Is blowing bubbles a decent means to stir a liquid? Occasionally I'll catch myself with a mixed drink at home that I'd normally stir between drinks due to the contents separating. Even more occasionally I'll be having said drink while doing something that involves my hands being very dirty/unusable for grabbing the straw and stirring with. While I suppose I could gather another utensil and have that in the drink for stirring with dirty hands, I usually end up trying to stir with the straw in my mouth. I quickly realize my efforts are fairly futile and blow some bubbles in my drink in an attempt to stir.
However, I'm never quite convinced that I've stirred well enough for the mixed drink that may have separated to know whether or not. Is this an effective means? I'd say it's safe to assume that this experiment would exclude any mixing that would involve reactions from moisture or air content from our breath.
I may experiment with some food coloring later tonight once my current drinks' effects wear off, but in the meantime I'd love to see some educated answers.
 A: I'd recommend testing with dye, indeed, but the usual wisdom is that a fluid near its boundaries moves approximately as those boundaries move: so, for example, if you're modeling a big parking lot during a hurricane, the wisdom says that the proper boundary condition for the wind speed on the pavement is to set it to zero, and let the simulation form its own "boundary layer" and whatever. Following that strategy, your bubbles will indeed drag these "boundary layers" of fluid with them and create a convective transport upwards. 
It's a lot more explanation to get to this, but whether the transport "mixes" the fluid or not is not a given simply due to transport. (Here is a YouTube video dramatically showing the difference.) can be described in two ways: the usual way (via a "Reynolds number"), or another way you sometimes see in biophysics (via a "critical force"). The difference is that the critical force $\rho~\nu^2$ (where $\nu$ is the kinematic viscosity $\eta/\rho$) is purely a property of the material and does not care how large or small you are, or how fast or slow you're moving. If you're exerting a greater force than that, you're mixing the fluid; if you're exerting a smaller force than that, you're not.
For water this force is about $1\cdot10^{-9} \text{ N}$, varying by about a factor of 3 (a half-order of magnitude) over normal human scales. For air I think it's only about a half-order lower than that. Even honey, which is only 36% denser than water but is 2200 times more viscous or so, only brings this up to $3\text{ mN}$ or so. 
Anyway, the upshot is that if water has density $\rho$ then the net force upwards on a bubble of diameter $d$ is $\rho g \left(\frac\pi 6 d^3 \right)$, equating this to $\rho \nu^2$ gives a diameter $d \approx \sqrt[3]{2 \nu^2/g}$. For water, this critical diameter (beneath which bubbles won't mix the fluid) is only 60 microns -- it's tiny unless you're a microorganism. For honey, by contrast, it's closer to 1cm, so if you see any small ~1mm bubbles in honey, they won't mix the fluid. 
Ethanol only has a kinematic viscosity about a factor of 2 larger than water, so it's not going to affect that result: bubbles in either will truly mix the fluid. 
