Investigation of Selected Wood Properties and the Suitability for Industrial Utilization of Acacia Seyal Var

Home , Balanites, Bandsaw box, Caragana korshinskii, Guaiacum officinale, Turraeanthus africana

Fakultät Umweltwissenschaften - Faculty of Environmental Sciences

Investigation of selected wood properties and the suitability for industrial utilization of Acacia seyal var. seyal Del and Balanites aegyptiaca (L.) Delile grown in different climatic zones of Sudan

Dissertation to achieve the academic title Doctor rerum silvaticarum (Dr. rer. silv.)

Submitted by MSc. Hanadi Mohamed Shawgi Gamal born 23.09.1979 in Khartoum/Sudan

Referees:

Prof. Dr. Dr. habil. Claus-Thomas Bues, Dresden University of Technology Prof Dr. Andreas Roloff, Dresden University of Technology Prof. Dr. Dr. h.c. František Hapla, University of Göttingen

Dresden, 07.02.2014

Acknowledgement

First of all, I wish to praise and thank the god for giving me the strength and facilitating things throughout my study.

I would like to express my deep gratitude and thanks to Prof. Dr. Dr. habil. Claus-Thomas Bues , Chair of Forest Utilization, Institute of Forest Utilization and Forest Technology, Dresden University of Technology, for his continuous supervision, guidance, patience, suggestion, expertise and research facilities during my research periods in Dresden. He was my Father and Supervisor, I got many advices from his side in the scientific aspect as well as the social aspects. He taught me how to be a good researcher and gives me the keys of the wood science.

I am grateful to Dr. rer. silv. Björn Günther for his help and useful comments throughout the study. He guides me in almost all steps in my study. Dr.-Ing. Michael Rosenthal was also a good guide provided many advices, thanks for him. I would also like to express my thanks to Frau Antje Jesiorski, Frau Liane Stirl, Dipl.- Forsting. Herrn Ernst Bäucker, and Herrn Ing. (FH) Harald Kirchner for their good cooperation and help during the period of my study.

I am also indebted to University of Khartoum for giving me this opportunity to study in Germany and to the Ministry of Higher Education and Scientific Research, Sudan for the scholarship. My thanks also extended to GFF (Gesellschaft Freunde und Förderer der TU Dresden) as well as DAAD (Deutscher Akademischer Austausch Dienst / German Academic Exchange Service) for their kindly support.

Very special thank to my husband Majdaldin for his support and patience during my study. Without his support I would not be able to finish my work. The last and not least, I would like to express my gratitude to the most important two persons in my life, my parents, they helped me a lot by their prays and advices.

Abstract

Sudan is endowed with a great diversity of tree species; nevertheless the utilization of wood resources has traditionally concentrated on a few species only. Despite of the richness of Sudan in most of basic factors required to establish forest based industries it still almost entirely dependent on imports to satisfy its needs of the products of such industries. There is an urgent need to assess the suitability of the local fibrous raw materials for industrial utilization, this would not only reduce imports, but they would also provide an economic incentive to the forestry and industrial sectors of Sudan.

Sudan has a wide variation of climatic zones, thus; great variations are expected in the anatomical and physical properties between and within species grown in each zone. This variation needs to be fully explored in order to suggest best uses for the species.

The present study was carried out to assess the suitability of Acacia seyal var. seyal Del and Balanites aegyptiaca (L.) Delile wood for pulp and paper making (PPM) and flooring industry, as well as to investigate the effect of rainfall zones on selected wood properties. For this purpose, a total of thirty trees per species were collected from four states in Sudan, namely: Blue Nile, North Kordofan, South Kordofan and White Nile. The study areas located in two precipitation zones. Zone one with 273 mm mean annual rainfall, and zone two with 701 mm mean annual rainfall. Wood samples in form of disc were obtained from two heights within each tree, which are 10 % and 90 % from the tree merchantable height. Anatomical, physical and mechanical investigations were conducted in order to test the wood properties of the study species. The studied anatomical properties were: fibre and vessel diameter, lumen diameter and wall thickness. In addition to fibre length and three fibre derived values, namely: flexibility coefficient, Runkel ratio and slenderness ratio. The trend of fiber length from pith to bark was determined. The anatomical composition was described. Wood density was investigated as a main physical property. Basic density as well as air dry density were measured in the current study. Additionally, the density was measured using X-ray densitometry method in order to assess its suitability as a valid tool for the study species density determination. The trend of wood basic density from pith to bark was also determined. Brinell hardness strength was measured in the transverse and radial sections.

According to the study results, the fibre length of both species considered as medium (900 - 1600 µm). However, Acacia seyal has longer fiber. Acacia seyal wood density considered heavy (≥ 720 kg/m³) while that of Balanites aegyptiaca is medium (500 - < 720 kg/m³). Depending upon the mean values of hardness strength in transverse as well as radial sections,

Abstract

the wood of both species can be classified as very hard (up to 146 N/mm 2 hardness strength). Fibre length and wood density for both species followed the increase trend from the pith to the bark. The X-ray densitometry technique is considered as a valid tool for wood density determination for both species. For each species, some wood properties (in mature wood) were significantly affected by the water stress in the drier zone. For instance, Acacia seyal fibre length was negatively affected, while vessel wall thickness, basic density as well as hardness strength of the radial section were positively affected. In case of Balanites aegyptiaca the following properties were affected: vessel dimensions (negatively) and basic density (positively). However, the water stress did not affect Acacia seyal fibre and vessel diameter and lumen diameter, fiber wall thickness, flexibility coefficient, Runkel ratio and hardness strength in transverse section. Balanites aegyptiaca fibre characteristics and hardness strength did not show any response to water stress as well.

In general, the overall wood properties of the study species considered compatible for PPM and flooring industry. However, trees growing in the more humid zone are preferable for both industries, due to their lower wood density and longer fibres in case of Acacia seyal and lower density in case of Balanites aegyptiaca.

III

Kurzfassung

Im Sudan kommen eine Vielzahl verschiedener Baumarten vor. Die Holzverwendung konzentriert sich jedoch traditionell nur auf wenige Arten. Trotz des sudanesischen Reichtums an forstlichen Ressourcen ist die holzverarbeitende Industrie nahezu vollständig von Holzimporten abhängig. Es ist dringend erforderlich, die Eigenschaften von lokal verfügbaren Faserrohstoffen im Hinblick auf ihre mögliche industrielle Nutzung zu untersuchen. Dies könnte nicht nur die Holzimporte verringern, sondern könnte darüber hinaus auch einen wirtschaftlichen Anreiz für die Holz- und Forstwirtschaft im Sudan liefern.

Der Sudan weist eine Vielzahl verschiedener Klimazonen auf. Dies lässt die Vermutung zu, dass eine große Variabilität der anatomischen und physikalischen Eigenschaften zwischen und innerhalb der in den Zonen vorkommenden Arten besteht. Um die optimale Nutzung dieser Arten zu ermöglichen, müssen diese Variationen umfassend untersucht werden.

In der vorliegenden Arbeit wurde die Eignung des Holzes von Acacia seyal var . seyal Del and Balanites aegyptiaca (L.) Delile für die Zellstoff- und Papierindustrie sowie für die Fußbodenindustrie bewertet. Des Weiteren wurde der Einfluss von Niederschlag auf ausgewählte Holzeigenschaften untersucht. Für die Untersuchung wurden insgesamt 30 Bäume je Baumart aus vier verschiedenen Bundesstaaten (Blauer Nil, Nord-Kordofan, Süd- Kordofan, Weißer Nil) beprobt. Die Untersuchungsgebiete befinden sich in zwei verschiedenen Niederschlagszonen. Die Zonen weisen einen jährlichen Gesamtniederschlag von 273 mm (Zone 1) und 701 mm (Zone 2) auf. Pro Baum wurden zwei Stammscheiben entnommen. Die Entnahme erfolgte in zwei unterschiedlichen Höhen, jeweils bei 10% und 90% der nutzbaren Stammhöhe. Um die Holzeigenschaften der untersuchten Holzarten zu bewerten, wurden anatomische, physikalische und mechanische Untersuchungen durchgeführt. Folgende anatomische Eigenschaften wurden untersucht: Faser- und Gefäßdurchmesser, Gefäßlumendurchmesser, Zellwandstärke, Faserlänge. Außerdem wurden faserbezogene Werte wie der Flexibilitätskoeffizient, die Runkel-Zahl und der Schlankheitsgrad abgeleitet. Für die Faserlänge wurde der Trendverlauf vom Mark bis zur Rinde bestimmt. Des Weiteren erfolgte die Beschreibung des holzanatomische Aufbau bzw. der holzanatomische Merkmale. Als wichtigste physikalische Größe wurde die Holzdichte untersucht. Sowohl die Raumdichte als auch die Normal-Rohdichte wurden in der vorliegenden Arbeit bestimmt. Zusätzlich erfolgten röntgendensitometrische Dichtemessungen, um die Verwendbarkeit dieses Verfahrens für die untersuchten Holzarten

Kurzfassung zu bewerten. Darauf aufbauend wurde der radiale Dichteverlauf ausgehend vom Mark bis zur Rinde bestimmt. Die Brinell-Härte wurde auf den Radial- und Querschnittsflächen gemessen.

Die Ergebnisse der Untersuchung zeigten für beide Arten mittlere Faserlängen (900 – 1600 µm), wobei Acacia seyal größere Faserlängen aufweist. Während das Holz von Acacia seyal eine hohe Dichte (≥ 720 kg/m³) besitzt, liegen die Dichtewerte von Balanites aegyptiaca im mittleren Bereich (500 – < 720 kg/m³). Aufgrund der bestimmten mittleren Brinell-Härten auf den Radial- und Querschnitten, kann das Holz beider Arten als sehr hart (Härte bis zu 146 N/mm 2) klassifiziert werden. Sowohl die Faserlänge als auch die Holzdichte weisen bei beiden Arten, ausgehend vom Mark bis zur Rinde, einen steigenden Trendverlauf auf. Das Verfahren der Röntgendensitometrie konnte bei beiden Arten erfolgreich für die Dichtebestimmung angewandt werden. Bei beiden Holzarten werden bestimmte Holzeigenschaften signifikant durch den in der trockneren Niederschlagszone herrschenden Trockenstress beeinflusst. Dies zeigt sich bei Acacia seyal in kürzeren Faserlängen, dickeren Zellwänden, in höheren Dichtewerten und in einer höheren Brinell-Härte auf den Radialflächen. Bei Balanites aegyptiaca verringerten sich die Gefäßgrößen und die Dichtewerte nahmen zu. Es zeigte sich jedoch, dass Wasserstress bei Acacia seyal keinen Einfluss auf die Faser- und Gefäßdurchmesser, den Gefäßlumendurchmesser, die Zellwandstärke der Fasern, den Flexibilitätskoeffizienten, die Runkel-Zahl und die Brinell- Härte auf der Querschnittsfläche hat. Bei Balanites aegyptiaca konnte sowohl für die Fasereigenschaften als auch für die Brinell-Härte keine Beeinflussung durch Trockenstress nachgewiesen werden.

Insgesamt kann gesagt werden, dass die Verwendbarkeit der beiden untersuchten Holzarten sowohl für die Zellstoff- und Papierindustrie als auch für die Fußbodenindustrie aufgrund der ermittelten Holzeigenschaften gegeben ist. Für beide Industriezweige sollte jedoch das Holz aus der niederschlagsreicheren Zone bevorzugt genutzt werden, da hier beide Holzarten geringere Holzdichten aufweisen und Acacia seyal zudem noch längere Fasern besitzt.

Table of contents

1 Introduction ...... 1 1.1 Background ...... 1 1.2 Problem setting ...... 3 1.3 Research objectives ...... 4 1.3.1 Main objective ...... 4 1.3.2 Specific objectives ...... 4 1.4 Research hypotheses ...... 5

2 Literature Review ...... 6 2.1 Wood properties and suitability for industrial utilization ...... 6 2.1.1 Anatomical properties ...... 6 2.1.1.1 Fibres characteristics ...... 7 - Fibre dimensions ...... 7 - Fibre derived values ...... 11 2.1.1.2 Vessels characteristics ...... 13 2.1.2 Physical properties ...... 16 2.1.2.1 Wood density ...... 16 2.1.2.2 Hardness strength ...... 19 2.1.3 Relations between anatomical composition and wood density ...... 22 2.1.3.1 Fibre wall thickness ...... 23 2.1.3.2 Cell frequency ...... 23 2.1.4 Wood properties variation ...... 24 2.1.4.1 Radial variation of fibres characteristics ...... 24 2.1.4.2 Radial variation of vessels characteristics ...... 27 2.1.4.3 Radial variation of density ...... 28 2.1.4.4 Vertical variation of density ...... 30 2.1.4.5 Trees, forests, and regions variation ...... 31 2.2 Influence of different climatic growth conditions on wood properties ...... 32 2.2.1 Water supply ...... 33 2.2.1.1 Anatomical properties ...... 33 - Vessels ...... 34 - Fibres ...... 36 2.2.1.2 Physical properties ...... 37 2.2.2 Air temperature ...... 38

Table of contents

2.2.3 Other factors ...... 38 2.3 Description, distribution and uses of the study species ...... 41 2.3.1 Acacia seyal var. seyal Del...... 41 2.3.1.1 Taxonomy and nomenclature ...... 41 2.3.1.2 Botanical description ...... 40 2.3.1.3 Wood properties ...... 42 2.3.1.4 Distribution and uses ...... 44 2.3.2 Balanites aegyptiaca (L.) Delile ...... 47 2.3.2.1 Taxonomy and nomenclature ...... 47 2.3.2.2 Botanical description ...... 47 2.3.2.3 Wood properties ...... 48 2.3.2.4 Distribution and uses ...... 52

3 Material ...... 54 3.1 Study area ...... 54 3.1.1 Zone one (with 273 mm mean annual rainfall) ...... 55 3.1.2 Zone two (with 701 mm mean annual rainfall) ...... 58 3.2 Trees selected ...... 61

4 Methods ...... 64 4.1 Sampling ...... 64 4.2 Sample processing ...... 71 4.2.1 Anatomical properties investigations ...... 71 4.2.1.1 Maceration ...... 72 4.2.1.2 Softening ...... 73 4.2.2 Density investigations ...... 75 4.2.2.1 Basic density ...... 75 4.2.2.2 Air-dry density ...... 76 4.2.2.3 X-ray densitometry ...... 77 Sample preparation ...... 77 - Fibre angle measurement ...... 77 - Sample sawing and conditioning ...... 77 X-ray of the sample material ...... 78 Analysis of X-ray film ...... 78 Correction factors for density equivalent values calculation ...... 80

VII

Table of contents

Possible source of error ...... 81 4.2.3 Hardness strength investigation ...... 81 4.3 Statistical analysis ...... 83 4.3.1 Data entry, cleaning and test of normality ...... 83 4.3.2 Independent-sample T-test ...... 84 4.3.3 Analysis of variance (One way ANOVA) ...... 85 4.3.4 Post-Hoc tests in ANOVA ...... 85 4.3.5 Descriptive statistics tests ...... 85 4.3.6 Trend line/ regression analysis ...... 86

5 Results and Discussion ...... 87 5.1 Anatomical properties ...... 88 5.1.1 Anatomical composition ...... 88 5.1.1.1 Acacia seyal var . seyal ...... 88 5.1.1.2 Balanites aegyptiaca ...... 92 5.1.2 Fibres characteristics ...... 95 5.1.2.1 Radial variation ...... 95 5.1.2.2 Trees, forests and regions variation ...... 104 5.1.2.3 Summarizing description ...... 108 5.1.3 Vessels characteristics ...... 115 5.1.3.2 Radial variation ……………...... ….....…....………………..…… 115 5.1.3.2 Trees, forests and regions variation …...…....…………………… 117 5.1.3.2 Trees, forests and regions variation ...... 117 5.1.3.3 Summarizing description ...... 119 5.2 Density ...... 120 5.2.1 Basic density ...... 120 5.2.1.1 Vertical variation ...... 120 5.2.1.2 Radial variation ...... 122 5.2.1.3 Trees, forests and regions variation ...... 128 5.2.1.4 Summarizing description ...... 130 5.2.2 Air dry density ...... 131 5.2.2.1 Vertical variation ...... 131 5.2.2.2 Radial variation ...... 132 5.2.2.3 Trees, forests and regions variation ...... 133 5.2.2.4 Summarizing description ...... 135 5.2.3 Density achieved by X-ray densitometry ...... 135

VIII

Table of contents

5.2.3.1 Correction factor ...... 135 5.2.3.2 Radial variation ...... 136 5.2.3.3 Trees, forests and regions variation ...………………………….… 137 5.2.3.4 Summarizing description ...………………………………………. 138 5.2.4 Comparison of density values derived from X-ray technique and air dry gravimetric method ...... 139 5.3 Hardness strength ...... 141 5.3.1 Vertical variation ...... 141 5.3.2 Radial variation ...... 142 5.3.3 Trees, forests and regions variation ...... 143 5.3.4 Summarizing description ...... 144 5.4 Variation of the wood properties due to climatic zones ...... 146 5.4.1 Anatomical properties variation ...... 146 5.4.1.1 Variation of fibres characteristics ...... 147 5.4.1.2 Variation of vessels characteristics ...... 149 5.4.2 Basic density variation ...... 153 5.4.3 Hardness strength variation ...... 154 5.4.4 Summarizing description ...... 155 5.5 The suitability of the study species for pulp and paper and flooring industries ... 156 5.5.1 Suitability for pulp and paper industry ...... 156 5.5.2 Suitability for flooring industry ...... 161 5.6 Proving hypotheses ...... 164

6 Conclusions and Outlook ...... 167 6.1 Conclusions ...... 167 6.1.1 Anatomical properties ...... 167 Anatomical composition ...... …….. 167 Fibres characteristics ...... 167 Vessels characteristics ...... 168 6.1.2 Wood density ...... 168 Basic density ...... 168 Air dry density ...... 168 Density achieved by X- ray densitometry ...... 168 6.1.3 Hardness strength ...... 169 6.1.4 Variation of the wood properties due to climatic zones ...... 169 6.1.5 Suitability of the study species for pulp and paper and flooring industries. 169

Table of contents

6.2 Research outlook ...... 170

7 Summary ...... 171 7.1 Anatomical properties ...... 171 Anatomical composition ...... 171 Fibres characteristics ...... 172 Vessels characteristics ...... 172 7.2 Wood density ...... 173 Basic density ...... 173 Air-dry density ...... 173 Density achieved by X- ray densitometry ...... 173 7.3 Hardness strength ...... 174 7.4 Variation of the wood properties due to climatic zones ...... 174 7.5 Suitability of the study species for pulp and paper and flooring industries ...... 174

8 Registers ...... 176 8.1 Sources ...... 176 8.2 Figures ...... 198 8.3 Tables ...... 203 8.4 Definitions of selected terms in wood science ...... 207 8.5 Abbreviations ...... 209

9 Appendices ...... 210

Declaration ...... 211

Introduction

1 Introduction

1.1 Background

Sudan’s forests cover about 11% of its total area of 1,882,000 square kilometres (GAFAAR 2011) (Figure 1.1). These forests include a great number of tree species: more than 3156 species belonging to 1137 genera and 170 families (BROUN and MASSAY 1929, ANDREWS 1950, 1952, 1956; EL AMIN 1990). Despite the diversity of trees found in Sudan, only a few are widely used, and despite the richness of Sudan in most of the basic factors required to establish pulp and paper and flooring industry, it is still dependent on import to satisfy its needs for pulp and paper. There is currently no flooring industry in Sudan. It is therefore evident that there is an urgent need to assess the suitability of local fibrous raw materials for paper and flooring industries. Domestic sources of such industries would not only reduce imports but also provide an economic incentive to the forestry and industrial sectors of Sudan.

Figure 1.1 : Sudan forest cover. Adapted by the author from DAWELBAIT et al. (2006)

Sudan has a wide variation of rainfall, from 0 mm/annum in the northern deserts to over 1250 mm/annum towards the southern border of the country (see Figure 1.2). The ecosystem classification and the vegetation distribution closely follow the isohyets. The effect of - 1 -

Introduction

topographical changes and soil variation on vegetation zones is much less pronounced than that of rainfall levels (GAFAAR 2011).

Figure 1.2: Sudan precipitation map. Adapted by the author from SOURCE1

Wood is a raw material of variable structure. Wood properties differ between and within species (PANSHIN and DE ZEEUW 1980). Research on wood has substantiated that the climatic conditions of the area in which the species grows has a significant effect on the properties of the wood (ALVES and ANGYALOSSY-ALFONSO 2000, ALVES and ANGYALOSSY-ALFONSO 2002, WIMMER et al. 2002, ROQUE 2004, AL-KHALIFA et al. 2006, MOYA and FO 2008). Understanding the extent of variability of wood is important because the uses for each kind of wood are related to its characteristics; furthermore, the suitability or quality of wood for a particular purpose is determined by the variability of one or more of these characteristics, which affect its structure and hence its physical properties (PANSHIN and DE ZEEUW 1980). However, to use wood to its best advantage and most effectively in the different applications, specific characteristics or physical properties must be considered (MILLER 1999). The versatility of wood is demonstrated by a wide variety of products. This variety is a result of a spectrum of desirable physical characteristics or properties among the many species of wood. Ideally, the extent of this wide variation needs to be fully explored. In many cases, more than one property of wood is important to the end

- 2 -

Introduction

product. Efficient utilization dictates that species should be matched to end-use requirements through an understanding of their properties (SIMPSON and TENWOLDE 1999).

The anatomical, physical, and mechanical properties of wood are considered the fundamental basis for timber utilization. Studies of wood properties have a special significance in countries such as Sudan where only a few timbers are well known. This research will help to promote quality research on wood products by studying some anatomical properties, density and hardness strength of wood species grown in Sudan. The tree species selected for study in this research are Acacia seyal var . seyal Del. and Balanites aegyptiaca (L.) Delile, because of their wide distribution and easy growth in large areas in Sudan, both of them found in almost all rainfall zones. Their uses are primarily for charcoal, firewood and fuel wood. The scientific information about the wood properties of Acacia seyal var. seyal and Balanites aegyptiaca grown in Sudan are still limited; the little available information (described in chapter 2.3) was derived mostly from trees collected from one zone.

This research attempts to provide information on important anatomical, physical and mechanical wood properties of Acacia seyal var. seyal and Balanites aegyptiaca and their variation between different rainfall zones, as well as to assess the suitability of the selected species for pulp and paper and flooring industries.

1.2 Problem setting

Little information is available on the wood properties of the tree species growing in Sudan. Studies of the anatomical and physical properties of wood are of special significance in countries like Sudan, where only a few timbers are well known. Such studies can suggest uses, especially for woods that are not in commercial demand.

With the great variation of the climatic zones of Sudan, great variations are expected in wood properties between and within species growing in these zones. This variation needs to be fully explored in order to suggest the best uses for the species.

Sudan, like most developing countries, is almost entirely dependent on imports to satisfy its needs for pulp and paper. There is thus a need to evaluate the locally available raw materials as potential sources for a pulp and paper industry by examining their suitability for pulping to meet the increasing cultural and industrial needs and to save the expenditure of foreign currency.

- 3 -

Introduction

1.3 Research objectives

1.3.1 Main objective

The main objective is to examine the suitability of the selected species for pulp and paper production and for flooring industry by providing basic information about some anatomical, mechanical and physical wood properties of the study species, as well as to investigate the effect of rainfall zones on selected wood properties.

1.3.2. Specific objectives

There are 5 specific objectives of this study:

Part 1 : Investigation of some anatomical properties with importance to pulp and paper making:

∑ Fibre dimensions (length, diameter, lumen diameter and wall thickness) and its derived values (flexibility coefficient, Runkel ratio and slenderness ratio).

∑ The trend of fibre length from pith to bark.

∑ Vessel element diameter, lumen diameter and wall thickness.

Part 2 : Investigation of wood density as the main physical property:

∑ Basic density and its trend from pith to bark.

∑ Density achieved by X-ray densitometry, to assess the suitability of this technique for the study species density determination.

∑ Air dry density, necessary for the calibration of X-ray density.

Part 3 : Investigation of hardness strength as one important mechanical property:

∑ Brinell hardness strength in transverse section.

∑ Brinell hardness strength in radial section.

Part 4 : Investigation of the possible effect of rainfall zone (climatic) on wood properties, by comparing the studied wood properties of trees grown under lower and higher precipitation.

Part 5 : Assessment of the suitability of the study species for pulp and paper and flooring industries, by comparing their wood properties obtained from parts 1, 2, and 3 with

- 4 -

Introduction

the acceptable values for each of the selected industries and with some hardwood species in commercial use for such industries.

1.4 Research hypotheses

The following hypotheses are proposed:

Hypothesis 1: The study species have anatomical and physical properties that qualify them for pulp and paper making either alone or by mixing with other pulping species.

Hypothesis 2: The study species have physical and mechanical properties that qualify them for use in the flooring industry.

Hypothesis 3: The different rainfall zones have a significant effect on the anatomical, mechanical and physical properties of the selected species.

Hypothesis 4: The expected modification of wood properties due to the selected rainfall zones may enhance the suitability of the selected species for pulp and paper production or for the flooring industry through its effect on their fibre dimension, hardness strength and density.

- 5 -

Literature Review

2 Literature Review

2.1 Wood properties and suitability for industrial utilization

Throughout history, the unique characteristics and comparative abundance of wood have made it a natural material for homes and other structures, furniture, tools, vehicles, and decorative objects. Today, for the same reasons, wood is prized for a multitude of uses (MILLER 1999). Wood is made up of cells of varying shapes and sizes (PANSHIN and DE ZEEUW 1980). Variations in the characteristics of those cells make woods heavy or light, stiff or flexible, and hard or soft (MILLER 1999).

Knowledge of the properties of any material is essential for its best utilization. This is especially true for wood because of its cellular nature and its complex cell wall structure. Wood can be used with intelligence only by understanding its properties (JOZSA and MIDDLETON 1994). The properties of woods are the fundamental basis of timber utilization; they are used to evaluate the suitability of a wood for a particular application. Researchers such as MILLER (1999) concluded that specific characteristics or physical properties must be considered in order to use wood to its best advantage and most effectively in the different applications. Efficient utilization dictates that species should be matched to end-use requirements through an understanding of their properties (SIMPSON and TENWOLDE 1999).

2.1.1 Anatomical properties

A tree trunk is composed of millions of individual woody cells. The cells of softwood species differ in appearance from those of hardwood species. These cells differ in size and shape, depending on their physiological role in the tree. Detailed descriptions of the anatomical features of softwoods and hardwoods can be found in HAYGREEN and BOWYER (1996) as well as TSOUMIS (1991).

Hardwood cells show greater variation in size and shape. The majority are long and narrow with closed and pointed ends. They have a general resemblance to softwood tracheids, but are much shorter. These cells, called fibres, may have thin or thick cell walls, depending on the tree species. Brick-like parenchyma cells are also present. Also included are a relatively small number of cells with open ends, these cells are called vessel members or vessel segments, and are generally shorter than fibres, but vary in size (diameter) and shape depending on the tree species.

