A winter wheat composite cross population (CCP), created in the UK in 2001, has been grown in Germany, Hungary and the UK since 2005 (F₅ generation). In 2008/9 (F₈), a cycling pattern for the populations was developed between partners to test the effects of rapidly changing environments on agronomic performance. One CCP was grown by eight partners for one year and subsequently sent to the next partner, creating “cycling CCPs” with different histories. In 2013, all eight cycling CCPs and the three non-cycling CCPs (from Germany, Hungary and UK) were included in a two-year experiment in Germany with three line varieties to compare agronomic performance and morphological characteristics. Differing seed weight of the F₁₃ at sowing affected some agronomic parameters under drought conditions in 2014/15, but not under less stressful conditions in 2013/14. In both experimental years, the CCPs were comparable to the line varieties in terms of agronomic performance, with some CCPs outyielding the varieties under drought conditions of 2015. The results highlight the potential of CCPs to compete with line varieties while the overall similarity of the CCPs based on their origin and cycling history for agronomic traits indicate a high buffering potential under highly variable environmental conditions.

Foliar diseases caused by fungal pathogens were assessed on 10 plants per plot at BBCH stages 45–55 (26 May 2014 and 29 May 2015) and BBCH stages 60–70 (16 June 2014 and 17 June 2015). Non-green leaf area (NGLA) was recorded as a percentage of the flag leaf (F), the leaf below the flag leaf (F-1), and, if possible, the F-2 leaf (first assessment only). The two main foliar leaf diseases observed were also recorded at each assessment date. Due to the strong similarity of symptoms between Stagonospora nodorum and Zymoseptoria tritici, these two pathogens were not distinguished separately from one another in the assessments. The occurrence of foot diseases was low in 2013/14 and close to zero in 2014/15, and as such, they were not considered relevant to the agronomic results.

Morphological assessments were performed on three 0.5-m rows of mature plants cut as close to the ground as possible from each plot (0.450 m² per plot). The number of ears for ear-bearing tillers per m² and the percentage of counted ears with awns (awned ears) were recorded. From these samples, total biomass and kernel weight were also recorded to calculate the harvest index (HI) and kernel number per ear. Combine harvested grain yields (t/ha) and thousand kernel weights (TKWs) (g) were adjusted to 14% moisture content. The TKW (g) was measured three times per plot (the weight of 1000 kernels), and the mean of these three measurements was used for analysis. Stem length (measured from base of stem to start of ear) and ear length (awn length not included) were recorded for 45 randomly chosen mature plants per plot (n = 180 plants in total per entry).

Protein content (%) was determined for all CCP entries and reference varieties in both generations and in both years. Protein content (%) was calculated according to seed nitrogen content (% N × 5.7), which was analysed using an Elementar Analysator vario MAX (Elementar Analysesysteme GmbH, Hanau, DE).