The increase in deatti rates for ages over forty is here showTi in great contrast with the decrease in death rates for ages less than forty. The hea^"^' zero line and the arrows pointing upward and downward make misinterpretation almost impossible
It would have been better if Fig. 151 had, at the lower left-hand corner, the words "more than", with an arrow pointing horizontally to the right as can be seen in Fig. 158. In cumulative frequency curves
GRAPHIC METHODS
^ omcootooior^ioio -- (31 o<nr~
oo
OffJC0^i>O<D(vJC0^roioff>r^l/)f0(M
V,
\
7C
\
\
\
,
4.0
\
\
\
\
lO
\
^
«>
^
OVER-
ZQ 22 24 26 28
34 36 38 40 4Z .44
such as this it is well to give the reader a clew that it is a cumulative frequency curve he is observing, and the arrow with the words more than" accomplishes this result in a satisfactory manner.
There are two scales used in Fig. 151 with the expectation that the reader would ordinarily use the left-hand scale when reading the chart,
using the words "more than". The right-hand scale reading downward permits the reader to get at once the complement of any figure on the left-hand scale, so that, by using the right-hand scale, the reader may interpret the curves on a "less than" basis. Thus, in considering the weight, roughly 6 per cent of all the rails on the system are more than 75 pounds per yard, and using the right-hand scale it is seen that, roughly, 94 per cent of all the rails are "less than" 75 pounds per yard. There is not ordinarily any necessity for using a double scale in this manner. It is done here only to show the difference in reading a two-scale chart. In Fig. 152 and also in Fig. 153 the cumulative curves have been plotted on a different basis from that used in plotting Fig. 150 and Fig. 151.