Australia: The Land Where Time Began

A biography of the Australian continent 

Ancient Human Influences on the Evolution of Baobab Trees and Distribution in Australia – New Genetic and Linguistic Analysis

This study investigates the role of human agency in the gene flow and geographical distribution of the Australian Baobab, Adansonia gregorii. The genus Adansonia is an iconic genus that is endemic to Africa, Madagascar and northwest Australia (Baum, 1995; Pettigrew et al., 1995) that has been valued by humans for its many uses. In Africa the distribution of genetic variation in baobabs has been partially attributed to dispersal that humans mediated over thousands of years, though this relationship has never been investigated for the Australian species. Rangan combined linguistic and genetic data in order to analyse geographic patterns of gene flow and movement of word-forms for A. gregorii in the Aboriginal languages of northwest Australia. Weak geographical structure and high gene flow was shown by comprehensive assessment of genetic diversity. Congruence of gene flow patterns and directional movement of Aboriginal loan words for A. gregorii was demonstrated by genetic-linguistic analysis. Along with previous archaeobotanical evidence from the Late Pleistocene and Holocene, these findings suggest that ancient humans influenced significantly the geographic distribution in northwest Australia of A. gregorii.

The role humans played in the shaping of crop diversity has always been considered an integral factor of the evolution of agriculture in different regions of the world (Donohue & Denham, 2009; Perrier et al., 2011; Roullier, Benoit, McKey & Lebot, 2013). Combining genetics, linguistics and archaeobotany with interdisciplinary research has enhanced further the understanding of the geographic patterns of the domestication of animals and crops and subsequent diffusion of humans (Donohue & Denham, 2009; Perrier et al., 2011; Roullier, Benoit, McKey & Lebot, 2013). Yet, very little comparable research has been carried out on how the evolution and distribution of uncultivated plants that have, nonetheless, had a long history of use by humans has been influenced by anthropogenic agency (Blench, 2007; Smith, 2011; Bostoen & Grollemund, 2013). Adansonia, the baobab tree, an iconic genus that is endemic to the Kimberley region of Australia, Africa and Madagascar, is a striking example of this.         These giant trees that are long lived, hold significant cultural symbolism, and as a food source that had multipurpose value, and also as medicine, water storage, shelter and raw material for artisanal [? artistic] purposes in all the places they are endemic to (Adanson, 1771; De Caluwé, Halamová & Van Damme, 2009; De Caluwé, Halamová, Van Damme, 2009; Gebauer, El-Siddig & Ebert, 2002; Livingstone, 1861; Patrut et al., 2013; Sidibe & Williams, 2002; Swart, 1993; Wickens, 1982; Wickens & Lowe, 2008). The distribution of the African baobab species, Adansonia digitata, L., has been linked closely to dispersals of humans and settlement patterns (Armstrong ed., 1979; Duvall, 2007), though there is no evidence of them being cultivated historically. This association is also recognised by the diversity and borrowing of terms for baobabs between language groups in Africa (Blench, 2007; Wickens, 1982). Previous research on the evolution and geographic distribution of the Australian baobab, Adansonia gregorii, F. Muell, in contrast, has been based on the assumption of long-term natural processes (Baum, 1995), with no significant influence by human agency. Rangan suggests the assumption may have been based on the view that is long-held of Aboriginal Australia as a ‘continent of hunter-gatherers’ (Barker, 2006; Bean, 2007; Bellwood, 2005; Denham, Donohue & Booth, 2009; Lourandos, 1997) where anthropogenic agency was limited to ‘firestick farming’ of landscapes for nomadic foraging and hunting (Bleige-Bird, Bird, Codding & Jones, 2008; Jones, 1969). In this study Rangan explored the role in humans in shaping the evolution of A. gregorii by the determination of whether the geographic distribution of genetic diversity, explained partially by patterns of human migration, as was inferred by linguistic analysis.

