Materials and Methods

This study screened some root and shoot traits in full grain maturity (GS 92) of 47 cultivars, lines, landraces, ancient wheat species and wild wheat relatives belonging to 14 different species under field conditions at Konya, Turkey during the 2013-2014 growing season. The soil medium consisted of a mixture of peat (70 %) and perlite (30 %). Soil samples were taken before sowing and analyzed for certain chemical and physical parameters. The soil at the experimental area has a loam texture and is slightly acidic, high in organic matter, and calcareous. It is adequate for K₂O, Zn, and Cu and high for Mg. In addition, P₂O₅, Ca, and Mn is found in the soil as very high. The climate of the Konya can be defined as semiarid continental. According to the meteorological data, the long-term (1980-2013) and average annual rainfall (2013-2014) is 310.9 and 301.1 mm, the average annual temperature is 10.3 and 12.5 °C, respectively.

In the study, 47 genotypes of Triticum aestivum L., Triticum Durum Desf., and Triticum compactum Host, ancient wheat species and wild wheat relatives such as Triticum spp. and Haynaldia spp. were studied (Table 1). Each genotype was sown in October toa cylindrical PVC tube that was 200 cm in height and 12 cm in diameter, which had previously been replaced to soil excavated by a backhoe (Figure 1). The tubes were established in 15×15 cm row and intra row spaces. The experimental design was a “randomized complete block design” with three replications.

Figure 1 PVC tubes were replaced to above 200 cm depth in soil under field environmental conditions 

After emergence, one seedling per tube was allowed to grow. At sowing, the fertilizer DAP (18 % N, 46 % P₂O₅) 130 kg ha^-1 was top-dressed on all plots. At the stem elongation stage (GS 31) and completing of flowering (GS 69), the plants were drip irrigated (141-tube) with a solution containing 37.5 g urea (46 % N), 64 g micro elements, and 11.8 cc humic acid. The plants were watered with tap water at three times, stages of tillering, stem elongation and completion of anthesis.

At GS 92 (middle of July), the plant roots were washed and cleaned on the sieve and the longest root length was measured on a flat surface (Figure 2). In addition, number of secondary roots per plant was counted. Shoot traits such as plant height per main stem, number of fertile tillers per plant, spike length per main spike, number of spikelets per main spike, number of kernels per main spike, kernel weight per main spike and grain yield per plant were determined.

Figure 2 Roots were washed on sieve after nylon bag were removed from root media 

The statistical significance of the means was determined by analysis of variance using the statistical packages, MSTAT-C followed by comparisons by LSD test.

Results and Discussion

Table 2 shows the results of variance analysis related to the root and shoot traits of cultivars, lines, landraces, ancient wheat species and wild wheat relatives. The average values and groups of significance are given in Table 3. A significant difference was observed between the cultivars, lines, landraces, ancient wheat species and wild wheat relatives with regards to investigated traits (P ≤ 0.01).

Shoot Traits. Plant height per main stem showed differences according to the genotypes. The plant height of all the genotypes ranged from 66.7 to 148.5 cm. The landraces, ancient wheat species and wild wheat relatives out of Aegilops biuncialis and Gır had long plant stem. The highest plant height of 148.5 cm was observed in two species, Triticum turgidum and Triticum boeoticum. The Australian wheat genotypes, Tamaroi (68.3 cm) and line 5,924 (66.7 cm) had the shortest plant height. Among the 222 winter wheat genotypes, the stem height varied between 110 and 133 cm and the most of landraces had a very long stem, however obsolete bred cultivars had a shorter stem (^{Dotlačil et al. 2003}). Similarly, the results indicated that most of wheat landraces, ancient wheat species and wild wheat relatives had longer plant height. Cultivars originated from abroad had shorter stem height than wheat landraces, ancient wheat species and wild wheat relatives.

The spike length per main spike varied from 5.7 to 11.8 cm in cultivars, lines and landraces and 3.8 to 10.8 cm in ancient wheat species and wild wheat relatives. Similarly, it was found that spike length of Aegilops biuncialis (3.5 cm) was shorter than that of Triticum dicoccon (7.3 cm) and Triticum monococcum (8.7 cm) (^{Karagöz et al. 2006}).

The number of spikelets per main spike ranged between 14.3 (Bayraktar 2000) and 23.7 (Çeşit 1252) in cultivars, lines and landraces and 3.3 (Aegilops biuncialis) and 36.3 (Triticum monococcum) in ancient wheat species and wild wheat relatives. There was a considerable difference between the ancient wheat species and wild wheat relatives, and the cultivated wheat genotypes in terms of spikelet number.

The number of kernels per main spike ranged from 23.7 to 68.0 in cultivars, lines and landraces and 4.7 to 36.5 in ancient wheat species and wild wheat relatives. PhaNIL (Triticum compactum) (68.0) had maximum number of kernels, while Aegilops biuncialis (4.7) had minimum number of kernels.

The maximum and minimum kernel weight per main spike was obtained in landraces; Among the genotypes, Kamut had the maximum kernel weight (2.46 g), while Ribasa 2 had the minimum kernel weight (0.50 g). Among the ancient wheat species and wild wheat relatives, Aegilops biuncialis had the lowest kernel weight per main spike (0.09 g) and Triticum vavilovii had the highest kernel weight (1.76 g).

