Introduction

In the breeding program aimed at the production of hybrids, plant breeders have two important decisions to make, namely: 1) choose the sources of germplasm; and 2) deciding on breeding methods to be used. As for the sources of germplasm, the greater the understanding and classification into groups and heterotic patterns, the greater the chances of success by making a better approach in the use of sources in generating hybrids (^{Goldman, 1998}).

Scientific information on corn (Zea mays L.) in temperate zones is greater than that generated in tropical environments (^{Paliwal et al., 2001}). Tropical maize genotypes have limited ability to set a high number of grains per square meter, unlike temperate. This difference is partly explained by the effect of genetic improvement that has contributed to the adaptation of the latter to high population densities; which it is reflected ingreater efficiency in the formation of grainsper unit area (^{Andrade et al., 1996}). The appearance of hybrids that combine the favorable characteristics of both types of germplasm, have allowed a better adaptation to tropical and subtropical areas, as well as in temperate zones (^{Nelson and Goodman, 2008}).

Generally tropical sources show good combination with hardened materials Stiff Stalk Synthetic group compared to other temperate US sources (^{Ron-Parra and Hallauer, 1997}). Tropical germplasm sources in the US have been used as a source of resistance to diseases and insects, mainly (^{Goodman, 1999}). ^{Holley and Goodman (1988)}, reported that crosses tropical tempered materials have proved competitive compared with commercial mild hybrids in the United States; likewise ^{Goodman et al. (1990)} and ^{Uhr and Goodman (1995)} say they have had favorable results with tropical combinations.

One of the ways that favor broadening the genetic base of tropical maize is favorable incorporate characters from exotic germplasm; which the implementation of efficient methods to help assess the appropriate percentage of each region more convenient for exotic germplasm, to be achieved quickly, it is necessary (^{Darsana et al., 2004}). There are several alternatives for the incorporation of useful characters in suitable materials and methodologies to use depend heritability, gene action, number of genes involved, and heterosis genotype environment interaction. Thus, when it has valuable germplasm, making improvements incorporating desirable traits of exotic germplasm sources adapted is usually a recommended technique (^{Eberhart et al., 1995}; ^{Ron-Parra and Hallauer, 1997}; ^{Hallauer and Carena, 2014}).

The use of temperate maize germplasm in tropical breeding programs has been poorly documented, but in general has been growing. Currently many tropical hybrids contain some temperate germplasm. In the experience of breeders of tropical environments, the use of temperate germplasm in crosses with tropical has helped improve heterosis for grain yield. This may be due to favorable alleles, or genomic regions of temperate materials in several cases found in crossbreeding with tropical complementary sources. It has also been clear that the hardened materials in tropical regions have shown susceptibility to pests, diseases and poor quality grain; therefore, it is necessary to design breeding programs to reduce these unfavorable aspects through combinations with tropical sources (^{Wen et al., 2012}; ^{Ramirez et al., 2013}).

Groups and heterotic patterns are concepts related to hybrid formation and the phenomenon of heterosis, the first is used to refer to a group of individuals, related or not, who come from the same or different population but show similar behavior in fitness combinatorics and heterosis when crossed with other individual’s genetically different group (^{Melchinger and Gumber, 1998}). Knowledge of patterns heterotic groups and facilitates the selection of lines to make crosses in breeding programs; due to, hybrids are formed across two complementary lines heterotic groups, which in turn represent a heterotic pattern. Initially, groups and heterotic patterns were developed empirically by observing crossbreeding with better performance. In the crop corn generally they must defined heterotic patterns as Iowa Stiff Stalk Synthetic (SSS) by Non-SSS for the United States, Crystalline by Jagged for Northern Europe and South America, and Tuxpeños by Non-Tuxpeños for regions of the Tropics and subtropics (^{Bernardo, 2010}).

^{Ron-Parra and Hallauer (1997)} and ^{Reif et al. (2003)} indicate that to improve heterosis in grain yield in tropical maize, it can be achieved with the incorporation of exotic germplasm

in order to increase the genetic distance between sources of opposite heterotic groups. Normally the performance of stocks that are used as sources of germplasm is affected when incorporation of exotic germplasm is made; because the populations are adapted to their environment and not exotic germplasm; to reduce this disadvantage, plant breeders make backcrosses to recover the good performance of the original populations taking care not to lose the desirable traits of exotic germplasm employee (^{Cowling, 2013}).

Taking into account that in breeding programs corn focused on the generation of hybrids, the success achieved is largely determined by the sources of germplasm that are used as the basis for generating parental lines, it is necessary to implement strategies that will generate populations with desirable traits and generate information about groups and heterotic patterns to help in planning for crossbreeding hybrids with high grain yield potential, health and stability. Therefore, the objective of this study was to form a heterotic pattern based on the development of maize populations generated with two yellow temperate lines (Stiff Stalk and Non-Stiff Stalk) in crosses with two tropical white lines (Tuxpeño and Hick) looking for the best tropical-temperate combinations to keep heterotic groups under the conditions that create a better productive potential in the formation of hybrids.

