Cercospora rosicola is a fungal pathogen that attacks roses and causes spotting on leaves, chlorosis and in severe cases, defoliation. This disease has become more prominent in roses after the widespread use of black spot resistant roses, most likely due to the reduction of fungicide applications on roses. The objective of this thesis is to evaluate garden roses and identify QTLs for resistance to cercospora leaf spot disease. Identification of susceptible and resistant roses to cercospora leaf spot is the first step for breeding cercospora resistant rose cultivars. Cercospora leaf spot was evaluated on a percentage-based rating scale of 0-9 (0=no disease and 9=foliage covered with lesions) during spring, summer and fall of 2016 through 2018 on 130 roses in Overton and College Station, TX. Average cercospora incidence in Overton increased from 2016 to 2017, but decreased in 2018, whereas in College Station, ratings increased three-fold from 2016 (0.6) to 2018 (2.2). Most cultivars showed tolerance to the disease, particularly Rosa rugosa hybrids. However, a number of cultivars including 'American Pillar', 'John Davis', 'Carefree Delight', 'Oso Happy Candy Oh', 'Oso Easy Cherry Pie' and 'Roxanne Veranda' were highly susceptible to cercospora. In College Station, moderately high repeatability and low coefficient of variance occurred in the later months of 2017 and 2018, while in Overton, all the evaluated months in 2016 except for April showed high repeatability and low coefficient of variance. This suggests that ratings during these months are consistent, and may be the more informative months for disease evaluation. Artificial inoculation in a greenhouse setting was also attempted to identify cercospora leaf spot resistant garden roses. Although the pathogen was successfully cultured, the low sporulation, high heat and low humidity during the inoculation period slowed disease development. Thus, severe disease symptoms were not observed for this experiment. In 2016, fifteen diploid inter-related rose populations were evaluated for cercospora leaf spot in June, September, October and November in College Station. The estimated broad sense and narrow sense heritability were 0.83 and 0.57, respectively. A pedigree-based analysis using Visual FlexQTL software was conducted on these populations. QTLs found on LG1, LG3 and LG7 appeared in different environments. In the overall analysis, QTLs found on LG1 (0-4cM) and LG3 (36-42cM) explained 8.5% and 7.7% of the total phenotypic variance, respectively. More studies are needed to improve the strength and consistency of QTL detection in this analysis.
