Introduction

In order to obtain quality seeds of hybrid maize, is required to detassel in time and in an appropriate manner to ensure that the male parent is combined with the female parent and prevent the latter from being contaminated with their own pollen. In thrashing activity are used 24 to 50 days per hectare, to obtain high quality seed with good genetic identity. In addition to the high economic cost, it represents thrashing risks losing quality made incorrectly. In response, the use of male sterility may be a viable option that favours obtaining adequate seed genetic quality and reduce production costs (^{Ramírez, 2006}; ^{Tadeo et al., 2007}; ^{Tadeo et al., 2014 b}).

Since 1970, in EE. UU the cytoplasmic sterility called cms-T stopped being used by the seed companies due to problems with susceptibility to leaf blight caused by the fungus Helminthosporium maydis race T. The disease caused an epidemic that affected 90% of the crops in the maize producing area of the country (^{Grogan et al., 1971}; ^{Airy et al., 1978}; ^{Tadeo et al., 1997}).

Since 1980 the investigation was resumed with male infertility, based on the discovery and development of new sources of male sterility. This new strategy, combined form cytoplasmic gene, mainly for the types C and S. Currently these types are used in about 25% of the area devoted to seed production in the US (^{Liu et al., 2002}; ^{Beck and Torres, 2005}). In all cases it has various sources within each type of sterility for not relying on only one (^{Partas, 1997}; ^{Tadeo et al., 1997}; ^{Thomson, 1979}; ^{Liu et al., 2002}; ^{Torres and Rodríguez, 2002}; ^{Tadeo et al., 2007}).

In the Faculty of Higher Studies Cuautitlan of the National Autonomous University of Mexico (FESC-UNAM) and the Valley Experimental Field of Mexico of the National Research Institute of Forestry, Agriculture and Livestock (INIFAP) have made studies for breeding male infertility include females of commercial and experimental hybrids developed by UNAM and INIFAP, with emphasis on the "C" source of hereditary interaction of cytoplasm and nuclear (^{Tadeo et al., 2007}; ^{Espinosa et al., 2009}; ^{Tadeo et al., 2010}). Since 1992, the character of sterility was added to the basic lines later (1995-1997) lines with restorative capacity were identified. The stability of male sterility source and the restorative capacity of developed lines, allowed a glimpse of the potential use of male sterility of the scheme, which could be limited to some extent, some problems related on thrashing and reduced costs (^{Tadeo et al., 2007}; ^{Espinosa et al., 2009}; ^{Tadeo et al., 2010}; ^{Tadeo et al., 2014 a}).

Product of work in male sterility, in 2011, INIFAP achieved registration with the National Catalogue of Plant Varieties (CNVV) of the maize hybrid H-51 AE. The FESC-UNAM won got the final register for 2014, before the CNVV, of the hybrid Tsiri Puma. Other varieties are in the last stage of research, validation and transfer before being released commercially (^{Espinosa et al., 2008}; ^{Espinosa et al., 2009}; ^{Espinosa et al., 2010}).

In this work we evaluated, in four uniform experiments, two in the CEVAMEX and other two in the Faculty of Higher Studies Cuautitlán (FESC-UNAM), with difference in both places on the date of planting, the productivity of the trilinear hybrids: Tsiri Puma 2, H-57 AE, H-47 AE, Puma 1183 AEC1, Puma 1183 AEC2. Each of the hybrid male-sterile and fertile versions. The aim of this study was to determine the productive capacity of the hybrids, considering the male sterile and fertile versions as well as the average of the four environments evaluated. The hypothesis that the introduction of the character of male sterility to five trilinear hybrids of the Plateau of Mexico, does not alter its productive potential, nor its phenology and morphology, not even the interactions with the environment under assessment.

Materials and methods

This research was conducted in four experiments established in the 2012 spring-summer cycle, two of them on the campus of St. Lucia de Prias in the Valley Experimental Field of Mexico of the National Research Institute of Forestry, Agriculture and Livestock (CEVAMEX-INIFAP), located in the municipality of Texcoco, State of Mexico, and two others at the experiment Experimental "RanchoAlmaraz" of the Faculty of Studies Cuautitlan (FESC), UNAM, located in Cuautitlan Izcalli, State of Mexico.

The four experiments were planted using the experimental design random complete blocks, with six repetitions, the analysis was performed in combination, calculating the interactions, to test the effect of environment the effects of repetitions x environment interaction were collected and combined with the effect of repetitions to integrate the error that was used for comparison of environmental effect hypothesis. The factors and levels comprise a full factorial 2 x 5 x 2, which factors are: two areas (A= 4), five trilinear hybrids (B= 5), the male sterile or fertile version (C= 2), which generates 10 treatments in each of the four environments. Each experimental unit consisted of a row of 5 meters long and 0.8 meters wide.

The five trilinear hybrids of maize of INIFAP assessed, identified in this study were Tsiri Puma 2, H-57 AE, H-47 AE, Puma 1183 AEC1, Puma 1183 AEC2, whose seed was obtained using single crosses in male sterile version (AE) and fertile (F), combined with its male parent during the spring - summer season, 2011.

