ORIGINAL ARTICLE
 Comparison of structure, regeneration and dead wood in virgin forest
 remnant and managed forest on Grmec? Mountain in Western Bosnia
 C? . VIS?NJIC? 1, S. SOLAKOVIC? 2, F. MEKIC? 1, B. BALIC? 1, S. VOJNIKOVIC? 1, M. DAUTBAS?IC? 3,
 S. GURDA1, F. IORAS4, J. RATNASINGAM5 & I. V. ABRUDAN6*
 1Faculty of Forestry, University of Sarajevo, Bosnia and Herzegovina; 2Unsko-Sanske Sume, Bosanska Krupa, Bosnia and
 Herzegovina; 3University of Sarajevo, Bosnia and Herzegovina; 4Centre for Conservation and Sustainability,
 Buckinghamshire New University, United Kingdom; 5Forestry Faculty, University of Putra, Malaysia and 6Faculty of
 Silviculture and Forest Engineering, Transilvania University of Brasov, Romania
 Abstract
 This paper compares the forest structure, regeneration and distribution of dead wood in a virgin forest remnant and a close-
 to-nature managed beech?conifer mixture situated on Grmec? Mountain in Western Bosnia. The investigations were carried
 out in a 1 ha permanent sample plot and 35 circular plots (20 m radius) in the virgin forest and in 17 circular plots (25 m
 radius) in managed forests. The number of trees in the managed forest was significantly ( p ? 0.05) higher than that in virgin
 forest and the distribution of the number of trees per diameter classes had a decreasing trend, but with a different shape in the
 virgin forest compared to the managed stands. In the lower diameter classes, the stock volume recorded in virgin forest was
 half of that in the managed forest, whilst for higher diameter classes the cumulated volume of the growing stock was almost
 double in virgin forest. The young crops had a significantly lower presence in the virgin forest and a larger volume of dead
 wood was identified in the virgin forest than in managed stands. The study results are important in assessing the
 consequences of close-to-nature management on the forest structure and regeneration when compared to the condition in
 virgin forests.
 Keywords: Dead wood, growing stock, managed forest, structure, virgin forest, Western Bosnia
 Introduction
 Bosnia Herzegovina?s (BiH?s) forests (1,266,000 ha)
 are part of the Piceo-Abieti-Fagetum association, with
 common beech, silver fir and Norway spruce
 representing more than 80% of BiH?s forest area
 (Pintaric? 1978; Ioras et al. 2009). Mixed forests of
 beech?fir and beech?fir?spruce account for around
 44% or more than 560,000ha (Matic et al. 1971;
 Visnjic et al. 2009).
 Virgin/pristine forests are important remnants of
 valuable and rare forest ecosystems in BiH. They
 provide a basis for close-to-nature silvicultural
 research and applications and for designing national
 networks of protected forest and they represent a
 reference for naturalness assessment of other
 managed forests (Abrudan and Mather 1999;
 Abrudan 2000; Ioras et al. 2009; Visnjic et al.
 2009; Dautbasic et al. 2010).
 In 1905, about 50% of the total forest area of BiH
 (2 million ha) was considered virgin forest (Fro?hlich
 1954), i.e. forest that has not had any human-
 induced treatment (Lund 2002). Nowadays, most of
 these forests have lost their virgin status, as in the last
 century they have been managed mainly for wood
 production, especially under the uniform or irregular
 shelterwood systems (Pintaric 1999). Twenty-seven
 ?strict forest reserves? totalling 3125 ha in area
 remained in Bosnia (Parviainen et al. 2000);
 however, only five of these reserves can be considered
 as ?virgin forest reserves? (Pintaric 1999). The total
 area of virgin forest remnants in Europe is difficult to
 estimate since this area depends on the definition of
 the notion ?virgin forest? (Lund 2002). Strictly
 protected forest areas cover more than 100,000ha in
 Austria, Belgium, the Czech Republic, France,
 Germany, Hungary, The Netherlands, Poland,
 Slovakia and Switzerland (Parviainen et al. 2000).
 ISSN 1126-3504 print/ISSN 1724-5575 online q 2012 Societa` Botanica Italiana
 http://dx.doi.org/10.1080/11263504.2012.751064
 Correspondence: I. V. Abrudan, Faculty of Silviculture and Forest Engineering, Transilvania University of Brasov, 1, Sirul Beethoven, 500123 Brasov, Romania.
 Tel: ? 40 723 533512. Fax: ? 40 268 476808. Email: abrudan@unitbv.ro
 Plant Biosystems, iFirst article, 2012, 1?10
 Downloaded by [193.63.226.141] at 06:55 14 February 2013
  
However, only a minor part of these forests can be
 classified to be virgin forests. In Scandinavia and
 Eastern Europe, larger patches of forest that have
 suffered human impact still exist (Parviainen et al.
 2000).
 Under the present ownership and socio-econ-
 omic context, the vast majority of forests in BiH are
 managed primarily for wood production (Ioras et al.
