The dead village sites are much alike along the whole river. They are generally located at the mouth, of some inland stream that carries clear fresh water, particularly if on the other side there is the protection of a hill. The dwellings were invariably on a flat and were throughout semisubterranean and of the same general type; which applies also to the larger communal houses or "cashims." The sites can often be told from afar in summer by the rich grass that covers them. [Illustration: FIGURE 4.--The Yukon from below Koyukuk to Lofkas] The burials were as a rule not far from a village and preferably on the slopes of the nearest hill. They were mostly above ground, but under the influence of Russians there were also shallow-ground burials. The latter can readily be told by the sawed planks of the coffins and the iron nails by which they are fastened. In many places no surface burials remain or there are mere traces. In such, places little mounds may betray old burials assimilated by the tundra. Trenching in likely spots would doubtless reveal others of which no trace remains on the surface. [Illustration: FIGURE 5.--Old map of the Nulato district] No excavations of any of these sites have ever been attempted, but many of the surface burials were disturbed or destroyed by seekers of relics and the curious vandal, who is present on the Yukon as in other parts of the country. [Illustration: FIGURE 6.--Map of Kaltag and vicinity. (By McLeod)] The maps shown here were made under my direction on the basis of maps and charts provided by the Geological and Geodetic Surveys, in Washington. Additional old sites will doubtless be located in the future and may be added to these records. PRE-RUSSIAN SITES [Illustration: FIGURE 7.--The Yukon from Bystraia to below Holy Cross] As already told in the Narrative, a search for truly ancient sites along the Yukon has proven largely negative. A more intense and prolonged archeological survey, with exploratory trenches wherever there is promise, may one day prove more fruitful. But, as pointed out before, much can never be expected. Man could at no time have occupied the Yukon Valley and watershed in large numbers. He would not have found enough sustenance. Even with fair resources he would hardly have tarried in these inclement regions as long as the ways toward the south were open. He never built here of lasting materials and had little chance to develop or even keep up any higher culture, and since he is gone the ever-cutting river has taken away whatever it could reach and scattered it through its silts and gravels. There is nevertheless a number of small elevated plateaus along the right bank that ought to be sounded by exploratory pits or trenches, particularly perhaps where there are traces of later habitations. [Illustration: FIGURE 8.--The Yukon from above Holy Cross to below Mountain Village] [Illustration: FIGURE 9.--The Yukon from below Mountain Village to near Marshall]