Correlations between Some Yield and Yield Components in Sunflower ( Helianthus annuus L . ) Hybrid , Line and Varieties *

This study was conducted on experimental field of Field Crops Department at Agricultural Faculty of Ankara University in 1995, 1997 and 1998. The trials were designed in randomized block design with three replications. Genetic male sterile lines, hybrids and Ekiz-1, Vn ıı mk-8931 and Super-25 varieties were used as a material. According to the results of correlation analysis; positive correlations between SeY, SY, HD and PH; between OR, TSW, SeY and HD; between SeY, TSW and SY were recorded.


Introduction Sunflower (Helianthus annuus L.) is widely grown in
The some of studies in this subject were giyen many parts of the world and one of the more important oil below.seed crops.
It is sown and produced in many regions of Anatolia, mostly in Thrace and in Marmara.80% of oil production in our country consists of vegetable oils.
Today, consumed vegetable oils are especially sunflower, cotton and olive oils.Sources of produced vegetable oils are sunflower, cotton, soybean, and olive.
Considering the proportion (%) of produced oil from oil plants in Turkey; the highest rate belongs to the sunflower of 57% followed by cotton of 21.4%, olive of 10.7%, soybean, sesame, opium poppy and rape of 7% (Coşge, 1999).The consumption of oil per person in our country is 14 kg/year and our annual total oil need is 840 000 ton.Our vegetable oil production is 380 000 ton/year.Cotton, corn, olive and soybean are grown in limited areas.Considering these circumstances, it is required to increase the production of sunflower.
New varieties of oil crops are resistant pests and diseases, productive and high quality.Producing of these ones will be assisted in solution to our shortage of oil which continues for years.
Correlations between " yield and some yield components are required for sunflower breeding works.
Prepared from Doctorate Thesis 'e-mail: b-cosge@yahoo.com 2 Ankara Univ.Faculty of Agriculture, Dept. of Agronomy-Ankara Kovac ı k and Skauld (1976) found that husk ratio and kernel ratio were negatively correlated.Zali and Somadi (1978) reported that there was a positive correlation between seed yield and head diameter.Alba andGreco (1979), andAlba et al. (1979) pointed out a direct effect of seed weight on seed yield of (-0.686), and (0.015), respectively.
It was found that seed oil concentration was most, highly correlated with seed yield (r= 0.715).Also, seed yield was related to 200-seed weight (r= 0.481) and seed number per plant (r= 0.626) (Benjamin and Geng, 1982).
There were positive correlations between seed yield per plant and thousand seed weight and oil ratio, a negative correlation between seed yield per plant and husk ratio recorded by Shr ı n ı vasa (1982( ). InCekara et. al. (1983) ) obtained oil ratio was positively correlated with seed yield, but negatively correlated with husk ratio, and both of the correlations were significant statistically at 0.01 level.The same fı ndings were observed by Martinez (1987) and Dedio (1993).Lakshmanrai et al. (1985) recorded that there were positive correlations between seed yield and head diameter, thousand seed weight and oil ratio.Tyagi (1985) stated that seed yield was significant statistically and positively correlated with plant height, head diameter and thousand seed weight.However, the correlations between oil ratio, plant height and head diameter were non-significant statistically.Gencer (1986) explained that there were indicated positive correlations between oil ratio and plant height at 0.05 level, between seed yield and head diameter at 0.01 level.
There was 'significantly positive correlation between seed yield and oil yield.Also, plant height and head diameter had a positive effect on oil yield (Vanozzi, 1987).
Results of the other researches (Marinkovic and Skoric, 1988) showed that there was signifıcantly a positive correlation between seed yield and plant height.Soltani and Archi (1988) reported that thousand seed weight and oil yield was positively correlated.Vanisree (1988) found that head diameter, thousand seed weight and plant height were positively related to seed yield.
Yield components such as thousand seed weight, oil ratio and head diameter had an significant and positive effect on oil yield (Visic, 1988).
Correlations between sunflower yield and its components were discussed by Pasda and Diepenbrock (1990) and that number of achenes/plant was concluded to be the most important yield component was recorded.Sezer (1991) stated that head diameter was highly and positively correlated with seed yield.Chaudhary and Anand (1993) mentioned that seed filling was positively correlated with plant height, stern diameter, head diameter and stern yield, with stern diameter and head diameter having the greatest direct contribution (0.6 and 1.74, respectively).
It was reported positive correlations between seed yield per plant and head diameter (r= 0.793), between head diameter and seed number per plant (r= 0.863) (Ünlü, 1994).Miralles et al. (1997) recorded that there were significant positive correlations between number of fertile achenes per head, concentration of oil in the achenes and yield of achenes.
Correlations between crop yield and yield components were determined for sunflowers grown at Navsari, Gujarat / lndia at different fertilizer and irrigation rates (Narayana and Patel, 1998).In their study, alt the parameters examined (plant height, stern width, head diameter, seed number/head, percentage filled seeds, seed index and seed weight/head) were significantly related to seed yield.
In this study, it was aimed to investigate correlations between yield and some yield components in sunflower.
Field experiments were conducted at the experimental fields of Field Crops Department at Agricultural Faculty of Ankara University (32 ° 51" E; 39° 5 7 " N ; 8 6 0 m a b o v e s e e l e v e l ) i n 1 9 9 5 , 1 9 9 7 a n d 1 9 9 8 .
The trials were designed in randomized block design with three replications.
Hundred plants were grown in each plot in sized 30 m 2 .The sunflowers were planted at a distance of 60 and 50 cm between and within rows, respectively.
The trial materials were sown by hand on 26 April 1995, 7 May 1997 and 8 May 1998.Before sowing, alt the seeds were treated with "Aprin" against downy mildew (Plasmopara halstedih.
To determine correlations between yield and some yield components of hybrids, lines and varieties; seed yield (SeY), stalk yield (SY), a thousand seed weight (TSW), head diameter (HD), plant height (PH), husk ratio .
( The fıfty plants without broomrape (Orobanche cumana Wallr.)parasite of each plot were selected as randomly for investigation.These plants were harvested separately and obtained seeds were cleaned and weighed and SeY (kg/da) was determined.
For PH (cm), stern length was measured from soil surface to tip of head.HD (cm) was obtained by measuring diameter of dried heads from selected plants in each plot.
Plants were cut from soil surface and weighed after they were dried on iand and SY (kg/da) was counted.4 x 100 seed was selected as randomly in each plot and weighed in the sensitive scale and TSW (g) was calculated over dry matter weight.Hybrid, line and varieties" Kernel of 4 x 100 seed, which was selected to calculate TSW, was squashed, OR (%) was determined with soxhelet method using "dry kernel of seed (%) x dry oil (%) / 100" formula giyen by İ lisulu (1968).
Husks of these seeds were dried at 105 ° C for 3 h, and then weighed and HR (%) was recorded over dry matter weight.

