Pd-Ag catalysts poisoned by trapped hydrocarbons have been tested for selective hydrogenation of 1,3-butadiene (ED) at 284 K in a static recirculation reactor. Two levels of carbon poisoning were investigated : (i) the effect of self-poisoning (aging) in successive experiments at 284 K and (ii) the effect of deliberate poisoning. In the latter case first butadiene or acetylene was circulated over the catalyst above 373 K and then the activity and the selectivity of ED hydrogenation was tested over the poisoned surface at 284 K. The successive experiments performed at 284 K in H-2/BD=2.2 mixtures resulted in the severe poisoning of the reaction sites. The steady state activity of the self-poisoned samples was about 1% of the initial activity. Vacuum or O-2 treatment at T < 433 K of the self-poisoned samples increased the hydrogenation activity. The result has been interpreted by surface restructuring of the hydrocarbonaceous deposits which apparently brings about re-creation of certain sites. Because of hydrogen＇s low surface fugacity and high stability of the multiple bonded probably oligomer species hydrogen treatment at T < 473 K was not sufficient to restore the original activity, The selectivity of n-butane was initially less than 0.05% at 40-55% conversions and was not affected by self-poisoning. However, as the amount of polymers built up upon deliberate poisoning at 385-488 following diene and acetylene treatments the n-butane selectivity gradually increased. The deliberate poisoning influenced only to a small extent the consumption of butadiene while the ratio : R1-B/R-BD (rate of l-butene consumption/rate of BD consumption) decreased from 3 to 0.1. Normal isotope effect of hydrogen addition was observed over both self-poisoned and deliberately poisoned catalysts : R-H/R-D was found to be less for n-butane formation than for ED consumption. Formation of n-butane on the catalyst has been interpreted by slow diene transport caused by the presence of oligomer layers on metal sites.