Materials and methods

The outstanding two-hybrid corn (fall-winter cycle, Sinaloa 2009-2010), one formed by two warm yellow lines (Stiff Stalk (SS) and Non-Stiff Stalk (NSS) and the other two were identified by tropical white corn (Tuxpeño (TUX) x Hick (CAT); and based on the performance of hybrids, the four lines for the formation of maize populations were used. For the formation of the sources of germplasm cross temperate lines were made by tropical lines for the purpose of incorporating favorable characters from exotic germplasm (tempered) in adapted materials (tropical) as indicated by ^{Ron-Parra and Hallauer (1997)}.

In the first instance it was not known the best combination to make the crosses of the two lines of yellow corn with temperate tropical two white lines; so, we proceeded to make the four possible crosses (SS/TUX, SS/CAT, NSS/ TUX and NSS/CAT). These crosses were performed in the cycle Spring-Summer 2010 in Zapopan, Jalisco (20° 45’ 30’’north latitude - 103° 32’ 50’’ west longitude).

In the autumn-winter 2010-2011 in Coquimatlan, Colima (19° 13’ 13’’ north latitude - 103° 47’ 50’’ west longitude), with four single crosses two double hybrids were unrelated (SS/TUX//NSS/CAT and SS/CAT//NSS/TUX).

In the spring-summer 2011 cycle double hybrids were evaluated in yield trials to identify the best performers; which would indicate the best temperate-tropical to identify the pattern and heterotic groups, and based on this form combinations maize stocks would be used as sources of germplasm. The trials were established in five subtropical environments: Zapopan, Jalisco (20° 45’ 30’’ north latitude - 103° 32’ 50’’ west longitude), Leon, Guanajuato (21° 03’ 55’’ north latitude - 101° 43’23’’west longitude), Silao, Guanajuato (20° 55’ 48’’ north latitude - 101° 27’ 22’’ west longitude), Irapuato, Guanajuato (20° 39’ 21’’ north latitude - 101° 17’ 18’’ west longitude) y Villagran, Guanajuato (20° 30’ 53’’ north latitude - 100° 58’ 43’’ west longitude). In performance tests, along with two double hybrid two-hybrid commercial maize as witnesses (2A120 Hybrid Dow yellow and white hybrid Pioneer 30P16) were included, for use as reference and evaluate the performance of experimental crosses.

The experimental design was the randomized complete block design with three replications in each of the towns, leaving the useful plot consists of two rows of 4 meters long. The variables studied were: grain yield (kg ha^-1), grain moisture at harvest (%), root lodging and stalk, ear rot, stalk rot Fusarium and leaf spot caused by Cercospora (these last 4 variables were recorded with rating from 1 to 9, where 9 is best).

In the spring-summer 2012 in Zapopan, Jalisco (0° 45’ 30’’ north latitude - 103° 32’ 50’’ west longitude) F2 of single crosses that gave rise to two-hybrid best performance was obtained (SS/TUXF2 and NSS/CATF2) to form maize populations that would be used as sources of germplasm in hybrid formation, taking into account the pattern and heterotic groups identified.

Results and discussion

In the Tables 1 and 2 show the results of temperate and tropical white yellow hybrids are shown; respectively, they were identified as outstanding materials; and of which the parental lines for generating populations of maize for use as germplasm sources were used.

Table 1 Results of evaluation of yellow corn hybrids where the SS/NSS crossing selected lines for use in forming germplasm sources. Sinaloa (Los Mochis, Juan Jose Rios, Angostura and Culiacan). Autumn - Winter 2009-2010. 

La muerte prematura se registró con calificación de 1 a 9, donde 9 es lo mejor. S= Stiff Stalk; NSS= Non-Stiff Stalk; TUX= Tuxpeño; CAT= Cateto; CAR= Caripeño; MAI= Maia y SUW= Suwan

Table 2 Results of evaluation of white maize hybrids where the TUX / CAT crosses was selected for use in forming lines of germplasm sources. Sinaloa (Los Mochis, Guasave and Culiacan). Autumn-Winter 2009-2010. 

La muerte prematura se registró con calificación de 1 a 9, donde 9 es lo mejor. SS= Stiff Stalk; NSS= Non-Stiff Stalk; TUX= Tuxpeño; CAT= Cateto; SUW= Suwan y MAI= Maia.

The results of grain yield of double hybrids and two witnesses established in the five locations are shown in Table 3. The combined analysis of variance for grain yield showed significant differences between hybrids (p≤ 0.01) and between localities (p 0.01), and showed no significant differences for the interaction Hybrid Towns (p= 0.36). The results indicate that the productive potential of hybrids is significantly different, and according to the comparison of means (DMS /= 0.05) two double hybrids have different production potential, with SS/TUX//NSS/CAT best; which, compared to controls employees (2A120 and 30P16) no observed significant differences. These results demonstrate that in the hybrid formation, combinations of tropical germplasm mild hybrids can generate high grain yield potential. In this regard, ^{Oyervides et al. (1985)}, the evaluation of crosses between populations developed in Mexico and US populations, found good heterotic response tests established in Mexico for combinations Stiff Stalk Synthetic and Lancaster crossed by ETO and Tuxpeño.

