Аннотация:Northern Asian populations are usually considered as a model for inferring fossil hominin adaptation to the glacial conditions of the Ice Age of Europe. But what is a suitable climatic model and what is a suitable human model? The climate of Northern Asia in winter is much colder but much drier than that of Northern Europe. The mean January temperature in Northern Karelia (64-66° North) is -12°C and mean precipitation is as high as 25mm in that month. In the area surrounding Lake Baikal (52-58°N) the same climatic parameters are -25-30°C and 5-6mm, respectively.Many previous studies have shown that humidity and precipitation are as relevant as air temperature in driving adaptation and may drive it in different ways. Thus it is important to compare possible adaptive changes in the mid-face in populations of North East Europe where climate is both cold and relatively humid with those of Northern Asia. Here we sampled nine populations from North East Europe: Finns, Karels (two samples), Komi, Letts, Mordvins, Russians (two samples) and Northern Ukrainians. For comparison eight groups from warmer regions of Europe were sampled: Armenians, Bulgarians, Italians, Norse, Ossetians, Romanians, Shapsugs and Turks. All the populations represent similar “Caucasoid” facial morphology but strongly differ in terms of language and origin.The total sample included 411 skulls. A set of 24 linear measurements was compiled to sample the different functional components of the facial skeleton. Nine environmental covariates were employed including air temperature, precipitation, vapor pressure and cloud fraction. Two-block partial least squares (PLS) analysis was used as an exploratory technique and univariate Spearman correlations and Mantel tests were used to assess the strength and significance of associations between climate and morphology. To compare patterns of association between North Europe and North Asia the same analyses were carried out using previously published data. The PLS revealed that overall squared covariance between form and climate was approximately twice as high among Asians (21.45%) as among Europeans (9.6%). Further, among Asians the matrix of morphological Euclidean distances among the populations was significantly correlated with the matrix based on all 9 climatic variables (0.66, p=0,03). This was not the case among Europeans and only matrices based on 14 measurements and the 6 climatic variables with the highest PLS loadings were significantly correlated (0.49, p=0.0012). That association is apparently driven by nine North East European groups. Loadings of the climatic variables on SV1 were generally similar for both Europeans and Asian with the biggest differences in the loadings of measures of precipitation and cloud cover. The presumably cold-adapted populations in both Europe and Asia tend to have a slightly wider piriform aperture in combination with a slightly narrower anterior part of the nasal cavity, broader and antero-posteriorly longer maxilla (and thus longer nasal cavity) compared to the groups from milder climate. But in contrast to the cold climate Asian populations, North East Europeans demonstrate a strong reduction in the length and protrusion of nasal bones as well as in facial protrusion in general and reduction of the piriform aperture and orbital heights. These differences could be either a result of adaptation to different climatic conditions or of marked differences in craniofacial morphology between Europeans and Asians. For instance, the facial skeleton in East andNorth Asian populations is generally very flat and thus it is difficult to expect it to get even flatter. But other features such as reduction of height of the nasal aperture in North East Europeans can hardly be explained by general morphological pattern alone. We discuss our findings in the context of published data on craniofacial morphology of Upper Palaeolithic H. sapiens.