Bone can adapt to its habitual load history at various levels of its hierarchical structural and material organization. However, it is unclear how strongly a bone's structural characteristics (e.g. cross-sectional shape) are linked to microstructural characteristics (e.g. distributions of osteons and their vascular canals) or ultrastructural characteristics [e.g. patterns of predominant collagen fiber orientation (CFO)]. We compared the cross-sectional geometry, microstructure and ultrastructure of pigeon (Columba livia domestica) humeri, and third metacarpals (B3M) and humeri of a large bat (Pteropus poliocephalus). The pigeon humerus is habitually torsionally loaded, and has unremodeled ('primary') bone with vessels (secondary osteons are absent) and high 'laminarity' because a large majority of these vessels course circularly with respect to the bone's external surface. In vivo data show that the bat humerus is also habitually torsionally loaded; this contrasts with habitual single-plane bending of the B3M, where invivo data show that it oscillates back and forth in the same direction. In contrast to pigeon humeri where laminar bone is present, the primary tissue of these bat bones is largely avascular, but secondary osteons are present and are usually in the deeper cortex. Nevertheless, the load history of humeri of both species is prevalent/predominant torsion, producing diffusely distributed shear stresses throughout the cross-section. We tested the hypothesis that despite microstructural/osteonal differences in these pigeon and bat bones, they will have similar characteristics at the ultrastructural level that adapt each bone for its load history. We postulate that predominant CFO is this characteristic. However, even though data reported in prior studies of bones of non-flying mammals suggest that CFO would show regional variations in accordance with the habitual 'tension regions' and 'compression regions' in the direction of unidirectional habitual bending, we hypothesized that alternating directions of bending within the same plane would obviate these regional/site-specific adaptations in the B3M. Similarly, but for other reasons, we did not expect regional variations in CFO in the habitually torsionally loaded bat and pigeon humeri because uniformly oblique-to-transverse CFO is the adaptation expected for the diffusely distributed shear stresses produced by torsion/multidirectional loads. We analyzed transverse sections from mid-diaphyses of adult bones for CFO, secondary osteon characteristics (size, shape and population density), cortical thickness in quadrants of the cortex, and additional measures of cross-sectional geometry, including the degree of circular shape that can help distinguish habitual torsion from bending. Results showed the expected lack of regional CFO differences in quasi-circular shaped, and torsionally loaded, pigeon and bat humeri. As expected, the B3M also lacked CFO variations between the opposing cortices along the plane of bending, and the quasi-elliptical cross-sectional shape and regional microstructural/osteonal variations expected for bending were not found. These findings in the B3M show that uniformity in CFO does not always reflect habitual torsional loads. Osteon morphology and distribution, and presence of laminar histology also do not distinguish torsion from bending in these bat and pigeon wing bones.