Table 3 Grain yield of double hybrids and witnesses set in five environments subtropics in the 2011 spring-summer cycle. 

Las letras A, B y C son utilizadas para señalar grupos en la comparación de medias, donde lestras iguales indican pertenencia al mismo grupo (estadisticamente iguales). SS= Stiff Stalk; NSS= Non-Stiff Stalk; TUX= Tuxpeño y CAT= Cateto.

Similarly ^{Wen et al. (2012)} to make combinations with lines from the project called Improvement of Maize Germplasm (GEM acronym) with lines developed at CIMMYT (CMLs) found that the incorporation of the temperate lines (GEM) in crosses with the tropical (CMLs) improved heterosis for grain yield; which may be due to favorable alleles, or genomic regions of temperate materials in several cases found in crossbreeding with tropical complementary sources. In this regard, also ^{Ron-Parra and Hallauer (1997)} and ^{Reif et al. (2003)} indicate that to improve heterosis in grain yield in tropical maize, can be achieved with the incorporation of exotic germplasm in order to increase the genetic distance between sources of opposite heterotic groups; which it is reflected in the results obtained in this study.

The results for the agronomic and phytosanitary characters combined analysis are shown in Table 4. Analysis of variance showed significant differences among hybrids for the variables evaluated: grain moisture (p≤ 0.01), lodging (p≤ 0.01), ear rot (p 0.01), Fusarium stalk rot (p≤ 0.01) and Cercospora leaf spot caused by (p≤ 0.01). Based on the comparison of means (DMS α= 0.05) of the two double hybrids, the SS/TUX//NSS/CAT showed better productive potential, also noted for its earliness (13.8% moisture at harvest) and his best behavior for variables flattens root and stem, ear rot, Fusarium stalk rot and leaf spot caused by Cercospora. The good performance of this hybrid combinations shows that in temperate tropical germplasm, as well as increasing heterosis for grain yield, also be achieved maintain good adaptation; which it is reflected in the results of agronomic and phytosanitary variables. With respect to the variable grain yield and earliness, it is confirmed that found by ^{Ramirez et al. (2013)} to evaluate the central-western region of Mexico you cross diallel formed between populations of tropical origin and tempered in the sense that the temperate germplasm can make important contributions to improve heterosis for grain yield and shorter cycle.

Table 4 Combined agronomic and phytosanitary characters of double hybrids and witnesses set in five environments subtropics in the spring-summer cycle of 2011 (Zapopan, Jalisco, Leon, Silao, Irapuato and Villagran, Guanajuato). 

Las variables acame de raíz y tallo, mazorcas podridas, pudrición de tallo por Fusarium y mancha de la hoja causada por Cercospora fueron registradas con calificación de 1 a 9, donde 9 es lo mejor. Las letras A, B y C son utilizadas para señalar grupos en la comparación de medias, donde lestras iguales indican pertenencia al mismo grupo (estadísticamente iguales). SS= Stiff Stalk; NSS= Non-Stiff Stalk; TUX= Tuxpeño y CAT= Cateto.

The good performance shown by two double hybrids with respect to witnesses confirms that there is generally good response to make crosses tropical germplasm tempered, as point out ^{Goodman et al. (1990)}, ^{Uhr and Goodman (1995)}, ^{Ron-Parra and Hallauer (1997)}, ^{De Leon et al. (2005)}, ^{Nelson and Goodman (2008)}, ^{Wen et al. (2012)}, ^{Ramirez et al. (2013)}, ^{Esquivel et al. (2013)} and ^{Hallauer and Carena (2014)}. In the case of this study, seeking the best combination in the crossing of maize lines selected for the formation of sources of germplasm, the good performance of two hybrid SS/TUX//NSS/CAT shown in the results indicates that the pattern heterotic formed is integrated by a heterotic group (a) made by combining lines Stiff Stalk x Tuxpeño and its counterpart (group B) by the combination of the Non-Stiff Stalk and Hick lines.

Conclusions

The analysis of the results of grain yield, agronomic and phytosanitary characters of the two double hybrids indicates that the incorporation of temperate germplasm in tropical can increase heterosis without losing adaptation, and the formation of maize populations for use as sources germplasm is important to seek the best combination of original materials to keep the heterotic groups under the conditions that will generate a better productive potential in the formation of hybrids.

Based on the hybrid double SS/TUX//NSS/CAT showed the best performance a heterotic pattern formed by an "A" group that integrates with the combination of lines Tuxpeño and Stiff Stalk and the opposite heterotic group defined "B" consists of the crossing of the line Non-Stiff Stalk and the line Hick.