Fiber Characteristics of White Poplar ( Populus alba L . ) Juvenile Wood along the Drava River Svojstva vlakanaca juvenilnog drva bijele topole ( Populus alba L . ) uz

Anatomical properties of white poplar (Populus alba L.) wood have not been suffi ciently investigated and quantifi ed, whereas data on fi ber characteristics will certainly defi ne its potential as a raw material. Fiber length, double cell wall thickness and fi ber lumen diameter of white poplar juvenile wood from two sites along the Drava River in Croatia were measured. Variations in fi ber characteristics in radial direction (from pith to bark), within and between white poplar populations from two sites were investigated. Results indicate that fi ber Rlength and double cell wall thickness varied signifi cantly from pith to bark, within and between sites. Variation in fi ber lumen diameter was non-signifi cant only between sites. The variation in radial pattern was characterized by increase in fi ber length with cambial age. Double cell wall thickness and fi ber lumen diameter varied in non-consistent radial pattern. In conclusion, variation signifi cance demonstrates nonhomogeneous wood fi ber characteristics at both sites.


INTRODUCTION 1. UVOD
Poplar wood is widely used for the manufacture of a large number and variety of products, especially for pulp, paper and other fi ber-based products.It is reported that approximately 70 countries worldwide grow poplars, both in mixtures with other natural forest species and in planted forests (Ball et al., 2005).However, the great majority of poplars still grow in natural forests.The value of diverse natural stands of poplars as a basis for tree improvement is well recognized (Ball et al., 2005).Acquiring genotypes of forest trees or trying to upgrade the existing properties is one of the main aims of tree improvement (Kajba and Ballian, 2007).It is often carried out in order to increase the productivity of the wood mass.The increase of wood quality is another important goal, which cannot be carried out without an understanding of wood structure.
Cell size and relative cell dimensions have a major infl uence on the quality of pulp and paper based products, as well as on solid wood products (Clark, 1985).Mechanical properties of wood stiffness and strength are, among others, also related to wood fi ber characteristics (Logan, 2006).
As they grow so rapidly, both natural and planted poplar stands can produce large volumes of wood in a short period of time (Dickmann and Kuzovkina, 2014).This suggests that they can be harvested within 10-20 years, having a high portion of juvenile wood.Juvenile wood is characterized by rapid changes in wood structure throughout the zone from the pith outward (Tsoumis, 1991).Maeglin (1987) reported shorter fi bers with generally thinner walls at or near the pith.According to Dadswell (1958), fi ber length is increased up to two times in the central core of hardwoods.
As one of the most distinctive poplar species, white poplar (Populus alba L.) is an indigenous wood species in Europe.In Croatia, white poplar can rarely be found in pure stands covering large areas, but forms mixed stands together with other species, such as white willow and black poplar.However, pure stands of white poplar are found along the Danube, Sava and Drava (Franjić and Škvorc, 2010).For a long period of time, white poplar has not been considered as a commercially important wood species.Due to climate changes and changed ecological conditions, it is becoming more interesting for breeding as a drought tolerant species (Eggens et al., 1972;Gilman and Watson, 1994;Isebrands and Richardson, 2014).
Within the Populus genus, white poplars have been least investigated.Horvat (1960) investigated some physical and mechanical properties of white poplar wood in Croatia.However, available data on anatomical properties of white poplar wood are limited (FAO 1958).The capacity of white poplar to grow in a wide range of soils and to propagate easily contributes highly to its cultivation.For that reason, the available data on its wood anatomical properties should be completed.Additionally, variability in juvenile wood characteristics signifi cantly affects product quality.There-fore, more emphasis should be put on determining the size of wood variability.
The aim of this study was: (a) to determine fi ber characteristics of white poplar juvenile wood from two sites along the Drava River in Croatia; (b) to investigate variations in fi ber characteristics in radial direction (from pith to bark), within and between white poplar populations from two sites.

