3.1. Relative performance of the provenances
Figure 3 presents the relative performance of 4 origins of P. maximinoi in 2, 4 and 8 years of evaluation, showing that the provenance of the San Jeronimo (SJ) maintained the same position qualifying of the relative performance in 2 and 4 years equal to 100% and having lowered its performance relative to 8 years for 93.79%. Another type of situation happens with the control (Common lot) that his relative performance is always increasing over the years (10.72%, 69.18% and 100%), having if detached like the second-best relative performance. Yuscaran (Yus) showed, on average, the third best relative performance, especially at 8 years.
-------------------------------------
Insert Fig. 3
--------------------------------------
The provenance of the Tatumbla (Tla) not only presented a performance very much down to 2 and 4 years but also tended to increase his performance with growth. For the original Coban (Cob), he presented the worst performance every year, except for 2 years, when it was 19.43%, which surpassed the control by nearly 9% to more.
3.2. Growth and adaptability of the provenances
The results of the ANOVA and mean test (Tukey) of different origins of P. maximinoi tested in Chimbonila are shown in Table 2. The degree of survival was significant at the 5% level at 2 years, and no differences were found in the last two years studied and ranged from 83.33 to 97.92% in the period. There were no marked differences from the heights in year 4, although the origins San Jeronimo (SJ) were higher than in the rest. At both two and eight years, there were no significant differences between the provenances. The DBH and volume parameters at 4 years or 8 years did not reveal significant differences between sources. The experimental coefficients of variation (CV%) in provenances were low for all parameters in the period studied, except for the volume to 4 years age.
-------------------------------------
Insert Table 2
--------------------------------------
The annual average increase (m3 * ha− 1yr− 1) from all sources was higher than that of the control except for the Yuscaran origin (Yus) at 4 years, and there were no differences after 8 years (Fig. 4). The Mean Annual Increments of Coban origins (Cob), Tatumbla (Tla) and Yuscaran (Yus) produced the equivalents of 113.65% and 114.34%, 102.8% and 112.06% and 100.46% and 106.2% of the Annual Average Increase control at 4 and 8 years old, respectively. San Jeronimo (SJ) had a higher mean annual increment than the other origins and the control (150.81% and 127.58%) for years 4 and 8, respectively, although there were no significant differences for 8 years.
-------------------------------------
Insert Fig. 4
--------------------------------------
3.3. Variation in growth among the progenies
The results of Tukey's test for averaging pairs among the progeny from all sources are shown in Table 3. The Tukey test was performed after the ANOVA assumed that the progeny showed significant values of the DBH and volume to 8 years old, as shown in the accompanying Table 8 in the appendix.
For the DBH parameter, the best progeny were Coban (15–884), San Jeronimo (15–868 and 15–875) and Tatumbla (15–962), whereas no statistically significant differences between them were observed, and they had a relative superiority over the other, contrary to Tatumbla (15–950), which had the worst performance of all progenies.
Regarding the volume parameter, the average progeny 15–962 is the best of all, and the worst progeny is 15–950 for both Tatumbla provenances. An important observation is in relation to the control, which showed a growth above the average of progeny.
-------------------------------------
Insert Table 3
--------------------------------------
Person's correlation coefficients (CCP) showed a strong positive (> 0.9) and significant (P < 0.01) correlation between the DBH and volume parameters for all progenies (see the last column of Table 3).
To better visualize the average of the test results of the origins of Table 2 and Table 3, the progeny of a cluster analysis were made, and subsequently, an ANOVA and mean test were performed on their respective clusters. The results are shown in Table 4 and Fig. 5 and it can be easily observed that the two-cluster solution was the best choice.
-------------------------------------
Insert Table 4
--------------------------------------
The first cluster comprised the 9 (45%) best progenies, mainly of origins Yuscaran (Yus) and San Jeronimo (SJ), and showed superiority in all parameters (height, DBH and volume). The last cluster contained 11 (55%) progenies with performance below the overall average and was mainly composed of the family of the Tatumbla (Tla) provenance. The control also belongs to the last cluster.
-------------------------------------
Insert Fig. 5
--------------------------------------
3.4. Selection of progenies based on phenotypic characters
Table 5 presents estimates of variance components and genetic parameters of the progenies of P. maximinoi at 8 years old. It is noted that by breaking the individual variance components, most of the total phenotypic variance of all the studied parameters was residual variance due to (\({\sigma }_{d}^{2}\)) and individual phenotypic variance (\({\sigma }_{f}^{2}\)). The individual heritability in the narrow sense (\({h}_{a}^{2}\)) allows us to predict moderately favourable conditions for the selection of families. However, the coefficient of determination of plot effects (\({\complement }_{e}^{2}\)) and the effect of provenances (\({\complement }_{p}^{2}\)) showed low magnitude values. The individual additive genetic variation coefficient (\({CV}_{gi}\)%) and genotypic variation coefficients among progenies (\({CV}_{gp}\)%) were high for all characters, and residual variation (\({CV}_{d}\)%) was of low magnitude for height and DBH and medium magnitude for the variable volume. Finally, the relationship between the genetic coefficient of variation and the residual coefficient of variation showed values close to and above one (1).
\({\sigma }_{a}^{2}\): additive genetic variance; \({\sigma }_{e}^{2}\): environmental variance between plots; \({\sigma }_{p}^{2}\): genetic variance among provenances; \({\sigma }_{d}^{2}\): residual variance; \({\sigma }_{f}^{2}\): phenotypic variance at the individual level in the experiment; \({h}_{a}^{2}\): individual heritability in the narrow sense; \({\complement }_{e}^{2}\): coefficient of determination of plot effects; \({\complement }_{p}^{2}\): coefficient of determination of the effects of origins; \({CV}_{gi}\)%: individual additive genetic coefficient of variation; \({CV}_{gp}\)%: coefficient of genotypic variation among progenies; \({CV}_{d}\)%: coefficient of residual variation.
-------------------------------------
Insert Table 5
--------------------------------------
Based on the predicted additive genetic effect (a) and genetic gain, nine (9) improved progenies were then classified, corresponding to 45% of the parent selection by volume, as shown in Table 6. It is noted that provenances Yuscaran (Yus) and Tatumbla (Tla) had a higher number of selected progenies, with three each. This was followed by the provenance San Jeronimo, with two selected progenies, and the Coban origin contributed only progeny.
-------------------------------------
Insert Table 6
--------------------------------------
Table 7 shows the desirable rating of the 17 (4%) best individuals selected from 20 progeny Pinus maximinoi continued for breeding. The classification criterion was based on the genetic improvement of the general average volume parameter. The San Jeronimo (SJ) origin contributed to the greater number of individuals selected, totalling 34% (6), followed by the Coban origin (Cob) to 29.4% (5) of selected individuals, all individuals of the progeny 15–884. The Tatumbla origin (Tla) contributed 23.5% (4) of individuals, and finally, the Yuscaran (Yus) contributed 11.8% (2) of individuals.
------------------------------------
Insert Table 7
--------------------------------------