Reader Comments

Post a new comment on this article


Posted by manueldr on 14 Oct 2012 at 19:01 GMT


Manuel Domínguez-Rodrigo1,2, Travis Rayne Pickering3,4,5, Fernando Diez-Martín6,
Audax Mabulla7, Charles Musiba8, Gonzalo Trancho9, Enrique Baquedano1,10, Henry T. Bunn3

1 IDEA (Instituto de Evolución en África), Museo de los Orígenes, Plaza de San Andrés 2, 28005 Madrid, Spain
2Department of Prehistory, Complutense University, Prof. Aranguren s/n, 28040 Madrid, Spain
3 Department of Anthropology, University of Wisconsin-Madison, 1180 Observatory Drive, Madison, Wisconsin, 53706, USA
4Institute for Human Evolution, University of the Witwatersrand, WITS 2050, Johannesburg, South Africa
5Plio-Pleistocene Palaeontology Section, Department of Vertebrates, Ditsong National Museum of Natural History (Transvaal Museum), Pretoria, 0002, South Africa
6Department of Prehistory and Archaeology, University of Valladolid, Plaza del Campus s/n, 47011 Valladolid, Spain
7Archaeology Unit, University of Dar es Salaam, Dar es Salaam, P.O. Box 35050 Tanzania
8Department of Anthropology, University of Colorado Denver, 1201 5th Avenue, Suite 270, Denver, CO 80217, USA
9Department of Anthropology, Complutense University, Prof. Aranguren s/n, 28040 Madrid, Spain
10Museo Arqueológico Regional, Plaza de las Bernardas s/n, 28801 Alcalá de Henares, Madrid, Spain

In their response to our paper [1], Crandall and Martin [2] argue that the porotic hyperostosis (PH) observed on OH 81, a fragment of subadult hominin parietal, might have resulted from hemolytic anemia (or other factors), and not from megaloblastic anemia. If correct, this diagnosis would diminish our emphasis on the importance of meat eating in the diets of early hominins as a relevant implication of our contradictory diagnosis that OH 81’s PH was probably caused by megaloblastic anemia. We welcome the opportunity to address the critiques of Crandall and Martin.

It is commonly assumed that hemolytic anemias , in general, produce PH. However, of the wide variety of hemolytic anemias, only thalassemia major and sickle anemia have been observed clinically to produce PH [3,4]. Both of these types of hemolytic anemias are found most commonly in areas that are malaria endemic (the presence of these conditions in malarial habitats suggests that they confer a selective advantage to carriers of the mutations that cause them). Given that it produces unique modifications of the splanchnocranium, thalassemia is relatively easy to diagnose by observing the bones of pathological individuals whose complete or nearly complete skeletons are preserved. In addition to the thinning of long bones (accompanied by extensive proliferation of subperiostial bone tissue along the shafts), the main osteological manifestations of thalassemia result from marrow hypertrophy producing enlarged maxillary and mid-facial bones with what is typically described (pejoratively) as “mongoloid” or “rodent” faces in subadults [3,5,6]. Thalassemia also commonly leads to enlargement of nutrient foraminae, obliteration of frontal, maxillary and sphenoidal sinuses, widening of ribs, formation of costal osteomas and early synosotosis of cranial sutures [6,7,8]. These osseous pathologies are exacerbated with increasing age of the individual.

Understanding these and other relevant issues (discussed below), we argued that it was extremely unlikely that hemolytic anemia was responsible for the PH documented on OH 81. For one thing, the overwhelming majority of archaeological evidence of PH is from geographic areas for which no direct evidence exists of endemic malaria. For instance, there are >70% occurrences of PH in some populations of pre-Columbian Anasazis [9,10], in which its presence is most commonly observed on the remains of individuals younger than two years of age when they died [10]. PH is also widely reported in the Old World archaeological record, in which its high frequency compared to the low probability of occurrence of inherited anemia today forced Angel (who promoted the idea that PH was caused by thalassemia [11]) to admit that many of the cases of PH in the Old and the New World archaeological records could have resulted from acquired, rather than inherited, anemia [12,13]. It has been argued that genetic anemias did not occur in the New World before Amerindian contact with Europeans [8]. Taken together, the low probable occurrence of thalassemia or sickle cell anemia in contrast to the widespread documentation of PH, especially in areas where neither endemic malaria occurs nor are hemolytic anemias known, supports the hypothesis that megaloblastic anemias can be responsible for a large portion of the documented PH. That said, we still acknowledged that this assertion does not falsify the hypothesis that genetic anemia caused the PH observed on OH 81; as we stated originally, “we cannot discard other causes for this pathology.” [1:4]. However, we are able to enumerate several reasons why causes other than megaloblastic anemia provide comparatively weaker hypotheses for the origin of OH 81’s PH. With regard to the weakness of specifically genetic anemia hypotheses, we note the following points:

