In Mexico, seeds for plant production programs are collected mainly from natural stands, which are not subjected to management for increasing the genetic gain of the seed, and, in a smaller scale, from seedbed areas where poorly formed trees are eliminated and only those trees that may have a better genetic quality are left (^{Prieto and Martínez, 2006}).

The analysis of cones and seeds makes it possible to assess the physical and biological characteristics of a specific batch in order to assign a value to it (^{Bonner, 1993}). Furthermore, it is a helpful tool for determining the amount and quality of the seeds produced in a particular area. This allows estimating the latter’s productivity for utilization as sources of seed-producing germplasm for forest plant production programs.

Since the national demand of forest seeds grows year after year, the production of high quality germplasm must be increased through the establishment of seedbed areas and seed orchards to ensure the availability of permanent sources of germplasm with a 5 to 50 % increase in genetic gain, according to the genetic improvement strategy implemented.

In the seed orchards, the materials for propagation ―seeds or shoots―, are obtained from superior (Plus or elite) trees. In general, seed orchards are generated from a selection of 40 to 60 trees (^{Droppelmann, 2012}).

Plus trees are selected according to various techniques, and their selection depends on such factors as: a) the characteristics of the growth of the species; b) the current forest management situation, and c) the variability and the hereditary pattern of the relevant features, according to the objectives of the forest genetic improvement program (^{Zobel and Talbert, 1988}).

The present study will allow comparing the seed production of three species of the genus Pinus of the state of Durango in two seed-producing years: 2013 and 2015. Its objectives were to determine which of these species has the highest seed production index, to describe the behavior of the dimensions of their cones, and to compare their biological potential.

Pinus engelmannii Carrière is a native species of Mexico, specifically of the Sierra Madre Occidental (Wester Sierra Madre). Its mean height ranges between 15 and 25 m, and its DBH, between 60 and 80 cm; it has perennial leaves; it blooms in winter, and its cones mature from late October to mid-December (^{Conafor, 2014}). It occurs as part of Quercus and conifer forests, between the geographical coordinates 21°50’ and 31°15’ N, and 103°45’ and 110°35’ W. Its natural distribution is limited to the states of Chihuahua, Sonora, Sinaloa, Durango, and Zacatecas (^{Conafor, 2014}).

Pinus durangensis Martínez is native to the Sierra Madre Occidental of Mexico. Its height ranges between 30 and 40 m, and its diameter, between 50 and 80 cm; it has perennial leaves that grow in fascicles of 6, 5, or 7, but only in rare occasions 4 or 8, in the same branchlet, and are thin or moderately thick, 14 to 30 cm long, 0.6 to 1 mm wide, and 0.6 to 0.7 mm thick. Its cones are reddish brown, slightly curved, of an ovoid shape, 6 to 10 cm long, generally occurring in groups of two or three but sometimes individually, and they are semi-persistent for several months. This species is distributed to the northeast of the Sierra Madre Occidental, between Chihuahua, Sonora and Durango, at an altitude of 2 500 to 2 700 m, on very deep, silty clayey loam soils of a brown to reddish brown hue, with a pH of 6 to 7, with a moderate to high organic matter content (5 to 10 %); soils rich in nitrogen, silicon and humus that are permeable and deep, and rich in calcium and potassium but poor in phosphorus; at a temperature of 9 to 17 °C; with precipitations of 600 to 1 200 mm. It is resistant to frost and intolerant to drought (^{Conafor, 2014}).

Pinus cooperi C. E. Blanco is found exclusively in the Sierra Madre Occidental of Mexico. Its height ranges between 15 and 35 m, and its diameter, between 30 and 80 cm; its leaves grow in fasciles of 3 to 5, mostly 5, and only rarely in fasciles of 6, and they are thick and of a dark or ashen green hue. This species is distributed in Zacatecas, Sinaloa, Durango, Chihuahua and Sonora (^{Conafor, 2014}).

The present study was carried out in the El Salto region, in Pueblo Nuevo, Durango (Figure 1). This region is located in the Sierra Madre Occidental, in the southwest of the state of Durango, north of the municipality of Pueblo Nuevo. It can be accessed from Km 100 of the Durango-Mazatlán road, at an altitude of 2 538 masl.

Figure 1 Location of the study area, El Salto, Pueblo Nuevo, Durango.  

75 Plus trees were selected during the execution of the genetic improvement program in the charge of the Instituto Nacional de Investigaciones Forestales Agrícolas y Pecuarias, INIFAP) (National Institute for Research on Forest, Agriculture and Livestock) in Durango, in 2008. For this purpose, the best phenotypic characteristics ―a straight, clean stem; thin branches, and a small crown; cone production; color of the leaves, and height and diameter― were considered and compared to those of their four neighbors within a radius of 10 to 15 m, with a spacing of over 100 m in order to reduce the likelihood of kinship between them (^{Callahan, 1964}).

