Abstract: One hundred and seventy one pieces of stone artifacts which were unearthed from the sand-gravel layer of 54:90 site near Dingcun village in 1954 have been studied here.
A. The general characters of the artifacts
The details of the observation, classification, measurements and statistics of the stone artifacts are shown in the text and tables 1-4. The general characters of the artifacts could be recognized as follows:
1. Most artifacts are large in size. On the basis of the measurements of the undamaged flakes, the large flakes make up 64.79% of the total and small tools are very rare (only 6.67% of total tools).
2. The flakes were mainly produced by direct percussion, occasionally by other flaking methods, such as throwing method, colliding anvil process etc. Though the cores were not prepared before producing the flakes, more flakes were regular in shape. It indicates that the "craftsman" who applied the primitive technique for getting flakes was more proficient.
3. The used flakes without retouch are more common and make up 17.71% of the total flakes.
4. Blanks of the tools have several kinds. The flakes constitute 76.66% of the total blanks. Tools made on the core are of less number (23.34%) (details sce table 4).
5. The tool type is simple and the tools include scrapers, chopping tools and pointed tools. The tools with two and more edges have higher percentage (70%) and the single edge tools are not so much (30%).
6. Tools were trimmed by direct percussion. The complex mode is the main way for retouching tools. Most of them were irregular in shape and the edges usually were blunt.
B. Industrial comparison
Because the artifacts from other sites in Dingcun region, except those from 54:100 site have not been studied in detail, the present author limited himself in comparing the component from 54:90 site with that from 54:100 site near Dingcun. The stone artifacts from these two sites have similarities and obvious differences. The similarities are: 1) The direct percussion is the most important method for producing flakes; 2) Though the cores are not prepared before chipping the flakes, most flakes are more regular in shape; 3) The tools are usually made by trimming the flakes (80.82% in 54:90 site and 71.8% in 54:100 site); 4) To trim tools is by direct percussion without exception. Retouching on the dorsal surface is the important mode for making tools.
The most important differences between the stone artifacts from the two sites are: 1) Most artifacts from 54:90 site are large; In 54:100 site the small stone artifacts have higher percentage: the small stone flakes make up 64.28% of the total flakes and the small tools make up 40.02%; 2) The used flakes from 54:90 site (17.71%) appear much more frequent than those from 54:100 site (8.46%); 3) The tools of two sites are obviously different in type and mode of manufacturing. In 54:90 site the chopping tools occupy higher proportion (36.67%), while no chopping tools were found in 54:100 site. The tools manufactured by complex mode hold large part in the component from 54:90 site while the tools trimmed on the dorsal surface occupy higher proportion (48.7%) in 54:100 site.
Besides, the tools found in two sites also exist other less important differences: 1) No graver were found in 54:90 site; 2) The pointed tools of B type are not only more in number but also better trimmed than that (just only one) from 54:100 site; 3) The tools with two and more edges from 54:90 site constitute the great majority (70%) and the single edge tools only make up smaller percentage (30%), while in 54:100 site they are 38.46% and 61.51% respectively.
Judging from the above mentioned similarities the assemblages from 54:90 site and 54:100 site should belong to a same culture tradition, but due to existing several differences of the artifacts from these two sites, we consider that they could represent different groups.

Key words: Paleolith; Dingcun; 54:90 site