The physical and optical properties of an atmospheric aerosol mixture depend on a number of factors. The relative humidity influences the size of hydroscopic particles and the effective radius of an aerosol mixture. In consequence, values of the aerosol extinction, the aerosol optical thickness and theÅngström coefficient are modified. A similar effect is observed when the aerosol composition changes. A higher content of small aerosol particles causes the effective radius of an aerosol mixture to decrease and theÅngström coefficient to increase. Both effects are analysed in this paper. The parameters of the size distribution and the type of components used to represent natural atmospheric aerosol mixtures are based on experimental data. The main components are sea-salts (SSA), anthropogenic salts (WS, e.g. NH 4 HSO 4 , NH 4 NO 3 , (NH 4 ) 2 SO 4

Abstract. Abed F, Bachir-Bouiadjra B, Dahloum L, Yakubu A, Haddad A, Homrani A. 2021. Procruste analysis of forewing shape in two endemic honeybee subspecies Apis mellifera intermissa and A. m. sahariensis from the Northwest of Algeria. Biodiversitas 22: 154-164. Honey bees play an important role as pollinators of many crops. Thus they are collectively considered as a veritable economic source. The present study was undertaken to describe variation in the right forewing geometry in two Algerian honeybee subspecies Apis mellifera intermissa and Apis mellifera sahariensis using landmark-based geometric morphometrics. A total of 1286 honeybees were sampled from 12 provinces in the northwest of Algeria. The forewing geometry was evaluated using 20 homologous landmarks by applying Procrustes superimposition analysis. The top four principal components accounted for only 41.1% of wing shape variation between the two subspecies. There was a significant difference in wing shape between the two subspecies (Mahalanobis distance = 1.0626 ; P<0.001), whereas their wing size seemed similar (P>0.05). Regarding the allometric effect, the percentage of variation in wing shape explained by size changes was relatively small, with 1.28% and 4.37% for A. m. intermissa and A.m sahariensis, respectively. The cross-validation procedure correctly classified 68.3% of specimens into their original groups. PERMANOVA test revealed significant differences in the right forewing shape among all geographic areas studied (P<0.001). The results clearly showed that the landmark-based geometric approach applied to forewings venation is a powerful and reliable tool in the discrimination of native honey bee subspecies and should be considered in local honey bee biodiversity improvement and conservation initiatives.