Introduction

Barley (Hordeum vulgare L.) takes fourth place in worldwide importance behind wheat (215 million ha), rice (155 million ha), and maize (139 million ha) (^{Langridge and Barr, 2003}). According to the FAO (2013), more than 49 million hectares were planted with an average yield of 2.9 t ha-1. Due to its great adaptability even in extreme situations and ecosystems, barley is a widely distributed crop worldwide (^{Poehlman, 1985}); around 89 countries produce this grain, from subtropical regions (Africa, Brazil) to cold regions (Norway, Alaska). Regardless of its wide distribution, its production is significantly centered in the European Union, being the top barley producer with 46.1%; Russia, Canada, Australia, and the Ukraine represent 73% of the global barley production.

Per country, 50% of the global production is centered in China, the United States, Germany, and Brazil with 18.5%, 17.7%, 8%, and 5%, respectively. Of the total production, it is estimated that 25% is destined for the production of malt, raw material for the elaboration of beer, and 75% is used as animal fodder. Although it has great potential due to its beta-glucans contents, its use for human consumption is limited (^{Newton et al., 2011}).

According to the SIAP in Mexico during the 2013 agricultural cycle, 355 782 ha of barley were planted, of which 320 946 ha corresponded to malting barley, 33 491 ha to fodder barley, and 1 345 ha for the production of seeds. Of the surface area planted with malting barley, 296,912 ha were harvested obtaining 594 437 t and an average yield of 2 t ha^-1. The main producing states in the Bajío region are Guanajuato, Querétaro, Michoacán, and Jalisco, whereas in the Altiplano region, the production is centered in the states of Hidalgo, Puebla, Tlaxcala, and Estado de México, with the latter region being where 75% of the surface area is planted under seasonal conditions in the summer.

The distinction between malting and fodder barley is mainly based on the protein content. For malting barley, the protein content must be below 12%; whereas if it is to be used as fodder, the protein content must be higher. However, international statistics do not establish differences between barley’s use as fodder or in malt production. The protein content of the grain depends on various factors among which can be found fertilization, type of soil, temperature, and variety (^{Pitz, 1990}).

The malting quality in barley is a complex characteristic that, in addition to the physical properties of the grain, depends on the enzymes synthetized during the germination process (^{Thomas et al., 1996}). Nevertheless, the grain of malting barley must meet specific parameters that involve physical characteristics (size and weight of the grain) and chemical properties (extract percentage, protein content, Kolbach index, diastase strength, among others) (^{Molina-Cano et al., 1986}; ^{Narziss, 1990}; ^{Mather et al., 1997}).

The characteristics of malting quality are quantitative; therefore, their expression does not only depend on the genotype but is also influenced by various environmental factors and by the interaction of the genotype with the environment, making its inheritance a complex matter (^{Sparrow, 1971}; ^{Mather et al., 1997}; ^{Iguarta et al., 2000}; ^{Zale et al., 2000}). Research on barley is based on aspects related to the yield and disease control. Investigations focused on evaluating the contribution of genes to the malting quality characteristics are limited; this research being of interest for malt and beer industries (^{Hockett et al., 1993}).

In the Mexican norm NMX-FF-043-SCFI-2003, the conditions and characteristics that malting barley must meet for its commercialization are established. The grain must have between 11.5 and 13.5% humidity, have a minimum germination of 85%, an 85% grain size for its use as malt (grain retained in a 55.5/64” x –” sieve), maximum 5% naked and/or broken grains, 2% impurities, maximum 10% damaged grain, mixtures up to 10%, weight per hectoliter (which is the weight of a hectoliter of grain of the original sample free of impurities expressed in kilograms; kg hL), and a 56 kg hL minimum on six-row barley, whereas on two-row barley it must have a value of 58 kg hL. Furthermore, the organoleptic characteristics of the grain must be appropriate making sure that the grain is not dirty, damaged, stained, painted or contaminated, and that it does not have a putrid, stale, alcoholic, or chemical odor, among others.

It is estimated that in Mexico, 80% of the national barley production is destined to be turned into malt, whereas at a global level the percentage is lower. The production of barley required by the malting industry is mainly centered on two varieties: the Esperanza variety for the irrigation zones of the Bajío, and the Esmeralda variety for the seasonal zones of the Altiplano. The Alina and Armida varieties constitute other options for irrigation conditions, whereas the Adabella variety is recommended for seasonal zones with good productivity environments (^{Zamora et al., 2008}; ^{Solano et al., 2009}; ^{Zamora et al., 2010}). It is calculated that the industry requires a volume of 750 000 t of raw material per year in order to satisfy demand and thus cover the requirements for the production of beer.

