DIGITALNA ARHIVA ŠUMARSKOG LISTA
prilagođeno pretraživanje po punom tekstu
|ŠUMARSKI LIST 1-2/2015 str. 52 <-- 52 --> PDF|
www.rinntech.com), with recordings and initial data processing undertaken using the TSAP Win software program (www.rinntech.de). The constructive features of the device allow a standard resolution of 1:100, while a Leica stereomicroscope (Leica Microsystems, Wetzlar, Germany, www.leica-microsystems.com) provided an enlargement factor of 6:1 (Badea 2008).
All statistical analysis was performed in the Statistica 8.0 software program, with graphs drawn using a combination of Excel and Statistica 8.0. Kolmogorov-Smirnov test was applied to check the normality of distribution and the assumptions of ANOVA were verified using Levene´s test. ANOVA was used to determine the variance components reflecting the influence of populations and repetitions, as well as that of residual variance. Considering the employed experimental design, the mathematical model selected for variance analysis was that recommended by Nanson (2004) and White et al. (2007):
Where: m = overall average value, αi = component of i populations (i = 1...a), βj = component of j repetitions (j = 1...b), εij = random error affecting ij plots.
The population × location interaction and the influence of test site variation were determined using the bifactorial ANOVA model (Nanson 2004):
Where: m and αi are as above, βj = component of j locations (j = 1...b), αβij = interaction of i populations with j locations, εijk = error caused by random events.
The level of significance was determined using Fisher’s (F) test, with population ranking and homogeneous groups determined using Duncan’s test for a transgression probability of 5%. In addition, Pearson correlations between traits and the geographical coordinates of the seed stand origins were also determined. For the latter, the ecophysiological latitude (Le = L + A / 100) was taken to represent the latitude (L) corrected by altitude (A) in such way that a 100 m difference in altitude is considered equal to one degree of latitude (Wiersma 1962).
Total height (TH) – Ukupna visina (UV)
Identical average TH value was observed in both trials established INR (17.0 m) which was similar to that seen in the Câmpina trial, situated ONR (16.9 m). In the second trial located ONR (Avrig) TH average value was significantly higher (11%) – Table 2.
ANOVA (Table 3) revealed significant differences within populations (P < 0.001) in all trials. Bifactorial analysis of variance (Table 4) demonstrated the existence of not only a strong site influence (P < 0.001), but also a highly significant population × site interaction, suggesting different reaction of populations to the change of environmental conditions. This conclusion is valuable also for AV/T, RI and LP (Table 4).
In the two trials established ONR, the most valuable seed stands were found to be those originating from Eastern and