Introduction

In breeding programs of any species of agricultural interest, the selection of genotypes is basically considering yield, disease resistance and agronomic value, then it is necessary to assess the consistency of their behavior being subjected to different environments for several years within a potential for adaptation region. Triticum aestivum L. is a species that has a wide range of adaptation, grows and develops in very different environments and can be grown both in winter and spring, which together with its large consumption has allowed its cultivation spread widely to many parts of the world and currently ranks first among the four major global production cereals: wheat, rice, corn and barley (^{Martin, 1990}).

Wheat is one of the most important worldwide for the area sown and harvested grain, because it is power base in many countries and being considered by FAO as one of the staple crops by providing essential amino acids humans (^{Moreno, 2012}). As mentioned by the United States Department of Agriculture (USDA, 2013), world production of wheat during the crop cycle 2011- 2012 was 695 million tons; the European Union contributed 20%, China 17%, India 13%, Russia and the United States of America 8%, 4% Australia and Canada, Pakistan and Kazakhstan 3%, and the remaining 20% was contributed by other countries. Mexico is at number 23 worldwide production with 3. 8 million tons per year, highlighting its unit yield of 5.7 t ha^-1 in irrigation and 2 t ha^-1 in time.

In wheat production in Mexico include the states of Sonora is the main producing state followed by Baja California and Guanajuato which provide 77% of national production (^{SIAP, 2014}). The area planted to wheat in the summer varies, and the main producing region is the high valleys of Mexico formed by the states of Hidalgo, Mexico, Puebla and Tlaxcala where it is planted during the months of May and June and harvested in November and December (^{Huerta and Singh, 2000}).

In sowing time, at all stages of growth wheat plants are exposed to numerous damages and stresses that interfere with normal functioning and development; any resulting abnormal condition is, in a broad sense, a disease (^{Wiesse, 1977}), fungal diseases being one of the most important causes that decrease cereal yields (^{Stubbs et al., 1986}).

The research has allowed free varieties by the National Institute of Agricultural and Livestock Forestry Research (INIFAP), which have been important for the growth of the planted area and productivity in the rainfed areas of the country factor, highlighting within their attributes the short cycle, the bearing of the plant, its greater adaptability and drought tolerance. However it is necessary to continue the genetic improvement year after year, because the varieties released eventually lose resistance to new pathogens (^{Villaseñor, 2000}), so the evaluation of varieties and experimental lines will identify which varieties are currently the more suitable for planting and lines are candidates for new varieties.

Diseases currently affecting more wheat rusts are: rust leaf Puccinia triticina, yellow rust Puccinia striiformisf. sp. tritici and recently in some countries in Africa and Asia stem rust Puccinia graminisf. sp. tritici. These pathogens tend to thrive mainly on rainy to rainy environments, as the relative humidity is the most important factor sporulation and development. It has been estimated that losses in grain yield can be up to 80% in susceptible varieties and quality deterioration due to the attack grain is sucked, drastically reducing the test weight, a feature that is strongly related to production and flour quality.

Control of these diseases can be performed with chemicals, however, through the years, within the national program to improve rainfed wheat and irrigation INIFAP has been observed that the best method of control is the use of varieties resistant. The use of such strains have been used in the country; however, varieties lose their resistance by the constant evolution of new virulent pathogen to physiological races or the incidence of races that were not present. Therefore, the constant formation and release of new varieties is required to replace those that are losing strength and help counteract the genetic variability of the pathogen present in different regions of production of this cereal. Moreover, with the constant evaluation of greenhouse genotypes they have been identified sources of genetic resistance that integrate themselves within groups of parents in the breeding program.

Based on this background it is that the present work was proposed in order to identify genotypes with higher yield potential and better phytopathological response in rainfed environments in the high valleys of Central Mexico and assess the progress of improvement in different environments temporary.

Materials and methods

The study included two years of evaluation corresponded to the 2012 and 2013 cycles spring-summer where eight varieties released by INIFAP and two candidate lines were evaluated varieties. The genotypes were established in towns of Valles Altos, comprising the states of Mexico, Puebla, Hidalgo and Tlaxcala, with temperate climate subhumid mostly to a lesser extent temperate rain. With variations in altitude relative to mean sea level ranging from 2 250 m to 2 811 m. As for precipitation, it varied in the towns of 500 mm in the most restrictive parts up to 840 mm in the best environments for this factor.

