HORTSCIENCE, VOL. 35(2), APRIL 2000 184 CROP PRODUCTION HORTSCIENCE 35(2):184–186. 2000. Received for publication 19 May 1998. Accepted for publication 14 Apr. 1999. Contribution number 335/98.05.02R of the Horticultural Research and Development Centre, Agriculture and Agri-Food Canada. The authors gratefully acknowledge the funding provided by Horizons sciences program of Agriculture and Agri-Food Canada. The cost of publishing this paper was defrayed in part by the payment of page charges. Under postal regulations, this paper therefore must be hereby marked adver- tisement solely to indicate this fact. 1 Postdoctoral Fellow, Physiologist. 2 Research Scientist and Adjunct Professor, Breeder and Physiologist; to whom reprint requests should be addressed (e-mail: khanizadehs@em.agr.ca). 3 Research Scientist, Postharvest Physiologist. 4 Research Scientist, Ornamental Nursery Manage- ment. Assessing Chilling Tolerance in Roses using Chlorophyll Fluorescence Nadia Hakam 1 , Shahrokh Khanizadeh 2 , Jennifer R. DeEll 3 , and Claude Richer 4 Agriculture and Agri-Food Canada, Horticultural Research and Development Centre, 430 Boulevard Gouin, St-Jean-sur-Richelieu, QC, J3B 3E6, Canada Additional index words. spring frost, hardiness, Rosaceae, low temperature injury, winter damage, nondestructive method, Rosa kordesii, R. hybrida, R. rugosa, R. arkansana Abstract. Chlorophyll fluorescence (CF) was evaluated as a technique to assess chilling injury of rose (Rosa sp.) leaves exposed to low temperatures. In the more susceptible genotypes, variable fluorescence (Fv) decreased dramatically as the temperature was lowered. In the less susceptible genotypes, Fv was more stable and decreased more slowly as temperature fell. Our results suggest that measurement of CF may provide a rapid method to prescreen genotypes for chilling susceptibility, as required in plant breeding. Chilling injury can be defined as damage occurring at temperatures near freezing, and can affect plants in the field in early spring and late fall. Plants sensitive to chilling must be grown in heated greenhouses with high energy costs. In order to reduce heating costs, plant breeding programs have been designed to se- lect genotypes more resistant to chilling. The usual selection procedure is growth analysis (Den Nijs and Smeets, 1987). Unfortunately, such selection procedures are time-consum- ing and the development of new genotypes usually requires many years. Commonly used techniques to measure the chilling tolerance of plants are visual expression of necrosis (VEN), vital staining (VSP), differential thermal analy- sis (DTA), and freeze-induced electrolyte loss (FIEL). Each of these methods has disadvan- tages, and the comparison of results among different methods is often difficult. The VEN method is subjective, the results may differ between observers, the appearance of injury generally requires an extended period of time, and the large sample size necessary is often impractical to obtain (Burr et al., 1990). How- ever, VEN requires no specialized equipment. The VSP method is difficult to use, time- consuming, and requires special equipment. The DTA technique is inaccurate when plants are in the process of hardening and dehardening, chilling (Hetherington et al., 1983a, 1983b; Neuner and Larcher, 1990; Smillie and Nott, 1979; Sthapit et al., 1995), freezing (Strand and Öquist, 1988), and heat, radiation, and drought stress (Havaux and Lannoye, 1985; Lichtenthaler, 1988). Several studies have in- dicated that reductions in CF occur in plants at temperatures near their chilling threshold. Smillie and Nott (1979) reported that the de- crease in CF of leaves during rapid chilling at 0 °C can be used as an index of the chilling sensitivity of plant species. Wilson and Greaves (1990) reported that CF offers