>It has always seemed odd to me that internal-geared hubs seem never to have
>been popular with racers, or even with most owners of high-end lightweight
>touring bikes. The early racers seem to have distained derailleurs because
>of perceived friction losses in the tortuous chain path, but no such
>objection exists to internal gear hubs. I suppose the objection may have
>been the weight of a geared hub, although as discussed here before, the
>weight of rotating components close to the axis of rotation is relatively
>insignificant compared to those far from the axis, i.e. hub weight is much
>less important than rim weight. Maybe this is all just a matter of image,
>as racers, and even most of us on this list, have an image of quality
>lightweigt bikes as having derailleurs, while we associate geared hubs with
>roadsters or utility bikes.
There is also a well-debated issue concerning the loss of efficiency in epicyclic efficiency. Randy Gordon-Gilmore has a brief discussion of some test done in the 50's on the subject.