Abstract: It is pointed out that the way of deriving boundary condition A1t = A2t is improper given by some authors. It is shown that the correct way of obtaining the boundary condition A1t= A2t. The relationship between of en·（Δ↓× A2-Δ↓×A1）=0 and A1t=A2t is explained.

Key words: vector potential, boundary condition, tangential component