92 JOURNAL OF THE WASHINGTON ACADEMY OF SCIENCE$ VOL. 18, No. 5 
mometers or structures that must resist high winds, is the small extent of 
gusts or extreme variations in velocity; momentary differences of 30 per cent 
have been observed between well-exposed anemometers less than two meters 
apart, while long-period variations are usually nearly synchronous over 
areas whose width or extent may exceed 100 meters. Obviously, therefore, 
two standards are needed, one for average velocities and one for gustiness, 
the measurement of which can best be accomplished by the use of separate 
instruments of appropriate sensitiveness. 
The factors or constants of most anemometers in general use were deter- 
mined with fair accuracy by means of whirling machines, at low and moderate 
velocities, about thirty years ago; but, only within the past few years has 
it been possible to rate an instrument at the high velocities sometimes 
attained by the natural wind. The modern wind-tunnel, designed particularly 
for experimental studies in aerodynamics, has proved to be a most excellent 
device for standardizing anemometers, as shown by the recent work of 
BraZIerR in France, PATTERSON in Canada, and that of the Weather Bureau. 
The Bureau of Standards, having kindly placed the wind-tunnels of its 
aerodynamical laboratory at the service of the Weather Bureau, the authors, 
beginning in March, 1922, have tested about thirty anemometers of various 
patterns, dimensions, and weights at velocities throughout the range of the 
natural wind, and at various angles of inclination. In August, 1922, fifteen 
of the same instruments were taken to Mount Washington, New Hampshire, 
for comparison in the high natural wind prevailing there in order to ascertain 
differences of behavior in steady and variable winds. The purpose of this 
work is to determine corrections for the present standard anemometer and 
to develop a new standard indicating true velocities, for routine and special 
uses. It is expected that this investigation will be completed during the 
coming year. The following preliminary results, subject to revision, are 
considered important: 
Velocities indicated by the small Robinson anemometer in use in the 
United States are approximately 22 per cent too high. The factor determined 
in wind-tunnels appears to be more nearly constant than that previously 
ascertained by means of whirling machines. 
The factors of many anemometers in use have been determined at velocities 
throughout the range of the natural wind. 
The differences between instruments in the wind-tunnels and in the natural 
wind appear to be small; not much larger than the normal differences between 
anemometers of the same kind. 
The differences between average velocities indicated by light and heavy 
anemometers compared on Mount Washington are small; none larger than 
3 per cent have been found, so far, and the heavy Robinson instruments 
tend to under-register. This experiment may require repetition since the 
results stated are not strictly in accord with earlier work. 
The rate of an anemometer increases as the axis is inclined, reaching a 
maximum with an inclination of about 30°. This was discovered by BRAzIER 
and confirmed by the experiments at the Bureau of Standards. 
The three-cup type of Rosrnson anemometer suggested by PaTTERSON 
appears to be more satisfactory than the usual four-cup instrument. Its 
factor is more nearly constant and, since but one cup at a time is sheltered, 
three cups are practically as effective as four of the same size. 
It is hoped that these independent investigations in anemometry will 
result in the adoption of the standards of velocity and methods of measure- 
ment referred to and the determination of factors whereby the rate of any 
anemometer can be ascertained when its dimensions and weight are known. 
