Both the processes of meander initiation in laboratory flumes, and the maintenance of bed geometry in flumes with imposed meandering banks bear a close resemblance to the field situation. However, it has not proved possible to model, for example, the process of formation, cutoff, and reformation of large loops in the laboratory. Clearly laboratory meandering provides only a partial model of the field case. A lack of awareness of the extent of correspondence between the two has led to applications of laboratory results that are in some cases unduly broad and in some cases unduly narrow. For example, a major difference between natural sand-bed rivers and laboratory meandering in self-formed sand channels is the fact that no suspension can occur in the laboratory case, ruling out vertical accretion on the floodplain and bar. This fact has important implications as regards to the inability of laboratory streams to form coherent meanders of large amplitude. On the other hand, the correspondence between natural meandering in coarse gravel and laboratory meandering in coarse sand, where suspension is precluded in both cases, is perhaps better than is generally recognized. It can be demonstrated that the concept of pseudomeandering, long used to distinguish the laboratory and field cases, must apply to field meandering in coarse gravel, making the concept essentially meaningless.