic data and found that there is no morphological difference and genetically bricenii is nested within rufina. This indicates that there is no basis to split bricenii and it is at best a subspecies of rufina.

The validity of other species has been backed up by genetic research as is the case for temama (Jorge & Benirschke, 1977), bororo (Duarte & Jorge, 2003) and nana (Abril et al., 2010). It is however good to note that sister species bororo and nana were nested with a few americana samples in one study (Duarte et al., 2008).

Mazama americana itself constitutes of at least two cryptic species in N and S Brazil (Duarte et al., 2008; Abril et al., 2010) and these are reproductively isolated, as hybrids bred in captivity were sterile or showed characteristics of subfertility (Cursino et al., 2014; Salviano et al., 2017). These cryptic species differ in chromosomes and as more populations with different karyotypes were found this raises the strong possibility there are more cryptic species of Mazama americana, e.g. in NE Brazil (Cursino et al., 2014). Genetic research across the whole range (not only Brazil) of Mazama americana s.l. is necessary to identify more cryptic species that are likely to exist.

Overall there is strong evidence that 6 species of Red brocket exist and that this is most probably an underestimate, the research of G&G does however not add much value and genetic research will be the way forward with these cryptic species.

Mazama americana s.l.

@Giant Eland , Parque Zoologico Huachipa, Peru

Mazama temama

@Giant Eland , Chapultepec Zoo, Mexico

Mazama rufina

@Giant Eland , Quito Zoo, Ecuador

Mazama rufina bricenii

@Giant Eland , Parque Jaime Duque, Bogota, Colombia

No pictures of bororo or nana seem to be present in the gallery.


References

Abril et al. (2010): https://www.researchgate.net/profil...a_Cervidae/links/0fcfd511e1497ea482000000.pdf

Cursino et al. (2014): The role of chromosome variation in the speciation of the red brocket deer complex: the study of reproductive isolation in females | BMC Evolutionary Biology | Full Text

Duarte & Jorge (2003): https://www.researchgate.net/profil...ocket-Mazama-bororo-Duarte-1996-in-Brazil.pdf

Duarte et al. (2008): https://repository.si.edu/bitstream...ising_evolutionary_history_of_South_Ameri.pdf

Gutierrez et al. (2015): The Taxonomic Status of Mazama bricenii and the Significance of the Táchira Depression for Mammalian Endemism in the Cordillera de Mérida, Venezuela

Jorge & Benirschke (1977): https://www.jstage.jst.go.jp/article/cytologia1929/42/3-4/42_3-4_711/_pdf

Salviano et al. (2017): Intraspecific chromosome polymorphisms can lead to reproductive isolation and speciation: an example in red brocket deer ( Mazama americana )

 

Gray brocket

The Gray brocket group is now normally thought to consist of four species:

Mazama gouazoubira (Gray brocket)
Mazama nemorivaga (Amazonian gray brocket)
Mazama chunyi (Peruvian dwarf brocket)
Mazama pandora (Yucatan brocket)

Genetic research does however indicate Yucatan brocket is more closely related to the Red brocket group (Mazama americana) (Escobedo-Morales et al., 2016).

nemorivaga and gouazoubira were long considered the same species but clear morphological (Gonzalez et al., 2018) and genetic differences have been found and both species are not even very closely related to each other, with gouazoubira more closely related to the Huemul species (Hippocamelus spec.) (Gutierrez et al., 2017). Mazama chunyi is the sister species of gouazoubira in genetic analyses.

G&G recognize seven additional species, which on the limited evidence proposed are of very questionable validity. A combination of very small sample size and very small differentation makes recognition of more species somewhat optimistic. Of two species they recognize, Mazama rondoni, no evidence whatsoever is provided as to why this is a species. Based on morphological characters rondoni was not recognized as being separate (see Rossi et al., 2010).

G&G recognize Mazama cita and this might be an as yet unrecognized species, as it has a tarsal tuft that the other Brocket deer lack (Rossi et al., 2010) and is described to be larger , but more evidence, including genetics, would be necessary to say more about the validity of this taxon.

Genetic work by Gutierrez et al. (2017) did find two separate clades of nemorivaga, one from the NE part of it's range and the other from the SW part of the range. There might thus be a cryptic species hiding in plain sight. Compared to the Red brocket complex, surprisingly little genetic work has been done on Gray brockets and work including samples throughout the range, especially countries other than Brazil, is long overdue.

For now The Gray brocket species complex consists of three clearly defined species, with the note that one or more cryptic species most likely exist. Mazama pandora should be considered part of the Red brocket group.


Mazama pandora

@Giant Eland , Xcaret Park, Mexico

Mazama gouazoubira s.l.