MATERIJAL I METODE
Five trees of white poplar (Populus alba L.) were collected from two sites with continental climate along the Drava River.The fi rst site is located near the city of Osijek, within Osijek Podravina forests.It is characterized by alluvial loamy-sandy soils, where humus is being formed, with mean annual temperature of 11.6 °C and total annual precipitation of 694.4 mm.
The second site is located near the city of Varaždin, within Varaždin Podravina forests.It is characterized by soil on alluvial pebbly-sandy shoals intersected with meanders, with mean annual temperature of 10.8 °C and total annual precipitation of 819.9 mm.Both units are managed by the company Croatian Forests Ltd.
Each test tree was marked on the side facing north because of further investigations on physical and mechanical properties.Relevant parameters on growth locations of all test trees were collected and measured (Table 1).
Disks, approximately 5 cm thick, were cut at breast height (1.3 m) from each tree.Radial segments (north-south orientation) were cut from each disk and annual growth rings were marked along one radius.On both sites, growth rings 2, 4, 6, 8 and 10 from the pith were selected from each tree for further anatomical analysis of wood.A series of radially oriented sample blocks, sized 10 (T) × 10 (R) × 20 (L) mm, was cut from the selected growth rings.
For measuring double cell wall thickness and fi ber lumen diameter, transverse sections (30 μm) were cut using Shandon Reichert sliding microtome.Sections were then stained with safranin and astra blue, while excess stain was removed by washing sections successively in 70 % and 96 % ethanol solution.Furthermore, they were mounted on microscope slides with coverslips using euparal.Photographs of samples were then taken at ×160 magnifi cation using stereo microscope, digital camera and Axio Vision program (Carl Zeiss).In each growth ring, fi ve images were taken as random replicates: one from the early wood region, one from late wood region and three from the middle part of the growth ring, according to Peszlen (1994).Fiber lumen diameter was measured by digital photographs, using ImageJ program (ImageJ 2014).
Double cell wall thickness was measured directly on light microscope, using an objective with measurement scale.The magnifi cation ×960 was used.From each growth ring, 30 tangential fi ber double cell wall thicknesses were measured (ten in the early wood region, ten in the middle part of each growth ring and ten in the late wood region).
The mean fi ber length of white poplar juvenile wood from Osijek site was 0.87 mm and from Varaždin site 0.84 mm (Table 2).Similar mean values of fi ber length were determined by Boessneck, Brossa, Viviani and Demi in white poplar wood from several locations in Europe, and they are all less than 1 mm (in FAO 1958).However, Demi also reported fi bers longer than 1 mm.The mean double cell wall thickness of white poplar juvenile wood from Osijek site was 3.72 μm and from Varaždin site 4.16 μm (Table 2).Those mean values are in the range of mean values reported by Demi (between 3.36 μm and 4.95 μm).Data on fi ber lumen diameter are not available in previously mentioned studies and thus could not be compared.Measured fi ber characteristics could be classifi ed according to the list of microscopic features for hardwood identifi cation, published by IAWA (Wheeler et al. 2007).This list is comprised of 163 anatomical features, including fi ber length and fi ber wall thickness.Fiber length was classifi ed into 3 categories and mean values from both sites fi t the fi rst category (fi bers ≤ 900 μm).Fiber wall thickness was also defi ned in 3 ways, by For fi ber length measurements, wood from each of the previously selected growth rings was separated with a razor blade and placed into an individual test tube.Maceration was performed according to Franklin's method (Franklin, 1945).Macerated material was stained using safranin, then placed on microscope slides and mounted in glycerin gelatin.From each growth ring, 40 unbroken fi bers were measured using light microscope, digital camera and DinoCapture 2.0 program (DinoCapture 2.0 2015).
Statistical analysis of results was carried out using Statistica 10.Repeated measures analysis of variance (ANOVA) was used to test signifi cance of differences in radial distribution of selected fi ber characteristics, as well as their variations within and between two white poplar populations.

