The Hidden History of Kiwi Diversification

Of the millions of Earth’s species that likely remain to be described, a majority is thought to be invertebrates, plants, fungi, or microbes. Nevertheless, the pace of species description in some vertebrate groups has not slackened over the past few decades. This hidden vertebrate diversity is surprising because it resides in the clades with which scientists and society are most familiar. It also speaks (at least in part) to the challenges of finding species boundaries in morphologically conserved taxa, wherein traditional characters provide few clues to true taxonomic diversity.
Besides an increasing amount of sequence data, recent progress in species delimitation is also being fueled by coalescent-based methods for modeling population histories. A new paper by Jason Weir et al. brings some of those tools to bear in attempting to resolve diversity and diversification of the kiwis (Order Apterygiformes, Genus Apteryx). Kiwi are flightless, morphologically conservative, forest-dwelling birds endemic to New Zealand (Fig. 1). Along with their ratite relatives, they are the conspicuous outgroup to all other birds. Any new species of apterygiform would therefore represent particularly disproportionate increases in global avian diversity. However, the systematics and phylogegography of kiwi have rarely been addressed with molecular data.

Left: Species tree of genus Apteryx inferred from 1,000 SNPs, with 11 major lineages colored (Weir et al. 2016, PNAS). Right: photo of a juvenile brown kiwi (Apteryx mantelli; photo courtesy of (http://www.nhc.net.nz).

Fig. 1. Left: Species tree of Apteryx inferred from 1,000 SNPs, with 11 major lineages colored (Weir et al. 2016, PNAS). Right: photo of a juvenile brown kiwi (Apteryx mantelli; courtesy of http://www.nhc.net.nz).


Weir et al. sampled 2 mitochondrial regions and >5kb of the nuclear genome (using a genotype-by-sequencing approach) of kiwi across their range. Species delimitation in BPP (Yang and Rannala 2010) supported existence of 16-17 Holocene lineages of kiwi, 11 of which are still extant (currently, only 5 species are recognized (!)). At least some of these lineages represent incipient species. Still, regardless of whether these lineages are best recognized at the specific or subspecific level, their conserved anatomy and plumage have resulted in them being overlooked by systematists for decades.
Most of Weir et al.’s paper is focused on contextualizing such high levels of cryptic diversity within a fresh phylogenetic framework. By taking a coalescent-based dating approach, they find a massive upturn in kiwi diversification rates over the past 780,000 years. This is coincident with the most severe of Pleistocene glacial periods, which also induced mountain glaciation in across New Zealand. It disagrees, however, with the hypothesis of some workers that recent bird diversification has been minimally affected by climatic cycling.
Fig. 2. Estimated effective population sizes of 11 extant lineages of kiwi over the last glacial/interglacial cycle. (Fig. 4 from Weir et al. 2016, PNAS)

Fig. 2. Estimated effective population sizes of 11 extant lineages of kiwi over the last glacial/interglacial cycle. (Fig. 4 from Weir et al. 2016, PNAS)


Weir et al. argue convincingly that kiwi underwent significant, recent diversification in allopatry – primarily as a result of Pleistocene climate and environmental change. This narrative is supported by the known geological history of the islands, the fossil record, and coalescent-based effective population size estimates (all lineages except 1 experienced strong late Pleistocene bottlenecks; Fig. 2). As the methods used by Weir et al. continue to be combined with genomic-scale data, expect hidden histories of diversification in other morphologically cryptic clades to be exposed.
 
Cited

Weir, J.T., Haddrath, O., Robertson, H.A., Colbourne, R.M., Baker, A.J. (2016). Explosive ice age diversification of kiwi. PNAS E5580–E5587. doi:10.1073/pnas.1603795113

Yang, Z., Rannala, B. (2010). Bayesian species delimitation using multilocus sequence data. PNAS 107, 9264–9269. doi:10.1073/pnas.0913022107

This entry was posted in genomics, phylogenetics, phylogeography, species delimitation, Uncategorized. Bookmark the permalink.