The structure consisted of treatments, as noted, for evaluation in the four environments (factor A), the five mentioned trilinear hybrids (factor B) in their fertile and male sterile versions (factor C).

Mechanical preparation consisted of a fallow field, a cross and a step of drag. The ridged was made at 0.80 m. Seeding in the experiment CEVAMEX-INIFAP F1 was made on May 16^th, 2012 and the planting environment CEVAMEX-INIFAP F2, held on June 7^th, 2012, in both cases using a shovel and watering planting. The FESC-UNAM F1, the first experiment was planted on May 17^th, 2012, with the assistance of irrigation and, the second experiment, FESC-UNAM F2, sowing was June 1^st, 2012, with rainfed humidity rain, in both experiments at the FESC-UNAM, planting was at "cap standing"

In all cases were deposited four seeds per plant every 0.50 m. After the seedling emergence, was thinning to leave 24 plants in five meters long, to achieve densities of uniform population at 60 000 plants ha^-1, which is the population density recommended in the high valleys of Central Mexico, to materials similar to those that were used in this study.

For weed control, two applications were made: first, a day after sowing, using per ha 1 L of Hierbamina(r) and 2 kg of Gesaprim(r) gauge 90; the second was made 40 days after planting, with 1 L of Sansón(r) 4 SC, 1 L of Hierbamina(r) and 2 kg of Gesaprim^(r) gauge 90. The harvest was performed manually and in the first half of December 2010, collecting all ears if only considered those containing healthy grain that covers the features for marketing, i.e. healthy grain in over 60% of the cob.

Field data were recorded: male flowering (when 50% of plants released pollen), silking (when 50% of plants were exposed stigmas at least three cm), plant height (from the base of the stem knot insertion of the spigot) and ear height (from the stem base to the node insertion on the upper ear). Data from other variables are taken: in five cobs the seed had moisture and was recorded with an electric moisture determiner Stenlite type, percentage of grain/cob (ratio of grain weight with the weight of more cobs grain); ear length, rows per ear and kernels per row were also measured.

For grain yield (RG) per plot, applying the following formula:

RG= (PC x (%) MS x (%) G} x F.C.)/8600

Where: PC is the weight of the total area, harvested ears in the lot, expressed in kg (%) MS is the percentage of dry matter calculated based on the sample of five freshly harvested grain cobs (%) G is the grain percentage obtained by dividing the weight of grain and ear weight, FC is the conversion factor to yield per hectare, which is obtained by dividing 10 000 m² between the size of the useful plot given in m² (4 m²⁾, 8 600 is a constant used to estimate yield with a grain moisture of 14%.

For statistical analysis, data recorded in each of the agronomic variables were subjected to analysis of variance, which was made considering three factors: environment (4), hybrid (5), version genotype androsterile or fertile (2), and the interactions among these factors; i.e. environments x hybrid, environments x andoesterilidad/fertility, hybrid x male sterility/fertility, environments x hybrid x male sterility/fertility. When the F values were significant, Tukey test was applied to compare the means of locations, maize genotypes version of male sterile or fertile genotype, at a significance level of 5%.

Results and discussion

For grain yield were found highly significant differences (p≥ 0.01) for hybrids and environments; as well as the interaction environment x hybrids and significant for the hybrids x male sterility/fertility, but not for the male sterility/fertility factor nor for the other interactions. The coefficient of variation for yield was 19.4% and the average was 7 460 kg ha^-1. For male flowering and plant height there were highly significant differences (p≥ 0.01) among environments (A), hybrids (B) and the male genotype sterile version or fertile (C), in the case of plant height there were highly significant difference for the environment x hybrids interaction.

In weight of 200 grains, length of the ear cobs and number of grains per row, there were highly significant differences for the factor of environmental variation. For the variable weight of 200 grains and grains per row, the difference were observed at a 5% level, for the environment x hybrids interaction (A x B) (Table 1).

Table 1 Mean squares and statistical significance of the values of F for grain yield (kg ha^-1) and other variables evaluated in four environments (A), five maize hybrids (B) on its fertile and male sterile versions (C) of the High Valleys. Spring-summer 2012. FESC-UNAM and CEVAMEX. 

*, ** valores significativos y altamente significativos al 0.05 y 0.01 de probabilidad; CV= coeficiente de variación. RG, rendimiento de grano; FM, floración masculina; FF, floración femenina; AP, altura de planta; AM, altura de mazorca; LM, longitud de mazorca; HM, hileras de la mazorca; GH, granos por hilera; GM, granos por mazorca.

200 grain weight, ear length and grains per row, there were highly significant differences for the factor of variation environments. For the variable weight of 200 grains and kernels per row, differences were observed at the 5% to the environments interaction hybrid x (A x B) (Table 1).

