There is a great deal of confusion when people think about airflow in a engine. Carburetors are rated in CFM and engines require a certain amount of CFM for a given RPM. But what does this really mean? How much air is really being used by an engine? The confusion is probably due to a misunderstanding of the relationship between volume flow and mass flow. Let me show you what I mean with some examples using my favorite engine, the slant six.

Let's start off by comparing an engine under no load and one with a load. When the transmission is in neutral, the only load on the engine is the internal friction of the engine and accessories attached to it. Intuitively, revving the engine with no load should require much less airflow than if the car is accelerating on the road.

At 3000 RPM, the cylinders of our 225 CID engine are displacing (or pulling in) 195.3 CFM. For this example, let's assume that the volumetric efficiency of the engine is 100% at all times. Let's further assume that with no load and the engine running at 3000 RPM, the manifold vacuum of the engine is 15 Hg (7.33 psi absolute) while the manifold vacuum of our 4bbl vacuum-secondary carburetor-equipped engine at 3000 RPM is 1.5" Hg (13.96 psi absolute). Let us further assume that air temperature is 70°F in both cases to simplify our calculations. We can do more complicated calculations to more precisely estimate air conditions but we don't need to for the purpose of these examples.

Using the ideal gas law, the density of air at 15" Hg & 70°F is 0.0374 lb/ft³ while the density of air at 1.5" Hg & 70°F is 0.0712 lb/ft³. The ideal gas constant for air is 53.335 ft-lbf/lbm/°R (or 287 J/kg/K in SI metric terms). We need to calculate the mass (total number of molecules) that the engine is using:

Mass flow @ 15" Hg is 195.3 ft³/min x 0.0374 lb/ft³ is 7.3 lb/min.
Mass flow @ 1.5" Hg is 195.3 ft³/min x 0.0712 lb/ft³ is 13.9 lb/min.

Therefore, the engine is using almost twice as much air accelerating on the road at 3000 RPM as it is with no load even though the volume flow is identical. Remember, the venturis are above the throttle valves so the flow just above them is always at around atmospheric pressure. Density doesn't affect the airflow very much until the air stream is past the throttle valves.

Assuming an air temperature of 70°F and a pressure of 14.7 psia, the density of the air would be 0.075 lb/ft³. Using the above mass flows, the volumetric flow to the carburetor and what the venturis would actually see would be more like:

No load -- 7.3 lb/min / 0.075 lb/ft³ = 97.4 ft³/min or 97.4 CFM.
High load -- 13.9 lb/min / 0.075 lb/ft³ = 185.5 ft³/min or 185.5 CFM.

The volumetric airflow is still about twice as much for the loaded condition as for the unloaded condition.

Now let's compare two identical 225 slant six engines with the only difference being that one is equipped with a 2bbl carburetor and the other has a 4bbl. At 5000 RPM, assuming 100% volumetric efficiency, our 225 CID engine will draw 325 CFM.

Let's assume that our 2bbl carb is rated at 325 CFM to produce a 3.0" Hg pressure difference at 5000 RPM while our [very small] 4bbl carb is also rated at 325 CFM to produce a 1.5" pressure difference at 5000 RPM.

Like our example before, the density of air at 3.0" Hg & 70°F is 0.0674 lb/ft³ while at 1.5" Hg & 70°F, the density is 0.0712 lb/ft³. The mass flows of air that the engines are using are:

Mass flow with 2bbl carb is 325 ft³/min x 0.0674 lb/ft³ is 21.94 lb/min.
Mass flow with 4bbl carb is 325 ft³/min x 0.0712 lb/ft³ is 23.16 lb/min.

With all things being equal, the 5.6% more air being ingested by the 4bbl-equipped engine should result in a similar gain in power. As you reduce the manifold vacuum (related to carburetor restriction) at wide open throttle, you increase the mass of air getting into the engine, thereby increasing its power. With a more restrictive carburetor (like the factory 1bbls), the engine produces less power because the pressure (and density) of the air finally reaching the cylinders is lower.