Perennial ryegrasses (Lolium perenne L.) are one of the most valuable forage grasses, characterized by high productivity and good feed quality. Intensive cultivation of ryegrasses has resulted in reduced genetic diversity. Wild ryegrass ecotypes can be used in breeding programmes as parental genotypes with the aim to neutralize this genetic erosion. The objective of this study was to evaluate wild perennial ryegrass ecotypes from Ukraine under Lithuanian climatic conditions. Wild ecotypes were collected in Western and Northern Ukraine in 2008-2009 and planted in 2009-2010 In the collection trials the following parameters were assessed: winter survival, re-growth in spring, beginning of heading, plant height, length and width of flag leaf, inflorescence length and biomass. The parameters were estimated using the standard UPOV methodology (Anonymous, 2006), where a score of 9 represent very high and 1 is a very low value for the trait.The spring of 2010 was warm, with sufficient available moisture in the soil. The weather in June was cool. There were days in July-April months when air temperature rose above 30°C. The entire growing season was characterised by abundant rainfall.The temperature in the spring of 2010 was variable. The summer was very warm and rainy. Mean temperature in summer was 1.7°C higher than the long-term mean temperature. Soil moisture was sufficient during the season, except in June.The coefficient of variation (CV) and standard error (Sd) of agro-biological traits were estimated using computer software STAT adapted by P. Tarakanovas in Visual Basic of Application as a macro programme to run in EXCEL (Tarakanovas and Raudonius, 2003 Principal component analysis was carried out based on agro-biological characteristics. For traits in 2010, the first three principal components (PC) accounted for 58.67% of the total variation among the 27 studied genotypes (Table 3). PC1 accounted for 22.36% of total variation, which according to eigenvector values, was best explained by beginning of heading and biomass. PC2 accounted for 18.86% of total variation and was related to winter survival and plant height. PC3 accounted for 17.45% of total variation and was explained by re-growth in spring and inflorescence length. The ecotypes were grouped according to these measurements. Genotypes No. 3811 and No. 3823 diverged from the rest (Fig. 1 In 2011, the first three PCs accounted for 69.83% of the total variation (Table 4). PC1 explained 38.17% of total variation and was related to biomass and beginning of heading. Factors explaining PC2 (16.95% of total variation) were plant height and width of flag leaf. PC3 accounted for 14.71% of total variation; factors associated with this axis were length of flag leaf and winter survival. Genotypes No. 3887 and 3886 formed a distinct group in the dendrogram (Fig. 2). Both started heading earlier than all other genotypes, had short and narrow flag leaves Genotype No. 3887 was tall and high-yielding, while No. 3886 was short and low-yielding. DISCUSSIONHigh genetic div...