Diagrammatic scale for severity evaluation of the Eremanthus erythropappus–Puccinia velata pathosystem

• Published in: Journal of phytopathology Vol. 168; no. 3; pp. 135 - 143
• Main Authors: Martins Pereira, Renata Cristina, Alencar Laignier Filho, Cézar, Amaral de Melo, Lucas, Alozen, Paulo Cesar, Gonçalves Mafia, Reginaldo, Ferreira Barros, Aline, Alves Ferreira, Maria
• Format: Journal Article
• Language: English
• Published: Berlin Wiley Subscription Services, Inc 01.03.2020
• Subjects: candeia; Cosmetics; Cosmetics industry; Epidemiology; Essential oils; Evaluation; Indigenous species; Leaf area; Lin's method; pathometry; Puccinia; Regression analysis; Reproducibility; rust; Rust fungi
• ISSN: 0931-1785; 1439-0434
• DOI: 10.1111/jph.12855

Candeia (Eremanthus erythropappus (DC.) MacLeish), a native forest species from South America, has garnered commercial interest due to its production of essential oil that contains alpha‐bisabolol. This compound is widely used in the pharmaceutical and cosmetics industry, with approximately 80% of Brazilian production being exported. Since candeia rust (Puccinia velata) has only been reported in Brazil, little is known about its epidemiology and control. There is no methodology to quantify rust severity in candeia, justifying the elaboration and validation of a diagrammatic scale containing eight levels of disease severity based on leaf area coverage (0.25%, 0.5%, 1%, 2%, 4%, 8%, 16% and 32%). In a natural sampling of disease in the field, 95% of the leaves showed severity below 16%, with the remaining 5% showing severities between 16% and 32% of leaf area. Validation of the proposed diagrammatic scale was performed by assessing the results from 10 inexperienced evaluators, performing evaluations of three leaves with different severity levels. The evaluations were performed at 7‐day intervals; in the first instance, severity values were assigned without the diagrammatic scale, and for the second attempt, the scale proposed in this study was used. The accuracy and precision of the severity estimates produced by each evaluator compared to the real severity was analysed by linear regression and by Lin's statistics. The reproducibility of the estimates was evaluated by analysing the coefficient of determination of linear regressions by pairs of evaluators. The scale provided adequate levels of accuracy, precision, repeatability and reproducibility, indicating the proposed scale was a suitable method for quantifying the severity of candeia rust.