Abstract: 【Objective】In order to determine the processing tomato varieties trait evaluation, a comprehensive evaluation system for processing tomato agronomic traits was established. The systern can be used for providing decision support for identification and screening of its excellent resources.【Method】Through a field experiment in Xinjiang in 2012, 22 processing tomato varieties (lines) were tested and analyzed, 17 agronomic traits of the 22 processing tomato varieties were measured. These 17 agronomic traits include fruit weight (FW), average yield (AY), lycopene (L), soluble solids (TSS), total acid (TA), total sugar (TS), sugar-acid ratio (SAR) , virus disease resistance (VDR), longitudinal diameter (VD), transverse diameter (TD), fruit shape index (FSI), color (CR), single fruit resistant to pressure (SFRP), the average spacing width (ASW), average height (APH), the average number of branches (ANB), and the growth period (GP). The principal component analysis and cluster analysis (PCA-CA) were also used to conduct a comprehensive analysis of the main agronomic traits of processing tomato after Matlab.2012a software programming for the sample data processing.【Result】The average coefficient of variation of 17 agronomic traits was 18.54%, the virus disease resistance was the largest and the coefficient of variation was 88.42%, and the smallest coefficient of variation of the color difference was 7.33%. The processing tomato evaluation variables were relatively independent but closely related. The 17 evaluation variables could be compressed into six indicators (the first six principal components cumulative contribution rate is 86.0369%, which reflected most of the 17 traits information) ,after analysis of 17 agronomic traits coefficient of the six main functional components corresponding to 17 agronomic traits could be summarized as fruit traits factor, quality factor inherent fruit, fruit appearance quality factor, yield factors, disease factors and the five major indicators could be used to analylize the quality of tomato, fruit resistance to pressure, weight, total acid, lycopene, longitudinal and transverse diameters, the average yield, disease-resistant viruses are the main characters of 8 traits. By using the class average of 22 different tomato varieties Q-type cluster analysis, all tomato varieties could be divided into three groups when Euclidean distance was 6.40.【Conclusion】The principal component analysis and cluster analysis were used to analyze 17 agronomic traits of the 22 processing tomato varieties (lines) to establish a comprehensive evaluation system for processing tomato. It can give a more comprehensive and objective evaluation and analysis from different perspectives and provide a reference for high-quality processing tomato cultivar selection.

Key words: processing tomato , agronomic traits , principal component analysis , cluster analysis , evaluation index