 2009; Avdibegovic et al. 2010); however, BiH has
 less than 0.3% of its forests classified as pristine
 (virgin) forests, namely Perucica, Janj, Lom and
 Trstionica forests (Ioras et al. 2009; Dautbasic &
 Ioras 2010). The only pristine forests in BiH that
 have been the subject of several scientific investi-
 gations in the past are Peruc?ica, Ravna vala, Mac?en
 do and Pljes?evica (Pintaric? 1978; Leibundgut 1993;
 Beus & Vojnikovic? 2002; Sebastia et al. 2005;
 Meskovic 2006; Visnjic et al. 2009).
 In this study, the authors conducted a vegetation
 survey in the managed beech?fir?spruce mixed
 forests on Mt. Grmec? in Western Bosnia. The
 obtained indicators were compared with structural
 indicators of ?Bobija? pristine forest that is located in
 the same area. The study endeavoured to assess the
 impact of the application of the group selection
 system on the structure and regeneration of mixed
 Figure 1. Mt. Grmec? location in Bosnia?Herzegovina.
 Figure 2. Location of the study area and sample plots.
 C? . Vis?njic? et al.2
 Downloaded by [193.63.226.141] at 06:55 14 February 2013
  
beech?fir?spruce managed forests compared to
 pristine forest.
 Material and methods
 Study area
 The research was conducted on Mt. Grmec? in
 Western Bosnia (Figure 1), which has a total forest
 area of 5473.8 ha. The study was carry out in a 17-
 km2 section of Mt. Grmec?, dominated by beech?fir?
 spruce mixed forests found on limestone?dolomite
 substratum (Stefanovic 1988) at an average altitude
 of 1200 m (latitude: N 44840027.1900; longitude:
 E 16821059.6200). These forests are relatively hom-
 ogenous according to their ecological and vegetation
 features (bedrock, soil, climate, forest type, structure
 and management).
 Mountain Grmec? is a Class 4 type (UNEP-
 WCMC classification) with a maximum altitude of
 1605 m above sea level (Anonymous 2006). Group-
 selection silvicultural system has been applied in the
 managed mixed forests in the last decades (Matic
 et al. 1977; S?GD Unsko-sanske s?ume 2010), whilst a
 small part of the forest ? Bobija pristine forest, with
 an area of 56 ha (most of it situated in compartment
 1 ? Figure 2) ? has remained untouched and was
 proposed for designation as a forest of special
 interest.
 The dominant soil types are black soils and
 limestone brown soils. The climate of the area is
 moderate continental with mild impact of the
 Atlantic climate (the main climate characteristics
 are summarized in Table I). However, as altitude
 increases above sea level, the climate becomes more
 continental.
 Methodology
 Species composition, distribution of the diameter
 classes and development stages, proportion of dead
 wood and dieback trees were assessed in a 1 ha (100
 ? 100 m) permanent sample plot (the sides of which
 had a north?south by east?west orientation) and in
 35 circular plots (with a radius of 20m) located at the
 intersection of a 100 m by 100 m overlaid grid in the
 virgin forest. This systematic sampling is rec-
 ommended for relatively small forest areas like the
 virgin forest investigated in this study, although
 similar studies show that a 20-m-radius plot size is
 sufficient to describe the structural characteristics of
 mixed stands (Leibundgut 1993; Abrudan and
 Mather 1999; Beus & Vojnikovic? 2002; Sebastia
 et al. 2005; Visnjic et al. 2009). In the managed
 stands, data were collected from 17 circular plots
 (25 m radius) located at the intersection points of aTabl
 e
 I.
 Cli
 m
 at
 e
 cha
 racteristic
 s
 o
 fth
 e
 st
 u
 die
 d
 a
 rea
 .a
 Mea
 n
 ai
 r
 temp
 eratur
 e
 (8C
 )
 M
 ea
 n
 ra
 infal
 l(mm
 )
 Lengt
 h
 o
 fth
 e
 gr
 owin
 g
 seaso
 n
 (d
 ays
 )
 Potentia
 le
 vaporatio
 n
 in
 A
 pri
 l?
 Septemb
 er
 (mm
 )
 An
 nua
 l
 A
 pri
 l?
 Septemb
 er
 A
 nnua
 l
 Apri
 l?
 Sept
 embe
 r
 7.
 6?
 8.
 7
 12
 .
 4?
 14.
 6
 1,
 19
 8?
 1,35
 0
 62
 5?
 63
 0
 15
 6?
 17
 0
 50
 3?
 52
 8
 a
 B
 as
 ed
 o
 n
 da
 ta
 pr
 ov
 id
 ed
 by
 D
 rin
 ic
 a
 n
 d
 B
 os
 an
 sk
 iP
 et
 ro
 v
 a
 c
 m
 et
 eo
 ro
 lo
 gi
 ca
 ls
 ta
 tio
 ns
 .
 Grmec? Mountain virgin forest in Bosnia 3
 Downloaded by [193.63.226.141] at 06:55 14 February 2013
  
1 km by 1 km overlaid grid, situated in the centre of
 the studied forest area (Figure 2).
 In each circular plot located in the virgin forest,
 the number of trees per species, diameter at breast
 height (dbh) for each tree (.5cm), basal area and
 growing stock were recorded/calculated. The per-
 manent plot was divided into 100 quadrates, 10 by
 10 m. At the crossing points of these quadrates, 81
 circular plots with a radius of 3m were established
 and young seedlings were recorded in each plot, per
 the method described by Matic? (1977). Recording of
 dead wood volume was conducted in the permanent
 plot from the virgin forest according to the
 methodology defined by Albrecht (1990).
