Plants of Rosa sp. cvs. Absolute Hit and Inter King (miniature roses) and Kalanchoe blossfeldiana cv. Debbie were used in the experiment that was carried out in a greenhouse at Bioforsk Særheim, Norway (about latitude 59˚ north). Pot roses with five cuttings per pot were cut back before start of the experiment, and the Kalanchoe plants were pretreated with short days (10 h·day⁻¹ light for induction of flowering) before the start of the experiment. The plants were grown in standard fertilised peat (Floralux) in 10 cm pots and were watered daily with a complete nutrient solution. Twenty-four pots of each of the cultivars were placed in a 5 m long row on four greenhouse tables. Altogether 192 pots with roses (two cultivars) and 96 pots with Kalanchoe were included in the experiment, which started on 14 December with the miniature roses and on 3 February with Kalanchoe. High- pressure sodium lamps (Philips SON/T) were placed above the tables, supplying from 60 to 670 µmol·m⁻²·s⁻¹ PFD from one end of the table to the other. The corresponding air temperatures around the plants varied from 20.0˚C to 24.0˚C (TEMP1) (Figure 3). Two of the tables were covered with cheese cloth, creating a PFD gradient from 50 to 370 µmol·m⁻²·s⁻¹ from one end of the table to the other, with corresponding air temperatures from 22.5˚C to 26.5˚C (TEMP2) (Figure 3). The mean daylight throughout the experimental period, calculated during the 20 h·day⁻¹ lighting period for miniature roses, was 8 and 16 µmol·m⁻²·s⁻¹ for TEMP1 and TEMP2, respectively (measured at the Meteorological Station at Særheim). For Kalanchoe the respective values were 14 and 27 µmol·m⁻²·s⁻¹ PFD. The outside light levels were reduced by approximately 50% due to shading by the greenhouse construction. The CO₂ concentration was maintained at 800 ± 100 µmol·mol⁻¹. Time until flowering (five open flowers), total number of open flowers and flower buds, plant dry weight per pot and a decorative value index were recorded for the pot roses. In addition, dry weight increase per day was calculated. In Kalanchoe time until flowering (visible flower colour), fresh weight of all flower stems and flowers, plant dry weight and a decorative value index were recorded. The decorative index was visually determined and ranged from 1 to 9, with 1 as the lowest and 9 as the highest value. Values below 5 were recognised as not saleable. The number

and colour of the flowers, leaf colour and plant compactness were taken into account while determining the index. The graphs are produced using Simaplot 10.0 and linear regression lines order 2 are generally included. All values are presented as means ± SE.

Plants of Rosa sp. cv. Roxy (miniature rose) and of Kalanchoe blossfeldiana cv. Debbie planted in 10 cm pots in standard fertilised peat (Floralux) were used in the experiment. The miniature roses were cut back and grown in a greenhouse compartment in sunlight until buds became visible. Kalanchoe plants were grown under natural light on long days and were vegetative. Homogenous plants were selected, and three pots (corresponding to a density of 12.5 pots per m² ground area) were placed in two separate cuvette units, as previously described [5] . The CO₂ exchange rate (CER) was recorded continuously. The cuvette units (293 litres in volume; 500 × 480 × 1220 mm and plant floor of 0.24 m²) were placed in a greenhouse compartment, and the carbon dioxide concentration was controlled by adding pure CO₂ from a container. Light was measured in each cuvette by a quantum sensor (Delta-T Devices, 400 - 700 nm). The CER was measured in two plant stands of Kalanchoe during the first half of May, and in two plant stands of pot roses in the second half of the month. The day length was 16 to 17 hours during the experimental period. The air temperature was 22.0˚C ± 0.5˚C, and the leaf temperature could increase to 2˚C above the air temperature at maximum photon flux density (PFD). The relative air humidity (RH) varied between 70% and 90%. The cooling system in the cuvette caused water vapour condensation during the day, resulting in a lower RH of 70%. One plant stand was exposed to 370 ± 30 µmol·mol⁻¹ CO₂ and the other to 800 ± 50 µmol·mol⁻¹ during five days of sunny weather. One day before and one day after the treatment period of five days, the two Kalanchoe stands were exposed to 370 µmol·mol⁻¹ CO₂. The same treatment (370 µmol·mol⁻¹ CO₂) was administered to the rose stands two days before the treatment started.

