Lying between these various population clusters were stretches of uninhabited or lightly occupied land. In possibly one or two cases, intervening areas may have supported populations almost as concentrated as Poverty Point territories but, for various reasons, these peoples did not participate regularly or intensively in Poverty Point culture. Our map of 1000 B.C. shows another interesting feature. The scattered Poverty Point population clusters were all linked by waterways. Every one was tied to the Mississippi River. Even though the Mississippi River did not run through every concentration, its major tributaries and distributaries did. These interconnected streams must have been the highways that carried people, trade goods, and ideas. Most of the population lived in permanent villages along these streams. There were small, medium, and large villages, ranging in size from less than an acre to over 100 acres. The smallest settlements probably housed only a few families, while residents at some of the larger ones must have numbered in the hundreds, possibly even more. One site among them was a veritable metropolis for the day; the population at the Poverty Point site itself has been estimated to number several thousands (Ford and Webb 1956; Gibson 1973). In addition to these stable villages, there were temporary campsites, where villagers evidently took advantage of seasonally available foods and other resources. Larger villages were often distinguished from smaller ones by more than population numbers. One or more villages in nearly every Poverty Point territory were set apart by public construction works, usually mounds and sometimes embankments. Mounds were made of dirt and were usually dome-shaped affairs constructed in several stages. Two unique mounds at the Poverty Point site have been identified as bird effigies (Ford 1955). Typically one mound stood at these villages, but two to eight mounds were present in some instances (Webb 1977:11-13). As a general rule, the number and size of these works varied directly with village size and population. Even though several of these mounds have been excavated, their purpose is still unclear. They superficially resemble mounds used as tombs by later cultures, but no burials have turned up in the Poverty Point structures. Beneath a mound of this type at the Poverty Point site was a bed of ashes and a burned human bone, suggesting that, at least in this example, it covered a cremation (Ford and Webb 1956:38). Embankments, or artificial ridges, were occasionally built at these bigger villages. In many cases, embankments seem to have been raised by a combination of construction and incidental accumulation of living refuse. Most of the giant ridges at Poverty Point seem to have grown this way (Ford and Webb 1956; Kuttruff 1975). However, not all of these ridges positively served as foundations for houses. Some served to connect mounds, others perhaps to mark alignments of some kind.