The method developing tor predicting !he values ot one or more variables from other variables is an imponant statistical problem in every branch of science This leads to the foundation ot relationship between two groups ot variables. The process ot maximizing the correlation between two linear function ot two sets of random variables is known as Canonical Correlation Analysis, Canonical Correlation Analysis is a statistical method which has been used to describe the re latıonshıp between a set o! independent and a set of dependent variables The aim of this study is estimate the canonical correlation coefficient between milk production trails (milk yield, lactation duration, dry period) and reproductive traits (age at first conception, calving interval, the number of insemination per conception, service period, calving season, calving year, gestation period) in 440 Holstein Cows. The data relating to milk and reproductive traits in the period of 1985-1992 ol a Holstein herd at the Bala State Farm was used. The STATISTICA 5.0 package programme was used tor applying this method The canonical correlation coefficient was estimated to be 0.931566 (p<0.0001) between milk and reproductive traits.