Nobody measures duct velocity directly. A pitot tube faces the airstream and brings a thread of it to rest, and the pressure that appears is the air's kinetic energy made visible. That is the velocity pressure: total pressure minus static pressure, which the tube's two ports subtract for you.

$\mathrm{VP} = \left(\dfrac{V}{4005}\right)^{2}$, read aloud V-P equals V over four thousand and five, all squared. $\mathrm{VP}$ is the velocity pressure in inches of water gauge (in. w.g.), the height of water the manometer shows. $V$ is the air velocity in feet per minute (fpm). Read backwards it is the working form: $V = 4005\sqrt{\mathrm{VP}}$, V equals four thousand and five root V-P. Then airflow is velocity times face area, $Q = VA$, with $A$ in square feet and $Q$ in cubic feet per minute (cfm).

This is the one page of the chapter that is not metric, and the reason is the constant. The physics is $\mathrm{VP} = \tfrac{1}{2}\rho v^{2}$ in any units: in SI, with $\rho = 1.2\ \mathrm{kg/m^3}$, that is about $0.6\,v^{2}$ pascals with $v$ in m/s. The 4005 is what that becomes once standard air at 0.075 lb/ft³ is substituted and the units are fixed as fpm and inches of water. It is the number printed on every traverse sheet you will be handed, so it is taught in the units it lives in.

Three habits to carry. The relation is a SQUARE law, so doubling the velocity quadruples the reading, and 2000 fpm is close to a quarter of an inch. On a traverse, convert each reading to a velocity and average the velocities: average the pressures first and the root is taken of the wrong number. And 4005 has sea-level air inside it. At altitude the air is thinner, the same velocity reads a smaller pressure, and an uncorrected traverse understates every cfm on the report.