1. Very low frequencies (5-20%) of PH are documented in cases of thalassemia [14,15]. These observations underscore the impressive phenotypic variability of individuals affected by thalassemia , as well as the difficulty of relating genotype to phenotype because of the complex interaction of the different allele variants of this disease and their environmental contexts [8]. In several clinical cases of the same type of thalassemia, the cranial bones of some individuals were pathological modified, whereas those of others were not; those individuals with cranial modifications were also malnourished [16]. How diet (and lack of crucial nutrients) influenced the observed bony modifications is unknown, because most of the clinical cases were not controlled for diet.

2. Point (1) demonstrates that clinicians often document correlations between PH and thalassemia, in the same way as other equally relevant studies document correlations between PH and iron deficiencies (see a brief review in [ref. 14])—despite knowing that low levels of iron inhibit marrow hypertrophy [14]. Correlation is not causation. Most of the classic clinical studies linking PH to thalassemia date to around the time of the Great Depression, when malnutrition was widespread. The contribution of diet to the development of the genetic anemia is thus understudied and poorly understood. However, recent studies suggest that malnutrition is a common factor affecting patients with thalassemia [17, 18]. A study of American and Canadian patients with thalassemia demonstrated that >30% of the affected individuals were deficient in vitamins A, D, E, K, as well as in folate, calcium and magnesium [17]. Folate (vitamin B9) deficiency is one cause of PH [14]. In contrast, results showed that consumption of vitamin B12 in several of these patients was adequate. These observations potentially explain the low incidence of PH in modern populations affected by thalassemia. Collectively, these data show that even today, with the luxury of direct observations, it is still not possible to discern the relative and potentially combined roles that intrinsic processes of genetic anemia and diet play in creating PH and other pathological skeletal modifications.

3. The ontogenetic onset of skeletal modifications induced by hemolytic anemias is an important variable to consider in evaluating the case of OH 81’s PH. In modern humans, the most common osteological manifestations of the first symptoms of the disease are commonly documented after two years of age [6] (although some pathological modifications of the long bones have been seen very rarely in affected individuals <1 year old [8]), and most of these earliest manifestations occur in the postcranial skeleton [8]. These first appearing skeletal pathologies are also very subtle; in most most clinical cases of genetic anemia, it is only children of greater than three years of age who show even just moderate cranial deformation [16,19,20]. Indeed, some researchers argue that the classic skeletal pathologies associated with hematopoietic diseases are very rarely observed in individuals younger than nine years old [21]. As to the order of the pathological modification of the skull, clinicians have documented genetic anemically induced changes in the maxilla and mandible at about two years of age [22]. Modification of the mid-face and the frontal bone follows after changes to the maxilla and mandible [16,19]. Specifically, skull modification follows a progression from the maxilla and mandible to the frontal, temporal, and finally, parietal and occipital bones. In classic radiological studies of genetic anemia-related pathologies, no changes were generally documented on the subjects’ parietals prior to four years of age, and most commonly appeared only substantially later in the development of individuals [16,19, 20] This documented sequence of pathological advancement explains why there is an abundance of thalassemia-induced pathologies on the human postcranial remains from the prehistoric site of Khok Phanom Di (Thailand), but none on the parietals of individuals <4 years old when they died (indeed, only 4 of 27 <4-year-olds from the site show any pathological modification of the neurocranium, aside from the frontal bone) [15].

