The dispersal of Homo sapiens / Southern Asia

The dispersal of Homo sapiens across southern Asia: how early, how often, how complex?

2012 Elsevier Ltd. All rights reserved.

Robin Dennell, Michael D. Petraglia; Department of Archaeology, University of Sheffield; School of Archaeology, University of Oxford

(I’m posting the discussion first because this is a long post) Essentially Dennell is politely refuting “Out of Africa” versions 1 and 2.

1.4. Discussion and implications

As with Out of Africa 1, the existing model of Out of Africa 2 has been a useful way of explaining a very small amount of information. Nevertheless, the present model unduly restricts the ways we can interpret the sparse, but recently much improved, data upon which it rests, and there are benefits to be gained from widening our range of possible hypotheses.

A useful initial step would be for us to be more explicit about just how few reliable observations we have of early H. sapiens in southern Asia; others would be to explore the likelihood that more than one donor population was involved in the earliest expansion of H. sapiens from Africa, that these expansions may have begun much earlier than currently envisaged, and dispersed via more than one entry point. As indicated above, there are good reasons for arguing that our species may have dispersed from Africa into Arabia and India earlier than previously thought, and therefore we should be open-ended as to how far east, and how early, they may have dispersed.

Although movement out of Africa was likely restricted during the driest parts of MIS 6 (MIS –Marine Isotope Stages: paleoclimate time scale of warm and cool periods) there is no obvious reason why H. sapiens could not have entered Arabia during the Last Interglacial or during moist episodes during MIS 6 and perhaps migrated further east. Additionally, we have yet to develop models that accommodate the probability that early H. sapiens in Asia interbred with both Neanderthals and Denisovans. Here, it would help to move discussion beyond an either/or approach to Out of Africa or multiregional models of modern human evolution, as both migration from Africa and inter-breeding with indigenous populations were likely involved, even if the role of each is impossible as yet to determine in most regions of southern Asia.

Investigations of how, when, and how often H. sapiens dispersed across southern Asia need also to be integrated with environmental and climatic studies so that the responses of human (and mammalian) populations to the changing conditions of MIS 3 and 4 (and perhaps earlier stages) can be properly assessed. We need to recognise that not all dispersal events are successful, and the first appearance of a species such as H. sapiens in a region does not necessarily indicate permanent colonisation. Under the challengingconditions of the Pleistocene, local contractions and even extinctions of populations are likely, just as with the Middle Pleistocene. In the same way that studies of Early and Middle Pleistocene hominin demography need to distinguish between short-term visitors and long-term residents, so too do those of hominin dispersal in the Upper Pleistocene. Finally, discussions of the expansion of early H. sapiens from Africa need to be kept distinct from discussions of the diffusion of “modern behaviour” however defined. Early H. sapiens populations were emphatically not “modern” in the same senses as Holocene or Late Pleistocene ones, and the expansion of H. sapiens across southern Asia (and other regions) before 50 ka may have stemmed as much from ecological opportunism as from behavioural superiority over other hominin species.

The timing and the paths of colonization of southern Asia by Homo sapiens are poorly known, though many population geneticists, paleoanthropologists, and archaeologists have contended that this process began with dispersal from East Africa, and occurred between 60,000 and 40,000 years ago. However, the evidence for this scenario is very weak, particularly the lack of human skeletal evidence between the Levant and Borneo before 40 ka and other explanations are possible. Here we argue that environmental and archaeological information is increasingly indicating the likelihood that H. sapiens exited Africa much earlier than commonly thought, and may have colonized much of southern Asia well before 60,000 years ago. Additionally, we cannot exclude the possibility that several dispersal events occurred, from both North and East Africa, nor the likelihood that early populations of H. sapiens in southern Asia interbred with indigenous populations of Neanderthals, Denisovans and Homo erectus. The population history of southern Asia during the Upper Pleistocene is likely far more complex than currently envisaged.

pleist_clim p5p2-lg


Although the expansion of our species from Africa into Europe between 30 and 40 ka has been extensively investigated for over a century, far less is known about its earlier expansion across southern Asia from Arabia to Australia. Over the last few years, a consensus has arisen that this expansion probably occurred between 40and 60 ka. Although opinion is far from unanimous, many palaeoanthropologists would support most components of the following consensus view of how and when this occurred:

