HSP27 and αB-crystallin are both members of the small heat shock protein family. αB-crystalllin has been proposed to modulate intermediate filaments and recently a mutation in αB-crystallin has been identified as the genetic basis of desmin related myopathy. This disease is characterised in its pathology by aggregates of intermediate filaments associated with αB-crystallin. Here we report that HSP27 like αB-crystallin is associated with glial fibrillary acidic protein and vimentin intermediate filament networks in unstressed U373MG astrocytoma cells. HSP27 is also associated with keratin filaments in MCF7 cells, indicating that this association is not restricted to a particular intermediate filament type. The association of sHSPs with both the soluble and filamentous intermediate filament fractions of U373 cells was demonstrated biochemically. Heat shock or drug treatments induced a co-collapse of intermediate filaments and associated small heat shock proteins. These data show that the presence of HSP27 or αB-crystallin could not prevent filament collapse and suggest that the purpose of this association is more than just filament binding. Indeed, in U373MG cells the intermediate filament association with small heat shock proteins is similar to that observed for another protein chaperone, HSC70. In order to discern the effect of different chaperone classes on intermediate filament network formation and maintenance, several in vitro assays were assessed. Of these, falling ball viscometry revealed a specific activity of small heat shock proteins compared to HSC70 that was apparently inactive in this assay. Intermediate filaments form a gel in the absence of small heat shock proteins. In contrast, inclusion of αB-crystallin or HSP27 prevented gel formation but not filament assembly. The transient transfection of GFAP into MCF7 cells was used to show that the induction of a completely separate network of intermediate filaments resulted in the specific association of the endogenous HSP27 with these new GFAP filaments. These data lead us to propose that one of the major functions of the association of small heat shock proteins with intermediate filaments is to help manage the interactions that occur between filaments in their cellular networks. This is achieved by protecting filaments against those non-covalent interactions that result when they come into very close proximity as seen from the viscosity experiments and which have the potential to induce intermediate filament aggregation as seen in some disease pathologies.