- 6 -

Literature Review

2.1.1.1 Fibres characteristics

Fibre dimensions

Hardwood fibres are typically long, slender, straight cells whose ends taper to a point. Fibres can be separated into intergrading types, libriform fibres and fibre tracheids . The differences between both are difficult to distinguish and therefore, this work will refer to fibres collectively; it will not distinguish between the two types. The range of fibre length and diameter in hardwood species are 400 –3,600 µm and 10 –60 µm, respectively (HARZMANN 1988 and ILVESSALO- PFAFFLI 1995). The fibre length classifications are provided in Table no 2.1, while Tables 2.2 and 2.3 provide examples for tropical hardwoods with short and long fibres, respectively.

Table 2.1: Fibre length (µm) classifications

METCALFE Classification and CHALK IAWA WAGENFÜHR (1989) (1989) (1983) Very short - - < 1000 Short < 900 < 900 1000 - 1500 Medium 900 - 1600 900 - 1600 1500 - 2000 Long > 1600 > 1600 2000 - 2500

Very long - - > 2500

Table 2.2: Examples for tropical hardwoods with short fibres (HARZMANN 1988)

Fibre length Species Origin (µm)

Aegiceras floridum ROEM. & SCHULT . 400 SE-Asia Camptostemon philippense BECC. 670 SE-Asia Cassia spectabilis L. 750 SE-Asia Guaiacum officinale L. 440 - 830 S-America Marquesia macroura GILG . 900 Africa Osbornia octodonta F. v. MUELL. 780 SE-Asia Schleichera oleosa MERR . 900 SE-Asia Sonneratia alba SM. 780 SE-Asia

The normal range for hardwood species fibre wall thickness is considered to be 3.0 –7.0 µm as cited by KHRISTOVA et al. (1998). CHATTAWAY (1932) classified fibre wall thickness to be very thin if the fibre lumina are three or more times wider than the double wall thickness, medium if the fibre lumina are less than three times the double wall thickness, and very thick

- 7 -

Literature Review

if the fibre lumina are almost completely closed. Measurement of the actual thickness of fibre walls usually involves an amount of work out of all proportion to the limited diagnostic value of the figure obtained. Therefore, the classes for fibre wall thickness are based on the ratio of lumen to wall thickness. See Tables 2.4 and 2.5 for examples of tropical hardwoods with thin and thick fibre walls, respectively.

Table 2.3: Examples for tropical hardwoods with long fibres (HARZMANN 1988)

Fibre length Species Origin (µm)

Anisoptera brunnea FOXW. 1650 SE-Asia Bischofia javanica BL. 2190 SE-Asia Ceiba pentandra GAERTN. 1790 Africa Dillenia philippensis ROLFE 2660 SE-Asia Dipterocarpus tuberculatus ROXB. 1660 SE-Asia Drypetes bordenii PAX & HOFFM. 2430 SE-Asia Lophira procera CHEV. 1830 Africa Mallotus cochinchinensis LOUR. 1830 SE-Asia Mimusops djave ENGL. 1870 Africa Ochroma pyramidale SW. 1900 - 3600 S- America Petersianthus quadrialata MERR. 2200 SE-Asia Rhizophora mangle L. 1950 Africa Uapaca kirkiana MUELL. ARG 1600 Africa

In wood industries, special attention is paid to the length of the fibre and the extent to which neighbouring fibres overlap and joined to one another. Fibre length affects the strength, surface, and bonding properties of fibre products and is therefore of interest. For many purposes, long fibres are more desirable than short ones (DADSWELL and NICHOLLS 1959).

Table 2.4: Examples for tropical hardwoods with thin fibre walls (HARZMANN 1988) Fibre wall Lumen Ø Species thickness Origin LW (µm) 2T (µm)

Dipterocarpus tuberculat . ROXB. 1,5 5 S-America Haplophragma adenophyll .WALL. 1,5 6 S-America Musanga cecropioides BR. 1,5 - Africa Ochroma pyramidale SW. 1,5 20 L.- America Sesbania sp. 1,4 48 Vietnam Toona febrifuga ROEM. 1,7 23 Vietnam

- 8 -

Literature Review

Table 2.5: Examples for tropical hardwoods with thick fibre walls (HARZMANN 1988) Fibre wall Lumen Ø Species thickness Origin LW (µm) 2T (µm) Dillenia philippens . ROLFE 14 27 SE-Asia Dipterocarpus gracilis BL. 11 5 SE-Asia Dipterocarpus grandiflorus BL. 10 8 SE-Asia Dipterocarpus kerrii FOXW. 10 8 SE-Asia Dipterocarpus tonkinensis CHEV. 13 12 Vietnam Dipterocarpus warburgi BRANDIS 11 8 SE-Asia Drypetes bordenii PAX&HOFFM. 10 7 SE-Asia Erythrophloeum fordii OLIV. 10-15 - Vietnam Garcinia fagreaoides CHEV. 15 - Vietnam Lophira procera CHEV. 10 - W-Africa Vatica dyeri KING 10 - Vietnam

Fibres are the principal component of paper and other products derived from pulp (MCMILLIN and MANWILLER 1980). The importance of fibre dimensions and their derived values (flexibility coefficient, Runkel ratio and slenderness ratio; see definitions later in the text) on pulp and paper properties are well documented. Investigators like WATSON and DADSWELL (1961), HORN (1978), ADEMILUYI and OKEKE (1979), HAYGREEN and BOWYER (1996) reported that the greater the fibre length the higher is the tear resistance. Others investigators reported that bursting strength of sheets made from unbeaten hardwood pulps also depend upon fibre length (HORN and SETTERHOLM 1990). Early research on the effect of fibre properties on paper strength (DADSWELL and WARDROP 1954 and BAREFOOT et al. 1964) led to the general belief that paper with desirable strength properties could only be made from long-fibre wood species. BAKER (1995) pointed out that in paper production, it is generally thought that the fibre length should be 2 –4 mm to obtain a good paper quality.

Woody species with short fibre length are less suitable for paper making than the long-fibre species (DICKMAN 1975 and ANONYMOUS 1978). The fibres obtained from hardwoods are mostly short in comparison with fibres from coniferous species (HURTER 1988). Despite this, numerous hardwood species with short or medium fibres are commercially used in pulp and paper making (PPM). For instance, Gmelina arborea Roxb . Roxb. is the prime source of pulpwood in West Africa and Brazil (ADEMILUYI and OKEKE 1979 and MARTIN 1984). Eucalyptus species is the preferred species worldwide for short-fibred pulp (DUTT and TYAGI 2011). Also, some Acacia species are used for PPM in many countries such as

- 9 -

Literature Review

Australia, Brazil, South Africa, Indonesia, Vietnam, and Malaysia (LOGAN and BALODIS 1982, YAHYA et al. 2010). Table 2.6 provides information about Gmelina arborea Roxb. wood properties as an example for hardwood species in commercial use for PPM.

Table 2.6: Gmelina arborea wood density and fibres characteristics (OGUNKUNLE and OLADELE 2008 )

Mean Property Remarks values

Basic density (Kg/m 3) 510 Fibre length (mm) 1.28 Trees of 1-16 years age collected from Flexibility coefficient (%) 73 different locations in Runkel ratio 0.39 Ogbomoso, Nigeria. Slenderness ratio 50

In Sudan, some studies confirmed the suitability of using some hardwood species with short fibres (such as Acacia species and others) for PPM with acceptable yields and pulp strength properties (see Table 2.7).

Table 2.7: Wood fibres characteristics of some species grown in Sudan

Fibres characteristics References Species Source Remarks FL FC RR SR Balanites aegyptiaca 1 1.11 25.3 2.9 69.1 KHRISTOVA Blue Nile Trees of 8-10 Eucalyptus tereticomis 0.80 36.5 1.7 56.2 et al. (1997) state years age Moringa oleifera 0.93 69.1 0.5 15.9 Acacia mellifera 1.2 36 1.8 66 KHRISTOVA A. senegal 1.1 28 2.5 63 Blue Nile Trees of 8-10 et al. (1998) A. seyal var. fistula 1.2 32 2.2 78 state years age A. seyal var. seyal 2 1.2 32 2.1 77 Acacia nilotica ssp: KHRISTOVA adansonii 1 41 - 46 Khartoum Trees of 10- and KARAR nilotica 1 43 - 55 state 12 years age (1999) tomentosa 1.3 46 - 67 KHRISTOVA Acacia seyal var. fistula 0.8 66 0.5 59 Blue Nile Trees of 5-6 et al. (2004) A. seyal var. seyal 1 0.8 77 0.3 64 state years age KHIDER Khartoum Trees of 5 Albizia lebbeck 0.98 61.3 0.6 38.4 et al. (2012) state years age

1 second study species, described in chapter 2.3.2, 2 first study species, described in chapter 2.3.1, FL = fibre length (mm), FC = flexibility coefficient (%), RR = Runkel ratio, and SR = slenderness ratio.

- 10 -

Literature Review

In addition to fibre length, fibre wall thickness has also significant effect on paper properties. HORN (1978) reported that fibres with thick walls result in paper having low burst and tensile strength, a high degree of resistance to tear and a very low folding endurance. The relationship of burst and tensile strengths to cell wall thickness is explained by the fact that these properties are very dependent upon a high degree of fibre-to-fibre bonding, which is affected by cell wall thickness (HAYGREEN and BOWYER 1996). These facts might lead to the conclusion that the thick-walled fibres are difficult to beat to a low freeness level as compared to the thin-walled fibres (see Figure 2.1).

Figure 2.1: The effect of fibre wall thickness in paper properties. 1. Thick walled fibres trend to retain tubular structure and provide less surface area, 2. Thin walled fibres are readily converted into ribbons and provide more surface contact. Adapted by the author from DUTT and TYAGI (2011)

Fibre derived values

Using fibre dimensions values; some researchers developed some fibre index or derived values which may be used as indicator for the suitability of the particular species for the industry. The most important fibre derived values on pulp and paper industry are flexibility coefficient, Runkel ratio as well as slenderness ratio. Those derived values are more useful for determining the paper making potential of a woody species than its absolute fibre length. DINWOODIE (1965) stressed the importance of the fibre derived values on pulp strength. Researchers like OGBONNAYA et al. (1997), SAIKIA et al. (1997) and VERVERIS et al. (2004) have successfully used those derived values to assess the suitability of various fibre raw materials for pulp and paper making.

Runkel ratio (after Dr. RUNKEL) or rigidity coefficient is defined as the ratio between twice the wall thickness and the lumen diameter (RUNKEL 1942). Fibres with a high Runkel

- 11 -

Literature Review

ratio are considered less desirable for papermaking (JANG and SETH 1998). The approximate limits of Runkel ratio appear to be from 0.25 to 1.5 for species that produce pulp of reasonable quality (VALKOMER 1969). Paper strength tends to improve with decreasing Runkel ratio (OKEREKE 1962 and RYDHOLM 1965). The high Runkel ratio gives low paper strength properties, specifically low burst, tear and tensile indexes (BEKTAS et al. 1999) and would produce porous papers (IWENOFU 1979).

Slenderness ratio or felting coefficient/power is the ratio between the length and the width of a fibre. The acceptable value of the slenderness ratio for PPM is more than 33 (XU et al. 2006). Slenderness ratio is an important factor in PPM having a positive effect on strength, tear, burst, breaking off, and double folding resistance (AKGÜL and TOZLUĞLU 2009). RYDHOLM (1965) reported that the higher the slenderness ratio, the stronger is the resistance to tearing. Normally, most softwood species have high values of slenderness ratio. For instance, Pinus kesiya has a value of 56.51 (DUTT and TYAGI 2011). However, some hardwood species have higher values of slenderness ratio than some softwood species. Table 2.7 provides information about the fibre characteristics of some hardwood species growing in Sudan. Most species provided in the table have higher slenderness values than Pinus kesiya of 56.51.

In addition, a high flexibility coefficient or elasticity coefficient (ratio of fibre lumen diameter and fibre diameter; WANGAARD 1962) is necessary for fibres used in papermaking (≥ 50 but preferably > 60). This is because paper strength tends to improve with increasing flexibility coefficient (PETRI 1952, OKEREKE 1962 and RYDHOLM 1965). According to BEKTAS et al. (1999), fibres are classified into four groups depending on the flexibility coefficient: High elastic fibres having elasticity coefficient greater than 75, elastic fibres having elasticity ratio of 50 –75, rigid fibres having elasticity ratio of 30 –50, and highly rigid fibres having elasticity ratio less than 30. Rigid fibres do not have efficient elasticity; they are used more on fibre plate and rigid cardboard production as sited by (AKGÜL and TOZLUOGLU 2009).

The improvement of paper strength with decreasing Runkel ratio and increasing flexibility coefficient may be explained by the fact that fibres with low Runkel ratio and high flexibility coefficient readily collapse and produce good surface contact in addition to fibre-to-fibre bonding. In contrast to this, fibre of high Runkel ratios and low flexibility coefficient tend to retain their pipe-like shapes to a large extent during the beating and sheet-forming process

- 12 -

Literature Review

resulting in minimal fibre-to-fibre bonding (NKAA et al. 2007). Some researchers confirmed the suitability of using species with higher Runkel ratios and lower flexibility coefficient than the acceptable ranges of PPM for such industry (see Table 2.7).

2.1.1.2 Vessels characteristics

In softwoods the functions of conduction and support are both carried out by the tracheids, whereas in most hardwoods these functions are performed by different cell types. Vessels or vessel members only occur in hardwoods. In wood tissue an indefinite number of vessel members are connected endwise to form a pipe-like structure of indeterminate length which is called vessel. In cross-sections, vessels appear as so called pores. The distribution of pores over the cross-section depends on the tree species and can be diffuse (homogeneously), or concentrated in ring-like structures called semi ring- porous or ring-porous. The size of vessel members varies widely (see Table 2.8 and 2.9); in ring-porous hardwoods, differences within a growth ring are much greater than differences between species. Vessel members are the most massive wood cells. Instead of tracheids, vessels serve to transport water and wood fibres assume the mechanical function of supporting the tree.

Pores (vessels) vary considerably in size, shape, arrangement, and number according to genera and species. Pores can be classified as small (≤ 100 µm), medium (100 - < 200 µm) or large (≥ 200 µm) according to NORMAND (1950), METCALFE and CHALK (1979), and WAGENFÜHR 1980, but typically range from 50 to >300 µm (HOADLEY 1990). Their length range from 100 to 1,200 µm.

Wood of different species varies in density and strength, due to the size and density of the vessel elements in the secondary xylem. For example, heartwood of the black ironwood (Krugiodendron ferreum , see Table 2.8) has very tiny vessel elements and is extremely dense. At the opposite extreme, the heartwood of balsa ( Ochroma pyramidale , see Table 2.9) has very large vessel elements, and is correspondingly light in density.

The vessels play a very important role as a conductive element in the living tree, but in a papermaking perspective they are not wanted. For many hardwood species, the vessel elements are a major problem in paper-making. When hardwood trees are developed to become a raw material for pulp and paper, the vessel elements should be fewer and smaller (LUNDQVIST 2002).

- 13 -

Literature Review

Table 2.8 : Examples for tropical hardwoods with small vessel diameters (<100 µm) (HARZMANN 1988)

Species Family Origin Adina cordifolia BENTH& HOOK Naucleaceae SE-Asia Aesculus indica COLEBR. Hippocastanaceae SE-Asia Altingia excelsa NOR. Altingiaceae SE-Asia Amyris balsamifera L. Rutaceae C.-America Bosqueia angolensis WELW. FIC. Moraceae W-Africa Brachylaena hutchinsii HUTCH. Compositacea E-Africa Chloroxylon swietenia DC. Rutaceae SE-Asia Cinnamomum camphora N&EB Lauraceae SE-Asia Cinnamomum cassia BL. Lauraceae SE-Asia Citrus limon BURM.f. Rutaceae SE-Asia Eurya acuminata DC. Theaceae SE-Asia Guaiacum officinale L. Zygophyllaceae L.-America Illicum verum HOOK Magnoliaceae SE-Asia Krugiodendron ferreum (VAHL.) URB. Rhamnaceae C.-America Liquidambar formosana HANCE Altingiaceae SE-Asia Nothofagus procera OERST. Fagaceae L.-America Pisonia spp. Nyctaginaceae L.-America Santalum album L. Santalaceae SE-Asia Tabebuia pentaphylla HEM. Bignoniaceae C.-America

In general, pulp with large vessel present more pronouncedly a defect known as "vessel picking " in printing paper (see Figure 2.2/1). The vessel picking trouble is a phenomenon by which some of the hardwood vessel elements in paper surface tend to be picked off by the ink tackiness of the printing press (OHSAWA 1988). The vessel picking can cause a blemish in the print, adhere to printing plates and lead to a further deterioration in print quality. They can also cause dust problems (LUNDQVIST 2002). Another defect which may occur in paper surface due to the existence of large vessel is the "star ground " defects (see Figure (2.2/2). Because the vessels are rich on pits, the ink can easily penetrate into them. Hence, there is an optimum ink acceptance by the vessels in the paper. Thus, the printed area becomes filled up with small brighter spots which correspond exactly to the vessel elements (FOELKEL 2007).

- 14 -

Literature Review

Table 2.9 : Examples for tropical hardwoods with large vessel diameters (>200 µm) (HARZMANN 1988)

Species Family Origin Albizzia zygia MACB. Mimosaceae E-Africa Anacardium excelsum Anacardiaceae S-America Anisoptera marginata KORTH. Dipterocarpac. SE-Asia

Bombax malabaricum DC. Bombacaceae SE-Asia Bombax spp. Bombacaceae Africa Ceiba pentandra GAERTN. Bombacaeae Africa Dalbergia latifolia ROXB. Fabaeae SE-Asia

Dalbergia retusa HEMSI. Fabaceae C.-America Dipterocarpus alatus ROXB. Dipterocarpac. SE-Asia Gmelina arborea ROXB. Verbenaceae SE-Asia Hopea odorata ROXB. Verbenaceae SE-Asia

Hura crepitans L. Euphorbiaceae L.-America Hymenaea courbaril L. Caesalpiniaceae L.-America Lophira procera CHEV. Ochnaeae W-Africa Mallotus cochinchinensis LOUR. Euphorbiaceae SE-Asia

Ochroma pyramidale SW. Bombacaceae L.-America Peltophorum tonkinensis CHEV. Caesalpiniaceae SE-Asia Prosopis africana TAUB. Mimosaeae E-Africa Pterocarpus macrocarpus KURZ. Fabaceae SE-Asia

Pycnanthus angolensis EXELL. Myristicaceae Africa Samanea saman MERR. Mimosaceae L.-America Spondias sp. Anacardiaceae SE-Asia Terminalia ivorensis CHEV. Combretaceae W-Africa

Terminalia superba E. & D. Combretaceae W-Africa Triplochiton scleroxylon SCHUM. Sterculiaceae W-Africa Turraeanthus africanus PELL. Meliaceae W-Africa

1 2

Figure 2.2: Defects in paper surface due to the large vessel elements. 1. Vessel picking area on a printed paper surface, 2. Printed sheet showing vessels covered with inks and therefore more glossy (star ground defect). FOELKEL (2007)

- 15 -

Literature Review

However, larger vessels have some advantages in pulp and paper processing. For instance, in the pulp converting processes they are excellent to favour chip impregnation by the cooking liquor. The existence of tyloses has a negative effect on the chip impregnation process. Tyloses make the chip impregnation process difficult and thus cooking woods from older trees is more difficult. Woods from older trees having a high amount of heartwood and being rich in tyloses, are more difficult to be converted into cellulose. Impregnation is worse, cooking must be more drastic, pulp yield is lower and the reject content increases, besides eventual later bleaching difficulties (FOELKEL 2007).

2.1.2 Physical properties

2.1.2.1 Wood density

In hardwood species the wood basic density ranges from 450 kg/m³ to 1,400 kg/m³ (KANEHIRA 1933 and TISSOT 1985; see Tables 2.11 and 2.12). It can be classified as light (≤ 500 kg/m³), medium (500 - < 720 kg/m³) or heavy (≥ 720 kg/m³) according to MELO et al. (1990). According to the Malaysian Grading Rules for Sawn Hardwood Timber (Anonymous 1984), hardwoods timbers may be classified as heavy (0.800 - 1.120 g/cm 3), medium (0.720 - 0.880 g/cm 3), and light (0.400 - 0.720 g/cm 3) depending upon air dry density. Definitions of wood density are given in Table 2.10.

Table 2.10: Definitions of wood density

Density Specific gravity

Density is the mass contained in a unit of a material:

Oven-dry density Air-dry density Basic density Ratio of the density of the material to that of water at Ratio of oven-dry Ratio of weight at 12 % Ratio of oven-dry the same temperature (4 weight to oven-dry moisture content to weight to green ºC). volume. volume at 12 % moisture. volume. Unit: without

Unit: g/cm³ Unit: g/cm³ Unit: kg/m³

Over the years, different methods have been developed to measure wood density and it variation. X-ray densitometry is considered as one of the high resolution methods used for density analysis in softwood. This method was first introduced to the field of wood analysis by POLGE (1963), and was further developed by various researchers (LENZ et al. 1976 and

- 16 -

Literature Review

SCHWEINGRUBER 1983). It is fully described by SCHWEINGRUBER (1988). The X-ray densitometry technique, traditionally applied for softwood species to assess the wood quality properties, due to its simple and relatively uniform wood structure. On the other hand, limited information is available about the validation of using this technique for hardwood species. However, some researchers succeed in using this technique in hardwood density analysis; for instance, ESPINOZA (2004) and MONTES et al. (2007) used this technique for Gmelina arborea and Calycophyllum spruceanum wood density radial variation analysis, respectively. Similarly, RIOG et al. (2008) used this technique in wood density assessment of Poplar (Populus x Canadensis and Populus deltoides ). GÜNTHER (2012), successfully used the X - ray densitometry for density analysis of Sessile oak (Quercus petraea ). The density values obtained using X-ray technique are not equivalent values. It is converted into equivalent values using correction factors (obtained by comparing the air dry density values with those of X-ray density). The correction factor of Pine and Spruce are found to be 0.886 and 0.933, respectively (SCHEINGRUBER 1988).

Some researchers found that the density values obtained by X-ray technique are slightly higher than those of air dry density. For instance, GÜNTHER (2012) revealed higher values of density achieved by an X-ray technique of 0.788 g/cm 3 in comparison with air dry density of 0.719 g/cm 3 in Sessile oak (Quercus petraea ). The author recorded a correction factor of 0.912.

Table 2.11: Some tropical hardwoods with low wood density (KANEHIRA 1933)

Species Family Density (g/cm 3) Aeschynomene hispida WILL. Fabaceae 0.044 Alstonia spathulata BL. Apocynaceae 0.058 Herminiera elaphroxylon GUILL.u.PERR. Fabaceae 0.065 Cavanillesia platanifolia H. B. K. Bombacaceae 0.103 Cavanillesia arborea K. SCHUM: Bombacaceae 0.106 Anonia palustris L. Anonaceae 0.116 Ochroma boliviana ROWL. Bombacaceae 0.116 Ochroma pyramidale Bombacaceae 0.160 Nyssa spp. Nyssaceae 0.124 Erythrina variegata L. Fabaceae 0.162 Chorisia spp. Bombacaceae 0.245 Paulownia tomentosa STEUD. Scrophulariaceae 0.260 Pisonia grandis R. BR. Nyctaginaceae 0.290

- 17 -

Literature Review

Density of wood is its single most important physical property. It is considered the best single index for overall wood quality as well as pulp yield and quality (BENDTSEN 1978). Most mechanical properties of wood are closely correlated with density. Density affects hygroscopicity, shrinkage, and swelling, mechanical, thermal, acoustic, electrical, and other basic wood properties, as well as properties related to the industrial processing of wood (DAVIS 1961, BAREFOOT et al. 1970, LEWARK 1979). DE GUTH (1980) added that wood density is an important wood property for both solid wood and fibre products in both conifers and hardwoods.

Table 2.12: Some tropical hardwoods with high wood density (ATIBT 1965 and FORS 1965)

Species Family Density (g/cm 3)

Guaiacum officinale L. Zygophyllaceae 1.15 - 1.30 Tabebuia ipe STANDL. Bignoniaceae 0.95 - 1.25 Krugiodendron ferreum (VAHL.) URB. Rhamnaceae 1.34 - 1.42 Zollernia parmensi Hub. Caesalpiniaceae 1.10 - 1.35 Schinopsis lorentzii ENGL. Anacardiaceae 1,15 - 1.30 Libidibia sclerocarpa BRITT. u. Rose Caesalpiniaceae 1.10 - 1.30 Swartziua tomentosa DC. Caesalpiniaceae 0,90 - 1.30 Astronium spp. Anacardiaceae 0.85 - 1.28 Dalbergia melanoxylon GUILL. u. PERR. Fabaceae 1.15 - 1.25 Colophospermum mopane KIRK ex LEONH. Caesalpiniaceae 1.15 - 1.25 Dalbergia retusa HEMSL. Fabaceae 0.99 - 1.22 Dalbergia granadillo PITT Fabaceae 0.99 - 1.22 Diospyros ebenum KOEN. Ebenaceae 1.00 - 1.20

Each end product has a unique set of requirements. For instance, both wood density and fibre characteristics determine whether the raw material is suitable for pulp and paper industry (IGARTÚA et al . 2003 and MONTEOLIVA et al. 2005). Basic wood density influences both the paper-making process and the properties of paper. It is an important economic indicator of pulpwood quality (BALODIS 1994). Density is related to the yield of paper per unit volume, paper resistance, optical properties and surface quality (IGARTÚA et al. 2003). The yield of pulp per unit volume is directly related to the basic density. A high basic density is therefore economically desirable as it means that the digester capacity required is lower. However, depending on the anatomical structure, it may inhibit impregnation and penetration with cooking liquor. Furthermore, too high density may also have a negative effect on the paper properties (KHRISTOVA et al. 2006). The usual density range of commercial temperate pulpwood is 350–650 kg/m³ according to CASEY (1980).

- 18 -

Literature Review

Wood density is acknowledged to affect most mechanical properties (HAYGREEN and BOWYER 1996, BARNETT and JERONIMIDIS 2003, BOWYER et al. 2003, KIAEI and SAMARIHA 2011). COWN et al. (1992) reported that the density of wood is recognised as the key factor influencing wood strength. According to SCHNIEWIND (1989) much of the variation in wood strength, both between and within species, can be attributed to differences in wood density. Research has shown that higher density species tend to have stronger timber than lower density species (TSEHAYE et al. 1995 and WALKER and BUTTERFIELD 1996).

A study conducted by KIAEI and SAMARIHA (2011) found positive correlation between wood density and MOE, MOR, and compression parallel to the grain. A significant linear relationship between wood density and mechanical properties of timber was reported by several researchers (SHEPARD and SHOTTAFER 1992, ZHANG 1995, IZEKOR et al. 2010). Therefore, density is considered as the best predictor of timber mechanical strength (DINWOODIE 2000). Earlier studies examined the predictability of some wood mechanical properties from density on various hardwood species such Celtis mildbraedii and Maesopsis eminii (ZZIWA et al. 2006), Eucalyptus globulus , E. nitens and E. regnans (YANG and EVANS 2003) and Hevea brasiliensis (GNANAHARAN and DHAMODARAN 1992). These studies reported density as a good estimator of mechanical properties in some species.

