Variation in Hay Yield and Quality of New Triticale Lines

Objectives of the present study are to determine the variations in forage yield and quality of new triticale lines developed by cross-breeding methods and to develop better lines with superior characteristics over the current ones. Experiments were carried out in randomized complete block design with 3 replications during the cropping years of 2011-2012 and 2012-2013. A total of 5 cultivar and 20 lines were used as the plant material of the experiments. Genotypes were harvested at milk stage and their herbage yield, hay yield, crude protein yield and chemical composition were investigated. Twoyears average results revealed that herbage yields varied between 36.44-48.47 t ha-1, hay yields between 12.77-18.68 t ha-1, crude protein yields between 1.02-1.80 t ha-1, acid detergent fiber (ADF) between 32.92-44.63%, neutral detergent fiber (NDF) between 63.72-78.47%, crude ash ratios between 5.06-7.87%, crude protein content between 6.21-11.36%, dry matter digestibility (DDM) between 54.14-63.25%, dry matter intake (DMI) between 1.528-1.881 and relative feed value (RFV) between 64.18-89.31. Current results revealed superior characteristics for new triticale genotypes developed with cross-breeding over the current standard lines with regard to investigated parameters. It was concluded that cross-breeding yielded positive outcomes and therefore, currently investigated high-yield and quality lines should be prepared for registration.


Introduction
Cool season cereals like oat, barley, wheat, rye and triticale are usually grazed during tillering period or commonly harvested as forage source for livestock (Uncuer 2011).Triticale is the cross-breed of wheat and rye and used both for kernel and herbage yields (Iğne et al 2007).While triticale provides at least 20% more hay yield than wheat, forage quality is also better than wheat and rye (Koch & Paisley 2002;Mut et al 2006).Triticale is also a good feed source for livestock because of its high protein yield and amino acid balance.
Breeding materials are usually evaluated with regard to yield characteristics, resistance against biotic and abiotic stress conditions and forage quality parameters.Forage quality is commonly evaluated by Relative Feed Value criteria developed in the USA for alfalfa and other coarse fodder (Rohweder et al 1978;Ball et al 1996;Bozkurt 2011).Such quality varies based on crop species, harvest or grazing durations, cultural practices and climate parameters.Besides, livestock productivity depends on the amount of feed consumed by the animals, availability and digestibility of the feed (Van Soest 1982;McDonald et al 1995;Lekgari et al 2008).In present study, variations in forage yields and quality of some triticale lines developed by crossbreeding methods were investigated and better lines with superior characteristics over the current registered ones were tried to be developed.This will be the first study in Turkey investigating forage yield and quality parameters of triticale lines.

Material and Methods
Experiments were carried out over the research fields of Eastern Mediterranean Transition Zone Agricultural Research Center during the cropping years of 2011-12 and 2012 -13.The genotypes used in experiments are provided in Table 1.
The research province, Kahramanmaraş is located in Eastern-Mediterranean Region between 37 o 38' North latitudes and 36 o 37' East longitudes and has an altitude of 568 m.Mediterranean climate is dominant in the province and day-night temperature difference is low.Climate parameters for research site are provided in Table 2 (TSMS 2012;TSMS 2013).
While the long-term average precipitation of the experimental site is 669.1 mm, annual total precipitations of the experimental years 2011-12 and 2012-13 respectively realized as 756.8 and 583.1 mm.The first cropping year had 87.7 mm higher precipitation and the second year had 86 mm lower precipitation than the long-term average.Beside the amount, distribution of the precipitation within the year also significantly varied between the years.Especially, the amount of precipitation during the germination, emergence and initial growth period of the second year (November-December) were relatively lower than the long-term average.The amount of precipitations during January and February of the first year were significantly higher than that of the second year and long-term average (Table 2).On the other hand, the amount of precipitations during plant generative development periods (booting, spiking, flowering) covering the months March and April of both years were below the long-term average.Long-term annual average protein verimi 1.02-1.80t ha -1 , asitte çözünmeyen lif (ADF) oranı % 32.92-44.63,nötrde çözünmeyen lif (NDF) oranı % 63. 72-78.47, ham kül oranı % 5.06-7.87, ham protein oranı % 6.21-11.36,sindirilebilir kuru madde (SKM) % 54.14-63.25,kuru madde tüketimi (KMT) 1.528-1.881ve nispi yem değeri (NYD) 64.18-89.31arasında değişmiştir.Araştırma sonuçlarına göre; melezleme ile elde edilen yeni tritikale genotiplerinin incelenen özellikler yönünden değerleri standart çeşitlerden daha yüksek olmuştur.Yapılan melezleme sonuçlarının olumlu neticeler verdiği ve bu hatların ot verimi ve kalitesi yönünden tescil için hazırlanması gerektiği düşünülmektedir.

