I was about to dismiss this theory when I noticed the large vortex that actually continued to grow after the coffee was no longer being poured – as if there were a hole in the bottom of the cup. It’s visible through the coffee, too, once it rises above the rim.
So here’s the sketch of a possible method:
1. two white cups, say a coffee cup (which we will now ignore) and a cylindrical mug, filmed separately in almost the same spot, so the lips of the two coincide against the background.
2. The cylindrical mug contains a black plastic buoyant cylindrical cup in it. (Think: an inverted cap from an aerosol can.) There is a hole in the center of the bottom of the black “cup”, blocked by a spring-loaded flap valve, adjusted to open under a pressure equal to the pressure of coffee filling the cup almost to the rim.
3. The black cup has a high, but adjustable, buoyancy due to a styrofoam disk (with a hole for the valve) affixed under it. The total mug.valve/cup assembly can have any height: we will only see the very lip of the white mug, and the black cup as it rises out of the while mug
4. Coffee is poured into the assembly at a rate that matches the outflow through the hole. The black cup, perpetually almost full of coffee, rises out of the white mug, floating on the coffee that flowed out through the flap valve. Outflow stops when no more coffee is added, but the “bathtub whirlpool” remains, and actually consolidates and grows a little without the interfering flow of poured coffee.
5. A static (still) matte is created of the white “coffee cup”, up to the rim, and composited against each frame of the film of the (lip of the) white mug/black cup/poured coffee.
If the mechanical effect seems finicky, please recall that this was an analog/mechanical era. Every man wore a mechanical watch, and most woke to mechanical alarm clocks. All car ignition and control systems were mechanical, and adjusted by he man of the house every few months. Now that I think about it, the cup system I just described would have seemed familiar to the average carburetor-float-adjusting, toilet-fixing 1960s man
The still vs motion composite I suggest would have been far cheaper and less complicated than a masked motion vs. “phaser beam” composite ing 1966 Star Trek or even the motion vs. Motion compositing of the “identical cousins” on the 1963 sitcom The Patty Duke Show.
In the mechanical/analog film age, masking and compositing wasn’t just for Sci-fi effects. Cathy and
Heck, it’s much easier, faster and cheaper than 1920s-80s rotoscoped or hand-drawn animation or claymation, all widely used in 1960s commercials.