Transmission and survival of perennial ryegrass endophyte during field based seed production: A thesis submitted in partial fulfilment of the requirements for the degree of Master of Agriculture Science at Lincoln University

Agriculture in New Zealand is heavily centred on the use of perennial ryegrass (Lolium perenne) based pasture systems. However, like all forage plant species, perennial ryegrass is susceptible to pests and diseases. To help overcome these pests and diseases, perennial ryegrass has developed a symbiotic relationship with the specific fungal endophyte species, Epichloë festucae var. lolii (syn. Neotyphodium lolii). The New Zealand pasture industry uses E. festucae var lolii which colonise perennial ryegrass (Lolium perenne) and this interaction has been extensively researched due to the positive and negative attributes they cause. This endophyte is only transmitted vertically (i.e. from parent to seed) and the transmission process is not always absolute which can result in a decrease in endophyte percentage. One such endophyte strain known to have transmission issues is AR37. This decrease commonly occurs during the seed production process. However, it is not yet fully known why this decrease occurs. The overall aim of this research (Objective 1) was to determine where in the seed production process the decrease in AR37 endophyte percentage in perennial ryegrass occurred. To facilitate this aim, the seed production processes of two farms in Canterbury (Farm A and Farm B) were followed during the 2017-2018 seed production season. The two farms were chosen based on their seed production history. Farm A was known to consistently produce seed with high (over 70%) endophyte levels, while Farm B was not as consistent. Both farms grew the same seed line (cv. Governor) containing the same endophyte strain (AR37). The management practices used to produce the ryegrass seed differed between the two farms. The most notable differences occurred during fungicide application and harvest. Farm A carried out six fungicide applications whereas Farm B only carried out three. Farm B also stored around 10% of the harvested seed in a tractor trailer for the duration of the harvest period (2 days). No significant decrease in endophyte percentage occurred for the seed stored in the trailer. This result can help assure seed companies that, if a grower must store seed under poor conditions for short periods, no loss in endophyte should occur. More research would be needed to further confirm this, such as repeating the experiment with various seed lines. No significant decrease in endophyte percentage occurred during the entirety of this experiment. However, it was a good harvest season for endophyte, in that no obvious loss in endophyte occurred compared to other seasons, so it is possible that losses may be seen during other harvest seasons. More research would be required to determine this. Fungicides are commonly used during ryegrass seed production to reduce the effect of pathogenic fungi such as Gloeotinia temulenta which causes blind seed disease. However, fungicides also have the potential to harm beneficial fungi such as endophyte. Two controlled experiments were set up for objective two to determine if fungicides commonly used during ryegrass seed production had any effect on endophyte level. The first experiment showed that application of benzimidazole (Protek), tebuconazole (Folicur), prothioconazole (Proline) and azoxystrobin (Amistar) fungicide to Alto AR37 ryegrass plants did not affect endophyte level. However, in this experiment the initial >70% endophyte level of seed (from laboratory analysis) was unexpectedly below 20% (on average) when tillers were tested as a control grow out during the experiment. Therefore, an additional experiment was set up to determine why the endophyte level dropped between the laboratory test and the tiller tests carried out during the experiment. In this second experiment the endophyte level was maintained at over 90% on average, so the variables tested (potting mix, fungicides, and pots used as well as area grown) were not those responsible for the drop in endophyte detection. Different seed lots were used for both experiments which could explain this difference seen in endophyte percentage. Another possible reason is that the plants in the first controlled experiment may have been exposed to different growing conditions, such as nutrient stress or water stress, which hindered colonisation of the endophyte in emerging tillers. The final objective involved observing the effect that temperature, humidity and storage length had on endophyte viability. It was found that as the storage length increased, to 16 weeks, the viability of the endophyte found in seed (seed endophyte) decreased by up to 66% (14% moisture), particularly at higher temperature (20.3-31.2°C) and humidity levels (65.1-71.1% RH). No major differences were observed in the first two experiments indicating that the key factors investigated (fungicide application and harvest methods) are not the sole reason for loss of endophyte transmission. This combined with other interesting but not tested observations has provided ideas for future research (such as testing the influence of climatic conditions on endophyte growth) and has helped to confirm that certain field environment and management factors, such as fungicide application, have no effect on endophyte viability