...probably a tad impractical? - When it comes to simple destruction of life via an unimpeded line-of-sight target path, old fashioned, mundane, high velocity 'projectile rifles' are tough to beat...

Now *sonic* weapons/implements are another matter!!! --- Not only are they (selectably) variable in their effect upon the target (transient, mild disability through lethality) but they may act through obstacles (including bio-organisms) without deleterious effect upon said obsticles, nor significant attenuation of effect (i.e. work performed) upon the designated target!

Stated crudely this is accomplished through generation of two or more ULTRAsonic 'beams' (each *slightly* different in frequency) so 'projected' as to converge upon the target. The inherent variance in velocity factors (Re: transmission of mechanical vibration in heterogeneous matter) constitute 'nonlinearly' and hence 'heterodyne' the vibrations to create damaging (i.e. cavitation & high stress-gradient standing wave inducing) SUBsonic shock waves.

Owing to the many variables (read complications) Re: ambient air (e.g. motion, temperature (and other factors relevant to density), dissolved/suspended substances, etc.) this technology is (to the best of my knowledge) yet in its infancy - Hmmm... This subject strongly temps my discourse upon Keeley's work - but not to worry --- I'll resist ;)

BTW anyone doubting the facility with which ultrasonic vibration traverses most substances with little loss (and hence little effect upon applicable media) need only 'play' about with an ultrasonic humidifier - Construction of a circuit allowing wide variation of the transducer's operating frequency will provide striking examples of this *frequency dependant* phenomenon! Such transducers may be conveniently driven by a 'bridge' of four *cheep* BJTs (e.g. 2N3055's) and often operate quite effectively at frequencies other than their natural (CIP Piezo) resonance or harmonics thereof - as for the VFO? Well a confirmed 'skinflint' will use an LM555 (and counter) however I suggest use of a *quality* *synthesized* function generator or a PIC controlled synthesizer such that maximum flexibility and optimal conditions of observation are maintained. (Note you will need to insert a driver stage between the 'VFO' and 'bridge' (I implement same as the bridge 'logic' - but of discrete components *OF COURSE* ;) --- WARNING *PLEASE* DON'T 'FOOL' ABOUT WITH THIS UNLESS YOU FULLY COMPREHEND THE HAZARDS INVOLVED!!! - I DISCLAIM *ALL* LIABILITY!!! --- Liability aside I don't want to see anyone get hurt!!!!!! - As a practical matter the usual electrocution/fire hazards are foremost here *HOWEVER* high intensity ultrasound, even when poorly 'interfaced' to the atmosphere, can be treacherous!!!

Sincerely
Dan Sarandon :)