considerable potential in accelerating the quantitative as- sessment of chilling injury, as it is rapid, sensitive, nondestructive, and relatively inex- pensive, and is able to detect injury before visible symptoms occur. It also can be used for estimating the chilling tolerance of plants. The objective of this study was to test whether in vivo measurements of CF could provide a rapid and simple method for assess- ing chilling tolerance in roses and to determine if there was any association between CF and chilling injury. In addition, in vivo CF of detached and attached leaves was measured to determine if detachment affected response. Materials and Methods Plant material. Thirteen genotypes with several degrees of chilling sensitivity were used; all were ranked using the VEN method (Table 1). Rosa kordesii ‘Captain Samuel Holland’, ‘George Vancouver’, ‘John Cabot’, and ‘William Baffin’; R. hybrida ‘JP Connell’; R. rugosa ‘David Thompson’; R. rugosa ‘F- 95’ and ‘Ottawa’; and R. arkansana ‘Prairie Joy’ (obtained from Agriculture and Agri- Food Canada, L’Assomption, Qué.) were evaluated by Reseau d’Essai des Plantes Ligneuses Ornementales du Québec (REPLOQ) (Conseil des Producteurs Végetale du Québec, 1990) using the VEN method. Rosa hybrida ‘Autumn Sunblaze’, ‘Sunny Sunblaze’, and ‘Betty Prior’; and R. arkansana ‘Prairie Fire’ (purchased from Bailey Nurser- and the results often vary considerably, mak- ing calibration difficult (Burr et al., 1990). Although FIEL is a very sensitive method, it generally is time-consuming and requires re- searchers to define an arbitrary point at which irreversible tissue damage is assumed to occur (DeHayes and Williams, 1989). Genotype selection based on chilling toler- ance could be improved by the availability of a screening test that is rapid, nondestructive, and applicable to young plants. Most chloro- phyll fluorescence (CF) studies on tempera- ture responses of plants have focused on the determination of low and high temperature limits of growth (Lichtenthaler, 1988; Weis, 1981). Recently, CF has been used to study cellular processes other than photosynthesis, the possible responses of plants to various stresses, including chilling/freezing tolerance in particular (Brennan and Jefferies, 1990; Sundblad et al., 1990; Walker et al., 1990). Several studies have suggested that in vivo CF can be used as a rapid screening technique for stress tolerance in crop plants, including Table 1. Fv measurements and F values for detached and attached leaves of Rosa genotypes, the decrease of Fv (β 1 ) over time when the leaves were subjected to 0 °C, and the chilling sensitivity of the genotypes in the literature as classified by visual expression of necrosis (VEN) method. Expt. 1. Chilling Fv Slope (ß 1 ) sensitivity Species Cultivar Detached Attached F-value z Detached (VEN method) Rosa rugosa Ottawa 146 159 1.00 NS –1.51 ** Very resistant y F-95 189 215 1.27 NS –1.64 ** Very resistant y David Thompson 121 114 1.03 NS –1.72 ** Very resistant y Rosa kordesii William Baffin 231 242 1.66 NS –1.57 ** Very resistant y George Vancouver 187 196 1.33 NS –1.91 ** Resistant x John Cabot 129 134 1.29 NS –2.06 ** Resistant y Capt.Samuel Holland 212 218 1.15 NS –2.16 ** Resistant y Rosa arkansana Prairie Joy 157 158 1.11 NS –1.79 ** Resistant x Prairie Fire 202 189 1.66 NS –2.28 ** Sensitive w Rosa hybrida JP Connell 198 218 1.27 NS –1.74 ** Resistant y Sunny Sunblaze 213 231 1.55 NS –2.13 ** Very sensitive w Autumn Sunblaze 162 170 1.21 NS –2.48 ** Very sensitive w Betty Prior 183 195 1.02 NS –2.31 ** Very sensitive w z All values nonsignificant at P ≤ 0.05. y Conseil des Producteurs Végetale du Québec (1990). x Richer et al. (1997). w Information obtained from Bailey Nurseries (St Paul, Minn.) based on VEN classification. ** Significant at P ≤ 0.01 by F test.