TOSSAL DE LA ROCA
EXPLOITATION OF NATURAL RESOURCES
The alimentary diet which as we see was based fundamentally on the consumption of lagomorphs and ibex, in addition to some partridges (SÁNCHEZ MARCO, A. 1995) during the Tardiglacial, is enriched with the advent of the Holocene. New species are now included such as chamois and wild boar and the importance of the deer increases considerably, at the same time as that of the rabbit and partridge diminishes. Few fish remains (eel, scorpion fish, trout and barbel) are witness to the existence of fishing as a food source, although this is a marginal element (MORALES, A. & ROSELLO, E., 1995).
This alimentary diet was completed with some plants, such as vetch (a type of Leguminosae) and Silene as well as fruits (acorns), the difficult conservation of which has meant that this vegetable consumption is only shown in the epipalaeolithic levels (ARNANZ, A. 1995).
On centering our attention on the most - hunted animals, such as the ungulates, we observe some significant differences, both in the age of the examples and in the animal parts shown. Thus, during the Magdalenian culture adult or sub-adult ibex were hunted. The importance of the remains of front and back limbs and the practical non-existence of remains of back legs indicates that the animal was not transported whole to the site, but only the parts richest in meat, rejecting due to their poor alimentary interest other portions such as the legs, which must have been left at the hunting site. There are few skull remains from which it is deduced that the brain was left behind or consumed at the time of hunting. There is also a moderate percentage of jaws, which implies their transportation to the site for later consumption.
During the Epipalaeolithic occupations sub-adult (about three years' old) deer were hunted specially and adult or sub-adult groups of ibex, although an important number of remains of young animals also exists. There are, therefore, certain differences in the hunting strategy compared with previous times also shown through the analysis of the different skeletal parts found. Thus, now, in addition to remains of front and back limbs being recorded, the remains of phalanxes are frequent and those of the axial part and the head increase. In short, and with the exception of feet, above all the back ones, all anatomical parts are shown. This indicates that the animal was transported practically intact to the site. A small variation exists within the Epipalaeolithic itself, for during the Epipalaeolithic with geometrics the presence of the animals' feet has been recorded, which would imply its transportation whole. However, there is no selective process at the time of hunting during these occupations with geometrics. The selection is slightly moderated in the first epipalaeolithic phases and specially marked during the Magdalenian. The absence of a selection process during the last occupations of the site (Epipalaeolithic with geometrics) is logical if we take into account the existence of greater stability during the Holocene occupations.
The study of the fragmentation of the bones has provided us with valuable information on the different uses of these food sources. In this sense, the state of fracture of the diaphyses in the Magdalenian levels has been interpreted as the result of marrow extraction, while the joint parts were crushed to extract the fat, which as also obtained in the phalanxes and the jaws after being beaten. The process observed in the Holocene levels appears to be very different, especially during the Epipalaeolithic with geometrics, where there are few fragments of diaphyses with direct fractures compared with a high percentage of those situated in joint parts. The explanation for this is not easy and perhaps the frequent bioturbations in the exterior sector make their reading more difficult (PÉREZ RIPOLL, M. & MARTÍNEZ VALLE, R.1995).
