Imagining the ‘Biochemical Race’: Sero-Anthropology and Concepts of Racial Purity in Portugal (1900s–1950s)

This article traces the reception of blood group research in Portuguese physical anthropology in the first half of the twentieth century and analyses its presence as ‘sero-anthropology’ within the context of the disciplinary and political dynamics of colonial and metropolitan Portugal and against the background of international developments on blood group research. It argues that Portugal, hitherto largely understudied in relation to the broader international picture, was in tune with these developments. The article argues further that Portuguese physical anthropology, particularly research based at the University of Porto, was deeply ingrained with the fear of ‘contamination’ of the ‘race’ by the colonialized ‘other’ and sought to differentiate the Portuguese from the peoples of Africa and the East where Portugal possessed colonies, while it also sought to place the Portuguese within the scale of racial hierarchies of ‘whites’ in Europe. The article elaborates on a number of central and marginal figures within Portuguese anthropology to illustrate these claims and argues that the discipline was in tune with wider European developments in the field but with specific colonialist and racialist inflections, some of which are still felt in Portuguese culture today.


Introduction
In this article we examine conceptions of racial purity in the works of early twentiethcentury Portuguese physical anthropologists who practised the science of serology or so-called sero-anthropology, that is, the set of methods and techniques aimed at collecting and classifying human blood groups as markers of 'race' in distinct human populations. 1 In the first half of the twentieth century, serology became increasingly popular among physical anthropologists in Europe and the United States and even beyond. Strong racial nationalism stimulated this growing international interest in serology, and Portugal was no exception. 2 Portuguese anthropologists similarly sought to isolate the supposed purity of the national 'race' as a 'biochemical' entity embodied in the physicality of human blood and expressed in statistical indexes; at the same time, and through the same methods, they sought to isolate the purity of other, supposedly inferior, nonwhite 'colonial races'. Hence, Portuguese researchers hoped that the statistical analysis of human blood groups would help discern the true biochemical nature of pure racial types, both among 'white Portuguese' at home and 'black' and 'primitive natives' in the country's African and Asian colonies. Always a diffuse concept, 'purity' was an artefact construed along an intentional and mutually exclusive black/white axis whereby a supposedly clear-cut and absolute distinction could be made between the two categories. In turn, the imagined purity of 'whites' became defined against the imagined purity (but also the inferiority) of 'blacks'. Such purity, nevertheless, was always an aspiration, an ideal, rather than a concrete attainment, but it was precisely its potency as a social and biological construct that made it so important for anthropologists in the early and midtwentieth century. The physical anthropologists' technical search of 'biochemical' racial purity was embedded in such dualistic imaginary. In tune with this premise, we argue that the nationalistic serology of the purity of 'white' Portuguese co-existed with and was mutually dependent on the colonial serology of those who were classified as 'pure' indígenas, the colonized 'natives' under Portuguese imperial rule.
Through an examination of purity as a powerful and organizational trope we contribute to decentring the commonly held narratives of Latin, and specifically Lusophone, racial exceptionalism in the twentieth century, by which an exclusively positive and celebratory view of race-mixing was (and still is) conveyed publicly. In the early modern period, a 'black legend' of the Portuguese empire developed around negative imaginaries first half of the twentieth century, more often than not, race-mixing was considered a vice rather than as a virtue of both 'nation' and 'empire'.
In this vein, this article provides an analysis of a set of contrary theories that both preceded and accompanied and possibly acted as a counter-weight to the dissemination of the race mixing thesis of luso-tropicalismo from the 1950s onwards, thus contributing to the enrichment of our understanding of the raciological discontentment that was present in Portuguese physical anthropology up to the 1970s. The celebration of racial purity and the rejection of miscegenation by these contrary theories, we suggest, inspired a set of Portuguese physical anthropologists based particularly at the University of Porto to embrace the promises of the science of serology, using blood-group tests, categories, and indexes, to discover the hypothetical reality of pure 'biochemical races' in Portugal and its African colonies (Angola, Mozambique, Portuguese Guinea, and São Tomé and Príncipe). We argue that this research was primarily associated with the enhancement of a double kind of biochemical purity: the supposed superior racial purity of the 'white nation' in Europe which ran parallel to, and in contrast with, the purity of lesser 'black' peoples of the Portuguese colonies, thus creating a dual interpretation that upheld racial hierarchies and undergirded notions of racial purity simultaneously within a nationalist-imperialist framework. 9 We would be naïve to assume, however, that such biological notions of purity were not also grounded on cultural understandings and on developmental schema that placed whites in the forefront and colonized blacks as only 'catching up'; hence the 'civilizing mission' of the Portuguese colonial state. 10 Portuguese nationalism, theories of the origin of the nation and colonialism drew deeply on these viewpoints and were deeply interrelated. In at least one case, as we will show, the allure of purity and the depreciation of mixing would bear the mark of the most hideous scientific racisms of the inter-war period: Nazi Germany's rassenkunde or 'racial science', which explicitly opposed racial mixing. 11 The article thus approaches the Portuguese focus on human blood as marker of race in connection with the international context of medical serology in the first half of the twentieth century. In doing so, we expand discussions on the trope of so-called Iberian 'blood purity' beyond the early modern era and beyond this phenomenon as mere Iberian exceptionality. Early modern historiography has long observed the centrality of the notion of pureza de sangue [purity of blood] in Catholic Spanish and Portuguese conceptions of the heredity of (im)pure status and religious conditions, crucial to regulate lineages and hierarchies both at home and in the colonies. 12 In the context of Iberian histories, then, obsessions with 'blood' and 'purity' (and with anti-Semitism, for example) have deep historical roots that stretch back to the sixteenth century; we would like to emphasize here the 'modern' configuration of this obsession in Portuguese theories and imaginaries of race, nation, and empire in scientific racial anthropology. Thus, although we acknowledge a possible early modern genealogy to the serologists' obsessions, we also consider earlier notions do not determine, and cannot be confounded with, the distinct materialist and biological conceptions of 'blood' and 'racial purity' that characterized the work of the Portuguese racial scientists discussed here. Therefore, in this exploration of the Portuguese serologists' work, we cast light on the enduring influence of imaginaries of human blood and purity in the construction of both the racial 'self' and the racial 'otherness' of Portugal and illustrate how such racialized imaginaries and hierarchies were embedded in nationalist and imperialist endeavours in the twentieth century.

