 So I'm going to talk a bit about a site that Sebastian Lacombe and I have been excavating for the past several years. And we've, interestingly, bridged the Atlantic. He's French. I'm American. We're married. The site of Père Blanc was discovered in the context of a long-term regional survey project started by a megconky in the early 1990s called Between the Caves. The archaeology of Paleolithic Western Europe was then, and still is, dominated by the archaeology of caves. Conky set out to demonstrate that long-term place to see human presence in the open air was visible if only we looked for it. She and her team used geological models and pedestrian survey of cloud fields to collect lithics in a zone between two cave sites in the central French Pyrenees, Marsoulis and Le Mans both of which are known for visual imagery. At the start of the project, Conky met with a great deal of skepticism. She would certainly find things in the fields since there's quite a lot of naturally occurring chert. But would it be possible to say anything systematic about the finds? Nevertheless, she received a high-risk project grant from the National Science Foundation in the United States. Sebastian joined the project as a graduate student at Université Toulouse de Meray, and I joined a few years later from Berkeley. Sebastian in particular was able to characterize a large proportion of the artifacts based on technological and petrographic traits that when combined with the geological models and observations from the survey revealed patterns in landscape use. Our particular interest was in upper Paleolithic use of the landscape. So after many years of survey, we began to excavate test units to understand the relationship between the surface finds and the possibility of preserved archeological levels underground. The early test units did not result in intact upper Paleolithic levels, but they did reinforce some of the geological observations we made, particularly reinforcing an upper Paleolithic preference for sandy sediments and higher relative elevations. All the while, we continued to survey plowed fields, especially fields that had not been surveyed before and those that were associated with sandy sediments. At one point, we followed a trail into the woods where we found a plowed field filled with upper Paleolithic objects. We continued to follow that trail uphill and found that on top of a crest in a spot called Père-au-Blanc, there were many lithic artifacts eroding out of the trail. The sandy trail was regularly disturbed by the passage of hikers, horseback riders, and motorcycles, but the sediments next to the trail were considerably less disturbed. Most of the lithics were flakes, but there were enough diagnostic objects such as end scrapers, burins, blade lids, and blades that we were able to determine that there was an upper Paleolithic presence and likely a Magdalenean presence. This time, when we excavated test units, we found that there was what appeared to be a single intact and very rich archaeological level, and we arranged to open broader horizontal exposure the following year. Over the next several years, we continually broaden the excavation. The open-air setting meant that there were some challenges that are not as problematic in caves and rock shelters, most notably preservation. And you have the two pieces of non-lithic tool that we have from the entire site on the left there, and you can see how big they are. And then other faunal remains, I found a fish vertebrae this summer in the wet screening, just to give you an idea of what we're dealing with. The sediment is very acidic, and due to the low level of sedimentation, the archaeological level is not deeply buried, so the organics are almost completely absent and we rely very heavily on stone to understand the site. The assemblage mostly consists of flaked chert, but the people of Père Blanc used other stone in intriguing ways. One example is this quartzite anvil that was deliberately modified in a way that does not affect its function, but reinforces its resemblance to a horse. So you can see, maybe not so well, but there are some flakes removed where the arrows are pointing, and that is on the face of the anvil that was not used. The opposite face does show signs of impacts, but it lines up very nicely with the contour de coupes of horses that are really common in the central Pyrenees at this time. Another more dramatic example of the use of stone is this structure, which is built of hundreds of sandstones of five different types, some of which occur naturally in the immediate surroundings, but others of which had to be carried uphill to the site. At present, we don't see any signs that these stones were modified, but the arrangement of stones is clearly patterned. The napped lithics provide us with a variety of information. The first surface finds indicated that the site dated to the upper Paleolithic, given the bladelets, bladelet cores, and burins. The court technology in particular suggested a late Paleolithic occupation compatible with the Magdalenian. As we continue to excavate, we reinforce the fact that there was just one archeological level, but it did not resemble the assemblages of Magdalenian cave sites in the central Pyrenees. While this assemblage had typical Magdalenian objects, it also contained tools often associated with earlier periods, such as large sidescrapers and carinated cores, as well as tools more common to later periods, such as triangles. This is where we ran into the problem of the tyranny of the typological approach. Regional typologies in France are well-established based on cave assemblages, and these typologies provide a very good shortcut in determining what period a site dates to before you can get a chronometric date. The assemblage at Pereblanc, taken together, did not correspond to the typology of Magdalenian cave sites in the central Pyrenees, or even more broadly in southwestern France. However, everything about the assemblage and the context told us that we had a single archeological layer. Fabric analysis, micromorphology, and the sharp edges and erases of the lithics told us that the objects were essentially in place. The thickness of the archeological level, the raw materials, and the technology were all consistent with a single component. However, we were constantly questioned about our identification of the site as Magdalenian. Surely the presence of large sidescrapers on flakes indicated a middle paleolithic presence. With care-nated cores, there must have been an