Experiment was conducted at 15 certified organic dairy farms in Eastern Finland during spring 2008. Farms were allocated to three balanced groups according annual milk yield, herd size and housing system. Three types of selenium (Se) supplementations: selenium yeast (SY), sodium selenite (SS) or their (50:50) combination (SYS) was used. Daily Se doses were adjusted according to milk yield basis (average 3 mg/d). Experimental model was 3 X 3 Latin square, consisting from 3 experiment periods. The length of each period was 28 days. RehuRaisio provided experimental feeds. In the end of each period bulk-tank milk samples were collected and analysed for Se at MTT Agrifood Research Finland. The Se supplementation form was crucial for milk Se concentration (P<0.001), when Se concentration of basic fodder was extreme low (<0.01 mg/kg DM). SS (n=15) resulted to milk Se concentration of 0.012 mg/kg, but SYS (n=15) resulted to 1.9-fold and SY (n=14) to 2.3-fold higher milk Se concentrations.