It is shown by levels of genetic diversity that A. gregorii separated from other species of Adansonia more recently than the breakup of Gondwana, though prior to the arrival in Australia of humans (Baum, 1995; Pettigrew et al., 2012). It was demonstrated (Leong-Pock Tsy et al. (Leong Pock Tsy et al., 2009) that seeds of A. digitata retain their viability in seawater, which makes dispersal by ocean currents feasible. It can be inferred from this that A. gregorii arrived in Australia before humans. It is also possible that the species arrived in Australia more recently from a population that is not known and is now extinct, though this is less parsimonious. One hypothesis that outlined about how A. gregorii may have arrived in Australia with humans has been explored in more detail Pettigrew (Pettigrew, 2011). Known as ‘boab’ in Australia, Adansonia gregorii is mainly distributed across the Kimberley region in northwestern Australia, and there are small extensions eastwards into the Victoria River District in the Northern Territory. The A. gregorii distribution in the Kimberley extends from the northern coastline the edge of the Great Sandy Desert and the Tanami Desert (Wickens & Lowe, 2008; Pettigrew, 2011; Baum & Handasyde, 1990; Brock, 1988; Gillison, 1983; Mueller, 1893). The westernmost extent of the Australian Monsoon Tropics (AMT) is represented by the Kimberley region. In urban centres of northern Australia this tree has been introduced more recently for ornamental purposes (Wickens & Lowe, 2008).

To the south the AMT is bounded by arid habitats, which began to develop in the Late Cainozoic and contain biota that is distinctly different (Bowman et al., 2010; Byrne et al., 2008). In northwestern Australia the major biogeographical divide is between the Kimberley to the west and Arnhem Land to the east, with barriers that are more localised and specific that were formed by major river drainage systems (Eldridge, Potter & Cooper, 2011; Hill & Johnson, 1995; Oliver, Adams & Doughty, 2010; Potter et al., 2012). Phylogeographic patterns for rock wallabies (Petrogale spp) and other species within the Kimberley suggest an East-West divide that runs through the central Kimberley (Bowman et al., 2010; Hill & Johnson, 1995; Potter et al., 2012). A previous detailed population genetic analysis of (A. gregorii) has demonstrated that there is little genetic structure, with F_st values that are non-significant between most populations (Bell et al., 2014), in spite of evidence of biogeographic barriers. An arrival in the Kimberley that is relatively recent, a recent genetic bottleneck, or high dispersal rates across the range of the species, could explain the low geographic structure. High dispersal is the most likely explanation, for reasons that are detailed in Bell et al., (Bell et al., 2014).

In this paper the aim was to evaluate the latter hypothesis – that the low levels of genetic structure within A. gregorii result from high levels of gene flow and, specifically, the dispersal of seeds being mediated by humans has been an important evolutionary factor in the history of the species. Occasionally, A. gregorii may be pollinated by birds or bats, though it is the hawkmoth that is the major pollination agent (Wickens & Lowe, 2008; Baum & Handasyde, 1990; Bowman 1997; Baum, 1995; Lowe, 1998). Gene flow that is mediated by pollination is limited to the paternal genome, and in species that are pollinated by insects, it is often a mechanism of some long distance gene flow that is less effective than fruit dispersal (Aldridge et al., 1998; Garcia, Jordano & Godoy, 2007; Oddou-Muratoria et al., 2001). Some dispersal by fruit (Baum, 1995; Wickens, 1982) could be explained by floodwaters, though this mechanism would probably not spread seeds beyond the edges of seasonal waterways and alluvial flats because of the fragile and dehiscent nature of the pericarp of A. gregorii (Wickens & Lowe, 2008; Bowman, 1997).  Mammals such as rock wallabies (Petrogale spp.), other wallabies and kangaroos (Macropus spp.), which eat the fruit and disperse the seeds in their scat (Wickens & Lowe, 2008), are other possible seed dispersal mechanisms. Phylogeographic studies of the short-eared rock wallaby (P. brachyotis) have shown, however, that there was strong genetic structure, which suggests that at least this species has limited ability to disperse seeds across environmental barriers in the Kimberley (Potter, Eldridge, Taggart & Cooper, 2012; Potter et al., 2012).

