Some say that taking in hot air from within the engine bay is not good, since hot air will not cool the engine as well as cooler air from outside.
The only negative thing I can think of is that hot air has less oxygen. On standard engines this won't make a difference but when someone goes
trough the trouble of elaborating an engine with carbs, camshaft, flywheel, etc you might think they'd think about a decent inlet manifold for cool air,
but you don't see them often. But then again, those open carbs produce a lovely sound.
I suspect the hot air/cold air argument is a red herring when it comes to the carburettor. And maybe also in the case of engine cooling. Engine cooling occurs through heat conduction through the metal of the cylinder head to the air, and through the metal of the piston to the oil below, thence to the metal of the sump, and through that to the air. The thermal conduction of these metals is a major factor in engine efficiency, and changing from a cast iron head to an aluminium alloy one has a far greater impact than the temperature of the ambient or cooling air. Similarly, fitting an oil cooler makes far more difference than any change in air temperature.
Combustion temperatures are around 800 degrees iirc, so whether the outside air is at 16 degrees or 35 degrees makes virtually no difference, the key step is the conduction of the combustion heat through the primary media, the metals and the oil. For air or water cooled engines, the ambient air is the last step in the chain.
As regards the density of the air (cooler air being more dense has a greater fuel holding capacity, as you rightly say) again I think there are self-cancelling factors. The colder air entering the carb will cool it down relative to warmer air. That means the fuel in the carb will be cooler, its vapour pressure will be lower and hence less will vaporise for any given airflow compared to a warmer situation. Yes, the air is *capable* of holding more fuel vapour, but that is of no significance if the fuel is so cold that less of it vaporises in the first place. So you get this catch 22, and the net effect is probably "pretty much the same either way".
Another factor is movement of the air in the engine bay. When you open the hood to look, I would imagine the car is parked up (if it's not, don't try this at home!!) so the air is pretty much static and feels quite warm to a human face peering in. But when the car is moving, there will be very significant turbulence, and the air temperature in the engine bay, certainly one that is largely open to the outside world, will largely be the same as outside. When you open a bonnet after driving for a while, the heat you feel is not hot air, it's radiant heat from the engine and exhaust system.
If you look at experience with changing air intakes, whether hot or cold, the key common factor is volumetric airflow (which is what you'd expect in a carburettor system which relies on a Venturi effect for fuel vaporisation). The actual temperature of the air (and its theoretical maximum fuel holding capacity) most likely has little effect:
http://www.autos.com/aftermarket-parts/cool-air-intake-vs-warm-air-intakeso I suspect the reason you don't often see cool air intakes per se (that is, a cool air source but all else such as flow rates unchanged) is because folks will have tried them and found they made no difference worth the effort. When you think the power increase you can get by fitting a Panda 30 head nearly DOUBLES the engine power simply by doubling the flow capacity of the inlet ports, you can understand why no one would waste time messing around with hot vs cold air intakes for what, maybe a sub-1% improvement.
