THE INFLUENCE OF POLLEN VIABILITY ON SEED SET AND FRUIT MASS IN STRAWBERRY ( FRAGARIA X ANANASSA DUCH . )

This research was conducted in 2006 in the Experimental Station of the Department of Genetics and Horticultural Plant Breeding in Felin near Lublin. It included 7 cultivars of strawberry (‘Pastel’, ‘Salut’, ‘Teresa’, ‘Jota’, ‘Paula’, ‘Senga Sengana’, ‘Plena’) and two selected clones: 2395 and 3995 that belonged to the Department. Analysis of pollen fertility was carried out on the basis of materials collected on fi ve occasions: on May 17 (at the beginning of fl orescence), May 20 and May 24 (the peak of fl orescence) and on June 2 and June 13 (the end of fl orescence). Smear preparations stained with 2% acetocarmin and glycerin solution (1:1) were used to test pollen viability. The highest percent of viable pollen in 2006 was in cultivar ‘Jota’ (70.88%). The lowest viability had pollen of ‘Teresa’ cultivar (33.83%). The average content of viable pollen grains was the lowest on May 20 (36.79%). The highest pollen fertility was noted on June 2 (62.15%), the percent of fertile pollen on a similar level was observed on May 17 and 24 and June13. The cultivar ‘Jota’ was characterized by an increased level of pollen fertility that was quite high during the whole period of fl orescence. The number of seeds has a great infl uence on the proper development of spurious strawberry fruit. The mass of seeds from a single fruit has a great infl uence on the mass of fruits in both large and small fruit categories. Pollen viability affected the mass of seeds on big fruits (r = 0.444), but there is no clear direct relationship between pollen fertility and mass of strawberry fruit(r = -0.193 and r = -0.052).


INTRODUCTION
Poland is one of European and world leading strawberry producers. Due to very favourable weather conditions during fruit ripening season, fruits produced in Poland develop a distinctive taste and aroma which make them very popular on the international market (R e j m a n , 1994).
For years, cv. 'Senga Sengana' has been the most widely grown strawberry cultivar in Poland. In recent years, a growing interest in dessert cultivars like 'Dukat' or 'Elsanta' is observed. The mentioned cultivars make up 16% of commercial strawberry production. It was marketing reasons that mainly infl uenced the interest in dessert cultivars (M a ł o d o b r y a n d B i e n i a s z , 1999).
Success in the modern strawberry industry largely depends on the cultivar used. Because of the growing interest in fresh fruit, it has become necessary to breed new cultivars characterized by resistance to low temperature, drought, diseases, pest as well as by good fl avour and great yielding potential (K a c z m a r s k a , 2000).
Many strawberry cultivars do not produce high yields in our climatic conditions or in the regions where they were developed. Thus, such cultivars are not recommended for fi eld production. Yet, they may be used as valuable material for further breeding (Ż u r a w i c z , 1997). This paper deals with the relation between the number and mass of seeds needed for the proper development of fruit and its mass. The infl uence of pollen fertility on this process is given special attention.

MATERIALS AND METHODS
The primary objective of the present research was to determine the quality of pollen and other utility traits. It was conducted in 2006 in the Experimental Station of the Department of Genetics and Horticultural Plant Breeding in Felin near Lublin. It included seven cultivars of strawberry ('Pastel', 'Salut', 'Teresa', 'Jota', 'Paula', 'Senga Sengana', 'Plena') and two selected clones: 2395 and 3995 that belonged to the Department. During fructifi cation of those cultivars, ten fruits of each cultivar were picked randomly. Then, the following characteristics were examined: the number of seeds of a single fruit, the mass of seeds and the mass of a single fruit. The analysis was carried out for the category of large fruits (the mass of a fruit was above 5 g) and small fruits (the mass of a fruit was below 5 g).
Statistical analysis of the data included arithmetic means, standard deviation, variance and coeffi cient of variation V (at p=0.05 confi dence level).
The tests of pollen viability were made on fi ve occasions: on May 7 (fl orescence onset), May 20 and 24 (the peak of fl orescence) and on June 2 and 13 (the end of fl orescence). Smear preparations stained with 2% acetocarmin and glycerin solution (1:1) were prepared. Fully coloured pollen was treated as viable, while uncoloured pollen was considered as non-viable. Correlations between the analyzed traits were estimated using the correlation coeffi cient (r).