The biological material in both crop cycles and evaluation consisted of commercial varieties of bread wheat Temporalera M87, Galvez M87, Romoga F96, Batan F96, Rebeca F2000, Tlaxcala F200, Nana F2007 and Altiplano F2007, released at different times by the program genetic improvement of rainfed wheat INIFAP and even lines Don Carlos "S" and Mona "S" candidates for release as new varieties. Such varieties are recommended for temporary crops in these areas in the Highlands of Central Mexico. The experimental design was randomized blocks with two replications; the size of the plot was four rows with spacing of 0.3 m and 3 m long, being the useful plot the total experimental plot of 4.5 m², using a seeding density of 110 kg ha^-1.

With the data obtained overall variance analysis was performed to detect differences in the variables evaluated in the different sources of variation, using the Statistical Analysis System 9.0 (^{SAS, 2002}) statistical package. Also, the categorization of localities assessment based on their potential grain yield obtained (favorable, intermediate and critics) was performed and proceeded to analysis by type of environment with the objective of evaluating the behavior of individual genotypes through thereof. Averages varieties per year or time of liberation and progress made through genetic improvement over time were compared. Additionally incidence readings were recorded rusts and foliar diseases complex to weigh the degree of resistance or tolerance of varieties and lines over time.

As variables were evaluated; days to spiking or flowering (DF), recorded when 50% of spikes emerged completely from the flag leaf; days to physiological maturity (DM), counting the days from sowing until 50% of spikes had yellowfin coloration at the base of the pin or golden color; plant height (ALT), measuring in centimeters from the base of it to the apex of the terminal spikelet; Grain yield (REND), once reached commercial maturity (13 to 14% moisture in the grain) proceeded to perform the calculation of the unit harvest and grain yield.

The modified Cobb scale (^{Roelfs et al., 1992}) was used to record the incidence of yellow rust (YR) and leaf rust (Lr), making the first reading at the stage of flag leaf and then taking readings every 10 15 days to physiological maturity scale from 0 to 9 Saari-Prescott was used (^{Eyal et al., 1987}) in the case of the incidence of foliar diseases, and scale double digits, where the first digit indicates the relative height reaching disease and the second indicates the severity of damage as a percentage. In each test site, once it reached commercial maturity of the grain we proceeded to make the harvest of the plots using a mini-thresher Wintersteiger small grain, threshing and separating the grain from each plot to be later taken samples cleanliness and heavy for analysis cabinet.

Data analysis was performed using the SAS program, where the variance analysis and comparison of overall mean (Tukey's test, with α= 0.05) was performed. According to the general analysis of the evaluated variables we proceeded to classified based on performance equally into three types called environments: favorable environments (4.2 to 6.2 t ha^-1), environments intermediates (3.4 to 4 t ha^-1) and critical environments (2.1 to 3.2 t ha^-1) using the criteria method proposed by ^{Villaseñor and Espitia (2000)}.

Results and discussion

The results of analysis of variance in general are presented in Table 1, which shows that in the case of the "locations", highly significant differences were presented for the four variables evaluated, indicating that the productive potential of each locality is different because the variables showed different degree of expression, which coincides with the expressed ^{Villaseñor and Espitia (2000)}, in the sense that in the areas of wheat (Triticum aestivum L.) temporary in Mexico, variability caused by differences between sites and years is large.

Table 1 Mean squares analysis of variance in general for the four variables evaluated in 10 wheat genotypes temporary, spring-summer, 2012 and 2013. 

gl= grados de libertad; DF= días a espigamiento; DM= días a madurez; ALT= altura; REND= rendimiento den grano; Loc= localidades; Var= variedades; C.V.= Coeficiente de variación en porcentaje; **= altamente significativo (α= 0.05); *= significativo; ns= no significativo.

In the case of the source of variation "varieties" highly significant differences for all evaluated variables (Table 1) they were also presented, indicating that genotypes were different from each other, a fact that can be attributed to genotypes have inherently different genetic heritage that makes them more or less productive and on the other hand the incidence of biotic and abiotic factors affecting these, i.e., the influence of the environment on them.

In the case of interaction "LocxVar" (Table 1), in this type of testing requires additional studies in order to clarify the selection of individuals with general adaptability and/or specific, as the in consistency of behavior among genotypes one environment to another, and when it occurs in large proportion, reduces the genetic progress of the selection (^{Yang and Baker, 1991}; ^{Magari and Kang, 1993}) together with the rainfed wheat is affected by biotic problems that detract from their performance; because the varieties released over time become susceptible to diseases, mainly yellow stripe rust (Puccinia striiformis f sp. tritici) and leaf rust (Puccinia triticina) because these pathogens present great variability of physiological races in Mexico (^{Singh, 1991}; ^{Huerta and Singh, 2000}) or because new races that break the resistance of varieties are presented.

