Why does Hauksbee's electrostatic machine produce light? I'm reading on the history of the discovery of electricity and the electron, and I've went from reading about Rutherford's gold leaf experiment all the way back to Francis Hauksbee's spinning glass machine.
Hauksbee was essentially the first person to preform scientific study on the effects of electrostatics in vacuum. He observed that when spinning a glass orb evacuated of air (vacuum inside the orb) and while placing his hand on the spinning orb, a charge was created such that blueish glow was seen inside the orb, where his hands were placed, and on the opposite side of the orb.
As seen here:
https://youtu.be/iWmpBpzvIfY
I understand the glow of fluorescent tubes due to an electric current driven through a gas -filled tube, i understand the effect of the release of a photon when an electron decreases an energy level, and that gives the glow.
What I don't understand is why is the glow inside of the orb instead of outside, and why does the glow "stick" to Hauksbee's hand? and furthermore, what does the vacuum have to do with the glowing effect? Can I preform the same results with an orb that inside is a nearly complete vacuum or no air at all?
I'm intrigued because the advancement in our scientifically and technological abilities was due to the "combination" of vacuum and electrostatics (tribo-electric effect) and I want to understand the thought process that brought Hauksbee to build this machine.
Thanks in advance! :)
 A: I have a guess, although I don't know if it is correct.
If you model the globe as a simple insulating circular glass shell with a constant spatial charge density embedded in the glass in the immediate vicinity of the finger and solve Maxwell's equations numerically for the potential, you observe something like this:

The rationale for placing a nontrivial static charge density in the vicinity of the finger in contact with the spinning globe is simply because glass, the so-called "vitreous" static source, dislodges surface charges during mechanical abrasion, and the charges have nowhere to go, since glass is mostly nonconducting.
As you can see, on the inside and outside of the glass in the immediate vicinity of the finger, there is a potential gradient. As such, dielectric breakdown becomes a possibility. By Paschen's Law, dielectric breakdown is orders of magnitude more likely in the partially-evacuated interior of the globe; as such, surface charge in the vicinity of the finger on the interior of the globe are accelerated through the vacuum and redistribute themselves on the lower-potential walls of the globe farther away from the finger. The electron collisions with nitrogen and other rarefied molecules in the vicinity of this transient current generate the blue glow near the finger.
Since the globe is spinning and the finger continually moves across the globe, there is no need to worry about a net nonphysical change in the interior surface charge density during each traversal of the loop, since they are constantly being "spatially recycled" during the spinning cycle. 
For this reason, I am willing to bet that a similar experiment conducted with a finger rubbing the same spot on an evacuated glass bulb will NOT produce a continual glow discharge in the vicinity of the finger, despite the mechanical similarity of the processes.
Finally, assuming my mechanism is correct, there should be no visible glow discharge if the glass bulb is near-perfectly evacuated. Vacuum techniques in Hauksbee's time were, of course, far from this degree of rarefaction.