Homo sapiens originated in East Africa by195-160 ka but did not enter Southwest Asia until the Last Interglacial, when it is first evidenced outside Africa by skeletal evidence from the caves of Skhul and Qafzeh in the Levant at 100 ka or perhaps as early as 125 ka, at the beginning of the Last Interglacial. It remained there until ca 65 ka when it was replaced by Neanderthals who may have been forced  southwards by the increasingly harsh conditions of MIS4. In this scenario, after being present in the Levant for ca 35-60 ka, these early populations of H. sapiens went extinct, thus leading to the argument that this was a failed dispersal by a groupof humans that were not as socially and cognitively advanced as later populations of H.sapiens.

Following an assumed demographic increase of populations in Africa, H. sapiens expanded across southern Asia between 40 and 60 ka and those that reached Sundaland, the landmass of modern island Southeast Asia that was linked to the Malaysian peninsula because of lowered sea-levels; they eventually crossed the open-sea to enter Australia. Some researchers have suggested that modern mans were the first to exploit systematically coastal resources (notably shellfish, sea mammals and inshore fish) and were thus able to disperse rapidly along a coastal route possibly by utilising coastal springs that emerged when sea levels fell and thus by passing indigenous populations that lived further inland. By this means, they were able to reach Australia by 40 ka or 60 ka, depending upon which dates are preferred from the Australian inland.

Although this consensus enjoys widespread support, we suggest that it is best regarded as a series of potentially falsifiable hypotheses, and we should thus be open-minded about alternative perspectives on what is likely to have been a much more complex process of colonisation. Leaving aside for the moment recent archaeological and skeletal evidence from southern Asia, there are two principal reasons why the current consensus may be overly simplistic. The first stems from its own internal weaknesses as a model, and the second is that the context in which early human dispersals across southern Asia is discussed is changing because of the implications of new fossil and archaeological evidence from Northeast Africa, and the evidence from ancient DNA (aDNA) studies that early H. sapiens outside Africa interbred with indigenous populations. To take each in turn:

1.1. Limitations of the current model (Out of Africa)

The current model for the expansion of H. sapiens across southern Asia is based on three lines of evidence: human skeletal material; genetic studies of modern populations; and the timing of arrival of H. sapiens in Australia. It has the following internal weaknesses:

1.1.1. The limitations of the human skeletal fossil record.

As indicated above, current evidence indicates that H. sapiens appeared in east Africa by ca 195-160 ka, whereas the earliest outside Africa are those from Israel at about 100-125 ka ago. At present, there are no hominin fossils 200 ka old from southern Asia, so it is an open question as to when H. sapiens first appeared in this region (or an open question as to when H. sapiens first appeared in this region y other type of hominin, from the Arabian Peninsula prior to the Holocene. All we know at present about southwest Asian hominins between 250and 125 ka ago is that Neanderthals were present in the northern Levant during part of MIS6. ”

Homo erectus was thought to have persisted into the last glaciation on the basis of isotopic dates for the Ngandong (Solo) assemblage of 28-54 ka, but a recent study now indicates that the likely age is between 143 and 546 ka. These estimates are consistent with faunal evidence that this hominin assemblage is late Middle Pleistocene in age (Storm, 2001). The earliest unambiguous evidence in Southeast Asia, for H. Sapiens is the cranium from Niah Cave, Borneo, dated to ca 35-37 ka.

The age of this specimen is comparable to that of the partial skeleton of H. sapiens from Tianyuandong, North China, and the 32 ka old H. sapiens cranium from Yamashita-cho, Okinawa (Kaifu and Fujita, 2012). Additionally, the 42 ka assemblages from Timor, which would most likely have been colonised by using boats or rafts with sails and paddles can also be attributed to H. sapiens.

Because the earliest evidence from the highlands of New Guinea now extends back to 49 ka, modern humans must have been on the Sunda Shelf and in mainland Southeast Asia by at least 50 ka. H. sapiens was absent from Southwest, South and mainland Southeast Asia before this time. As noted in relation to the evidence for Out of Africa 1, absence of evidence is not evidence of absence.