In general, the higher the density the harder is the wood (TSEHAYE et al. 1995 and WALKER and BUTTERFIELD 1996). However, in some cases, species with high density may have lower hardness strength than other species with lower density. For instance, Millettia laurentii (Wengé), with wide banded parenchyma, has wood air dry density of 0.860 g/cm 3 and end hardness strength (HB) of 39 N/mm 2. On the other hand, Quercus petraea (Sessile oak) has an air dry density of 0.690 g/cm 3 and a range of 50–60 N/mm 2 end hardness strength (HB) (WAGENFÜHR 2007). Thus, in this case, species with the lower density (Sessile oak) has the higher hardness strength. This may be due to the existence of the wide banded parenchyma in case of Wengé species.

2.1.2.2 Hardness strength

Hardness is a measure of the resistance of wood to the entrance of foreign bodies in its mass (TSOUMIS 1991). The hardness of wood varies with the direction of the wood grain. Testing on the surface of a plank, perpendicular to the grain is said to be of "side hardness". Testing

- 19 -

Literature Review

the cut surface of a stump is called a test of "end hardness". This resistance is higher up to about double in the axial direction than sidewise, but the difference between radial and tangential surfaces is seldom important (KOLLMANN 1951). Hardness is related to the strength of wood, to abrasion and scratching with various objects, as well as to the difficulty or ease of working wood with tools and machines (TSOUMIS 1991).

Generally, Brinell, Janka, and Monnin (Chalais-Meudon) are the three most used tests for hardness strength measurements applied for wood (see Table 2.13). Tropical woods include a range (see Table 2.14, for hardness classification) from extremely soft (e.g. balsa, Ochroma pyramidale , see Table 2.11 for density) to extraordinarily hard species (e.g.iron wood, Guaiacum officinale , see Table 2.12 for density).

Table 2.13: Definitions of hardness tests applied for wood

Method of hardness measurement after

Brinell Janka Monnin

Measured by pressing a chromium- Measured by pressing an Measured by pressing a steel or tungsten-carbide ball (1 11.28 mm steel into the wood cylinder with 30 mm centimetre or 0.4 inch in diameter) to half the ball's diameter. The diameter into the wood with a force between 30 to 300 N for results are stated in various perpendicular to the grain 5 to 30 seconds). The hardness is ways (United States pounds- with a force of 2000 N for expressed in Brinell Hardness force (lbf); Sweden kilograms- 5 seconds. The depth t of Number (BHN) computed by force (kgf); Australia Newtons the depression left by the dividing the load in N by the area of (N) or kilonewtons (kN); cylinder is measured in indentation made by the ball sometimes the results are mm. Monnin hardness N is measured in square millimeters treated as units, for example calculated as 1/t (N/mm²) "660 Janka"

Wood mechanical properties are usually the most important characteristics of wood product to be used in structural applications such as flooring and rafters, structural panel roof, wall sheathing, etc. (HAYGREEN and BOWYER 1996).

Quercus rubra (Red oak) wood is the American’s flooring industry benchmark for comparing the relative hardness of different wood species, because red oak is a popular and widely selected wood for flooring that performs well residentially. Hardwood flooring species are usually compared to Red oak as the basis. In Africa, Milletia laurentii De Wild. (wengé) is considered as one of the most widely used hardwood species for flooring industry.

- 20 -

Literature Review

Table 2.15 provides information about wood air dry density as well as Brinell hardness strength of Milletia laurentii and Quercus rubra according to WAGENFÜHR (2007).

Table 2.14: Hardness classifications

Hardness classifications according to the method of hardness measurement after Janka 2 Classification 1 Brinell 3 HB Monnin BHN Level Pounds (lbf) Extremely soft - 1 up to 300 - Very soft up to 35 2 301 - 600 up to 1.5 Soft up to 49 3 601 - 900 1.5 - 3 4 901 - 1,200 Somewhat hard - - 5 1,201 - 1,500 Medium 6 1,501 – 1,800 up to 59 3 - 6 (moderately) hard 7 1,801 - 2,100 8 2,101 - 2,400 Hard up to 65 6 - 9 9 2,401 - 2,700 10 2,701 - 3,000 Very hard up to 146 9 - 20 11 3,001 - 3,300 12 3,301 - 3,600 Real hard - - 13 3,601 - 3,900 14 3,901 - 4,200 Extremely hard over 150 - 15 4,201 - 4,500 16 4,501 - 4,800 Extraordinarily hard - - 17 4,801 - 5,100

1 MÖRATH in LOHMANN (1991) 2 ASTM (2005) 3 TROPIX 7 (2013)

Table 2.15: Air dry density and Brinell hardness strength of Milletia laurentii and Quercus rubra

Property Milletia laurentii Quercus rubra 0.869 Air dry density (g/cm 3) 0.700 (0.810 - 0.950 ) Hardness strength (N/mm 2) 39 Transverse section 53 - 66 (30 - 45) 24 Mean of radial and tangential sections 29 - 36 (21 - 27)

- 21 -

Literature Review

Janka as early as 1906 and V. LORENZ (1909) stated that the hardness is approximately proportional to the density of wood (KOLLMANN and CÔTÉ 1968). YLINEN (1943) showed that for practically the whole range of densities of commercial timbers a linear relationship between Brinell hardness and over dry density is applicable. WIEMANN and GREEN (2007) developed a compatible relationship between basic specific gravity and Janka hardness for softwoods and temperate as well as tropical hardwoods (see Figure 2.3). Positive relationship between specific gravity and both of end and side Brinell hardness was found in hardwood as well as softwood (MÖRATH 1932) see Figure 2.4.

HB Brinell hardness

Specific gravity Q 0

Figure 2.3: Relationship between Janka side Figure 2.4: Relationship between Brinell hardness and basic specific gravity for hardness (end and side) and specific geenwood (WIEMANN and GREEN 2007) gravity. Measurement by MÖRATH (1932). Source: KOLLMANN and CÔTÉ (1968 )

2.1.3 Relations between anatomical composition and wood density

Wood is a natural material of biological origin. It is an extremely versatile material with a wide range of properties (NASROUN and ALSHAHRANI 1998). This variability is attributed mostly to variations in the anatomical composition of wood (PANSHIN and DE ZEEUW 1980, ZOBEL and VAN BUITENEN 1989 and VALENTE et al. 1992). Wood density considered as the most important wood physical property for optimizing the end use (see chapter 2.1.2.1). VALENTE et al. (1992) reported that the wood density is a complex

- 22 -

Literature Review

feature influenced by cell wall thickness, the proportion of the different kind of tissues, and the percentages of lignin, cellulose and extractives. Similarly, ZOBEL and VAN BUITENEN (1989) pointed out that the density is not a simple characteristic and that it determined by several wood anatomical characteristics such as cell size and wall thickness, the ratio of earlywood to latewood, the amount of ray cells, the size and amount of vessel elements.

2.1.3.1 Fibre wall thickness

Xylem fibres with thick walls and small lumina add to the increase of the specific weight, while wide lumina and thin walls decrease it. WIEDENHOEFT (2010) reported that density is dependent largely on fibre wall thickness. Researchers such as, SCARAMUZZI and FERRARI (1963) concluded that wood density can be estimated from fibre wall thickness. NASROUN and ALSHAHRANI (1998), in a study of the relationship between anatomical structure and density of five hardwood species, found positive relationship between density and some fibre characteristics such as, fibre wall thickness, Runkel ratio, coefficient of cell rigidity and fibre length. Their study also revealed negative relationship between density and all off: vessels, parenchyma proportions and fibre lumen diameter.

The cell wall represents the woody material, therefore, the density increases as the proportion of cell wall increases and vice versa. This fact agrees with those pointed out by LEWIN and GOLDSTEIN (1991) that there is positive relationship between wood density and cell wall proportion. Similarly, CHOWDHURY et al. (2012) found positive relationship between air dry density and fibre wall thickness in Casuarina equisetifolia growing in Bangladesh.

2.1.3.2 Cell frequency

The frequency of cell types differs from one wood species to another (see Table 2.16). The typical composition of every species determines peculiar technical and physical properties of the wood. High wood density is the result of a high amount of thick-walled fibres (e.g. Lophira procera in Table 2.5 and 2.16, as well as Ocotea rodiaei in Table 2.16); additionally many vessels or a high portion of parenchyma cells lead to lower density values (e.g. Liquidambar formosana in Table 2.16 and Turraeanthus africanus in Table 2.9).

In all woods, density is related to the proportion of the volume of cell wall material to the volume of lumina of those cells in a given bulk volume. In hardwoods, density is dependent

- 23 -

Literature Review

on fibre wall thickness, the amount of void space occupied by vessels and parenchyma (WIEDENHOEFT 2010).

Table 2.16 : Frequency of cell types in some tropical hardwoods (HARZMANN 1988) Axial Ray Species Vessels Fibres paren- paren- Density (%) (%) chyma chyma (g/cm³) (%) (%) Ceiba pentandra GAERTN. 8 30 41 21 0.26 Gossweilerodendron balsami f.HA. 10 47 27 16 0.49 Lophira procera CHEV. 9 58 20 13 1.12 Millettia laurentii DE WILLD. 6 44 31 19 0.75 Ochroma pyramidale SW. 3-5 4 7 17 - 19 0.12 Ocotea rodiaei MEZ. 6 75 4 13 0.93 Pycnanthus angolensis EXELL. 4 8 7 34 0.44 Rhodognaphalon brevicuspe ROB. 17 32 4 24 0.34 Triplochiton scleroxylon SCHUM. 9 28 - 22 0.35 Turraeanthus africana PELLEGR. 15 75 41 7 0.51 Dipterocarpus tuberculatus ROXB. 17 45 3 25 0.73 Haplophragma adenophyllum WA. 12 51 13 12 0.75 Lannea grandis ENGL. 13 69 25 13 0.55 Terminalia tomentosa W. &. A. 12 51 5 12 0.70 Liquidambar formosana HANCE 30 54 25 6 0.63

2.1.4 Wood properties variation

Wood is a raw material of variable structure. Its wood properties differ between and within species, and between and within trees (PANSHIN and DE ZEEUW 1980).

2.1.4.1 Radial variation of fibres characteristics

Understanding the radial variation of wood fibre characteristics, and related anatomical features provides a basis for improved wood utilization. The patterns of radial variation are not the same for all wood characteristics. PANSHIN and DE ZEEUW (1980) reported that the systematic variation in characteristics of wood over a series of growth rings from pith to bark is the result of several influences. The first influence is the change in the cambial initials themselves as they continue to function in aging trees, and the second influence is the postcambial development of cells derived from the cambium.

In general, the fibre length increases from pith to bark. This fact has been confirmed by the pioneering wood scientists PANSHIN and DE ZEEUW (1980) as well as ZOBEL and VAN BUJITENEN (1989). The same trend has been observed in several species, such as in 9-

- 24 -

Literature Review

years-old seedling of Eucalyptus camaldulensis (VEENIN et al. 2005), Eucalyptus regnans (BISSET and DADSWELL 1949), Eucalyptus grandis (MALAN and HOON 1992), E. globulus of different tree ages (JORGE et al. 2000, MIRANDA et al. 2001, 2003, MIRANDA and PEREIRA 2002), Tectona grandis (IZEKOR and FUWAPE 2012), and Quercus garryana Dougl. (LEI et al. 1996).

In some species, the fibre length increases gradually from pith to bark and then become more or less constant near the bark. This pattern has been found in some Acacia species such as Acacia auriculiformis (CHOWDHURY et al. 2009), A. mangium of different tree ages (SAHRI et al. 1993 and HONJO et al. 2005) and A. melanoxylon (TAVARES et al. 2011). Similarly, in Balanites aegyptiaca tree species , PARAMESWARAN and CONRAD (1982) registered radial increase of fibre length, starting sharply in the first few centimetre from the pith, attaining a constant values at a distance of 2.5 cm from the pith. Variation in fibre length within trees, including branches has been studied by BHAT et al. (1989) in 11 tropical Indian hardwoods growing in Kerala. Radial pattern of fibre length variation in their study, in the majority of the species studied showed a decline in fibre length near the bark after an initial increase from the pith to outwards. However, BRASIL and FERREIRA (1972) did not observe significant radial variation in fibre length at breast height level in 3-year-old trees of Eucalyptus grandis .

BUTTERFIELD et al. (1993) studied the wood of Hyeronima alchorneoides and Vochysia guatemalensis growning in natural stand as well as in plantation. Their results revealed a radial increase of fibre length from pith to bark for natural and plantation trees of both species; ranging from 1.92 to 2.98 mm (natural) (155 % increase) and 1.45 to 2.71 mm (plantation) (187 % increase) for Hyeronima; and from 0.63 to 1.75 mm (natural) (277 % increase) and 0.69 to 1.47 mm (plantation) (213 % increase) for Vochysia. Eucalyptus regnans recorded an increase of 225 % for fibre length from pith to bark which varied from 0.6 mm near the pith to 1.35 mm outward (BISSET and DADSWELL 1949). JORGE et al. (2000) found in their study of the variability of fibre length in wood and bark in Eucalyptus globules an increase of 0.70 mm to 1.20 mm from 10 % to 90 % distance from the pith at 15 % height level (with 171.4 % increase) versus an increase of 0.70 mm to 1.13 mm at 75 % height level (with 161.4 % increase). TAVARES et al. (2011) recorded percentage of fibre length increase from pith to bark of 141.3 % in Acacia melanoxylon with values of 0.75 mm fibre length in 10 % of the distance from the pith to 1.06 mm at 90 % distance from the pith. CHOUDHURY et al. (2009), in a study of the radial variations of wood properties in 11-year

- 25 -

Literature Review

old Acacia auriculiformis grown in Bangladesh, reported that fibre length and fibre length increments increased up to about 80 mm radial distance from pith and then were almost constant towards the bark. BISSET and DADSWELL (1949) reported that fibre length varied from 0.6 mm near the pith to 1.35 mm in the outermost growth rings in one tree of Eucalyptus regnans . The increase was rapid for ten rings, then became more or less constant outwards.

In addition to fibre length, an increase of fibre diameter and wall thickness was observed in different hardwood species in previous studies. For instance, on Eucalyptus grandis × E. Urophylla (QUILHÓ et al. 2006), Eucalyptus tereticorn (SHARAMA et al. 2005 and UNIYAL 2012) and Neolamarckia cadamba Roxb. (ISMAIL et al. 1995). Similarly, in Rubber wood ( Hevea brasiliensis ), NAJI et al. (2012) found that fibre dimensions (length, diameter, lumen diameter and wall thickness) increase from pith to bark. In Sudan , OSMAN (2001) studied the fibre dimensions radial variation in ten hardwood species including Acacia seyal (the first study species) as well as Balanites aegyptiaca (the second study species). The author found a radial increasing trend in the entire studied fibre dimension (fibre length, diameter, lumen diameter, and wall thickness) in all the studied species.

RAO et al. (2003) found significant pith to periphery variations in fibre length, diameter, lumen diameter of plantation grown Tecomella undulate . BAKHSHI et al. (2012) revealed an increase in Quercus castaneaefolia fibre length, diameter, lumen diameter, double wall thickness, and flexibility coefficient along the radial direction from the pith to bark. KIAEI (2012) studied the influence of cambial age on fibre dimensions of Maple Wood ( Acer velutinum Boiss). The study results revealed that the values of fibre length, cell wall thickness, slenderness ratio and Runkel ratio increased while fibre width, lumen diameter and flexibility ratio decreased from pith to bark. Similar to the radial decrease trend of fibre lumen diameter found by ISMAIL et al. (1995), SHARMA et al. (2005) and KIAEI (2012), CHOWDHURY et al. (2012) observed in their studies radial decreasing trend in fibre lumen diameter. UNIYAL (2012) found a gradual increase from pith to bark in fibre length, diameter, wall thickness as well as the slenderness ratio of Eucalyptus tereticorn.

An increase of fibre wall thickness was detected on Acacia auriculiformis (CHOWDHURY et al. 2013), Acacia melanoxylon (TAVARES et al. 2011), Casuarina equisetifolia (CHOWDHURY et al. 2012) and on Eucalyptus globules (JORGE 1994). Other authors found no radial pattern of wood wall thickness (RAMIREZ et al. 2009).

- 26 -

Literature Review

Some researchers did not found significant radial increase in fibres characteristics of their study species. For instance, PANDE and SINGH (2009) found a non-significant increase in fibre length, fibre diameter and fibre wall thickness in 4-years-old clonal ramets of Eucalyptus tereticornis Sm. Similarly, UNIYAL (2012) found non-significant differences in Runkel ratio from pith to bark in Eucalyptus tereticornis Sm . However , OLUWAFEMI and TUNDE (2008) concluded that fibre diameter, lumen diameter, flexibility coefficient and Runkel ratio did not show a distinct radial pattern in Sterculia setigera growing in Nigeria.

2.1.4.2 Radial variation of vessels characteristics

Vessel lumen diameter generally shows a juvenile to mature pattern of variations from the pith to the bark (radial variation), where vessel lumen diameter is smaller in the inner parts of the stem, and gradually increase in size outwards before leveling off in the outer parts of the stem (FURUKAWA and HASHIZUME 1987, OHBAYASHI and SHIOKURA 1990, PESZLEN 1994, GARTNER et al. 1997, BHAT et al. 2001). UNIYAL (2012) in a study of the variation in wood anatomical properties and specific gravity in Eucalyptus tereticornis, revealed significant radial variation in vessel element length as well as vessel diameter. In Neolamarckia cadamba Roxb., ISMAIL et al. (1995) found radial variation in vessel diameter as well as vessel element length.

Vessel diameter increased gradually up to about 40–45 % distance from pith and then levelled-off to bark in Acacia auriculiformis (CHOWDHURY et al 2013). Similarly, other authors found a significant radial increase of vessel diameter in different hardwood species. For instance, in Eucalyptus grandis by TAYLOR (1973) and MALAN and GERISCHER (1987), Hevea brasiliensis by NAJI et al. (2012), Hyeronima alchorneoides and Vochysia guatemalensis by BUTTERFIELD et al. (1993), Melia dubia by SWAMINATHAN et al. (2012), Quercus garryana Dougl. by LEI et al. (1996) and Shorea acuminatissima by ISHIGURI et al. (2012).

GARTNER et al. (1997) concluded in their study on Alnus rubra , that radial variation was small, although it was significant for vessel diameter and proportion vessel. Radial variation of vessel lumen diameter in 30 hardwood species was studied by TSUCHIYA and FURUKAWA (2009). They found a juvenile-mature pattern in radial variation of vessel lumen diameter in earlywood in almost all ring porous woods and in the central 10 portion of the annual rings in the diffuse porous woods.

- 27 -

Literature Review

Radial variation in anatomical properties was studied in an 8-year-old plantation grown Tecomella undulata . Significant pith to periphery variation in vessel frequency, vessel diameter, percentage of solitary vessel was reported. Among the different correlation studied vessel frequency, vessel diameter and vessel types were found to be inter-related (RAO et al. 2003). XIAOMEI et al. (2003) reported in their study on I-214 poplar ( Populus×canadensis cv. “I-214") in Beijing, that all anatomical properties including vessel length and tangential diameter and vessel ratio showed rapid and then gentle increase from pith to outwards.

In contrary to the above mentioned literature, some authors reported non-significant radial variation in vessel diameter. For instance, VEENIN et al. (2005) in Grandis camaldulensis , PANDE and SINGH (2009) in Eucalyptus tereticornis Sm. and ISHIGURI et al. (2009) in Paraserianthes falcataria.

In several hardwood species, the number of vessels per square millimeter is decreasing along the radial position. For instance, in Casuarina eguietifolia (CHOWDHURY et al. 2012), Eucalyptus globules (RAMÍREZ et al. 2009), Hyeronima alchorneoides and Vochysia guatemalensis (BUTTERFIELD et al. 1993), Neolamarckia cadamba Rox (ISMAIL et al. 1995) and Shorea acuminatissima (ISHIGURI et al. 2012).

2.1.4.3 Radial variation of density

In hardwood species, all possible patterns of wood density radial variations can be found. Generally, the middle to high density diffuse porous hardwood species follow a pattern of low density near the pith and then an increase, followed by a slower increase or levelling off toward the bark (ZOBEL and BUIJTENEN 1989). PANSHIN AND DE ZEEUW (1980) classified the radial patterns of specific gravity variations into three general types: type one showing an increase from pith to bark, type two showing a decrease outward from pith, then increase toward the bark and finally type three where the specific gravity decreases from pith to bark.

OSMAN (2001) found a radial increase trend in air dry and oven dry densities for ten hardwood species grown in Sudan including Acacia seyal as well as Balanites aegyptiaca . The basic density increased up to 80 mm radial distance from the pith, then was almost constant towards bark in 11-years-old Acacia auriculiformis species (CHOUDHURY et al. 2009). ZEIDLER (2012) studied the variation of wood basic and oven dry densities in Turkish hazel ( Corylus colurna L.) grown in the Czech Republic. The author found an initial

- 28 -

Literature Review

increase followed by a moderate decrease at 4/5 of the radius replaced again by an increase, reaching the highest value close to the cambium for both densities.

Radial variation in basic specific gravity was determined for three natural and nine plantations grown tress of Light Red Meranti ( Shorea leprosula and S. parvifolia ). In both natural and plantation grown trees significant radial increase in specific gravity has been detected. The plantation grown trees have slightly less variable wood than the wild trees. Within and among trees specific gravity parameters vary considerably (BOSMAN et al. 1994).

Variation in wood specific gravity from pith to periphery was examined in Gmelina arborea plantation in Venezuela. The specific gravity was obtained using an X-ray densitometer. The results showed that there was an increase in specific gravity from pith to bark (ESPINOZA 2004). In the same species radial increase in basic density has been reported by LAMB (1968). A radial increase of specific gravity in natural and plantation trees of Hyeronima alchorneoides as well as Vochysia guatemalensis has been detected by BUTTERFIELD et al. (1993). In Calycophyllum spruceanum, a radial increase trend has been observed in wood air dry density using X-ray densitometry (MONTES et al. 2007). SETTLE et al. (2012) revealed a radical increase in wood basic density of Endospermum medullosum .

The radial increase trend in wood density has been observed in Eucalyptus species as well. For instance, HEIN et al. (2012) studied the radial variation of wood basic density in 14-years old Eucalyptus urophylla S.T. In their study, a significant linear increase from pith to bark was confirmed. Studying the radial variation of specific gravity in 4-years old Eucalyptus tereticornis Sm ., UNIYAL (2012) revealed a radical increase from pith to bark. Similarly, the radial increase of basic density has been reported in Eucalyptus regions by FREDERICK et al. (1982). However, SADEGH (2012) revealed a decreasing trend from pith to bark in wood basic density of Eucalyptus camaldulensis grown in Iran. Similarly, HIETZ et al. (2013) studied wood density of several species. The authors found radial increasing as well as decreasing trends in basic density in their study species. GARTNER et al. (1997) reported that specific gravity remained constant along the radial direction from pith to bark on six 40- year-old Alnus rubra trees.

- 29 -

Literature Review

2.1.4.4 Vertical variation of density

ZOBEL and BUIJTENEN (1989) stated that almost all possible pattern of longitudinal variation can be found in either conifer or the hardwood. But the decline pattern is evident in hardwood; however most diffuse porous hardwood has little variation in wood density from the base to the top. They highlighted the importance of studying the vertical trend in juvenile and mature woods separately. PANSHIN AND DE ZEEUW (1980) reported that in hardwood species the vertical variation in specific gravity follows one of three general patterns. It may decrease upward; it may decrease in the lower trunk and increase in the upper trunk; or increase upward.

Variations in density along the stem are less consistent than those in the radial direction. As the cylinder of juvenile wood extends from the base of the stem to the top, the proportion of juvenile wood over the cross-section of the stem increases. As a result of this a vertical decrease trend in density often occurs (ZOBEL and BUIJTENEN 1989).

A decreasing trend from the base to the top has been detected in wood basic density and wood oven dry density in Corylus colurna L. by ZEIDLER (2012). UNIYAL (2012) found vertical decline in specific gravity of Eucalyptus tereticornis Sm. However, it was not significant. Similarly, LAMB (1968) and LEI et al. (1996) found vertical decreasing trend in specific gravity of Gmelina arborea and Quercus garryana Dougl., respectively. However, some authors revealed an increasing trend from the base to the top in their studied species; for instance, GÖHRE (1960) in Populus spp, DARGAVEL (1968) and FREDERICK et al. (1982) in Eucalyptus regnans .

SADEGH (2012) studied the variation of basic density in Eucalyptus camaldulensis grown in Iran. The study results showed a vertical increase trend of wood basic density. Physical and mechanical properties of Grevillea robusta were studied by KAMALA et al. (2000). Results indicated that average specific gravity increases with height . Other authors reported no differences in wood density with stem height like MCELWEE and FAIRCLOTH (1966) in Nyssa silvatica , PURKAYASTHA et al. (1982) in Eucalyptus tereticornis and JAIN and ARORA (1995) in Eucalyptus camaldulensis.

Generally, the mechanical wood properties do not vary at different height as concluded by CHUDNOFF (1961) in Eucalyptus camaldulensis . The variability of most mechanical and elastic properties can be estimated by referencing to the variation in specific gravity as sited

- 30 -

Literature Review

by PANSHIN and DE ZEEUW (1980). Therefore, the radial and vertical variation in hardness strength will not be discussed in this section.

2.1.4.5 Trees, forests and regions variation

Difference in wood properties occurs in the same site, as well as between sites in the same or in different geographical localities and altitudes. In general, the tree-to-tree differences in wood properties are large and differ considerably among species. While no species has uniform wood, some are much more uniform than the others (ZOBEL and VAN BUIJTENEN 1989).

ZOBEL and VAN BUIJTENEN (1989) stated that the tree-to-tree variability in wood properties within a species or within a provenance is large. CHUDNOFF (1961) concluded that there was more variability among trees at the same site than between sites in Eucalyptus camaldulensis . TAYLOR (1973) reported significant differences between tree in fibre length and non-significant differences in fibres diameter. However, ISHIGURI et al. (2009) found significant differences among trees in wood fibre diameter of 13-year-old Paraserianthes falcataria planted in Indonesia.

CHOWDHURY (2013) found significant differences among tree in fibre and vessels diameter (mean of radial and tangential diameter) in 11-years-old Acacia auriculiformis growing in Bangladesh. On the other hand, the fibres tangential diameter and wall thickness did not vary significantly. ISMAIL et al. (1995) found significant variation between trees in all the studied anatomical properties of Neolamarckia cadamda , including fibre length, diameter, lumen diameter, wall thickness and vessel diameter.

Considerable differences in fibre length among trees were reported in Carya pecan by TAYLOR (1969) and in Eucalyptus grandis by HANS et al. (1972). However, BHAT et al . (1990) observed non-significant difference in fibre length between trees in Eucalyptus grandis . Similarly, LEI et al. (1996) observed no significant difference among trees in fibre length, vessel diameter, cell proportions as well as specific gravity in Quercus garryana .

