Anaplasma marginale infection in Europe has been limited to the Mediterranean and eastern countries, to Austria and to very sporadic cases in Switzerland. There are no reports of its occurrence in the countries north of Switzerland. A severe outbreak of anaplasmosis in August 2002 in a cattle farm in the canton Grisons, Switzerland, north of the Alps, with more than 300 cattle that had to be culled, came unexpected and gave reason to hypothesize presence of an increased yet undetected prevalence of A. marginale in Switzerland. Randomly selected bovine serum samples collected in 1998 and 2003 were tested using a competitive inhibitory ELISA (cELISA) to test the hypothesis. Our validation of the diagnostic sensitivity and specificity of this test, done in the outbreak herd, yielded 99.2 and 83.3%, respectively, probably underestimating the true specificity. The true seroprevalence of anaplasmosis in Swiss cattle determined by cELISA was likely to be zero with upper 95% confidence limits of 2.49% in the canton Grisons and 1.17% in the rest of Switzerland, respectively, in 1998. For 2003, these estimates were even lower. There was no significant difference in apparent prevalences between 1998 and 2003. In search of a possible reservoir, three chamoises out of 46 free ranging wild ruminants from the Swiss National Park, Grisons, tested positive in the cELISA. This reaction is in accordance with A. marginale or a cross reacting agent such as Anaplasma ovis. From our results we conclude that the hypothesis of an increased prevalence of anaplasmosis in cattle in Switzerland must be rejected.