RESULTS
In 2006 the highest percent of viable pollen was found in the clone 3995 in which 70.88% of seeds were fully fi lled with cytoplasm. Cv. 'Plena' showed the lowest viability; only 33.83% of its pollen was able to pollinate.
Comparing the average content of viable pollen grains collected at different dates, the lowest content was on May 20 (36.79%), the highest on June 2 (62.15%). On the other occasions (May 17 and 24 and June 13), the percentage of viable pollen grains remained similar.
When analysing the fl orescence stages (from May 17 to June 13) of different cultivars, a characteristic pattern of the pollen fertility curve was observed. The results presented in Table 1 show that in the case of cultivars 'Senga Sengana', 'Plena', 'Teresa' and 'Salut' pollen fertility in the initial period (May 17 and 20) is very low to reach its highest level around June 2 and it decreases again in the later period (June 13).
The cultivar 'Pastel' and clones 2395 and 3995 represented a different fertility curve than the one described above. It started from a very high value at the beginning of fl orescence (May 17), then decreased in the mid period (June 2) and rose again at the end of fl orescence (June 13). Only the cultivar 'Jota' demonstrated an increased pollen fertility level that was quite high during the whole period of fl orescence. Table 2 shows the test results of pollen fertility and chosen traits of fruits for nine cultivars of strawberry.
There was a strong relationship between pollen fertility and other features for cv. 'Jota'. The average Table 1 The content of fertile pollen tested between 17 May -13 June 2006 in nine different cultivars (%    content of pollen grains in the cultivar 'Jota' was high (70.88%), the average mass of a single fruit was the heaviest of all the analyzed genotypes (16.42 g). The number of seeds in the fruit (377.7) and their mass (0.26 g) were higher than in other cultivars. A similar relationship was observed for the cultivar 'Salut': with its pollen fertility at the level of 59.19%, the mass of seeds was not that heavy (0.17 g), but the number of seeds and the mass of a single fruit (315.73 and 14.47 g, respectively) were one of the highest and not much less than for the cultivar 'Jota'.
In contrast to clone 2395, the average mass of a single fruit (15.01g) and the number of seeds (340.25) were high despite of the fact that the average content of pollen grains was low (45.72%).
Comparing the average numbers for the analyzed characteristics of all the tested genotypes, we can state that the number of seeds from a single fruit as well as their mass in the category of large fruits (the average mass of a fruit was 20.95 g) was considerably higher x compared to the corresponding numbers for small fruits (the average mass of a fruit was 5.66 g) (Tab. 3 and 5). It should be emphasized that the coeffi cient of variation for the mass of a single fruit in the category of large fruits was the highest (V = 0.2466). The lowest coeffi cient of variation concerns the mass of a single fruit in the category of small fruits (V = 0.0901).
On the basis of the estimated correlation coeffi cient, a high correlation was found between the average number of seeds in a fruit and the average mass of seeds in the category of large fruits (r =0.807). The mass of a single fruit in both the large and small fruit categories was evidently infl uenced by the number of seeds (r=0.585 in large fruit category, r = 0.634 in small fruit category). Pollen viability had a benefi cial effect on the mass of seeds in large fruits (r = 0.444). However, the direct relationship between pollen fertility and the mass of strawberry fruit cannot be confi rmed (Tab. 4 and 6).

DISCUSSION
The appropriate development of strawberry fruit, which in this case is a spurious fruit that formed from the growing receptacle, is to a large extent infl uenced by the number of seeds formed after pollination and fertilization of ovules. The size of the fruit depends mainly on the position of fl owers in the capitulum -cyme. The growth of the receptacle is stimulated by hormonal substances that appear after fertilization of the ovules (S z c z y g i e ł and P i e r z g a , 1999).
Studies of D e c (2006) have shown that the number of seeds from a single large fruit is sometimes even lower than from a small fruit. However, the research presented in this paper does not confi rm it. In all the tested cultivars, the number of seeds from a large fruit was higher than from a small fruit. According to Dec, the mass of seeds from large fruits was almost always higher than of those from small fruits. This was also confi rmed in our research. In our observations, it was proved for each of the tested cultivars and clones.
A study of pollen grains by W y k a (1984) shows a slight difference in strawberry pollen viability. The content of pollen that were coloured (which meant it was viable) varied from 72 up to 98%.
Ż e b r o w s k a and H o r t y ń s k i (1992) tested viability of fresh pollen of fi eld plants. In their experiment, pollen of 'Teresa' and 'Senga Sengana' cultivars was coloured in 65% and 49%, respectively. In our research conducted in May and June 2006, the percentage for these cultivars was lower ('Teresa'-33.83%, Senga Sengana' -40.09%).
According to K a c z m a r s k a (2000), pollen fertility of 'Senga Sengana' in the fourth week of May was 15%. In the present research, pollen fertility was checked 5 times, and the lowest level for this cultivar was 25.9% and the highest 40%. It was similar in the case of cultivar 'Teresa'.
The differences found in our research prove the opinion of many authors concerning an enormous impact of weather conditions, especially temperature and rainfall, on blooming and consequently on the setting and development of fruit of strawberry grown in fi eld conditions (H o r t y ń s k i and H u l e w i c z , 1992; Ż e b r o w s k a and H o r t y ń s k i , 1992; K o s t us i a k , 2001).
The following conclusion may be drawn from our fi ndings : 1. In 2006 the highest content of viable pollen (70.88%) was observed in 'Jota' cultivar 3995, and the lowest (33.83%) in 'Teresa' cultivar. 2. 'Jota' cultivar is characterized by having a great infl uence of pollen quality on the number and mass of seeds and also on the average mass of a single fruit. 3. There was a strong relationship between the number of seeds and the average mass of strawberry fruit. It was statistically proved in the category of large fruits in which the average mass of a fruit is 20.95 g. 4. Correlation coeffi cient data reveal a strong relationship between the mass of seeds and the mass of fruit in both categories (r = 0.585 for big fruits, r = 0.634 for small fruits). 5. Analysis of the correlation coeffi cient shows a quite strong relationship between pollen fertility and the mass of seeds on large fruits (r = 0.444). 6. No correlation between pollen viability and the mass of strawberry fruit was found, the correlation coefficient for large fruits was -0.193 and for small fruits -0.052.