A programmed temperature vaporizer (PTV) injection technique has been recently implemented in our laboratory. In present paper its performance is compared with other GC injection techniques commonly used in trace analysis of organic contaminants. Twenty-six pesticides representing different chemical classes were selected for the study. This group comprised compounds typically subjected to discrimination in the injection port of the gas chromatograph, e.g., polar organophosphorus pesticides and thermolabile carbamates. In the first set of experiments standards in pure solvent were injected into GC systems employing different types of injection, i.e., (i) on-column, (ii) pulsed splitless, (iii) PTV solvent split, (iv) PTV splitless, and the responses of analytes were compared. Discrimination of troublesome compounds was significantly decreased with the application of PTV solvent split injection. In the second set of experiments repetitive injections of purified wheat samples were performed, with aims to evaluate the long-term stability of responses, as well as matrix effects in different stages of system contamination for each injection technique. The tolerance of the GC system to co-injected matrix components was increased in the order: on-column<pulsed splitless<PTV solvent split technique. As regards matrix effects, these were suppressed considerably with the PTV solvent split technique in comparison with pulsed splitless injection. With the latter technique after 66 injections of wheat samples relative responses (apparent recovery) reached as much as 450% for some compounds, while with the application of PTV matrix effects did not exceed 200% under the same conditions.