Matskási István (szerk.): A Magyar Természettudományi Múzeum évkönyve 85. (Budapest 1993)

Finnegan, M., Tóth, T. , Ferencz, M. , Fóthi, E. ; Pap, I.: Biological distance during the Avar period based on non-metric cranial data

182 M. Finnegan, I T. Tóth 1

. M. Ferencz, E. Fóthi Sc 1. Pap sively elaborated, but is here represented by 43 unsexed individuals. Three cemetery samples under current elaboration by PAP were also studied from this region: the Tiszavasvári-Béke Tsz cemetery represented by 9 unsexed individuals, the Tiszavasvári-Zöldmező cemetery represented by 7 individuals and the TiszavasváriKoldusdomb cemetery represented by 4 individuals, all from the 7th century Avar period. The Avar cemetery from Toponár (elaborated by WENGER 1974) is represented our sample by 14 male, 22 female and 3 unsexed crania, all rom the Avar period. The final sample comes from the Solymár cemetery which has been elaborated elsewhere by FERENCZ (1983) and is represented in our sample by 16 male, 9 female and 8 unsexed crania derived from the 7th and 8th century of the Avar period. The number of sexed remains in this total sample is relatively balanced with 207 males and 189 females. The majority of the unsexed crania used in this study come from the various Tiszavasvári cemeteries as noted above. Each cranium in each sample was scored for the 42 cranial non-metric traits following the methodology of FINNEGAN (1972) and FINNEGAN & MARCSIK (1979,1989a, 1989b). One of us (MF) was responsible for all of the scoring of these traits in the order to eliminate interobserver error (FINNEGAN & RUBISON 1980a) and periodic rescoring of some crania was used to control the extent of intraobserver error on a day to day basis (MOLTO 1979). Seven of the traits in this battery were midline traits while the remaining 35 traits had the possibility of bilateral expression and therefore the maximum possible number of sides rather than crania is used as the sample size for these traits. Trait descriptions and examples of expression are presented in FINNEGAN (1972) and FINNEGAN & MARCSIK (1979). As well, the general statistical procedure is presented in FINNEGAN (1972), FINNEGAN (1978) and FINNEGAN & COOPRIDER (1978). Because of the small sample size for samples from cemetery Tiszavasvári-Béke Tsz, Tiszavasvári-Zöldmező, Tiszavasvári-Koldusdomb and at the fact the that they are of the same time period and closely associated geographically, it was felt that these three samples could either be coalesced into one sample (Tiszavasvári Group) or would have to be eliminated from further analysis due to the low number of individuals represented in each of the samples. We decided to coalesce these samples and thus have a resultant group, Tiszavasvári Group, represented by 20 unsexed. RESULTS Summary demographic data for the original 13 population samples are presented in Table 1. Because of small sample sizes, samples from Tiszavasvári-Béke Tsz, Tiszavasvári-Zöldmező, Tiszavasvári-Koldusdomb cemeteries were coalesced into one samples, for the remainder of the analysis. The frequency for each of the 42 traits in each of the resultant 11 samples is presented in Table 2. Due to breakage, incomplete, or obscure trait areas, not each cranium could be scored for all 42 traits. As such, the sample size per trait in a given sample may vary greatly, but the actual sample size is presented in Table 3. The minimum and maximum frequency, as well as the range of frequency for each trait over the samples collected at the Natural History Museum, are presented in Table 4. Side asymmetry Chi square values (derived from the theta chi square statistic, FINNEGAN 1972) were generated for the 35 bilateral traits used in this study of 11 population samples, in testing left to right side differences. Male and female samples were run separately so that any sex dimorphism in raw frequencies would be eliminated. Nine of the 11 samples (where sex was ascertained) were utilized in this analysis. With this number of traits over this number of samples, we would expect 15.5 significant differences at the .01 level of confidence. We found 20 significant differences at the .01 level does not exceed chance expectation and as such, we will not further consider that level here. However, 20 comparisons were significant at or above the .05 level