@Giant Eland , Buenos Aires Zoo, Argentina

Mazama gouazoubira murelia (split by G&G)

@Giant Eland , Bioparque Los Ocarros, Villavicencio, Colombia

No pictures of nemorivaga and chunyi have been uploaded. The status of Gray brocket in Faunia, Madrid, is unclear to me, I don't know whether they have been confirmed to be nemorivaga or gouazoubira.


References

Escobedo-Morales (2016): https://www.researchgate.net/profil...mplications-on-Neotropical-deer-evolution.pdf

Gonzalez et al. (2018): https://www.researchgate.net/profil...-of-gray-brocket-deer-species-from-Brazil.pdf

Gutierrez et al. (2017): A gene-tree test of the traditional taxonomy of American deer: the importance of voucher specimens, geographic data, and dense sampling

Rossi et al. (2010): https://www.researchgate.net/public...WN_BROCKET_DEER_Mazama_nemorivaga_Cuvier_1817

 

Here follows a wrap up post containing the remaining S-American deer species. Note that though there is no indication for possible splits here, the taxonomy at genus level is highly confused and most probably differs hugely from the current status quo.

Pampas deer

The Pampas deer (Ocetoceros bezoarcticus) is a species that occurs in Brazil, Uruguay, Argentina and Paraguay. Based on morphological and genetic differences (see Gonzalez et al. 2002) five subspecies are currently recognized:

O.b. bezoarcticus E & C Brazil
O.b. celer Buenos Aires Province, Argentina
O.b. leucogaster Mato Grosso, Brazil, Paraguay & Argentinian Chaco
O.b. uruguayensis E grasslands of Rocha Department, Uruguay
O.b. arerunguaensis NW Uruguay


@Giant Eland , Bioparque M'Bopicua, Fray Bentos, Uruguay


@Giant Eland , Bioparque M'Bopicua, Fray Bentos, Uruguay


Marsh deer

The Marsh deer (Blastocerus dichotomus) is a large species of deer that has a patchy distribution in E Peru, Brazil, Paraguay, Bolivia and Argentina and is extinct in Uruguay. No subspecies are currently recognized, but genetic work by Marquez et al. (2006) indicates that based on genetic data there are potentially 3 different subspecies.


@Saro , Santa Cruz Zoo, Bolivia


@Tim May , Zoo Berlin, Germany


Northern huemul

The Northern huemul or Taruca (Hippocamelus antisensus) is a monotypic deer species that occurs in the Andes from N Peru to W Bolivia and south to N Chile and NW Argentina. There has been debate as to whether Northern and Southern huemul are indeed close relatives or whether their similar appearance is a result of convergent evolution. Genetic research has sometimes placed them as sister species and sometimes not.


@devilfish , Parque de las leyendas, Lima, Peru


@Giant Eland , Parque Zoologico Huachipa, Lima, Peru


Southern huemul

The Southern huemul (Hippocamelus bisulcus) is the rarest S-American deer and has a highly fragmented distribution in S Chile and SW Argentina. This species is currently considered monotypic.


uploaded by @UngulateNerd92 , Zoo Buenos Aires, Argentina


Southern pudu

The Southern or Chilean pudu (Pudu puda) is a small deer species from SC Chile and W Argentina.


@Tim May , Hamerton Zoo, UK


@ro6ca66 , Hamerton Zoo, UK


Northern pudu

The Northern pudu (Pudu mephistophiles) is the smallest deer species and occurs in the Andes of N Peru, Ecuador and SW Colombia. Multiple genetic studies have shown that Northern and Southern pudu are actually not sister species. The Southern pudu being more closely related to Marsh and Pampas deer than to Northern pudu. The correct scientific name for Northern pudu would thus be Pudella mephistophiles.

No pictures of this enigmatic species have been uploaded to the gallery.

References

Gonzalez et al. (2002): https://watermark.silverchair.com/8...TINWBOYX8eugebEPG4X45Dz9X96Z_9FlO3MXXQ4SiRTZk

Marquez et al. (2006): Phylogeography and Pleistocene demographic history of the endangered marsh deer ( Blastocerus dichotomus) from the Río de la Plata Basin

 

Reindeer / Caribou

The Reindeer or Caribou (Rangifer tarandus) is a deer species with a holarctic distribution. A large number of subspecies is currently recognized and although G&G call for a taxonomic revision, genetic research indicates that there is relatively little divergence between different populations (Cronin et al., 2003). Morphologically described subspecies do however not correspond with genetic groupings. Up to 14 subspecies are recognized, of which two, eogroenlandicus and dawsoni, are extinct. These are the 12 provisionally listed extant subspecies:

R.t. tarandus S Norway, introduced to Iceland
R.t. fennicus E Finland, extreme NW Russia (Karelia)
R.t. platyrhynchus Spitsbergen
R.t. buskensis Russia
R.t. pearsoni Nova Zemlya Archipelago
R.t. valentinae Ural mountains, Russia, and Altai Mountains, Mongolia
R.t. sibiricus Siberia
R.t. phylarchus Kamchatka Peninsula and E Russia
R.t. granti Alaska, USA and Yukon, Canada
R.t. osborni British Columbia, Canada
R.t. caribou NW USA and S Canada
R.t. groenlandicus Northwest Territories and Nunavut, Canada and W Greenland


Rangifer tarandus fennicus

@KevinB , Burgers' Zoo, Arnhem, Netherlands


@KevinB , Gaiazoo, Kerkrade, Netherlands

Rangifer tarandus sibiricus

@ThylacineAlive , San Diego Zoo, USA

Rangifer tarandus granti

@Pleistohorse , Alaska, Usa


@Pleistohorse , Alaska, USA

Rangifer tarandus caribou

@Daniel Sörensen , Zoo Hannover, Germany


@Maguari , Zoo Hannover, Germany


Cronin et al. (2003): Error - Cookies Turned Off

That means that Cervidae is now done.

 

GIRAFFIDAE

The family Giraffidae consists of two genera, which historically were thought to consist of just one species each. Recent genetic research has indicated there are 3-8 different species of Giraffa, but there is still only one species of Okapia:

Okapi

The Okapi (Okapia johnstonii) is a large ungulate species endemic to the rainforests of C and E DR Congo. This species is currently considered monotypic.


@Tim May , Marwell Wildlife, UK


@ro6ca66 , Marwell Wildlife, UK


@Tim May , ZSL London Zoo, UK


@KevinB , Zoo Antwerpen, Belgium

 

In North America, I've commonly seen pearyi (Ellesmere and other nearby islands) recognized as well. The Peary caribou is very distinct morphologically from typical barren ground caribou, but wasn't included in the Cronin paper on genetics. The more interesting thing with caribou in North America is that eastern woodland caribou (R. t. caribou) are apparently descended from populations that were south of the Ice Age glaciers in North America, while the various barren ground subspecies are descended from populations that were north of the glaciers in Beringia and the western woodland caribou are a genetic mix of the two (Genetic diversity in caribou linked to past and future climate change). They make a good example of subspecies in the process of merging.

 

Giraffe

Giraffes (Giraffa camelopardalis s.l.) are possibly the most recognizable mammals of them all and traditionally only one species was recognized with up to nine subspecies. In 2007 this traditional view was challenged by two papers, one by @DavidBrown et. al. and one by Hassanin et al., showing large differences between Giraffes across Africa and redrawing subspecific distributions. These two papers gave rise to follow up research and the debate in the scientific community is not whether there is more than one species of Giraffe, but rather "how many?". Depending on whom you ask the answer of "how many giraffe species are there?" ranges from 3-8.

The traditional view of Giraffe taxonomy was the following:

G.c. camelopardalis Nubian giraffe, E Sudan & W Ethiopia
G.c. peralta, Peralta giraffe/W-African giraffe, W Africa
G.c. antiquorum, Kordofan giraffe, including congoensis, S Chad, NE DR Congo and Central African Republic
G.c. reticulata, Reticulated giraffe, S Ethiopia, SW Somalia, N Kenya
G.c. tippelskirchi, Masai giraffe, S Kenya & Tanzania
G.c. rothschildi, Rothschild's giraffe, South Sudan, Uganda & W Kenya
G.c. thornicrofti, Thornicroft's or Rhodesian giraffe, Luangwa Valley, Zambia
G.c. giraffa, Cape or South African giraffe, SW Mozambique, Malawi, Zimbabwe, E South Africa, extreme S Zambia, E Botswana
G.c. angolensis, Angolan giraffe, Namibia, W Botswana

Bock et al. (2014) showed that Giraffes in parts of Botswana & Zimbabwe, as well as S Zambia, did not belong to angolensis as was previously assumed, but to giraffa.

G&G recognize eight species of Giraffe and only lump rothschildi with camelopardalis. Their morphological research does not add a lot to the current discussion. G&G do a good job in describing differences in pelage between the different taxa, but their analysis of skulls does not give much information. It has been noted however that differences in pelage within one taxa can be as big as differences between taxa, but other authors. The G&G analysis is hampered by small sample size and one of the conclusions they draw is that there is little size difference between male and female Reticulated giraffes (and that Reticulated giraffes are thus smaller than closely related subspecies). Given obvious size differences I have seen both in wild and in captive Reticulated giraffes, this conclusion seems more driven by an inadequate sample size than an actual difference.