Barley is one of the crops that offers a better production alternative in the seasonal areas of Mexico. Therefore, in addition to the varieties needing good agronomic attributes, the grain production must meet the required quality for the malting industry. Given the importance of the quality parameters of the barley meant to be used as malt, the Instituto Nacional de Investigaciones Forestales, Agrícolas y Pecuarias (INIFAP) carries out investigations in order to create genotypes that satisfy these quality parameters, are tolerant to diseases, and have an excellent yield.

In order to know the behavior of the genotypes generated in the research stage, agronomic evaluations and physical tests were carried out on advanced lines of malting barley with six-rows of grain in the ear, with the objective of determining the productive potential and physical quality based on the variables of weight per hectoliter and the percentage of malting grain, parameters which are included in the norm for the commercialization of malting barley and that will allow for an efficient selection of the best malting genotypes.

Materials and methods

The research project was carried out during the agricultural cycles of spring-summer 2012 and spring-summer 2013, under seasonal conditions in five contrasting localities: Experimental Field Mexico Valley (CEVAMEZ-INIFAP), Tlacatecpan and Polotitlán, Estado de México; Calpulalpan, Tlaxcala; and Experimental Site North of Guanajuato (SENGUA-INIFAP). 16 malting barley genotypes were evaluated (14 advanced lines and the Esmeralda and Adabella varieties as control). The tests were established in each locality under a 4 x 4 simple Lattice design with four replications. The dates of planting and fertilization doses were done according to the technical recommendations for the cultivation of seasonal malting barley provided by the INIFAP. The planting density was 80 kg ha^-1. Data was recorded on the days of flowering, days of physiological maturity, height of the plant, yield, weight per hectoliter, and percentage of malting quality grain. The yield was determined harvesting the total plot (3.6 m²), which consisted of four furrows 3 m in length with 0.3 m separating the rows.

In order to determine the weight per hectoliter, the method using a funnel and tray was considered; for this purpose, a 110 g sample free of impurities was taken and its volume was determined. The corresponding hectoliter weight was assigned to the obtained value with the aid of the table of values indicated in the Mexican norm NMX-FF- 043-SCFI-2003. The percentage of malting grain (PMG) was determined using a 200 g grain sample, passing it through two sieves, one of ⁶/₆₄” x ³/₄” and one of ^5.5/₆₄” x ³/₄” in order to see how much grain was retained in the sieve with a larger mesh. The quantity of grain retained by each sieve was weighed. Both quantities were added and the percentage was determined, given that the norm specifies that for this variable the test must be carried out using a ^5.5/₆₄” x ³/₄” sieve.

An analysis of variance was carried out per locality for the variables measured with the aid of the statistical program SAS9.0 (^{SAS, 2002}). Subsequently, a combined analysis of the localities was carried out for each evaluated agricultural cycle. The comparison of averages was done with the least significant difference (LSD, α= 0.01).

Results and discussion

The results obtained from the analysis of variance per locality showed significant differences for the majority of the evaluated variables. When a combined analysis was carried out, these differences were greater when interacting with genotypes with the environment, resulting in a higher significance for all the evaluated variables.

The behavior of the evaluated lines was mainly influenced by the environmental conditions that were presented in each agricultural cycle. For the 2012 cycle, highly significant statistical differences were detected (Table 1) between localities and genotypes for all the evaluated variables. These results could be attributed to the particular conditions of each locality where the materials were established, in addition to the specific behavior of each line in each locality.

The genotype interaction per locality showed a high statistical significance for the yield and the physical quality variables (weight per hectoliter and percentage of malting grain); however, the variables of days of flowering, days of maturity, and plant height were not significant for this interaction.

The results obtained from the analysis of variance for the 2013 agricultural cycle showed a greater statistical significance for the evaluated agronomic and quality variables (Table 2). For the interaction of genotypes per locality, height was not significant, maturity was significant, and the rest of the variables showed values with a high statistical significance (p≤ 0.01), which evidences the environmental variation between localities, the existence of genetic variability in the response of the barley lines, and the influence of environmental variation on the expression of each one of the evaluated genotypes.

