Perry's Perennial Pages -- 2003 Freezing Research Update

Andrea Luchini, Graduate Research Assistant

In the winter of 2003, several container perennial plants were subjected to controlled freezing procedures to determine their relative hardiness (see more detailed methods at end). Different treatments were applied to the plants in the fall prior to the freezing experiments to determine if these treatments had any effect on the plant's cold hardiness. Plants were frozen to -2 (control), -5, -8, -11, and -14°C and then allowed to regrow throughout the spring at which time visual ratings as well as dry weight data was taken. This research expands upon and continues research done for several years by Leonard Perry and his graduate research assistants and will continue in 2004.

Plant Age

Plants used in this experiment were of 3 "ages". One was severely pot-bound, having survived for several years in the same pot; these were labeled "established". Another set of plants had recently been divided from pot-bound plants and repotted; these were labeled "division". The third set of plants were new cuttings and had minimal root growth; these were labeled "plugs". The objective in this experiment was to determine if the plugs and/or divisions were less able to survive freezing temperatures than plants that were established in their pots. Dianthus deltoides ‘Vampire’ was used for this experiment.


There was little difference in both the ratings and dry weights for the three treatments. However, the divisions, as well as the plugs were slightly better than the established plants. This may be because the roots of established plants would be extended to the side of the pot and more exposed to the cold temperatures. Plants with more soil in pots were more protected by that soil, which acted as an insulator. The established plants were also already more stressed and, it seems, were less able to handle the additional stress of cold temperatures.

Length of Outdoor Acclimation

Plants for this experiment were left outdoors for different lengths of time in the fall prior to the freezing experiments. Geranium ‘Dilys’, Geranium x cantabrigiense ‘Cambridge’ and Dianthus deltoides ‘Vampire’ were used. G. ‘Dilys’ and G. ‘Cambridge’ plants were brought into a 4°C (40°F) greenhouse from an outdoor nursery area on October 18, November 1, and November 15. Some Geranium ‘Dilys’ remained inside the greenhouse for the entire fall. The Dianthus were brought in on November 1 and November 15. The plants brought into the greenhouse at the earliest date will experience more constant and slightly warmer temperatures than the plants outside. This might allow them to grow and become more established and therefore better able to survive freezing temperatures. An extra set of Dianthus were brought inside on November 1 and placed in a warm house (14-24°C, 57-75°F) for 7 days prior to freezing to determine if this would induce a deacclimation effect and render the plants unable to survive the freezing temperatures.


The results varied by cultivar. Overall, G. ‘Dilys’ plants did not survive well below -5°C (23°F) and almost all plants died at and below -11°C (12.2°F). The plants with the least amount of acclimation were the least able to survive the freezing treatments. The G. ‘Dilys’ plants that did the best and were the biggest were brought in on November 1. Therefore, the G. ‘Dilys’ plants did not seem to benefit from the added warmth for growth in the fall and were better able to acclimate to cold temperatures outside. With G. ‘Cambridge’, we found that the plants brought in on October 18 did the best and were larger than plants brought in on the other two dates, although the results for all three were similar. The G. ‘Cambridge’ may have benefited from the lengthened growing season in the greenhouse. Dianthus plants brought in on November 1 did slightly better and were bigger than those brought in on November 15. The deacclimation seemed to have little effect on freezing survival of Dianthus. It is possible that the plants did deacclimate while in the warm house but the acclimation period in the freezer was enough for them to reacclimate and therefore withstand the freezing temperatures almost as well as the plants that did not experience deacclimation.

Temperature Cycling

Plants for this experiment were treated to different levels of temperature cycling prior to the controlled freezing procedures. Plants used for this study were, Geranium ‘Dilys’, Geranium x cantabrigiense ‘Karmina’ and Iris sibirica ‘Pirate Prince’. The Geranium ‘Karmina’ and the Iris were subjected to 1 cycle of -3°C/+3°C, with 24 hours at each temperature in January and the Geranium ‘Dilys’ were subjected to 2 cycles of -3/+3°C with 24 hours at each temperature in both January and March. After the temperature cycles, plants were left in the freezer for two days to acclimate to -3°C and then subjected to controlled freezing on day seven.