In contrast, the pathologically modified OH 81 parietal comes from an individual much younger than 4 years-old when he/she died; we estimate, based on comparisons to modern humans, he/she died at or less than two year of age [1]. And given that early Pleistocene hominins probably matured more rapidly than do modern human children (see arguments and references in [1]), the OH 81 individual might have been only slightly older than one year old when he/she died. If so, this means an even greater discrepancy between the child’s age at death and clinically and archaeologically based estimates for the earliest age at which parietal modification induced by thalassemia occurs. In contrast, in the prehistoric New World, where anemias were acquired and not inherited, PH is not uncommonly observed on the crania of prehistoric Amerindians <2 two years old when they died [9,10], including on their parietal, as well as frontal bones. The difference in the estimated ontogenetic onset of the pathological modification of the OH 81 parietal and that reported for thalassemia-induced modification of human parietals weakens the case that OH 81’s PH was caused by hemolytic rather than by megaloblastic anemia.

4. PH induced by thalassemia most commonly appears and expands from the ossification center of the parietal eminence [23]. The same pattern of the spread of PH is also documented in cases in which it is induced by sickle cell anemia [21] (although sickle anemia-induced PH never shows the dramatic expression of hair-on-end phenomenon typical of PH induced by thalassemia). PH on OH 81 is located around the rim of the posterior sutures (lambdoid, parieto-mastoid). Although the parietal eminence is not preserved, the portion of parietal most distant from the sutures shows no porosity, suggestive of a PH-free parietal eminence and a PH pattern different from that typically induced by thalassemia and more similar to those of other similarly-aged modern human individuals [24] who may have died of acquired anemia.

5. In some clinically investigated cases of PH resulting from thalassemia, both the outer and inner tables of the neurocranium are thinned pathologically [20]. In contrast, the inner table of OH 81 is normal and unmodified.

Crandall and Martin argue a hypothesis of hemolytic anemia-induced PH is preferred for OH 81 because the fossil was discovered at a site in a region that is today malaria endemic. As Lewis [6] stressed, a diagnosis of genetic anemia in archaeological contexts that is based on the geography of the case is unwarranted for various reasons [25]. Contrary to Crandall and Martin’s argumentation [2], there is no conclusive evidence that “malaria has long been a major issue in the region where OH 81 was uncovered.” Indeed, there is no evidence at all that malaria that was/is harmful to hominins appeared anywhere in the world prior to 10, 000 years ago [26]. Plasomodium falciparum evolved from P. reichenowi by a single host transfer, but contrary to Crandall and Martin’s assertion, there is no reason to believe that the latter was a health threat to early hominins, since the inactivation of the CMAH gene in humans makes us resistant to P. reichenowi and it is not known when this mutation occurred [26]. In fact, the scarcity of polymorphisms in P. falciparum probably indicates a recent expansion of this species, which likely occurred only a few thousands of years ago rather than millions of years ago [26]. In sum, there are no available data that support the hypothesis that malaria that was harmful to hominins existed in the early Pleistocene, when OH 81 lived and died. In this absence of such data, there is no empirical support for hypotheses that hemolytic anemias were part of the adaptive repertoire of early Pleistocene hominins. A lack of fitness-impacting malaria and associated anemias might explain, in part, the generally rare occurrence of PH on hominin crania throughout most of the Pleistocene record.
And, even in the extreme unlikelihood that P. falciparum did appear in the Pleistocene, “the threat malaria posed to early hominids would have been minor, since hominid populations were sparse. Because the major blood source for the Anopheles was nonhuman primates, the mosquito may have adapted to grassy woodlands but still not have been a threat to early hominids. It was only with the advent of slash-and-burn (swidden) agriculture, which creates a breeding ground for the mosquito near human settlements, that malaria became a constant threat. The increase in population size and density as well as the presence of a sedentary and dependable pool of human hosts would have allowed the mosquitoes to shift from nonhuman primates to hominids as a food source.” [27:31]. This would also explain why sickle cell anemia may have been a recent response to malaria, according to the gene frequency of this disease [28].

