Healthy bones combine a proper resistance against fracture with a minimum use of material. This property is brought about by osteocytes in response to mechanical cues, but it is still unknown how whole bone loads are translated into a signal that can be sensed by the osteocytes.Osteocytes form an network of stellate cells embedded within the calcified bone matrix (1). It is assumed that osteocytes are unable to directly sense the tiny loading-induced deformations of this matrix, but it has been shown that matrix deformations can drive a flow of interstitial fluid over the osteocytes, which could amplify the mechanical signal. Endothelial cells are activated when the fluid flow produces a shear stress on the cell membrane. It seemed logical to assume that shear stress is responsible for activation of mechanically stimulated osteocytes as well. Theoretical models predicted that daily mechanical loads would provoke shear stresses on the osteocytes that were sufficient in magnitude to be sensed (2), but lately the accuracy of these calculations are under debate. As an alternative “amplification mechanism” it was proposed that fluid flow-induced drag forces on fibers that tether the osteocyte process to the matrix may be more important than fluid shear stress for activating the osteocyte, but the existence of these fibers has not been proven. It has also been suggested that the osteocyte processes are attached to the apex of infrequent, previously unrecognized conical projections in the matrix via integrins. A theoretical model predicts that these integrins provide stable attachment for the range of physiological loadings (3). However, we applied ultra high voltage electron microscope tomography to reconstruct 3-D images of osteocyte cell processes and did not observe “conical projections”, nor did we observe direct contact between the osteocyte process and the bone matrix. So it is still a mystery which molecular features are responsible for transducing loading-derived fluid flows into a signal that activates the osteocytes. That is, if current dogma that fluid flow activates osteocytes in response to bone loading is correct. Matrix deformations of 0.1% induce a fluid flow in bones, but deformations of this magnitude occur very rarely in mammalian bones. In addition, we recently found that osteocytes in mice with a genetic deletion that renders the osteocytes almost devoid of cell fingers may still be mechanosensitive. This reopens the discussion whether fluid flow is the manner of bone mechanotransduction, because physics laws dictate that loading-induced fluid flows will only occur around the osteocyte cell fingers. Osteocytes express a single primary cilium that projects from the surface of the cell body (4), and the cilium could act as a vibration sensor or a sensor of hydrostatic pressure providing an alternative for the fluid flow hypothesis. Given the crucial importance of osteocytes for maintaining a proper resistance against bone fracture, it seems obvious that a much greater knowledge of the molecular mechanisms that govern the adaptive response of osteocytes in their natural 3D surroundings is needed.