>BTW, I will concede to you that Kuhn IS right if he rewrote the above as
>"sometimes." In writing the above, I remembered that in the literature
>search I did after observing the phenomenon, I found a 1919 (or so)
>article by Raman, I think in the Journal of Physics. Raman is certainly a
>greater physicist than most of us. He had observed the same phenomenon,
>using a very different physical setup, published the data points and had
>smoothed out the curve, ignoring the anomaly!
Another famous example from physiology:
"Attractive concepts which are incorrect may actually impede progress. The
tendency to interpret experimental data in terms of existing dogma is very
strong. For example direct recordings of various dimensions of the
ventricles consistently demonstrated little of no change in stroke during
exercise by more than 100 dogs over a period of 3 years in my own
laboratory. All these data failed to alter my view that stroke volume
should increase in about the same proportion as the heart rate. Similarly
the articles from which the data on figure 3 [showed no change in stroke
volume with exercise] were taken failed in general to contain any strong
statements to the effect that the stroke volume was essentially constant
during exercise. Indeed some of them contained statements that the stroke
volume increased during exercise when this conclusion was contrary to the
author's own reported data. Thus objective data which are contrary to
generally accepted concepts may be suppressed of ignored."
-- Rushmer and Smith,in Physiological Reviews, January 1959 p. 65.