Crandall and Martin argue further that we overestimate the importance of meat in early hominin diets in our underestimation of the relevant roles played by other factors that may trigger ameloblastic anemia. Although Crandall and Martin assert that “[i]nadequate dietary uptake of vitamin B12 is only one of several avenues by which an individual may exhibit a B12-deficiency,” they fail to demonstrate, in turn, what other processes are able to create a deficit in vitamin B12 that is prolonged enough to result in pathological bone modification. Crandall and Martin’s list of alternative factors (padded in places by separate consideration of synonymous conditions) includes, prominently, achlorhydria (very frequently caused by Helicobacter pylori infection) or homocystinuria. Achlorhydria is a disease producing low stomach acid, which can lead to nutritional deficiencies through malabsorption of basic electrolytes and vitamins. It is very uncommon in infants <2 years old and its frequency increases with age, with its greatest prevalence in people >45 years of age [29,30]. In children, it is mostly documented in association with pernicious anemia or iron-deficient anemia [29]. Homocystinuria is an uncommon homozygous recessive disorder affecting metabolism of the amino acid methionine. Homocystinuria is frequently linked to megaloblastic anemia and can occur without affecting the vitamin B12 concentrations in the body [31]. This all said, even Crandall and Martin acknowledge that achlorhydria and homocystinuria are rare diseases, and therefore, very unlikely candidates to explain PH because of their low probability of occurrence. Moreover, we emphasize neither of these diseases has been shown clinically to produce PH or even be a relevant contributor to the formation of PH. One of the reasons may be because both diseases are easily treated with supplements of vitamins B12 and B6 respectively [29].

Crandall and Martin’s mistake is in confounding a number of disorders that inhibit vitamin B12 digestion (worsening previous conditions or creating new ones) with clinical studies of metabolic processes that produce PH. Crandall and Martin’s assertion that PH is created by multifactorial processes, which include food treatment (cooking), hyperthyroidism, pellagra, cancer and parasites, is not only inaccurate (i.e., no clinical evidence exist of the link between any of these diseases and PH as defined in our paper), but it also dismisses microscopic evidence that each of them is structurally different and can thus be accurately identified and distinguished [24]. Further, logic positivists would label Crandall and Martin’s assertions as “doubtful propositions,” since they are based on a priori synthetic statements [32]. In other words, arguing that PH could be caused by all these processes has similar epistemic value to ambiguous propositions (a) arguing that these processes intervene in anemias resulting in PH; but (b) it is unknown the role that they play in anemic conditions and the long-term effects that they have on skeletal modification, given the absence of clinical evidence thereof.

In our paper [1], we cited thoroughly the recent debate over the mutiple factors cited by Crandall and Martin that may o may not contribute to human anemias, and emphasized that none, alone, is probably able to trigger and sustain the conditions over long periods. For instance, parasites contribute to degradation of human health only when the anemic process is already active. Parasites alone have not been documented to create an anemic condition which will last enough to produce pathological bone modifications (see also [ref. 14]). Giardia and other parasites can interfere with the metabolism of vitamin B12 by preventing its absorption or depleting it, but “in people with normal vitamin B12 intake, such intestinal parasite-associated losses may not be large enough to stimulate the large-scale marrow hypertrophy that causes porotic hyperostosis” [14: 115]. This is probably one of the reasons why the modern Hadza foragers of northern Tanzania, who both regularly eat meat and are frequently infected with Giardia, do not show signs of chronic anemic conditions, malnutrition or PH. It seems that as long as regular intake of vitamin B12 is maintained PH will not develop, despite any anemic condition that might be initiated a priori by parasites (similar relationships probably exist between parasite loads and others of the diseases mentioned above regarding their limited impact on health deterioration provided vitamins B are regularly taken). So, while we know that health problems related to weaning and anemia are multifactorial, we also know that not all factors contribute equally to the occurrence of the profound nutritional deficiencies that underlie PH. By referring diffusely to all factors contributing to anemia as multifactorial, Crandall and Martin deconstruct the hierarchy of factors that cause anemia—and, in doing so, they blur the distinction of those variables that contribute to sustaining or exacerbating anemia from those that have been demonstrated clinically to directly cause skeletal pathologies.