1.1.2. Stone tools are poor indicators of population changes

In the Levant, both Neanderthals and H.sapiens used Mousterian stone-tool assemblages, and in East and North Africa, early H. sapiens used the same lithic tool-kits as their predecessors. This indicates that stone tools are a poor indicator of the species of the hominin that made them: changes in hominin type did not necessarily result in changes in lithic technology. Conversely, unchanging lithic traditions need not imply that the type of hominin that used them remained the same.

As example, in mainland and island Southeast Asia, there is no equivalent of the Upper Palaeolithic, and Mode 1 technologically-simple flake and core assemblages persist into the late Pleistocene and even in places the Holocene, even though the species of hominin that made them changed from H. erectus to H. sapiens. This is in sharp contrast to a region such as western Europe, where the shift from the Middle to Upper Palaeolithic was associated with the replacement of Neanderthals by H. sapiens leaving aside a much-contested debate over the status of so-called transitional industries that could have been made by either. Changes in lithic technology in southern Asia are likely to have been more subtle, and discernible only through more rigorous quantitative analysis than merely noting the size and frequency of a few common elements.

1.1.3. DNA studies of modern populations do not detect extinct ones, or ones older than 60 ka.

Analyses of the DNA of modern populations across southern Asia have made important contributions towards understanding the demographic history of H. sapiens outside Africa, and convincing arguments have been made that they first appeared in southern Asia ca 60ka. In China, similar studies indicate that the earliest populations of our species were present ca 50-60 ka or perhaps 27-56 ka. However, these claims are based on studies of surviving extant populations, and cannot take into account those that became extinct during the last 60 ka. Additionally, coalescence ages from mitochondrial DNA and Y chromosome studies of contemporary Eurasian populations do not extend as far back as125 ka, when H. sapiens first reached the Levant. By the same line of argument, any expansion of H. sapiens that might have occurred across southern Asia before 60 ka is currently undetectable through analyses of modern populations. This is particularly relevant to the Arabian Peninsula, which H. sapiens probably reached at the same time, or even perhaps earlier than, the Levant, as its presence there in MIS 5 is also undetectable through genetic analyses of modern populations (Cabrera et al., 2009).

Moreover, given significant environmental changes in Arabia over the course of the Late Pleistocene and Holocene, and the vulnerability of populations to even minor decreases in rainfall, we should anticipate that there were contractions and even local extinctions of hunter-gatherers, followed by population expansion and immigration from neighbouring regions. Thus the genetic make-up of contemporary populations in Arabia is likely only a partial and recent representation of the entire demographic history of the region, particularly because none has roots deeper than the late Pleistocene. Indeed, new genetic studies of Arabian populations indicate different source areas for indigenous peoples, with some groups representing much later Holocene expansions. The absence of any human skeletal evidence prior to the Holocene and the lack of ancient DNA studies continue to be major impediments to understanding the colonization and dispersal of populations of this critical region.

1.1.4. DNA studies of modern populations do not detect their residential histories

A point often overlooked in studies of the DNA of living populations is that they show their genetic, but not their residential history. As example, the Andaman Islanders are commonly regarded as the oldest surviving population in southern Asia, with a genetic history extending back 60 ka, and this in turn has been used as evidence in favour of an early coastal dispersal of H. sapiens across southern Asia. However, the fact that they have a genetic history extending back 60,000 years does not necessarily mean that they have inhabited the Andaman Islands, or have even beena coastal population, for that length of time. There is in fact noevidence that the islands were inhabited in the Pleistocene, as the earliest indications of occupation extend back only two millennia. This might indicate that either the Pleistocene (and early Holocene) occupation of the islands has yet to be detected, or that the modern islanders were until recently living on the main-land, and not even necessarily on the coast for much of their history. The same point can be made about the 30 ka Denisova individual, known only from its ancient DNA: its origin is probably African or Eurasian, but its distribution and the point at which “Denisovans” first (or how often) inhabited Siberia remain unknown.

1.1.5. The timing of arrival of humans in Australia

A conservative estimate for the arrival of humans in Australia is ca 40 ka, when it was conjoined by lowered sea levels to New Guinea and Tasmania. This estimate is consistent with the date of the Niah cranium in Borneo, and the evidence from Timor, and with a narrative that envisages modern humans arriving in Southeast Asia ca 40 ka and shortly afterwards using boats or rafts with sails and paddles to colonise islands such as Timor and the Australian landmass. Recent evidence from New Guinea indicates that humans were already in the highlands by ca 49 ka, which would imply an earlier entry into Southeast Asia that preceded the earliest dates for Niah and Timor. If one accepts the earliest dates proposed, humans reached Australia ca 60 ka. However, a recent genetic study of a100-year old lock of Aboriginal hair implies that Aboriginal Australians are descended from a dispersal of H. sapiens into SE Asia ca 62-75 ka.