 In the managed forest, all trees above 5 cm dbh
 were recorded in each of the 17 circular plots.
 Seedlings were recorded in circular plots of 3 m
 radius located at the intersection of the diagonals of
 the 1 km by 1 km squares. For all seedlings/trees with
 a diameter less than 5 cm dbh and a height greater
 than 10 cm as the recording method described by
 Matic? (1977) was applied. The volume of dead wood
 was recorded, in circular plots of 25 m radius, with
 the same point of origin as the 3-m-radius plots.
 Recording of dead wood volume was conducted in
 the permanent plot from the virgin forest as well as in
 the managed forest according to the methodology
 defined by Albrecht (1990). Thus, in each circular
 plot dead wood that was thicker than 7 cm was
 recorded and diameter and height/length was
 measured to calculate the volume. Also, the category
 of dead wood (lying dead wood, standing dead wood,
 broken dead wood, stump and assortment) and the
 degree of decomposition (freshly dead wood, rotten,
 and decomposed) were recorded.
 To assess the statistical significance of differences
 between the mean number of trees (N), basal area
 (G) and volume of growing stock (V) in virgin and
 managed forests, an independent samples t-test was
 applied.
 Results and discussion
 Number of trees
 The total number of trees per hectare recorded in the
 virgin forest was 429 trees. Of these, 201 were silver
 fir (Abies alba Mill.), 1 Norway spruce (Picea abies
 L.), 165 common beech (Fagus sylvatica L.), 23
 sycamore (Acer pseudoplatanus L.), 8 ash (Fraxinus
 excelsior L.), 26 elm (Ulmus glabra Huds.) and 5 lime
 (Tilia cordata Mill.) trees. The average number of
 trees in the managed forest was 1001 trees per
 hectare, of which 445 were silver fir, 94 Norway
 spruce, 386 common beech, 40 sycamore, 20 ash, 11
 elm, 4 rowan/mountain ash and 1 lime tree. The
 independent samples t-test revealed a statistically
 significant difference between the number of trees in
 managed forest and virgin forest (Table II).
 Distribution of the number of trees per diameter
 classes is mainly used as an indicator of the structural
 composition of the stands. Early research by
 Leibundgut (1945), Pintaric and Izetbegovic (1980)
 and Parviainen (2005) indicated that in some cases
 virgin forests could have a structure typical to a forest
 that has undergone selection system.
 The hyperbolic distribution in Figure 3 is clearly
 reflecting the structure of a forest managed under the
 selection system, with the number of trees decreasing
 in direct correlation with dbh increase. Despite the
 Table II. Basic statistical parameters for the number of trees (N), basal area (G) and growing stock (V) of the investigated managed forest
 (MF) and virgin forest (VF).
 Parameter/Statisticstatistic Variable
 N_MF N_VF G_MF G_VF V_MF V_GF
 Number of plots 17 35 17 35 17 35
 Minimum value 713 162 32.99m2/ha 19.12m2/ha 457.51m3/ha 340.42m3/ha
 Maximum value 1518 1158 52.44m2/ha 62.61m2/ha 844.88m3/ha 2040.51m3/ha
 Mean value 1001 429 37.99m2/ha 45.28m2/ha 611.53m3/ha 725.23m3/ha
 Coefficient of variation 22.8% 78.6% 15.7% 29.2% 16.4% 46.8%
 t 9.74 2.49 3.32
 P p value 0.05 0.01 0.01
 0
 50
 100
 150
 200
 250
 300
 350
 400
 450
 0 10 20 30 40 50 60 70 80 90 100 110 120 130 140
 Diameter class (cm)
 Managed forest
 Virgin forest
 Number of trees (N/ha
 )
 Figure 3. Distribution of the number of trees per diameter classes
 (virgin forest and managed forest).
 C? . Vis?njic? et al.4
 Downloaded by [193.63.226.141] at 06:55 14 February 2013
  
fact that the two curves follow a similar trend it does
 not automatically follow that virgin forest has a
 structure typical for a forest managed under the
 selection system (Anic and Mikac 2008). It can be
 seen in Figure 3 that the number of trees in the 5- to
 30-cm diameter classes is significantly smaller in the
 virgin forest than in the managed forest, whilst the
 distribution curve shows several peaks for diameter
 classes bigger than 31 cm. In the case of managed
 forest for diameters classes above 90 cm, the curve is
 actually relatively flat, without significant variations.
 Distributions of the number of trees by diameter
 classes in virgin forest and in managed forest
 followed different patterns for the same tree species
 (Figure 4). A high decrease in the number of trees as
 the diameter increased was recorded for silver fir in
 virgin forest. It can be noticed in Figure 4 that for
 diameter class 1?10 cm, the proportion of beech in
 managed forest was much higher than that in virgin
 forest; this might be due to the application of the
 group selection cutting correlated with a mast year.