The comparison between hybrids showed differences in yields, three groups of significance was defined, the best yield corresponded to Tsiri Puma 2 (8 930 kg ha^-1) and the lowest was the hybrid Puma 1183 AEC1, with 6 321 kg ha^-1 (Table 2). The five hybrids evaluated, obtained with the scheme male sterility and commercially used mixtures scheme and fertile male sterile seed, which is how the male sterility is used in other materials (^{Espinosa et al., 2009}; ^{Tadeo et al., 2010}). The similar yield of the hybrids H-57 AE, H-47AE, and Puma 1183AEC2 and Puma 1183AEC2, may be because of the two materials there are inconsistencies in one or two lines that make up its structure as hybrids, these materials that have been evaluated for several years and in different jobs have excelled, which allowed for 2014 signed one of the hybrids registered by the FESC-UNAM in the National Catalogue of Plant Varieties (CNVV), which means the first male sterile hybrid commercially released by UNAM (^{Espinosa et al., 2008}; ^{Espinosa et al., 2010}; ^{Tadeo et al., 2014}; ^{Tadeo et al., 2014 b}.)

Table 2 Comparison of average on five maize hybrids on the fertile and male sterile versions, considering the average of four environments, for the yield (kg ha^-1) and other variables evaluated. Spring-summer 2012. FESC-UNAM and CEVAMEX. 

*Las medias con la misma letra dentro de cada columna son iguales estadísticamente (Tukey, p≥ 0.05). RG= rendimiento de grano; FM= floración masculina; FF=floración femenina; AP= altura de planta; AM= altura de mazorca; LM= longitud de mazorca; HM= hileras por mazorca; GH= granos por hilera; GM= granos por mazorca.

The comparison of means for grain yield showed the male sterile production version presented statistically similar to the fertile (Table 3). This result can be explained as the male-sterile versions are isogenic fertile; they differ only in the production or not of pollen grains, as noted in several studies (^{Martínez et al., 2005}; ^{Ramírez, 2006}; ^{Tadeo et al., 2007}; ^{Tadeo et al., 2014}; ^{Tadeo et al., 2014 b}). Nevertheless, these results contrast with those observed in other studies that have found that male-sterile versions reach statistically higher than the yields of fertile (^{Martínez et al., 2005}; ^{Tadeo et al., 2007}). The comparison of means for male flowering, defined as delayed the male sterile with respect to the version fertile in plant height and number of rows per ear, version fertile male-sterile was higher version, while for the other variables: silking, ear height, ear length, kernels per row, grains per ear, to compare means with no significant differences between the male sterile and fertile versions, a situation attributable to the fact, as previously stated that are genetically similar, except for those related to male sterility cytoplasmic C-type gene, which cause no production of pollen in the male-sterile version, while the genes in the fertile did (^{Tadeo et al., 2007}; ^{Tadeo et al., 2010}).

Cuadro 3 Comparación de medias para las versiones androestéril ó fértil considerando la media de cinco híbridos y cuatro ambientes de evaluación, para rendimiento (kg ha^-1) y otras variables evaluadas. Primavera-verano 2012. FESC-UNAM y CEVAMEX. 

*Las medias con la misma letra dentro de cada columna son iguales estadísticamente (Tukey, p≥0.05). RG= rendimiento de grano; FM= floración masculina; FF= floración femenina; AP= altura de planta; AM= altura de mazorca; LM= longitud de mazorca; HM= hileras por mazorca; GH= granos por hilera; GM= granos por mazorca.

In the comparison of means between environments, the second planting date on the FESC-UNAM F2, exhibited the highest average grain yield (9 892 kg ha^-1) significantly different to the average yield of the other three environments, possibly because the other planting date on the FESC-UNAM and the two planting dates in CEVAMEX were made by the precipitation conditions, which probably affected the yield (Table 4).

Table 4 Comparison of the average of the four evaluation environments, considering the average of five hybrids and male sterile and fertile versions to yield (kg ha^-1) and other variables evaluated. Spring-summer 2012. FESC-UNAM and CEVAMEX. 

*Las medias con la misma letra dentro de cada columna son iguales estadísticamente (Tukey, p≥0.05). RG= rendimiento de grano; FM= floración masculina; FF= floración femenina; AP= altura de planta; AM= altura de mazorca; LM= longitud de mazorca; HM= hileras por mazorca; GH= granos por hilera; GM= granos por mazorca.

Conclusions

The FESC-UNAM 2 environment, with an average grain yield of 9 892 kg ha^-1, and was statistically superior compared to the other three evaluation environments.

The average yield of the five hybrids, considering the four evaluation environments and male sterile and fertile version, the highest production corresponded to Tsiri Puma, with 8 930 kg ha^-1, four hybrids were statistically similar (H-57 AE, H-47 AE, Puma 1183 AEC2).

The male-sterile version, considering the average of the hybrids and the four environments, presented statistically similar yield (7 627 kg ha-¹⁾ with respect to the fertile version (7 294 kg ha^-1).

Highly significant statistical difference was found in the environment x hybrid interaction, indicating the desirability of evaluating genotypes in more environments.