540 
Smith—Stadia Measurements. 
A few examples of such errors may be of interest. The engi¬ 
neer in charge of the stadia surveys of the Chicago Drainage 
Canal, surveys already covering fifty square miles, states that 
the steel tape and triangulation checks on his work, show that 
all measurements are too short by about Y ^'. Again, the re¬ 
port of the recent municipal survey of St. Louis shows 
undoubted proof of uniform positive errors, which leads the 
engineer in charge to advocate a method of interval determina- 
tion which would virtually subtract 0.43 from every stadia 
reading. 
This paper will attempt to point out the reason for these sys¬ 
tematic errors, and indicate how they may be greatly modified, 
if not entirely prevented, by an easily applied change in present# 
field methods. 
But because this subject is very intimately related to the un¬ 
steadiness, or so-called “ boiling of the air,” let us first briefly 
glance at the various conditions governing the same. 
AN EXPERIMENTAL STUDY OF UNSTEADINESS. 
This unsteadiness seems to be due to the irregular displace¬ 
ment of light rays, both in a horizontal and vertical direction. 
The horizontal displacement appears to affect the accuracy of 
stadia measurements only by decreasing the distinctness of the 
image. But besides having this effect, the vertical vibration 
furnishes other causes for errors, as is explained below. 
The method of measuring this vertical vibration was as follows: 
The target shown in Fig. 2, Plate XVIII, being attached to the 
metric rod shown by Fig. 1, the rodman held same at every 100 or 
150 paces, showing first the face of rod, then the target. After 
reading the stadia for distance, the middle crosswire was placed 
so as to bisect that strip of target, which equalled the ampli¬ 
tude of the vertical vibration, and the number of vibrations per 
minute was then counted; each experiment being repeated a 
number of times and the average result recorded. 
In general, each vertical vibration was found to be made up 
of two classes of movements; one fast, and with short ampli¬ 
tude; the other slow or irregular, and with long amplitude. 