The genetic story on how many giraffe species exist, is much more interesting. Although in the wild different giraffe (sub)species are distinct genetic units, they interbreed readily in captivity. Thomassen et al. (2013) give evidence that what keeps the different taxa separated in Eastern Africa is differences in climatic patterns, more specifically different timings of main the rainy season. Dispersal barriers and distance played far smaller roles in explaining the separation of the taxa. Three recent genetic studies have given strong evidence for splitting Giraffe into four species (Bock et al., 2014; Fennessy et al., 2016; Winter et al., 2018). Fennessy et al. (2016) found that there are four clear lineages that are monophyletic and genetically quite distant. Winter et al. (2018) built on this dataset and found the same results and additionally showed that there is only very limited gene flow between the four groups. The four groups are (1) Northern giraffe, including camelopardalis, antiquorum, peralta, rothschildi, (2) Reticulated giraffe, (3) Masai giraffe, including tippelskirchi, thornicrofti, (4) Southern giraffe, including giraffa, angolensis. Fennessy et al. (2016) did not find any genetic difference between rothschildi and camelopardalis and proposed lumping both taxa. This has given rise to several zoos now naming their Rothschild's giraffe, Nubian giraffes. G&G support this lump as they did not find a clear morphological (or skull) difference between these taxa. Muller (2018) argues that Brown et al. (2007) found that rothschildi was genetically distinct, but unless I am mistaken Brown et al. (2007) did only include rothschildi and not camelopardalis, so there seems to be no disagreement between Brown et al. (2007) and the later studies. There has also been considerable debate about whether thornicrofti is different from tippelskirchi, with the majority of studies finding no Mtdna difference between these taxa. Fennessy et al. (2013) have however argued that thornicrofti represents a unique isolated location, also based on Mtdna, and nuclear dna results find some minor differences between both taxa (Fennessy et al., 2016), so there is some ground to tentatively recognize this subspecies, especially as G&G describe some morphological differences.

Using a more conservative approach Petzold & Hassanin (2019) argue for recognition of 3 Giraffe species, lumping Reticulated giraffe with Northern giraffe. Although their results seem pretty robust, as far as I can tell, I disagree with this lump. Earlier research has suggested very limited gene flow between Reticulated and Northern giraffe, as well as a climatic explanation as to what kind of dispersal barrier exist. Additionally Reticulated giraffe are morphologically quite distinct and are almost completely restricted to the Somali arid zone, whereas Northern giraffe are present mostly in the Sahelian & Sudan savanna ecosystems.

The debate on Giraffe taxonomy will likely continue in the near future, but what everybody agrees on is that there is more than one species of giraffe. I personally think the best case made so far is to recognize 4 species, following Fennessy et al. (2016) & Winter et al. (2018).

Giraffa camelopardalis camelopardalis "rothschildi"

@Hanjo , Ree Park Safari, Ebeltoft, Denmark


@Jogy , Lake Nakuru NP, Kenya


@KevinB , Gaiazoo, Kerkrade, Netherlands

Giraffa camelopardalis antiquorum

@Maguari , Zoo Doue-la-Fontaine, France

Giraffa reticulata

@lintworm , Samburu NR, Kenya


@lintworm , Samburu NR, Kenya


@KevinB , Zooparc Overloon, Netherlands

Note the variation in leg colour and overall pattern.

Giraffa tippelskirchi tippelskirchi

@lintworm , Mkomazi NP, Tanzania


@Hix , Serengeti NP, Tanzania


@Hix , Tarangire NP, Tanzania

Giraffa giraffa giraffa


@Gondwana , Chobe NP, Botswana


@bongowwf , Biblical Zoo, Jerusalem, Israel


Giraffa giraffa angolensis

@Tim May , Zoo Dortmund, Germany


@Bele , Zoo Lisboa, Portugal

No pictures of thornicrofti or peralta have been uploaded to the gallery.

References

Bock et al. (2014): Mitochondrial sequences reveal a clear separation between Angolan and South African giraffe along a cryptic rift valley

Brown et al. (2007): Extensive population genetic structure in the giraffe

Fennessy et al. (2013): Error - Cookies Turned Off

Fennessy et al. (2016): Multi-locus Analyses Reveal Four Giraffe Species Instead of One - ScienceDirect

Hassanin et al. (2007): https://s3.amazonaws.com/academia.edu.documents/38517741/CRBiologies-Giraffa2007.pdf?response-content-disposition=inline; filename=Mitochondrial_DNA_variability_in_Giraffa.pdf&X-Amz-Algorithm=AWS4-HMAC-SHA256&X-Amz-Credential=AKIAIWOWYYGZ2Y53UL3A/20191109/us-east-1/s3/aws4_request&X-Amz-Date=20191109T115451Z&X-Amz-Expires=3600&X-Amz-SignedHeaders=host&X-Amz-Signature=ed713d031ec3fdd1083f835e49988ae370717d27461bd34121bd7d30b2fe9e66

Muller (2018): Error - Cookies Turned Off

Winter et al. (2018): Error - Cookies Turned Off