The aim of this paper is to investigate whether characteristics of the first lactation (FL) curve of Iranian Holstein cows are associated with survival. Cows with least 10 test-days of milk production in their FL were used. The persistency of lactation (PL) and survival were estimated using a random regression model by restricted maximum likelihood with the ECHIDDNA software. We also used the Wood model to parameterize each individual lactation curve and then analyzed various curve characteristics using an animal model. The predicted breeding value (EBV) of the characteristics of the lactation curve of the cows from day 40 to 305 was predicted. The EBV of the production range (PR) and the slope of line in increasing phase (m_40,Peak) of production curve of sires with higher survival EBV were lower than other sires (P < 0.05). The estimates of PL were independent of survival estimate. Therefore, the PR from 40th day after calving can be considered as a definition of PL because the lower the PR, the flatter is the milk production curve. Genetic evaluation of young bulls for survival needs the data of death or culling of their daughters. Therefore, the bulls can genetically be evaluated for survival according to the PL and m_40,Peak of FL information of their daughters.