'Race' Between Purity and Miscegenation
During the first half of the twentieth-century, purity was an obsessive motif of scientific research in the racial sciences in Europe. In the early 1900s physical anthropologists followed the then classical presupposition of French anthropology. According to mid-nineteenth-century theorists such as Gobineau and Agassiz, the mixture of races provided a mongrel product that was a 'corruption of the originals, degenerate and degraded', an amalgamation that threatened 'to subvert the vigour and virtue of the pure races with which they came into contact'. 13 In late-nineteenth-century physical anthropology, nostalgia for a now lost idea of 'pure racial types'an idea articulated, among others, by French anthropologist Paul Topinard who stated in 1876 that 'il n'y a plus des races pures sur la terre' [there are no longer any pure races on the earth]did not mean that the quest for such purity died with the assumption of the universality of racial mixing. In effect, even upon such premises the search for purity persisted in physical anthropology for anthropologists still believed a kind of pure raciological abstraction could be distilled into racial 'types' by means of anthropological methods and instruments. 14 Such a sentiment suggested that although any desirable past racial purity had been lost, it could still be recovered, and anthropologists sought eagerly to discriminate the essence of racial 'types' from the confusing 'mixtures'. Such a negative understanding of the consequences of miscegenation prevailed particularly in those scientific communities of countries that still maintained colonial empires. Europeans, in such scenarios, were defined in contradistinction to the colonized 'other', particularly where the other race was deemed to be culturally and biologically far removed from the 'white' European 'standard type'. In fascist Italy and Nazi Germany, most notably, racial mixing was prohibited or discouraged, and, in 13  the former case, hybridism and incrocio were seen to 'irrevocably alter the racial constitution of the nation'. 15 Against the excesses that this obsession entailed, many anthropologists increasingly targeted and criticized purity as an unattainable illusion, emphasizing mixing and unity as a de facto reality or as a desirable occurrence, in an attempt to salvage racial analysis from the dangers and horrors of scientific racism that prevailed in Nazi Germany. In the inter-war years, for example, the study of mixed-races and 'hybridity' in the South Seas was especially pursued by certain prominent North American physical anthropologists with a view to countering the obsession with racial purity and oppose the rising racism of many hardline racial anthropologists. 16 Restrictive and prejudiced concepts of race, which sought to maintain the supposed qualities of racial purity, led many anthropologists throughout the 1930s and 1940s to popularize the terms 'ethnicity' and 'ethnic group'. These descriptors were presented as less biologically rooted, more culturally inflected and largely devoid of political overtones. A landmark text in the anglophone world that attempted to consolidate this shift was the book by Julian Huxley and Alfred C. Haddon: We Europeans, A Survey of 'Racial' Problems. 17 Such a work did not necessarily represent a clear break with the previous conceptualizations. Huxley and Haddon's book has been signalled by Marek Kohn, nevertheless, as a 'transitional work' as it moved between older and newer conceptualizations of the question: 'in relegating race to history, it showed that it was still rooted in the race concept'. 18 While the Nazi policies of marriage restriction, the glorification of the supposed qualities of the 'Aryan race', and the incarceration and extermination of particular racial and other groups were opposed well before the end of the world war in Germany and elsewhere, 19 such deeds confirmed the need to revise the relationship between biology and environment, as well as between science and politics more generally.
In the midst of these debates in anthropology, the consensus that human beings constituted a single species but with plural origins, rather than constituting several races descending from a common origin gained more adherents. 20 This single race, nevertheless, was deemed to be endowed with a large degree of variability in respect of its cultural and biological influences and there was constant linguistic slippage between the acceptance that there was one race and variability being put down to the existence of 'different races'. Models explaining the inheritance of traits, whether positive or negative differed also according to locality and scientific tradition. As stated above, it was anxieties over the mixing of these different human varieties that never left some anthropological circles and which served to renew biological claims of difference. Although it was admitted that unions between allied races were fertile, they were held to be not necessarily desirable. Unions between 'distant' races were held to be either infertile or to 'tend to degeneration'. 21 This latter understanding became the dominant view that, having emerged in the 1850s, prevailed in European physical anthropology until the 1930s. In the ensuing years, scientific theories that decried 'racial mixing' or that saw mixes as degenerative or socially undesirable became less fashionable, entailing 'the collapse of the consensus that miscegenation was unhealthy', a stance in part brought about by the UNESCO statements on race in 1950 and 1951. 22 The divisions in the anthropological camp were not, however, so marked as to be able to identify two clear groupsthose against racial mixing and those in favour of itbut rather formed a graded set of positions that became attached to miscegenation and the future healthiness of each national race. Indeed, this general shift away from biological concepts of multiple races and in favour of a more open acceptance of 'mixed-race' relations was far from hegemonic and had certainly not prevailed across the whole of Europe by the late 1930s.
It was, we argue, the more restrictive and anti-miscegenation stance that prevailed certainly in the post-war period and at least until the early 1950s, despite some exceptions, in Portugal. 23 The rise of ethnic nationalism and the desire to preserve a large overseas empire infused most anthropological work in the country from the 1880s onwards. In the late nineteenth century a dominant racialist discourse on Portugal and its empire combined a rejection of miscegenation, abhorrence of 'black African blood', and a celebration of the European 'whiteness' of the Portuguese metropolitan population against the possibility of hybridity. 24 This racialist viewpoint gained momentum with the creation of the Estado Novo, the fascistic nationalist-imperialist dictatorship headed by Salazar, which was consolidated in 1933. It was, in this sense, the representation of the colonial 'other' that shaped the 'self image of the imperial power' in a process of transculturation 21 Young, Colonial Desire, 18; cf. Stepan, The Idea of Race in Science, 105-6. 22 Kohn, The Race Gallery, 44-7 (at 47), drawing on Robert Proctor, 'From Anthropologie to Rassenkunde in the German Anthropological Tradition', in Bones, Bodies, Behavior, 138-79. 23 Exceptions include the work of the already mentioned haematologist and serologist Almerindo Lessa. 24 Roque, Antropologia e Império; Ricardo Roque, The Colonial Anthropometrist (manuscript in preparation). that operated not only from metropole to colony but also in reverse. 25 This 'pure' Portuguese race, in addition to allowing for the differentiation and subsequent hierarchization of regional types in Portugal, 26 and for the construction of an 'original' racial type that held pride of place in the historic foundation of the nation, 27 was also to be distinguished from both Portugal's colonized Africans and Asians and other 'races' in Iberia and Europe. Purity of race, followed by negative views on the impact of cross-racial miscegenation, in other words, was the dominant motif of Portuguese racial imagination until the mid-twentieth century. It was the tension between purity and miscegenation, and between Europeanness expressed as whiteness and the colonized as black, which was to guide the early Portuguese reception of the new science of blood.

Blood Science in Portugal: a Research Hypothesis
Basing their work on the identification of red cell types in research into compatibility for blood transfusions undertaken in the early 1900s, a number of scientists across Europe posited that it was possible to identify races or sub-races by reference to blood. In 1900 Kurt Landsteiner defined three different blood groups and these were complemented in 1901 when Decastello and Sturli discovered a fourth. 28 There was no unanimity, however, on the designation of the blood types identified, with some scientists favouring a I, II, II, IV system of classification. In the late 1920s, the attempt to rationalize the classification of human blood types into A, B, AB and O (what became known and internationally adopted in 1927 as the ABO system) 29 expanded the clinical uses of blood transfusions dramatically, and came to configure one of the first known demonstrations of the application of the principles of Mendelian genetic inheritance in humans. Soon anthropologists in Europe and the United States looked at the new so-called serological methodologies as a promising pathway in racial analysis. In 1919, Ludwik and Hanka Hirschfeld (also written Hirzsfeld) famously argued for geographical patterns in ABO blood-group variation corresponding to distinct national and racial populations; they claimed a 'biochemical racial index' could be established on the basis of such distributions. 30 Literature on the subject grew exponentially in the inter-war years. 25 We draw on the interpretation of Mary Louise Pratt's Imperial Eyes: Travel Writing and Transculturation  The researchers Von Dungern and the Hirschfelds found that of the four classes, A, B, AB and O, it was the A group that was the predominant type in Northern Europeans. For these scientists, the inheritance of the dominant characters, A and B, depended on two pairs of independent allomorphic genes while the O group had no specific qualities and was, in a sense, a group without genes. Felix Bernstein argued against this notion, however, and suggested a gene for the O group, which would be re-named R. Despite these developments, the lack of international nomenclature and different classifications of blood types hindered research and the actual application of knowledge on blood types, particularly in the field of transfusions. Such doubts, however, did not impede many anthropologists seeking to isolate 'racial types' in accordance with the distribution of different blood types. Scholars of race eagerly sought to develop the hypotheses, methods and taxonomies of blood groups and apply them to racial analysis, in the expectation that the 'biochemical' order of blood types would be made to correspond to distinctively 'regional' or even 'national' types of human races.
