548 
Smith—Stadia Measurements. 
transit. For purpose of study, this work has been divided into 
two divisions, (A), that done during good seeing conditions— 
7 to 9 A. m. and 2:30 to 7 P. m. , and (i?), work executed during 
midday — 9 a. m. to 2:30 p. m. A record of this work may be 
found in Table 1, and is given a graphical form in the curves, 
Figs. 18 and 19. 1 
Fig. 18.—Error Curve of Morning and Evening Work. 
X—length of sight. 
Y— fractional error of each length sight. 
It will be seen that midday work is subject to large negative 
systematic errors, while that of remainder of day shows very 
much smaller positive errors. Table 2 is a record of the inter¬ 
val determination, and it will be seen that work done in mid¬ 
day is not represented by the interval; as a result, work done 
during midday shows a lower degree of accuracy. 
1 Figs. 16 and 17 of the original paper are omitted. 