Significant variations among trees were reported in basic density, fibre and vessel element length by CHOUDHURY et al. (2009). In Alnus rubra, there was significant variation in specific gravity and vessel diameter between trees, although the magnitude of these differences were not large (GARTNER et al . 1997). Variations in specific gravity were found significant in different clones of Delbergia sissoo (PANDE and SINGH 2005).

- 31 -

Literature Review

Wood specific gravity as well as other wood qualities will vary greatly from tree to tree, regardless of the species or where the trees are grown (ZOBEL and TALBERT 1984). HANS et al . (1972) observed significant variation among trees for wood density in Eucalyptus grandis . Similarly, TAYLOR (1973) observed that specific gravity varied significantly from tree-to-tree in Eucalyptus grandis. Basic density varied significantly among the clones of Eucalyptus tereticornis (SHASHIKALA and RAO 2005). FIMBEL and SJAASTAD (1994) reported that for wood specific gravity, tree-to-tree variation accounted for 76–90 % of the total variability in Ceiba pentandra trees collected from four Costa Rican life zones. THORBJORNSEN (1961) and TYLOR (1968) reported tree-to-tree variation in Liriodendron tulipifera .

PANDE et al. (2008) studied the intra and inter-tree variations in physico-chemical and wood anatomical features in Leucaena leucocephala (Lam.) De Wit. The authors concluded that inter-tree variations of wood anatomical properties were significant. Location also affected anatomical properties and pulping and paper quality ratios significantly. However, in a study of the radial variation trend in the wood of Aspidosperma quebracho blanco, MOGLIA and LOPEZ (2001) found that the wood anatomical features (vessles area, fibre diameter, lumen diameter, cell wall thickness, percentage of vessels, fibres, parenchyma and rays) were not affected by locations difference. The variability of wood density and fibre length was determined in six 13-year-old willow ( Salix spp.) clones growing under different site conditions in Argentina by (MONTEOLIVA et al. 2005). Significant differences were found between clones as well as sites.

Influence of provenance variation on wood properties of Tectona grandis from the Western Ghats region in India was studied. Three major teak provenances (locations) were characterized in terms of mechanical and anatomical wood properties (BHAT and PRIYA 2004). The study results revealed significant differences among the selected locations in air dry density, some of the studied anatomical properties and in mechanical properties.

2.2 Influence of different climatic growth conditions on wood anatomical and physical properties

Research on wood has substantiated that the climatic condition where the species grow has a significant effect on wood properties (ALVES and ANGYALOSSY-ALFONSO 2000, BEADLE et al. 2001, ALVES and ANGYALOSSY-ALFONSO 2002, WIMMER et al. 2002,

- 32 -

Literature Review

ROQUE 2004, AL-KHALIFA et al. 2006, MOYA and FO 2008). Understanding the extent of variability of wood is important because the uses for each kind of wood are related to its characteristics; furthermore, the suitability or quality of wood for a particular purpose is determined by the variability of one or more of these characteristics (PANSHIN and DE ZEEUW 1980).

Therefore the effect of the climatic growth condition on wood properties will be discussed in the current section. More focus will be provided on the effect of water supply as it is usually the most limiting factor for plant growth. The effect of air temperature will be discussed also in this section. Brief discussion of the effect of the other factors will be provided also.

2.2.1 Water supply

Water is the most limiting factor responsible for the distribution of higher plants (HSAIO 1973 and ENU-KWESI et al. 1986). If plants do not receive adequate rainfall or irrigation, the resulting drought stress can reduce growth more than all other environmental stresses combined. It is well known that water stress affects every aspect of plant growth, modifying anatomy, morphology, physiology and biochemistry (HSAIO 1973, ENU-KWESI et al. 1986, AUGE et al. 1998). Plants, by nature, possess remarkable adaptive mechanisms to tolerate drought stress (LEVITT 1972). However, the response patterns to the same environment can differ among species as concluded by NOSHIRO and SUYUKI (1995). SASS and ECKSTEIN (1995), in their study of the variability of vessel size in Fagus sylvatica , reported that the formation of vessels during the beginning of the cambial activity was controlled by internal factors (not specified), while adult wood formation was affected by external ones (e.g. rainfall).

2.2.1.1 Anatomical properties

The anatomical characteristics of the trees are determined by the climatic condition of the sites in which they are grown, especially in aspects related to water availability (WIEMANN and WILLIANSON 2002). The anatomical features are modified within trees during their growth in order to adjust physiologic and water stress, then they maintain the existence of the species. AKACHUKU and BURLEY (1979), AKACHUKU (1985), and OHTANI et al. (1989) demonstrated that the vessels and parenchyma are the main elements of secondary xylem that reflect ecological variations. Moreover, several researchers found significant

- 33 -

Literature Review

correlation between fibre dimensions and the climatic growth condition (FAHN et al. 1986, OGBONNAYA et al. 1997, ALVES and ANGYALOSSY-ALFONSO 2002, AL- KHALIFAH et al . 2006 and MOYA and FO 2008).

Vessels

In hardwood, the function of water conduction is performed by vessel elements. Therefore, it is considered as the most sensitive anatomical features for water stress. Among the different vessel's characteristics, vessel diameter is probably the most important anatomical variable in angiosperm. Wood species from drier habitats have narrower vessels than species from more humid habitats (CARLQUIST 1988).

Several authors have pointed out that water availability is one of the most important environmental sources of variation of the vessel anatomy of different species such as Eucalyptus globulus, Eucalyptus grandis and Eucalypt hybrids (LEAL et al . 2003, MALAN 1991, 1993, respectively), Fagus sylvatica L. (SASS and ECKSTEIN 1995). Previous research has shown that in general vessel diameters and vessel element lengths decrease while vessel frequencies increase with decreasing water availability (CARLQUIST 1975, BAAS and SCHWEINGRUBER 1987, VAN DER WALT et al. 1988, ZHANG et al. 1988, WILKINS and PAPASSOTIRIOU 1989, FEBRUARY 1993 and LENS et al . 2004).

Moreover, some researchers confirmed that vessels frequency are not affected by water availability (FEBRUARY et al . 1995 and MOYA and FO 2008) and some confirmed that even the vessels diameter and length are not affected by water availability (MOYA and FO 2008).

MOYA and FO (2008) in their study of the variation in the wood anatomical structure of Gmelina arborea , found that the annual average precipitation affected vessel proportion, as the precipitation increases, vessel proportion decreases. While the vessels diameter, vessel frequency, vessels multiple percentages have not significantly affected by precipitation. Al- KHALIFAH et al. (2006) investigated the impact of water stress on the sap wood of Callgonum comosun , and found that the drought causes narrower vessels both in earlywood and latewood, thicker vessel walls, longer vessel elements, a higher frequency of small latewood vessels and a lower frequency of large earlywood vessels, narrower growth rings, a lower total fraction of vessels per xylem area.

- 34 -

Literature Review

ZHANG et al. (1988) revealed a strong positive correlation between rainfall, altitude and element size in Syringa oblate. WILKINS and PAPASSOTIRIOU (1989), indicated a trend towards decreasing vessel diameter and vessels length and increasing vessels frequency with increasing dryness for Acacia melanoxylon and FEBRUARY (1993) demonstrated the same trend in Combretum apiculatum and Protea caffra.

FEBRUARY et al. (1995) studied the relationships between water availability and selected vessel characteristics in Eucalyptus grandis and two hybrids and reported that anatomical responses to water treatment differed between taxa. They found that vessel diameter and vessel element length in Eucalyptus grandis and E. grandis x camaldulensis were positively correlated with increases in water used. However, neither E. grandis nor E. grandis x camaldulensis showed significant correlation between water treatment and vessel frequency. In E. grandis x nitens only vessel frequency responded to water treatment. LINDORF (1994), studying several species of the dry tropical regions of Venezuela, found that increasing number of vessels, small vessels diameters, short vessels elements and minute intervessels pits were observed in species of dry forest.

Vessels have been reported to show a tendency towards grouping in dry environments, whereas they are more often solitary and only rarely grouped in humid environments (CARLQUIST 1966, BARAJAS-MORALES 1985, CARLQUIST and HOEKMAN 1985, FAHN et al. 1986, LINDORF 1994). Such a trend could functionally be explained by a clustering strategy to improve safety for hydraulic conduction, during periods of physical or physiological drought. According to ZIMMERMANN (1983), vessel multiples are safer, since they provide alternative paths for the xylem sap to bypass embolisms. In a group such as Asteraceae, the degree of vessel grouping rises markedly in relation to dryness of the habitat (CARLQUIST 1966). Likewise, LIPHSCHITZ and WAISEL (1970) and SCHUME et al. (2004) demonstrated that the porosity and vessel grouping of Populus were correlated to water availability. Mature poplars reacted to a decrease in water availability by forming vessels with smaller diameters and increasing vessel density (SCHUME et al. 2004). Eucalyptus grandis, grown in South Africa showed a trend of increasing vessel area with increasing precipitation (NAIDOO et al. 2006).

The effect of water stress on some vessels characteristics has been summarized in Table 2.17.

- 35 -

Literature Review

Table 2.17: The effect of water stress on some vessels characteristics Vessel s References Characteristics L D F G CARLQUIST (1966) ** ** ** + CARLQUIST (1975) - - + ** BARAJAS-MORALES (1985) ** ** ** + CARLQUIST and HOEKMAN (1985) ** ** ** + FAHN et al. 1(986) ** ** ** + BAAS and SCHWEINGRUBER (1987) - - + ** VAN DER WALT et al. (1988) - - + ** ZHANG et aI. (1988) - - + ** WILKINS and PAPASSOTIRIOU (1989) - - + ** FEBRUARY (1993) - - + ** LINDORF (1994) ** ** ** + LENS et al. (2004) - - + ** Al-KHALIFAH et al. (2006) + - ** ** MOYA and TOMAZELLO (2008) 0 0 0 **

L = length, D = diameter, F = frequency, G = grouping, + = increase, - = decrease, 0 = no change and ** = no information is available

Fibres

The effect of water stress on fibre dimensions has been studied by several researchers (FAHN et al. 1986, OGBONNAYA et al. 1992 and 1997, ALVES and ANGYALOSSY-ALFONSO 2002, AL-KHALIFAH et al . 2006 and MOYA and FO 2008).

OGBONNAYA et al. (1992 and 1997) have shown that water stress adversely influenced the wood fibre dimension of Kenaf ( Hibiscus cannabinus L.) and Gmelina arborea , respectively. Similarly, MUKHTAR (2008) in a study of the variation in wood anatomy of Acacia senegal grown in different rainfall zones in Sudan, found significantly lower values of fibre diameter, lumen diameter and wall thickness in the rainfall zone of <300 mm annually than in the rainfall zone of 300–500 mm annually. The reduced fibre dimensions by water stress would be due in part to the role of water in turgidity maintenance necessary for cell enlargement (KRAMER 1963), because cell expansion is achieved when loosening of the cell wall yields under stress of the internal turgor pressure (SCHULTZ and MATTHEW 1993).

On the other hand some authors found thicker wall fibres in drier environments (FAHN et al. 1986 and ALVES and ANGYALOSSY-ALFONSO 2002) and some did not found any significant effect for water stress in fibres wall thickness (NKAA et al. 2007 and MOYA and FO 2008). This confirms the conclusion of NOSHIRO and SUZUKI (1995) that the response patterns to the same environment can differ among species. AL-KHALIFAH et al . (2006)

- 36 -

Literature Review

investigated the impact of water stress on the sap wood of Callgonum comosun , and found that the drought makes longer and narrower fibres. MOYA and FO (2008) studied the variation in the wood anatomical structure of Gmelina arborea grown at different ecological condition, and found that precipitation has significant effect on fibre diameter and lumen diameter. As the precipitation increases fibre diameter and lumen diameter increase while no effect on fibre length and fibre wall thickness has been found.

2.2.1.2 Physical properties

Water supply may also affect wood density. Generally the differences in wood density are associated with differences in wood anatomical structure. Several authors (ZHANG and MORGENSTERN 1995 and HANNRUP et al. 1998) found genetic and environmental variation in basic density due to differences in cell sizes and cell wall thickness.

Some authors such as BEADLE et al. (2001), WIMMER et al. (2002), ROQUE (2004), Al- KHALIFAH et al. (2006) and NAIDOO et al. (2006) found that water stress positively affects the wood density in their studied species. For instance, the Gmelina arborea wood produced in the dry tropical region has higher specific gravity than wood growing in the humid region as reported by ROQUE (2004). Likewise, AL-KHALIFAH et al . (2006) in their study of the impact of water stress on the sap wood of Callgonum comosun found that drought leads higher density. NAIDOO et al. (2006) found the same trend in Eucalyptus grandis . In the genus Eucalyptus , an increase in wood density has been associated with decreased soil water supply in Eucalyptus globulus Labill. (WIMMER et al. 2002) and E. nitens (Deane&Maiden) Maiden trees (BEADLE et al. 2001).

SEARSON et al. (2004) found significant increase of wood density due to water stress in both Eucalyptus grandis and E. sideroxylon seedlings, while no effect appeared in E. occidentalis . However, when ethanol and water-soluble extractives were removed from the wood, there was no difference in wood density between the well-watered and water-limited E. sideroxylon seedlings. Other authors found no relationship between wood density and water stress in Grandis globulus (MACFARLANE and ADAMS 1998 and CATCHPOOLE et al. 2000).

- 37 -

Literature Review

2.2.2 Air temperature

Temperature is also considered to be an important factor influencing wood properties. The response of wood anatomy and density to the elevated temperature is variable (KOSTIAINEN 2007). For instance, a negative relationship between temperature and vessel frequency have been observed in Eucalyptus camaldulensis (Dehn.) seedling (THOMAS et al. 2004) and E. grandis (W. Hill ex Maid.) seedling (THOMAS et al. 2007), while in European beech ( Fagus sylvatica L.) no change has been observed (OVERDIECK et al. 2007). Similarly, the temperature has negative effects on the wood vessel area, as observed in European chestnut ( Castanea sativa ) (FONTI and GARCIA-GONZALES 2004) and in Eucalyptus grandis seedlings (THOMAS et al. 2007). In Eucalyptus grandis seedlings, an increase of fibre wall thickness and a decrease of fibre lumen diameter with increasing temperature were also observed (THOMAS et al. 2007).

Temperature is considered to be an important factor influencing wood density. Positive relationships between wood density and temperature have been observed by several researchers in several species. For instance, in Eucalyptus camaldulensis and E. grandis seedlings (THOMAS et al. 2004, 2007), Eucalyptus dunnii mature maiden trees (MUNERI et al. 2004), Larix sibirica Ldb. (Larch) (ANTONOVA and STASOVA 1997), Picea abies (L.) Karst.) (Norway spruce) (KOSTIAINEN 2007), Pinus sylvestris L . (Scots pine) (ANTONOVA and STASOVA 1993 and KILPELÄINEN et al. 2005) and Pinus radiata (HARRIS 1965 and WILKES 1987).

Temperature is thought to influence wood density via the general influence of temperature on plant growth (CREBER and CHALONER 1984). More recently, it has been proposed that the mechanism by which temperature influences wood density is via its impact on the viscosity of water (RODERICK and BERRY 2001 and THOMAS et al. 2004). RODERICK and BERRY (2001) suggested that fewer voids would be required to transport water to the canopy at higher temperatures, when water has a lower viscosity. As a result, the volume of lumen (voids) per unit volume of wood would be reduced, resulting in an increase in wood density.

2.2.3 Other factors

In addition to water supply and temperature, there are many other factors which can affect the wood properties such as site, geographical location (latitude, longitude and altitude), growth rate, etc. For instance, in Gmelina arborea , many studies indicate that the variations in wood

- 38 -

Literature Review

anatomy occurred in relation to tree age (AKACHUKU and BURLEY 1979, AKACHUKU 1985 and OHBAYASHI and SHIOKURA 1989), growth rate (ESAN 1966, LAMB 1968, HUGHES and ESAN 1969, OHBAYASHI and SHIOKURA 1989), differences in site fertility (AKACHUKU and BURLEY 1979 and OGBONNAYA 1993).

Researchers such as AKACHUKU and BURLEY (1979) and AKACHUKU (1985) demonstrated that vessels percentage was negatively correlated with latitude, longitude and growth rate. Growth rate affects fibre dimensions too. OHBAYASHI and SHIOKURA (1989) carried out a study on fibre length in 15-year-old trees and found that a high growth rate was strongly correlated with short fibre length. ROQUE and FO (2007), in their study of wood density and fibre dimensions of Gmelina arborea in fast growing trees in Costa Rica, found that the minimum and mean density, cell wall thickness, fibre length, width and lumen diameter decreased with increase in growth rate. ZOBEL and VAN BUIJTENEN (1989) established that cell dimensions in many wood species decreased because of rapid growth in trees. This is caused by an increase of cell division rate in the cambium. However, studies in eucalyptus species disagreed with these findings (WILKES and ABBOTT 1983).

Geographical location may also have effect on wood anatomical properties. For instance, NOSHIRO and BAAS (2000) found that vessel diameters in two of three species of Cornus were negatively correlated with altitude (wood samples from higher elevations had narrower vessels). Likewise, FISHER et al. (2007) found that high elevation plants have significantly narrower vessels than lower elevation plants in Metrosideros polymorpha . Among 31 species of Symplocos, there was a similar but weaker negative correlation between a specimen’s altitude and its vessel diameter (VAN DEN OEVER et al. 1981). DICKISON and PHEND (1985) associated multiseriate rays to higher latitudes in Styracaceae, while CARLQUIST (1966) associated it to lower latitudes in Compositae. VAN DEN OEVER et al. (1981) found thinner-walled fibres at higher latitudes in Symplocos . ZHANG et al. (1988) related thicker wall fibres to lower altitudes in Syringa oblata .

Regarding the effect of growth rate upon wood density, institution or common sense might suggest that wood density should decrease if the rate of growth of a tree increases. This is not necessarily or even usually the case. When wood density is related to growth rate, the response depends on the species and the range of growth rates involved. Other factors such as age, vitality of the tree when the wood was produced, and location of wood in the tree are more closely correlated to density than the rate of growth. The only group of tree species, in which density are closely related to growth rate are ring-porous hardwoods. In this species

- 39 -

Literature Review

the density tends to increase as the growth rates increases. Diffuse-porous hardwoods do not exhibit a consistent relationship between density and growth rate. In softwoods density decreases with increasing growth rate.

Faster growth rates in gamhar (LAURIDSEN and KJAER 2002) and silver birch, Betula pendula , (DUNHAM 1999) have been correlated with lower wood density. Teak ( Tectona grandis ) (BHAT and PRIYA 2004) and White ash ( Fraxinus pennsylvanica) (BLANKENHORN et al. 2005) produced denser wood when grown slowly. No relation between growth rate and wood density has been observed in Young Populus (DEBELL et al. 2002) nor in seven year old Red alder trees (LEI et al. 1997).

The factors which may affect the wood properties are multiple and could not be covered in this section; the above mentioned correlations provide just some examples.

Despite the abundance information found in literature about the effect of climate on the wood properties of several hardwood and softwood species, no information was found about the effect of the climate on the wood properties of Acacia seyal and Balanites aegyptiaca .

- 40 -

Literature Review

2.3 Description, distribution and uses of the study species

2.3.1 Acacia seyal var. seyal Del.

2.3.1.1 Taxonomy and nomenclature

The definition of a taxon is encapsulated by its description. In the fields of botany, phycology, and mycology, the naming of taxa is governed by the International Code of Nomenclature for algae, fungi, and plants. The taxonomic descriptions of Acacia seyal var. seyal can be taken from the following overview:

Kingdom: Plantae Division: Magnoliophyta Class: Magnoliopsida

Order : Fabales Family: Fabaceae Subfamily: Mimosoideae

Genus: Acacia Species: A. seyal Varity: seyal

Note: The generic name ‘ acacia ’ comes from the Greek word ‘akis’, meaning ‘point’ or ‘barb’.

In biological science, at least, nomenclature is regarded as a part of (though distinct from) taxonomy. ORWA et al. (2009) gave the following definitions for Acacia seyal var. seyal :

Local names: English names: Trade name:

Talh, Suffar ahmer Red Acacia, Shittim wood, White-galled Gum talha, Shittimwood acacia, Whistling thorn, White thorn, White (used in Sudan), Soffa whistling

2.3.1.2 Botanical description

Acacia seyal var. seyal is a deciduous, small to medium-sized tree with 3 - 10 m height (EL AMIN 1990 and ORWA et al. 2009); it may reach a height of 12 - 17 m (VON MYDELL 1990, HALL and MCALLAN 1993, MCALLAN 1993, and ARBONNIER 2004). At maturity Acacia seyal usually reaches 9 – 10 m in height, its diameter at breast height reaches 30 cm (MUSTAFA 1997) or 60 cm under favorable conditions (VON MYDELL 1990, ARBONNIER 2004, and ORWA et al. 2009). Crown is umbrella shaped (see Figure 2.5). Its bark is powdery, smooth or sparsely flaking, orange red, with a green layer beneath (EL AMIN 1990, VON MYDELL 1990, and ORWA et al . 2009).

- 41 -

Literature Review

Leaves are bipinnate, dark green with 4 – 12 pairs of pinnae; alternate, each pinnae with 10 –12 leaflets (ORWA et al. 2009). Other author cited that its leaves have 3 – 9 pinnae (EL AMIN 1990). Acacia seyal has straight thorns in pairs, up to 8 cm long. Flowers clustered in shining, yellow, globose heads. Fruits are falcate, dehiscent pods, light browm in colour having 6 – 10 seeds per pod (EL AMIN 1990 and ORWA et al. 2009).

Figure 2.5: General features of Acacia seyal var . seyal. 1. Adult tree, 2. Powdery red bark, 3. Leaves, fruits and flowers, 4. Wood (cross section)

2.3.1.3 Wood properties

The wood colour varies from white-cream (THIRAKUL 1984) to pale yellow or medium brown, with localized pinkish-brown patches and some dark mahogany-red heartwood in larger or older individuals (ORWA et al. 2009). NEUMANN et al. (2000) described the anatomical features of Acacia seyal as follow:

- Growth ring boundaries: Distinct.

- Vessels: Diffuse-porous. Vessels arranged in no specific pattern. Vessels in multiples. Vessels commonly short (2–3 vessels) radial rows. It is rounded in shape. Average tangential vessel diameter 30–170 µm. Average tangential diameter of vessel lumina

- 42 -

Literature Review

is large. Average number of vessels/mm² 4–20 (classified as “few”). Average vessel element length short. Perforation plates simple. Intervessel pits alternate. No tyloses in vessels.

- Tracheids and fibres: Vascular or vasicentric tracheids sporadic to absent. Fibres of medium wall thickness; or very thick-walled. Fibre pits mainly restricted to radial walls. Fibre pits simple to minutely bordered. Fibres non-septate.

- Axial parenchyma: Present, banded, bands marginal (or seemingly marginal, much wider than rays. Axial parenchyma apotracheal (diffuse); or paratracheal (vasicentric; or aliform; or confluent; or unilateral paratracheal). There are Crystals located in axial parenchyma cells.

- Rays: Multiseriate (1–10 cells in width). Rays commonly 5–10 seriate; or commonly more than 10 seriate. Height of large rays commonly 500 to 1000 µm; or commonly over 1000 µm. Rays composed of a single cell type. Homocellular ray cells procumbent.

- Storied structures: Storied structure absent.

- Secretory structures: Oil and mucilage cells absent. Intercellular canals absent. Laticifers or tanniniferous tubes absent.

- Cambial variants: Included phloem absent. Other cambial variants absent.

- Mineral inclusions: Crystals present. Crystals prismatic. Crystals located in axial parenchyma cells. Crystal-containing axial parenchyma cells chambered. Number of crystals per cell or chamber: one. Crystal containing cells of normal size; or enlarged (idioblasts). Cystoliths absent. Crystal diameter 7–50 µm. Silica not observed.

In addition to NEUMANN et al. (2000), some other authors confirm the existence of growth ring boundary in Acacia seyal (GEBREKIRSTOS et al. 2008, NICOLINI et al. 2010, MBOW et al. 2013). The previously mentioned authors concluded that the growth ring boundary of Acacia seyal was delimited by marginal parenchyma. TARHULE and HUGHES (2002) have identified three categories of species in terms of their potential for dendrochronology: (1) potentially useful, (2) problematic, and (3) poor. The categorization is based on the distinctiveness of the annual ring boundaries, the ability for cross dating, ring circuit uniformity, ring wedging, and ring width variability. Those for which one can expect

- 43 -

Literature Review

to obtain good results fall into the potentially useful category. Acacia seyal was classified as potentially useful species by NICOLINI et al. (2010) and MBOW et al. (2013). The authors concluded that Acacia seyal forms one tree ring per year, i.e. annual rings. Various tree ring width has been detected for Acacia seyal. For instance , annual ring width varies from less than 2 mm to more than 16 mm with an overall mean of 6.47 mm as revealed by NICOLINI et al. (2010). Range of 0.27– 9.12 mm with a mean of 2.32 ± 0.66 mm according to GEBREKIRSTOS et al. (2008). NICOLINI et al. (2010) reported that the variation in tree ring width may be attributed to the precipitation.

NEUMANN et al. (2000) observed crystals in Acacia seyal . Several authors such as FAHN et al. (1986), CARLQUIST (1988) and JOHN (1990) concluded that the existence of crystals may be due to the influence of aridity. The existence of tyloses is also apparently a response to a loss of water in the vessels (ZIMMERMANN 1983).

SHAWGI (2007) and YOUSIF (2000) reported that Acacia seyal has non septate fibres. Some authors studied the fibre characteristics of Acacia seyal and found different values as shown in Table 2.18. Acacia seyal’s wood basic density found to be 649 kg/m³ or 669–692 kg/m³ as revealed by KHRISTOVA et al. (1998) and KHRISTOVA et al. (2004), respectively. Its air dry density is found to be 0.66 g/cm 3 and its oven dry is 0.60 g/cm 3 as revealed by OSMAN (2001).

Table 2.18: Fibres characteristics of Acacia seyal var. seyal

References Fibres characteristics 1KHRISTOVA 1YOUSIF 2OSMAN 1KHRISTOVA *1SHAWGI et al. (1998) (2000) (2001) et al. (2004) (2007) Length (mm) 1.2 1.090 1.070 0.8 1.086 Diameter (µm) 14.8 18.0 14.7 13.8 18.8 Lumen diameter (µm) 4.7 - 12.2 - 7.87 Wall thickness (µm) 5.0 - 2.6 2.3 5.49 Flexbility coefficient (%) 32 - - 77 42 Runkel ratio 2.1 - - 0.3 1.65 Slenderness ratio 77 - - 64 59.9

1 = trees collected from Blue Nile state in Sudan, 2 = trees collected from White Nile state in Sudan, *SHAWGI (2007) measured the fibre dimension in the middle point of fibre length, thereby, the obtained values considered as maximum values.