 Also, for the diameter class 11?20 cm, the decrease
 of the number of fir trees was much higher in virgin
 forest than in managed forest. For common beech,
 the trend was different in the studied plots. The
 number of trees per diameter classes was less variable
 in the virgin forest. Similar to common beech, ash
 had a relatively even distribution. In the managed
 forest, the distribution of the number of trees of silver
 fir and common beech followed a hyperbolic pattern
 similar to the one in Figure 3, namely the higher the
 diameter class was, the lower the number of trees
 present.
 Basal area
 Basal area in the investigated virgin forest was
 45.28 m2/ha, whilst in the managed forest it reached
 37.99 m2/ha. The independent samples t-test indi-
 cated a statistically significant difference between the
 basal area in the managed forest and virgin forest
 (Table II).
 Figure 5 shows a mild left-skewed distribution of
 basal area, which for the virgin forest is more
 accentuated. This is due to a higher share of larger
 size trees. It is also evident that as a result of the
 presence in the managed forest of a large Norway
 spruce tree (diameter class 121?130 cm), both the
 basal area and the volume for the respective diameter
 class was higher in managed forest than in virgin
 forest (Figures 5 and 6). In the virgin forest, the basal
 area pick was recorded in the 71- to 80-cm diameter
 class whilst in the managed forest it was recorded in
 the 31- to 60-cm diameter class, which is under-
 standable, since the increment actually happens in
 this diameter class. In virgin forest, there is no felling
 and in managed forest a selective silvicultural system
 has been conducted, so the basal area of managed
 Managed forest
 0
 100
 200
 300
 400
 500
 600
 700
 10 20 30 40 50 60 70 80 90 100 110120 130
 Diameter class (cm)
 lime
 rowan
 elm
 ash
 sycamore
 beech
 Norway spruce
 silver fir
 Virgin forest
 0
 50
 100
 150
 200
 250
 10 20 30 40 50 60 70 80 90 100 110
 Diameter class (cm)
 Number 
of 
trees 
(N/ha
 )
 Number 
of 
trees 
(N/ha
 )
 lime
 elm
 ash
 sycamore
 beech
 Norway spruce
 silver fir
 Figure 4. Distribution of the number of trees per tree species and
 diameter classes (virgin forest and managed forest).
 0
 1
 2
 3
 4
 5
 6
 7
 8
 9
 Diameter class (cm)
 Managed forest Virgin forest
 Basa
 l are
 a 
(m
 /ha
 )
 0 10 20 30 40 50 60 70 80 90 100 110 120 130 14
 Figure 5. Distribution of the basal area per diameter classes in
 virgin forest and managed forest.
 0 10 20 30 40 50 60 70 80 90 100 110 120 130 14
 Diameter class (cm)
 Managed forest Virgin forest
 Growin
 g 
stoc
 k 
 
(m
 3 /ha
 )
 0
 10
 20
 30
 40
 50
 60
 70
 80
 90
 100
 110
 120
 130
 140
 150
 Figure 6. Distribution of growing stock per diameter classes in
 virgin forest and managed forest.
 Grmec? Mountain virgin forest in Bosnia 5
 Downloaded by [193.63.226.141] at 06:55 14 February 2013
  
forest recorded a lower value than the basal area of
 virgin forest. The results of the study show that in the
 virgin forest the trees with a diameter of up to 50-cm
 diameter classes accounts for 33% of the total basal
 area whilst those over 51 cm in diameter account for
 67%; in the managed forest, the proportion was 69%
 and 31%, respectively, of the total basal area.
 Growing stock
 The average volume of the growing stock in the virgin
 forest was 725.23m3/ha, of which silver fir rep-
 resented 14%, Norway spruce 0.8%, common beech
 76% and other broadleaves trees 9%. In the managed
 forests, the average volume of the growing stock was
 significantly lower ? 611.53 m3/ha (Table II).
 Figure 6 shows an almost normal Gaussian
 distribution, with a slight left-skewed shape,
 especially in the case of virgin forest. In the virgin
 forest, the highest volume was recorded in the 61- to
 90-cm diameter class, whereas in the managed forest
 it was recorded in the 31- to 60-cm diameter class,
 the distribution for the latter showing a lower pick.
 As shown in Figure 7, in the case of the virgin
 forest, common beech was dominant in all diameter
 classes, whilst in managed forest, silver fir was the
 Table III. Abundance of young crops in virgin forest and managed forest.
 Abundance in virgin forest Abundance in managed forest
 Tree species Height (cm) dbh (cm) Total Height (cm) dbh (cm) Total
 10?50 51?130 , 5 10?50 51?130 , 5
 Silver fir 51 55 82 188 3284 581 303 4168
 Norway spruce 677 102 4 783 884 101 202 1187
 Beech 27 8 4 39 859 606 936 2401
 Sycamore 337 207 8 552 2526 126 101 2753
 Ash 1519 70 4 1593 859 202 253 1314
 Elm 51 43 8 102 76 25 75 176
 Lime 67 121 23 211 0 0 0 0
 Rowan 0 0 0 0 682 177 25 884
 Total 2729 606 133 3468 9170 1818 1895 12,883
 Managed forest
 0
 20
 40
 60
 80
 100
 120
 140 lime
 rowan
 elm
 ash
 sycamore
 beech
 silver fir
 Virgin forest
 0
 20
 40
 60
 80
 100
 120
 140
 160
 10 20 30 40 50 60 70 80 90 100 110
 Diameter class (cm)
 lime
 elm
 ash
 sycamore
 beech
 Norway spruce
 silver fir
 Norway spruce
 Growing stock (
 m3
 /ha
 )
 Growing stock (
 m3
 /ha
 )
 10 20 30 40 50 60 70 80 90 100 110 120 130
 Diameter class (cm)
 Figure 7. Distribution of growing stock per tree species in virgin forest and managed forest.
