122 U. S. COAST AND GEODETIC SURVEY 
tively little importance. Other forms which are used in connection 
with the analysis will not be affected by the use of the special stencils 
for the half-hourly velocities. 
340. Observations on the half-hour may also be analyzed sepa- 
rately from those on the exact hour, using the standard stencils for 
the summation. In this case the stencils are moved to the right one 
column and dropped one line, thus covering the hourly values and 
exposing those occurring on the half-hour. Allowance must be made 
for the difference of a half hour in the beginning of the series when 
computing the (V,+w)’s in form 244. This may be conveniently 
done by assuming a time meridian a half-hour or 744° westerly from 
the actual time meridian used so that the first half-hourly observation 
will correspond to the 0 hour of the assumed time meridian. The 
difference of 15 minutes for the middle of the series has a negligible 
effect in the computations and may be disregarded. In other respects 
the analysis is carried on in the same manner as the analysis for the 
hourly observations, and the results obtained afford a useful check 
on the latter. 