2.3.1.4 Distribution and uses

ORWA et al. (2009) described the Acacia seyal ’s distribution over the world as native or exotic tree as follow:

- 44 -

Literature Review

Native: Egypt, Eritrea, Ethiopia, Ghana, Iran, Israel, Kenya, Malawi, Mali, Mozambique, Namibia, Niger, Nigeria, Saudi Arabia, Senegal, Sudan, Syrian Arab Republic, Tanzania, Uganda, Yemen, Republic of, Zambia, Zimbabwe.

Exotic: Afghanistan, Bangladesh, Bhutan, India, Nepal, Portugal, Sri Lanka, USA (see Figure 2.6).

Figure 2.6: Acacia seyal var. seyal’s distribution over the world (after ORWA et al . 2009)

In Sudan, it is found on dark cracking clays on higher slopes of rivers and valleys on the hard clay plains of Central Sudan and on clay of seasonally wet depressions (EL AMIN 1990). It also extends from Gadarif, Blue Nile, and White Nile to clay plains around Nuba Mountains and the Darfur Region (SAHNI 1968, EL AMIN 1990, and MUSTAFA 1997) (see Figure 2.7).

Acacia seyal var . seyal tree requires annual rainfall of 250–1000 mm (NAS 1980, VON MYDELL 1990, and ORWA et al. 2009); and annual temperature of 18 –28ºC (VON MYDELL 1990), or 15 –35ºC as cited in VOGT (1995). The species thrives in most soil types, but normally it prefers heavy, clayey soils, stony gravely alluvial soils or humic soils (NAS 1980, McALLAN 1993 and ORWA et al. 2009).

- 45 -

Literature Review

Figure 2.7: Acacia seyal var. seyal’s distribution in Sudan. Adapted by the author from SOURCE2

Acacia seyal var. seyal ’s wood is used widely throughout its range as an important source of rural energy as both firewood and charcoal (VOGT 1995 and ORWA et al. 2009). The best fuelwood in Chad is provided from A. seyal var. seyal’s wood, also in Sudan it provides the best charcoal, and used to make a fragrant fire over which women perfume themselves (ORWA et al. 2009). The timber is used in construction, light furniture in rural areas, but is very susceptible to insect attack and is little used (VOGT 1995 and WICKENS et al. 1995). The timber works well and is hard and tough. It produces a hard, dark wood, called shittim wood, with interlocked, irregular, and coarse textured grain. It was used by ancient Egyptians for pharaohs’ coffins (ORWA et al. 2009).

Bark and pods contain around 18 –20% tannin, and is used for tanning leather in Sudan (WICKENS et al. 1995). Also it produces gum which is called talha gum; it is darker and inferior in quality to that of Acacia senegal (gum arabic). However, it forms 10 % of the annual Sudanese exported gum (BARBIER et al. 1990, McALLAN 1993, ORWA et al. 2009). The bark, leaves, and gums are used for the treatment of colds, diarrhea, hemorrhage, jaundice, headache, and burns. A bark decoction is used against leprosy and dysentry, it is a stimulant, and acts as a purgative for humans and animals. Exposure to smoke is believed to relieve rheumatic pains. A root decoction mixed with leaves of Combretum glutinosum and curdled milk causes strong diuresis (ORWA et al. 2009).

- 46 -

Literature Review

2.3.2 Balanites aegyptiaca (L.) Delile.

2.3.2.1 Taxonomy and nomenclature

The taxonomic descriptions of Balanites aegyptiaca can be found in the following overview:

Kingdom: Plantae Division: Magnoliophyta Class: Magnoliopsida

Order: Zygophyllales Family: Balanitaceae Subfamily: Tribuloideae (previously placed in Zygophyllaceae , Simaraoubaceae ) (NAS 1983)

Genus: Balanites Species: B. aegyptiaca Varity : aegyptiaca

ORWA et al. (2009) gave the following definitions for Balanites aegyptiaca:

Local names: English names: Trade name: In Sudan its name is Heglig Desert date, soap berry tree, Desert date (dried fruit); or Hejlij, fruit names is Jericho, balsam and thorn heglig berries (in the Sudan), lalob tree Egyptian myrobalan)

Note: The name Balanites (from the Greek for acorn, referring to the fruit) was given in 1813 by Alire Delile and replaced Agialid (derived from the Arabic name for the tree, 'heglig'). There are five varieties of Balanites aegyptiaca distributed throughout Africa (SANDS 2001). The only variety recorded to appear in Sudan is variety aegyptiaca (EL FEEL 2004).

2.3.2.2 Botanical description

Balanites aegyptiaca is an evergreen species, which loses its leaves only during very dry period (SULIMAN and JACKSON 1959). It is a multi-branched and armed tree, that varies in height from 8 to 10 m (EL AMIN 1990) or 15 m and in diameter from 30 to 50 cm (THIRAKUL 1984). Crown spreading usually with dense drooping branches (VOGT 1995). The bole is mostly fluted. Bark grey to dark brown; older trees are recognizable by the dark bark, which has deep vertical fissures running the length of the trunk (THIRAKUL 1984 and EL AMIN 1990). Its leaves are with two separate leaflets; leaflets obovate, asymmetric, 2.5– 6 cm long, bright green, leathery. Flowers in fascicles in the leaf axils, fragrant, yellowish- green (THIRAKUL 1984 and EL AMIN 1990); some trees have thorns with leaves (THIRAKUL 1984 and EL AMIN 1990). Fruits are egg-shaped, green turning yellow or brown when mature, Pulp bitter-sweet and edible (EL AMIN 1990 and VOGT 1995); see Figure 2.8.

- 47 -

Literature Review

DDC C B

C Figure 2.8: General features of Balanites aegyptiaca. 1. Adult tree; 2. Fissured bark; 3. Leaves and flowers; 4. Green immature fruit and yellow mature fruit, the photo shows also the thorns; 5. Wood (cross section showing the fluted stem)

2.3.2.3 Wood properties

Wood pale yellow or yellowish-brown, coarse grained; hard; durable and resistant to insects (THIRAKUL 1984 and ORWA et al. 2009). Heartwood and sapwood are not clearly differentiated (ORWA et al . 2009).

NEUMANN et al. (2000) described the anatomical features of Balanites aegyptiaca as follow:

- Growth ring boundaries: Indistinct or absent. Zones with different vessel density present but not clearly representing growth rings.

- Vessels: Diffuse-porous. Vessels arranged in no specific pattern; Vessels in multiples; rounded in shaped. Average tangential vessel diameter 50–200 µm. Average tangential diameter of vessel lumina medium; or large. Vessels per square millimetre

- 48 -

Literature Review

few. Perforation plates simple. Intervessel pits alternate. Intervessel pits average diameter (vertical) 3 - 5 µm. Vessel-ray pits with distinct borders. Helical thickenings absent. Tyloses in vessels absent.

- Tracheids and fibres: Vascular or vasicentric tracheids commonly present. Fibres of medium wall thickness. Fibre pits common in both radial and tangential walls. Fibre pits distinctly bordered. Fibres non-septate.

- Axial parenchyma: Axial parenchyma present; apotracheal (diffuse; or diffuse-in- aggregates); or paratracheal (scanty).

- Rays: Rays 2–3 per tangential mm; multiseriate; composed of 10–18 cells in width. Height of large rays commonly 500 to 1000 µm; or commonly over 1000 µm. Rays composed of two or more cell types. Heterocellular rays square and upright cells restricted to marginal rows.

- Storied structures: Storied structure present in axial parenchyma, vessel elements and fibres. - Secretory structures: Oil and mucilage cells absent. Intercellular canals absent. Laticifers or tanniniferous tubes absent. - Cambial variants: Included phloem absent. Other cambial variants absent. - Mineral inclusions: Crystals present. Crystals prismatic. Crystals located in ray cells; or axial parenchyma cells. Crystal-containing ray cells upright and/or square; or procumbent. Crystal-containing upright and/or square ray cells chambered. Crystals in procumbent ray cells in radial alignment. Crystal-containing axial parenchyma cells chambered. Number of crystals per cell or chamber: one. Crystal containing cells of normal size. Cystoliths absent. Silica not observed.

Also an anatomical description depending upon IAWA List of Microscopic Features for Hardwood Identification (IAWA committee), has been provided in Inside wood database by FAHN et al. (1986) as follows:

2 Growth ring boundaries indistinct 5 Wood diffuse-porous 7v Vessels in diagonal and / or radial pattern 11 Vessel clusters common 13 Simple perforation plates 22 Intervessel pits alternate

- 49 -

Literature Review

24 Minute - <= 4 µm Vessel-ray pits with distinct borders; similar to intervessel pits in size and shape 30 throughout the ray cell 42 100 - 200 µm 46 <= 5 vessels per square millimetre 47 5 - 20 vessels per square millimetre 52 <= 350 µm 60 Vascular / vasicentric tracheids present 61 Fibres with simple to minutely bordered pits 62 Fibres with distinctly bordered pits 63 Fibre pits common in both radial and tangential walls 66 Non-septate fibres present 70 Fibres very thick-walled 72 900-1600 µm 76 Axial parenchyma diffuse 77 Axial parenchyma diffuse-in-aggregates 78 Axial parenchyma scanty paratracheal 86 Axial parenchyma in narrow bands or lines up to three cells wide 90 Fusiform parenchyma cells 91 Two cells per parenchyma strand 99 Larger rays commonly > 10-seriate 102 Ray height > 1 mm 104 All ray cells procumbent 106 Body ray cells procumbent with one row of upright and / or square marginal cells 110 Sheath cells 114 <= 4 / mm 120 Axial parenchyma and / or vessel elements storied 131 Intercellular canals of traumatic origin 136 Prismatic crystals present 140 Prismatic crystals in chambered upright and / or square ray cells 142 Prismatic crystals in chambered axial parenchyma cells 154v More than one crystal of about the same size per cell or chamber *Notes: number followed by the letter v means that the feature is variable

In contrary with the anatomical description provided by FAHN et al. (1986) and NEUMANN et al. (2000), other authors detected growth ring boundary in Balanites aegyptiaca (PRARAMESWARAN and CONRAD 1982, GEBREKIRSTOS et al. 2008, MBOW et al. 2013). The growth ring in Balanites aegyptiaca is delimited by a combination of marginal parenchyma and accumulation of vessels according to GEBREKIRSTOS et al. (2008) and MBOW et al. (2013). GEBREKIRSTOS et al. (2008) revealed range of 0.34–5.25 mm

- 50 -

Literature Review

growth ring width in Balanites aegyptiaca with overall means of 1.93 ± 0.30 mm. Concerning the Dendrochronological potential, MBOW et al. (2013) classified Balanites aegyptiaca as problematic and poor where the tree-ring boundary detection is difficult and not useful for dendrochronology.

GEBREKIRSTOS et al. (2008) as well as MBOW et al. (2013) observied that the growth ring boundry is more distinctive in Acacia seyal in comparison with Balanites aegyptiaca. A tendency for distinct ring formation by deciduous species compared to evergreen species was reported by (WORBES 1999), and distinctiveness was related to wood structure differences, which is genetically controlled (DÉTIENNE 1989 and WORBES 1995). GEBREKIRSTOS et al. (2008) cocludied that frequent occurrences of wedging rings in Balanites aegyptiaca made the determination of growth boundaries more difficult. In some species the wedging rings occured in the outer part near the bark as observed in Terminalia guianensis by WORBES (2002) where the wedging rings were observed in the outer part when the tree starts to form buttresses.

PARAMESWARAN and CONRAD (1982), in their study of Balanites aegyptiaca reported values of 105 µm vessels diameter and a range of 4–10 µm fibre wall thickness. In studies of Balanites aegyptiaca ’s fibre characteristics , some authors found different values as shown in Table 2.19. SHAWGI (2007) reported that Balanites aegyptiaca has non septate fibres. Its oven dry density estimated by 0.79–0.84 g/cm³ (PARAMESWARAN and CONRAD 1982). KHRISTOVA et al. (1997) found the wood basic density to be 619 kg/m³, while OSMAN (2001) found its air dry and oven dry densities to be 0.66 g/cm 3 and 0.60 g/cm 3, respectivley.

Table 2.19: Fibres characteristics of Balanites aegyptiaca

References 1 PARAMESWARAN 2 3 4 Fibres Characteristics KHRISTOVA OSMAN * SHAWGI and CONRAD et al. (1997) (2001) (2007) (1982) Length (mm) 1.160 1.11 1.175 1.038 Diameter (µm) - 16.07 12.3 20.6 Lumen diameter (µm) 2.5-8.00 4.06 10.5 10.04 Wall thickness (µm) 2.5-6.00 6.01 1.9 5.30 Flexbility coefficient (%) - 25.3 - 49 Runkel ratio - 2.9 - 1.14 Slenderness ratio - 69.1 - 51.9 1 = trees collected from Ouagadougou in Burkina Faso, 2 = trees collected from Blue Nile state in Sudan, 3 = trees collected from White Nile state in Sudan, 4 = trees collected from South Kordofan state in Sudan, *SHAWGI (2007) measured the fibre dimension in the middle point of fibre length, thereby, the obtained values considered as maximum values

- 51 -

Literature Review

2.3.2.4 Distribution and uses

According to ORWA et al. (2009), the distribution of Balanites aegyptiaca over the world can be summarized as follow (see Figure 2.9):

Native: Algeria, Angola, Benin, Burkina Faso, Burundi, Cameroon, Chad, Cote d'Ivoire, Democratic Republic of Congo, Djibouti, Egypt, Eritrea, Ethiopia, Gambia, Ghana, Guinea, India, Israel, Kenya, Libyan Arab Jamahiriya, Morocco, Myanmar, Nigeria, Saudi Arabia, Senegal, Somalia, Sudan, Tanzania, Uganda, Yemen, Republic of, Zambia, Zimbabwe.

Exotic: Cape Verde, Dominican Republic, Puerto Rico.

Figure 2.9: Balanites aegyptiaca’s distribution over the world (after ORWA et al. 2009)

Balanites aegyptiaca is an indigenous species in Sudan with a wide range of natural occurrence over diverse climatic and edaphic conditions (SULIMAN and JACKSON 1959). It is an evergreen tree adapted to various climatic conditions especially in arid regions with extremely high temperatures and scarce water; thus it was advised to be promoted for combating desertification (GOUR and KANT 2012). So, the tree is widely distributed in the Sudan (see Figure 2.10) under rainfall of less than 1000 mm, with higher occurrence in zones of 200–800 mm (BADI et al. 1989) and requires mean annual temperature of 20–30ºC (ORWA et al. 2009). It grows in sand, clay, cracking clay and gravel soils (SULIMAN and JACKSON 1959 and ORWA et al. 2009). It is found in Central Sudan, often associated with Acacia seyal on short grass Savanna (EL AMIN 1990).

- 52 -

Literature Review

Figure 2.10: Balanites aegyptiacal’s distribution in Sudan. Adapted by the author from SOURCE2

The wood makes excellent firewood and good quality charcoal (THIRAKUL 1984); it is also used for local furniture, agricultural implements, joinery, and walking sticks. The fruit is edible and the maceration of the fruit is used against constipation and as anti-diabetic. The sapogenin, yamogenin, and diosogenin, which are extracted from all parts of the plant, are used in partial synthesis of steroidal drugs (VOGT 1995).

- 53 -

Material

3 Material

This chapter provides an overview of the study areas, highlighting the locations, and providing information about their climatic condition. Description of the selected trees is also provided in this chapter.

3.1 Study area

The trees species were collected from 11 natural forests located in four states in Sudan. The selection criteria’s for forests within each state were the availability and accessibility of the Acacia seyal and Balanites aegyptiaca trees. The location of the study areas are illustrated in Figure 3.1.

Figure 3.1: Locations of the study areas.1. Al Tainat Forest; 2. Al Ein Forest; 3. Habila Forest; 4. Al Homora Forest; 5. Aum Top Forest; 6. Goz Fagor Forest; 7. Tawla Forest; 8. Khor Donia Forest; 9. Al Sheheata Forest; 10. Al Homara Forest; 11. Al Bardab Forest. Adapter by the author from SOURCE3.

- 54 -

Material

According to the mean annual rainfall for ten years (2000–2009) (Table 3.1), the study areas were divided into two zones:

- Zone one: the overall mean annual rainfall for ten years (2000–2009) in this zone is 273 mm, calculated as an average of the mean annual rainfall of the two selected locations within this zone (see Table 3.1).

- Zone two: this zone has a relatively high rainfall in comparison with zone one. The mean annual rainfall in this zone is 701 mm calculated as an average of the annual rainfall for ten years (2000–2009) of the two selected locations within this zone.

Table 3.1: Mean annual rainfall (mm) and temperature (ºC) for ten years (2000-2009) and their average in the study locations. R = Mean annual rainfall (mm); T = Mean annual temperature (ºC); N.K = North Kordofan State; W.N = White Nile State; B.N = Blue Nile State; S.K = South Kordofan State and Loc. = Location.

Years Average Location Per Per Zo ne Zone Zone 2000 2001 2002 2003 2004 2005 2006 2007 2008 2009 loc. R T Al Obeid R 314.5 348.7 216 406.6 285.9 254.1 490.7 557.7 411.2 291.2 357.7 (N.K) T 27.5 27.5 28.2 28.3 28.2 28.4 27.8 27.6 27.7 28.6 28.0 273 28.9

One Al Duwaim R 194.5 270.2 263.3 275.8 219.6 172.2 165.8 313.7 0.0 0.0 187.5 (W.N) T 29.0 29.5 29.7 29.9 30.0 30.0 30.0 29.5 29.7 30.6 29.8 Al Damazien R 874.8 774 585.3 696.7 691.6 684.1 632.9 762.3 877 529 710.8 (B.N) T 28.4 28.4 29.1 29.1 29.0 29.7 28.8 28.7 28.7 29.5 28.9 701 32.4

Two Kadugli R 463.2 810.2 828.4 670.7 657.7 658.1 712.5 772.3 642.2 539.1 675.4 (S.K) T 35.2 36.1 36.0 35.9 36.0 36.0 35.3 35.2 35.3 36.6 35.8

Note: Information on rainfall amounts and temperature were obtained by the author in the year 2010 from Khartoum meteorological sation for Al Obeid and Al Duwaim locations, from Kadugli meteorological station for Kadugli location and from the Ministry of Agriculture, Livestock and Irrigation General Department of Agriculture - Department of Planning and Information, Blue Nile, for Al Damazien location.

Figures 3.2, 3.3, 3.5 and 3.6 depict the climatic diagrams (2000–2009) for the study locations. These climatic diagrams show the curves for average monthly temperatures in °C versus the average monthly rainfall in mm with a ratio of 1: 2. This means, for instance, that the distance along the ordinates is the same for 20 mm precipitation and 10°C temperature. At this ratio, times during which the precipitation curve is above the temperature curve are considered humid, while the remaining periods are considered arid as described by SCHULTZ (1995).

3.1.1 Zone one (with 273 mm mean annual rainfall)

This zone is represented by North Kordofan State as well as White Nile State. In North Kordofan State , the tree species were collected from two forests located in Al Obeid city. The rainfall in this city commences in May and lasts until October, with a peak in July and August. The dry season extends from November to April (Figure 3.2).

- 55 -

Material

The two selected forests in this city are:

‹ Al Ein Forest: it is a reserved forest located 26 km south west of Al Obeid city, between latitudes 12° 54′ 58" and 12° 59′ 37" N and longitudes 30° 14′ 19" and 30° 18′ 51" E. The forest is dominated by Acacia mellifera , A. nilotica and A. seyal var. seyal. Terminalia brownie , and Adansonia digitata are also found .

‹ Al Tainat Forest: a reserved forest located in the northern part of Al Obeid city (approximately 42 km from Al Obeid), at 13° 32′ 03" and 13° 35′ 16" N latitude and 30° 10′ 50" and 30° 13′ 54" E longitude. The main tree species in this forest are Acacia senegal , A. tortilis , A. nubica , Balanites aegyptiaca , Faidherbia albid, leptadenia pyrotechnica and Ziziphus spina-christi.

Figure 3.2: Climatic diagram (2000-2009) for Figure 3.3: Climatic diagram (2000-2009) for Al Obeid locality, North Kordofan Al Duwaim locality, White Nile

The White Nile State was represented by four reserved natural forests located in Al Duwaim city. A small amount of rain falls in May and June, becoming more intense in July and August, decreasing again in September and October, and the dry season starts in November and extends until April (Figure 3.3).

The four selected forests in this city are:

‹ AL Homora Forest: located in the western part of Al Duwaim.

‹ Aum Top Forest: located southwest of Al Duwaim.

‹ Goz Fagor Forest: located in the south western part of Al Duwaim.

- 56 -

Material

‹ Habila Forest: located northwest of Al Duwaim.

The dominant trees species in these four forests are Acacia seyal var. seyal , A. tortilis , Balanitea aegyptiaca and Zizphus spina-christi . Figure 3.4 presents some of the selected forests in zone one.

Figure 3.4: Different forests in zone one. 1; 2. Al Tainat Forest; 3. Al Ein Forest in Al Obeid, North Kordofan State, and 4. AL Homora Forest; 5. Goz Fagor Forest; 6. Aum Top Forest in Al Duwaim, White Nile State.

- 57 -

Material

3.1.2 Zone two (with 701 mm mean annual rainfall)

South Kordofan state and Blue Nile state were selected to represent this zone. In South Kordofan State , the tree species were collected from three forests located in Kadugli city. The rainy season starts in April and May with a relatively light rain becoming intense from June to September, decreasing in October, and then the dry season lasts for five months, starting from November until March (Figure 3.5). The selected forests are:

‹ Al Sheheata Forest: it is a reserved forest located 40 km northeast of Kadugli city, at 11° 13′ 56.4" and 11° 14′ 07.1" N latitude and 30° 05′ 5.3" and 30° 05′ 25.1" E longitude. The forest includes many species, some of which are planted, such as Azadirachta indica , Ceiba pentandra , Tectona grandis , Khaya senegalensis. Dalbergia session , Eucaliptus spp. etc., and others of which are growing naturally, such as Acacia seyal var . seyal , Ailanthus exelsa and Balanites aegyptiaca.

‹ Al Homara Forest: a non-reserved natural forest located 15 km southeast of Kadugli (10° 56′ N latitude and 29° 49′ E longitude), dominated by Acacia seyal var. seyal and Balanites aegyptiaca . Other tree species such as Acacia nilotica and A. senegal are also present .

‹ Al Bardab Forest: a non-reserved natural forest located 20 km northwest of Kadugli at 11° 12′ N latitude and 29° 38′ E longitude. It is dominated b y Acacia seyal var. seyal and Balanites aegyptiaca. Anogeissus leiocarpus , Combretum hartmannianum and Sclerocarya birrea are also present.

Figure 3.5: Climatic diagram (2000-2009) for Figure 3.6: Climatic diagram (2000-2009) for Al Kadugli Locality, South Kordofan Damazien locality, Blue Nile

- 58 -

Material

In Blue Nile State, Al Damazien city was selected to collect the tree species from two reserved natural forests. The rainfall in this city continues from May until October with 6 months of drought extending from November until April (Figure 3.6).

The selected two forests are:

‹ Khor Donia Forest: a reserved forest located southwest of Al Damazien city at latitude 11º 00' and longitude 34º 05'. It is dominated by Acacia seyal var . seyal , A. Polycantha , A. senegal and Balanites aegyptiaca .

‹ Tawla Forest: it is located northeast of Al Damazien city (12º 17' N latitudes and 34º 21' E) and is dominated by Acacia seyal var. seyal , A. senegal and Balanites aegyptiaca .

The selected five forests in this zone are presented in Figure 3.7.

- 59 -

Material

Figure 3.7: Different forests in zone two. 1. Al Bardab Forest; 2. Al Sheheata Forest; 3. Al Homara Forest in Kadugli, South Kordofan Stae, and 4. Tawla Forest; 5. Khor Donia Forest in Al Damazien, Blue Nile State.

- 60 -

Material

3.2 Trees selected

ZOBEL and BUITENEN (1989) stated that the variation in wood properties from the tree to tree within species is large. While no species has uniform wood, some are much more uniform than others. Therefore, they suggested sampling at least 30 trees to get a reliable estimate of the average properties of a stand. HAPLA (1993) in a study of the experimental planning in wood research concluded that the recommended sample tree size could be considerably reduced from 25 to 10 trees per clone (13 in the case of density) while maintaining the necessary accuracy of the statistical results. The author studied the oven dry density and some morphological fibre characteristics (length, lumen diameter and wall thickness) in Balsam poplar ( Populus balsamifera ). However, other authors such as SABOROWSKI and HAPLA (1985) suggested that collecting 5 trees per experimental area is sufficient to achieve reliable estimation of some physical properties (specific gravity, tensile strength parallel to fibre, bending strength, compressive strength parallel to fibre and impact bending strength) in Douglas fir species ( Pseudotsuga menziesii ).

However, in the study in hand, the author would not be able to collect the number of trees recommended in the literature, for several reasons. For instance, the high cost of tree harvesting, sample cutting and processing. There were also difficulties in transportation of samples from the study area into Khartoum (Sudan’s capital city) and then to TU Dresden, Germany, where the wood properties investigations were conducted.

Therefore, 30 trees for each species were considered in the current study, collected as follows: 15 trees from Acacia seyal and Balanites aegyptiaca were collected randomly from each rainfall zones, giving a total of 60 trees for both species from both zones (15 trees/species/zone × two zones × two species = 60 trees). The trees were harvested using a chainsaw in Al Duwaim, an axe in Al Damazien, and an axe and two-man saw in Kadugli and Al Obeid. The trees total height, merchantable height, and diameter at breast height (DBH) were measured using tape and callipers (Tables 3.2 and 3.3).

- 61 -

Material

Table 3.2: Description of tree samples of Acacia seyal and its growing conditions

Tree Total Merchantable DBH Zone State City Forest number height (m) height (m) (cm) 1 8.30 2.14 20.95 North Al Obeid *Al Ein 2 7.70 1.70 22.70 Kordofan 3 8.50 1.20 27.00 4 6.56 1.57 19.90 Habila 5 4.80 1.30 17.35 6 6.20 1.40 25.00 7 5.40 1.40 25.00 Al One 8 6.10 1.65 21.00 White Homora Al 9 8.00 1.77 20.15 Nile Duwaim 10 6.50 1.35 20.25

Aum Top 11 4.70 1.77 19.50 12 6.37 1.48 25.00 13 7.42 1.60 19.85 Goz 14 8.23 1.45 17.10 Fagor 15 7.00 1.35 22.25 1 9.00 2.54 22.85 Tawlla 2 9.16 2.39 23.00 Al 3 8.72 1.60 19.65 Blue Nile Damazien 4 8.30 1.60 22.20 Khor 5 8.50 2.23 21.00 Donia 6 7.30 1.55 19.80 7 12.50 2.86 25.65 Al Two 8 11.16 2.16 26.55 Sheheata 9 11.09 2.25 25.70 10 8.26 1.85 24.75 South Kadugli Al 11 10.60 1.40 29.15 Kordofan Homara 12 10.43 1.54 25.30 13 10.35 1.76 30.00 Al 14 7.67 2.44 19.80 Bardab 15 10.20 1.90 20.70 *Al Tainat forest not exists in this table because no Acacia seyal trees were collected from this forest .

- 62 -

Material

The total height measurement was performed after tree felling by measuring the linear distance from base to tip in addition to measuring the length of the remaining part on the ground (stump).

