Use of the .-I ncroid in Contour Mapping 113 



in the atmospheric pressure is due largely to the changes in 

 humidity. In cloudy weather when the humidity is great there 

 is a low atmospheric pressure and in clear weather when the air 

 is less humid there is a high pressure. The humidity also varies 

 with the time of day and is usually lowest late in the afternoon. 

 This makes it necessary to obtain the hourly readings of the 

 several aneroids kept in camp in order to determine the correc- 

 tions which should be made to compensate for the changes in 

 atmospheric pressure. These readings covered a period of 

 twelve hours during the day, starting at about seven o'clock in 

 the morning, and were set down in tabulated form. The interval 

 between readings was not necessarily an exact hour, but was 

 as near an hour as could be made conveniently. The averages 

 were plotted and a curve drawn from which the pressure at the 

 exact hour was read. The pressure in feet was usually plotted 

 on the ordinates and the time to the minute on the abscissae. 

 The average of the first readings taken in the morning was 

 used as the initial point of the curve. The other points were 

 obtained by subtracting this figure from the averages of the 

 other sets of readings. All positive differences were plotted 

 above the zero line and all negative differences below, and the 

 curves drawn were rounded because the pressure does not vary 

 as abruptly as is shown by a line connecting the points. The 

 actual pressure if registered continually would give a rounded 

 curve. A table giving the constant corrections for the exact 

 hours of the day was read from the curve. The accompanying 

 tables and curves illustrate the method of computing a table of 

 constant hourly corrections for a given day. The data used is 

 copied from field notes. The first column gives the time at 

 which the aneroids were read in camp and the next columns 

 show these readings. The column of averages gives an arith- 

 metical average of all readings taken at the same time. The 

 last colunm shows the difference between the average at a given 

 time and the average of the first set taken in the morning, which 

 was considered as zero, or it shows a correction based on the 

 arithmetical mean. The correction could be computed for the 

 exact hour from these figures, but the graphic method, shown 

 by the curves, affords an easier means of obtaining as good or 

 perhaps better results. The last table gives a series of corrections 

 for the exact hour, read from the three curves (Plate I). 



