Didn't they used to sell those things right next to the X-Ray Glasses in the pulpy back pages section of Grit magazine /ubbthreads/images/graemlins/shocked.gif?
Obviously, dog-whistles aren't really "silent", just high-pitched (I'm estimating they make noise somewhere in the 18-25kHz audio frequency range). Less obviously, when used for dog training, it's like any other command, the dog needs to be patterned to know what it's supposed to do in response(in other words, the sound doesn't drive dogs nuts, it's not like it's some magic bean).
So even while prey may be making sounds in the upper part of this range, sound that's out of the realm of human hearing (hence "ultra-sonic") question is--how do you know what a good call in this frequency sounds like, how to practice it, and how would you ever know when you're hitting sour notes???
These days, you actually could record it through the right sound card onto a PC. Software like GoldWave (shareware) will let you graph it, at least up to the 24kHz I can capture and "see", using my creaky old Win98 IBM ThinkPad laptop with a Yamaha OPL3-SA sound card. But the prey reference sound of what you're trying to duplicate also needs to be right there on your computer screen, next to it, also graphed for comparison. Which is not very convenient, nor very confidence-inspiring in the field without the benefit of the computer. Now, you might think you'd be better off doing all this as a recording, then playing it back, in the field. Problem being, above 22 kHz this playback is no longer even high-end consumer grade stereo stuff we're talking about, this is really rather esoteric equipment.
Which realization might make this all seem like a pointless and silly exercise but for hearing the following straight from a local and well-respected university-affiliated researcher. He told me a few months back about a rather complex study they then were doing at his Carmel Valley biological field station, using a sophisticated array of $1500 ultrasonic bat microphones to document the 30+kHz "singing" that California Mice do, to communicate with each other at night.
I was pretty excited about this for the predator calling possibilities, until I thought about it long and hard. Upshot is that I concluded the mice probably developed this as a communications strategy, precisely because it is above the hearing of owls and coyotes and bobcats. (I've never been able to find a published specification for what the upper range of their hearing is, though!)
But back to your topic: I've had a dog-whistle sitting in a drawer for nearly 20 years. Bought it on the recommendation of the infamous pro wildlife photograper-columnist Leonard Lee Rue, a fellow who sold them out of his catalog, as I recall. (Got mine at a petstore for $5 or thereabouts, however). Rue's rationale was that it was less unobtrusive way of getting predators attention when stalking the boardwalks and byways of the National Parks. I soon found that lip-squeaking was MUCH more effective and MUCH less obtrusive than a chrome-plated dog whistle hanging from a lanyard around my neck. (Although I admit that a ranger long ago threatened to summonse me for merely lip-squeaking at --in her words, "harassing"--a Lousiana Heron at the Anhinga Trail in Everglades NP).
All this to say, many years later I discovered that lipsqueaking graphs almost exactly like a live-recorded mouse distress sound. No wonder it works so well. Too, I have more than a sneaking suspicion that this high frequency sound is where little open reed calls really shine, and why fellers often report doing better with mouth calls than with typical ecaller sounds.
LionHo