Table 3.3: Description of tree samples of Balanites aegyptiaca and its growing conditions

Tree Total Merchantable DBH Zone State City Forest number height (m) height (m) (cm) 1 7.66 2.25 29.45 North *Al Al Obeid 2 6.80 2.80 31.90 Kordofan Tainat 3 6.45 2.35 25.50 4 6.90 2.30 23.20 Habila 5 7.95 1.94 27.00 6 7.5 1.80 21.25 7 6.80 1.40 23.20 Al One 8 6.40 1.45 20.50 White Homora Al 9 6.50 1.30 28.55 Nile Duwaim 10 7.29 2.20 25.00

Aum Top 11 6.20 1.57 24.75 12 7.25 1.40 25.75 13 5.00 1.45 21.35 Goz 14 6.90 1.78 19.50 Fagor 15 8.00 2.50 32.00 1 6.90 1.40 21.50 Tawlla 2 7.00 1.40 22.00 Al 3 4.60 1.30 20.80 Blue Nile Damazien 4 7.80 1.87 25.35 Khor 5 8.10 2.30 31.50 Donia 6 6.50 1.90 27.15 7 9.37 3.15 33.00 Al Two 8 8.60 1.90 33.45 Sheheata 9 8.70 2.30 25.00 10 9.00 1.96 29.05 South Kadugli Al 11 7.35 3.46 32.30 Kordofan Homara 12 11.17 2.06 35.25 13 9.16 3.14 33.30 Al 14 10.60 3.50 29.50 Bardab 15 8.40 2.00 29.90 *Al Ein forest not exists in this table because no Balanites aegyptiaca trees were collected from this forest .

- 63 -

Methods

4 Methods

This chapter provides a detailed description of sampling and sample processing. The statistical analysis is also described in this chapter.

4.1 Sampling

Two samples were obtained from each tree described in chapter 3.2 at 10 % and 90 % of the merchantable height in the form of a disc of approximately 30 cm in height. For each disc, the number and the north direction were marked in the upper part. The bark of Acaica seyal samples was removed (debarking), because of its susceptibility to mite attack, and then the samples were wrapped in plastic bags (Figure 4.1).

Figure 4.1: Acacia seyal samples’s debarking and storage.1, 2. Samples debarking, 3, 4. Samples wrap by plastic bags, 5. Samples storage and covering by Azadirachta indica leaves.

- 64 -

Methods

Samples of both species ( Acacia seyal and Balanites aegyptiaca ) were stored under shade and covered by leaves of Azadirachta indica (famous for expelling insects) to prevent insect attack (Figure 4.1). Afterwards the discs were transferred to sawmill to obtain small samples (strips) of 3 cm × 3 cm × tree diameter in cm (include tree’s pith) using a circular saw, narrow band saw and thickness planer (Figure 4.2).

Figure 4.2: Samples processing. 1, 2. Cutting the discs edges using circular saw; 3. Reduction of samples thickness using thickness planer; 4, 5. Final obtained samples (strips).

- 65 -

Methods

Five defect-free samples (strips) were taken from each disc of 10 % height and three samples from the 90 % height, giving a total of 8 strips for each tree and of 480 strips for the 60 trees collected from both study species. One sample from the 10 % disc and one from the 90 % disc for each tree were wrapped in plastic bags and immediately stored in a refrigerator to keep it wet (these were later used in basic density investigation). The remaining strips were air dried.

The log processing, including details about tree collection, areas and number of samples collected from each tree, are presented in Figure 4.3.

Figure 4.3: The study species logs processing flowchart (Acacia = Acacia seyal ; Balanites = Balanites aegyptiaca ; C = City and F = Forest)

- 66 -

Methods

For the anatomical properties investigations , one air dried sample from 10 % stem height of 20 trees per species (10 per zone) was used for the maceration and softening.

The samples′s selection criteria for the anatomical properties investigation was as follows: two trees were selected from each forest depending upon the tree pith eccentricity. The selected trees have the lower pith eccentricity values to avoid the possibility of tension wood existence. Tables 4.1 and 4.2 represent the selected trees within each forest.

Table 4.1: Acacia seyal selected trees for anatomical properties investigations

Zone State City Forest Total trees Selected trees

1 1 North Al Obeid *Al Ein 2 2 Kordofan 3 x 4 3 Habila 5 x 6 4 7 5 Al One 8 6 White Homora Al 9 x Nile Duwaim 10 x

Aum Top 11 7 12 8 13 9 Goz Fagor 14 x 15 10 1 1 Tawlla 2 2 Al 3 x Blue Nile Damazien 4 3 Khor Donia 5 x 6 4 7 5 Two Al Sheheata 8 6 9 x 10 x South Kadugli Al Homara 11 7 Kordofan 12 8 13 x Al Bardab 14 9 15 10 *Al Tainat forest is not exist in this table because no Acacia seyal trees were collected from this forest. x = the excluded trees. Trees numbers in the colum selected trees are changed according to the total number of trees selected for anatomical properties investigations (10 trees per zone)

- 67 -

Methods

Table 4.2: Balanites aegyptiaca selected trees for anatomical properties investigations

Zone State City Forest Total tree Selected trees

1 1 North Al Obeid *Al Tainat 2 2 Kordofan 3 x 4 3 Habila 5 x 6 4 7 5 Al One 8 x White Homora Al 9 6 Nile Duwaim 10 7

Aum Top 11 8 12 x 13 9 Goz Fagor 14 x 15 10 1 1 Tawlla 2 2 Al 3 x Blue Nile Damazien 4 3 Khor Donia 5 4 6 x 7 x Two Al Sheheata 8 5 9 6 10 7 South Kadugli Al Homara 11 x Kordofan 12 8 13 9 Al Bardab 14 x 15 10 *Al Ein forest is not exist in this table because no Balanites aegyptiaca trees were collected from this forest. x = the excluded trees. Trees numbers in the colum selected trees are changed according to the total number of trees selected for anatomical properties investigations (10 trees per zone)

A total of 40 samples were used in the anatomical properties investigations for both species from both zones. The selected samples were cut into two radiuses, one of which was used to prepare the anatomical investigation samples. This selected radius was then separated into two parts in middle length. The upper parts were used for maceration by setting sampling points at centimetre intervals from pith to bark and cutting small slivers of 10 mm in length from each sampling point in order to investigate the trend of fibre length from pith to bark. The lower parts meanwhile were used for softening by setting two sampling points, one at 10 % and the second at 90 % distance from pith to bark. One specimen of 0.5 × 0.5 × 1 cm was cut from each sampling point to prepare a cross section in order to measure fibre and vessel diameter and lumen diameter. One sample per species was selected to be used for the

- 68 -

Methods

anatomical composition description by taking a specimen from 50 % of the distance from pith to bark to prepare cross, radial and tangential sections.

The samples for fibre length investigations were taken at each 1 cm interval, as previously mentioned. The selected trees are of different diameters; consequently the strips taken from each tree are of a different diameter from those taken from other tree. Also the number of the radial samples taken from each strip is different from the others. Therefore, the values of the measured samples were divided into 5 radial portions from pith to bark, which are 10 %, 30 %, 50 %, 70 % and 90 %. Thus, they will be comparable with each other and with other studies.

For basic density investigation, one wet sample from each stem height (10 % and 90 %) were taken from each tree, giving a total of 120 samples for both species, and were cut from the pith (centre) into two parts (radiuses). Each part was then sawn into small specimens of 3 × 1 × 3 cm dimension each 1 cm interval from pith to bark. The obtained values of the measured samples were divided into 5 radial portions from pith to bark, which are 10 %, 30 %, 50 %, 70 % and 90 %.

For air dry density investigation, one sample was chosen from each height (10 % and 90 %), giving a total of 120 samples for both species. The samples were then cut from the pith (centre) into two parts (radiuses). One radius was chosen for air dry density determination. Small specimens of 3 × 1 × 3 cm were cut from each radius at three portions representing the distance from pith to bark (10 %, 50 % and 90 %).

For X-ray densitometry investigation, one air dried sample (stripe) from each tree (from 10 % height) was taken, giving a total of 60 samples for both species. Each sample was cut from the pith (centre) into two parts (radiuses). The measured values were divided into 5 radial portions from pith to bark, which are 10 %, 30 %, 50 %, 70 % and 90 %. For more details about the sample preparation for X-ray densitometry see section 4.2.2.3.

Finally, for hardness strength investigation, one sample (stripe) from each stem height (10 % and 90 %) was taken from each tree, giving a total of 120 samples for both species. Each sample was split from the pith (centre) into two radiuses. In the transverse section, the hardness test was performed along the radial distance from pith to bark of each sample with 2 cm intervals between the measured points. In the radial section, the test was performed in the middle of the sample at two points with 2 cm interval.

- 69 -

Methods

Details of tree sampling are illustrated in Figure 4.4.

Figure 4.4: Tree sampling

Table 4.3 summarized the samples position within the tee (radically and vertically) for each of the studied wood properties.

- 70 -

Methods

Table 4.3: The samples position of each wood property along the radial and vertical axis

Samples position Property Vertically (along the stem Radially (over the radius) merchantable height) Fibre length 10 %, 30 %, 50 %, 70 % and 90 % 10 % dics Other fibres characteristics 10 % and 90 % 10 % dics (D, LD, WT, RR, SR and FC) Vessels characteristics 10 % and 90 % 10 % dics (D, LD and WT) Basic density 10 %, 30 %, 50 %, 70 % and 90 % 10 % and 90 % dics Air dry density 10 %, 50 % and 90 % 10 % and 90 % dics X- ray density 10 %, 30 %, 50 %, 70 % and 90 % 10 % dics Hardness strength in each 2 cm intervals from pith to 10 % and 90 % dics transverse section bark Hardness strength in radial two points in the middle of the 10 % and 90 % dics section radius with 2 cm interval

D = diameter, LD = lumen diameter, WT = wall thickness, RR = Runkel ratio, SR = slenderness ratio and FC = flexibility coefficient.

4.2 Sample processing

This section provides detailed description of the investigation methods used in respect of the anatomical properties, density and hardness strength. Therefore, it is divided into anatomical properties investigations, density investigations and hardness strength investigation.

All the investigations were conducted in Wood Laboratory at the Institute of Forest Utilization and Technology (Forstnutzung), Faculty of Environmental Sciences, Dresden University of Technology, Germany, except the investigation of wood basic density, which was conducted in the Wood Laboratory at the Department of Forest Products and Utilization, Faculty of Forestry, University of Khartoum, Sudan.

4.2.1 Anatomical properties investigations

In this section, two procedures were conducted: maceration procedure to measure the length of fibres, and softening procedure to measure the fibres and vessels diameter and lumen diameter and for the anatomical composition description.

- 71 -

Methods

4.2.1.1 Maceration

The maceration procedure developed by Shultze as cited in JANE (1970) was adopted to macerate the woody materials into individual cells using a strong agent (nitric acid). About 3 slivers were placed in test tubes, to which 65 % nitric acid with 2–3 crystals of potassium chlorate were added and then warmed up (100 ºC) in a water bath for about 5–10 minutes. The maceration process was stopped as soon as the slivers exhibited a white ragged appearance.

The test tubes were filled by distilled water in order to stop the action of the nitric acid. The macerated material was washed twice with distilled water and then left for about 10 minutes in distilled water to remove any traces of nitric acid. When the loosened sections settled down, excess water was gently drained away, and then the macerated material was stained with 2–3 drops of safranin dye for 5 minutes and then rewashed twice with distilled water. The prepared macerated material was transferred to a slide surface (76 × 26 mm) using a needle and about 2 drops of Kaiser’s glycerol gelatine were added to each slide. Each slide was then covered gently with a cover slip measuring 24 × 46 mm. The prepared slides were left to dry gradually for 24 hours. The length of 40 fibres, chosen randomly from each sample, were measured using light microscope (JENAMED VARIANT – Carl Zeiss Jena) with an 10x ocular lens provided with a measuring scale divided into ten equal segments and each segment divided into ten sub-segments (see Figure 4.5).

Figure 4.5: Fibre length measurement using measuring scale

- 72 -

Methods

4.2.1.2 Softening

Specimens of 0.5 × 0.5 × 1 cm were softened by boiling in water for 8–10 hours, in the case of Balanites aegyptiaca species. Due to the high wood density of Acacia seyal , softening in an autoclave device method, as described in GROSSER (1971) for bamboo species, was adopted to soften specimens. The softening time was 3 hours using a temperature of 140°C and a pressure of 4–5 bar.

The cooked Acacia seyal specimens were then immersed in 75 % ethanol (JAGIELLA and KÜRSCHNER 1987) for a week. Cross sections of 10–15 µm in thickness from Balanites aegyptiaca specimens were cut using a freezing microtome (REICHERT-JUNG 1206) with movable blade at an angle of about 15 degrees. While the GSL1 microtome (invented by H. Gaertner, F.H. Schweingruber and S. Luccinetti) was used to cut the cross sections of Acacia seyal specimens, the immersion in alcohol creates difficulties in samples’ installation by freezing microtome. The thicknesses of Acacia seyal cross sections were 15–20 microns. Figure 4.6 shows the two different microtomes used in the current study.

1 2

Figure 4.6: The used Microtomes: 1. GSL microtome; 2. Freezing microtome (REICHERT- JUNG 1206).

In addition to the cross sections prepared from each specimen taken at 10 % and 90 % distance from pith to bark, one specimen from 50 % distance was selected from one tree per species for the anatomical composition description. For this purpose cross, radial as well as tangential sections were prepared. The prepared sections from both species were dehydrated using various ethanol concentrations (25 %, 50 % and absolute, respectively). The

- 73 -

Methods

dehydration time was about 3–5 minutes for each concentration. The sections were then stained with safranin dye, washed in distilled water and then dehydrated in 25 %, 50 % and absolute ethanol, respectively, for 3–5 minutes each. They were then transferred carefully onto slides of 76 × 26 mm, to which about 2 drops of Kaiser’s glycerol gelatine were added, and then covered gently with a cover slips measuring 24 × 46 mm. The slides were left to dry for 24 hours. A total of 40 slides were prepared from each species. The cross sections in each slide were divided into four equal quarters (see Figure 4.7).

Figure 4.7: Microscopic slides including cross sections divided into four equal quarters

A Nikon Coolpix 990 Camera fixed in light microscope (JENAMED VARIANT- Carl Zeiss Jena), and connected to a computer, was used to take photos from the prepared slides (from each square separately). ImageJ software (version 1.45b) was used to measure the fibres and vessels diameter and lumen diameter from the photos (Figure 4.8).

Figure 4.8: Fibre and vessel’s diameter and lumen diameter measurements using ImageJ

software program

- 74 -

Methods

A number of 40 fibres were selected randomly to measure fibre diameters and lumen diameters (10 fibres were measured from each quarter); while 30 vessels were selected to measure vessel diameter and lumen diameter (10 vessels were measured from three quarters).

Both fibres and vessels’s wall thickness was calculated using the following formula:

D − LD WT = (F 4.1) 2

Where: WT is wall thickness; D is diameter and LD is lumen diameter.

Using the fibre dimensions measured values, three derived values were calculated:

- Slenderness ratio as fibre length/fibre diameter,

- Flexibility coefficient as fibre lumen diameter /fibre diameter × 100 and

- Runkel ratio as 2 × fibre cell wall thickness/fibre lumen diameter (SAIKIA et al. 1997, OGBONNAYA et al. 1997).

The anatomical composition description was performed on the basis of the IAWA List of Microscopic Features for Hardwood Identification (IAWA committee 1989). Quantitative and qualitative anatomical properties were observed. For this purpose, solid wood and micro- slides of cross, radial and tangential sections were used. Hand lens of 10 x magnifications, light microscope (JENAMED VARIANT – Carl Zeiss Jena) and The ImageJ software (version 1.45b) were used to measure the wood anatomical properties.

4.2.2 Density investigations

In the current study, three densities were investigated: basic density, air dry density as well as density achieved by X- ray densitometry.

4.2.2.1 Basic density

The wood density of the tree species was determined using the dry weight and green volume.

The prepared wood specimens (described in section 4.1) were marked using permanent ink, then immersed in water for a few hours in order to compensate the water lost during the sawing process.

- 75 -

Methods

The specimen’s green volume was determined using the water displacement method by placing a beaker full of water on a digital balance. The balance was then re-zeroed and the specimens (fixed in needle) were carefully submerged in the water. The specimens should not contact the sides or bottom of the beaker. The reading weight is equal to the specimen volume. The specimens were then immediately transferred into an oven (Model: Binder) and dried for 48–72 hours until the constant mass was attained. After oven drying, the specimens were weighted using sensitive balance (Model: AAA16ØL). This weight was taken as the oven-dried weight.

The basic density was calculated using the following formula:

woven dry (F 4.2) D = vgreen

Where: D basic density in Kg/m³ Woven dry dry weight in Kg Vgreen green volume in m³.

4.2.2.2 Air-dry density

The air dry gravimetric method was conducted on the basic of DIN 52 182 (ANONYMOUS 1991). Wood specimens of 3 × 1 × 3 cm were taken from three portions representing the distance from pith to bark (10 %, 50 % and 90 %) as previously described in section 4.1. These were then conditioned to a constant mass at 20 °C air temperature and 65 % relative humidity. The specimens’ weight was measured using a sensitive digital balance (Model: Sartorius BP 210S). The volume was measured using micro callipers (Mitutoyo Digimatic, Model CDN-P30). The air dry density was calculated using the following formula:

w air dry (F 4.3) D = vair dry

Where:

D air dry density in g/cm³ Wair dry air dry weight in g Vair dry air dry volume in cm³.

- 76 -

Methods

4.2.2.3 X-ray densitometry

The X-ray densitometry method described in SCHWEINGRUBER (1988) was adopted to measure the density as follows:

Sample preparation

One air dried sample (strip) from each tree (from 10 % height) was taken for X-ray densitometry investigation. Each sample was cut from the pith (centre) into two parts (radiuses).

- Fibre angle measurement

The fibre angle measurement is considered one of the most important steps in the preparation of samples for X-ray analysis. If this step is not performed with high accuracy, the anatomical structures of the wood on the radiograph will be reproduced in poor quality (SCHWEINGRUBER 1988). The samples for X-ray analysis should be cut at a right angle to the fibre orientation (which is 90º). To achieve this, a thin layer was split from one of the radial sides of each sample to make the fibre angle more pronounced, and then the samples were glued on wooden supports from the other radial side. The fibre angles for each sample were measured using a Nikon SMZ-1B DENDROSCOPE (WALESCH ELECTRONIC) (see Figure 4.9/2).

- Sample sawing and conditioning

Thin cross sections (laths) of 1.20 ± 0.05 mm were cut from each fibre angle using a double pleated circular saw DENDROCUT 2003 (WALESCH ELECTRONIC) (see Figure 4.9/3). The thickness of each lath was measured three times (in the inner, middle, and outer parts) using a Mitutoyo Digimatic indicator; the names and thickness were recorded in protocol. The laths thickness should not exceed 1.20 ± 0.05 mm. The maintenance of a constant lath thickness is of fundamental importance for the accuracy and comparability of the density measurement (SCHWEINGRUBER 1988).

The samples were placed in a cellophane carrier, fixed in plastic baskets and stored in a conditioning chamber (20 ºC and 65 % atmospheric relative humidity) for 48 hours.

- 77 -

Methods

X-ray of the sample material The X-ray of the samples was carried out under defined conditions at a temperature of 20 ºC and 65 % atmospheric relative humidity. The wood samples and a calibration wedge of cellulose acetate with steps of known density were placed on an X-ray film and then radio graphed with the following irradiation characteristics:

Accelerating voltage: 12.0 kV Electric current: 20 mAs Irradiation time: 56 minutes Distance between the samples and radiation source: 185 cm

The radiation was performed with the aid of a stationary X-ray structure of coarse plant “Baltographe” (BALTEAU). The film was developed using an automatic processor.

Analysis of the X-ray film The density of the radiographed samples was then measured using a DENDRO 2003 (WALESCH EL ECTRONIC) (see Figure 4.9/7), which consists of the following main parts:

1. Portable measuring table for receiving and manipulating the X-ray film 2. Light source for radiography X-ray film. 3. Optics with different magnifications (10 -, 25 - or 50 times). 4. Projection display with seven built-in photo sensors. 5. Computer for controlling the individual modules and data processing.

The film was placed on a movable table that passes over a light source; light signals of varying intensity were transformed by a potentiometer into electrical impulses. During this process comparisons were made between the grey scale of the wood samples and the calibration wedge. The optics magnification used in the current study was 25 times. The DENDRO 2003 records one value for each 20 micrometre distance. Then the data was imported into Excel files for calculation and analysis.

The measured values were divided into 1 cm intervals to be comparable with those of basic density. In this case each 500 values are equal to 1 cm. Then the mean of the first 500 values from the pith were calculated and considered as the first centimetre from the pith, and so on.

- 78 -

Methods

Figure 4.9: X-ray densitometry methodology: 1. Wood samples glue in wooden supports, 2. Fibre angles measurement using DENDROSCOPE, 3, 4.Thin cross-sections (laths) cutting using double parallel circular saws, 5. The resulted 1.20 ± 0.05 mm thickness laths, 6. Wood radiographic image and 7. Density measurement using DENDRO 2003.

- 79 -

Methods

Corrections factors for density equivalent values calculation

The obtained density values from the X-ray technique are not equivalent values. In order to obtain equivalent density values, corrections factors were calculated by calibrating each value obtained by X-ray technique by its counterpart in air dry density, and then an average was calculated (see Figure 4.10). A total of 81 samples were used for this purpose in case of Acacia seyal and 90 samples in case of Balanites aegyptiaca.

Figure 4.10: Correction factor calculation

The factors were used to convert the values obtained by X-ray densitometry technique into equivalent values using the following formula:

RD = XRD × F (F 4.4) Where: RD real density in g/cm 3 XRD density achieved by X ray F correction factor

- 80 -

Methods

Possible source of error

There are many different problems that may be unavoidable during the preparation of the samples and that may lead to unreliable results. SCHWEINGRUBER (1988) divided the main sources of error into technical errors and biological faults. Table 4.4 shows some of the technical and biological errors described by SCHWEINGRUBER (1988).

Table 4.4: Possible sources of error in X-ray densitometry analysis

Sources of error

Technical Biological - Splits in the samples - Missing, double rings - Overlaps in samples which have not - Non-extractable, strongly- been registered it appears as if an reflecting substances, such as annual ring is missing fungi - Very narrow ring may not be - Tangential traumatic resin ducts registered when the measuring slit is - Irregular ring boundaries too wide - Reaction or compression wood - Where the sample has been badly - Biological decomposition oriented the anatomical properties will

come out blurred on the X- ray picture

SCHWEINGRUBER (1988) also mentioned that the heartwood substance may lead to error in density measurement. The heartwood substances vary enormously in terms of the rate at which they absorb X-rays. For instance, resins absorb relatively little compared to cellulose, water absorbs slightly more, and certain other cell substances display considerably higher rates of X- ray absorption.

In the current study, pre test was performed in three samples for the heartwood substance by measuring the density before and after substance extraction by water and ethanol. The density values before and after the extraction were compared. No differences were found, thus no extraction were performed for the rest of the samples.

4.2.3 Hardness strength investigation

Brinell hardness test was conducted on the basis of DIN EN 1534 (ANONYMOUS 2000) to measure the hardness of the studied species. The selected samples for this investigation were split from the pith (centre) into two radiuses. In order to obtain a soft surface, the four sides of each radius were sanded using a sanding machine (Scheppach, Model BTS 900x) at 80

- 81 -

Methods

grade, then two sides (one transversal and one radial) were sanded using 120 and 180 grades (the hardness test was later performed on these two sides). After sanding, the samples were conditioned at 20 ºC air temperature and 65% relative humidity.

The TIRA test 28100 machine provided by a hardened steel ball with a diameter of 10 ± 0.01 mm was used to perform the hardness test. The applied force must be increased so that the nominal value of 1 kN is achieved after 15 ± 3 seconds. This force has to be maintained for 25 ± 5 seconds. Then the diameter of each indentation point was measured twice using micro loupe. Lead graphite was used to make the edges of the indentation points more visible (see Figure 4.11). The following formula has been used to get the hardness:

2F HB = (F 4.5) g ⋅π ⋅ D []D - (D² - d²)1/2

Where:

HB Brinell hardness strength (N/mm 2) g Acceleration due to gravity (m/s 2) π Factor „pi“ (≈ 3.14) F Nominal force (N) D Ball diameter (mm) d Diameter of the residual indentation (mm)

In the transverse section, the hardness test was performed along the radial distance from pith to bark of each sample with 2 cm intervals between the measured points. In the radial section, the test was performed in the middle of the sample at two points with 2 cm interval.

- 82 -

Methods

Figure 4.11: Hardness test methodology: 1. TIRA test 28100, 2. Sander machine, 3. Press of the steel ball into wood sample, 4. Micro loupe, 5. and 6. Indentation points in cross and radial sections, respectively.

4.3 Statistical analysis

4.3.1 Data entry, cleaning and test of normality

The data entry and mathematical processes were conducted using an Excel spreadsheet in Microsoft Office Excel (2007). The samples for fibre length and wood basic density investigations were taken at each 1 cm interval, as previously mentioned in section 4.1. Also, the values obtained from X-ray densitometry are divided into 1 cm intervals (see section 4.2.2.3). The selected trees are of different diameters, so the strips taken from each tree are of a different diameter from those taken from other trees. Even within the tree, samples (strips) taken at 10% height have a larger diameter than from those taken at 90 % height. Then, the number of radial samples taken from each strip is different from the others. Therefore, the values of the measured samples were divided into 5 radial portions from pith to bark, which are 10 %, 30 %, 50 %, 70 % and 90 %. Thus, they will be comparable with each other and with other studies.

After data entry, the outliers were searched for by checking box plots and were then deleted from the data (see Figure 4.12). For Acacia seyal , values of some trees were omitted from the data due to abnormality, such as tree number 1 from Habila forest and trees number 2 and 3 from Al Homora forest.

- 83 -

Methods

Test of normality was then conducted using histograms to test whether the data came from a normally distributed population. The importance of this test is that it determines the type of test that should be used to test the significance between two parameters or groups of parameters. In the current study, the test of normality showed that the data are normally distributed. Thus, several statistical tests were conducted according to the tested properties and the goal of the tests.

Figure 4.12: Outliers searching SPSS Box plots. In the left side Box plots before outliers’ removal, and in the right side Box plots after outliers’ removal. (VD = vessels diameter in µm, S.N = radial sample number, where 1 near the pith and 2, near the bark and TR = tree). The values next to each outlier represent the case number in excel sheet.

The statistical data analysis was performed using the computer statistical program PASW (SPSS, statistics processor, version 18.0) for Windows.

4.3.2 Independent-sample T-test

The independent sample T-test compares the means of two variables. It answers the question of whether or not the difference between means is statistically significant in the population of interest. In this test, two hypotheses were tested (not significant or significant) between the means of the two variables.

