Daggers, awls, and needles were made of bone; also combs apparently used for stretching sinews into long threads. The flint implements are rudely chipped, never polished. We find long flakes used as knives, and numerous scrapers and borers.[32] The axe, if we may call it so, was of peculiar form, approaching the triangular and looking as if made out of a circular disk of flint by breaking away two sides of the periphery, leaving a somewhat flaring cutting edge. The middle was thick, the edge tapered somewhat rapidly, making a rough but quite durable instrument. Longer implements in the form of chisels or picks were also roughly flaked with skilfully retouched edges, often with one end narrowed or bluntly pointed. In all cases the work is very rude compared with the best specimens of Paleolithic time. Arrow-heads are common, usually with a broad edge instead of a point, well suited to killing birds and small mammals. The bone harpoon seems to have gone out of use. The pottery is thick, heavy, crude, with practically no ornament, except finger-prints around the upper edge. The jars are sometimes of large size; often the base is pointed instead of flat or rounded. Hearths of calcined stones are abundant. Sometimes these are surrounded by circular depressions in the heaps, which may mark the form and position of huts or shelters; or these may have been placed under the lee of the near-by forests. No graves or human remains of this period have been found. Shell-heaps quite similar to those of Denmark were discovered at Mugem, in Portugal, in the valley of the Tagus, twenty-five to thirty metres above sea-level, and thirty to forty miles from the mouth of the river. The shells are of marine origin, and indicate a considerable elevation of land since their accumulation. The stone implements are very primitive and of Azilian-Tardenoisian type. Large flat stones, perhaps for grinding, perhaps for dressing skins, occur. Pottery occurs only in the upper layers, where the bones of mammals increase in number. There are no polished implements, no traces of domesticated animals, not even of the tame dog. Graves were found here and there; and while the skulls were badly contorted, they seemed to show that the inhabitants were partly long-headed, partly broad-heads. Remains, apparently of the same age, have been found in Great Britain. Even the Danish shell-heaps are not all of the same age. According to Forrer, Havno is ancient; Ertebolle is also old, but was long inhabited, and some of its uppermost layers may be full Neolithic; Aalborg and others are younger. Mugem strikes us as more ancient than the similar Danish remains. Other remains near the Baltic suggest very strongly quite marked differences in age or in the culture of their inhabitants, or in both these respects. We can notice only two of these.