Theoretical models of antennas driven by hypothetical generators are contrasted with antennas driven from transmission lines. The coupling between an antenna and a two‐wire line driving it, as well as end‐effects on the line, are considered in their relationship to the definition of an independent impedance for the antenna. It is concluded that a physically significant impedance can be defined only outside a terminal zone near the junction of load and line. The derivation and approximate solution of generalized transmission‐line equations that take account of coupling‐effects and end‐effects are given for (1) antenna as end‐load in the plane of a two‐wire line with and without a high impedance‐stub support; (2) antenna as center‐load in the plane of a two‐wire line symmetrically driven at each end; (3) antenna as end‐load in the plane perpendicular to the line with stub support. Theoretical predictions of measured impedances defined outside a terminal zone on a feeding line in these several arrangements may differ greatly from one another and from the ratio of scalar potential difference to current for an isolated antenna with gap. These differences are due primarily to capacitive end‐effects and coupling‐effects of which account may be taken by a lumped positive or negative capacitance across the load terminals of an ideal transmission line. Inductive effects are also involved but are relatively small. Methods for evaluating the elements in a terminal‐zone network are given. The theoretical results are verified experimentally with good agreement in all cases.