The data were analysed using the SAS-GLM procedure (SAS Institute Inc., Cary, USA) with days as replicates. The analysis of the correlation between the CER and the PFD was conducted for low and high CO₂ concentrations separately, and hourly mean measurements were taken throughout the five-day treatment period.

Miniature roses: Time until flowering decreased from about 50 to 25 days when the supplementary PFD increased from 50 to 370 µmol·m⁻²·s⁻¹ and the mean air temperature increased from 22.5˚C to 26.5˚C in the TEMP2 treatment (Figure 3). The same increase in PFD accompanied by a temperature increase from 20 to about 23˚C (TEMP1 treatment) decreased the time until flowering from about 50 to 35 days. The total number of flowers increased up to about 300 µmol·m⁻²·s⁻¹ PFD and was significantly decreased by increased temperature (Figure 3). The mean dry weight increase per day (growth rate) until flowering increased progressively with PFD up to about 400 µmol·m⁻²·s⁻¹. The rate started at a lower level but increased somewhat faster in the TEMP2 treatment as compared with the TEMP1 treatment. Above 400 µmol·m⁻²·s⁻¹ PFD the time until flowering, total number of flowers and dry weight were relatively unaffected. The leaf colour changed from dark green to bright green and the flowers from red to pink at PFD >500 µmol·m⁻²·s⁻¹ (data not presented). The decorative value of the plants at the lowest PFD levels was very low due to poor growth and few flower buds (not saleable), and it increased up to 200 - 300 µmol·m⁻²·s⁻¹ PFD in the two temperature treatments(data not presented). Kalanchoe blossfeldiana: Increasing the supplementary PFD from 50 to 370 µmol·m⁻²·s⁻¹ decreased the time until flowering by about 15 days during both temperature treatments (Figure 4). The dry weight was strongly decreased (by about 60%) by a temperature increase from 20.0˚C to 23.5˚C (the difference between treatments TEMP2 and TEMP1) at the lowest PFD. As the PFD increased the dry weight of the TEMP2 treatment increased significantly, and the dry weight at 370 µmol·m⁻²·s⁻¹ was only 20% lower than the dry weight measured during the TEMP1 treatment. The fresh weight of flowering stems and flowers followed the same pattern as the plant dry weight. The decorative value reached its maximum at a PFD of about 150 µmol·m⁻²·s⁻¹, while light levels above 250 µmol·m⁻²·s⁻¹ caused leaf yellowing (data not presented).

The photosynthetic capacity of the miniature roses was saturated at about 300 µmol·m⁻²·s⁻¹ PFD at 370 µmol·mol⁻¹ CO₂, and it was saturated at 400 to 500 µmol·m⁻²·s⁻¹ PFD at 800 µmol·mol⁻¹ CO₂ (Figure 5). In Kalanchoe blossfeldiana saturation levels were reached at roughly 400 and 500 µmol·m⁻²·s⁻¹ PFD, respectively (Figure 6). Increasing the CO₂ concentration increased CER in the miniature roses and Kalanchoe blossfeldiana by 48% and 69%, respectively. The CER was the same in both plant stands when both species were exposed to the same CO₂ concentrations (Table 1, Table 2; Figure 5, Figure 6). When the CER and PFD were calculated during a seven-hour period in the middle of the day (10:00-17:00), the CER increased by 40% in the miniature roses and by 64% in Kalanchoe. During this period, the plants received around 70% of the PAR, corresponding

to 70% to 80% of the daily CER. A significant correlation was found between the CER and the PFD (p < 0.001, n = 24) in miniature roses (r² = 0.91 - 0.94) as well as in Kalanchoe (r² = 0.90) at both CO₂ concentrations.