Forty-eight primiparous and pregnant Holstein cows at 202 (+/-59) days in milk and average daily milk production (DMP) of 22 (+/-5.3) kg at the beginning of the trial were used to study the possibility of identifying animals that are more responsive to recombinant bovine somatotropin (rbST). Animals were housed in free-stall barns, milked three times a day and fed a total mixed ration. Animals were injected with one dose (500 mg each) of rbST every 21 days during a 105-day total period (five doses in all). Daily milk production was recorded to measure average apparent response (AAR) and peak of apparent response (PAR). Average apparent response was defined as the average increase in milk production, expressed as kilogram per cow per day, for days 0-18 after dosing, as compared to the average production calculated for days -1, -2, -3, 19, 20 and 21 after dosing. Peak of apparent response was calculated as the increase in milk production for each day from 0 to 18 days after dosing as compared to the average production for days -1, -2 and -3. There was a positive linear correlation (r = 0.608; p = 0.0001) between AAR for the first 21 days and AAR for the total 105-day period (AATR). A correlation coefficient was calculated between each individual PAR value and AATR. The highest correlation coefficient (r = 0.511; p = 0.0002) was observed for days 8, 9 and 10, and therefore these days were considered the best estimates for calculating AAR when DMP records are not available. Also, high somatic cells count had a negative impact on milk yield response to treatment (p = 0.0162).