Introduction

The annual consumption of popcorn (Zea mays L.) in Mexico is estimated at 80 000 t, of which only 596 t are produced in the country (^{SIAP, 2019}) and the rest is imported. This situation makes Mexico dependent on imports of popcorn, so the generation of improved national materials of this type of corn is urgently required. The lack of ‘popping’ varieties and technological lag have made it difficult to structure a national popcorn production program that would satisfy domestic consumption (^{Valadez-Gutiérrez et al., 2014}).

Almost all the popcorn consumed in the country is imported, mostly from the United States of America. Nonetheless, in Mexico, there are breeds of corn that produce ‘popping’ grains, which, in theory, are suitable for producing ‘popped corn’, but with a low capacity for expansion, so the market and consumers do not accept them. Such corn could be used in breeding programs to obtain popcorn varieties and hybrids in Mexico.

However, in this regard, there are precedents that the first hybrid of popcorn (H 367 P) was released in Mexico in 1977 (^{Miranda, 1977}; ^{Ángeles, 2000}) and the popcorn variety V 460 P was released in 2012 (^{Valadez et al., 2014}), which is an incentive. The center of origin of popcorn is Mesoamerica and this type of corn is considered the most primitive (^{Mangelsdorf and Smith, 1949}); numerous evidences show that its degree of rusticity and ability to pop are related to teocintle (^{Piperno and Pearsall, 1993}), but its relationship with other groups of corn is still under discussion (^{Ziegler, 2001}).

This origin gives popcorn a high adaptation to the ecological conditions of Mexico and a great diversity that can be used for genetic improvement. Among the methods to study the qualities of a set of popcorn parents are the diallel designs proposed by ^{Griffing (1956}), which allow estimating their genetic parameters, such as their general combining ability (GCA) and specific combining ability (SCA); in addition, it is possible to define the most appropriate method of genetic breeding to predict superior crosses and combine the best characteristics of agricultural importance (^{Melani and Carena, 2005}; ^{Cai et al., 2012}).

SCA is more important than GCA in a breeding program aimed at obtaining hybrids (^{Hoegemeyer and Hallauer, 1976}) as better use of dominance and epistasis can be made with SCA. Given that Mexico needs to generate outstanding popcorn materials to reduce its dependence on foreign grain and in view of the strategic interest in offering improved varieties using the available national native germplasm, for more than 20 years, the Cuautitlán Faculty of Higher Studies (FESC, for its acronym in Spanish) of the National Autonomous University of Mexico (UNAM, for its acronym in Spanish) and the Valle de México Experimental Field (CEVAMEX, for its acronym in Spanish) of the National Institute of Forestry, Agriculture, and Livestock Research (INIFAP, for its acronym in Spanish) have work on this matter.

This work consisted of the development of popcorn varieties to offer competitive alternatives to producers; in the genetic improvement process applied during all this time, emphasis was placed on the yield criterion and the popping trait was somewhat left aside. As a result of these works, there are experimental varieties and elite populations, in which the technology for their production is generated (^{Espinosa et al., 2018}).

Since 1997, quality protein sources (QPM) have been combined with native popcorn varieties and lines brought from Tamaulipas and abroad with the intention of adding expansion quality to the popcorn, then several backcrossing cycles were carried out towards popcorn quality (^{Espinosa et al., 2018}). This research aimed to determine the genetic components of GCA, SCA, and reciprocal and maternal effects involved in the yield expression of six varieties of popcorn adapted to the High Valleys of Mexico and their respective crosses.

Materials and methods

The study was carried out in six environments located in the State of Mexico. During the spring-summer cycles of 2018, 2019 and 2020, it was sown on FESC-UNAM lands located in Cuautitlán Izcalli (19° 41’ north latitude, 99° 11’ west longitude, 2 274 m altitude), which have a clayey loam soil; in 2018, it was also evaluated on lands of the Huexotla ejido, municipality of Texcoco (19° 27’ north latitude, 98° 51’ west longitude, 2 326 m altitude), which have a silty loam soil; in 2019 and 2020, it is found in Santa Lucía de Prías, Coatlinchán Texcoco (19° 29’ north latitude, 98° 52’ west longitude, 2 300 m altitude) with sandy loam soil.

The genotypes evaluated included six populations of popcorn: PB1, PB2, PB3, PB4, PB5, and PB6, which were integrated since 1997, for which native popcorn varieties, lines from Tamaulipas, and lines brought from the United States of America were combined with the intention of adding quality to the popped corn, then several backcrossing cycles were carried out towards popcorn quality. Four controls and 15 direct and 15 reciprocal crosses corresponding to a complete diallel were also included [^{Griffing (1956}) Method I] (Table 1).

In each locality, an experiment was established where the genotypes were distributed in a randomized complete block design with three replications at a density of 65 000 plants ha^-1, and the experimental plot consisted of a furrow 5 m long by 0.8 m wide. The sowing was carried out in June of the three years in the three localities. In 2018, in Huexotla, there was moisture in the soil at the time of sowing and three supplemental irrigations were applied, in 2019 and 2020, in Santa Lucía, sowing irrigation was applied and then two supplemental irrigations; in the FESC-UNAM, in the three years, only one irrigation was applied to the sowing and the subsequent moisture was met with rainfall.

In the growing period to the harvest date, information was obtained on 18 quantitative traits related to morphological characteristics and yield composition. The harvest was manual between November and December. In each plot, all the ears of corn were harvested, and their field weight (FW) was recorded.

The statistical genetic analysis was performed with the SAS^® v 9.0 program (SAS, 1996) for the yield variable with ^{Griffing’s (1956}) Model I (fixed-effect), Method I (complete diallel), which examines parental lines and direct and reciprocal crosses, through the Diallel-Sas program proposed by ^{Zhang and Kang (2003}), which allowed the split of reciprocal effects (RecE) into maternal (MatE) and non-maternal effects (NMatE). The relative importance of GCA and SCA was assessed with the formula proposed by ^{Baker (1978}): [2xMS_GCA]/ [2xMS_GCA + MS_SCA] where MS_GCA is the mean square of the GCA and MS_SCA is the mean square of the SCA.

Results and discussion

Conclusions

In the popcorn corn materials evaluated, the effects of general combining ability (additive effects) were more important than those of specific combining ability for yield, male flowering, and volumetric weight. Maternal and reciprocal effects were found in the crosses, so the traits evaluated are determined by nuclear and cytoplasmic inheritance, which allows crossing parents directly and reciprocally for the development and use of PB6 x PB3 and PB3 x PB6, which stood out with the greatest effect of SCA.

The use of parents of contrasting GCA (high and low) allowed their progeny to express favorable yields. Popcorn populations with high GCA effects can be used to develop synthetic varieties or drive more breeding cycles, while crosses with high SCA levels can be used to generate single-cross hybrids.