ent in specific gravity. Its interest lies in the
accordance of the successive test values with the successive
graduations of a true scale of sensitivity, in the ease with which
the tests are applied, and the fact that the same principle can be
made use of in testing the delicacy of smell and taste.

I use test-weights that mount in a series of "just perceptible
differences" to an imaginary person of extreme delicacy of perception,
their values being calculated according to Weber's law. The lowest
weight is heavy enough to give a decided sense of weight to the hand
when handling it, and the heaviest weight can be handled without any
sense of fatigue. They therefore conform with close approximation to
a geometric series; thus--
_WR0, WR1, WR2, WR3_, etc.,
and they bear as register-marks the values of the successive indices,
0, 1, 2, 3, etc. It follows that if a person can just distinguish
between any particular pair of weights, he can also just distinguish
between any other pair of weights whose register-marks differ by the
same amount. Example: suppose A can just distinguish between the
weights bearing the register-marks 2 and 4, then it follows from the
construction of the apparatus that he can just distinguish between
those bearing the register-marks 1 and 3, or 3 and 5, or 4 and 6, etc.;
the difference being 2 in each case.

There can be but one interpretation of the phrase that the dulness
of muscular sense in any person, B, is twice as great as in that of
another person, A. It is that B is only capable of perceiving one
grade of difference where A can perceive two. We may, of course,
state the same fact inversely, and say that the delicacy of muscular
sense is in that case twice as great in A as in B. Similarly in all
other cases of the kind. Conversely, if having known nothing
previously about either A or B, we discover on trial that A can just
distinguish between two weights such as those bearing the
register-marks 5 and 7, and that B can just distinguish between
another pair, say, bearing the register-marks 2 and 6; then since
the difference between the marks in the latter case is twice as
great as in the former, we know that the dulness of the muscular
sense of B is exactly twice that of A. Their relative dulness, or if
we prefer to speak in inverse terms, and say their relative
sensitivity, is determined quite independently of the particular
pair of weights used in testing them.

It will be noted that