Nibler and Turner (1929) and Turner et al. 1930 have reported relatively high biological estrogenic activity in urine from pregnant cows. Smith et al. 1956 have shown that fluorescence-producing compounds increased from early to late stages of gestation. They also showed that this chemical estimate for phenolic steroids was approximately four times that shown by bioassay for estrogenic activity on the same aliquots of urine. Turner et al. 1930 reported an average of 57 rat units of estrogenic activity per day in the urine of cows preceding conception. Asdell et al. 1945, using a bioassay sensitive to 18 rat units, was unable to find estrogenic activity in the urine of heifers in estrus. Since Smith (1954) and Gorski (1956) have previously observed variable quantities of fluorescence-producing compounds in non-pregnant cow urine, it was the purpose of this experiment to estimate the presence of biologically active estrogens in the urine of non-pregnant cows. Twenty-four hour samples of urine were collected, using the apparatus described by Gorski et al. 1957. The sources of the urine samples are listed in table 1. Aliquots of the urine samples were acid hydrolyzed and extracted with ether. The acidic elements were removed by methods adopted from Brown (1955). The neutral steroids were then separated according to the method of Friedgood (1948) from the phenolic estrogen-containing fraction (phenolic steroids). Preliminary work showed that further purification was necessary to remove certain toxic elements. The steam distillation procedure developed by Bassett et al. (1955) to remove p