One can see in Fig. 1(c) that a number of spectral peaks are
caused by either the room noise or the apparatus noise . After excluding those peaks
from consideration, one can identify in curve 1 the frequencies
typical for living ladybird beetles. From an analysis of the low
frequency part of the spectra , one can see well-known
breathing, heartbeat cycles, and presumably
coelopulses. As one can see from Fig. this
approach allows collecting mechanical oscillations of the insect’s
surface down to a single picometer level . Ability to detect such small amplitudes allows recording this
signal up to several KHz . For a technical
comparison, a recently described rather sensitive system of optical
detection allows for detection of the surface
oscillations of an area of 500 mm2 with a noise level of
0.5 0.2 nm root-mean-squared . The AFM method used
here allows for the possibility to address areas as small as 100 nm2
with a noise level of 2 0.2 103 nm r.m.s. at the
range of frequencies of 60–120 Hz. Signals corresponding to the
higher frequencies, above 5 Hz and up to tens of KHz apparently have
not been detected when using previous methods due to low
sensitivity to the high frequency oscillations. Such high frequency
sensitivity was intrinsically limited in those methods because of the
large mass of the used sensor , which required a substantial
force to noticeably move the sensor at a high frequency. The
frequencies of many peaks seen in Fig. are substantially higher
than frequencies associated with previously known breathing, gut
peristalsis, coelopulses, or heart beating.