b'crossings. The breeding is much more interesting compared to breeding other crops because we get much more surpris-ing results, he says. Its also interesting because there are many open questions in this species which are not completely explained yet through research. Intensive evaluation is also the norm at Columbia River Seeds/Novel Ag. Gyllenberg notes that every line has dif-ferent levels of apomixis. It is more diffi-cult with some accessions than others to generate hybrid filial genera. Breeding P. pratensis is not for the faint of heart, he says. We require huge amounts of space and are unable to achieve as much genetic diversity per crossing event than some of the less-complicated crosses. This makes the breeding decisions much higher stakes, since the opportunity costIsolation cages with Poa pratensisin lost time and space requirements can. Source: DLFbe very real. Brentano adds that when workingally-reproducing plants. Grieder publishedwith other turf species, these two traits toward development of any new culti- a paper on this with colleagues in 2016 inare often negatively correlated.var within a facultative apomict species,the journal Agrarforschung Schweiz (only each and every new single plant neces- available in German or French).MEETING DUS STANDARDSsarily must be better than the parents,He adds that because they have trans- With all the breeding challenges presented as compared to open-pollinating speciesformed their P. pratensis gene pool toby P. pratensisand its large degree of where you target the shift of a populationsexual reproduction, for them breeding isvariabilityits not easy for breeders to bell curve. Essentially, youre going toquite similar to other crops except thatmeet the Uniformity and Stability require-see many individual plants that appearthe introgression of new genetic resourcesments for DUS purposes. desirable and do this or that, he notes.is more challenging. Plants from new- According to van der Aa, varieties But many fall short and are dropped out.ly-collected ecotype populations first havealso have to pass DUS registration and Austin Fricker says he and his colleaguesto be checked for their type of reproduc- because Smooth meadow grass is apom-at Pure Seed Testing use certain motherstion, he says. Interesting traits fromict, the varieties have to be very homo-that produce more apomictic hybrids. apomictic materials can only be intro- geneous. That requires a lot of selection However, van der Weijde does notgressed by using the apomictic plant aswork and single plants are controlled for see apomixis as necessarily a complica- father, followed by intensive re-selectionapomixis and uniformity, he says.tion, but very attractive in that a breederof offspring for sexual seed formation.Gyllenberg notes however that insta-can pick a single desirable plant andDemands regarding homogeneity for sex- bility in P. pratensis is quite obvious, and produce seed from it to make a variety.ually-reproducing varieties are lower com- genetic stability can, for the most part, be Because its not fully apomictic, you canpared to apomictic varieties.quickly identified early on in the breeding find attractive off-types and you can evenprocess. make crosses between elite parents andOTHER BREEDING CHALLENGES For his part, Schumann notes that if find some offspring from sexual reproduc- Poa annua weed infestation also presentsyou have a high apomictic type with good tion, he explains.a major challenge in P. pratensis, in bothquality, the variety will be stable. Van der established turf trials and in the breedingWeijde and his team, to test the final vari-USING SEXUAL REPRODUCTION nurseries, explains van der Weijde. Onlyeties for uniformity and stability, exam-DSP/Agroscope has gone a different route.one herbicide (Hussar) is effective againstine 50 plants and advance only those with Over the last two decades, P. pratensisPoa annua and not against P. pratensis,fewer than two off-types.breeders there have shifted towards usingand it can only be used once a year. This is sexual-reproducing maternal plants (iden- especially a problem because P. pratensisEditors Note: Part 2 of this article will tified by the high level of variability inis a bit slow to establish, while Poa annuaexamine breeding P. pratensis for dis-their offspring). Working within this geneis a notorious weed that spreads its seedsease resistance, access to genetics and pool, recurrent selection can be performedvery fast and start growing at low tem- innovation.as it is for other grass species to continu- peratures.ously increase the frequency of favourableIndeed, van der Aa considers the alleles, Grieder explains. Analysing dataslow establishment/germination of Poa to from the past 30 years, weve shown that nobe its biggest breeding challenge. Good breeding gain was obtained when workingestablishment is very important for theIf you wish to learn more about breeding for with apomictic materials (its like search- use of this grass in sport mixtures withclimate change and the growth markets of ing for the needle in haystack), but a con- turf Perennial ryegrass, and to combinesmooth stalked meadow grass, make sure tinuous increase in performance has beenexcellent turf quality with high seed yieldto check out the full version of this article on gained since we began working with sexu- is also a challenge, he explains. And, asour website: www.european-seed.comEUROPEAN-SEED.COMIEUROPEAN SEED I 17'