M. Járó - L. Költő szerk.: Archaeometrical research in Hungary (Budapest, 1988)

Analysis - BARTOSIEWICZ László: Water-sieving experiment at örménykút, site 54

The results of these calculations are summarized in Table 2. As may be seen from the coefficients of correlation, the logarithmic transformation increased the accuracy of curve fitting so that statistically significant linear relationships could be evaluated on the at least 5% level of probability. The correlation coefficients are high by the standards set by Guilford (1956). All of them are negative, which is a consequence of the fact that the number of fragments increases with the decrease of size categories. This latter, however, represents increased cost, that is labour/energy investment. The same interpretation holds true for the "negative " exponents obtained by the calculation. Naturally, the mean size of bone splinters is largest in the case of hand collection (x = 323 cm, SE = ± 0.04 cm). The exponent of the curve fitted to the data set, on the other hand, is the lowest. In the case of water-sieving using a 2.5 mm mesh the mean size declines (X = 226 cm, SE = ± 0.40 cm) but the exponent shows that the increase in the number of bones is highest per refinement unit (05 cm). The use of a 0.8 mm mesh further decreases the mean value (x = 1.06 cm, SE = ± 0.02 cm), however, refinement in mis size interval no longer results in a spectacular increase in the number of splinters recovered. Thus from a purely quantitative point of view, energy investment is best returned by water-sieving with 25 mm mesh size. Within the 65 to 05 cm size interval studied in this paper no intersection of the three exponential functions could be observed. Conclusions The beneficial effect of water-sieving on the efficiency of recovery has unambiguously been demonstrated. A few biasing factors as well as difficulties in practical application, however, must briefly be reviewed. As far as the experiment is concerned, although bone splinters served as perfect standardized subjects for measurement, they do not act in every respect as a number of others of similar size would. In addition, pieces in the lower size ranges may be slightly overrepresented due to secondary fragmentation during the course of the water-sieving process itself. A major cause of resistence against the widespread use of this technique in Hungary is related to reservations concerning its labour efficiency. In fact, with reasonable sampling methods and the application of a multi-purpose flotation machine, the energy investment may be expected to be fully returned. As was shown by the example of the Neolithic excavations at Öcsöd — Kováshalom, although flotation/water sieving practically would have required a full time technician (Rasson, personal communication), the quality and quantity of small finds recovered was remarkable. After all, while this study allowed a quantitative approach to be made to the problem, the main purpose of water-sieving is the recovery of a qualitatively different group of artifacts which would not en masse be available by hand collection as the only way of recovery. This fact, however, creates an other aspect to the problem of labour efficiency: while the introduction and use of this technique may be solved by unskilled personnel trained at the site, the evaluation of finds requires increased time input from a number of experts.-These include archaeozoologists specialized in the identification of various taxa, and botanists as well as archaeologists who are interested in small artifacts or even by-products of certain craft activities. At this point, however, the necessity of water-sieving is not a matter of labour efficiency but a question of reliable and detailed documentation versus information loss.