**4. Discussion**

The farming system affected the seed yield of mixtures by approximately 24% in favor of the mixtures grown conventionally, compared to those grown organically. These findings are consistent with several other studies [4,37–43] and result mainly from the direct growth- and yield-promoting effects of mineral nutrition of crops in the conventional system. However, Schram et al. [44] underline that crop yield differences between farming systems diminish with time; after 13 years, they amount to only 13% in favor of the conventional system over the organic one.

A detailed analysis revealed that the mixtures' components, namely oats and common vetch, reacted differently to agricultural production intensification. The share of vetch seeds in the seed yield, number of pods per m2, and the thousand-seed mass of vetch were higher in an organic farming system. Reversely, oats yielded well in the conventional system. Under stressful conditions of a limited supply of soil resources, the legume component performs better than the cereal one, leading to the resilience of a total mixture yield [45]. Due to an extensive root system, legumes can activate phosphorus from organic

compounds in the soil, mostly unavailable to cereals [46]. Moreover, they also use biologically bound nitrogen assimilated by the *Rhizobium* bacteria [47]. This effect clearly shows a complementarity of the components of the oat-vetch mixture. A proper selection of cereal components for mixture with a legume is of significance in this context. The interaction of oat cultivars and farming system variant, and the oat cultivar and year were observed in our study for almost all of the analyzed plant and canopy traits. In general, oats cv. Celer turned to be more competitive toward vetch in the mixture as compared to oats cv. Grajcar. Interestingly, both oat cultivars tested in this study were characterized by their breeder as having a very similar set of traits, i.e., time to ripening, thousand-grain mass, and plant height. The competitive effort of oat cultivars toward vetch was related to their productivity traits—specifically, even though oats cv. Grajcar developed more tillers in the mixture, as compared to cv. Celer they were less productive, i.e., displaying a lower density of panicles per m<sup>−</sup>2, a lower number of grains per panicle, and a lower thousand-grain mass. Contrarily, vetch was more productive in the mixture with cv. Grajcar as reflected by a higher number of vetch pods per m2, seeds per pod, and a share of vetch seeds in the mixture's yield, compared to the mixture with cv. Celer. As a result, even though both mixtures had a similar total yield during the years of study, the mixture of oats cv. Grajcar and vetch cv. Hanka had a more optimal share of oats/vetch seeds in the yield than the mixture with cv. Celer. Noteworthy was the finding of the negative correlation of the mixture yield with the number of pods and percentage of vetch seeds in the mixture yield. The greater the yield of the mixture, the lower the percentage of vetch seeds. Contrarily, the lower the mixture's yield, the greater the number of vetch pods per unit area. Both findings indicate strong competitive effects of oats toward vetch. Only a few studies discuss the influence of oat cultivar selection on the yield of the oat-vetch mixture, e.g., [48]. In our previous studies, we have shown that the oat cultivar is crucial for a good vetch yield, which is also influenced by the type of soil [49]. The share of vetch seeds in the mixture with oats is variable and influenced by several factors [50–52]. The main restrictions are weather conditions during the growing year. With low rainfall, vetch cannot withstand competition for water with oats, and its share in the yield is smaller [51–53].

In general, the leaf area index, which relates to the leaf assimilation area, and the leaf relative chlorophyll content (SPAD) were higher for the mixtures grown in the conventional system compared to the organic one. The LAI and chlorophyll content measured in SPAD values are good indices of the crop canopy status; many authors confirm their usefulness for estimating crop yields [54,55].

The results of the canonical analysis, performed for all of the tested factors, revealed that weather conditions were the main driver affecting the performance of the mixtures. The best year turned out a regular year, namely 2014. In 2012, a severe drought occurred in May and later in July, whereas June was very humid. In that year, regardless of oat cultivar and farming system variant, oats over-compete vetch by developing a significantly higher number of panicles than in 2013 and 2014. This shows that both oat cultivars tend to redistribute assimilates to produce higher grain yields in stressful conditions. This finding agrees with Zao et al. [56], who found a similar phenomenon in oats cv. Bia. According to those authors, under moderate drought stress there is a decreased biomass distribution to stems and leaves and a greater grain yield of oats. On the other hand, in 2013, when an excess of precipitation occurred in May and June and a severe drought in July and August, the share of vetch seeds in the mixtures' yields was the highest and for oats this yield was the lowest. These results confirm the benefits of cultivating mixtures, namely maintaining a high yield of at least one mixture component in years with weather unfavorable for the other component of the mixture [57].