 C? . Vis?njic? et al.6
 Downloaded by [193.63.226.141] at 06:55 14 February 2013
  
dominant species and showed a left-skewed distri-
 bution of volume by diameter classes. The higher
 proportion of silver fir and Norway spruce in
 managed forests is a clear result of the previous
 management, which favoured these more economi-
 cally valuable conifer species compared to beech, as
 was common practice in many European beech?
 conifer mixtures (Abrudan 2000). Norway spruce,
 common beech and sycamore maple were present in
 almost all diameter classes in the managed forest.
 Common beech, similar to silver fir, in the managed
 forest also showed a left-skewed distribution of
 volume by diameter classes, being present together
 with ash in almost all diameter classes in the virgin
 forest.
 Juvenile trees
 Table III presents the abundance of juvenile crops
 (dbh , 5 cm) per species and height, per hectare in
 the virgin forest and managed forest.
 In the studied forests, the most common species
 in the juvenile category were silver fir, sycamore, ash,
 common beech, Norway spruce and elm. Rowan
 seedlings were not found in virgin forest because no
 mature trees were present, whilst lime was present
 only in the upper standing layer of managed forest.
 Amongst the abundance of the juvenile crops,
 within the virgin forest, ash was the dominant species
 (55%) in the height class of 10?50 cm, followed by
 Norway spruce (25%) and sycamore. In the juvenile
 category, the least represented tree types were
 common beech, elm and silver fir. Silver fir and
 sycamore were found in a significantly higher
 number in the managed forest, with silver fir
 seedlings accounting for 35% and Sycamore maple
 for 27%. The least present juvenile crop in the
 managed forest in the 10?50 cm height category was
 elm. In the height class of 51?130 cm, this was not
 so. In virgin forest, the highest number of juvenile
 crops belonged to sycamore (34%), then lime and
 Norway spruce and the lowest presence was recorded
 of common beech. On the other hand, in the same
 height class in managed forest, the highest number of
 juvenile crops belonged to common beech (33%)
 and silver fir (32%), and the lowest presence was for
 elm. For juvenile crops with the dbh less than 5 cm
 and the height greater than 130 cm, in virgin forest
 the highest number belonged to silver fir (62%) and
 lime (17%), and in managed forest it was common
 beech (49%) and silver fir (16%). This could be a
 direct consequence of the close-to-nature approach
 adopted in managed forest, which favoured common
 beech and silver fir as the main component species of
 the local natural forests (Ciancio and Nocentini
 2011).
 Some of the early reports on Bosnian virgin
 forests mention competitive interactions between
 and within tree species. For instance, the observed
 shade tolerance of seedlings and saplings of A. alba
 Mill. and F. sylvatica L. was used to explain their
 ability to outcompete light-demanding species
 Table IV. Total volume of dead wood in virgin forest and managed forest.
 Form of dead wood Virgin forest Managed forest
 Type of decomposition (m3/ha) Total Type of decomposition (m3/ha) Total
 Freshly d.w. Rotten Decomposed Freshly d.w. Rotten Decomposed
 Lying d.w. 0 80.97 22.42 103.39 2.45 2.79 26.22 31.46
 Standing d.w. 0 30.09 4.19 34.28 0.09 2.58 0.43 3.10
 Broken h . 1.3 cm 0 16.87 0.98 17.85 0 1.96 0.52 2.48
 Stump h , 130 cm 0 0.60 0.88 1.48 0.60 1.90 12.98 15.48
 Assortment 0 0 0 0 1.14 3.08 4.92 9.14
 Total 0 128.53 28.47 157.00 4.28 12.31 45.07 61.66
 Note: d.w., dead wood.
 Table V. Total volume of living and dead wood in virgin forest and managed forest.
 Tree species Virgin forest Managed forest
 Living wood Dead wood Total Living wood Dead wood Total
 Conifers m3/ha 105.1 40.9 146.9 451.3 43.9 495.2
 % 71 29 100 91 9 100
 Broadleaves m3/ha 620.2 116.1 736.3 160.0 17.7 177.7
 % 84 16 100 90 10 100
 Total m3/ha 725.2 157.0 882.2 611.3 61.6 672.9
 % 82 18 100 91 9 100
 Grmec? Mountain virgin forest in Bosnia 7
 Downloaded by [193.63.226.141] at 06:55 14 February 2013
  
(Cermak 1910). These facts were studied and very
 evident in the investigated virgin forests, where stand
 structures often differed from those found in
 managed forests.
 Dead wood
 The presence of dead wood is important to biological
 diversity within a forest, and it is a critical factor for
 the development of particular species like mush-
 rooms, lichens, mosses, arthropods, rodents and
 birds. On the basis of research focused on optimising
 tree stand diversity, it has been suggested that the
 proportion of dead wood in relation to the total wood
 mass should be around 20?25% (Siitonen 2001;
 Alexander 2003; Diaci et al. 2010; Keeton et al.
