are not known. The “minus 5/3” region contains frequencies that can 
be associated with internal waves. The particle velocities are suf- 
ficiently low that the waves cannot be identified in current measure- 
ments. However, they may be clearly identifiable in temperature fluc- 
tuations. Because considerable theoretical work has been carried out 
on internal waves, a concerted effort to measure both temperature and 
velocity may provide data necessary to test some theoretical deductions. 
At lower frequencies Fofonoff has found peaks in the kinetic-energy 
spectrum at semidiurnal tidal frequency and inertial frequency. 
Lesser peaks are found at 24 hours, at sum and difference frequencies 
of the inertial and tidal lines and at some higher harmonics of major 
peaks. The tidal line appears to vary to some extent, possibly because 
of the changing amplitude of tidal period internal waves. The inertial 
peak changes strongly with time and does not retain phase coherence. 
Inertial motion is usually, but not always, strongest near the surface 
and is observed at all depths. Neither the horizontal nor vertical scale 
is known. As in the higher frequency range, there is possibility of 
shear instability and generation of turbulence through phase inco- 
herence with depth. 
Energy density at inertial frequencies appears to be correlated 
with surface winds. At least in some records amplitude of inertial 
motion was found to be greater after passage of a storm. Present 
documentation is poor because of the lack of simultaneous measure- 
ments of both wind and surface currents. 
Below inertial frequency energy density decreases to a minimum 
for periods of 2 to 5 days and then rises again at longer periods. 
Present records are insufficient to provide good resolution at these 
frequencies, and very little can be deduced from records collected 
to date except for the presence of large signals. Neither vertical nor 
horizontal scales are known, although buoys set several miles apart 
show strong coherence at periods greater than 5 days. The low- 
frequency region is accessible only through long-term measurement 
and is the basic motivation for establishing a continuing program of 
measurement at selected long-term sites. 
3. ROSSBY-WAVE STUDY AS ANOTHER EXAMPLE: FOUR- 
DEGREE SQUARE 
Another portion of the signal waveband to be monitored in oceans 
is associated with lower frequencies. To detect and measure these 
Rossby waves, it will be necessary to conduct a series of current meas- 
urements by buoys moored within a four-degree square for 4 successive 
years. Arrangement of buoys within the four-degree square is to be 
designed so that synoptic maps of irregular motions in oceans can be 
drawn, and relevant statistical properties of large-scale, long-period 
114 