Beyond their doubt about the decisiveness of vitamin B12 deficiency in the formation of PH, Crandall and Martin discuss insects as an alternative dietary source of vitamin B12 for early hominins. This assertion, that insects are a good source of vitamin B12, is also unsupported empirically. Animals obtain vitamin B12 directly from bacteria. Certain insects, including some species of termites, contain vitamin B12 that is produced by their gut bacteria, but most other species of insects analyzed contain low levels or no vitamin B12 [33]. This could explain why hominins invested costly into hunting mammals instead of just foraging for insects.

Finally, Crandall and Martin argue that “abundant evidence of PH has been documented among other hunter-gatherers,” and that from this assertion it follows that we should necessarily question the hypothesis that a deficit of meat induced the PH on OH 81. As with their previous assertions discussed above, this one is also unsupported. The “abundant evidence of PH” to which Crandall and Martin refer is observed on the remains of a few (n=6 [34]) prehistoric “transitional” foragers (late Woodland sedentary foragers from 900-1250 AD, in transition to agriculture) [34], the dietary make-up of whom, in contrast to that of modern Hadza (with meat contributing >50% of their annual energy intake [35]) is unknown. The use of these Late Woodland foragers is a poor proxy for representing the diets of most hunter-gatherers, especially those of immediate-return foragers (such as the Hadza), more adequate as analogues for early Pleistocene hominins. As a matter of fact, PH is widely unreported among hunter-gatherers and when it occurs, it is mainly documented in sedentary foragers, such as late Mesolithic hunter-gatherers. Several of the Mesolithic transitional foragers became increasingly more reliant on marine resources, with less contribution to their diets from terrestrial mammal food resources , especially as compared to the Hadza [36,37,38]. That these prehistoric transitional foragers settled for prolonged periods near marshes and in other near-water locations, and during a time in which there is strong evidence that P. falciparium malaria was now well-established, suggests the relatively greater spread of PH-related diseases in these populations. Even so, PH still occurs in much lower frequencies on skeletons in this earlier transitional period than is observed in more recent skeletal populations from the sedentary, agrarian Neolithic. And, we re-emphasize that relative commonness of PH in the Mesolithic and Neolithic is in stark contrast to its virtual absence during most of the Paleolithic. Whether the elevated frequencies of PH in the Mesolithic and Neolithic is related to megaloblastic anemic conditions [e.g., meat-eating was less important to semi-sedentary hunters and farmers] or to hemolytic anemias [e.g., human-deleterious malaria was established by the Holocene] remains to be discerned through detailed pathological studies of those populations, which are focused on identifying the diagnostic bone pathology manifestations of each type of anemia. Given that long bones of some Mesolithic children seem to show modifications typical of thalassemia, hemolytic anemia could be responsible for the PH documented during this period [39].)

In closing, we note that sound scientific interpretations require empirically-grounded referents. Crandall and Martin claim that “absence of evidence in paleopathology is not sufficient evidence of absence” [2]; but, in science “absence of evidence” means that certain analytical propositions cannot be made (e.g., process x plays a causal role in effect z) since “meaningful” propositions can only be scientifically sustained if empirically supported. Currently, clinical studies only show direct links between some hemolytic and megaloblastic anemias and the formation of PH [14]. Given their common occurrence of both types of anemia, especially of the latter, they are the only empirically proven candidates to explain the formation of PH unless other agents/processes are eventually demonstrated clinically to induce PH. Until such time, we continue to support our original hypothesis, that the PH on OH 81 was most likely induced by megaloblastic anemia, and that this, in turn, suggests that hominin digestive physiology was adapted to a diet of which meat was a major component. Here we have demonstrated that this hypothesis, although potentially falsifiable, is currently the most heuristically supported alternative as compared to more speculative scenarios proffered by our colleagues.