These developments were followed attentively in Portugal. On the basis of the French model of anthropology as a naturalist 'science of race', the first university course in anthropology was created in 1885 at the University of Coimbra, the oldest (and, at that time, the only) Portuguese university, to be followed by the University of Porto, established in 1911 under the new Republican regime. For the next four or five decades, Porto academics and their political networks would dominate the anthropological establishment and would place their fixation with the essential purity of raceits genesis, its perpetuation, and its contamination or degenerationat the centre of this scientific field.
It must be noted, however, that concerns in Portugal over racial purity and the supposed dangers of racial mixing were not static phenomena. There were challenges to such ideas, not least from the Luso-tropicalist positions of Gilberto Freyre, as discussed above. Miguel Vale de Almeida has argued that discourse on miscegenation in Portugal passed through three broad periods from the early 1930s, from more or less complete opposition through to greater acceptance by the 1960s and 1970s. 31 Pro-miscegenation ideas slowly, but also ambivalently and never completely, penetrated Portuguese universities, colonial institutes and the higher offices of the state from the late 1940s onwards. Such a periodization may be useful but it presents an overly homogenizing and progressive narrative that irons out important aspects of the variability of twentieth-century Portuguese racial conceptions. Yet in many scientific circles, most importantly in the influential school of colonial anthropology championed by Mendes Correia and his pupils at the University of Porto, scepticism, if not open opposition, towards miscegenation was dominant well into the 1970s. 32 In order to examine this fundamental dominant adherence to understandings of racial purity and opposition to racial mixing, as espoused primarily by the Porto school of anthropology, we now examine three interconnected tendencies within serology and blood research emanating from this locality. First, the article discusses the birth of 'sero-anthropology' and attempts to isolate pure races at the University of Porto. Second, it traces the search for 'bio-chemical' purity in the late 1930s and early 1940s particularly in the work of the figure of Dr Ayres de Azevedo, also of the University of Porto, who worked under the direct supervision of race scientist Otmar von Verschuer in Germany. Third, the article discusses the programme of colonial serology as elaborated by Dr Amílcar Magalhães Mateus in the form of his 'haemato-anthropological index'. In analysing these scenarios, we argue that the mark left by this hardline raciological programme in the study of human blood as a pathway to defining pure 'biochemical races' was a significant one and one which only perhaps became marginal to Portuguese anthropological imagination once the very structures of the empire and the regime that supported it, the Estado Novo, lay in ruins. 33 Imagining 'Serological Races': The 'Birth' of Sero-Anthropology at Porto Almerindo Lessa evocou os tempos em que iniciei em Portugal os estudos de serologia antropológica. Eram os tempos heróicos. Era o começohá 25 anos! Agora, há muitos materiais acumulados, mas ainda não possuímos dados suficientes para ter uma ideia de conjunto. Eu não sou um feiticista da estatística, mas acredito nas suas possibilidades para nos dar uma certa margem de segurança. 34 [Almerindo Lessa evoked the period when I began serological studies in anthropology in Portugal. They were heroic times. This was the beginningsome 25 years ago! Now, there are large amounts of materials that have been amassed, but we still do not have sufficient data in order to gain a comprehensive overview. I am not a statistical magician but I do believe in the ability of statistics to provide a certain degree of certainty.] Thus in 1951 physical anthropologist António Augusto Mendes Correia recalled his role and that of his research group at Porto University as the self-proclaimed pioneer in the origin-story of 'anthropological serology' in Portugal. 35 He spoke briefly as a 33 Such a raciological and often racist imaginary continues to operate, nevertheless, in democratic Portugal, a country that still struggles to come to terms with its imperial, racial, and ideological past. special guest at the ambitious First Colloquium on African Haematology, the international scientific meeting organized in Lisbon by Almerindo Lessa (the rising national authority on medical serology and haematology) to foster research on the blood of 'normal black Africans'. 36 Correia's invitation to make a statement at the congress, at a moment when serology was in its heyday, confirmed his position of authority in the field of human sciences in Portugal. By then, aged 65, he was at the height of his academic and political power: besides holding the Professorship of Anthropology at University of Porto (continuously since the 1910s), he was simultaneously President of the Government Overseas Research Board, the High Colonial School, and the Geographical Society, positions he kept almost until his death in 1960. Certainly, Lessa's evocation and Mendes Correia's self-praise, when sero-anthropology claimed victory and institutional success within Correia's programme of 'colonial anthropology' of the indigenous 'natives', were self-serving statements that simplified a more complex historical process. Yet, Correia's brief retrospection in fact refers to a set of pioneering initiatives in serology developed at the University of Porto in the 1920s-30s under his very own guidance. Correia's statement is also significant for another reason. It combines a denial of the 'fetishism of statistics' with an appraisal of their significance in guaranteeing 'safer' results in race studies. The statement of 1951, made in the midst of European and wider attemptsfor example, those of UNESCO referred to aboveto deconstruct race and question it as a viable scientific and cultural category, thus reveals the original ambivalence, the combination of pragmatic hope and critical scepticism, which characterized Mendes Correia's early engagement with racial serology 25 years earlier and which was never extinguished throughout his life.
The so-called 'heroic times' of Mendes Correia's 'anthropological serology' began at the northern city of Porto in the 1920s soon after he had been appointed Professor of Anthropology and director of the Institute of Anthropology at the University of Porto. He was eager to make his name in anthropology, both nationally and internationally, inspired by a racialist and conservative form of right-wing imperial-nationalism which would lead to a lifelong and profound political association with the Estado Novo regime. He was a prolific writer on criminology, eugenics, archaeology, anthropometry, and pre-history, initially concerned with the origins and racial constitution of the Portuguese metropolitan population and, increasingly, with the origins and racial constitution of the 'native races' under Portuguese colonial dominationa sub-field he would baptize as 'colonial anthropology' in 1934. In the wave of the white racialist theories of a preceding generation of (physical) anthropologists, 37  affiliation of Portuguese people with a pure type of the white European race (the 'Mediterranean'), and vehemently rejected the possibility of 'black' African mixtures. The Portuguese population, he argued, was primordially and primarily a pure-breed type of the 'Mediterranean' or 'Ibero-insular' white European race. 38 Indeed, throughout his life, he was a strong defender of the intrinsic value of aboriginal racial purity as part of a broader imagination of the potentially nefarious and 'degenerative' consequences of racial métissage upon Portugal as a nation of imperial vocation; races should be kept apart and each in their 'own place'. The imagined interdependence between the purity of Portugal's white European-Mediterranean race, the investigation of the colonized peoples as distinctive pure 'native races', and finally the vitality of Portugal as an imperial-colonial nation guided his visions and inspired his anthropology students. It was with a view to pursuing this programme that Mendes Correia embracedsceptically and pragmaticallythe promises of blood-group classification in unveiling the existence of 'serological races'.
