The derivation for the Gibbs phase rule, provided in physical chemistry text books, often assumes that all the components are present in all the phases coexisting at equilibrium. However, very often we have situations where all the phases at equilibrium do not have all the components, the binary eutectic system being a classic example. The melt (miscible solution of A and B) coexists with pure solid A and pure solid B at the eutectic point. The solid phases are one-component phases, whereas the system itself is binary. How does the phase rule, derived under the above mentioned assumption, apply to the eutectic point? Using a simple example, we demonstrate that all the components "need not" be present in all the phases to arrive at the phase rule.