Seed relative weight, morphometric variation, and germination

Seeds of A. americana from Sauz Xathe provenance (AmerSauz) exhibited the maximum relative weight and had the highest values (α ≤ 0.05) in all the morphometric variables, except for roundness; seed length appeared as a distinctive feature of AmerSauz, while roundness was uniformly higher in all the A. salmiana seed origins relative to A. mapisaga and A. americana (Table 1). Multivariate analysis of the seed morphology results demonstrated divergence of A. americana respect to A. salmiana and A. mapisaga according to Andrews curves (Figure 1).

Table 1. Relative weight (RW), morphometric evaluation, and initial maximum germination (Max. Germ.) of seeds collected in 2012.

Note. Mean values with different letters within columns are statistically different according to Tukey’s test (α ≤ 0.05). Seed origin = species × provenance, described in Section 2.

Abbreviations: MSD, minimum significant difference; RW, relative weight.

Figure 1. Andrews curves constructed with seed morphometric variables evaluated from A. americana (blue), A. mapisaga (red), and A. salmiana (purple) seed origins.

AmerSauz germination of 89% was inferior relative to the rest of the seed origins (range from 93 to 100%; Table 1). AmerSauz seed capsules were brown color at the collection time, indicative of a dry capsule and the actual moment of flowering stalk maturation was unknown. The variable time that the seeds remained under the abiotic elements might influence on seed’s relative moisture content possibly affecting embryo viability, or enabling dormancy induction; as consequence, reducing germination regardless of its highest relative seed weight (McDonald, Reference McDonald2007; Peña-Valdivia et al., Reference Peña-Valdivia, Sánchez-Urdaneta, RJR, Trejo, Cárdenas and Villegas2006). In contrast, capsules of the rest of the seed origins were green yellowish, the green color was indicative of physiological activity of the capsules’ outer tissue and the yellow color exhibited chlorophyll denaturation part of the aging process; thus, the capsules of these seed origins had a similar maturation time and less affectation by the abiotic elements.

Within A. salmiana seed origins the seeds from Tlaxiaca provenance (SalmTlax), had the highest relative weight, greatest area, and width, which were proportional with high germination of 99%, exhibiting intraspecific variation given by anthropogenic and/or environmental factors. Seed origin of A. mapisaga from Metepec (MapsMet) showed a greater germination of 100%, which was proportional to higher weight, similar to SalmTlax (Table 1). These results indicated that higher relative seed weight was proportional to elevated germination (Penfield, Reference Penfield2017); also, pinpointed the importance of capsule maturation stage for seed harvest, which is an anthropogenic factor that should be considered to standardize Agave seed processing and storage.

Seedlings genetic variation

Seedling DNA amplification exhibited 70 fragments which 57 (81.42%) were polymorphic. Asexual propagation, used traditionally, can reduce genetic variation by 8% for A. salmiana and A. mapisaga populations (Alfaro et al., Reference Alfaro, Legaria and Rodriguez2007), and by 10% for Agave tequilana plantations (Abraham et al., Reference Abraham, Ramírez and Gil2009), which can lead to genetic erosion (Gil-Vega et al., Reference Gil-Vega, González, Martínez, Simpson and Vandemark2001).

The dendrogram (Figure 2) shows AmerSauz isolated from the rest of the seed origins, which is consistent with the differences observed in the seed morphometric results (Table 1 and Figure 1). These results exhibited the interspecific variation of AmerSauz with respect to the plants from the Salmianae section (Gentry, Reference Gentry1982); given by geographic isolation of Sauz Xathe provenance located in the Meztitlan Canyon (INEGI, 1997), reducing its distribution and interaction with other Agave species. Also, the anthropogenic influence can accentuate AmerSauz characters, subjected to less domestication that induce smaller plants with less sap production, compared to A. salmiana and A. mapisaga landraces that have been intensively selected for those purposes (Mora-Lopez et al., Reference Mora-Lopez, Reyes-Agüero, Flores-Flores, Peña-Valdivia and Aguirre- Rivera2011; Trejo et al., Reference Trejo, Reyes, Cortes-Toto, Romano-Grande and Muñoz-Camacho2020).