In the inter-war years, as stated above, the enthusiastic incorporation of blood grouping into the study of human racial difference gave rise to a new sub-disciplinary variant in physical anthropology termed sero-anthropology. In Portugal, Mendes Correia, trained in the classic French and German physical anthropologies and an advocate of a physicalsomatological definition of race, 39 followed the international developments in racial serology attentively but was cautious about embarking on the serology trend. 40 Thus in 1926, in a set of articles published almost simultaneously in Portuguese and in French (in the reputed racial anthropology journal, L'Anthropologie), he extensively discussed the new scholarship on serology and the possibility of grounding the concept of 'race' and racial taxonomy in blood-based 'biochemical' characters. 41 Could race concepts and taxonomies be grounded upon pure 'biochemical' evidence and statistical tests on human blood? Could 'races' be established on purely serological terms? Mendes Correia attempted a response to these then pressing questions. He acknowledged the 'difficulties' of physical anthropologists in defining race as simply morphological, based on the study of 'external' bodily characters, and he was open to discussing the possibility of grounding race upon the 'chemical analysis' of internal bodily constituents. Yet, against excessive enthusiasm for blood-group interpretations, Mendes Correia doubted that serological methods of 'iso-agglutination' and blood-group categorization, as developed by the Hirschfelds, could in effect lead to the isolation of a 'chemical substratum' of race, 38  and to differentiating real 'biochemical races' (races biochimiques). 42 In the mid-1920s, he wrote critically about the incongruence between groupings of morphological characters and haematological characters, declaring 'strong reservations regarding the attribution of an important taxonomic value to those [iso-agglutination] reactions'. 43 He pointed out 'imprecisions' concerning the 'supposed' existence of 'serological races'; 'chemical methods' in blood-grouping, he claimed, exciting as they might seem, ought not to be prioritized over 'morphological observations' in 'determining the human races', an appreciation that would hold strong for him into the 1950s. 44 'L'importance de l'étude des phénomènes d'iso-agglutination est evidente dans la clinique' [The importance of the study of the phenomena of iso-agglutination is evident in the clinic], Mendes Correia remarked, 'mais l'interpretation théorique du fait et son emploi comme un élément de classification ethnologique sont, jusqu'à aujourd'hui, très loin de pouvoir braver avec succèss une critique sereine, voire une critique trop modérée' [but the theoretical interpretation of this fact and its employment as a marker of ethnological classification are, to date, far from heralding a successful and reliable critique or even a fairly moderately reliable one]. 45 In addition, he reacted sceptically to excessive 'enthusiasm' surrounding the statistical making of 'biochemical indexes', criticizing the Hirschfelds' dualistic biochemical index and engaging in a brief polemic with American anthropologist Laurence Snyder, whose 'seven types' of serological races were described as mere 'mathematical speculation' by Correia. 46 In spite of his marked scepticism and the 'insufficiencies' he saw in serology, Mendes Correia accepted that in the future an improved raciology could contribute importantly to racial taxonomy in the work of the physical anthropologist. He also wrote that serology is not a 'campo fechado a investigações fecundas, nem é lícito … pôr em dúvida as largas perspectivas das indagações bioquímicas com o objectivo da determinação mais ou menos satisfactória da raça e da individualidade' [field closed to profound research; neither is it acceptable … to question the extensive perspectives provided by biochemical testing which has the objective of determining more or less satisfactorily race and individuality]. 47 And, in concluding his L'Anthropologie article he declared: 'Les méthodes sérologiques de classement des races attendent encore un perfectionnement qui seul leur permettra peut-être de remplacer ou d'aider efficacement les classifications morphologiques' [Serological methods of racial classification currently await a degree of perfection that will allow them to replace, perhaps, or significantly aid morphological 42 Correia, 'As tentativas de definição bioquímica da raça e do indivíduo', 6. 43 Correia, 'Sur les prétendues "Races" Sérologiques', 439. 44  classifications]. 48 Thus Mendes Correia himself would never become a practitioner of serological methodologies; but, at the same time that he criticized the excesses of racial serologists in the 1920s, he encouraged his disciples to pursue the new science of blood in the analysis of race.

'The Most European of Europeans'
In 1926, the same year he voiced his sceptical views on 'serological races', Mendes Correia launched the raciological study of blood-groups at his Institute of Anthropology. At the time that he was writing about 'supposed serological races' it was under Mendes Correia's supervision at Porto University that serological methods were in fact first applied to anthropological research on Portuguese metropolitan populations, a development that apparently fuelled the early wave of sero-anthropological research in Portugal between 1926 and 1950. Precisely in 1926, Correia instructed his young anthropology student, Adélia Seirós da Cunha, to pursue a sero-anthropological study on a population of 459 Portuguese army soldiers from northern Portugal, a regional population then imagined as the 'modern' representative of the most authentic 'aboriginal' European inhabitants of the country. The result of Cunha's work was a doctoral thesis entitled Grupos hemáticos nos Portugueses (Haematic Groups in the Portuguese) that bears the mark of Correia's raciological visions. 49 Cunha's thesis was motivated by the need to fill the 'lacuna deploravel' [lamentable gap] in serological studies on Portugal, 50 and it seems to have constituted a testing-ground for Correia's theories in at least two senses. 51 On the one hand, it came to offer proof to her mentor's visions about the pure racial European nature of the Portuguese people; on the other hand, it was a means to address Correia's trouble with the relative value of blood-based 'races'. For, observing that the biochemical race types had nothing in common with the anthropological classifications based on somatological characters, Cunha's work proposed to figure out the differences, comparing the Portuguese soldiers' serological results with classical physical charactersthe cephalic index, pigmentation, and stature.
In the context of our analysis, Cunha's conclusions as regards the racial purity of the Portuguese as white 'Europeans' is especially worth considering. The work began with a standard synthesis of the medical 'history' of blood transfusion, and with an account of 48 Correia, 'Sur les prétendues "Races" Sérologiques', 443. 49 The use of army recruits for physical anthropology was widely disseminated from the nineteenth century, and it was also on soldiers that Hirschfeld conducted his pioneering studies on the 'racial biochemical index' in 1919 a precedent that was explicitly taken into consideration by Cunha. Adélia Seirós da Cunha, 'Grupos how the findings on blood transfusion and iso-agglutination began to be applied to the racial study and classification of distinct human populations in the early twentiethcentury. Echoing the questions raised by her supervisor, Seirós da Cunha intended to investigate the Hirschfelds' and Landsteiner's claim that a 'novo processo taxonómico para raças bioquímicas' [new taxonomic process of biochemical races] in humans could be established on the basis of the differentiation of blood-types. 52 Guided by this question, and following 'Hirschfeld's [sic] technique' of in situ blood-group determination, Cunha applied serological methods of classification and statistical analysis to a vast number of blood samples taken from Portuguese army recruits. In particular, Cunha tested the Hirschfelds' hypotheses of 1919 that the two main human blood-group types, A and B, were to be found in different degrees among human populations, according to 'race' and geography. According to the Hirschfelds (and followers), the predominance of blood-group A was 'uma característica das nações europeias' [a characteristic of the European nations]; this would gradually reverse in African and Asian peoples where group B, in contrast, was more prevalent than group A. 53 The Hirschfelds had also proposed translating these differences statistically, creating a mathematical formula, christened an 'índice bioquímico de raças' [racial biochemical index], 54 which, in their view, could position every human population according to blood-group variation in a threefold typological racial scale: the 'European', the 'intermediary', and the 'African-Asian' types.