These uncertainties over the timing of arrival of humans in Australia have wider but unaddressed implications for the colonisation of southern Asia. An exit from Africa ca 60 ka and an arrival in Australia at 40 ka implies a slow rate of dispersal, and low rates of population increase in the donor population. In contrast, a near simultaneous departure from Africa and arrival in Australia implies an almost epidemic rate of population growth in the donor region (presumably East African) and ensuing daughter populations, for which there is no evidence, anda dispersal event of H. sapiens as early as 62-75 ka into SE Asia implies that the date of dispersal from East Africa may have been under-estimated.

1.2. Emerging problems: donor regions, points of entry, and inter-breeding

In addition to the above problems, three recent developments pose further diffculties for the existing model.

1.2.1. How many donor regions?

Emerging evidence from Northeast Africa indicates that the Jebel Irhoud (Morocco) cranium is that of sapiens rather than a Neanderthal, and is ca 160 ka old (Smith et al., 2007). Associated Middle Stone Age a date back ca 145 ka at Ifrin Anmaar. These include perforated beads from the Grotte des Pigeons (Morocco) that are ca 82 ka old, and thus predate the 75 ka old beads from Blombos, South Africa (Balter, 2011), which were previously the oldest known examples. Similar and even earlier examples are now known from Qafzeh, Israel, and Oued Djebbana, Algeria, where they are ca 100-135 ka. All this leads to the possibility not only that H. sapiens may have entered Southwest Asia from either North or East Africa, or both, but also that there may have been several dispersal events,involving different populations. As example, the 80 ka skull from Dares-Soltan, Morocco, shows affinities with both those from JebelIrhoud and Qafzeh (seeBalter, 2011). Although these early populations of H. sapiens in Morocco may have been marginal to developments in Asia, the greening of the Sahara in the Last Interglacial and early part of the last glaciation makes it more probable that they could have dispersed eastwards towards the Levant and Arabia during and after the Last Interglacial.

1.2.2. How many entry points from Africa into Asia?

Because the Levant is often envisaged as a crossroad that links Africa to Europe and continental Asia, it is frequently depicted as the principal entry point into Asia for the earliest anatomically modern humans that left Africa via the Sinai Peninsula or theNegevDesert at the north end of the Red Sea. This view is of course reinforced by the presence of H. sapiens in the northern Levant a searly as 115-125 ka. It is not, however, the only route out of Africa, or even the most likely one. An alternative is the Bab al Mandab at the southern end of the Red Sea between theHorn of Africa and southern Yemen, and 1200 miles south of the Sinai Peninsula. These straits are only 20 km wide, and at times of low sea level, perhaps only12-15 km wide, and cannot be counted as a major sea barrier. With island hopping (especially at times of lowered sea levels), the distances across open water are reduced still further, although they were never totally eliminated. We should note also recent indications from Crete, that humans, presumably H. sapiens arrived there during the Last Interglacial aftera sea-crossing of at least 70 km.

A third route out of Africa that avoids the need to cross any open water is along the western and then eastern coasts of the RedSea; after that, entry into the interior of the Arabian peninsula,where there were large river courses would havebeen relatively easy. Alternatively, hominins might have continued along the coast and dispersed along the coast of the Indian Ocean towards Oman, following lowered spring lines. There is no reason to exclude any of these routes; indeed it is likely that all were used, although not necessarily at the same time or by the same populations. Nevertheless, the Arabian Peninsula was probably a more important faunal corridor than the Levant. Besides being geographically closer to East Africa than the Levant, the faunaof the Arabian Peninsula has few endemic mammalian species.

Irrespective of dispersal route(s), topographic and favourable environmental settings in Arabia indicate that populations could have penetrated and expanded in these zones. If H. sapiens was present in the Levant for a period spanning 60 ka (from 130 to70 ka), it seems inconceivable that similar populations were not also present in the well-watered and vegetated zones of Arabia at the same time.