REZULTATI I RASPRAVA
Statistical values of fi ber dimensions of white poplar juvenile wood from both sites are given in Table 2. fi ber lumina in relation to double wall thickness.Based on that relation, fi bers from both sites are very thin walled (fi ber lumina was 3 or more times wider than the double wall thickness).Signifi cant differences in fi ber length and double cell wall thickness were detected between sites (Table 3).Due to soil type and fl ooding effect on Varaždin site, Osijek site is considered to be more favorable for poplar growth.In this case, the better site accelerated the growth rate and produced trees with longer fi bers in the fi rst ten years that were analyzed.This is in agreement with Kennedy and Smith (1959), who found greater fi ber length in poplar wood from good sites.According to Peszlen (1994) and Pliura et al. (2007), those differences are most likely related to site edaphic and climatic conditions.This may be explained by several factors.The mean length of the fusiform cambial initials increases gradually as the tree matures.Considerable fl uctuation in their length may occur at any time as a result of fl uctuating environmental conditions that govern seasonal growth (Panshin and de Zeeuw, 1980).However, fast growth of hardwoods does not always result in meaningfully shorter fi bers if wood of equal age is compared (Zobel, 1989).The results did not sup-Table 3 Results of repeated measures analysis of variance for fi ber characteristics of white poplar juvenile wood from two sites and in direction from pith to bark Tablica 3. Rezultati analize varijance ponovljenih mjerenja svojstava vlakanaca juvenilnog drva bijele topole s dva staništa i u smjeru od srčike prema kori   (Petrić et al., 1991;Šefc et al., 2009;Zhao et al., 2014).They reported about non-signifi cant difference in selected wood anatomical properties between sites.In this study, difference was non-signifi cant only in fi ber lumen diameter (Table 3).According to the results, the change in fi ber diameter would be mainly the result of As expected, fi ber length signifi cantly increased linearly from pith to bark in white poplar juvenile wood from both sites (Figure 1 and Table 3).This suggests that cambial age has an important effect on the length of wood fi bers.Longer fi bers were measured in all fi ve annual rings from Osijek site (Figure 1).Signifi cant variation in fi ber length from pith to bark of poplar wood was reported by many authors over the years (DeBell et al., 1998;Koubaa et al., 1998;Huda et al., 2012).Double cell wall thickness also varied signifi cantly from pith to bark, with more pronounced change between annual rings in white poplar juvenile wood from Varaždin site (Figure 2 and Table 3).Results from Varaždin site are only partially in agreement with Peszlen (1994) and Huda et al. (2012).Both reported about continuous increase in cell wall thickness with age.Furthermore, fi ber lumen diameter showed similar pattern of radial distribution in white poplar juvenile wood from both sites (Figure 3).However, differences in fi ber lumen diameter were most pronounced in growth ring 6 from the pith.

Fiber lumen diameter
There was highly signifi cant difference in fi ber length, double cell wall thickness and fi ber lumen diameter among trees within both white poplar populations (Table 4).Signifi cant tree-to-tree variation in fi ber length of poplar clones was reported by Klašnja et al. (2003), and additionally in fi ber lumen diameter and fi ber lumen area by Peszlen (1994).This may be explained by response of each tree to differences in its microenvironment, or may be due to large within-tree variation (Zobel and van Buijtenen, 1989;Tsoumis, 1991).Subsequently, variations in a certain site are considered to be higher than between sites.More precisely, variations in wood properties are generally more pronounced within the tree in comparison to variations between trees of one population or between populations (Larson, 1967;Tsoumis, 1991).

ZAKLJUČCI
As expected, the measured juvenile wood fi ber characteristics were determined to be highly variable.Fiber characteristics of white poplar juvenile wood from two sites differed.Signifi cant site effect was detected for fi ber length and double cell wall thickness.Trees within both white poplar populations had highly signifi cant effect on fi ber length, double cell wall thickness and fi ber lumen diameter.
There was a general trend in radial direction in both white poplar populations, in which fi ber length increased from pith to bark.Double cell wall thickness and fi ber lumen diameter also varied signifi cantly with age, but no consistent radial pattern was observed.
For the present study, variations in each site (both within and among trees) were generally more significant in comparison to variations between sites.
Forest product manufacture tends for wood uniformity.Due to highly signifi cant variability, nonhomogeneous wood fi ber characteristics were determined within both white poplar populations.In conclusion, variation in wood characteristics provides the possibility to select the breeding stock with desirable wood quality.

Table 4
Results of repeated measures analysis of variance for fi ber characteristics of white poplar juvenile wood within population from each site Tablica 4. Rezultati analize varijanci ponovljenih mjerenja svojstava vlakanaca juvenilnog drva bijele topole sa svakog staništa unutar iste populacije