Measurements of genotype composition in a long-term experiment (182 days) with the marine phytoplankton species Chaetoceros affinis and Emiliania huxleyi, each consisting of nine genotypes (http://roscoff-culture-collection.org/; Emiliania huxleyi: GC22: RCC6383, C42: RCC6376, C41: RCC6375, C47: RCC4546, C96: RCC6381, C91: RCC6380, C35: RCC6374, C30: RCC6371, C48: RCC6377, Chaetoceros affinis: B63: RCC6347, B67: RCC6348, B68: RCC6349, B74: RCC6350). The species were cultivated together at three different nutrient regimes (10 N, 20 N, 30 N) with increasing nitrate supply. Nutrient concentrations were added as 9.0 ± 0.7 (10 N), 19.6 ± 0.8 (20 N), or 29.9 ± 0.9 μmoL-1 (30 N) nitrate and 0.9 ± 0.1 μmoL-1 phosphate. Silicate concentrations were aligned to reflect a 4:1 N:Si ratio. Cultivation was done in a semi-continuous batch cycle system with transfer of 2.88e7 ± 1.57e7 µm³ biovolume into a new batch every 7 days. Prior to the experimental start, genotypes were separately acclimated to experimental conditions with 20 μmol L−1 nitrate for 7 days. The genotype composition was assessed for each species after batch cycles 7,13, and 26 by using microsatellites.