In this source of variation, the differences for days to flowering and grain yield were highly significant revealing that the varieties showed different amplitude of the biological cycle when changing environments indicating that the same variety modify their life cycle when exposed to different performance environments and behaved in the same way, that is, the performance of a genotype was different when changing environment. While the differences for plant height and days to physiological maturity were only significant indicating that genotypes also showed differential to be changed from one locality to another answers.

These results agree with those reported by ^{Hortelano et al. (2013)} in the sense that production environments bread wheat are very contrasting temporal, that varieties behave differently to be changed environment and interactions can be exploited by detecting the adaptation that certain varieties have.

The above expressed undertakes to set tests under different conditions to estimate yield potential and phenotypic stability of varieties and provide a reliable guide to select the best genotypes for new locations or future years as mentioned ^{Crossa (1990)}.

The general behavior of genotypes are presented in Table 2, which shows that the outstanding lines Don Carlos "S" and Mona "S" obtained the highest yields of 4.8 and 4.7 t ha^-1 respectively, and in this case they had short cycles as physiological maturity was 127 and 123 days, respectively, in both lines. Here again it highlights the fact that in these lines short cycles and high yields are brought together, a situation that places these lines as good prospects to be released as new varieties for rainfed environments High Valleys. For varieties that are the most recent will be more productive as reported by ^{Villaseñor and Espitia (2000)}.

Table 2 Average overall variables by variety and incidence of diseases, evaluated in the spring-summer cycles, 2012 and 2013 in towns of Valles Altos. 

DF= días a espigamiento; MAD= días a madurez; ALT= altura de planta en cm; REND= rendimiento en t ha^-1; DMG= diferencia en (%) con respecto al mejor genotipo; YrE= roya amarilla en la espiga; Yr= roya amarilla en la hoja; Lr= roya de la hoja; r= resistente; mr= moderadamente resistente ms= moderadamente susceptible; s= susceptible; Fol= complejo de enfermedades foliares.

In the case of varieties Nana F2007 and Altiplano F2007 yields ranged from 4.5 to 4 t ha^-1, respectively, however their differences with the best line were the order of 239 kg (5%) and 737 kg (15.3), respectively (Table 2). Lower yields presented them the variety Galvez M87 and Batan F96 with 3 and 3.1 t ha^-1 respectively, reaching differences in grain yield with the best line was Don Carlos "S" of 1671 kg (35.8%) and 1722 kg (34.7%), respectively.

Other varieties also showed lower yields were Rebeca F2000 is a variety which is strongly affected by yellow rust, Romoga F96, Temporalera M87 and Tlaxcala F2000 in the same conditions as their genetic resistance to yellow rust was defeated a few years ago and whose differences from the best line ranged from 990 kg (20.6%) to 1453 kg (30.2%). In this regard, until 2008 Rebeca F2000 was virtually immune to yellow rust, even in the presence of the 219MEX0 race, which in the summer of 2004 in the high valleys of Mexico became completely susceptible varieties Galvez M87, Pavon F76, Temporalera M87, and Juchi F2000 ^{Rodriguez et al. (2009)}, as mentioned ^{Villasenor et al. (2012)} to yellow rust, Tlaxcala F2000 is moderately resistant to moderately susceptible and not even the strength of Altiplano F2007 and Nana F2007.

The yellow stripe rust in the tang (Puccinia striiformis f. sp. tritici) (YrE) impacted more severely than older varieties, corresponding to the genotypes: Galvez M87 and Temporalera M87, like Batan F96 which reached 70% infection. The other genotypes were maintained with 5 to 20% of infection in the ear, being Altiplano F2007 variety with better resistance to the incidence of yellow rust in the ear. With the planting of the variety Altiplano F 2007, which is more resistant to disease and with higher yield potential, you can increase performance and profitability of rainfed wheat planted in rainy rainy environments and a half, coupled with that in a very short time time new lines and as varieties are available and are adopted in an agile by wheat farmers temporarily Los Valles Altos.

The grain yield of genotypes released in different years is presented graphically in Figure 1, which shows how the trend line of regression in the variable grain yield per hectare have increased from varieties released in 1987 where they reached in this study 3.3 tons per hectare, until 2014 the experimental lines that exceed 4.7 t ha^-1, validity supported by the coefficient of determination of the regression is 0.85.

Figure 1 Progress in performance evaluated bread wheat varieties in rainfed environments in Valles Altos, spring-summer, 2012 and 2013. 

