The spreading of urban sewage sludge is a constant cause of controversy, in particular as this practice is known to supply amounts of potentially toxic trace metals to the soil, particularly cadmium. In order to clarify some parts of this debate, this article presents a synthesis of the results of many studies carried out in France on the impact of sewage sludge spreading: old experiments as well as more recent trials. This paper deals mainly with cadmium but sometimes some interesting results relating to other trace metals are reported. Presented data are relative to: (i) the soil quality such as total trace metal contents and results of partial extraction aimed at determining the most phytoavailable fractions; (ii) the composition of cereal grain, chiefly wheat. Depending upon the period, the experiments were completely different as far as the tonnages of sludge applied and the quantities of cadmium involved. Three categories of trials stand out: (1) During the 1970s and 80s, sludges with a high trace metal content, especially cadmium, were spread at one experimental farm on sandy soils and in the Vexin area on silty topsoils. In all cases, a notable long-lasting impact was found on the total Cd content of the topsoil and the composition of cereal grains. The quantities of applied Cd were very high, from 3,600 g ha -1 to 641,000 g ha -1. (2) Sludges containing high amounts of industrial Ca were spread on acid soils in the Limoges region for more than 20 years. Increases of the cadmium content in the topsoil were only measured in fields where the cadmium input was the highest, of 300 to 600 g per ha. A clear increase in the Cd content of cereal grains has been noted as well. (3) During the 1990s and 2000s, numerous experiments with sewage sludge applications compatible with the French regulations were implemented. Local authorities launched many, relatively light, experiments in diverse areas and on diverse soil types. Research institutions carried out other, more exacting, trials. The quantities of Cd input varied from 0.6 to 270 g ha -1. In these cases, no impact was measurable on the composition of cereal grains.