The number of tillers per plant changed from 3.7 to 17.0 in cultivars, lines and landraces and 4.7 to 30.0 in ancient wheat species and wild wheat relatives. Genotypes that had more tiller per plant were not always high yielding because the grain yield was affected by yield components such as number of spikelets, number of kernels, and kernel weight per main spike.

The grain yield per plant ranged from 2.11 to 12.30 g in the cultivars, lines and landraces and 0.23 to 6.49 g in the ancient wheat species and wild wheat relatives. In the study, Triticum aestivum genotypes, Vanlı (12.30 g), Tosunbey (9.27 g), Esperya (8.74 g) and Sönmez 2001 (8.45 g), Triticum durum genotypes, Berkmen 469 (9.44 g) and Triticum compactum genotype, AK 702 (11.68 g) had higher grain yield per plant. Triticum turgidum (5.90 g), Triticum turanicum (6.49 g), and Triticum vavilovii (6.30 g) had higher grain yield among the ancient wheat species and wild wheat relatives. However, Triticum boeoticum, Triticum monococcum (Kelcyras), Triticum monococcum (982) and Aegilops biuncialis had very low grain yield.

Root traits. The secondary root number widely varied in the evaluated genotypes, ranging from 42.5 to 98.3 for wheat cultivars, lines, and landraces and from 25.3 to 86.3 for ancient wheat species and wild wheat relatives. ^{Manske et al. (2002)} observed that there are two types of root in cereals, i.e., primary and secondary roots. The primary roots are called the first root or seminal root, and the secondary roots are known adventitious root, coleoptilar root, or nodal root. The secondary roots develop from first leaf node under 1-2 cm of soil when the leaf of the fourth main stem appears. ^{Pinthus (1969)} showed that late cultivars have not only larger number of secondary roots than early cultivars, taking a long period between germination and heading and but they have also more tillers. To some extent, the number of roots increases in proportion to the number of tillers (^{Roasti 2005}). However, in the study, Aegilops biuncialis had the highest tiller number among the genotypes, while it had the lowest secondary root number.

Wheat cultivars, lines, landraces, ancient wheat species and wild wheat relatives showed significant differences in terms of root length, which varied from 134.7 to 250.7 cm for cultivars, lines and landraces and from 170 to 240 cm for ancient wheat species and wild wheat relatives. A study on the drought tolerance of wild barley in the early growth stages has indicated that the most significant trait is root length, followed by shoot length and root shoot length-1 ratio (^{Tyagi et al. 2011}). The root length of wild barley (Hordeum vulgare L. ssp. spontaneum) was up to 91 % greater than the spring barley cultivar, Scarlett (^{Sayed 2011}). The average root length of wheat cultivars, lines and landraces was 216.8 cm, however ancient wheat species and wild wheat relatives had 212.2 cm. Root length has been shown to reach up to 2 m in soil (^{Gregory 1976}, ^{Hoad et al. 2001}, ^{Botwright Acuña & Wade 2012}), and up to 5 m in sandy soil (^{Zhang & Hu. 2013}). Here, the wheat root reached up to 2.5 m under favorable conditions. A landrace genotype, Gır had minimum root length, however Sönmez 2001 had the maximum. In addition, Aegilops biuncialis had the shortest root system among the wild wheats and wheat relatives. Genotypes with deeper root system may have adaptation mechanisms. Deep-rooted cultivars absorb water and nitrogen from deep soil (^{Smika & Grabouski 1976}). Genes controlling root length may become drought tolerant by avoiding or delaying the drought effects (^{Ober 2008}). The results of study indicated that among the cultivars, lines and landraces, Sönmez 2001, line 5924, AK 702 and ARS Amber that had a root length of 240.0 cm and above can be used in breeding programs to obtain deep-rooted genotypes. Triticum turgidum and Triticum vavilovii that had higher grain yield and longer root length comparing to ancient wheat species and wild wheat relatives can be considered to improve superior cultivars.

Conclusions

The evaluated cultivars, lines, landraces, ancient wheat species and wild wheat relatives showed wide range of genetic variation in terms of root and shoot traits. The average root length of wheat cultivars, lines and landraces was 216.8 cm, while that of wild wheats and wheat relatives was 212.2 cm. In the study, Sönmez 2001, Bezostaja 1, 5924 (line), AK 702, ARS Amber and Triticum turgidum that had up to 240.0 cm root length could be considered in breeding programs to improve deep rooted genotypes.

The study showed that ancient wheat species and wild wheat relatives such as Triticum monococcum, Triticum boeoticum and Haynaldia villosa resulted lower grain yield than other genotypes. However, the ancient wheat species and wild wheat relatives are known to be the most important sources of genetic wealth, providing resistance to biotic and abiotic stresses. Furthermore, Triticum vavilovii and Triticum turgidum that had higher root length and grain yield among the ancient wheat species and wild wheat relatives can be evaluated in breeding programs to improve the genotypes with high yield and deep-root system. Among the cultivars, lines and landraces, Sönmez 2001, Bezostaja1, AK 702, ARS Amber, and line 5964 that had longer root system can be considered to improve deep-rooted genotypes. In addition to deep-rooting system of genotypes, more study should be performed at field conditions where plants are compared with grain yield in large plots.