Results and discussion

In the experiments to determine grain productivity, the results of the variables analyzed in the evaluation of the 10 hybrids, across three environments, showed for the variation factor, environments, highly significant (p< 0. 01) differences, except in ear height. For the variable grain yield, the coefficient of variation was 9.2% and the overall mean was 6 742 kg ha^-1 (Table 2). For the variation factor, genotypes (hybrids), highly significant differences in the variables yield and plant height were found. In the interaction (G x E), significant differences were identified only for the variable, plant height (Table 2).

In the second experiment, to determine forage productivity and quality, the analysis of variance of the evaluation of the 10 hybrids, across three environments, detected for the variation factor, environments, significant (p< 0.05) differences for the variable green matter yield, as well as highly significant (p< 0.01) differences for the variables, dry matter yield, dry matter percentage, ear percentage, ear weight, protein percentage and digestibility percentage (Table 3). For the genotype variation factor, highly significant (p< 0.01) differences were detected in the variables green matter yield and dry matter yield. For the variables ear percentage and ear weight, significant (p< 0.05) differences between the genotypes evaluated were defined, while for the variables % dry matter, % protein and % digestibility, there were no significant differences (Table 3).

For the variation factor of the interaction (environment x genotype), highly significant (p< 0.01) differences were found for the variables: ear percentage and ear weight. In the rest of the variables, no statistical significance was detected (Table 3).

For the first experiment, that is, to determine grain productivity, in the test of comparison of means in the three environments evaluated (Table 4), in the variable of grain yield, two groups of significance were defined, placing Santa Lucía 1, first sowing date, with the highest grain yield (7 282 kg ha^-1), 11.2% higher than the yields obtained in FES Cuautitlán, UNAM and Santa Lucía, second date (6 467 kg ha^-1). These results are like those reported by ^{Espinosa-Calderón et al. (2018)}, for the locality of Santa Lucía (CEVAMEX).

These results were probably a consequence of the 10 days of difference in flowering that occurred, compared to FES Cuautitlán, UNAM, which possibly caused a greater use of the wet period in the locality of Santa Lucía, possibly the 18 days of difference in sowing between the first date and second date of the locality of Santa Lucía is also the consequence of the difference in yield between the two environments; likewise, in Santa Lucía, supplemental irrigations were applied compared to FES Cuautitlán, UNAM, where the water requirements of the crop were covered with rainfall.

In the variable plant height, the Santa Lucía 1 environment was superior and significantly different from the other two environments, which probably influenced grain and forage yield (^{Zaragoza-Esparza et al., 2019}). For the variable percentage of grain, three groups of significance were defined; that is, significant differences were obtained in the three localities, on the other hand, in the variables volumetric weight and weight of 200 grains, two groups of significance were defined, where the highest values were obtained in FES Cuautitlán, UNAM, compared to the environments of Santa Lucía. This could be a consequence of an environment that favored the growth and development of the crop and therefore, the formation of the ear (Table 4).

For the second experiment; that is, to determine forage productivity and quality, in the test of comparison of means in the three environments evaluated (Table 5), in the variable for green matter yield, two groups of significance were defined, placing Santa Lucía 1 (first sowing date), with the highest yield of green forage (64 392 kg ha^-1 ), which was 9.5% higher compared to the yield of Santa Lucía 2 (second date) and this in turn 14% more compared to FES Cuautitlán, UNAM (55 383 kg ha^-1). These results are higher than those reported by ^{Elizondo and Boschino (2001)}; ^{Hernández-Sánchez et al. (2011)}.

It can be considered that the percentage of ears was the variable that directly affected the yield of green matter, this is because when the population density increases (from 65 000 to 80 000 plants ha^-1), the interception of photosynthetically active radiation by the canopy increases, thus promoting the accumulation of dry matter at the population level, especially for early varieties (^{Li et al., 2020}), in this sense, the first sowing date of the genotypes, with a difference of 28 days with respect to the second date in the first locality and 10 days with respect to the second, had a direct impact on the forage yield.

For the comparison of means in the variable dry matter yield, two groups of significance were defined, the highest yield occurred (17 458.3 kg ha^-1) in Santa Lucía 2 (second date), this may be due to the lower amount of humidity, due to the low rainfall compared to FES Cuautitlán, UNAM and the late sowing compared to the first date. The dry matter yield is related to the percentage of dry matter, so the results in these variables are similar (Table 5).

For the variable percentage of protein, two groups of significance were defined, placing in the highest value the environment of Santa Lucía 2 (second date), which had 9.82% protein. The other two environments showed no significant differences (Table 5). Another important variable for forage quality is the digestibility percentage, where no significant differences were found between FES Cuautitlán, UNAM (69.56%) and Santa Lucía second date (70.94%).

In the comparison of means for the 10 genotypes evaluated, in the first type of experiments, (determination of grain productivity), considering the mean of the three evaluation environments, in the variable grain yield, two groups of significance were defined, placing the hybrid Atziri Puma with the highest grain yield, at 12% moisture (8 044 kg ha^-1), statistically superior to hybrid H 51 AE and H 47 AE. These results are like those reported in other research papers (^{Tadeo et al., 2016}; ^{Tadeo et al., 2018a}; ^{Tadeo et al., 2018b}; ^{Tadeo-Robledo et al., 2020}).