1.2.3. The implications of ancient DNA (aDNA) studies

One of the most important revelations of the recent analysis of the aDNA of Neanderthal fossils is that they interbred with non-African modern humans before the divergence of modern Europeans, East Asians and Papuans, who thereby acquired ca 4-8% of their genes from Neanderthals. This most likely occurred through contact in Southwest Asia between 100 and125 ka (when H. sapiens is first recorded outside Africa) and 65 ka (when they became locally extinctin the Levant). Likewise, analysis of the aDNA of the Denisova individual also indicates inter-breeding with H. sapiens, hence the presence of some “Denisovan” genes in modern-day inhabitants in Melanesia, and the possibility that Denisovans once inhabited Southeast Asia. Both discoveries imply complex population histories in southern Asia, and the likelihood of hybridisation between Neanderthals, Denisovans and H. sapiens. By extension of the same reasoning, inter-breeding may also have occurred in East Asia between incoming populations of H. sapiens, and resident populations of H. erectus, whose genetic make-up remains unknown. This point has relevance to some of the specimens from South China, as indicated below.

1.3. Archaeological and human skeletal evidence

Recent evidence from Arabia, India, mainland and island Southeast Asia imply that the population history of southern Asia may have been more complex than envisaged under the current view, and that our species may have left Africa before 40-60 ka.

1.3.1. Arabia

Recent analysis indicates that some Arabian Middle Palaeolithic assemblages show affinities with both the Levant and EastAfrica.At Shi’bat Dihya in Southwest Yemen (opposite the Horn of Africa), assemblages with Levallois cores and tool forms date to ca 55 ka, and are said to be reminiscent of both the Levantine Mousterian and East African MSA, but are also idiosyncratic. As mtDNA analysis indicates that H. sapiens was in southern Asia by 60 ka, it seems reasonablycertain that any Middle Palaeolithic assemblage younger than that from southern Asia (including Arabia) would also have been made  sapiens, rather than Neanderthals. In contrast with Shi’bat Dihya, a lithic assemblage dated to ca 125 ka from Jebel Faya in Oman is claimed to have affinities with Northeast Africa, suggesting a population in flux of H. sapiens from Northeast Africa during the Last Interglacial. (Opinion remains sharply divided on this claim.

Nevertheless, given that so little is known about the early prehistory of the million-square miles of the Arabian Peninsula, we should not dismiss out of hand the possibility that modern humans were resident there before 60 ka, particularly because they were present scores of millennia before that date in the adjacent regionsof Northeast Africa and Levant.

1.3.2. India

The crucial link between the western and eastern parts of southern Asia is India, which has traditionally played little part indiscussions of modern human origins in Asia though this has been changing in recent years. In northern Pakistan, a small blade assemblage was excavated at site 55, Riwat,and dated to ca 45 ka (Dennell et al., 1992) but most blade- and micro-blade dominated assemblages appear after ca 35 ka in peninsular India (Clarkson et al., 2009), given the caveat that there are more reliable dates available for sites such as Hayonim, Israel, than for the whole of India! Moreover, most radiocarbon and luminescence dates from India are single determinations, oftenmade many years ago and long before laboratory and field tech-niques improved to their current standards.Recent research in peninsular India has produced a great deal of new evidence that substantially revises previous knowledge of the Indian Palaeolithic sequence. Much of this new evidence has comeabout through intensive study of the Younger Toba Tuff (YTT) that resulted from the volcanic eruption of Toba in Sumatra ca 74 ka. As this was by far the largest eruption of the Pleistocene, a great deal of ash fell across India, and thus it providesa clear marker horizon  Most significantly, recent analyses of core reduction techniques show that these most resemble those of the African Middle Stone Age made by H. sapiens. Such assemblages can be distinguished from those made in the Levant by Neanderthals. This strongly implies that these Indian Middle Palaeolithic assemblages were also made by H. sapiens, and this suggestion further implies that that it entered India earlier than the dates estimated from mtDNA data. Even more exciting is the recent discovery that stone assemblages from above and below the Toba ash are extremely similar technologically to those from early modern humans sites in sub-Saharan Africa, southeast Asia and Australia, suggesting modern humans may have entered India before the Toba eruption as part an early eastward dispersal from Africa.

