The formation of late Pleistocene Yedoma in western Beringia (Siberia) is still widely debated. Moreover, different geological and cryostratigraphical views on Yedoma exist between researchers focusing on western or eastern Beringia (Alaska and Northwest Canada). These differences largely concern the prominence of the role of eolian processes. In particular, previous studies on Yedoma in the Yukon Territories and Alaska interpret these deposits as being largely loess or re-transported loess (muck). In contrast, several hypotheses have emerged over decades of research in the extensive Siberian Yedoma region, including (1) alluvial genesis, (2) ice-sheet-dammed basin sediments, (3) deltaic formation, (4) proluvial and slope deposits, (5) cryogenic-aeolian deposits, (6) nival deposits, and (7) polygenetic origins. Characteristics that most studies agree on include the dominance of large syngenetic ice wedges, mainly allochthonous silty to sandy sediment deposition in low-center polygons in combination with deposition of mainly autochthonous organic remnants from plants and animals, very harsh continental, glacial climate conditions. In terms of landscape and relief characteristics, various Yedoma types seem to exist across the extensive region where Yedoma does occur, ranging from spatially confined Yedoma valley fills including slopes to vast accumulation plains on Arctic lowlands and shelves. Accordingly, we here support the notion that Yedoma may have different depositional properties and genetic origins under a common frame of similar environmental and climatic conditions during the Late Pleistocene. This hypothesis is known as polygenetic formation of Yedoma. An important aspect of Yedoma is the dominating presence of excess ground ice. Ice wedges and segregated intra-sedimentary ice constitute the majority of this deposit by volume (50-80%) in most Yedoma regions and are thus one of the most critical factors in deposit genesis in contrast to accumulations of silty to sandy deposits in temperate regions. The Yedoma Ice Complex formation includes cryogenic processes such as cryogenic weathering, ice segregation, syngenetic ice wedge formation and growth, secondary sediment deformation and reworking due to ground ice, and cryosol formation (including phases dominated by orthels, turbels, or histels). All these processes were promoted by long-lasting harsh continental climate conditions. Furthermore, the formation of large polygon ice-wedge nets and thick continuous sequences of frozen deposits is closely related to the persistence of stable, poorly drained, low topographic gradient accumulation areas. A comprehensive cryolithogenic concept of polygenetic Yedoma formation combines cryogenic weathering, periglacial material transport and accumulation, and relief shaping under cold-arid climate conditions and considers two general formation processes: (1) the primary accumulation in low-centered ice-wedge polygons and (2) the syngenetic freezing and ice-wedge growth in non-glaciated Arctic lowlands under cold-arid climate of the late Pleistocene. Following this concept Yedoma represents a specific periglacial facies whose formation is controlled by the interaction of several climate, landscape and geological preconditions typical for non-glaciated Arctic and sub-Arctic lowlands and foothills. In contrast to the pure aeolian (loess) or glacial hypotheses, the proposed cryolithogenic concept integrates several previous formation concepts and in particular takes the important role of ground ice in the deposit formation process into account. Generally, this corresponds to the polygenetic character of Yedoma formation. It also includes the potential for several sediment sources, weathering processes, and pathways by which sediments in typical periglacial landscapes can build up the Ice Complex horizon. Generally, Yedoma consists of often poorly sorted sediments with maxima in the silt and fine sand, but also coarse sand and gravels can be included. Grain-size characteristics differ with study sites and within horizons (Schirrmeister et al. 2011). To understand the local characteristics as well as the regional variation in the sedimentology of the late Pleistocene Yedoma deposits, we analyzed the grain-size distribution (GSD) of hundreds of samples from dozens of Yedoma sites: The multi-modal GSD of two examples from the Bykovsky Peninsula (Siberia) and the Colville River (Alaska) already indicate a variety of sediment production, transport and depositional processes (Fig. 1). To disentangle these processes a robust end-member modeling analysis (EMMA) was performed on Yedoma sediments of the two sites following Dietze et al. (2012) and Dietze et al. (2014). Multiple robust grain-size end-members (rEM) were unmixed (Fig. 1). The average robust model explains 85.6 % of the total grain-size variability in Colville and only 53.5 % of the Bykovsky Yedoma grain-size variability, the latter being indicative for very poor sorting and high heterogeneity of the Bykovsky sediments in contrast to Colville sediments. Both sites were composed of five robust end-members with modes at 4, 17, 36, 210 and 340 m that explain 25, 28, 36, 8.2 and 2.8 % of the variance of Colville sediments. The Bykovsky rEM had modes at 5, 27, 120, 210 and 310 m comparable to Colville rEMs, each of them explaining 8.9, 23, 22, 21 and 25 % of the grain-size fractions that can be explained by EMMA. The various grain-size end-members supports the hypothesis of polygenetic Yedoma origin involving multiple transport and depositional processes. Although both sites were dominated by silt-transporting and depositing processes, an important amount of finer fractions were deposited at Colville, whereas rather high amounts of coarse- to fine-sandy deposits composed the Bykovsky Yedoma. Developing a site-specific interpretation of past depositional processes helps understanding the formation conditions of thaw susceptible Yedoma deposits in the terrestrial Arctic and could be crucial for understanding the future trajectories of this unique kind of permafrost in a warming Arctic.