52
Ta r ı m B i l i m l e r i D e r g i s i -J o u r n a l o f A g r i c u l t u r a l S c i e n c e s 21 (2015) 50-60 temperature of Kahramanmaraş Province is 12.6 º C. Annual average temperatures of the experimental years respectively realized as 12.5 and 14.0 º C. Temperature in April of the first year was higher and the temperature in May was lower than the second year and long-term average.Annual average relative humidity of the province is 62.0% and the relative humidity of the experimental years respectively realized as 57.2 and 61.9%.
The soil texture is loamy with a soil pH of 7.61 in the first year and 8.00 in the second year.Lime contents were respectively observed as 12.55% and 24.59%.Available phosphorus varied between 0.046 -0.080 t ha -1 , available potassium between 0.459 -1.270 t ha -1 .Organic matter contents of the soils varied between 1.22 -0.97.
Experiments were carried out in randomized complete block design with 3 replications.Seeding rate was 500 seeds m -2 and seeding was performed with a plot-drill over 6 x 1.5 m size plots.There were 6 rows in each plot with row spacing of 20 cm.In both years, 0.08 t N and 0.08 t P2O5 ha -1 were applied to soil during sowings and additional 0.1 t ha -1 N was supplied during tillering period.Irrigation was not performed in both years and herbicide (Grand Star) was used for broad-leaf weeds.Plants were harvested at milk-stage.Side rows and 1 m strips at top and bottom of the plots were omitted as side effects.A total of 500 g fresh sample was taken from harvested plants and dried at 70 ºC for 48 hours.Then, dry matter ratios and hay yields were determined.Hay samples were grinded in a hand-mill with 1 mm sieve.Crude ash contents of the samples were determined by burning the samples at 550 °C for 8 hours.Kjeldahl method was used to determine nitrogen (N) contents of dry samples.Crude protein ratios were calculated by using the equation of N x 6.25 (AOAC 1990).NDF (Van Soest & Wine 1967) and ADF (Van Soest 1963) contents were analyzed with an ANKOM 200 Fiber Analyzer (ANKOM Technology Corp. Fairport, NY, USA) device.Dry matter digestibility, dry matter intake and relative feed value (RFV) were calculated by using the following equations (Rohweder et al 1978): To calculate relative feed value, initially dry mater digestibility (DMD) was calculated from ADF value by; DMD% = 88.9 -(0.779 x ADF%) (1) Dry matter intake (DMI) based on animal liveweight was calculated from NDF value by; Then, relative feed value was calculated from DMD and DMI by; Relative feed values were evaluated by using the values provided in standards for hays in Table 4.
Data variance analyses were performed by using SAS (SAS Inst., 1999) statistical software.Duncan's multiple range tests was employed to compare the treatment means.Cluster analysis of 25 triticale lines was performed by using DICE similarity index and UPGMA method and dendrograms were prepared for yield, chemical composition and digestibility parameters.