[1] Domínguez-Rodrigo M, Pickering TR, Diez-Martin, F, Mabulla A, Musiba, C, et al. (2012). Earliest porotic hyperostosis on a 1.5-million-year-old hominin, Olduvai Gorge, Tanzania. PLOS ONE 7(10):e46414. Doi:10.1371/journal.pone.0046414
[2] Crandall, JJ, Martin DL (2012). On porotic hyperostosis and the interpretation of hominin diets. Comment on Domínguez-Rodrigo M, Pickering TR, Diez-Martin, F, Mabulla A, Musiba, et al. 2012. Earliest porotic hyperostosis on a 1.5-million-year-old hominin, Olduvai Gorge, Tanzania. PLOS ONE 7(10):e46414. Doi:10.1371/journal.pone.0046414
[3]Caffey, J (1937). The skeletal changes in the chronic hemolytic anemia. Am. J. Roentgenol Radium Ther. Nucl. Med. 37: 293-324.
[4]Sebes JI, Diggs, LW (1979). Radiographic changes of the skull in sickle cell anemia. Am. J. Roentgenol. 132: 373-377.
[5]Moseley,JE (1965). The paleopathologic riddle of “symmetrical osteoporosis”. Am. J. Roentgenol. 95: 135-142.
[6]Lewis, ME (2010). Thalassemia: its diagnosis and interpretation in past skeletal populations. Int. J. Osteoarchaeol. DOI:10.1002/oa.1229.
[7]Resnick D (1995). Diagnosis of Bone and Joint Disorders, W.B. Saunders Company: Philadelphia.
[8] Lagia A, Eliopoulos C, Manolis S (2006). Thalassemia: Macroscopic and radiological study of a case. Int. J. Osteoarchaeol. 17: 269–285.
[9] Martin, DL, Goodman, AH, Armelagos, GJ, Magennis, Al (1991). Black Mesa Anasazi health: reconstructing life from patterns of death and disease. Southern Illinois University, Center for Archaeological Investigations, Occasional Paper 14.
[10] Stodder, ALW (1993). Bioarchaeological investigations of protohistoric Pueblo health and demography. In (Larsen, CS, Milner, G, eds) In the wake of contact: Biological responses to Conquest, pp. 97-107.
[11] Angel JL (1966). Porotic hyperostosis, anemias, malarias, and marshes in the Prehistoric eastern Mediterranean. Science 153: 760–762.
[12] Angel JL (1984). Health as a crucial factor in the changes from hunting to developed farming in the eastern Mediterranean. In (Cohen MN, Armelagos GJ, eds). Paleopathology at the Origins of Agriculture, Academic Press: New York; 51–73.
[13] Angel JL, Kelley JO, Parrington M, Pinter S (1987). Life stresses of the free black community as represented by the First African Baptist Church, Philadelphia, 1823–1841. Am. J. Phys. Anthro. 74: 213–229.
[14] Walker P, Bathurst R, Richman R, Gjerdrum T, Andrushko, V (2009). The causes of porotic hyperostosis and cribra orbitalia: A reappraisal of the iron-deficiency-anaemia hypothesis. Am. J. Phys. Anthropol. 139:109–125.
[15] Tayles N. (1996). Anemia, genetic diseases, and malaria in prehistoric mainland Southeast Asia. Am. J. Phys. Anthropol. 101: 11–27.
[16] Caffey, J. (1951). Cooley´s erythroblastic anemia. Some skeletal findings in adolescents and young adults. Am. J. Roentgenol Radium Ther. Nucl. Med. 65: 547-560.
[17] Fung, EB, Xu, Y., Trachtenberg, F, Odame, J., Kwiatkowski JL, et al. (2012). Inadequate dietary intake in patients with thalassemia. J. Acad. Nutr. Diet 112: 980-990.
[18] Tienboon, P, Sanguansermsri, T, Fuchs GJ (1996). Malnutrition and growth abnormalities in children with beta-thalassemia major. Southeast Asian J Trop. Med. Public Health 27: 356-361.
[19] Caffey J (1957). Cooley’s anaemia: A review of the roentgenographic findings in the skeleton. Am. J. Roentgen. 78: 381–391.
[20] Mandeville, FB. (1930). Roengent-ray findings in erythroblastic anemia. Radiology 15: 72-84.
