Scientia Horticulturae, 51 (1992) 139-144
Influence of bud chilling on subsequent reproductive fertility in roses
Serge Gudin

Pot plants of Rosa hybrida L. cultivar 'Meijikatar' were submitted or not to a 1 month cold treatment, just after cut-back. After subsequent culture in a greenhouse, the treatment resulted in an improved production of flowers, hips and seeds resulting either from self- or cross-pollinations. It also determined a specific morphology of the styles and an improved in vitro elongation of the pollen tubes. The cold pre-treatment effect could be explained by its influence on fertility. This influence, although diminishing, could still be demonstrated after pollination on the second flush of flowers.

INTRODUCTION

The effect of bud chilling on flowering has been widely studied with many plants (Heller, 1985), including roses (Asaoka and Heins, 1982; Khayat and Zieslin, 1982; Zieslin and Tsujita, 1990). Furthermore, bud chilling requirement has often been related to seed (Westwood and Bjornstad, 1968; Kester, 1969; Kester et al., 1977; Pasternark and Powell, 1980; Chang and Werner, 1984; Rodriguez and Sherman, 1985) and even seedling chilling requirement (Kester et al., 1977; Rodriguez and Sherman, 1985) in numerous woody members of the Rosaceae family (pear, almond-peach hybrids, almond, apple, peach, nectarine). However, the possible relation between bud chilling and resulting fertility has, to our knowledge, never been investigated in this family. We therefore studied the effect of a cold treatment of cut-back rose plants on subsequent flowering and fertility, resulting either from cross- or self-pollination. The in vitro germination of pollen collected from cold-treated and untreated plants was also observed. Finally, fertility resulting from self-pollination of the second generation of flowers (following cut-back and eventual cold treatment) was recorded.

MATERIAL AND METHODS

Meijikatar = Orange Sunblaze
Meimutin = Meitinirol = Galia

Eighty 2-year-old pot plants of Rosa hybrida L. cultivar 'Meijikatar', propagated by cuttings on the same day by a nursery (Rosaflor France, Le Cannet des Maures) were used. The plants were grown in a greenhouse at Selection Meilland (Antibes) in 18-cm diameter containers in a peat moss-pouzzolane mixture. They received, per plant and per week, the equivalent of 3 g of a complete soluble fertilizer Kristallon (N, P, K equilibrium: 17-6-18 + microelements) and 7 l of water by daily distribution through a drip irrigation system.

*Fertilization + irrigation

On 24 April 1989, 40 plants were cut back (two to three buds per axis remaining on the plant) and stored in a cool room at 4°C for l month. On 24 May 1989, the 40 plants that had remained in the greenhouse were identically cut back and disconnected from the fertigation* system for 2 weeks. The plants that had been cold-treated were placed back in the greenhouse on the same day and re-connected to the fertigation system. The average minimum night temperature varied from 15 to 20°C and the average maximum day temperature varied from 25 to 32°C, until the flush, which occurred from 27 June to 7 July in these conditions.

The flowers were counted. In each lot (cold-treated and untreated plants), a section was covered with transparent paper cones (Chrystal 6 x 10 with flap) and left to self-pollinate, while the other part was pollinated 48 h after emasculation with fresh pollen of R. hybrida cultivar 'Meimutin', as described by Gudin et al. (1990), on the same day (30 June 1989). Prior to cross-pollination, a sample of 50 flowers per case (cold-treated and untreated plants) was observed with a micrometric lens (Haas 8x) in order to determine the averages of style number per flower and style length (five styles of the outer whirl measured per flower). Furthermore, the anthers castrated on cold-treated and untreated plants while they were being prepared for the forthcoming cross-pollination, were collected for an in vitro pollen germination test, as described by Gudin etal. (1991).

The second flush of flowers, originating from the first flush flower stems, occurred from 25 July to 30 July. The flowers were counted and were all left to self-pollinate as previously described. For both flushes, the hips were counted and collected, their seed contents counted 4 months after the pollination periods.

RESULTS

Table 1 shows that cold pretreated plants, compared with untreated ones, produced more flowers (12.25 instead of 5.62 per plant), hips and seeds, resulting either from self- or cross-pollination. It is noticeable that no hip formation occurred in the case of untreated plants. Hip formation (occurring on cold pretreated plants) was significantly higher in the case of self-pollination than in the case of cross-pollination, whereas a Student's t test showed the mean numbers of seeds per hip not to be significantly different.

TABLE 1

Effect of a 1 month cold pretreatment at 4°C of cut-back plants of R. hybrida 'Meijikatar' on subsequent
hip and seed formations resulting from self-pollination or cross-pollination with R. hybrida 'Meimutin'1

  Cold-treated plants Untreated plants
Self-pollination
Number of flowers pollinated 175 100
Rate of hip formation (%) 17.1 c 0 a
Mean number of seeds per hip (±SD) 2.7 ± 1 -
Cross-pollination
Number of flowers pollinated 315 125
Rate of hip formation (%) 3.2 b 0 a
Mean number of seeds per hip (±SD) 2.5 ± 0.5 -

1Rates followed by an identical letter do not differ significantly, for P=0.05 (Chi-square test).

