Since the second half of the 19th century that researchers had a special interest in the topic of the last hunter-gatherers in Portugal. The main reason was the discovery of large shellmiddens sites in the Muge area. Carlos Ribeiro found the sites of Arneiro-do- Roquete (also known as Ribeira da Sardinha) and Cabeço da Arruda as early as 1863 (Cardoso e Rolão, 1999-2000). In that same year, Ribeiro started excavations in Cabeço da Arruda and at the same time was able to locate other sites, such as Moita do Sebastião, Cabeço da Amoreira e Fonte do Padre Pedro (Cardoso and Rolão, 1999-2000; Rolão, 1999). The first Muge results are published in 1865 by Pereira da Costa in collaboration with Ribeiro, refering the existence of a burial ground with at least 45 individuals, as well as an important collection of lithic artifacts, bone tools and fauna. In the 80’s, Ribeiro and Nery Delgado reevaluate the area and state that more sites should be present. In 1884, Ribeiro documents the presence of over 120 individuals present in Cabeço da Arruda and Moita do Sebastião. In 1884 and 1885, by request of Nery Delgado, Francisco Paula Oliveira resumes the work with the excavation of Moita do Sebastião, Cabeço da Arruda and Cabeço da Amoreira retrieving more than 50 skeletons. The Muge complex is left alone for 5 decades. In early 30’s, Mendes Corrêa goes back to Cabeço da Amoreira (Mendes Corrêa, 1933; Cardoso e Rolão, 1999-2000). During his excavations about a dozen more burials were found, as well as a series of habitat structures. In the summer and fall of 1937, there is a new intervention in Cabeço da Arruda, where they recover between 12 and 14 more skeletons (Rolão, 1999). Twenty years later, Muge was again the centre of attentions. O. Veiga Ferreira and Jean Roche restart the work in 1952, lasting more than 10 years (Roche and Veiga Ferreira, 1967). There are excavations in Moita do Sebastião, Cabeço da Amoreira and Cabeço da Arrudda, clearly the most important sites in the Muge area. In Moita do Sebastião, another 33 burials were found with possibly as many as 44 individuals (Newell, et al, 1979). In the base of the shellmidden, Roche and Veiga Ferreira excavated a series of structures, such as hearths, post holes, and storage pits (Roche, 1954 and 1972). In Cabeço da Amoreira, Roche uncovers close to 40 archaeological layers divided in 3 main occupational phases of the site. The result seems to have been the recovery of at least 17 more burials. The work in Cabeço da Arruda took place in 1964-65, with the recovery of more than 10 skeletons and close to 90 layers (Roche, 1972). More recent archaeological work was carried out by José Rolão. This work has taken place in 2000 with excavations in Cabeço da Arruda e Cabeço da Amoreira, and site preservation at Cabeço dos Morros. Finally, Bicho was funded by FCT in 2006 and survey and excavation took place, mostly at Amoreira, producing new results and new middens. Although there is a long history of research in Muge, the results are still scarce and, frequently, equivocal. Most papers and published results have focused on detailed descriptions of lithic artifacts and fauna.
In the last case, the fauna is reduced a list of species, with the exception of the work of Lentacker (1986) and the doctoral research of Cleia Detry (2007). In any case, it is clear that a large diversity of fauna, both terrestrial and aquatic (fish and shellfish) were present and made up the human diet during the Atlantic period in Muge. In the mid 80’s, David Lubell and Mary Jackes started a program of isotopic analysis of the human skeletons with the aim of studying the diet of those populations (e.g., Jackes, 1988; Lubell et al., 1994). The same work has more recently been followed by Umbelino in her doctoral dissertation (2006). The initial results showed that the diets had mixed components, with both marine and terrestrial origin. But the results obtained by Umbelino are more interesting and more complex. Not only the trace element and isotopic analyses have corroborated the earlier results, but have also shown that the diet was made up of an important segment of terrestrial animals and of plants and fruits. But the complexity of the results did not end there. In fact, since Umbelino’s sample, though not as large as one would want it to be, was larger than that from Lubell and Jackes, the results show that there is an important variability in the diet from site to site as well as within each site. This last element suggests thus, that there was some diversity in the access to the different foods available to the community. In addition, new data resulting from a recent project coordinated by the present PI showed that site location in the Muge valley is organized so every major midden site (5 old and 3 new ones found by the team of the above mentioned project) are almost at the same distance on both sides of the valley (Gonçalves, 2009). This fact, suggests a division of space based on emerging social complexity with clans, lineages and moities divisions (Bicho in press), similar to that known for the Anazasi of Chaco Canyon (Van Dike, 1999; Vivian, 1970). New analysis of 1950’s data (Roche, 1972) show that the Moita do Sebastião necropolis is marked spatial organized with two areas, one with burials with goods and the other without any artifacts or faunas associated to the human skeletons. Also, in the first, there seems to be an internal division between burials with special shells and one with special artifacts (Bicho, 2009b). This type of spatial patterning is very similar to another well known Mesolithic case, that of the Oleni Ostrov, in the Baltic Sea (Jacobs, 1995; Price and Jacobs, 1991), reflecting social complexity patterns.