EFFECTS OF ABIOTIC FACTORS ON FULFILLMENT OF THE POTENTIAL OF PEA CULTIVARS

The results of analysis of studies carried out in the Laboratory of Peas Breeding of the Plant Production Institute named after VYa Yuriev are presented. Effects of abiotic factors on the yield and protein content in pea seeds during interphase periods are described. Determination of the effects of abiotic factors on the yield and protein content in pea seeds on average during the vegetation period does not give objective information. Determination of the effects of Σ t and Σ Р on the traits, depending on developmental phases, is more informative. The aim and tasks of the study was to evaluate the effects of the abiotic factors (temperature regime and precipitation amount) on the yield and protein content in seeds of pea cultivars. The experiments Breeding of different eco-geographical origin. and HTC) on the yield and other traits on average over the study period may not be sufficiently informative. Evaluation of the effects of Σ P and Σ t during specific phases of vegetation on the traits may be more informa-tive.

conditions for a long time -from onset of active growth to bean filling [11]. Omelianiuk LV et al marked that both overwetting and lack absence of precipitation during mass anthesis and at the beginning of bean formation in the southern forest-steppe of the Omsk region significantly reduced the crop potential [12]. Lozhkina OV established that the performance was influenced both by varietal characteristics and by agroecological conditions, including water regime throughout the vegetation period. Mid-ripening locally-bred pea accessions with normal leaves were the most adapted to the southern forest-steppe in the Omsk region [13]. High drought-tolerance of local pea cultivars created in the Bashkir Research Institute of Agriculture (the western steppe piedmont of the Ural Mountains) by Popov BK is attributed to their biological characteristics -accelerated rate of growth and accumulation of macro nutrientsduring an extended vegetative period [14].
Summarizing the literature data [8-10, 12, 13, 15, 16], we can assume that the highest performance in an area can be obtained from local cultivars and breeding material, although sometimes there are cultivars with a wide norm of reaction. Assessment of adaptability not only of one's own breeding material but also of collections allows identifying valuable genotypes of different eco-geographical origin, which are further involved in breeding [17]. However, use of nothing but high-yielding accessions upon creation of new material can lead to loss of environmental stability. Since the averages of the -susceptibility to the environment‖ traits are relatively independent and genetically determined separately, breeding for environmental stability should be monitored continuously [18].
Our purpose was to determine the effects of abiotic factors (temperature regime and precipitation amount) on the yield and protein content in pea cultivars.
Materials and methods. The experiments were carried out in the experimental field of the Plant Production Institute named after VYa Yuriev (PPI nd. a. VYa Yuriev) in 2008-2017. In the competitive trial, seeds were sown with a seeder "Klen" equipped with a batch apparatus. The seeding rate was 1,200,000 germinable seeds per hectare. The plot area in the competitive trial was 20 m 2 . The experimental variants were arranged in a random design and carried out in 4 replicas [19]. Phenological observations were conducted; uniformity of plants in the plot was evaluated.
The Laboratory of Genetics, Biotechnology and Quality of the PPI nd. a. VYa Yuriev NAAS analyzed protein content in pea seeds on an Infralium FT-10.
The hydrothermal coefficient (HTC) was calculated, as Selianinov described [20]. The experimental data were statistically processed in standard Microsoft software, as described in [19].
Results and discussion. In the study period, the hydrothermal regime significantly fluctuated during the ontogenesis of peas plants. Thus, analyzing the temperature sum (Σ t ), precipitation amount (Σ P ), HTC and yield, we found that the smallest Σ t was in 2008 and 2017 -1,003.2°C and 1,092.2°C, respectively, with Σ P of 119.2 mm in 2008 and only of 54.2 mm in 2017, nevertheless, the yield was the highest in 2008 -4.31 t/ha ( Table 1). The largest Σ t was recorded in 2013 (1,349.2 С), with Σ P of 97.1 mm, however the yield in 2013 was the lowest during the study period -1.45 t/ha.
When we assessed the effect of Σ P on the yield, we observed neither visible nor statistically significant dependencies. Thus, for example, in 2011 Σ Р was 211.2 mm, and the yield was 2.29 t/ha, and in 2012 and 2017 Σ Р was 75.5 mm and 54.2 mm, respectively, with higher average yields of 2.54 t/ha and 2.41 t/ha, respectively.
The statistical analysis confirmed this by a negligible correlation coefficient between Σ P and yield (r = -0.10), and a similar insignificant coefficient was between the HTC and yield (r = 0.02). The correlation coefficient between Σt and yield (r = -0.64) was significant and negative; this is quite consistent with other authors'data reporting a greater impact of temperature regime during ontogenesis on the pea yield, although there is a completely opposite opinion [1, 6 (p. 257)]. The analysis of sowing dates and some phases of vegetation showed that the crop was sown in April throughout the study years, and, depending on the conditions of a year, shoots appeared 8-18 days later ( Table 2). In general, in 2008-2017, the anthesis onset was within the first 10 days of June, except for 2014, when this phase was within the third 10 days of May. Bean filling began 7-10 days after the anthesis onset, i. e., throughout the study period, except for 2014, this phase occurred within the second 10 days of June.