Figure 2. Dendrogram constructed according to the binary matrix of DNA bands amplified from seedlings of Agave seed origins (species × provenance) using Ward’s method (city block distance).

SalmMet and MapMet were paired in the closest cluster, as both came from the same plantation with similar inflorescence maturation which inferred that hybridization might occur. SalmTlajo and SalmTlax were paired, as both might share genetic traits as a consequence of similar environmental factors, which provenances are in the northern convergent plateau between Mexico City, for Tlajomulco in the northeast toward Tulancingo Valley, and for Tlaxiaca in the northwest toward Actopan Valley, both provenances are communicated by a natural corridor that passes through Pachuca city possessing xerophytic vegetation with sub-humid to semi-arid conditions; contrasting to SalmMetβ located in mild conditions of Tulancingo Valley (INEGI, 1992; Jimenez-Torres et al., Reference Jimenez-Torres, Peña-Valdivia, Padilla-Chacon and Garcia-Nava2021).

Germination after a 10-year storage period

After a 117-month storage period (≈ 10-year storage period), when the lids were removed from the glass jars it was broken a vacuum seal making a characteristic sound while releasing the lid, although these containers were only airtight sealed not vacuum sealed, which indicated that seed respiration during storage consumed the oxygen present into the jar generating a vacuum by itself. The maximum germination for A. salmiana seeds from Metepec (SalmMetβ) and Tlaxiaca (SalmTlax) after a 10-year storage period was 96 and 95%; respectively (Table 2), with a germination reduction of 1 and 4% compared to the initial germination observed at 2012 (Table 1).

Table 2. Evaluation of maximum germination (Max. Germ.), SM content, RW, and DW of seeds after 55-month (SalmTlajo17) and 10-year storage period (SalmTlax and SalmMetβ).

Note. Mean values with different letters within columns are statistically different according to Tukey’s test (α ≤ 0.05). Seed origin = species × provenance, described in Section 2.

Abbreviations: DW, dry weight; MSD, minimum significant difference; RW, relative weight; SM, seed moisture.

Seeds of A. salmiana collected in September 2017 from Tlajomulco provenance (SalmTlajo17) after a 55-month storage period, exhibited no statistical differences in maximum germination results with respect to the other seed origins, regardless of a longer storage period of SalmMetβ and SalmTlax (Table 2). These results pinpointed the importance of the capsules’ maturation stage for seed preservation, as SalmTlajo17 capsules were brown color fully mature, having similar germination results as AmerSauz (Table 1).

After a 10-year storage period, no statistical differences were observed between SalmMetβ and SalmTlax in seed moisture content, relative weight, and dry weight (Table 2). SalmTlajo17 exhibited significant differences for all the variables evaluated (except maximum germination) respect SalmTlax (Table 2).

These values demonstrated that A. salmiana seeds can withstand desiccation and long storage periods (>5 years), such as orthodox seeds, which maintained relatively adequate germination under the given storage conditions. Seed moisture content, storage temperature, and oxygen exposition are key factors that must be considered when working with species without known adequate storage parameters (Bewley et al., Reference Bewley, Bradford, Hilhorst, Nonogaki, Bewley, Bradford, Hilhorst and Nonogaki2013; De Vitis et al., Reference De Vitis, Hay, Dickie, Trivedi, Choi and Fiegener2020).

The results obtained in this work can be used as reference parameters to determine the optimum Agave’s seed moisture content threshold for adequate storage conditions. In this work, the storage temperature was not controlled nor recorded, but according to normal climatic data of the nearest weather station (Zempoala, Hidalgo) from the storage facility, the storage temperature did not surpass the range of 24–26°C during the warmer season of March–April, and the rest of the year was below that temperature range (CONAGUA, 2021). Oxygen exposition was controlled by airtight containers, the seeds consumed the air that remained in the glass jar and generated a vacuum seal.