There is an urgent need for agricultural systems to intensify sustainably, increasing crop productivity, farmer livelihoods and soil health while using fewer resources. Crop perennialization, the conversion of especially annual grains to perennial forms, has shown such possibility. Here we report the successful breeding of perennial rice and assess its performance and potential. Domesticated, annual Asian rice (Oryza sativa) was hybridized with its perennial African relative Oryza longistaminata. From a single planting, irrigated perennial rice produced grain for eight consecutive harvests over four years, averaging 6.8 Mg ha⁻¹ harvest⁻¹ versus the 6.7 Mg of replanted annual rice, which required additional labour and seed. Four years of cropping with perennial rice resulted in soils accumulating 0.95 Mg ha^–1 yr^–1 organic carbon and 0.11 Mg ha⁻¹ yr⁻¹ nitrogen, along with increases in soil pH (0.3–0.4) and plant-available water capacity (7.2 mm). Perennial cultivars are strongly preferred by farmers; growing them saves 58.1% of labour and 49.2% of input costs in each regrowth cycle. In 2021, perennial rice was grown on 15,333 ha by 44,752 smallholder farmers in southern China. Suited to a broad range of frost-free environments between 40° N and 40° S, perennial rice is a step change with potential to improve livelihoods, enhance soil quality and inspire research on other perennial grains.

Unilateral introgression from diploids to tetraploids has been hypothesized to be an important evolutionary mechanism in plants. However, few examples have been definitively identified, perhaps because data of sufficient depth and breadth were difficult to obtain before the advent of affordable high-density genotyping. Throughout Japan, tetraploid Miscanthus sacchariflorus and diploid Miscanthus sinensis are common, and occasionally hybridize. In this study, 667 M. sinensis and 78 M. sacchariflorus genotypes from Japan were characterized using 20 704 SNPs and ten plastid microsatellites. Similarity of SNP genotypes between diploid and tetraploid M. sacchariflorus indicated that the tetraploids originated through autopolyploidy. Structure analysis indicated a gradient of introgression from diploid M. sinensis into tetraploid M. sacchariflorus throughout Japan; most tetraploids had some M. sinensis DNA. Among phenotypically M. sacchariflorus tetraploids, M. sinensis ancestry averaged 7% and ranged from 1–39%, with introgression greatest in southern Japan. Unexpectedly, rare (~1%) diploid M. sinensis individuals from northern Japan were found with 6–27% M. sacchariflorus ancestry. Population structure of M. sinensis in Japan included three groups, and was driven primarily by distance, and secondarily by geographic barriers such as mountains and straits. Miscanthus speciation is a complex and dynamic process. In contrast to limited introgression between diploid M. sacchariflorus and M. sinensis in northern China, selection for adaptation to a moderate maritime climate probably favoured cross-ploidy introgressants in southern Japan. These results will help guide the selection of Miscanthus accessions for the breeding of biomass cultivars.

Miscanthus is a perennial C4 grass that is a leading potential feedstock crop for the emerging bioenergy industry in North America, Europe and China. However, only a single, sterile genotype of M. × giganteus (M×g), a nothospecies derived from diploid M. sinensis (Msi) and tetraploid M. sacchariflorus (Msa), is currently available to farmers for biomass production. To facilitate breeding of Miscanthus, this study characterized genetic diversity and population structure of Msi in its native range of East Asia. A total of 767 accessions were studied, including 617 Msi from most of its native range in China, Japan and South Korea, and 77 ornamental cultivars and 43 naturalized individuals from the USA. Accessions were evaluated with 21 207 restriction site-associated DNA sequencing single nucleotide polymorphism (SNP) markers, 424 GoldenGate SNPs and ten plastid microsatellite markers. Six genetic clusters of Msi from geographically distinct regions in Asia were identified. Genetic data indicated that (1) south-eastern China was the origin of Msi populations found in temperate eastern Asia, which is consistent with this area probably having been a refugium during the last glacial maximum (LGM); (2) Msi migrated directly from south-eastern China to Japan before migrating to the same latitudes in China and Korea, which is consistent with the known sequence of warming post-LGM; (3) ornamental Msi cultivars were derived from the southern Japan population, and US naturalized populations were derived from a sub-set of the ornamental cultivars; and (4) many ornamental cultivars previously described as Msi have hybrid ancestry from Msa and Msi, whereas US naturalized populations of Msi do not. Population structure of Msi was driven by patterns of warming since the LGM, and secondarily by geographical barriers. This study will facilitate germplasm conservation, association analyses and identification of potential heterotic groups for the improvement of Miscanthus as a bioenergy crop.

Miscanthus × giganteus (Mxg) is an important bioenergy feedstock crop, however, genetic diversity among legacy cultivars may be severely constrained. Only one introduction from Japan to Denmark of this sterile, triploid, vegetatively propagated crop was recorded in the 1930s. We sought to determine if the Mxg cultivars in North America were all synonyms, and if they were derived from the European introduction. We used 64 nuclear and five chloroplast simple sequence repeat (SSR) markers to estimate genetic similarity for 27 Mxg accessions from North America, and compared them with six accessions from Europe, including the species’ type‐specimen. A subset of accessions was also evaluated by restriction‐site associated DNA sequencing (RAD‐seq). In addition, we assessed the potential of new crosses to increase Mxg genetic diversity by comparing eight new triploid Mxg progeny grown from seed, along with samples of the parental species M. sacchariflorus and M. sinensis. Estimates of genotyping error rates were essential for distinguishing between experimental error and true genotypic differences among accessions. Given differences in estimated error rates and costs per marker for SSRs and RAD‐seq, the former is currently more cost‐effective for determining if two accessions are genetically identical. We concluded that all of the Mxg legacy cultivars were derived via vegetative propagation from a single genet. In contrast with the Mxg legacy cultivars, genetic similarity to the type‐specimen of eight new triploid Mxg progeny ranged from 0.46 to 0.56. Though genetic diversity among the Mxg legacy cultivars is critically low, new crosses can provide much‐needed variation to growers.

A network of 21 experiments was established across nine countries on four continents and spanning both hemispheres, to evaluate the relative performance of early generation perennial cereal material derived from wheat, rye, and barley and to inform future breeding strategies. The experimental lines were grown in replicated single rows, and first year production and phenology characteristics as well as yield and persistence for up to three years were monitored. The study showed that the existing experimental material is all relatively short-lived (≤3 years), with environments that are milder in summer and winter generally conferring greater longevity. No pedigree was superior across this diverse network of sites although better performing lines at the higher latitude sites were generally derived from Thinopyrum intermedium. By contrast, at lower latitudes the superior lines were generally derived from Th. ponticum and Th. elongatum parentage. The study observed a poor relationship between year 1 performance and productivity in later years, highlighting the need for perennial cereal material with greater longevity to underpin future experimental evaluation, and the importance for breeding programs to emphasize post-year 1 performance in their selections. Hybrid lines derived from the tetraploid durum wheat generally showed greater longevity than derivatives of hexaploid wheat, highlighting potential for greater use of Triticum turgidum in perennial wheat breeding. We advocate a model in future breeding initiatives that develops perennial cereal genotypes for specific target environments rather than a generic product for one global market. These products may include a diversity of cultivars derived from locally adapted annual and perennial parents. In this scenario the breeding program may have access to only a limited range of adapted perennial grass parents. In other situations, such as at very high latitude environments, perennial crops derived from barley or rye may have a better chance of success than those derived from wheat. In either case, development and selection of the perennial parent for adaptation to local environments would seem fundamental to success.