The results obtained are within the range of grain yield (3.89 to 11.71 kg ha^-1) reported by ^{Velasco-García et al. (2019)}, for High Valleys of the Mexican Tropics. At CEVAMEX, the hybrid Atziri Puma exhibits yields above what was obtained in the studies of (^{Tadeo et al., 2018a}; ^{Tadeo et al., 2018b}; ^{Tadeo-Robledo et al., 2020}), which showed yield ranging from 6 876 to 7 528 kg ha^-1.

The second hybrid with the highest grain yield was Tsíri Puma with 7 783 kg ha^-1, which is important, since this hybrid was recently released for commercial use, is registered in the National Catalogue of Plant Varieties (CNVV), has breeder’s title (^{Tadeo-Robledo et al., 2016}). The other genotypes exhibited statistically similar yields. The hybrids Tsíri Puma, H49AE and H51AE presented plants with greater height. In the variables male flowering, female flowering and ear height, there were no significant differences (Table 6).

In the comparison of means for the variables, dry matter percentage, grain percentage and volumetric weight, there are no significant differences between the genotypes evaluated. For the weight of 200 grains, differences between the hybrid H 51 AE were observed, which was higher than H 62 (Table 6).

For the second type of experiment, forage productivity and quality, in the test of comparison of means between the 10 hybrids for the three environments evaluated, in the variable green matter yield. Four groups of significance were detected, with the hybrid Tlaoli Puma being placed with the highest yield of green forage (70 202 kg ha^-1), statistically higher than H 51 AE and H 47 AE.

For its part, the hybrid H 47AE presented the lowest yield compared to the other hybrids, being lower by 46.4% than the hybrid Tlaoli Puma, the above is important since Tlaoli Puma was registered in the National Catalog of Plant Varieties (CNVV) and has a breeder’s title in the name of UNAM (^{Tadeo-Robledo et al., 2018}; ^{Tadeo-Robledo et al., 2020}). In second place was the hybrid Tsíri Puma with 68 750 kg ha^-1. The hybrid Tlaoli Puma had an average yield of dry matter of 19.6 tonne ha^-1, being higher (p< 0.05) than the hybrids H 61, H 51AE and H 47AE by 34.1%, 37% and 59%, respectively. In ear weight, the hybrid H 53AE was superior (p< 0.05) to the hybrids H 61 and H 47AE (Table 7).

The three best materials in dry matter yield were Tlaoli Puma, Tsíri Puma and H 49 AE, with 19 615 kg ha^-1, 17 567 kg ha^-1 and 17 474 kg ha^-1 respectively, all three are in the same group of significance and surpass the hybrid H 50 (Table 7). This hybrid exhibited a dry matter yield of 15 278 kg ha^-1, indicating that the materials indicated with better yields could be disseminated for extensive use, which could well replace H 50, which has been the most sown corn in the High Valleys of Mexico for 22 years (^{Espinosa-Calderón et al., 2019}).

In the comparison of means of the 10 genotypes for the variables of protein percentage and digestibility percentage, which represent forage quality, they did not show significant differences, which is a good indicator of the forage quality that each of them has, since all were within acceptable values for these variables (Table 7). Under arid conditions such as those of the Lagunera Region, Coahuila, using the same population density that was used in this research (80 000 plants ha^-1) and increasing irrigation as evapotranspiration increased (from 60 to 100%), yields from 47 665 kg ha^-1 to 57 665 kg ha^-1 were obtained (^{Yescas et al., 2015}), which shows that the amount of water supplied to the crop causes a sufficiency or lack of moisture to the soil, which is directly reflected in the yield of the crop and in its different variables such as green forage and dry matter yield.

The mean obtained for protein percentage was 9.12%, a figure higher than that reported by ^{Núñez-Hernández et al. (2015)}, who also obtained an average percentage of digestibility of 8.6%. ^{Yescas et al. (2015)} reported that in case of an increase in the irrigation sheet and population density (from 80 000 to 16 0000 plants ha^-1), the percentage of protein does not increase significantly, finding values of 8.51 to 8.6% under a density of 80 000 plants ha^-1, a similar case with % expressed by the hybrids evaluated in this research (from 8.7 to 9.7%); however, it should be noted that in that research only one hybrid was evaluated. The differences found here can be attributed to the genetic constitution of each hybrid evaluated (Arellano-Vázquez et al., 2017), remembering that these are single crosses, trilinear crosses and double crosses.

In the comparison of means between genotypes, no significant (p< 0.05) differences are found. An average of 68.2% in vitro digestibility was obtained for the hybrids evaluated. For the comparison of means between the hybrids, there are no significant (p< 0.05) differences, but the single-cross hybrid (ENS4@QRL-#) xMIA47 had the highest value of this characteristic.

These results are like those obtained by ^{Núñez-Hernández et al. (2015)}, who report that in vitro digestibility values range from 62.6 to 67.8% in intermediate cycle hybrids and from 67.2 to 73.2% in early hybrids.