CORRELATION ANALYSIS IN BREAD WHEAT (Triticum aestivum L.) GENOTYPES

The research was carried out 2017-2020 in the Absheron Scientific - Experimental (irrigation) area of the Institute of Genetic Resources of ANAS. In the study were used 90 bread wheat ( Triticum aestivum L. ) genotypes of Azerbaijani origin, kept in the National Genebank, which 13 of them are commercial varieties. This study aims to identify selection criteria to be considered in plant selection studies with Correlation, PCA analysis and Statistical analysis of a number of indicators of bread wheat genotypes. In practice were studied the biomorphological indicators of wheat samples, assessment of yellow rust infection in bread wheat genotypes with a modified Cobb scale under a natural epidemic and and statistical analysis of the data was performed in SPSS computer programs. The correlation and PC analysis were computed between plant height, number of plants, peduncle length, spiklets per spike, spike weight, number of grains per spike, grains per spike, number of grains per plant, grain yield per plant, 1000 grain weight and the rate of yellow rust infection. KMO and Barlett tests are used to perform PC analysis correctly, and KMO = 0,786 was obtained in the study. According to our results, the total variation of the five components was 77,5%. The prices of the first three PCs, which accounted for 61.4% of the variability, were 38,5%, 14,2% and 8,7%, respectively. Correlation was calculated for two traits and as a result of this analysis, highly significant correlation was found between grain yield per plant and other following components : plant height ( r=0,332), spiklets per spike ( r=0,678), number of grains per spike (r=0,813), grains per spike (r=0,868). Although there is no high correlation between the data of spike and the rate of yellow rust infection, it has been shown negative significant correlation ( r =-0,288 ). The information obtained as a result of the assessment may be the starting material for research in this area.