One of the important aspects to consider wheat producers, both rainfed and irrigated, is the grain yield obtained per unit area of the variety of choice and these yields the same variety must be more or less similar to through environments and for a certain period of time. This leads us to determine that at a certain time some varieties are more productive than others and that the passage thereof, the first and second obsoleted be mostly planted. Genetic resistance to rusts, through the generation of resistant varieties, is far more safe, economic and environmental control (^{Ma et al., 1997}), although often the protection is short-lived as populations of pathogenic fungi respond to selection pressures generated by producing new strains resistant varieties or biotypes maturing varieties resistance. Under this circumstance, it is that breeding for genetic resistance to rusts of wheat should be an ongoing activity (^{Schafer, 1987}), as illustrated in Figure 1, since progress in grain yield through time is attributed to both genetic potential of varieties as genetic resistance to rusts and diseases in general.

Genetic improvement observed based on grain yield, seen as progress in it, we can see in Figure 1, where the varieties released in the decades of the nineties eighties and two thousand are lower yielding comparatively regarding the lines of 2014, showing the trend in increasing yield of genotypes, since the progress of performance gain will increase from varieties 1987, where performance gains are lower in more than one ton per hectare regarding genotypes 2014, this is consistent with that reported by ^{Hernandez (1988)}, who mentions that progress in grain yield achieved by the program wheat Mexico, via genetic improvement has been linear and upward, trend it has most clearly observed in wheat yield obtained commercially in our country. While comparing 1987 with 2007 gains are slightly less than the latter because of a difference barely reached 990 kg, which are considered good performance.

Progress in increasing grain yield several causes among which genetic gain in yield potential are attributed; in resistance to factors such as disease, reduced lodging, tolerance to various environmental stresses; and improvement in production technology (^{Villaseñor and Espitia, 2000}). A particular case occurred with the varieties released in 1996 compared to 1987 as the first performance declined compared with the second (Figure 1). This behavior can be explained on the one hand that on average had a higher percentage of damage by yellow rust in the ear compared to those of 1987 since this pathogen to attack the pin direct impact on grain yield by modifying characteristics such as weight grain, grain filling etc.

When susceptible varieties are planted, in addition to direct losses in grain yield, disease decreases the test weight (grain filling) and flour yield, so that the harvested grain would feed use only (^{Rodriguez et al., 2009}).

The incidence of rusts and foliar diseases is presented in Figure 2, where it is clearly seen as different graphs show a downward trend to negative and the incidence of these pathogens.

Figure 2 Incidence of rusts and foliar diseases in wheat genotypes evaluated in time in Valles Altos, spring-summer, 2012 and 2013. 

Is observed as genotypes generated in 2014 are more resistant to yellow rust in the ear, yellow rust in the basin, leaf rust and leaf diseases complex (Figure 2a, b, c, and d). Also, it is clear that according to the trend lines there is further progress in genetic improvement in resistance to yellow rust on the blade (Figure 2b) and has not been so prominent progress in the generation of genotypes resistant complex foliar disease (Figure 2d).

Taking into account the genetic improvement of wheat before the attack of yellow rust, a short cycle favors when the crop is established temporarily, as mentioned ^{Carrasco (2009)} as “six hours of free water are needed on the leaf surface production of continuous reinfection” based on this, there is a high possibility that high this moisture during the growing season causing the chances of infection by yellow rust, which causes a decrease in performance, fewer grains per spike, less filling grain and decreased quality (^{Carrasco, 2009}.)

The behavior of genotypes across the three environments are shown in Figure 3, which shows that the trend line, according to the regression is positive slope, indicating that advances in grain yield were gradually and significant, since the varieties released in the decades of the nineties eighties and two thousand are lower yielding comparatively with respect to the lines candidates for new varieties (to 2014), Don Carlos "S" and Mona "S".

Figure 3 Trend in grain yield per type of environment in wheat genotypes evaluated in time in Valles Altos, spring-summer, 2012 and 2013. 

These results indicate that the experimental lines, candidates for new varieties for Los Valles Altos del Centro in Mexico, maintained their high performance in the three aforementioned types of environments, so it is confirmed once again that the strategy to try to genotypes in a wide range of contracted environments enables the choice of the most suitable genotypes for each type of environment where this crop is practiced or detect materials that respond well or maintain their good behavior across environments and years, as in this if all environments, a situation that would be ideal in the case of conditions of rainfed agriculture, where variations occur in the expression of the variables of interest depending on the year, locality and genotype, even within the same type of environment (^{Crossa, 1990}).

Conclusions

The outstanding lines, candidates to be released as new varieties, Don Carlos "S" and Mona "S" were the genotypes showed the highest yields in general and in the three types of environment as well as lower incidence of yellow rust and leaf, followed by the varieties Altiplano F2007 and Nana F2007, less yielding and more susceptible varieties were invariably Temporalera M87 and Galvez M87.

Advances in grain yield, achieved over time, have been ascending and steadily, being genotypes recent formation which had the highest returns, regardless of the environment and the year, assemblage in them, biological cycles short and high productivity.