[21] Hershkovitz I, Rothschild BM, Latimer B, Dutour O, Leonetti G, et al. (1997). Recognition of sickle cell anemia in skeletal remains of children. Am. J. Phys. Anthropol. 104: 213–226.
[22] Poynton, HG, Davey, KW (1968). Thalassemia: changes visible in radiographs used in dentistry. Oral Surgery, Oral Medicine and Oral Pathology 25: 564-576.
[23] Herrerín, J, Baxarias, J, García-Guixé, E, Nuñez, M, Dinarés, R (2010). Betatalasemia en un niño de una necropolis del Imperio Nuevo (Luxor, Egipto). Estudio macroscópico y radiológico. Imagen Diagnóstica 1: 61-66.
[24] Schultz M (2003) in Identification of Pathological Conditions in Human Skeletal Remains, ed Ortner DJ, (Academic Press, London), pp73-106.
[25] Keenleyside A, Panayotova K (2006). Cribra orbitalia and porotic hyperostosis in a Greek colonial population (5th to 3rd centuries BC) from the Black Sea. Int. J. Osteoarchaeol. 16: 373–384.
[26] Rich SM, Leendertz FH, Xu G, LeBreton M, Djoko CF, et al. (2009) The origin of malignant malaria. Proc Natl Acad Sci USA 106(35): 14902-14907.
[27] Armelagos, GJ, Harper, KN (2010). Disease globalization in the third epidemiological transition. In (Gust, G., ed.) Globalization, Health and the Environment, Altamira Press, New York, pp. 27-53.
[28] Livingston, FB (1983). The Malaria Hypothesis. In Distribution and Evolution of Hemoglobin and Globin Loci, 4. J. Bowman, ed. Pp. 15-44. New York: Elsevier.
[29] Hamksley, JC, Lightwood, R., Bailay, UM (1934). Iron-deficient anemia in children: its association with gastrointestinal disease, achlorhydria and haemorrage. Archives of Disease in Chilhood 9,:359-372.
[30] Svendsen, JH, Dahl, C., Svendsen LB, Christensen, PM (1986). Gastric cancer in achlorhydria patients: a long term follow up study. Scand. J. Gastroenterol., 21: 16-20.
[31] Rosenblatt, DS, Thomas, IT, Watkins, D., Cooper, BA, Erbe, RW (1987). Vitamin B12 responsive homocystinuria and megaloblastic anemia; heterogeneity in methylcobalamin deficiency. Am. J. Med. Genet. 26: 377-383.
[32] Carnap, R. (1937). The Logical Syntax of Language. Kegan Paul.
[33] Wakayama, EJ, Dillwith, JW, Howard, RW, Blomquist, GJ (1984). Vitamin B12 levels in selected insects. Insect Biochem. 14: 175-179.
[34] Armelagos GJ (2010) Health and Disease in Prehistoric Populations in Transition. In (Brown PJ, Barrett R, eds.) Understanding and applying medical anthropology. Boston: McGraw Hill. pp. 50-60.
[35] Marlowe F (2010). The Hadza: Hunter-gatherers of Tanzania. Origins of
Human Behavior and Culture. Los Angeles: University of California Press. 325
[36] Lubell, D., Jackes, M., Schwarcz, J., Knyf, M., Meiklejohn, C (1994). The Mesolithic–Neolithic transition in Portugal: Isotopic and dental evidence of diet. J. Archaeol. Sci. 21:201–16.
[37] Schulting, RJ (2011). Mesolithic-Neolithic transitions: an isotopic tour through Europe. In (R. Pinhasi, J. Stock, eds.), The Bioarchaeology of the Transition to Agriculture, 17-41. New York: Wiley-Liss.
[38] Anders Fischer, Jesper Olsen, Mike Richards, Jan Heinemeier, Árny E. et al. (2007). Coast-inland mobility and diet in the Danish Mesolithic and Neolithic evidence from stable isotope values of humans and dogs. J. Arch. Sci. 34: 2125-2150.
[39] Angel, LJ (1984). Health as a crucial factor in the changes from hunting to developed farming in the eastern Mediterranean. In (Cohen, MN, Armelagos, GJ, eds.) Paleopathology at the Origins of Agriculture (proceedings of a conference held in 1982). Orlando: Academic Press., pp. 51-73.

No competing interests declared.