ordnation occupation, or shouldn't the number of microliths indicate mesolithic use of the site? Because of the state of organic preservation of Pereblanc, radiocarbon dating has not yet been possible. We sought out a date through optically stimulated luminescence dating and obtained a date of about 17,000 years, which was more or less consistent with our determination that the site dated to the early middle Magdalenian shortly after the last glacial maximum. We think that this date is in fact a bit young due to the low rate of sedimentation on the site. We also sought out similar assemblages elsewhere and found the most convincing analogs in cave sites in Cantabria, and these sites also dated to the early middle Magdalenian. I'd like to turn to the assemblage for a bit now and describe and highlight some of the more interesting aspects of it. One of the first things to note is the high percentage of objects made of local raw material. The site is located approximately 300 meters up slope from a good quality lithic source, and nearly 80% of the raw material is from this source, with another 10% or so coming from within 40 kilometers away. The most distant materials, the remaining 10%, come from the Béron and Dordogne, about 150 and 200 kilometers away, respectively. This is completely different from middle Magdalenian cave assemblages in the central Pyrenees, where the majority of raw materials, around 65% at a site like on land, for example, typically come from distant sources. Another distinctive aspect of the lithic assemblage at Père Blanc is the presence of the complete chenneau brottoir, compared to cave sites that are usually lacking the early stages of core reduction, even in the case of local raw materials. This is clear in the case of the chert collected from the nearby source, but also for other types of chert collected from the broader region. The most distant material, however, is only represented by tools or blades and bladelets with a few occurrences of bladelet cores, a practice that will be reproduced during the subsequent phases of the regional Magdalenian. Looking at the cores specifically, the 73 cores currently in the assemblage show some interesting traits. They're largely dominated by bladelet cores at about 44%, followed by blade cores and flake cores in roughly equal proportions. If the dominant presence of laminar cores in the broad sense is the usual characteristic of the Magdalenian lithic assemblages, the high proportion of flake cores at Père Blanc is quite unusual and signals the clear intention to produce flakes in various shapes and sizes that the discoidal core type is the most frequent. This intention translates into the tool assemblage as we'll see in a moment. Within the bladelet cores, we can identify at least four distinct chenneau brottoir. These are linked to the production of wide, thin and moderately curved lengths, or others that are narrow and thicker depending on the architecture of the core. One specific type is the so-called carinated core that yields thin and slightly twisted bladelets. A similar diversity exists within the blade cores where we see three types of chenneau brottoir but with less variation in the products. This diversity distinguishes Père Blanc from the regional Magdalenian cave sites in which the laminar norm is aimed at the production of narrow and thicker blanks as mentioned above. We can also see differing levels of expertise in core preparation, in part because we have the entire chenneau brottoir represented. The dominance of laminar cores is echoed in the tool assemblage where bladelets and blades constitute the most frequent types of blanks for making retouched implements. The assemblage in general is dominated by microliths comprised of an almost equal number of retouched bladelets and triangles. If the former are not always an essential component in the regional Magdalenian tool kits, the latter definitely are not and provide further evidence to place Père Blanc at an earlier phase of the middle Magdalenian. Blades are the most frequent blanks used to craft various types of burins, usually dihedral and end scrapers. Along with the retouched blades, some drills and a few composite tools are present and they constitute the bulk of the tool kit commonly found in the Magdalenian. These implements are often linked to the processing of various materials such as antlers, bones, wood, meat, hides and probably plants, none of which is preserved. Another unusual aspect of the lithic assemblage is to have a significant proportion of so-called archaic tools such as side scrapers, notches, piècesquiet and denticulates, as well as a high number of retouched flakes. Almost exclusively crafted on deliberately produced flakes rather than the byproducts of blade production and particularly on cortical flakes, these tools are rarely found in such abundance in Magdalenian cave sites. This is especially true of the side scrapers that constitute a very particular type of tool with pair of long. In the context of the site and based on use wear analyses and comparisons to other assemblages, these side scrapers are likely associated with the processing of fresh hides. Beyond this, the lithic assemblage is also indicative of another activity that is seldom described in cave sites. The collecting and processing of natural pigments, mostly iron and manganese oxides and with colors ranging from yellow, orange, red and dark purple. Besides the presence of the pigments themselves, this activity is visible through the presence of massive hammerstones and anvils, both whole and fragmented, as well as through the existence of many diverse implements, including blades, bladelets and flakes, retouched or not, that bear clear traces of pigments and adhesive on various edges or on the active ends. The discovery of the site of pair of long resulted from a desire to expand our knowledge of human presence outside of caves. So it should come as no surprise that people engaged in different activities in the open air than inside caves and rock shelters. The fact that our assemblage is different from those in cave sites does not mean that it was produced by succession of different archeological cultures. In fact, if caves were not the preferred sites for everyday life and shelter, but more often used for special purposes, a site like Pereblanc is probably more typical of everyday life in the Magdalenaean and the prehistory of caves that forms the vast majority of our understanding is really just one chapter of a larger book. Thank you.