In the current study, the independent sample T-test was conducted to test the variation between trees within each forest in the case of the anatomical properties where two trees were selected to conduct anatomical investigations. It was also conducted to test the variation between regions within each zone as well as between the two zones themselves. Radial and/or vertical variations of the entire studied wood properties were also tested using

- 84 -

Methods

independent sample T-test, with the exception of fibre length and wood density radial variation. It was also used to compare the air dry density values with those obtained from X- ray densitometry method after calibration. This test was conducted using 0.05 probability level in all cases.

4.3.3 Analysis of variance (one way ANOVA)

The one-way analysis of variance (ANOVA) is used to determine whether there are any significant differences between the means of two or more independent (unrelated) groups (although it tends only to be used when there are a minimum of three, rather than two groups).

In the present study, this test was conducted to test the variation among trees within each forest in the case of wood density and hardness strength investigations, where three trees were selected to conduct the investigations. It has also been used to test the variation among forests within each region as well as the radial variation of both fibre length and the three measured wood densities. The probability level 0.05 was used in all cases.

4.3.4 Post-hoc tests in ANOVA

Post-hoc tests are generally performed only after obtaining a significant difference among the means and where additional exploration is needed to provide specific information on which means are significantly different from each other. In the current study Post-hoc (Tukey) test was performed to test the variation among the selected portions in the case of fibre length and wood air dry and basic densities. No Post-hoc tests were conducted in the case of density achieved by X-ray densitometry due to the non-significant differences that were observed among the radial positions.

4.3.5 Descriptive statistics tests

Descriptive statistics provide simple summaries about the sample and the observations that were made. These summaries may either form the basis of the initial description of the data as part of a more extensive statistical analysis, or they may be sufficient for a particular investigation in and of themselves.

- 85 -

Methods

In the present study, descriptive statistics tests were performed in order to obtain information about mean, median, maximum, minimum as well as standard deviation values for each tested property. This test was conducted for all the studied wood properties using all measured values in order to obtain general information about the statistics of the wood properties. It was also conducted in the inner sample values (taken at 10 % distance from pith to bark) and the outer sample values (taken at 90 %) separately, in order to obtain general information about the expected statistics for juvenile and mature wood.

4.3.6 Trend line/ regression analysis

Trend line or regression analysis is a statistical process for estimating the relationships among variables. In the current study, this test was performed to determine the trend of fibre length and wood basic density from pith to bark. Microsoft Office Excel (2007) was used for this purpose. The trend was tested using the mean values of each of the selected five portions (10 %, 30 %, 50 %, 70 % and 90 %). Several equation forms were examined (e.g. exponential, linear, logarithmic, polynomial, power and moving average). Selection of the representative equation depends on the R-squared obtained in each of the tested equations. R-squared is a statistical measure of how close the data are to the fitted trend/ regression line. A trend line is most reliable when its R-squared value is at or near 1. Thus, in the current study, the trend line with higher R-squared values was chosen.

- 86 -

Results and Discussion

5 Results and Discussion

The results and discussion in this study are divided into 5 parts (main chapters) in order to achieve the objectives of the study (mentioned in section 1.3) as follows:

Part 1: Investigation of anatomical properties of the study species with importance for pulp and paper making.

This part describes the anatomical aspects of the study species, including their fibres and vessels characteristics as well as the trend of fibre length from pith to bark. The anatomical composition is also described briefly in this part.

Part 2: Investigation of wood density as the main physical property.

Information about the study species’ air dry density, basic density as well as X-ray density values are provided in this part. Determination of the trend of wood basic density from pith to bark is also included. Brief discussion about the suitability of using X-ray technique as a valid tool for the study species density determination is provided in this part.

Part 3: Investigation of hardness strength as an important mechanical property.

This part provides information about the study species hardness strength in transverse and radial sections.

Part 4 : Investigation of a possible effect of rainfall zones (climatic zones) on wood properties.

This part discusses intensively the effect of rainfall zones on the selected wood properties and compares it with the results obtained from the literature.

Part 5 : Assessment of the suitability of the study species for pulp and paper and flooring industries.

From the results presented in Parts 1, 2, and 3, the suitability of the study species for pulp and paper and flooring industries will be assessed. Additionally, depending on Part 4 the effect of rainfall zones on the suitability of the study species for the selected industries is briefly discussed.

The within tree variation (radial and/or vertical) and between trees, forests and regions variations for the selected properties are described briefly at the beginning of Parts 1, 2 and 3 in order to provide an overview of the study species behaviour. These results are described in the form of figures and tables. Moreover, full statistics are provided in appendices, which can be find on the attached CD.

- 87 -

Results and Discussion

5.1 Anatomical properties

This section begins with a brief description of the study species’ anatomical composition, providing information about the obtained fibres and vessels characteristics values as well as the trend of fibre length from pith to bark.

5.1.1 Anatomical composition

In this section, the study species’ wood anatomical composition is described depending upon microscopic slides using a light microscope (Model: Variant Jenamed), as well as solid wood using a hand lens of 10 × magnifications. The anatomical description was based on IAWA List of Microscopic Features for Hardwood Identification (IAWA committee 1989). The anatomical description provided by the current study results and those of the literature are also compared in this section.

5.1.1.1 Acacia seyal var . seyal

Table 5.1 presents the anatomical composition description of Acacia seyal obtained from the current study in comparison with that provided by NEUMANN et al. (2000). The description of NEUMANN et al. (2000) was modified to be comparable with the IAWA List of Microscopic Features for Hardwood Identification. No anatomical description depending on the IAWA List (IAWA committee 1989) was found for Acacia seyal in the literature.

- 88 -

Results and Discussion

5.1: Anatomical composition description of Acacia seyal in comparison with literature

Description sources (*) Feature NEUMANN et al. Study results (2000) Growth rings (1) distinct V (5) Wood diffuse-porous V (10) Vessels in radial multiples common V (11) Vessel clusters common x (13) Simple perforation plates V (20) Intervessel pits alternate V (25) Small 4 – 7 µm (24) medium 7-10 µm (30) Vessel-ray pits with distinct borders; similar V to intervessel pits in size and shape throughout

Vessels the ray cell (42) 100 - 200 µm (Mean tangential diameter of V lumina) (47) 5 - 20 vessels per square millimeter V (52) <= 350 µm mean vessels element length V (short) (56) Tyloses common x (58) Gums and other deposits in heartwood V vessels (61) Fibres with simple to minutely bordered pits V Tracheid and (66) Non-septate fibres present V Fibres (70) Fibres very thick-walled V (72) 900-1600 µm fibre length (short) V (76) Axial parenchyma diffuse V (77) Axial parenchyma diffuse-in-aggregates x (79) Axial parenchyma vasicentric V (80) Axial parenchyma aliform V (81) Axial parenchyma lozenge-aliform V (83) Axial parenchyma confluent V Axial parenchyma (84) Axial parenchyma unilateral paratracheal V (85) Axial parenchyma bands more than three V cells wide (89) Axial parenchyma in marginal or in V seemingly marginal bands (90) Fusiform parenchyma cells V (91) Four (3-4) cells per parenchyma strand V (99) Larger rays commonly > 10-seriate V (102) Ray height > 1 mm V Rays (104) All ray cells procumbent V (110) Sheath cells V (115) 4-10 / mm (Ray per millimeter) V (136) Prismatic crystals present V (138) Prismatic crystals in procumbent ray cells Mineral inclusions V (142) Prismatic crystals in chambered axial V parenchyma cells

(*) = numbers between brackets refer to the number of the anatomical features observed in the study species according to IAWA list 1989, V = the anatomical feature is exist, X = the anatomical feature is absent.

From Table 5.1, it is obvious that the anatomical description of Acacia seyal provided by the current study is similar to that provided in the Sahara database by NEUMANN et al. (2000) with the exception of the absence of tyloses, diffuse in aggregates axial parenchyma and

- 89 -

Results and Discussion

clusters vessels in the description of NEUMANN et al. (2000) and the small vessel pits diameter in the current study vs. the medium pits in NEUMANN et al. (2000). This may be due to the differences in region of origin for the collected samples. Figure 5.1 shows the differences in anatomical features between the current study anatomical description and that of NEUMANN et al. (2000).

Figure 5.1: Acacia seyal anatomical structure. A and B in photo 1 show the existence of clusters vessels and axial parenchyma diffuse in aggregates, respectively, C in photo 2 shows the existence of tyloses and D in photo 3 represents the small pits diameter in vessel element.

In general, the anatomical structure of Acacia seyal is typical to tree species growing in topical and sub-tropical areas, where the trees have mostly wood with diffuse- porous, vessels in multiple or clusters, simple perforation pate, etc. Also the occurrence of numerous crystals is more common in wood from the tropics. In addition to their existence in axial parenchyma as well as rays, the crystals are also present in fibres. The influence of aridity on the occurrence of crystals in wood was observed by several authors (e.g FAHN et al. 1986, CARLQUIST 1988, JOHN 1990). The presence of tyloses is also apparently a response to a loss of water in the vessels (ZIMMERMANN 1983).

- 90 -

Results and Discussion

In the current study, growth ring boundaries of Acacia seyal were demarcated by a thin marginal band of parenchyma (see Figure 5.2). Similarly, GEBREKIRSTOS et al. (2008), NICOLINI et al. (2010) and MBOW et al. (2013) had success with the ring boundary distinctiveness of A. seyal which was delimited by marginal parenchyma. TARHULE and HUGHES (2002) have identified three categories of species in terms of their potential for dendrochronology: (1) potentially useful, (2) problematic, and (3) poor. The categorization is based on the distinctiveness of the annual ring boundaries, the ability for cross dating, ring circuit uniformity, ring wedging, and ring width variability. Those for which one can expect to obtain good results fall into the potentially useful category. Depending on the above mentioned categories, NICOLINI et al. (2010) and MBOW et al. (2013) classified Acacia seyal as potentially useful species. The authors concluded that Acacia seyal forms one tree ring per year, i.e. annual rings.

Figure 5.2: Acacia seyal growth ring boundary. 1. The arrows indicate growth ring boundaries with different width, 2. The arrow indicate magnified thin band of marginal parenchyma which determine the growth ring boundary.

Variation in rings width of Acacia seyal was observed in the current study (see Figure 5.2/1). Similarly, NICOLINI et al. (2010) found that Acacia seyal annual ring widths vary from less than 2 mm to more than 16 mm with an overall mean of 6.47 mm, while GEBREKIRSTOS et al. (2008) found rings of 0.27–9.12 mm with a mean of 2.32 mm. NICOLINI et al. (2010) reported that the variation in tree ring width may be attributed to precipitation. This may explain the big differences between the mean growth ring widths found by NICOLINI et al. (2010) of 6.47 mm in comparison with the 2.32 mm found by GEBREKIRSTOS et al. (2008).

- 91 -

Results and Discussion

5.1.1.2 Balanites aegyptiaca

Description of the anatomical composition of Balanites aegyptiaca according to the current study is provided in Table 5.2. A comparison with the anatomical description of Balanites aegyptiaca obtained from FAHN et al. (1986) (in Insidewood database) and NEUMANN et al. (2000) is also provided in Table 5.2. As for Acacia seyal , the description provided by NEUMANN et al. (2000) was modified to be comparable with the IAWA List of Microscopic Features for Hardwood Identification and was then compared with the current study results. The anatomical description provided by FAHN et al. (1986) was obtained from the Insidewood database where the anatomical descriptions are based on the IAWA List of Microscopic Features for Hardwood Identification (IAWA committee 1989).

- 92 -

Results and Discussion

Table 5.2: Anatomical composition description of Balanites aegyptiaca in comparison with literature

Description sources (*) Feature FAHN et al. NEUMANN et al. Study results (1986) (2000) Growth rings (1) distinct (2) indistinct (2) indistinct (5) Wood diffuse-porous V V (7v) Vessels in diagonal and / or radial pattern V x (11) Vessel clusters common V x (13) Simple perforation plates V V (20) Intervessel pits alternate V V (24) Minute - <= 4 µm V V (30) Vessel-ray pits with distinct borders; similar to Vessels intervessel pits in size and shape throughout the ray V V cell (42) 100 - 200 µm (Mean tangential diameter of V V lumina) (46) <= 5 vessels per square millimeter, V V (47) 5 - 20 vessels per square millimeter V x (52) <= 350 µm mean vessels element length (short) V ? (56) Tyloses common x x (60) Vascular / vasicentric tracheids present V V (61) Fibres with simple to minutely bordered pits V x (62) Fibres with distinctly bordered pits V V Tracheid and (63) Fibre pits common in both radial and tangential V V Fibres walls (66) Non-septate fibres present V V (70) Fibres very thick-walled V x (72) 900-1600 µm fibre length (short) V ? (76) Axial parenchyma diffuse V V (77) Axial parenchyma diffuse-in-aggregates V V V V Axial (78) Axial parenchyma scanty paratracheal (86) Axial parenchyma in narrow bands or lines up to V x parenchyma three cells wide (90) Fusiform parenchyma cells V V (91) Two cells per parenchyma strand V x (99) Larger rays commonly > 10-seriate V V (102) Ray height > 1 mm V V (104) All ray cells procumbent V V Rays (106) Body ray cells procumbent with one row of V V upright and / or square marginal cells (110) Sheath cells V V (114) <= 4 / mm (Ray per millimeter) V V Storied (120) Axial parenchyma and / or vessel elements V V structure storied (121) Fibres storied Secretory elements and (131) Intercellular canals of traumatic origin V x cambial variants (136) Prismatic crystals present V V (139) Prismatic crystals in radial alignment in V V Mineral procumbent ray cells (140) Prismatic crystals in chambered upright and / or V V inclusions square ray cells (142) Prismatic crystals in chambered axial V V parenchyma cells (*) = numbers between brackets refer to the number of the anatomical features observed in the study species according to IAWA list 1989, V = the anatomical feature is exist, X = the anatomical feature is absent, ? = no available information .

- 93 -

Results and Discussion

The anatomical description provided by FAHN et al. (1986) appeared more similar to the current study description than that provided by NEUMANN et al. (2000) according to Table 5.2. However, the tyloses are absent and the growth ring boundary is indistinct in the descriptions of both FAHN et al. (1986) and NEUMANN et al. (2000). This may be due to the differences in the region of origin of the collected samples.

Figure 5.3 highlights some of the differences in anatomical features between the current study anatomical description and that of FAHN et al. (1986) as well as NEUMANN et al. (2000).

Figure 5.3: Balanites aegyptiaca anatomical structure. A and B in photo 1 show the accumulation of vessels in the beginning of the growth ring and the clusters vessels, respectively, C in photo 2 shows the existence of tyloses and D in photo 3 shows the resin canals.

The growth ring boundary of Balanites aegyptiaca wood is distinct according to the current study results. It is delimited by a thin band of marginal parenchyma combined with accumulation of vessels in the beginning of the growth ring (see Figure 5.4).

- 94 -

Results and Discussion

Figure 5.4: Balanites aegyptiaca growth ring boundary. 1. The arrows indicate growth ring boundaries with different width, 2. Growth ring boundary with a combination of marginal parenchyma and accumulation of vessels cells.

Similarly, PRARAMESWARAN and CONRAD (1982), GEBREKIRSTOS et al. (2008) as well as MBOW et al. (2013) observed growth ring boundary in Balanites aegyptiaca , which is delimited by a combination of marginal parenchyma and accumulation of vessels (GEBREKIRSTOS et al. 2008 and MBOW et al. 2013). The width of ring of Balanites aegyptiaca in the present study is variable (but no special width-analysis of the increment zones was carried out). Similarly, GEBREKIRSTOS et al. (2008) found a range of 0.34–5.25 mm growth ring width in Balanites aegyptiaca with overall means of 1.93 mm. Concerning the dendrochronological potential, MBOW et al. (2013) classified Balanites aegyptiaca as problematic and poor where the tree ring boundary detection is difficult and not useful for dendrochronology.

Generally, it is observed from the present study that the growth ring boundary in Acacia seyal (deciduous species) is more distinct than that of Balanites aegyptiaca (evergreen species) . This observation is confirmed by GEBREKIRSTOS et al. (2008) as well as MBOW et al. (2013). Moreover, a tendency for distinct ring formation by deciduous species compared to evergreen species was reported by WORBES (1999). The frequent occurrences of wedging rings in Balanites aegyptiaca made the determination of growth boundaries more difficult, as observed in the present study and also reported by GEBREKIRSTOS et al. (2008). Generally, the wedging rings in Balanites aegyptiaca were observed in the outer part near the bark in the present study. The same behaviour was observed in Terminalia guianensis by WORBES (2002) where the wedging rings were observed in the outer part when the tree starts to form buttresses. The stem of Balanites aegyptiaca also forms buttresses near the bark (see Figure

- 95 -

Results and Discussion

2.8 in section 2.3.2.2). The occurrence of wedging rings in the outer part is hence believed to be related to the onset of buttresses formation.

5.1.2 Fibres characteristics

5.1.2.1 Radial variation

Understanding the radial variation in wood fibre characteristics and related anatomical features provides a basis for improving wood utilization. In this part, the fibre diameter, lumen diameter, wall thickness, flexibility coefficient and Runkel ratio radial variations are discussed according to the variation between sample one (the inner part sample, taken at 10 % distance from pith to bark) and sample two (the outer part sample, taken at 90 % distance). The fibre length radial variation and its trend from pith to bark are discussed according to the variation among the five selected portions (10 %, 30 %, 50 %, 70 % and 90 %). No radial variation test was conducted in the case of slenderness ratio because it is calculated from the mean values of fibre length and diameter. However, its behaviour is discussed briefly in this section.

The results showed significant differences for all fibre characteristics, in all the studied trees in the case of fibre length and in most of the studied trees in the case of fibre diameter, lumen diameter, wall thickness, flexibility coefficient and Runkel ratio for Acacia seyal and Balanites aegyptiaca (see Appendix 9.1, 9.13 and 9.2, 9.14, respectively). Acacia seyal fibre length, diameter, wall thickness and Runkel ratio showed an increasing trend from the inner to the outer samples, while fibre lumen diameter as well as flexibility coefficient displayed a reverse trend. Balanites aegyptiaca showed an increase from the inner to the outer samples in all fibre characteristics, with the exception of Runkel ratio, which showed a decreased trend (Figures 5.5–5.9).

- 96 -

Results and Discussion

Figure 5.5: The study species fibre length box plots

Figure 5.6: The study species fibre diameter and lumen diameter box plots. Values for 10 % and 90 % radial distance (D = diameter and LD = lumen diameter)

- 97 -

Results and Discussion

Figure 5.7: The study species fibre wall thickness box plots. Values for 10 % and 90 % radial distance

Figure 5.8: The study species Runkel ratio box plots. Values for 10 % and 90 % radial distance

- 98 -

Results and Discussion

Figure 5.9: The study species flexibility coefficient box plots. Values for 10 % and 90 % radial distance

In general the slenderness ratio is expected to increase from pith to bark due to the increasing pattern detected in fibre length as well as fibre diameter for both species. From the above mentioned results it is obvious that the study species have different behaviour from each other in terms of lumen diameter, flexibility coefficient and Runkel ratio radial variation pattern. This confirmed the fact that the pattern of radial variation is not the same in different species. In Acacia seyal the radial decreasing trend in fibre lumen diameter leads to an increase in Runkel ratio and decrease in flexibility coefficient in the radial direction.

An increase of fibre characteristics was also observed in different hardwood species in previous studies (see Table 5.3). In contrast to the fibres characteristics significant increasing trend observed in the present study and confirmed in other studies (Table 5.3), PANDE and SINGH (2009) found a non-significant increase in fibre length, fibre diameter and fibre wall thickness in 4-year-old clonal ramets of Eucalyptus tereticornis Sm. Similarly, UNIYAL (2012) found non-significant differences in Runkel ratio from pith to bark in Eucalyptus tereticornis Sm . However , OLUWAFEMI and TUNDE (2008) concluded that fibre diameter, lumen diameter, flexibility coefficient and Runkel ratio did not show a distinct radial pattern in Sterculia setigera growing in Nigeria.

- 99 -

Results and Discussion

Table 5.3: Some hardwood species with fibre characteristics radial increase pattern

Species Increased characteristics References Acacia auriculiformis FWT CHOWDHURY et al. (2013) Acacia melanoxylon FWT TAVARES et al. (2011) Acer velutinum FL, FWT,RR, SR KIAEI (2012) Casuarina equisetifolia FWT CHOWDHURY et al. (2012) Eucalyptus globules FWT JORGE (1994) Eucalyptus grandis × E. Urophylla FL, FD, FWT QUILHÓ et al. (2006) Eucalyptus tereticornis Sm. FL, FD, FWT SHARAMA et al. (2005) Eucalyptus tereticornis Sm. FL, FD, FWT,SR UNIYAL (2012) Hevea brasiliensis FL, FD, FLD, FWT NAJI et al. (2012) Hyeronima alchorneoides, FL BUTTERFIELD et al. (1993) Vochysia guatemalensis Neolamarckia cadamba Roxb. FL, FD, FWT ISMAIL et al. (1995) Quercus castaneaefolia FL, FD, FLD, FWT, FC BAKHSHI et al. (2012) Tecomella undulata FL, FD, FLD RAO et al. (2003)

FL = fibre lengtgh, FD = fibre diameter, FLD = fibre lumen diameter FWT = fibre wall thickness, FC = flexibility coefficient, RR = Runkel ratio and SR = slenderness ratio.

ISMAIL et al. (1995), SHARMA et al. (2005), CHOWDHURY et al. (2012), and KIAEI (2012) found a radial decreasing trend in fibre lumen diameter, which is in agreement with that found in Acacaia seyal and in disagreement with that of Balanites aegyptiaca . The increasing trend in fibre lumen diameter found by RAO et al. (2003), BAKHSHI et al. (2012) and NAJI et al. (2012) (see Table 5.3) is in disagreement with those found in Acacia seyal . Contrary to the radial increasing trend of fibre diameter found in the current study for both species, KIAEI (2012) and CHOWDHURY et al. (2013) reported a decreasing trend of fibre diameter from pith to bark.

Generally, the study results are in agreement with those of OSMAN (2001), who found radial increasing trend in fibre dimensions (length, diameter, lumen diameter, and wall thickness) in ten hardwood species grown in Sudan, including Acacia seyal and Balanites aegyptiaca . However, the study results disagreed with those of OSMAN (2001) in respect of the decreasing trend found in fibre lumen diameter of Acacia seyal vs. the increasing trend observed by OSMAN (2001). In agreement with the decreasing trend found in the flexibility coefficient of Acacia seyal , KIAEI (2012) found a radial decreasing trend in the flexibility coefficient of Acer velutimum . In contrast to the current results, some authors found no radial pattern of fibre wall thickness (RAMIREZ et al. 2009).

The trend of fibre length from pith to bark was also determined in the current study. According to the study results, a significant increase trend in fibre lengths along the radial positions from pith to bark was observed in all the studied trees for both species. Figures 5.10

- 100 -

Results and Discussion

and 5.11 show the general radial trend for all of the studied trees (see also Appendix 9.13 and 9.14). Figures 5.12 and 5.13 show the fibre length radial variation using the means of the studied 20 trees.

Figure 5.10: Acacia seyal fibre length radial variation trend by tree

- 101 -

Results and Discussion

Figure 5.11: Balanites aegyptiaca fibre length radial variation trend by tree

This pattern has been reported previously by the pioneering wood scientists PANSHIN and DE ZEEUW (1980) as well as ZOBEl and VAN BUJITENEN (1989). The same trend has been found in several hardwood species (see Table 5.4). However, BRASIL and FERREIRA (1972) did not observe significant radial variation in fibre lengths at breast height in 3-year- old Eucalyptus grandis trees.

- 102 -

Results and Discussion

Table 5.4: Some hardwood species with fibre length radial increase pattern

Species References Acacia auriculiformis CHOWDHURY et al. (2009) Acacia mangium of different tree ages SAHRI et al. (1993) and HONJO et al. (2005) Acacia melanoxylon TAVARES et al. (2011) Eucalyptus camaldulensis (seedling of 9 years old) VEENIN et al. (2005) JORGE et al. (2000), MIRANDA et al. (2001, Eucalyptus globulus of different tree ages 2003), MIRANDA and PEREIRA (2002) Eucalyptus grandis MALAN and HOON (1992) Eucalyptus regnans BISSET and DADSWELL (1949) Quercus garryana Dougl. LEI et al. (1996) Tectona grandis IZEKOR and FUWAPE (2012) Vochysia aurifera, Hyeronima alchorneoides BUTTERFIELD et al. (1993)

In accordance with the results of the current study, other authors observed radial increase of fibre lengths in Balanites aegyptiaca (PARAMESWARAN and CONRAD 1982 and OSMAN 2001) and Acacia seyal (OSMAN 2001). The increase in fibre length could be explained on the basis of the increase in length of cambial initials with increasing cambial age.

The fibre length of both species showed an increase in the first three portions from pith to bark (10 %, 30 % and 50 %), and then became more or less stable in the last two portions (70 % and 90 %) (see Figures 5.12 and 5.13). A similar conclusion was drawn in studies of other hardwood trees species; for instance, in 11-year-old Acacia auriculiformis by CHOUDHURY et al. (2009), in Casuarina equisetiflolia by LEI et al. (1996), in five different poplar clones by ZHA et al. (2005) and in I-214 poplar ( Populusx. canadensis cv. “I-214”) by JIANG et al. (2003). However, BHAT et al . (1989) in 11 tropical Indian hardwoods growing in Kerala found that fibres length variation, in the majority of the species studied showed a decline near the bark after an initial increase from the pith outwards.

The fibre length radial trend of the study species may be explained using the polynomial equations described in Figures 5.12 and 5.13. The R-squared values of the used polynomial equations for both species are high, the thing which enhance their reliability.

- 103 -

Results and Discussion

Figure 5.12: Acacia seyal fibre length radial variation trend. mean ± Std (bars with different letters are significantly different from each other at 0.05 probability level)

Figure 5.13: Balanites aegyptiaca fibre length radial variation trend. mean ± Std (bars with different letters are significantly different from each other at 0.05 probability level)

- 104 -

Results and Discussion

According to Figure 5.12, Acacia seyal fibre length ranged from 1.07 mm near the pith (portion 10 %) to 1.35 mm near the bark (portion 90 %), with a 126.17 % increase. Balanites aegyptiaca fibre lengths ranged from 1.00 mm to 1.23 mm, with a 123 % increase (Figure 5.13).

In general, the increase in fibre lengths along the radial position for both species is considered small in comparison with that recorded in other hardwood species such as Vochysia guatemalensis , in which an increase of 277 % for fibre length from pith to bark is recorded for natural stand while an increase of 213 % is recorded for plantation (BUTTERFIELD et al. 1993). Eucalyptus regnans fibre length varied from 0.6 mm near the pith to 1.35 mm outward (225 % increase) (BISSET and DADSWELL 1949). JORGE et al. (2000) in their study of the variability of fibre length in wood and bark in Eucalyptus globules found an increase from 0.70 mm to 1.20 mm for 10 % and 90 % distance from the pith at 15 % height level (with 171.4 % increase) versus an increase from 0.70 mm to 1.13 mm at 75 % height level (with 161.4% increase). In Hyeronima alchorneoides , an increase of 155% is recorded for natural stand and 187 % for plantation (BUTTERFIELD et al. 1993). TAVARES et al. (2011) recorded a slightly higher percentage of fibre length increases from pith to bark (in comparison with the study species) of 141.3 % in Acacia melanoxylon with values of 0.75 mm fibre length in 10 % of the distance from the pith to 1.06 mm at 90 % distance from the pith.