Unlike the «yield» trait, the variability range for the «protein content» trait over the study years was insignificant: V = 4.02-8.64%. Such values the variation coefficient can be accounted for by low phenotypic variability of the «protein content» trait, which is encoded in the cultivargenotypes in this sample, in contrast to the variability range for the «yield» trait, which is more influenced by the vegetation conditions (each plant responds differently to this factor), and this is reflected in the overall performance of an agrocenosis.
The overall ranges of phenotypic variability over the study years for the «yield» and «protein content» traits were within 0.92 t/ha-4.80 t/ha and 15.59%-26.01%, respectively. The environmental variability of the yield was within 1.45-4.31 t/ha and of the protein content -within 17.90-24.04%. The genotypic variability for the «yield» trait was within 1.85-2.42 t/ha, with significant variation of the trait between cultivars (V = 31.03%-47.72%) (see Table 4); the genotypic variability for the «protein content» trait was within 19.45%-23.41%, with slight or medium variation (V = 7.24-10.34%) (see Table 5).  Over the study years, during the third 10 days of April (sprouting phase), the effects of Σ P and HTC on the yield were significant in three cultivars only: Tsarevych (r = 0.65 and r = 0.69, respectively), Orlovchanin (r = 0.71 and r = 0.74, respectively) and Damir 2 (r = 0.65 and r = 0.69, respectively). During the first 10 days of May, the correlation coefficients were significantly negative for two cultivars: Intensyvnyi 92 and Damir 2 (r = -0.64 and r = -0.68, respectively). We observed no significant influence in the other cultivars either in April or in May. Table 4 presents the results of correlation analysis between the «yield» trait, Σ P and Σ t only for the periods (10-day periods) with the largest numbers of significant indices. The correlation coefficients between the HTC and yield are not presented, because they were practically identical to the correlation coefficients between Σ P and yield.
The HTC effect on the yield during the second 10 days of June (bean filling phase) was significant only in cultivar Ramon 77 (r = 0.64), with r = 0.63 for Σ P .
It is noteworthy that the effect of Σ t on the yield during the first 10 days of June (anthesis onset) was significant in 11 cultivars, and later the situation changed, as duringthe second 10 days of June (bean filling) the effect of Σ P became significant in the vast majority of cultivars (20). In addition, over the study years, the effect of Σ t on the yield was significant in 9 cultivars only.
Thus, it was established that on average over the study years and during the anthesis phase Σt negatively affected the yield with various levels of significance and that Σ P during the bean filling phase significantly affected on the yield in the vast majority of cultivars.
The effect of HTC on the protein content in seeds in the study years was significant only in three cultivars-Kharkivskyi Etalonnyi, Deviz and Oplotwith r = 0.66, r = 0.66 and r = 0.69, respectively. Kharkivskyi Yantarnyi was the only cultivar in which the protein content in seeds was affected by Σt during the second 10 days of May (r = 0.74). Table 5 presents the results of correlation analysis between the «protein content in seeds» trait, Σ P , Σ t and HTC for the periods (10-day periods), during which the largest numbers of significant indices was obtained.
In contrast to the «yield» trait, the protein content was significantly influenced by Σ P in eight cultivars only over the study years. Among them, there were cultivars, in which Σ t significantly influenced the yield over the study years -Kharkivskyi Etalonnyi, Oplot and Damir 2 (see Table 4).
The effect of Σ P during the first 10 days of May on the protein content was significant in ten cultivars, and the effect of the HTC remained significant in six cultivars. During the second 10 days of May, the effect of the HTC became significant in nine cultivars, however, it changed the value from positive to negative. It should be noted that in cultivars Vusatyi 90 and Oplot the effect of HTC on the protein content persisted during the first and second 10-day periods of May.
In the first 10 days of June (anthesis onset), both Σ t and Σ P significantly influenced the «protein content in seeds» trait in two cultivars only -Efektnyi and Tsarevych. In 12 cultivars from the sample under investigation, the effects of Σt and Σ P on the protein content in seeds were insignificant. The effect of Σt was significant in cultivars Kharkivskyi 302, Intensyivnyi 92, Rezonator, Kharkivskyi Etalonnyi, Modus, Oplot, Orlovchanin, Ramonskyi 77 and Kamelot, and, on the contrary, in cultivars Kharkivskyi Yantarnyi, Korvet, Zekon and Damir 2 the effect of Σ Pwas significant. Thus, the effects of Σ t, Σ P and HTC on the protein content in pea cultivars were not unidirectional, did not depend on the plant morphotype, cultivar origin and were most likely to be determined by the cultivar genotype.