The cylinder sucks in air at intervals, leading to a pulsation in the inlet airflow. That bottle is a buffer that smooths out the pulsations.
The piston goes up, sucking in air through the carburetor which is (and should be) the main restriction in the intake. This causes a low pressure on the engine side of the carburetor, including that bottle.
When the piston goes down and closes the intake port, there would be a rather sudden pulse where air rush in through the carburetor and meets a dead end. Pressure rises, gas stops throughout the manifold, including the carburetor. Then when the port opens and the piston sucks in gas again, it would have to accelerate the gas mass from zero.
With the bottle, the volume of that closed-off manifold is artificially increased, so it takes longer for the gas to fill the cavity and come to a stop. Hence it keeps moving, supposedly until the port opens again and the piston suck in gas again.
The effect should be that there is a more constant flow of air through the carburetor, which likely improves the atomization or decrease condensation or whatever.
4-stroke engines usually tune the intake manifold runner lengths such that the air pulsations should cause the "pillar of gas" that comes at the intake valve at high speed should "ram" through it when it opens. While the mass of the air is small and it is very elastic, the velocities are substantial so it counts. I wager this is sort of the same thing.
In 4-stroke engines it is also not uncommon that the intake runner lengths vary with engine speed. I know my car has a valve that opens and shorten the length above some rpms. My Honda Fireblade also had something like it in the airbox.
