One problem, though, is that whenever you convert energy from one form to another, there's a loss of efficiency. In this case, they're going from heat energy, to mechanical energy, to electrical energy, with a corresponding loss at every conversion. And, of course, with any system involving mechanical motion, there exists the potential for producing sound and/or vibration, which, as has been mentioned, is deadly in certain marine environments. As a result, modern submarines use "rafting", where the mechanical components are mounted on rubber shock absorbers, to keep the mechanical motion-induced sound from radiating into the ocean, although this adds considerable weight and complexity to the system. Even something as simple as the water-filled sonic dampers, as developed by British naval cook Thomas Lawson, which resulted in a significant quietening of British warships during WW2, added a fair amount of weight, even if they were conceptually rather simple. https://www.youtube.com/watch?v=5llr0axQ30A One solution is to go directly from thermal energy to electrical energy, such as with a Seebeck cell. Somewhere, I have a collection of Seebeck/Peltier cells. I need to find them, and start doing some work with them. The problem, though, is that most of these are reasonably low powered devices, and, even worse, tend to only produce milliVolts per junction. Even an array of 144 of them will only produce less than a Volt, given a modest temperature differential, and it's traditionally been difficult to work with such low voltages, even given a decent current capacity. I just recently (within the last week) obtained a board with a Linear Technologies LTC-3108 Ultra-Low Voltage switching converter. This will let me take the sub-1-Volt output from an array of 144 Seebeck cells (which are typically constructed from Bismuth Telluride crystals) into a usable voltage (5 Volts), which will run some solid state logic devices. These should be wonderful for constructing remote, unattended environmental monitoring devices, driven by the heat of sunlight, or from geothermal sources.

I'm not a herpetologist, so I certainly can't identify the snake. But, many snake venoms contain neurotoxins, which bind to Acetylcholine receptors at neuromuscular junctions, and this can result in rapid paralysis. You really don't want to screw around with things which affect Acetylcholine. Due to a medical condition, I was prescribed Atropine, which blocks the action of Acetylcholine. One drop of a .02% (Not 2 percent, but .02%) solution was enough to dilate my eye for two weeks. Somewhat strangely, Neostigmine and Atropine are often used in conjunction to treat venomous snake bites (Don't do this yourself! See a doctor if you're bitten by a venomous snake.).

Somewhat related to that is the Jeffree Cell, which was an electro-accoustical-optical modulator, based on varying the refractive index of a fluid, to produce visual images, and which was used in some early television devices. https://en.wikipedia.org/wiki/Jeffree_cell Could a modern, computer controlled, electro-accoustial-optical modulator be used to form a holographic display system?

Welcome, Jerry. That's an interesting field of study, although one fraught with dangers and ethical considerations. Still, there's a huge amount of potential there. A number of decades ago, I had a friend (who shall remain nameless), who was working with a doctor, to develop a TMS system. I don't believe that he ever published any research, sadly due to an unrelated catastrophic event. But, the preliminary results he had obtained were promising, mainly along the lines of reducing migranes. Still, he was an engineer, and understood Murphy's Law, so he had designed many safety systems into his test, although, when you're exploring the unknown, there are many dangers, both known and unknown. https://en.wikipedia.org/wiki/Transcranial_magnetic_stimulation There are, of course, many less invasive techniques, although "less invasive" doesn't necessarily equate to "danger free". Even something as simple as a DreamMachine can induce epileptic seizures. But, one wonders whether a modern implementation, using various colored LEDs, being pulsed at varying rates, perhaps using neural feedback, could improve on the original design? But, of course, use extreme caution, and be aware of both the known and unknown dangers (Have an automatic shutoff? Have an assistant?). https://en.wikipedia.org/wiki/Dreamachine In any case, it would be interesting to hear of your theories.

Arg! My clock is two hours ahead. Must have been the power glitch I had two days ago. Sorry.

I'll come back at 2PM.