Anthropogenic agency has not been formally considered as a vector of gene flow for A. gregorii, in spite of archaeological evidence of its long-term use by Aboriginal groups in the region (Pettigrew, 2011; Boland & Brooker, 1985; McConnell & O’Connor, 1997; McConnell & O’Conner, 1999; Wallis, 2001). This omission may have resulted from the fact that the boab was not cultivated historically, or not considered to be part of the food crops that were culturally harvested by Aboriginal people. Also, the presence of the remains of boab fruit at 1 or 2 archaeological sites is not enough to consider dispersal by humans across the geographic range of the species. Evidence is also required to evaluate whether humans have played a role in gene flow of the boab. According to Rangan they reasoned that historical linguistics, which has been used combined with data sources to trace geographic diffusion patterns of domesticated species (Donohue & Denham, 2009; Perrier et al.,  2011; Roullier, Benoit, McKey & Lebot, 2013), could be applied to a tree species that was not domesticated such as A. gregorii (Bostoen, Grollemund & Muluwa, 2013). Specifically, in this study, they investigated the role human agency played in gene flow of A. gregorii by testing for congruence between the spatial distribution of genetic variation in A. gregorii trees and associated word forms of Aboriginal languages of northwest Australia. They demonstrated that there was a high degree of spatial overlap between the genetic and linguistic data. It is indicated by the results of this study that, as has been shown previously for A. digitata in Africa (Duvall, 2007; Assogbadjo et al., 2010; Leong Pock Tse et al.,  2009), the genetic diversity of A. gregorii in northern Australia has been influenced by ancient humans, probably by acting as agents of seed dispersal over long distances.

The hypothesis that dispersal by humans has played a role in shaping the geographical distribution of A. gregorii is supported by the results of this study. In northwestern Australia the limited morphological and genetic divergence of the genus Adansonia can be attributed to gene flow within A. gregorii, in spite of the presence of biogeographic barriers. The lack of evidence of barriers to gene flow is explained most easily by a long history of humans moving boab seeds, given the lack of other seed dispersal agents. The hypothesis of gene flow by human-mediated dispersal of A. gregorii is supported further by the concordance between gene flow and loanword diffusion. An indication of patterns of interaction between nPNy language families and PNy subgroups that would have influenced the distribution of A. gregorii in the Kimberley region is provided by the movements of loanwords across all boab populations. At Carpenter’s Gap Shelter in the Napier Ranges of the Kimberley, which is within the Banuba language area (McConnell & O’Connor, 1997; McConnell & O’Connor, 1999), Additional evidence of anthropogenic agency is facilitated by gene flow. Continuity of human occupation that extended over 40 ka into the Late Pleistocene was shown by the presence of lithic and macrobotanical remains. Fragments of boab pods recorded dates of 39 ka, 20 ka and 18 ka and 15 ka (McConnell & O’Connor, 1997; McConnell & O’Connor, 1999) and from 3 ka onwards, peaking at 650 years ago, a substantial increase in deposition of pods (McConnell & O’Connor, 1999). Further evidence of long-term consumption by humans of the fruit (McConnell & O’Connor, 1997), was provided by the presence of old boab trees near Aboriginal middens in western Kimberley. In the northern Kimberley prehistoric rock art also shows depictions of the tree, which indicates its cultural significance to ancient groups that may have occupied this region (Pettigrew, 2011).

It was postulated by Rangan that, based on the above evidence, that recent evolution and geographic distribution have been shaped primarily by ancient human agency. It is implied by the phylogenetic tree of boab populations and the predominant direction of gene flow together, that the source populations of A. gregorii gene flow were most likely in the extreme northwestern Kimberley, which potentially overlaps with the inferred NC population area. According to Ragan it is likely that the range of this source population extended beyond the coastline of the present at the Last Glacial Maximum (LGM, roughly 20 ka) when the sea levels were more 120 m below what it is at present, and the northwest continental shelf was exposed to the maximum extent (Kershaw, 1995). The increased exposure of land surfaces of both the Sahul and Sunderland shelves, lower sea surface temperatures (SST), and oceanic currents that were altered as the result of the closure of several shallow seas and passageways between these continental shelves contributed to a northwards shift of the Inter-Tropical Convergence Zone (ITCZ), and thereby reducing seasonal precipitation levels (Decker, Tapper & van der Kaars, 2002), resulting in the formation of semiarid savannah conditions (Bird, Taylor & Hunt, 2005) in which ancestral populations of boab would have existed in northwest Australia. Subtropical desert conditions would have prevailed across much of the exposed continental shelf beyond the Kimberley of the present (Reeves et al., 2013; Hesse, Magee & van der Kaars, 2004; Fitzsimmons et al., 2013; Wyrwoll & Miller, 2001), with rainfall as much as 30-50% below those of the present (De Dekker, Tapper & van der Kaars, 2002) and higher levels of aeolian landform activity (Bowler, Wyrwoll & Lou, 2001). The distribution of A. gregorii would have been limited by these arid conditions with low levels of rainfall in particular, as the current distribution of the species coincides with areas that receive at least 700 mm of annual seasonal rainfall (Beard, 1967). It is likely, therefore, that under the drier climatic conditions during the LGM that resulted from the northwards shift of the ITCZ (De Dekker, Tapper & van der Kaars, 2002), that populations of boab would have been limited to the extreme northern coast of the Kimberley of the present, as well as to the continental shelf.