The selected individuals corresponded to 25 specimens of each taxon. In the samplings, it was observed that not all individuals produced cones and seeds; therefore, the number of individuals was smaller (Table 1).

A random sample of 12 cones was obtained from each tree; the characterization and dissection of the cones, the extraction of the seeds, and the analysis of the seeds and scales were carried out according to the methodology of ^{Bramlett et al. (1977)}, adapted by ^{Alba-Landa et al. (2000)}. The fertile scales and the developed seeds were identified; the biological potential (BP) was estimated using the following formula:

The length and diameter of the cones were measured using a FW-V8 digital caliper calibrated in millimeters, with a precision of 0.001 mm.

The biological potential is the maximum capacity of a cone to contain fully developed seeds (Figure 2).

Figure 1 Classification of the scales of a dissected cone. Lower infertile scales (A), fertile scales (B), and upper infertile scales (C). 

The seed production efficiency (DS) was estimated according to the formula:

Where:

DS = Total number of developed seeds

BP = Biological potential

The values of the means were considered in order to make a comparison and determine the differences between two seed-producing years in the same production areas for the three studied species and even at specimen level.

Furthermore, this study detected significant differences between the seed-production variables (Developed Seed DS), since it determines the actual number of seeds, unlike the Biological Potential, which refers to the maximum capacity of each species to develop full seeds in each cavity or shell in the fertile scales. However, a cone with a high biological potential may not always yield a large number of fully developed seeds.

An ANOVA variance analysis was performed using the Statistical Package for the Social Sciences (SPSS), version 24. The Kolmogorov-Smirnov normality test was applied to the seed production data per species for the two study years.

Table 1 summarizes the descriptions of the variables for the studied species for the years 2013 and 2015.

Table 2 shows the (ANOVA) statistics produced in order to determine the differences between the species in terms of seed production (DS) from 2013 to 2015.

In general, the biological potential (BP) of the three Pinus taxa increased from 2013 to 2015. When it was compared across time, the BP of Pinus durangensis increased by 23.76 %, and that of Pinus cooperi, by 21.61 %, while Pinus engelmannii exhibited no increase from one year to the other (Table 1).

When comparing only the biological potential (PB), regardless of the time, Pinus engelmannii showed the highest value in the comparison of means, with a

difference of 21.80 %, with respect to the other species (Table 1).

In terms of the developed seed (DS), there was a significant difference in both Pinus cooperi and Pinus durangensis between 2013 and 2015, with an increase in seed production of 51.98 % for P. durangensis and of 49.67 % for P. cooperi, while Pinus engelmannii exhibited no differences (Table 2).

In their comparison of three regions of the state of Durango, ^{Bustamante-García et al. (2012)} registered a significant difference in seed production for Pinus engelmannii.

^{Alba-Landa et al. (2003)} compared the biological potential and seed production efficiency of Pinus hartwegii Lindl. from two provenances in Mexico and found differences in both variables; this confirms that the differences in seed production and biological potential may be affected by various environmental, phenotypic of genotypic factors.

According to ^{Bramlett et al. (1977)}, specific clones in seed orchards exhibit little variation in seed-producing potential; however, this potential differs from the average for the species. ^{López-Upton and Donahue (1995)} point out significant differences between sites for the seed-producing potential of Pinus greggii Engelm.

^{Alba-Landa et al. (2000)} document no significant differences between sites for Pinus oaxacana; however, the present study determined that the seed-producing potential may vary significantly on an annual basis at the level of specimens. ^{Boyer (1987)} registered considerable variation in seed production in Pinus palustris Mill. between years and localities.

This agrees with the findings of ^{Spurr and Barnes (1982)}, who state that the characteristics of the seeds of a given species vary significantly according to the habitat. Thus, the differences are determined by the environmental conditions of each site and to the genetic conditions of each taxon.

Since the studied taxa are located in a region that shares the same environmental and topographic characteristics, the behavior of Pinus engelmannii in regard to all the variables may be ascribed to its adaptability to environmental changes, as it is widely distributed in Zacatecas, Sinaloa, Durango, Chihuahua and Sonora, unlike the other studied species.

This adaptability of P. engelmannii (^{Conafor, 2014}) renders it more tolerant to the modification of the various factors that affect or enhance seed production, which include the level of moisture, temperature, rainfalls, soil nutrients and pollination.

Pinus engelmannii is a species with a stable seed production, unlike Pinus cooperi and Pinus durangensis.