Skip to main page content
Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
, 8 (20), 10156-10165

Limited Introgression Supports Division of Giraffe Into Four Species


Limited Introgression Supports Division of Giraffe Into Four Species

Sven Winter et al. Ecol Evol.


All giraffe (Giraffa) were previously assigned to a single species (G. camelopardalis) and nine subspecies. However, multi-locus analyses of all subspecies have shown that there are four genetically distinct clades and suggest four giraffe species. This conclusion might not be fully accepted due to limited data and lack of explicit gene flow analyses. Here, we present an extended study based on 21 independent nuclear loci from 137 individuals. Explicit gene flow analyses identify less than one migrant per generation, including between the closely related northern and reticulated giraffe. Thus, gene flow analyses and population genetics of the extended dataset confirm four genetically distinct giraffe clades and support four independent giraffe species. The new findings support a revision of the IUCN classification of giraffe taxonomy. Three of the four species are threatened with extinction, and mostly occurring in politically unstable regions, and as such, require the highest conservation support possible.

Keywords: conservation; gene flow; giraffe; hybridization; speciation.


Figure 1
Figure 1
Reticulated giraffe (Giraffa reticulata) in the Samburu National Reserve, Kenya (©GCF)
Figure 2
Figure 2
Map of Sub‐Saharan Africa with giraffe (sub)species distributions and sampling locations. Geographic ranges (colored shadings) of giraffe as identified by the Giraffe Conservation Foundation (2017) were plotted on a map of Sub‐Saharan Africa. Numbered circles represent sampling locations (for details see Supporting information Table S1). Species and common names as per Fennessy et al. (2016)
Figure 3
Figure 3
Nuclear phylogeny and population structuring of giraffe. (a) Bayesian multi‐locus tree from 21 nuclear loci and 137 giraffe individuals reconstruct four significant supported (p ≥ 0.95) giraffe clades, corresponding to the four giraffe species (Fennessy et al., 2016). The okapi is used as the outgroup. The asterisks indicate branches with statistical significant support (p ≥ 0.95). The red frame indicates the potential hybrids. (b) STRUCTURE analysis of the dataset, excluding the okapi. The colors indicate the membership in a cluster for each sampling location and individual. K = 4 shows four well‐resolved groups and is supported as best fitting number of clusters by several statistical methods (see Supporting information Figure S2). The grouping into four clusters is consistent with the Bayesian multi‐locus analysis: yellow: northern giraffe, orange: reticulated giraffe, green: Masai giraffe, and blue: southern giraffe. Three individuals within the reticulated giraffe cluster (red arrowheads) indicate potential hybridization with admixture from the northern giraffe. K = 3 merges northern and reticulated giraffe, and at K ≥ 5 no further clustering is evident. (c) PCA axes 1‐2 and axes 1‐3 for four distinct giraffe clusters (1: northern; 2: reticulated; 3: Masai; 4: southern). Colors as in Figure 3b. The 95% confidence intervals are shown as oval outlines. Note that the nonoverlapping confidential intervals in the PCA axes 1‐2, as well as, axes 1‐3 indicate significantly different clusters. Potential hybrids are indicated by black circles. Note. The drawing by Jon B. Hlidberg shows a Nubian giraffe
Figure 4
Figure 4
Circular migration plot of recent migration rates among four giraffe clades. Recent directional migration rates (m) as estimated by BayesAss and indicated by ribbons connecting one species to another. The color coding of the four species is according to the STRUCTURE clusters (Figure 3b). Peripheral concentric stack bars show relative migration rates in percent. Whereas the inner stack bar shows the outgoing ribbon sizes, the middle stack bar the incoming ribbon sizes and the outer stack bar the combination of both

Similar articles

See all similar articles

Cited by 2 articles


    1. Arnold M. L. (2016). Divergence with genetic exchange. Oxford, UK: Oxford University Press.
    1. Avise J. C., & Ball R. M. Jr (1990). Principles of genealogical concordance in species concepts and biological taxonomy In Futuyama D., editor; & Antonovics J., editor. (Eds), Oxford surveys in evolutionary biology (pp. 45–67). Oxford: Oxford University Press.
    1. Beerli P. (2006). Comparison of Bayesian and maximum‐likelihood inference of population genetic parameters. Bioinformatics, 22, 341–345. 10.1093/bioinformatics/bti803 - DOI - PubMed
    1. Beerli P. (2012). Migrate documentation version 3.2.1. Tallahasee, FL: Florida State University.
    1. Beerli P., & Felsenstein J. (2001). Maximum likelihood estimation of a migration matrix and effective population sizes in n subpopulations by using a coalescent approach. Proceedings of the National Academy of Sciences, 98, 4563–4568. 10.1073/pnas.081068098 - DOI - PMC - PubMed

LinkOut - more resources