With the Iris cultivar used, we found that cycling had no effect on survival or dry weight. All plants seemed to survive all treatments with equal success. The average dry weight for the control plants was 1.05 g and the average for the cycling plants was 0.94 g. There were few visual differences between the treatments so a rating scale was not used. There also did not seem to be any practical effect on the Geranium ‘Karmina’ plants. The average dry weights for the control plants and the treatment plants were very similar and no practical difference was reflected in the ratings.

Geranium ‘Dilys’ that were treated in January did show some statistical as well as practical significance. The only treatment with plants that averaged a rating of over 3.0 (marketable plants) was the 1 cycle treatment. The other treatments had a few individual plants that rated 3.0 or greater but the averages for both two cycles and for the control plants was below 3.0. Dry weight data did seem to reflect this difference as well, although the small plant size makes the difference questionable. The dry weights for plants exposed to only one cycle was over two times the average dry weight for both the control and the two cycle treatment. In March, when theoretically the plants would be less hardy because ambient temperatures are warming, few plants had ratings over 3.0. The averages for all three treatments were well below 3.0 and, in fact, most plants did not survive the freezing temperatures. The plants that did the best were in the two cycle treatment, but again, even these plants were not marketable.


The research was done in insulated chest freezers with temperature controllers attached. These controllers monitored the temperature in the freezer and maintained it at the appropriate setting. Small circulating fans were placed in the bottom of the freezers in order to increase air circulation to ensure an even temperature throughout the freezer. Temperatures were also monitored using several thermocouples (thermometers), both in the air of the freezer as well as in the soil of a few pots. Another data logger thermocouple was also placed in the freezer. This unit recorded both pot and air temperature every five minutes. Plants were loaded into the freezer on flats, 15 4-inch pots per flat.

Plants were loaded, from a greenhouse chamber set at ±4°C (40°F) into the freezer and the temperature controller was then set to achieve -2°C (28.4°F). These plants were left in the freezer for two days in order for all plants to freeze to -2°C. Plants of different moisture levels take different amounts of time to freeze. Plants usually froze after about 12 hours of -2°C but the 48 hours ensured that all plants were at the same temperature. This was previously determined to be the best acclimation period for these freezing procedures. After the 48 hours, 6 plants of each treatment/cultivar were removed. These plants served as the control plants. The temperature in the freezer was then reset to -5°C. The freezer cooled at about 2°C per hour so plants were allowed two hours to get to this temperature and then left for another 1/2 hour to ensure they were all frozen to -5°C (23.0°F). Another 6 plants per treatment/cultivar were removed and the temperature was reset to -8°C (17.6°F) and the process was repeated. The same occurred for -11°C (12.2°F) and -14°C (6.8°F). All plants, immediately upon removal from the freezer, were transferred back to the ±4°C greenhouse chamber.

Plants were allowed to regrow throughout the spring. Once plants had reached sufficient size, plants were rated on a visual scale and dry weight data was taken. Plants were rated on a scale of 1 to 5, with 1 being the lowest (dead) and 5 the highest. The rating scale varied for each cultivar but in general, ratings of 3 and above were plants that would be considered saleable in a commercial setting and those below 3 were considered unmarketable (unusable for a commercial grower). After the visual ratings were taken, plants were cut back to about 1 cm from the soil level and all living parts were placed in a paper bag. Dry weight was recorded in grams for each plant. The rating and dry weight results were subjected to analysis of variance (ANOVA) to determine if there were any statistical differences between treatments and between temperature settings.

Copyright 2004, not for publication without permission from author.

Return to Perry's Perennial Pages | Sustainable Perennials Research