Breeding perennial grasses for forage usage: An experimental assessment of trait changes in diploid perennial ryegrass (Lolium perenne L.) cultivars released in the last four decades

International audience

English. Breeding to improve perennial grasses for forage usage aims to increase the whole-plant aerial biomass. However, the ability to increase the whole-plant yield by breeding may be questionable in grasses which are expected to reach an optimal leaf area index during grass stand growth rather rapidly. We carried out a field experiment aiming to compare herbage yield and other performances of seven natural populations and 21 cultivars of diploid perennial ryegrass registered on European National lists in the last 40 years. Cultivars were sampled in the spike emergence earliness range used for pasture usage, i.e. in the semilate and late earliness range according to the French classification into spike emergence earliness groups. Forage performances of populations and cultivars were tested in four locations in monthly cut dense sward plots. Morphological and phenological traits were recorded in two spaced-plant experiments, and seed yield was assessed in two locations in standard seed production conditions. Trait regression on cultivar registration year pointed out that total dry-matter yield of cultivars increased by +3.2% per decade. However, dry-matter yield was not improved in the most favourable period for grass growth, i.e. in spring. On the other hand, it was improved in summer (+2.8% per decade) and autumn (+7.4% per decade) when the aerial biomass production is on average rather small. Breeders succeeded in significantly reducing aftermath heading. Reduction in aftermath heading is likely to have contributed to the noted increase in summer and autumn dry-matter yields, and to a moderate improvement in herbage feeding value. Breeding was also efficient in improving rust resistance and in bringing forward the start of spring growth. For monthly cut swards, we evidenced a clear association of leaf and lamina lengths with spring and summer dry-matter yields. Breeding for long leaves, or for high leaf elongation rate, should contribute to improve the cumulated intercepted radiation during re-growths, and consequently could be a way of improving the spring dry-matter yield. Seed yield did not improve with breeding for forage performances. However, we did not notice any negative association between the seed yield and traits related with forage usage. Direct selection pressures on seed yield criteria applied at the core of the breeding process should enable to improve the seed yield without any negative impact on forage performances. (C) 2011 Elsevier B.V. All rights reserved.