Seirós da Cunha then experimentally applied the Hirschfelds' biochemical index to the blood-group results of the Portuguese soldiers. And she came to a self-satisfying conclusion: the Portuguese sample was predominantly classified as blood-group A; it revealed a very high 'racial index' of Europeanness, which placed the Portuguese at the top of the white European biochemical type devised by Hirschfeld. 'Em virtude deste valor' [Given these values], Cunha concluded, 'Portugal ocuparia o primeiro lugar da coluna dos povos do tipo europeu. Confirma-se a opinião de Hirschfeld, segundo a qual o grupo A iria aumentando, à medida que nos aproximássemos do Sudoeste da Europa' [Portugal would occupy the top place in the list of peoples of European type. Hirschfeld's opinion, according to which the propensity for Group A will increase as we come closer to the south west of Europe, is thus confirmed]. 55 In addition, Cunha decided to test the Portuguese results against Laurence Snyder's formulas and she came to a similar conclusion: 'Parece-me então acertado colocar Portugal, não junto da Itália, mas no princípio da lista de Snyder, para o tipo europeu' [It seems to me correct to place Portugal, not alongside Italy, but at the top of Synder's list of European types]. 56 In sum, according to Seirós da Cunha's results, the Portuguese did not simply belong to the European type. Because their index was situated at the extreme of the Western 52 Cunha, 'Grupos hemáticos', 35. Cunha referred to 'Hirschfeld's work', in the singular, possibly not appreciating that the publications emanated from both Ludwig and Hanka Hirschfeld. 53 Ibid., 55. 54 Ibid., 56. 55 Ibid., 77. 56 Ibid., 78. European series they could configure the ultimate example, the prototype, of purity of the European biochemical race revealed by serology. Portugal could be considered 'uma região biologicamente pura' [a biologically pure region], wrote Cunha, 'onde a mutação se daria com maior intensidade' [where the mutation [from the 'primitive' Group O to Groups A and B] would take place with greater intensity]. 57 It was certain, Mendes Correia also admitted, and Seirós da Cunha reiterated, that to presume the 'pureza original' [original purity] of race (rather than the mixing, as most classical physical anthropologists assumed) was an 'hipótese errónea' [erroneous hypothesis], a bias, that pervaded the biochemical constructs of the Hirschfelds and Snyder; for this reason, Mendes Correia and Cunha expressed scepticism with respect to the overambitious indexes, as 'ousadias do espírito, que não dispensam uma confirmação [exalted expressions of the spirit that do not allow for confirmation].' 58 Nevertheless, these were rhetorical artifices that did not preclude Correia from praising Cunha's 'scientific results' as a confirmatory demonstration of the racialist thesis of a Portuguese white European superiority. As late as 1955, for instance, Mendes Correia's disciple Amílcar Magalhães Mateus (whose work we analyse below) would refer to Cunha's thesis as the first authoritative demonstration of Portugal's European racial purity. Mateus wrote: Apesar de ser uma tese de licenciatura [sic], o trabalho da Dra D. Adélia Seirós da Cunha teve e tem grande valor, dado que, nas publicações de serologia, Portugal não era citado por falta de estudos dessa natureza. Deu-se o caso que as determinações feitas para aquele trabalho colocavam o nosso País em situação privilegiada, pois que, além de apresentarem a percentagem do grupo A superior à do B, como é próprio de povos europeus, a superioridade daquela em relação a esta era maior do que nos restantes grupos europeus até então estudados. Quere dizer, estas determinações colocavam os Portugueses metropolitanos como os mais europeus dos povos europeus, contra a tese da existência de grande infusão de sangue negro, tese seguida por alguns autores estrangeiros, mas sem base sólida. Trabalhos posteriores a este não fizeram mais que confirmar as conclusões a que chegou aquela senhora. 59 [Despite being an undergraduate thesis [sic], the work of Dr Adélia Seirós da Cunha was of great value given the fact that, in terms of publications on serology, Portugal was not cited as a result of a lack of publications on this subject. It is the case that the results found in this study place this country in a privileged position. This is because, in addition to finding that the percentage of Group A was superior to that of Group B, as is normal in European populations, the difference between the former and latter was far greater than that detected to date in other remaining European groupings. This means that these findings placed the metropolitan Portuguese as the most European of the European, contrasting with the thesis defended without solid foundation by many foreigners that posits a significant infusion of black blood. Later studies only confirmed the conclusions arrived at by this researcher]. 57 Ibid., 78. 58 Ibid., 90. 59 Mateus, Os grupos sanguíneos e as suas aplicações, 21.

Biochemical Purity: Ayres de Azevedo and Germanic Science in Portugal
For our next example we remain in Porto but consider the international and more overtly political dimensions of racial science and blood research within the context of flows of knowledge between Portugal, Germany, and Nazi Germany in the 1940s. The links between Portuguese and German science, in general, were strong in the period discussed and, in respect of physical anthropology, the 1930s were the key period for the forging and consolidation of such relations. The steady internationalization of Portuguese science, letters, and culture was facilitated by the concession of grants for study and research abroad via the prestigious Institute for High Culture (Instituto para a Alta Cultura, IAC). Between 1929 and 1950, France was the most popular destination for such academic and research visits, with some 360 grants awarded; Britain came second with 90 and Germany third of all countries with 77. 60 In turn, a potent cultural policy was executed by Nazi Germany in Portugal, 61 which was echoed in the creation of organizations that mirrored Nazi creations such as the 'Joy at Work' initiative. The Coimbra anthropologist, Eusébio Tamagnini, was in receipt of many grants from the IAC and was also active in dialoguing with certain branches of German science, particularly physical anthropology and race science. Evidence of such collaboration can be seen, for example, in the inaugural meeting of the Portuguese Society for Eugenic Research (Sociedade Portuguesa de Estudos Eugénicos) in Coimbra in 1937. 62 Here, the German race scientist Eugen Fischer, among other foreign guests, was invited to address the assembled eugenicists. On the same occasion, he was awarded an honorary doctorate from the University in recognition of his research. 63 Fischer, after the Coimbra event, travelled to Porto, where he inaugurated the facilities at the Institute of Anthropology. His presence in Portugal confirmed an evolving history of the reception of German and Nazi eugenics in the country that would last beyond the Second World War. 64 Tamagnini's inaugural speech at the Eugenics Society was reported on amply in the German race hygiene journal, the Archiv für Rassen-und Gesellschaftsbiologie. 65 Although not central to the work of anthropologists in the city of Porto, the Porto-based medical doctor José Ayres de Azevedo Novais Basto can be placed within this growing Germanic scientific and political web of relations. 66 While any actual links between Ayres de Azevedo, Mendes Correia and the Porto Institute of Anthropology remain unclear, two issues need to be highlighted. The first is that Ayres worked on his thesis under the aegis of the Faculty of Medicine, not the Institute of Anthropology. The second qualification to be made is that, as is to be expected, Ayres de Azevedo was clearly familiar with the on-going anthropological work that was being undertaken in Porto and other centres of anthropological research in the country. What we argue here is that there were epistemological and, to some degree, methodological connections between the work of Ayres and Mendes Correia in respect of the search for racial purity in white Europeans, but we have not detected any explicit working together or collaboration between the two scientists.