During the postglacial period and the Pleistocene-Holocene transition, the subsequent sea level rise and restoration of monsoonal activity between 17 and 6 ka the Sahul Shelf was flooded and established the coastlines of northern Australia of the present (Kershaw, 1995; Hesse, Magee & van der Kaars, 2004). The flooding of the continental shelf beyond the Kimberley coast of the present, as well as increased monsoonal rainfall over inland Kimberley, would have altered the distribution of ancestral populations of A. gregorii and possibly led to a genetic bottleneck from which the current populations would have expanded (Bell et al., 2014). According to Rangan it was proposed that ancient human groups living on the coast along the shelf that had previously been exposed during the Late Pleistocene would have retreated from the inundated areas, carrying boab fruit with them as they migrated further south and east.

Records from the Pleistocene of boab remnants at the Carpenter’s Gap archaeological site (McConnell & O’Connor; McConnell & O’Connor, 1999) may represent sporadic visits of human settlers from the north (McConnell, 1997). The site records the presence of shells and beads from the Early Holocene, which suggests the movement of high value goods from the coast (O’Connor et al., 2014). Evidence of occupation at Carpenter’s Gap, in common with other archaeological sites in Australia, increases sharply from the Middle to Late Holocene, which   possibly reflects a demographic expansion in southern Kimberley (Williams, 2013).This  population increase, along with other factors such as climate and vegetation change (McConnell & O’Connor, 1999; Wallis, 2001), growth in the local populations of boab, and the more frequent use of the site for cultural ceremonies and exchanges (McConnell, 1997), could explain the increase in the remnants of boab pods, as well as other foods and seeds from about 3 ka to 650 years ago (McConnell & O’Connor, 1997).

It is likely that in the Late Holocene the increased mobility of Aboriginal groups in the southern Kimberley could have been influenced by greater climatic variability in northern Australia. The Late Holocene (~1,000 BCE – 500 CE) the climate in the Australian Monsoon tropics was marked by periods of increased seasonality and aridity (Lees & Clements, 1987; Brockwell et al., 2013; Shulmeister, 1999). Rangan suggests these conditions may have contributed to increased mobility of Aboriginal groups between different parts of the Kimberley and would have contributed to higher gene flow of A. gregorii by the dispersal of fruit seeds, which would have been accompanied by diffusion of word forms.