 2010) in unmanaged forests. In managed forests that
 share is much lower, at around 1?10% (Meyer
 1999b; Stancioiu and O?Hara 2006).
 Table IV shows the structure of dead wood per
 form of occurrence and type of decomposition,
 whilst Table V presents the total volume of living and
 dead wood, by conifers and broadleaves, in both
 virgin forest and managed forest.
 In virgin forest, no fresh dead wood was found
 due to the absence of any silvicultural operations or
 recent natural disturbance. The dead wood present
 was in the form of lying wood (due to wind-throws
 and parts of broken trees), with rotten dead wood
 accounting for 81% of the dead wood present. In the
 managed forest, the most common dead wood was
 represented by wood left behind after silvicultural
 interventions (small and even larger branches as a
 result of felling). It is important to notice that a
 significant proportion of dead wood (15%) is
 represented by various wood assortments left in the
 forest after felling (15%). Other studies in similar
 forests in Austria and Slovakia describe variation in
 the rate of dead wood due to disturbance events
 (Splechtna and Gratzer 2005; Kucbel et al. 2010). In
 support of this, dendroecological reconstructions of
 disturbance history in old-growth fir?beech forests
 in Slovenia and Austria found peaks in the
 disturbance chronology likely related to past storm
 events that caused intermediate levels of mortality
 (Splechtna et al. 2005; Nagel et al. 2007), and direct
 observations of the storm consequences were made
 in such old-growth stands (Nagel & Diaci 2006).
 Total wood volume in the investigated virgin
 forest was 882.2 m3/ha. Of this wood volume, living
 wood represented 725.2 m3 (82%) and dead wood
 accounted for 157.0 m3 (18%). In virgin forest, 29%
 of the total conifers volume is represented by dead
 wood whilst the proportion is much smaller for
 broadleaves (16%). These volume proportions of
 dead wood for conifers and broadleaves are close to
 those described in studies carried out in other
 European forest reserves (Saniga & Schu?tz 2001;
 Christensen et al. 2005; Pasierbek et al. 2007; Mu?ller
 and Bu?tler 2010). In managed forest, the total stock
 of living and dead wood was 672.9 m3/ha. Of this
 wood volume, living wood represented 611.3 m3
 (91%) and dead wood 61.6 m3 (9%).
 Conclusion
 This study identified that the number of trees in
 managed forest was significantly higher (almost 2.5
 times) than in virgin forest. The distribution of the
 number of trees, per diameter classes, had a
 decreasing trend, but with a different shape in virgin
 forest compared to the managed stands. The
 distribution of growing stock by diameter classes
 was different in virgin forest compared to the
 managed stands; for lower diameter classes (up to
 50 cm), the stock volume in virgin forest was half that
 of the volume of managed forest, whilst for higher
 diameter classes (over 51 cm), the cumulated volume
 of the growing stock was two times higher in the
 virgin forest than in the managed stands. The
 juvenile crops had a significantly lower presence in
 virgin forest compared to managed forest (almost
 four times). Also, the study identified a larger volume
 of dead wood in the virgin forest than in managed
 forest. The volume of dead wood in the investigated
 virgin forest was slightly lower than the volume
 reported in other similar studies, whilst in the
 managed forest under the group selection system the
 proportion of dead wood was similar to the one
 reported in the existing literature.
 Forest structure cannot be considered as static,
 and natural forest structure cannot be limited to a
 stable environment (Boncina 2000). Thus, the
 structure of a forest at any given moment in time is
 always determined by current environmental factors
 (including management) as well as of the natural
 processes underway in the forest (Schnitzler and
 Borlea 1998; Boncina 1999; Travaglini et al. 2012).
 The study revealed that the group selection system
 still had a significant impact on the stand structure
 and regeneration, compared to virgin forests, and
 therefore in cases where virgin remnant forest is
 surrounded by managed forest it might be prudent
 for the management to let the natural processes
 dominate and where possible reduce the intervention
 in the course of forest development.
 Notes
 1. Email: c.visnjic@sufasa.org
 2. Email: solakovic@bih.net.ba
 3. Email: f.mekic@sufasa.org
 4. Email: balicbesim@yahoo.com
 5. Email: mirzad@bih.net.ba
 C? . Vis?njic? et al.8
 Downloaded by [193.63.226.141] at 06:55 14 February 2013
  
6. Email: s.gurda@sufasa.org
 7. Email: florin.ioras@googlemail.com
 8. Email: jegaratnasingam@yahoo.com
 References
 Abrudan IV. 2000. Natural and semi-natural mixed stands in the
 Romanian Carpathians. In: Price MF, Butt N, editors. Forests
 in sustainable mountain development. A State of Knowledge
 Report for 2000. CABI Publishing. pp. 208?209.
 Abrudan IV, Mather RA. 1999. The influence of site-factors on the
 composition of semi-natural mixed-species stands of beech
 (Fagus sylvatica L.), silver fir (Abies alba Miller) and Norway
 spruce (Picea abies (L.) Karsten) mixtures in the Upper
 Draganul Watershed in North-West Romania. Forestry 72(2):
 87?93.