Although only a single find, the recovery of a tanged point under the Toba ash, also suggests an African affiliation.The earliest micro-blade industries excavated fromrock sheltersin India and Sri Lanka firmly date to ca 35 ka. Although it has been suggested that the tradition of making micro-blade tool forms and small crescentic tools was derived from the considerably older Howieson’s Poort assemblages from southernAfrica, dated to ca 65-60 ka, the most parsimonious explanation is that these are local innovations that may denote changes in hunting or plant processing activities, rather than the revival of a technique first initiated several thousands of kilometres distant and scores of millennia earlier. In support of this suggestion, there is currently no strong or convincing evidence for “Upper Palaeolithic” industries or symbolic assemblages, or microliths in early MIS 3 > 40 ka or MIS 4 in either the Zagros (which is fairly well-documented for the last glaciation) or Arabia. If the micro-blade industries of the Indian subcontinent are local in origin, their wider significance is that H. sapiens did not arrive in southern Asia with a behavioural “package” of useful items and techniques (e.g. microliths, bone points, fishing equipment), but developed these as and when they were locally useful, in the same way that has been proposed for Australia.

1.3.3. Southeast Asia

There are two slight hints that H. sapiens might have been present in island SE Asia and the Philippines before 50 ka. The first is a premolar from Punung, Java, dated on faunal grounds to 80-126 ka. Although identified as H. sapiens, Bacon et al. were not convinced that this constitutes convincing proof for the presence of modern humans in Java at such an early date. The second and stronger piece of evidence is a hominin 3rd metatarsal dated to ca 67 ka from Callao Cave, Luzon,the northern Philippines.

1.3.4. South China

As most readers will be aware, most Chinese palaeoanthropologists maintain that H. erectus evolved locally in East Asia into H. sapiens, with the early forms of our species best described as “archaic” H. sapiens” This leads to an immediate confusion in that the same term is sometimes given to the earliest examples of H. sapiens in Africa. This point aside, there is still considerable uncertainty over the dating and context of these “archaic” specimens prior to 30 ka. The main problems in evaluating the Chinese evidence (besides the obvious one over language for non-Chinese readers) are that excavations are not always fully published, some claims are based solely on isolated teeth, and there are doubts over the context and dating of finds, particularly in these complex cave environment. The hominin specimens from several cave sites are particularly problematic. At Liujiang, the breccias containing a cranium and post-cranial bones attributed to H. sapiens have been dated to a minimum age of 61-68 ka, but a probable age of 111-139 ka; they also suggest that the maximum age could be 153 ka.

The layer containing H. sapiens teeth at Bailongdong cave was previously dated to 30-14 ka, but a stalagmite contemporary with the same layer has recently been dated to 160 ka. At Tubo Cave, stalagmites above and below a layer with human teeth were dated to between 94 and 220 ka, and dating by 230-234 and 227-230 of two non-human teeth in the same layer to 85-139 ka. At Laibin, human teeth have been dated to a minimum of 39-44 ka, and a maximum of 112 ka. More recently, a mandible from Zhirndong, South China, has been attributed to H. sapiens and dated by U-series dating to ca 125 ka. Interestingly, because the mandible shows a mixture of characteristics associated with both H. sapiens and H. erectus, it is interpreted as showing inter-breeding between the two species at an early date, rather than purely in situ evolution (An alternative interpretation is that it indicates a gracile late H. erectus.  

The Chinese evidence outlined above is clearly at odds with the conventional scenario for the colonisation of Southeast Asia by H. sapiens, with Niah as the earliest indication at ca 43 ka, and western scholars have tended to ignore these early dates, or reject them because of uncertainties over age and context. However, itmight be advisable to consider alternative interpretations. As Kaifuand Fujita (2012: 8) suggests “a compatible scenario is that there was an earlier MIS 5 eastward dispersal of early modern humans but their genes did not significantly contribute to the genetic make-up of contemporary Asia people.” Because one implication of the aDNA evidence for both Neanderthals and “Denisovans” is that they interbred with H. sapiens, there is no a priori reason why H. erectus might not also have interbred with H. sapiens (and other extinctspecies), and produced off-spring with hybrid features (as suggested for Zhirendong). Additionally, if, Denisovans inhabited Southeast Asia and interbred with H. sapiens, the population history of this region is likely much more complex than currently envisaged.”


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s