Results and Discussion
Herbage, hay and protein yields of triticale genotypes are provided in Table 5.While only the differences between protein yields of the years were not significant, the differences between the entire parameters of the years, genotypes and year x genotype interactions were found to be significant (P<0.01).The lowest herbage yield of the first year was observed in line L9 (30.09 t ha -1 ) and the highest in L19 (49.56 t ha -1 ).During the second year, the lowest and the highest values were respectively observed in L21 (32.84 t ha -1 ) and Melez 2001 (54.69 t ha -1 ).With regard to average of years, the lowest herbage yield was obtained from L21 (36.44 t ha -1 ) and the highest from L4 (48.47 t ha -1 ).The lowest hay yields were observed in cultivar Karma with 12.44 t ha -1 in the first year, 13.09 t ha -1 in the second year and 12.77 t ha -1 as the average of years.The highest hay yield was obtained from L19 (21.13 t ha -1 ) in the first year and from L8 in the second year and as the average of years (respectively with 20.36 and 18.68 t ha - 1 ).The lowest crude protein yield was observed in L1 (0.96 t ha -1 ) in the first year, in L7 (0.99 t ha -1 ) in the second year and in L12 (1.02 t ha -1 ) as the average of years.The highest crude protein yield of the years and average of the years were all observed in L12 respectively as 1.77, 1.80 and 1.78 t ha -1 .
Chemical composition of the triticale genotypes are provided in Table 6.The differences between entire chemical composition parameters were found to be significant (P<0.01).ADF ratios varied between 30.97-46.45% in 2011 with the lowest ratio in L24 and the highest in L12.The ADF values of the year 2012 varied between 33.48-44.44%with the lowest value in the cultivar Alperbey and the highest value in L7.With regard to average of years, the lowest value was seen in L24 (32.92%) and the highest value in L12 (44.63%).The NDF ratios of the first year varied between 59.84-79.28%with the lowest value in L1 and the highest value in L12.During the second year, NDF values varied between 59.52-77.66%with the lowest ratio in cultivar Karma and the highest ratio in L12.Considering the average of years, the values varied between 63.72-78.47%with the lowest value in L2 and the highest value in L12.The lowest crude ash ratio of the first year was observed in L18 (4.85%), the lowest value of the second year in L21 (4.88%) and the lowest average of years in L21 (5.06%).The highest values of the first and second year and the average of years were respectively observed in L9 (8.89%), L3 (7.82%) and L9 (7.87%).Relatively larger differences were observed in crude protein ratios of the triticale genotypes.While L1 yielded the lowest crude protein ratio in both years and average of years, the highest value was obtained from L12 (12.48%) in the first year, from cultivar Mikam (10.69%) and as the average of years from L12 (11.36%).Ta r ı m B i l i m l e r i D e r g i s i -J o u r n a l o f A g r i c u l t u r a l S c i e n c e s 21 (2015) 50-60    Digestibility parameters of the triticale genotypes are provided in Table 7.The differences between entire digestibility parameters were found to be significant (P<0.01).DMD values of the first year varied between 52.71 -64.78% with the lowest value in L12 and the highest value in L24.The values of the second year varied between 54.28 -62.82% with the lowest value in L7 and the highest value in cultivar Alperbey.With regard to average of years, DMD values varied between 54.14 -63.25% with the lowest value in L12 and the highest value in L24.The lowest DMI ratios were obtained from line L12 as of 1.510% in the first year, 1.547% in the second year and 1.528% in the average of years.The highest DMI value was observed in L1 (2.006%) in the first year, in cultivar Karma (2.017%) in the second year  and in line L2 (1.881%) as the average of years.
Considering the RFV values of the genotypes, the lowest values were obtained from L12 in both years and the highest values were obtained from L24 in the first year, cultivar Karma in the second year and L2 in the average of years.
Similarity levels varied between 0.06-1.90and there were two main groups (A and B).The first group (A) was composed of 12 triticale lines and the lines in this group (Line2, Line16, Melez 2001, Line14, Alperbey, Line4, Line18, Line12, Line22, Line24, Line6 and Line8) were separated from the second group (B) with a similarity level of 0.78.Genetically, the Line14 and Alperbey genotypes were found to be 99% similar with each other.The first group was divided into two sub-groups (A.1 and A.2) with a similarity level of 0.70.The subgroup A.1 was composed of Line8 and Line 6 with a similarity level of 0.53.
The second group (B) was composed of 13 triticale lines and separated from the first group (A) with a similarity level of 0.55.The second group was also divided into two sub-groups (B.1 and B.2) with a similarity level of 0.57.The first sub-group of the second group (B.1) was separated from the second sub-group (B.2) with a similarity level of 0.57.Only the Line 23 was placed into the B.1 and the lines Line1, Line13, Line3, Line9, Line11, Line19, Line21, Line7, Karma, Mikam 2002, Tatlıcak, Line17 and L23 were placed into the B.2. Within B.2, the lines Line11 and Line19 were the closest lines to each other and they separated from each other with a similarity level of about 0.10 (Figure 1).
Precipitations of the experimental years 2011-2012 and 2012-2013 were significantly different from each other.Especially the lower precipitations during March and April of the first year and higher precipitations of the second year significantly effected yields and chemical compositions and resulted in significant differences between these parameters.Different plant growth levels in March and April and different responses against water deficits resulted in significant year x genotype interaction.
Herbage values of the present study were higher than the values reported by Kaplan et al (2011) and similar to the values reported by Lithourgidis et al (2006) andSurmen et al (2011).Hay yields of the current study were similar to the ones reported by Delogu et al (2002); Albayrak et al (2006); Mut et al (2006) and Kaplan et al (2011); and higher than the values of Lithourgidis et al (2006).Such differences in hay yields were mainly due to differences in climate conditions and different responses of genotypes against different conditions.These differences may also result from higher nutrient accumulation levels of early-spiking plants (Delogu et al 2002).
Crude protein content is an essential parameter to evaluate the quality of forages (Caballero et al 1995;Assefa & Ledin, 2001).Differences in dry matter and crude protein contents usually come from the genetic characteristics of plants but spike-shoot ratio, growing period, temperature and fertilizers are also effective on both parameters (Ball et al 2001).While protein yields of the present study were similar to values reported by Kaplan et al ( 2011

