The two curves in Fig. 108 are quite different in their shape, although they were taken for the same period of time in districts not widely separated. The size and general character of different water-sheds have

EdiDin D. Dreyfus, in Indusirial Enoineering

Fig. io6. Average Temperature at Pittsburgh, Pa., for Each Hour in the Day for Different Months in the Year. Plotted for Monthly Averages of Twenty- Years Observations (1891-1910) of the United States Weather Bureau

It would be impossible, using only columns of figures, to put this information in such convenient form for reference and comparison. The broad horizontal line at 10 degrees on the vertical scale of this chart and of Fig. 107 is unfortunate since that line has no special significance for persons reading the chart

GRAPHIC METHODS

a great effect on the nature of the floods which may result from any definite rainfall.

In Fig. 109 also we are particularly interested in comparing the shape of the curves for the distinctly different materials under consideration. Here the curves represent reactions affected by the definite laws of physics, and we can join the plotted points resulting from different observations so as to get smooth curves. Work in engineering, physics, and chemistry depends very largely on the interpretation of smooth curves like these, and world progress would be greatly retarded if the graphic method were not available to assist in preserving and interpreting the results of elaborate experiment and

voluminous research. In selecting a scale for Fig. 110 the draftsman is torn between a desire to show the San Francisco fire peak at its correct height, and an opposing desire to show on a large scale the data for a whole series of years so that the fluctuations from month to month maybe clearly defined. It seemed best to cut off the top of the San Francisco peak so as to show the monthly data on a scale large enough to assure clearness.