Additional indication of patterns of migration and social interaction that could have contributed to the spread of A. gregorii is provided by linguistic data. During the Late Holocene the periods of aridity may have affected the survival of desert-based Marngu (PNy subgroup) speakers, leading to the migration into southern and western Kimberley. Rangan suggests the word form larrkarti was probably coined by Marrngu speakers during this period when they would have encountered the boab tree in the landscape of the Kimberley. The mobility and cultural interactions with the mobility and cultural interactions of the PNy groups with neighbouring nPNy language groups such as the Nyikina and the Banuba groups would have increased at the same time. Evidence that the loanword larrkarti is of a relatively recent nature is demonstrated by the way in which it has been incorporated into other languages, generally in a form that has been unchanged (McConvell & Laughren, 2004). The Gija language (Southern Jarragan has retained its inherited word jumulu for the tree and adopted the larrkarti for the fruit pod, possibly indicating its salience and the portability of the edible seed pod in the more recent borrowing. Likewise the Jaru language (PNy Ngumpin subgroup) has no inherited words for boab and uses both Jamula (modified from Gija) and larrkarti for the tree. Other examples of recent expansion of the boab and loanword diffusion further east is present in the Ngarinyman language (Ngumpin) in the Northern Territory, where the words jang-nge (borrowed from miriwong, meaning for eating) and jumulu (borrowed from Gija) are used for the fruit or its edible pith, and the tree respectively. In the Kimberley, the possible climatic influence on human migration and movement of the boab loanword echoes some aspects of Bostoen et al. (Bostoen, Grollemund & Muluwa, 2013) of climate-induced dynamics and expansion of Bantu in Africa. The PNy Marrngu term larrkarti, as well as other pre-existing words, such as wajarr, jumulu, and kertewun moved across the Kimberley in patterns that corresponded with multidirectional boab gene flow as was shown in this study, though there is no analogous evidence of a large-scale expansion of a single linguistic group into the Kimberley. Rangan suggests the high gene flow in A. gregorii appears to be similar to the case of A. digitata in Africa, where human agency has been involved in dispersal of the species (Duvall, 2007; Assogbadjo, 2010; Leong Pock Tsy, 2009). The boab loanword movement in the Kimberley may be compared with Blench’s (Blench, 2007) account of the spread of Bantu words for A. digitata. He noted that the genetic diversity of baobabs in the ecological zones of West Africa and the diversity of vernacular names for the African tree suggests considerable antiquity as well as well as significant east-west movement along trade routes and exchanges of associated terms and terminology. In spite of the diversity of baobab names, however, he points out that 2 competing Bantu roots, #mbuyu and #muramba and variations of these, are present in the Bantu languages of southern and eastern Africa. According to Rangan it is said that the Bantu expansion from the tropical forests of West Africa to have begun from about the middle of the first millennium BC (~2,500 BP onwards) and by 500 CE reached southern Africa (Bostoen, Grollemund & Muluwa, 2013; Bostoen, Grollemund & Muluwa, 2013; Ehret, 2001). It is argued by Blench that the Bantu would not have been familiar with the baobab as it doesn’t grow in areas of tropical forest in Cameroon, Gabon and Congo where the protolanguage is believed to have originated. The tree would have been encountered by them as they expanded eastwards and emerged into the savannah, where they developed new terms by borrowing from resident hunter gatherer groups or comparing it with some species of trees they already knew. In the Bantu languages if eastern and southern Africa the loan or variations of baobab words mbuyu and muramba would therefore indicate the movement of Bantu into these Areas (Blench, 2007). The expansion of Bantu over this 3,000 year period and their contribution to high levels of gene flow in baobabs across these regions may have resulted in the lower levels of diversity of A. digitata in eastern and southern Africa that was detected by Leong Pock Tsy et al. (Leong Pock Tsy, 2009).

Conclusion

The limited intraspecific divergence within A. gregorii in Australia is probably the result of high gene flow that was mediated by human agency, similar to that inferred by A. digitata in continental Africa (Leong Pock Tsy, 2009; Assogbadjo, Gele Kakai, Kyndt & Sinsin, 2010), combined with suitable habitat shifts and a weak bottleneck following the end of the LGM (Bell et al., 2014). The use of Adansonia by humans over many thousands of years on both continents would have contributed to gene flow over long distances and across biogeographical barriers. It could be, in contrast, that the divergence of Adansonia into 6 species in Madagascar was possible partially because humans were not present until about 2 ka. This hypothesis can only be tested by investigation of the ecological, physiological and biogeographical processes that contributed to the speciation within the Adansonia clade from Madagascar.

According to Rangan et al. this study has contributed new evidence to the role of humans in influencing the evolution and distribution in Australia of a non-domesticated plant species. For a long time Australia has been viewed as a continent of hunter gatherers (Denham, Fullagar & Head, 2009), where prehistoric and pre-contact Aboriginal populations played minimal roles in selecting and dispersing useful plants (Denham, Donohue & Booth, 2009; Lourandos, 1997). This assumption has been challenged by some recent studies, however, by providing evidence that ancient Aboriginal groups had a role in the dispersal of food plants across the continent (McConnell, 1997; Kondo et al., 2012; Gott, 1983). Included among these are bananas, (Musa spp.) taro (Colocasia esculenta), in northern Australia (Denham, Donohue & Booth, 2009), Livistona (Kondo et al., 2012) palms in Central Australia and some yams (dioscoria spp.) in northern Australia (Kondo et al., 2012), and yam daisy (Microseris scapigera) in southeastern Australia (Gott, 1983). New insights regarding the influence of human agency in the evolution and distribution of the boab, an important non-cultivated food plant species that shaped the long-term landscape and environmental history of northwest Australia, by the findings of this study.

Rangan, H. (2015). "New genetic and linguistic analyses show ancient human influence on baobab evolution and distribution in Australia." PLoS ONE 10.