53
Ta r ı m B i l i m l e r i D e r g i s i -J o u r n a l o f A g r i c u l t u r a l S c i e n c e s 21 (2015) 50-60

54
Ta r ı m B i l i m l e r i D e r g i s i -J o u r n a l o f A g r i c u l t u r a l S c i e n c e s 21 (2015) 50-60 with common superscripts do not differ (P>0.05);NS, non-significant; Sig., significance level; *, P<0.05; ** , P<0.01; HeyY, , herbage yield; HY, hay Yield; CPY, crude protein yield Ta r ı m B i l i m l e r i D e r g i s i -J o u r n a l o f A g r i c u l t u r a l S c i e n c e s 21 (2015) 50-60 with common superscripts do not differ (P>0.05);NS, non-significant; Sig., significance level; * , P<0.05; ** , P<0.01; CP, crude protein; NDF, neutral detergent fiber; ADF, acid detergent fiber; CA, crude ash 57 Ta r ı m B i l i m l e r i D e r g i s i -J o u r n a l o f A g r i c u l t u r a l S c i e n c e s 21 (2015) 50-60

58
Ta r ı m B i l i m l e r i D e r g i s i -J o u r n a l o f A g r i c u l t u r a l S c i e n c e s 21 (2015) 50-60 ); Lithourgidis et al (2006); Surmen et al (2011) and Mut et al (2006), crude protein ratios were similar

Table 4 -Relative feed value standards a
Çizelge 4-Nispi yem değeri standartları a feed value is assumed to be 100 when the ADF is 41% and NDF is 53%(Rohweder et al 1978) a, relative