Dr Ayres de Azevedo began his career in 1939 as an adjunct at the Júlio Dinis Maternity Hospital in Porto. 67 Shortly afterwards, he was granted a scholarship by the IAC in order to work with the 'race hygienist' Otmar von Verschuer, director of the Kaiser Wilhelm Institute, first in Frankfurt and then in Berlin between May 1941 and August 1943. 68 Even though, as we will see, his work, much of which went towards the elaboration of his doctoral thesis on blood types, was later subject to disapproval from his Porto colleagues, Ayres de Azevedo was highly praised initially by the director of the KWI's anthropology, human heredity, and eugenics section. According to Verschuer, Ayres de Azevedo had obtained 'new and valuable scientific data' from his work on hereditary factors among twins, a subject that had long interested German and, increasingly, Nazi scientists. 69 Verschuer's report also noted that Ayres's work on twins would allow for important conclusions on the question of heredity and serology.
The 'Bio-chemical Purity' of the Portuguese Race Before his cool reception in Porto on his return from Nazi Germany, Ayres de Azevedo had made his name in a number of high-profile scientific events, which had been organized with the approval of the Estado Novo. Notable among these was the 1940 Congress on Population Sciences where he spoke of the 'bio-chemical purity' of the Portuguese race. 70 This congress was held as part of the wide-ranging programme of events under the umbrella of the Congresses of the Portuguese World, inaugurated in 1940 by the Salazar regime in order to showcase the historic and contemporary achievements of the New State. 71 The date chosen by the dictatorship for these events was significant: 1940 represented the eighth centenary of the establishment of the Portuguese nation (1140) and the third centenary of the recovery of Portuguese national sovereignty from the Spanish (1640). The year 1940 represented the consolidation of the Salazar regime.
The paper on blood chemistry by Ayres de Azevedo was certainly up-to-date in its survey of thinking on the distribution and nature of blood groups. In a brief overview of research, following the work of the already mentioned figures of Hirschfeld and Hirschfeld, 72 Von Dungern, and Landsteiner, the Portuguese researcher noted that despite recent warnings against racial differentiation on the basis of blood analysis, the field was still promising for biology, therapeutic work, legal medicine, and anthropology. It had been possible, according to the above authors, to divide humanity into three principal groups, followed by a fourth (drawing on the work by Decastello and Sturli), on the basis of blood types. 73 According to Ayres de Azevedo, these authors had proved that blood types were stable and could only be altered if crossings between individuals of different groups took place. Any crossings followed the patterns set out by Mendel, obeying with 'maravilhosa precisão' [marvellous precision] his laws of inheritance. 74 Ayres de Azevedo noted that all racial groups were diverse in their characteristics and, for example, some 90 per cent of Amerindians were deemed to belong to the O group and some 80 per cent of Inuit ('Eskimos' in the text) belonged to the same group. 75 Certain qualities, nevertheless, were, according to Ayres de Azevedo, attached to the O group. He noted that his fellow Portuguese researcher, J.R. Santos Júnior, had traced the incidence of the O group blood in people in the Tete region in Mozambique (with a presence of 92.7 per cent), allowing him to extrapolate certain lessons in respect of the purity and status of African populations in the racial hierarchy. 76 Such a predominance of the O group allowed Ayres de Azevedo, in turn, to argue that this showed that 'primitive humanity', in this case the colonized Mozambicans, belonged to this very type. 77 By contrast, in Europe, it was to be expected that the highest incidence of the AB group occurred as this represented the most recent peoples to have undergone racial mixing. Similar conclusions on the racial purity of the northern Portuguese had been arrived at, as we have seen, by Seirós da Cunha and Mendes Correia. In other words, Azevedo extrapolated conclusions on the contrasting racial purity of 'white Portuguese' and 'black Africans' from metropolitan and colonial sero-anthropology then developed by his Porto contemporaries, under Mendes Correia. Finally, endorsing René Martial's understanding of the 'anthropological frontier' of Europe, traced by blood types rather than political divisions or territorial lines, 78 such research showed that it was the 'bio-chemical index', following the method suggested by L. Hirschfeld and Landsteiner, which was most important for the differentiation and classification of races.
The mapping of blood types would allow for the entire population of any given country to be surveyed, as Seirós da Cunha and others had begun to do in Portugal, and for details of blood type, family health, educational details, ancestry, fertility, and police records to be correlated in order to determine the health and characteristics of a national or regional group, as was being undertaken in Nazi Germany. 79 Ayres de Azevedo, like Seirós da Cunha and others such as Duarte Santos working at the same time, 80 was keen to explore the racial heritage and blood type distribution of different populations within Portugal itself. He noted that the proportion of type B grew the further south one went, reflecting the predominance of the 'Arab race' over the presence of the 'Nordic race', which was most influential in the North. 81 Ayres de Azevedo drew on a range of findings provided by the researchers Alberto Saavedra (1932), Waldemar Teixeira (1932), António de Matos Júnior (1939), M. Prates, and Fraga de Azevedo (1940), and a number of further studies in what was demonstrably a vibrant field of research in different parts of the country. 82 In Coimbra, the historic seat of anthropology in the country, as well as the work undertaken by Duarte Santos, the Germanophile Eusébio Tamagnini had conducted research into some 2968 individuals in a study that was also presented at the Congress of the Portuguese World in 1940. 83 The amalgam of these pieces of research on a large number of individuals across different localities allowed Ayres de Azevedo to calculate the precise 'bio-chemical index' of the Portuguese race. This was a figure arrived at by adding the percentage of the A type to the AB type and then dividing this by the sum of the B and AB types. The index came out at 4.3 on the basis of some 3757 Portuguese individuals studied by these various researchers. The relative value of this figure in terms of the most European and, even though Ayres de Azevedo does not explicitly say so, the 'best' racial types, became clear. With the Portuguese race indexed at 4.3, it was only surpassed by the English at 4.5. Spaniards scored 3.9, Italians 2.8 and Jews languished at 1.3. 84 In a table included in the paper, he detailed the percentages and frequencies of the various blood types. Out of the Portuguese, Spanish, Arabs from Morocco, Berbers from Morocco, Blacks from Congo and Indians, it was the Portuguese who apparently possessed the highest percentage of blood type A (48.76%). Blacks and Indians possessed the lowest. 85 This supposed 'purity' of the Portuguese (a mere 48 per cent, we should note), Ayres de Azevedo argued, had been maintained because the Portuguese had not mixed to such a large degree with the 'colonial races', unlike the Spaniards. Not only did his assertion affirm the purity of the Portuguese; it also countered the myth of the Portuguese propensity to assimilate and mix with other races and firmly judged such mixing as undesirable. Like Mendes Correia, Ayres de Azevedo coincided in repudiating mixed-race relations and he articulated an explicitly racist warning against such mixtures: other racial types were deemed to be 'contaminadoras possíveis da nossa [raça]' [possibly contaminating influences on our [race]]. 86 The result was a Portuguese race 'de grande pureza, maior que a da quási totalidade dos povos da Europa' [of the highest purity, greater than that of nearly all European peoples]. 87 Despite this display of confidence in the purity of the Portuguese, there remained, for Ayres de Azevedo, the spectre of contamination from outside sources. Could such levels of purity be maintained? If contamination did occur, would the extraneous elements be dominated by those of the local race, allowing the balance to be re-set? If so, by what mechanism would this occur?