 Albrecht L. 1990. Grundlagen. Ziele und Methodik der
 waldo?kologischer Forschung in Naturwaldreservaten. Natur-
 waldreservate in Bayern. Schriftenreihe des Bayerischen
 Staatministeriums fu?r Erna?hrung, Landwirtschaft und Forsten
 gemeinsam mit dem Lehrstuhl fu?r Landschaftechnik, Band I,
 Mu?nchen : 221.
 Alexander KNA. 2003. British saproxilic invertebrate fauna.
 Proceedings of the second pan-European conference on
 saproxilic beetles. London:.
 Anic I, Mikac S. 2008. Struktura, tekstura i podmladivanje
 Dinarske bukovo-jelove pras?ume C? orkova Uvala S?umarski list
 Zagreb 11-12, CXXXII., pp. 505?515.
 Anonymous. 2006. Statistical Yearbook of the Federation of
 Bosnia and Herzegovina. Sarajevo:.
 Beus V, Vojnikovic? S. 2002. Floristical characteristics of the virgin
 forest of beech and fir in Ravna vala on mountain Bjelas?nica,
 Rasprave IV. Razreda SAZU, Ljubljana, XLIII 3: 63?78.
 Boncina A. 1999. Stand dynamics of the virgin forest Rajhenavski
 Rog (Slovenia) during the past century. In: Diaci J, editor.
 Virgin forests and forest reserves in Central and East European
 countries. Ljubljana: COST E4 Management Committee and
 Working Groups. pp. 95?110.
 Boncina A. 2000. Comparison of structure and biodiversity in the
 Rajhenav virgin forest remnant and managed forest in the
 Dinaric region of Slovenia. Global Ecol Biogeogr 9(3):
 201?211.
 Christensen M, Hahn K, Mountford EP, Odor P, Standovar T,
 Rozenbergar D, et al. 2005. Dead wood in European beech
 (Fagus sylvatica) forest reserves. For Ecol Manage 210:
 267?282.
 Ciancio O, Nocentini S. 2011. Biodiversity conservation and
 systemic silviculture: Concepts and applications. Plant Biosyst
 145(2): 411?418.
 Dautbasic M, Ioras F. 2010. Conservation of forest biodiversity in
 Central and Eastern Europe - A case study of Romania and
 Bosnia Herzegovina. International Forestry Review Proceed-
 ings of XXIII IUFRO World Congress, 23?28 August 2010.,
 p.103.
 Dautbasic M, Ioras F, Abrudan R, IV, atnasingam J. 2010.
 Funding biodiversity protection in Central and Eastern Europe
 - A case study of Bosnia Herzegovina. Not Bot Hort Agrobot
 Cluj 38(1): 252?256.
 Ioras F, Abrudan IV, Dautbasic M, Avdibegovic M, Gurean D,
 Ratnasingam J. 2009. Conservation gains through HCVF
 assessments in Bosnia-Herzegovina and Romania. Biodivers
 Conserv 18(13): 3395?3406.
 Kucbel S, Jaloviar P, Saniga M, Vencurik J, Klimas? V. 2010.
 Canopy gaps in an old-growth fir-beech forest remnant of
 Western Carpathians. Eur J Forest Res 129: 249?259.
 Leibundgut H. 1945. Waldbauliche Untersuchungen u?ber den
 Aufbau von Plenterwa?lden. Mittgl. D. Schw. A.f.d.f.V.w.
 Leibundgut H. 1993. Europa?ische Urwa?lder. Bern, Stuttgart:
 Verlag Paul Haupt.
 Lund HG. 2002. Coming to terms with politicians and definitions.
 IUFRO Occasional Paper 14: 23?24.
 Matic? V. 1977. Metodika izrade s?umskoprivrednih osnova za s?ume
 u drus?tvenoj svojini na podruc?ju BiH. S?umarski fakultet i
 Institut za s?umarstvo u Sarajevu, Posebna izdanja broj 12,
 Sarajevo. p.254.
 Matic V, Drinic P, Stefanovic V, C? iric M. 1971. Stanje s?uma u SR
 Bosni i Hercegovini prema inventuri s?uma na velikim
 povrs?inama u 1964?1968. godini S?umarski fakultet i Institut
 za s?umarstvo u Sarajevu, Posebna izdanja broj 7, Sarajevo.
 p.639.
 Meskovic E. 2006. Analiza strukture prirodnog podmaltka u
 pras?umskom rezervatu ?Mac?en do? (BiH) Jasrebarsko: Institut
 za s?umarstvo., pp. 1?9.
 Meyer P. 1999a. Totholzuntersuchungen in nordwestdeutschen
 Naturwa?ldern Methodik und erste Ergebnisse Forstw Cbl 118,
 167?180. Berlin:.
 Meyer P. 1999b. Bestimmung der Waldentwicklungsphasen und
 der Texturdiversita?t in Naturwa?ldern. Allg. Forst-u. J.-Ztg. 170
 (10?11): 203?211.
 Meyer P, Pogoda P. 2001. Entwicklung der roemlichen
 Strukturdiversitaet in nordwestdeutschen Naturwaeldern.
