The experiment was carried out at the Department of Agriculture at the State University of Londrina, Londrina, Paraná, in June and July 1999, to develop a separation method for camomile seeds using the mass difference between empty and fully grained seeds. Solutions of nonil phenoxi poli (ethilenoxi) ethanol (200 g i.a/l) non-ionic spreading adhesive in deionized water at concentrations of 2500; 625; 156.25; 39.0625; 9.76563; 2.44141; 0.61035; 0.15259 and 0.0mg/l (blind test). The supernatant (empty seeds and impurities) and the submerged mass (only fully grained seeds) were assessed. A complete randomized design was used with four replications. The experimental data was submitted to analysis of variance, polynominal regression and means comparison by the Tukey test. The concentrations only differed statistically (p<0.01) for the blind test showing the high efficiency of the product in camomile seed separation, even when working with the minimum dose (0.15259mg/l).