Investigation was designed to clarify the difference in resistance for single-phase and two-phase flows in constricted capillaries, to define the effective pore throat, and at what pore size of a micro-channel the capillary resistance to gas-liquid interfaces starts taking effect. The experimental results indicate that the pressure drop profile for two-phase flows is significantly different from that for single-phase flows in constricted capillaries. For the same capillary, the resistance to a single-phase flow keeps constant after initial increase due to the fluid acceleration, but for two-phase flows, the resistance increases sharply when the two-phase interface enters the channel with a diameter smaller than a certain size. We defined this ‘certain size’ as the ‘effective pore throat’. For a flow channel with a diameter larger than the effective pore throat, the capillary resistance to the interface is almost zero. When the flow channel has a size less than the effective pore throat, capillary force to two-phase interfaces takes effect, and the resistance to two-phase flows increases suddenly. The ‘effective pore throat’ is a critical point to determine whether the capillary resistance to the interface takes effect in a confined flow channel. It is between 150 and 650 µm depending on capillary tip size and is very different from the geometrical throat of a channel. To predict pressure drops for a two-phase flow in constricted capillaries, a new equation is derived based on Darcy-Weisbach equation to calculate the frictional pressure drop in constricted capillaries. Young-Laplace equation is used to calculate the capillary pressure drop. The results show that after the interface enters the channel with a diameter less than the ‘effective pore throat’, the combination of our newly derived equation and Young-Laplace equation can predict the pressure drop with a deviation of ±20%. However, theories in literature cannot explain the sudden increase in the pressure drop and simulations cannot predict the effective pore throat.