Given the directions that the research by Ayres de Azevedo would take in Germany in the field of genetics and eugenics, this work was but a stepping stone on the way to the construction of an explicitly hierarchical and racist typology. This racial order, facilitated by Ayres de Azevedo's particular usage of research on the bio-chemical index, reasserted the existence of independent, identifiable races and gave credence to the idea that it was possible to 'contaminate' one race with the blood of another. In the words of Pauline Mazumdar, the German project that Ayres de Azevedo would eventually subscribe to 84 Ayres de Azevedo, 'A pureza bioquímica', 558. 85 Ibid., 560. 86 Ibid., 560. Lest we assume that Mendes Correia was immune from similar prejudices, a reading of A.A. Mendes Correia, 'Escravos africanos em Portugal e no Brasil', Boletim Geral das Colónias, Vol. 14, No. 157 (1938), 3-30, where 'estranhos encrespamentos de cabelos' [strange curls in the hair] were identified as traits to be avoided (12), will illustrate how such racist interpretations were publicly aired by the Porto anthropologist. 87 Ayres de Azevedo, 'A pureza bioquímica ', 560. was 'a model system for human genetics, and for its practical arms, eugenics and race hygiene. The blood groups themselves provided a race-marker that attracted the attention of the German völkisch anthropologists. It promised to be a new and scientific way to define populations, to distinguish races from each other, and to trace their origins, migration routes, and boundaries'. 88 While the questions he posed on the mechanisms of 'race recovery' when 'contaminated' by other blood groups could not, by his own admission, be answered at this stage, the serological journey embarked upon in Germany in the 1940s would provide, he hoped, definitive evidence.
In December 1940, Ayres de Azevedo applied for funds from the IAC in order to undertake research at the Kaiser Wilhelm Institute. The study plan outlining the areas he would be working on included racial hygiene, eugenics, demography, blood group research, matters pertaining to sterilization, birth control, the struggle against miscegenation and the competencies of the state with respect to these issues. 89 It was particularly blood group research, inheritance in twins, and the medico-legal purview of doctors acting under the Nuremberg Laws that were to capture Ayres de Azevedo's interest most. Ayres de Azevedo's thesis written during his German experience places these concerns at its centre.

Blood Types, Twin Research and the KWI
Blood group research, racial mixing, twin research and work on inheritance were all integral parts of work undertaken in the German race hygiene and eugenics movement. Verschuer had requested Reich Research Council funding for twin studies several times during the 1930s and on one occasion he requested funds for work on the inheritance and distribution of eye colour in the 1940s. The infamous Dr Josef Mengele worked as an assistant at Verschuer's Frankfurt Institute for Genetic Biology and Race Hygiene in 1937 and returned to work with him at the KWI in January 1943, being named camp doctor at Auschwitz in May 1943. 90 At Auschwitz, Mengele continued his twin research, used live subjects for research, killed others for similar purposes and had blood and other samples sent to Verschuer at the KWI. 91 There were two particular projects that Mengele would pursue at Auschwitzon specific proteins and eye colour, activities that were approved in August and September 1943. 92 The Jewish slave-assistant, Dr Miklós Nyiszli, reported on these activities after the war and noted that four pairs of twins were killed by Mengele, the heterochromatic eyes of which were sent to the KWI's anthropology section. 93 With respect to the blood samples derived from Auschwitz, Verschuer acknowledged in March 1944 that these had been sent to the KWI for processing. 94 It is almost certain that Ayres de Azevedo worked on at least some of these human samples. 95 In his 1944 thesis he remarked that the assessment of the relative importance of hereditary and environmental factors could be undertaken through twin studies as well as research into the pigmentation of the iris, hair and skin, and the measurement of eyebrows, nose, ears, as well as blood type. 96 In addition to these activities, researchers at the KWI routinely adjudicated in so-called racial ancestry or paternity cases in accordance with the Nuremberg Laws, that is, on whether 'racial contamination' had occurred between Aryans and Jews. 97 If in paternity cases a child was found to have a Jewish father, the latter would 'almost certainly be deported to the East to an uncertain future and, presumably, a violent death'. 98 In order to complete his research on Nazi extermination, Müller-Hill asked as many German human geneticists as he could who had been involved in this work and all of them claimed that they had always related the scientific truth: 'it was not their business to worry about the fate of the person'. 99 Such practices did not prevent Ayres de Azevedo from praising the population policies of Nazi Germany or from advocating sterilization and the elimination of the descendants of 'dysgenics'. 100 Ayres de Azevedo's doctoral thesis comprised two main parts. The first was a history and systematization of serology, followed by a technical exposition on the determination of blood types. The second part focused on the hereditary transmission of different blood types. This part also contained a section on heredity in twins and twin research. The question of the awareness or otherwise of Ayres de Azevedo with respect to the samples that he employed for his research is posed immediately in the preface to the thesis. Here, he recognized that blood used for research came from many Germans, some of whom were children, and he thanked the hospitality of what he termed this culturally advanced and naturally affable people. The question of how voluntary this participation was or where exactly the samples came from, however, is not clarified in the preface. 101 There were two principal findings that Ayres de Azevedo advanced in his thesis in respect of his twin studies. The first was that there was a strict correlation in terms of blood types in identical twins of 100 per cent and in non-identical twins, this percentage was 70 per cent. Second, for Ayres de Azevedo, this proved beyond doubt Mendelian processes of inheritance. Such information could be employed fruitfully in the paternity cases that the KWI was asked to adjudicate. 102 Despite the detailed nature of the research presented in his thesis and elsewhere in the Porto-based Jornal do Médico, as we have said, the reception of Ayres de Azevedo in this city was far from warm. This arose from three main factors. First, the kind of politicallydriven scientific research that Ayres de Azevedo undertook in Germany, with a focus on strong interventionist eugenics and an approval of sterilization, was already out of favour in Portuguese milieus by the mid-1940s. The praise by Ayres de Azevedo for National Socialist science and politics, and his past as a National Syndicalist sympathizer, were apparently not shared by his peers and assessors in Porto and with the tide of the war having turned against Nazi Germany, any approval of this country's research may have been politically dangerous. Certainly, the Portuguese state had begun a rapprochement with the Allied forces. 103 Second, and related to this, even though such research met with a certain positive reception in Coimbra and among some anthropologists in Porto, at the University's Faculty of Medicine where the thesis was presented, a more social hygienic orientation in medicine prevailed. 104 Third, a question mark had been raised over the integrity of Ayres's work in Germany and there was knowledge of these doubts in the Porto medical circle. 105 It remains unclear whether the doctoral degree was in fact actually ever awarded. 106 blood groups of the 276 Africans and Asians who had come from the Portuguese colonies as performers in the Colonial Exhibition. 109 A couple of years later, he likewise instructed the same Santos Júnior, whom he designated in 1936 to become the chief investigator of the Mozambique Anthropological Mission, to incorporate the study of blood groups in field research. In 1936, blood groups were collected from 138 'natives', and due to their 'great anthropological importance', during the Mission's many field campaigns (which lasted until 1959), Santos Júnior took pains to collect over 2000 blood samples for the determination of ABO blood groups, from different 'ethnic groups' of 'blacks [pretos] from Mozambique'. 