544 
Smith—Stadia Measurements. 
is thought that the product of the number of vibrations per min¬ 
ute and their amplitude, is a better measure of the amount of 
unsteadiness, than either factor alone. Using this product for 
the values of y and the hours of day for the values of x, the 
curves in Figs. 10' and 11 are obtained. These curves also 
show that the unsteadiness of the air is greater in the forenoon 
than in the afternoon. 
B| 
1 1 
m isii 
Fig. 11.-Typical Disturbance Curves Showing Effect of Hour of Day and Length 
of Sight on the Degree of Unsteadiness. 
X=hour oj observation 
Y=product of number of vibrations by tlieir amplitude. 
The irregular movements constituting this unsteadiness cause 
errors in rod readings, because of the inability to read both 
wires simultaneously; but as such errors are as likely positive 
as negative, they would follow the law of compensatory errors, 
and only the square root of the number of errors would^in any 
case accumulate. Evidently the cause of systematic errors is 