Results and Discussion
The average values which were obtained from trial materials in 1995, 1997 and 1998 are giyen in Table 1.
We obtained a low negative correlation (r=-0.102) between SeY and PH in 1995.But, this correlation was a positive effect in 1997 and 1998 (r= 0.337 and r= 0.402, respectively).It was found that SeY was highly significantand positively correlated with PH (Tyagi 1985 Marinkoviç and Skoric 1988, Chaudhary and Anond 1993, and Narayana and Patel 1998).
As known, seed yield is a product of seed weight and number of seeds (Benjamin and Geng, 1982), Miralles et al. (1997) explained that there was a significant positive correlation between number of fertile achenes per plant and yield of achenes.Also, that number of achenes/plant was concluded to be the most important yield component was reported by Pasda and Diepenbrock (1990).
Negative correlations which were recorded in our study were not significant statistically.That there will be negative correlations between yield and yield components in sunflower was explained by Martinez (1987).For example, Kovac ı k and Skould (1976) obtained there was a negative correlation between HR and kernel ratio.Also, a negative correlation between SeY per plant and HR was recorded by Shrı n ı vasa (1982).
According to our results, it was observed a low correlation between PH andOR in 1995 and1997 (r= 0.144 andr= 0.105, respectively).In 1998, this correlation was a negative effect and non-significant statistically.Gencer (1986) stated that OR was correlated with PH at 0.05 level.
3. Correlations between SY, TSW and SeY were positive in 1997 and 1998.But in 1995, these two correlations were low and negative effects 4. In 1997, correlation between PH and HD had high and positive effect (P<0.01).In 1995 and 1998, this correlation was also positive, but a low effect.5. On the contrary correlation in 1998, PH was negatively associated with HR (r= -0.398) in 1995.
6.In 1995, there was a significant (P<0.05) and positively effect correlation between TSW and HR.

Conclusion
The important yield components of sunflower consist of SeY, SY, PH, HD, TSW, HR, OR and oil yield.These characters affect positively or negatively on each other.For example, SeY, PH, HD, OR and TSW had positive effects on oil yield (Vanozzi 1987, Soltani and Archi 1998and Visic 1998).According to results of correlations analysis in the study; HD, SY, PH and TSW; PH; HD and SY had a positive effect on SeY; HD; TSW respectively.In addition, OR and HD were adversely correlated in 1997.The development of new varieties with high seed yield and oil concentration, and low husk ratio is an important goal among sunflower breeders.To attain this aim, it is required that bilateral relations in between yield and its components are known.For example, Toms and Pooni (1995) used analysis of correlations between the characters to investigate whether there were barriers to obtaining hybrids with a favorable combination of traits for early flowering, short stature and high yield.Therefore, we think that these findings obtained from our study may be useful for researchers who want to study in sunflower breeding.
R ) a n d o i l r a t i o ( O R ) v a l u e s w e r e r e c o r d e d .

,
Tablel.The yield and yield components values of hybrid, line and varieties