TABLE 2

Effect of a 1 month cold pretreatment at 4°C of cut-back plants of R. hybrida 'Meijikatar' on
quantities and lengths of subsequently formed styles. For each plant lot, 50 flowers were observed1

  Cold-treated plants Untreated plants
Mean number of styles per flower (±SD) 23.2 6.2 a 28.8 7.2 b
Mean length (mm) of styles (±SD) 4.4 0.63 a 3.3 0.36 b

1For each line, means followed by a different letter vary significantly for P=0.05 (Pearson's conformity test).

The above differences concerning comparative flower and seed productions of cold pretreated and untreated plants were accompanied by differences in style number and morphology (Table 2). Style number was significantly higher in the case of untreated plants while style length was significantly greater in the case of cold pretreated plants.

The cold treatment of cut-back plants also had an influence on the in vitro germination of subsequently formed pollen, as shown in Table 3. The mean length of pollen tubes produced by pollen collected on cold pretreated plants was significantly higher (three times) than the mean length of pollen tubes produced by pollen from untreated plants, the germination percentages remaining statistically similar in both cases.

As with first flush flowers, the flowers of the second flush (following cutback), developed from cold pretreated plants, produced significantly more hips and seeds (resulting from self-pollination) than the flowers developed from untreated plants (Table 4). Once again, no hip formation occurred in the latter lot. Both hip formation (occurring only on cold pretreated plants) and number of seeds per hip were significantly lower than in the case of self-pollination of first flush flowers (Chi-square and Student's t tests, respectively).

TABLE 3

Effect of a 1 month cold pretreatment of cut-back plants of R. hybrida 'Meijikatar' on in vitro germination of subsequently formed pollen.
For germination percentage determination, 10 x 50 pollen grains per case were considered; for mean length determination, 25 pollen grains per case were observed1

  Pollen collected on plants
pretreated 1 month at 4C
Pollen collected on
untreated plants
Mean percentages of germinated pollen grains (SD) 9.6 1.8 a 8.8 1.4 a
Mean length (um) of emitted pollen tubes (SD) 71 30 a 23.3 11 b

1For each line, means followed by an identical letter do not differ significantly for P=0.05 (Student's t-test).

TABLE 4

Effect of a 1 month cold pretreatment at 4°C of cut-back plants of R. hybrida 'Meijikatar' on
subsequent hip and seed formations resulting from self-pollination of the second flower flush1

  Cold-treated plants Untreated plants
Number of flowers 365 275
Rate of hip formation (%) 2.7 a 0 b
Mean number of seeds per hip (±SD) 1.5 0.5 0

1Rates followed by a different letter vary significantly for P=0.05 (Chi-square test).

DISCUSSION

It could be argued that the grouping of the pollinations during the same period of time (for self-pollination) or on the same day (for cross-pollination) unavoidably influenced the difference in timing of the cutting back of the two plant lots (reference and cold pretreated). However, this choice made as a difference in timing of pollination in roses was recently shown to have a dramatic influence on resulting fertilization, through pollen quality and female fertility (Gudin et al., 1991).

The promotion of flowering observed on cold-treated plants confirms previous results obtained on another rose cultivar (Asaoka and Heins, 1982) and on the same one as used here (Zieslin and Tsujita, 1990). It can be easily explained by the bud dormancy breaking effect of cold (see Heller, 1985), already demonstrated in roses (Khayat and Zieslin, 1982).

Hip and seed production results demonstrate that cold pretreatment also greatly influences pollination efficiency, as each of these two criteria have been shown to represent a good index of fertilization in roses (Gudin and Arene, 1991; Gudin et al., 1991). It is in that respect interesting to note that cold pretreatment resulted in a specific morphology of female sexual parts of the flower such as the styles, affecting them in exactly the same way as described by Semeniuk (1964) under low temperature growth conditions. Cold pretreatment had previously been demonstrated to influence vegetative traits such as flower stem length (Asaoka and Heins, 1982).

In the case of self-pollination, fertility improvement due to cold pretreatment could be at least partly explained by its positive influence on pollen tube elongation. Length of pollen tubes developed in vitro has been shown to be related to in vivo pollen fertilizing efficiency (Gudin and Arene, 1991; Gudin et al., 1991). The results observed on cross-pollinated plants (with fresh unstored pollen) demonstrate that cold pretreatment also influences female sexual receptivity. It would be of great interest to determine which female parts (stigmatic or stylar tissues, ovules, etc.) are particularly affected by cold pretreatment, and by which physiological process (hormonal balance, male-female interaction in cell nutrition during pollen tube progression in the stylar tissues, etc.).

The results obtained with the second flush of flowers clearly demonstrate the positive influence of cold pretreatment on fertility, even though it is less efficient than in the first flush.

REFERENCES