110 In Mozambique as well as in Correia's subsequent 'colonial anthropological missions' sponsored by the Portuguese imperial state, in Angola and Timor (headed by António de Almeida) and in Guinea-Bissau (headed by Magalhães Mateus), serological methods were used extensively as a means of cracking the long-lasting problem of racial taxonomies and genealogies. Indeed, for Mendes Correia and the chief anthropologists of these Missions, relations between anthropometry and serology should be complementary and cumulative. Beyond anthropometry and blood groups, a plurality of bio-anthropological methods was adopted and different techniques appliedsuch as psycho-technical tests, dermatoglyphs, arterial tension, and constitutional biotypology. In addition, and importantly for our purposes, in the emerging colonial anthropology, as in the study of the Portuguese people, racial purity, the inescapable fact of mixing notwithstanding, was, again, the magnetic centre of research activity. In the coming years, then, the 'colonial anthropology' of the 'natives' would follow in parallel to the anthropology of the 'Portuguese'separate and yet intimately connected by the obsessive trope of 'purity'. It is thus significant that a primary concern of physical anthropologists doing blood-grouping fieldwork in the colonies was to guarantee the 'racial' or 'ethnic' purity of the samples. Santos Júnior assured his readers in 1957, Registe-se que os pretos das nossas séries, são de relativa pureza tribal. No inquérito preliminar eram aparta dos aqueles que se diziam filhos e netos, pelo lado paterno e materno, de indivíduos pertencentes à mesma tribo. Só esses eram estudados. Era a maneira de se apurar uma relativa pureza, no meio dos muitos cruzamentos entre individuos de tribos diferentes' [It is recorded that blacks in our samples are of relatively high tribal purity. In the initial questionnaire, those who told us they were children and grandchildren, whether on the paternal or the maternal side, of individuals that belonged to the same tribe, were segregated. It was only these that were studied. This was how we arrived at a relative degree of purity among the multiple crossings of individuals from different tribes]. 111 Cunha in the 1920sthat the Portuguese were positioned as the purest European types. According to Mateus, the Portuguese possessed 'um dos mais elevados da Europa' [one of the highest [haemato-anthropological indexes] in Europe]. 114 He argued additionally that blood groups and his new index could help solve the problem of the 'origins of present-day human groups', thus coming to the aid of Professor Mendes Correia's ambitious hypothesis for locating in Southeast Asia and, specifically in Portuguese Timor, the birthplace of mankinda point that possibly further raised his superior's hopes for the advantages of serology in the Timor expedition. 115 Statements such as Mateus's, providing serological grounds for his ethnogenic theories on Timor and Oceania, possibly strengthened his superior's embracing of sero-anthropology.
In the 1950s, in just a few years in fact, Portuguese anthropological literature on blood groups in colonial populations expanded rapidly. In 1955, the same Amílcar Mateus emphasized the momentum it had obtained. He wrote enthusiastically about the 'applications' of blood groups in the anthropology of the 'Portuguese Overseas provinces', and the numerous results already obtained. He referred to the 'thousands' of observations in individuals of various tribes, the very large series then already obtained by António de Almeida's field campaigns in Angola (6000) and, more recently, in Timor (more than 5600) for ABO blood types. 116

Conclusion
The openly racialist and colonialist tone of some of the work discussed above may cause surprise, but the integration of serology in such a colonial and overseas form of anthropobiology was not dissonant with broader post-war trends in the systematic study of blood groups in anthropology. In the early 1950s the study of blood groups gained further momentum, both nationally and internationally. The escalation of field campaigns within the framework of the Portuguese Anthropological Missions that took place in this decade was part of this development. Notwithstanding the anti-racist climate and a steady broader scientific and cultural acceptance of racial mixing which followed World War II, international interest in the study of race and human diversity persisted, particularly under the guise of blood-group research and the patterning of its geographical and 'population' variability. 117 In Portugal, interest in determining blood groups for clinical purposes as well as in crossing serology with anthropology apparently grew exponentially, even independently and outside the purview of Mendes Correia's programme. This renewed interest in human blood for anthropological purposes was accompanied by a continuing attachment to Portugal's metropolitan white European colonial venture, both ideologically and practically. On the one hand, simultaneously, but also more or less independently from the Overseas Research Council's and Mendes Correia's jurisdiction, a number of colonial medical doctors in Angola, Mozambique, and Goamost notably Alexandre Sarmento and Germano Correia (whose work in Goa may have preceded serological research at Porto) 118 were undertaking their own surveys, determining (especially ABO) blood groups in distinct ethnic groups with a view to racial classification and/or the drawing up of 'anthropological charts'. 119 On the other hand, by the late 1950s, interest in blood-group racial research bifurcated into two principal expressions. Mendes Correia's followersmost prominently António de Almeida and his collaborators at the Centre of Anthropobiology Studies in Lisboncontinued the study of blood with a view to producing discrete racialized biological types until well into the 1980s. 120 At the same time, some Portuguese scholars developed a special interest in the study of mestiços, inspired by Freyre's Luso-tropicalism.
The Luso-tropicalist orientation in Portuguese human biology was championed by Dr Almerindo Lessa from the late 1950s onwards. Lessa was a central protagonist of haematology and racial serology research at home and in the colonies, having organized the IV International Congress on Blood in Lisbon in 1951. 121 By the late 1950s, however, he embraced positive views of miscegenation and became an avowed advocate of Freyre's visions in the field of human biology in Lisbon. 122 Lessa could argue in favour of blood as a marker of individuality and individual 'race', yet he also viewed racial mixing favourably, for example, in the localities of Cape Verde and Macau. 123 In 1956, sponsored by the state, Lessa led a sero-anthropological mission to Cape Verde. In the following years, in the context of a nascent Luso-tropicalist human sciences hub at the Centre of Political and Social Studies in Lisbon, led by Adriano Moreira, Lessa expanded his serological researches on 'Portuguese mestiços', with a view to demonstrating the biological excellence of the 'mestiço luso-tropical' [luso-tropical mestizo] idealized by Gilberto Freyre. 124 Lessa's positioning in fact expressed a Portuguese Luso-tropicalist trend in the fields of social and political sciences (including an emerging cultural anthropology) that accepted more openly racial mixing and did not view it as a negative phenomenon.
Thus, despite the ambivalence of one of the main advocates of serology, Mendes Correia, as the decade of the 1950s began, sero-anthropology prospered by directing its gaze away from Europe as a science that sought out pure-race indígenas in the Portuguese colonial world, as much it had prospered as a science of pure-race white Europeans in the metropole in an earlier period. For a moment, racial purity seemed to be consolidated as a programme, a shared obsession, of all those who took human blood as a pathway to raciological analysis. For many, it appeared to resolve the fundamental question of the purity of the Portuguese 'at home' and the integrity of the 'native races' in the colonial empire. Yet, as just our brief overview of Almerindo Lessa's work above reveals, this 'consensus', if that is what it was, would fast wane. Until then, it was no wonder that Mendes Correia, converted warily to the promises of blood-group classification, would speak so authoritatively at the 1951 congress to praise his own contribution to the occasion as the outcome of the 'heroic times' of racial, and racist, serology in Portugal that stretched back in the now distantbut still resonantpast.