 Allg. Forst- u J.-Ztg. 172(12): 213?220.
 Meyer P, Tabaku V, von Luepke B. 2003. Die Struktur albanischer
 Rotbuchen - Urwaelder - Ableitungen fuer eine naturnahe
 Buchenwirtschaft. Forstw Cbl 122: 47?58.
 Mu?ller J, Bu?tler R. 2010. A review of habitat thresholds for dead
 wood: A baseline for management recommendations in
 European Forests. J Forest Res 129(6): 981?992.
 Nagel TA, Diaci J. 2006. Intermediate wind disturbance in an old-
 growth beech-fir forest in Southeastern Slovenia. Can J Forest
 Res 36: 629?638.
 Nagel TA, Levanic T, Diaci J. 2007. A dendroecological
 reconstruction of disturbance in an old- growth Fagus-Abies
 forest in Slovenia. Ann Forest Sci 64: 891?897.
 Parviainen J. 2005. Virgin and natural forests in the temperate
 zone of Europe. Joensuu: Finnish Forest Research Institute.
 p.18.
 Parviainen J, Kassioumis K, Bucking W, Hochbichler E, Paivinen
 R, Little D. 2000. Final report summary: Mission, goal,
 outputs, linkages, recommendations and partners. In: Euro-
 pean Commission, editor. EUR 19550 ? COST Action E4 ?
 Forest reserves research network. Luxembourg: Office for
 Official Publication of the European Communities. pp. 9?38.
 Pasierbek T, Holeksa J, Wilczek Z, Z? ywiec M. 2007. Why the
 amount of dead wood in Polish forest reserves is so small? Nat
 Conserv 64: 65?71.
 Pintaric? K. 1978. Urwald Peruc?ica als naturlicher Forschungsla-
 boratorium. Muenchen 24: 702?707.
 Pintaric K. 1999. Forestry and forest reserves in Bosnia and
 Herzegovina. In: Diaci J, editor. Virgin forests and forest
 reserves in central and east European countries: History,
 present status and future development. Proceedings of the
 invited lecturers? reports presented at the COST E4 Manage-
 ment Committee and Working Groups meeting in Ljubljana,
 Slovenia, 25?28 April 1998. Ljubljana: Biotechnical Faculty,
 Department of Forestry and Renewable Forest Resources.
 pp. 1?15.
 Pintaric K, Izetbegovic S. 1980. Prouc?avanje metoda obnove u
 bukovim s?umama i mjes?ovitim s?umama bukve, jele i smrc?e.
 Radovi S?umarskog fakulteta i Instituta za s?umarstvo u
 Sarajevu, knj. 25: 3.
 Saniga M, Schu?tz JP. 2001. Dynamic changes in dead wood share
 in selected beech virgin forests in Slovakia within their
 development cycle. J Forest Sci 47(12): 557?565.
 Grmec? Mountain virgin forest in Bosnia 9
 Downloaded by [193.63.226.141] at 06:55 14 February 2013
  
Schnitzler A, Borlea F. 1998. Lessons from natural forests as keys
 for sustainable management and improvement of naturalness
 in managed broadleaved forests. Forest Ecol Manage 109
 (1?3): 293?303.
 Sebastia M-T, Casalas P, Vojnikovic? S, Bogunic? F, Beus V. 2005.
 Plant diversity and soil properties in pristine and managed
 stand from Bosnian mixed forests. Forestry 78(3): 297?303.
 S?GD Unsko-sanske s?ume. 2010. FMP ? Forest Management Plan
 for 2010. Bosanska Krupa:.
 Siitonen I. 2001. Forest management, coarse woody debris and
 saproxilic organisms: Fennoscandian boreal forests stand in
 Northeast Sweden. Biol Conserv 45: 169?178.
 Splechtna BE, Gratzer G. 2005. Natural disturbances in Central
 European forests: Approaches and preliminary results from
 Rothwald. Austria. For Snow Lands Res 79: 57?67.
 Splechtna BE, Gratzer G, Black BA. 2005. Disturbance history of
 a European old-growth mixed-species forest - A spatial dendro-
 ecological analysis. J Veg Sci 16: 511?522.
 Stancioiu PT, O?Hara KL. 2006. Leaf area and growth efficiency
 of regeneration in mixed species, multiaged forests of the
 Romanian Carpathians. Forest Ecol Manage 222(1?3):
 55?66.
 Stefanovic V. 1988. Pras?umski rezervati Jugoslavije, dragulji
 iskonske prirode. Biolos?ki list, Sarajevo 9?10: 1?5.
 Travaglini D, Paffetti D, Bianchi L, Bottacci A, Bottalico F,
 Giovannini G, et al. 2012. Characterization, structure and
 genetic dating of an old-growth beech-fir forest in the northern
 Apennines (Italy). Plant Biosyst 146(1): 175?188.
 Visnjic C? , Vojnikovic S, Ioras F, Dautbasic M, Abrudan Gure, IV,
 an D, et al. 2009. Virgin status assessment of Pljes?evica Forest
 in Bosnia-Herzegovina. Not Bot Hort Agrobot Cluj 37(2):
 22?27.
 C? . Vis?njic? et al.10
 Downloaded by [193.63.226.141] at 06:55 14 February 2013