On average, the agronomic and physical variables evaluated (Table 3) had flowering characteristics with a similar behavior between years in each locality. For the 2012 cycle, the values of flowering days (FD) were distributed on an average range of 46 to 57 days, with CEVAMEX being the most premature locality, although statistically it had a similar behavior to the localities of SENGUA and Tlacatecpan. During the 2013 cycle, SENGUA was the most premature locality, with a statistical value similar to CEVAMEX. The locality of Calpulalpan was the slowest for the two evaluated cycles.

In regard to the time required in order to reach physiological maturity, this variable behaved differently between each cultivation cycle. During the 2012 cycle, an average maturity of 98 days after planting was observed, whereas for the 2013 cycle, this increased to 106 days after planting. The locality of Tlacatecpan had an atypical behavior due to the climatic conditions that prevailed during the 2013 agricultural cycle (drought before tillering and rain during tillering and the beginning of flowering), causing the cultivation cycle to be prolonged for almost a month. In the other localities, the seasonal effect between the 2012 and 2013 cycles was about 5 days, being more delayed during the second cycle. Excluding the locality of Talcatecpan-2013, Calpulalpan was the slowest in the two agricultural cycles. This behavior is related to the climatic conditions of the region, considered semi-cold humid and characterized by its altitude (2 600 meters above sea level), high humidity, and low temperatures. The humidity and temperature conditions promote a slower development of the crop than in the other localities.

On average, the genotypes reach maturity around 100 days after planting (98 days after planting in 2012 and 106 days after planting in 2013); the required time from planting to harvest for the cultivation of malting barley in Mexico is much shorter in comparison with the data reported for this crop at other latitudes, where the environmental conditions cause the materials to mature much slower, even reaching maturity at up to 200 days after planting (^{Eshghi and Akhundova, 2009}; ^{Fedaku et al., 2014}).

Plant height was also affected by the climatic conditions of each locality, particularly due to the lack of water at the early stages of crop development resulting in a height difference for the plant between each cycle. Greater sizes were observed for the 2012 agricultural cycle (88.4 cm), whereas for the 2013 cycle, the height decreased by more than 20 cm (65.7 cm). The differences found between localities decreased when the genotypes within the localities were compared, observing in this manner similar heights between them (^{Tamm, 2003}). The plant height for malting barley is a highly important aspect, given that if it is too short, mechanical harvest becomes much more difficult. When these plants are very tall, however, it causes a greater susceptibility to being flattened (^{Eshghi and Akhundova, 2009}), which can gradually affect the final grain yield and furthermore it also makes harvest, given that it is mechanized, much more difficult. Plant height as observed in the evaluated lines can be considered adequate for the barley producing regions in Mexico.

The grain yield variable, as well as with the other variables, showed variations between localities and cycles; an average yield of 3.8 and 3.1 t ha^-1 was obtained for 2012 and 2013, respectively. During the 2012 agricultural cycle, the locality established in the municipality of Calpulalpan had the least grain yield with 2.1 t ha^-1. The best locality for this cycle corresponded to CEVAMEX whose average yield was 6.9 t ha^-1; in this locality, the most efficient lines surpassed 7 t ha^-1. In the 2013 cycle, the grain yields decreased up to 50% (SEGUA) with regard to the previous cycle; CEVAMEX, the best locality in this agricultural cycle, decreased by more than 1.5 t h^-1. This decrease was due to the lack of water, which caused a smaller grain size, less grain weight, and a smaller number of grains per ear (^{Ataei, 2006}).

Conclusions

The evaluated genotypes had premature behavior, on average flowering at 50 days and reaching maturity at around 100 days after planting.

The yields of the experimental lines were greater than the controls, with the average yield being 3.5 t ha^-1. The lines M176, M177, M178, and M184 behaved in a stable manner in the two agricultural cycles for all the evaluated localities; their average yields were 3.8, 3.5, 3.9, and 4 t ha^-1, respectively.

The volumetric weight was superior to the value specified in the norm, 56 kg hL. The lines with greater hectoliter weight surpassed the controls by at least 1 kg hL.

In various localities drought problems were presented; the lack of water caused low yields due to the fact that the grain did not fill-out adequately, given that when the water deficit was presented the genotypes were in the grain fill stage. This water deficit was an important factor, given its limitation affected the final yield of the evaluated experiments.

Through the application of this evaluation strategy under various test environments, the genetic improvement program for barley of the INIFAP has managed to obtain lines with good potential yield, tolerance to diseases, and good malting quality.