5.1.2.2 Trees, forests and regions variation

Due to the significant differences observed among the radial samples in section 5.1.2.1, the variation between trees, forests and regions were conducted separately for each sample. As previously mentioned in section 5.1.2.1, no variation test was conducted in the case of slenderness ratio because it is calculated from the mean values of fibre length and diameter.

Tables 5.5–5.6 and 5.7–5.8 represent briefly the between trees, forests and regions variation results for Acacia seyal and Balanites aegyptiaca , respectively (see also Appendix 9.3–9.7 for Acacia seyal and 9.8–9.12 for Balanites aegyptiaca ). The variation of fibres characteristics between the two climatic zones is presented in chapter 5.4.

- 105 -

Results and Discussion

Table 5.5: Acacia seyal fibres characteristics variation (without fibre length) (ANOVA/T-test) Variation sources T Tree Forest Region Region Forest N S L W F R S L W F R S L W F R Zone D D D N D T C R N D T C R N D T C R 1 S N S N N Al Obeid Al Ein 2 2 S S N S N 1 S S S S S 1 N S S S S Habila 2 2 S N S S S 1 S S S S S Al 1 N N S S S Al 2

One Homora 2 N N N S S Duwaim 1 N N N N N Aum Top 2 2 S S N N N 2 N N N N S 2 N N S N N Goz 1 N S N S S 2 Fagor 2 N N N S S 1 N S N S S Tawla 2 1 N N N N N Al 2 N S S S S Damazien Khor 1 N N N N S 1 S S N N S 2 2 N N N N S Donia 2 N N N S N Al 1 N N N N N 2

Two Sheheata 2 N N N N S 1 N S S S S Al 1 N N S S N Kadugli 2 Homara 2 N N N N S 2 S S S N N 1 N N N N N 2 S S N S S Al Bardab 2 2 N N N N S TN = number of trees, SN = sample number (1 from 10% and 2 from 90%), D = diameter, LD = lumen diameter and WT = wall thickness, FC = flexibility coefficient, RR = Runkel ratio, S = significant difference, N = non significant difference at 0.05 probability level, X = no statistics are available.

According to Table 5.5, Acacia seyal between trees variation showed no significant differences in fibre diameter, lumen diameter and wall thickness, with the exception of a few cases. More variation between trees was detected in the flexibility coefficient and Runkel ratio. The variations among forests and between regions were significant in some cases and non-significant in others.

Regarding fibre length, between trees and among forests variations were observed in almost all cases, while significant differences were observed between regions in some cases (Table 5.6).

- 106 -

Results and Discussion

Table 5.6: Acacia seyal fibres length variation (ANOVA/T-test)

Variation sources T Region Forest N Tree Forest Region

Zone 10 30 50 70 90 10 30 50 70 90 10 30 50 70 90 % % % % % % % % % % % % % % % Al Obeid Al Ein 2 S S S S S Habila 2 S S S N N Al Al Homora 2 S S S S S N N N S S

One Duwaim S S S S S Aum Top 2 S S S S N Goz Fagor 2 S N S S S Al Tawla 2 S S S S S S S S S S Damazien Khor Donia 2 S N S N S Al Sheheata 2 S S S S S S S N S S Two Kadugli Al Homara 2 N N S N S S S S S S Al Bardab 2 S S S S S TN = number of trees, S = significant difference, N= non significant difference at 0.05 probability level, X = no available statistics.

Balanites aegyptiaca trees exhibit different behaviour than Acacia seyal . According to Table 5.7, the between trees variation was much greater than that observed in Acacia seyal in respect of fibre diameter, lumen diameter, wall thickness, flexibility coefficient and Runkel ratio. This result confirms the fact reported by ZOBEL and VAN BUIJTENEN (1989) that the amount of tree-to-tree variation differs considerably among species. While no species has uniform wood, some are much more uniform than others. Significant differences between forests and regions were observed in some cases in Balanites aegyptiaca (Table 5.7).

Similar to the fibre length variation results for Acacia seyal , the between trees and forests variation in Balanites aegyptiaca fibre length were observed in almost all cases, while significant differences were observed between regions in some case (Table 5.8).

In general, the tree-to-tree differences in wood properties are large, as cited in the literature (ZOBEL and VAN BUIJTENEN 1989). In agreement with the significant difference detected between trees in fibre length for both species, considerable differences in fibres length were reported among trees for several hardwood species; for instance, in Carya pecan by TAYLOR (1969), in Eucalyptus grandis by HANS et al. (1972) and in Acacia auriculiformis by CHOUDHURY et al . (2009). However, BHAT et al . (1990) and LEI et al. (1996) observed no significant difference in fibre length between trees.

ISMAIL et al. (1995) found significant variation between trees in all the studied anatomical properties of Neolamarckia cadamda , including fibre length, diameter, lumen diameter wall thickness and vessel diameter. The authors’ results agreed with the findings of the current study for Balanites aegyptiaca and disagreed with those for Acacia seyal .

- 107 -

Results and Discussion

Table 5.7: Balanites aegyptiaca fibres characteristics variation (without fibre length) (ANOVA/T-test) Variation sources T Tree Forest Region Region Forest N S L W F R S L W F R S L W F R Zone D D D N D T C R N D T C R N D T C R 1 S S N S S Al Obeid Al Tainat 2 2 S S N S S 1 S N N N N 1 N N N S N Habila 2 2 S S S S S 1 S S S N S 1 S S S N N Al Homora 2

One Al Duwaim 2 S N S N N 1 S S S S N Aum Top 2 2 S S S S S 2 N N N N N 2 S S S S S 1 N N N N N Goz Fagor 2 2 N S S S S 1 N S S S S Tawla 2 1 N S N N S Al 2 N N S N N Damazien Khor 1 S S N S S 1 S S S N N 2 2 S S N S S Donia 2 N S N S S Al 1 N S N S S 2

Two Sheheata 2 N S N S S 1 S S N N N 1 S S N S S Kadugli Al Homara 2 2 S S N S S 2 S N S N N 1 S S N S S 2 N N N N N Al Bardab 2 2 N S N S S TN = number of trees, SN = sample number (1 from 10 % and 2 from 90 %), D = diameter, LD = lumen diameter and WT = wall thickness, FC = flexibility coefficient, RR = Runkel ratio, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

Table 5.8: Balanites aegyptiaca fibres length variation (ANOVA/T-test)

Variation sources T Region Forest Tree Forest Region N Zone 10 30 50 70 90 10 30 50 70 90 10 30 50 70 90 % % % % % % % % % % % % % % % Al Obeid Al Tainat 2 S S S S S Habila 2 S S S S N Al Al Homora 2 S S S S S S S S S S

One Duwaim S S S S S Aum Top 2 S S S N S Goz Fagor 2 S N N S S Al Tawla 2 N S S S S S N N S N Damazien Khor Donia 2 S N N N S Al Sheheata 2 N S N N S S S S S S Two Kadugli Al Homara 2 S S N N S S S S S S Al Bardab 2 S S S S N TN = number of trees, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

In Acacica seyal , the no significant differences detected in fibre diameter, lumen diameter and wall thickness vs. significant differences in fibre length corresponds with the findings of TAYLOR (1973) who reported significant differences between trees in terms of fibre length and non-significant differences in fibre diameter. The current study results agreed with those

- 108 -

Results and Discussion

of CHOWDHURY (2013) who found no significant differences among trees in fibre tangential diameter as well as wall thickness in Acacia auriculiformis growing in Bangladesh.

The detected variations between forests and regions in all the studied fibre characteristics for both species were also expected according to the literature (ZOBEL and VAN BUIJTENEN 1989), and may be due to many factors such as the environment, geographical factors, soils, genetics, etc. These results agreed with the findings of PANDE et al. (2008) that the locations significantly affect the anatomical properties and pulping and paper quality ratios of Leucaena leucocephala (Lam.) de wit. Similarly, MONTEOLIVA et al. (2005) found differences between clones as well as sites in fibre length of 13-year-old willow ( Salix SPP.) in Argentina. However, the current study results disagreed with those of MOGLIA and LOPEZ (2001), which held that fibre diameter lumen diameter and wall thickness were not affected by locations.

From Tables 5.5–5.8, it is obvious that the variation in fibre length is greater than the other studied fibre characteristics in all variation levels (trees, forests and regions) for both species. This result was expected due to the high sensitivity of fibre length to tree age.

5.1.2.3 Summarizing description

The fibre characteristics mean values are provided in Table 5.9. Additionally, information about the fibre characteristics mean values for inner and outer samples are provided separately in Table 5.10 in order to provide a general idea about the expected mean values for juvenile and mature wood.

Table 5.9 : The study species fibres characteristics. Mean ± Std (min.-max.), *Slenderness values doesn’t include Std., Min., and Max.values because it is calculated from the mean values of fibre length and diameter.

Species Fibre Characteristic Acacia seyal Balanites aegyptiaca Length (mm) 1.26 ± 0.19 (0.60 - 1.90) 1.15 ± 0.14 (0.60 - 1.60) Diameter (µm) 10.70 ± 2.52 (4.85 - 20.91) 13.01 ± 2.57 (6.00 - 22.55) Lumen diameter (µm) 3.71 ± 1.37 (1.09 - 9.64) 5.28 ± 1.62 (1.45 - 11.64) Wall thickness (µm) 3.46 ± 0.82 (1.70 - 7.09) 3.81 ± 0.86 (1.64 - 6.65) Flexibility coefficient (%) 34.44 ± 7.99 (14.29 - 62.86) 41.09 ± 8.37 (19.23 - 67.66) Runkel ratio 1.99 ± 0.68 (0.58 - 4.83) 1.48 ± 0.49 (0.42 - 3.33) Slenderness ratio* 117.76 88.39

According to the study species fibre characteristics mean values provided in Table 5.9, fibre lengths and diameters of both species are within the range of hardwood of 0.7–2.0 mm and

- 109 -

Results and Discussion

10–60 µm, respectively (ILVESSALO-PFAFFLI 1995). The study species fibres can be classified as short according to WAGENFÜHR (1989), medium according to IAWA (1989) and METCALFE and CHALK (1983) classifications (see Table 2.1 in section 2.1.1.1). The study species fibre lengths are comparable with the values provided in Tables 2.2 and 2.3 for tropical hardwood species. Both species have thick wall fibres according to CHATTAWAY (1932) classification, and are also within the range for hardwood species (3.0 – 7.0 µm) as cited by KHRISTOVA et al. (1998) (see also Table 2.4 and 2.5). The thick fibre wall leads to obtain a high Runkel ratio (>1) for both species. Fibres of both species have a high slenderness ratio compared to those of some softwood and certainly to most hardwoods.

Fibres are classified into four groups according to the flexibility coefficient (elasticity coefficient) (BEKTAS et al. 1999): High elastic fibres have an elasticity coefficient greater than 75, elastic fibres have an elasticity ratio of 50–75, rigid fibres have an elasticity ratio of 30–50, and highly rigid fibres have an elasticity ratio of less than 30. According to this classification, the study species belong to the rigid fibres group.

Table 5.10 : The study species inner and outer wood fibres characteristics. Mean ± Std (min.-max.), *Slenderness values don’t include Std., Min., and Max.values because it is calculated from the mean values of fibres length and diameter. Species Fibres Characteristics SN Acacia seyal Balanites aegyptiaca 1 1.17 ± 0.19 (0.60 - 1.86) 1.07 ± 0.14 (0.60 - 1.54) Length (mm) 2 1.34 ± 0.16 (0.94 - 1.90) 1.23 ± 0.11 (0.90 - 1.60) 1 10.36 ± 2.34 (4.85 - 17.94) 12.53 ± 2.52 (6.18 - 20.61) Diameter (µm) 2 11.05 ± 2.64 (5.27 - 20.91) 13.49 ± 2.52 (6.00 - 22.55) 1 3.80 ± 1.34 (1.21 - 8.28) 4.92 ± 1.44 (1.45 - 8.91) Lumen diameter (µm) 2 3.61 ± 1.39 (1.09 - 9.64) 5.64 ± 1.70 (1.47 - 11.64) 1 3.23 ± 0.72 (1.70 - 5.39) 3.70 ± 0.86 (1.64 - 6.65) Wall thickness (µm) 2 3.69 ± 0.84 (1.73 - 7.09) 3.91 ± 0.85 (1.82 - 6.31) 1 37.10 ± 7.99 (17.30 - 62.86) 39.62 ± 8.02 (20.51 - 67.26) Flexibility coefficient (%) 2 31.78 ± 7.06 (14.29 - 57.81) 42.51 ± 8.45 (19.23 - 67.66) 1 1.75 ± 0.55 (0.58 - 4.07) 1.56 ± 0.51 (0.49 - 3.33) Runkel ratio 2 2.22 ± 0.71 (0.73 - 4.83) 1.39 ± 0.47 (0.42 - 3.04) 1 103.28 79.81 Slenderness ratio* 2 122.18 91.18 SN = sample number, for all fibre characteristics sample 1 is taking from portion 10 % and sample 2 from 90 % with the exception of fibre length where sample 1 is the mean of portions 10 % and 30 % and sample 2 is mean of portions 70 % and 90 %.

Table 5.10 provides information about the fibre characteristics expected mean values for juvenile and mature wood. For all the studied fibre characteristics (except length), the values of juvenile and mature wood were obtained from mean values of the inner wood samples (taken at 10 % distance from pith to bark) and outer wood samples (taken at 90 %),

- 110 -

Results and Discussion

respectively. As shown in Figures 5.12 and 5.13 in section 5.1.2.3, fibre length mean values of portions 10 % and 30 % are significantly different from each other, while those of portions 70 % and 90 % are not significantly different from each other for both species. Therefore, portions 10 % and 30 % may be considered as juvenile wood and those of 70 % and 90 % as mature wood. Thus, in Table 5.10, fibres length values were obtained from the mean values of portions 10 % and 30 % in the case of juvenile wood and 70 % and 90 % in the case of mature wood, in order to provide more reliable values for fibre lengths for juvenile and mature wood.

Tables 5.11 and 5.12 provide comparisons between the study species fibre characteristics obtained in the current study and those in the literature.

Table 5.11 : Acacia seyel fibres characteristics in comparison with literature

Fibres Study results References KHRISTOVA YOUSIF OSMAN KHRISTOVA SHAWGI Characteristics IN OT M et al. (1998) (2000) (2001) et al. (2004) (2007) Length (mm) 1.17 1.34 1.26 1.2 1.09 1.07 0.8 1.09 Diameter (µm) 10.36 11.05 10.70 14.8 18.0 14.7 13.8 18.8 Lumen 3.80 3.61 3.71 4.7 - 12.2 - 7.87 diameter (µm) Wall thickness (µm) 3.23 3.69 3.46 5.0 - 2.6 2.3 5.49 Flexibility coefficient (%) 37.10 31.78 34.44 32 - - 77 42.40 Runkel ratio 1.75 2.22 1.99 2.1 - - 0.3 1.65 Slenderness ratio 103.3 122.2 118.8 77 - - 64 59.9 IN = inner samples, OT = outer samples and M = mean of inner and outer samples

According to Table 5.11, the fibre lengths of Acacia seyal obtained from the current study are greater than those obtained by KHRISTOVA et al. (2004), OSMAN (2001), SHAWGI (2007) and YOUSIF (2000). The fibre length values obtained by the previously mentioned authors are more comparable with those of the inner samples of the current study, which are considered to be juvenile wood. On the other hand, the fibre diameter mean values obtained in the current study are smaller than those obtained by the previously mentioned authors. However, KHRISTOVA et al. (1998) reported a similar value of fibre length as that obtained in the current study.

The fibre lumen diameter of Acacia seyal is extremely small in comparison with those obtained by OSMAN (2001) and SHAWGI (2007) but comparable with those of KHRISTOVA et al. (1998). Acacia seyal fibres wall thickness revealed in the current study is bigger than those revealed by both KHRISTOVA et al. (2004) and OSMAN (2001) but smaller than those revealed by KHRISTOVA et al . (1998) and SHAWGI (2007).

- 111 -

Results and Discussion

Acacia seyal slenderness ratio calculated in the current study (124) is much bigger than those recorded by KHRISTOVA et al. (1998), KHRISTOVA et al. (2004), and SHAWGI (2007) of 77, 64 and 59.9 respectively. KHRISTOVA et al. (2004) and SHAWGI (2007) recorded higher values of flexibility coefficient (77 % and 42 %) and smaller values of runkel ratio (0.3 and 1.65) than those estimated in the current study of 35.6 % and 1.98, respectively. While the flexibility coefficient and runkel ratio values recorded by KHRISTOVA et al. (1998) (32 % and 2.1) are close to that estimated in the current study.

KHRISTOVA et al. (2004) collected samples from 5- and 6-year-old-trees (previously mentioned in Table 2.7). This may be the reason for the small obtained values of fibre length, lumen diameter and wall thickness in comparison to the values obtained from the current study. However, the reasons for the higher values of fibre diameter, lumen diameter and wall thickness obtained by SHAWGI (2007) may be due to the fact that the author measured the fibre dimension in the middle point of fibre length, and therefore the obtained values can be considered as maximum values.

Table 5.12: Balanites aegyptiaca fibres characteristics in comparison with literature Study results References Fibres PARAMESWARAN KHRISTOVA OSMAN SHAWGI Characteristics IN OT M and CONRAD et al. (1997) (2001) (2007) (1982) Length (mm) 1.07 1.23 1.15 1.160 1.11 1.175 1.038 Diameter (µm) 12.53 13.49 13.01 - 16.07 12.3 20.6 Lumen diameter (µm) 4.92 5.64 5.28 2.5 - 8.00 4.06 10.5 10.04 Wall thickness (µm) 3.70 3.91 3.81 2.5 - 6.00 6.01 1.9 5.30 Flexibility coefficient (%) 39.62 42.51 41.09 - 25.3 - 48.80 Runkel ratio 1.56 1.39 1.48 - 2.9 - 1.14 Slenderness ratio 79.81 91.18 89.16 - 69.1 - 51.9 IN = inner samples, OT = outer samples and M = mean of inner and outer samples

From Table 5.12, the mean fibre length of Balanites aegyptiaca obtained in the current study is similar to that obtained by PARAMESWARAN and CONRAD (1982) and close to that of OSMAN (2001). In the other side, it is larger than those of KHRISTOVA et al. (1997) and SHAWGI (2007). Balanites aegyptiaca ’s fibre diameter mean value obtained in the current study is similar to that obtained by OSMAN (2001) and smaller than those of KHRISTOVA et al. (1997) and SHAWGI (2007), while the lumen diameter value is much lower than those obtained by OSMAN (2001) and SHAWGI (2007) and bigger than those obtained by KHRISTOVA et al. (1997) and PARAMESWARAN and CONRAD (1982).

- 112 -

Results and Discussion

Regarding the fibre wall thickness of Balanites aegyptiaca , its mean value is smaller than those described by KHRISTOVA et al. (1997) and SHAWGI (2007), and within the range of the values reported by PARAMESWARAN and CONRAD (1982). SHAWGI (2007) found smaller values of both Runkel ratio (1.14) and slenderness ratio (51.9) than those calculated in the current study, while that for flexibility coefficient was greater than that estimated in the current study. Moreover, KHRISTOVA et al. (1997) found larger values of Runkel ratio and smaller values of flexibility coefficient and slenderness ratio.

A comparison between the studied species fibres characteristic mean values of trees growing in Blue Nile, White Nile as well as South Kordofan states and those obtained from literature for the same regions was conducted (see Tables 5.13 - 5.16).

Table 5.13: Acacia seyel fibres characteristics of tree growing in Blue Nile state in comparison with literature

Study results References Fibres Inner Outer KHRISTOVA YOUSIF KHRISTOVA SHAWGI Characteristics Mean sample sample et al. (1998) (2000) et al. (2004) (2007) Length (mm) 1.12 1.30 1.24 1.2 1.09 0.8 1.086 Diameter (µm) 10.29 10.51 10.40 14.8 18.0 13.8 18.8 Lumen diameter (µm) 3.71 3.33 3.52 4.7 - - 7.87 Wall thickness (µm) 3.27 3.57 3.42 5.0 - 2.3 5.49 Flexibility coefficient (%) 36.11 31.28 33.66 32 - 77 42.40 Runkel ratio 1.84 2.17 2.00 2.1 - 0.3 1.65 Slenderness ratio 108.84 123.69 119.23 77 - 64 59.9

According to Table 5.13, Acacia seyal ’s fibres characteristics mean values obtained in the current study are close to those obtained by KHRISTOVA et al. (1998) and higher than those obtained by KHRISTOVA et al. (2004). As previously mentioned in Table 2.7, KHRISTOVA et al. (2004) collected samples from 5- and 6-year-old-trees. This may explain the smaller obtained values of fibres characteristics mean values in comparison to those obtained from the current study. YOUSIF (2000) detected a mean of 1.09 mm fibre length which is more comparable with those of the inner sample values (considered as juvenile wood) obtained in the current study results. However, the fibre diameter mean values obtained by YOUSIF (2000) are higher than those obtained in the current study.

SHAWGI (2007) detected higher values of fibre diameter, lumen diameter and wall thickness in comparison to the current study results. This my be explained by the fact that the author measured the fibre dimension in the middle point of fibre length, and therefore the obtained values can be considered as maximum values.

- 113 -

Results and Discussion

Table 5.14: Balanites aegyptiaca fibres characteristics of tree growing in Blue Nile state in comparison with literature Study results Fibres KHRISTOVA Inner Outer Characteristics Mean et al. (1997) sample sample Length (mm) 1.04 1.22 1.17 1.11 Diameter (µm) 12.21 13.01 12.60 16.07 Lumen diameter (µm) 4.78 5.52 5.16 4.06 Wall thickness (µm) 3.62 3.75 3.69 6.01 Flexibility coefficient (%) 38.98 43.41 41.23 25.3 Runkel ratio 1.58 1.36 1.47 2.9

Slenderness ratio 85.18 93.77 92.86 69.1

From Table 5.14, KHRISTOVA et al. (1997) detected lower values of fibre length, lumen diameter, flexibility coefficient and slenderness ratio than those of the current study. On the other side, the authors detected higher mean values of fibre diameter, wall thickness and Runkel ratio than those detected in the current study. In general the mean values of fibre characteristics obtained by KHRISTOVA et al. (1997) considered more comparable with those of the inner samples mean values of the current study. This may be explained by the young age of the tree collected by KHRISTOVA et al. (1997) (8-10 years old, as previously mentioned in Table 2.7).

Table 5.15: The study species fibres characteristics of tree growing in White Nile state in comparison with literature Study results Fibres OSMAN Species Inner Outer Characteristics Mean (2001) sample sample Length (mm) 1.15 1.32 1.27 1.070 Diameter (µm) 10.04 10.81 10.42 14.7 Acacia seyal Lumen diameter (µm) 3.57 3.48 3.52 12.2 Wall thickness (µm) 3.16 3.62 3.39 2.6 Length (mm) 1.09 1.25 1.17 1.175 Balanites Diameter (µm) 12.30 13.68 12.99 12.3 aegyptiaca Lumen diameter (µm) 4.82 5.76 5.29 10.5 Wall thickness (µm) 3.62 3.91 3.76 1.9

According to Table 5.15, Acacia seyal ’s fibre length mean values obtained by OSMAN (2001) are comparable with those of the inner samples mean values obtained in the current study (considered as juvenile wood). However, the fibre diameter, lumen diameter and wall thickness mean values obtained by OSMAN (2001) are not comparable with those obtained in the current study.

- 114 -

Results and Discussion

In case of Balanites aegyptiaca , the fibre length and diameter mean values obtained in the current study are comparable with those of OSMAN (2001). However, fibre lumen diameter as well as wall thickness mean values are not comparable with those of OSMAN (2001). OSMAN (2001) used the stereological count techniques to measure fibre dimensions (with the exception of fibre length). This technique depend upon equations to obtain the fibre dimensions. This may explain the differences in mean values of some fibres characteristics between OSMAN (2001) finding and those of the current study for both species.

Table 5.16: Balanites aegyptiaca fibres characteristics of tree growing in South Kordofan state in comparison with literature Study results Fibres SHAWGI Inner Outer Characteristics Mean (2007) sample sample Length (mm) 1.05 1.22 1.13 1.038 Diameter (µm) 12.97 13.54 13.26 20.6 Lumen diameter (µm) 5.15 5.59 5.37 10.04 Wall thickness (µm) 3.82 4.02 3.92 5.30 Flexibility coefficient (%) 40.69 41.80 41.27 48.80 Runkel ratio 1.52 1.41 1.47 1.14 Slenderness ratio 80.96 92.01 85.22 51.9

From Table 5.16, Balanites aegyptiaca ’s fibres characteristics mean values obtained in the current study are not comparable with those of SHAWGI (2007) with the exception of fibre length which is considered comparable. As previously mentioned in Table 2.19, SHAWGI (2007) measured the maximum values of fibre diameter, lumen diameter and wall thickness, therefore, their mean values are much bigger than those of the current study.

5.1.3 Vessels characteristics 5.1.3.1 Radial variation

The vessel characteristics radial variation was discussed according to the variation between the inner part samples and the outer part samples. The results showed significant differences for all the studied vessel dimensions for both species (see Appendix 9.15 and 9.16).

Balanites aegyptiaca vessel dimensions showed an increase from the inner to the outer samples (Figures 5.14 and 5.15). These results are similar to those obtained in other studies (see Table 5.17). Similarly, TSUCHIYA and FURUKAWA (2009) studied the radial variation of vessel lumen diameter in 30 hardwood species. They found a juvenile-mature pattern in radial variation of vessel lumen diameter in early wood in almost all ring porous woods and in the central 10 portions of the annual rings in diffuse porous woods.

- 115 -

Results and Discussion

Figure 5.14: The study species vessels diameter and lumen diameter box plots. Values for 10 % and 90 % radial distance (D = diameter and LD = lumen diameter)

In disagreement with the current study, some authors reported no significant radial variation in vessel diameters; for instance, VEENIN et al. (2005) in Eucalyptus camaldulensis , PANDE and SINGH (2009) in Eucalyptus tereticornis Sm. and ISHIGURI et al. (2009) in Paraserianthes falcataria .

Figure 5.15: The study species vessels wall thickness box plots. Values for 10 % and 90 % radial distance

- 116 -

Results and Discussion

Table 5.17: Some hardwood species with vessels dimensions radial increase pattern

Increased vessels Species References dimension Acacia auriculiformis VD CHOWDHURY et al. (2013) Alnus rubra VD GARTNER et al. (1997) Casuarina equisetifolia VLD CHOWDHURY et al. (2012) TAYLOR (1973), MANLAN Eucalyptus grandis VD and GERISCHER (1987) Eucalyptus tereticornis Sm. VD, VEL UNIYAL (2012) Hevea brasiliensis VD NAJI et al. (2012) Hyeronima alchorneoides, VD BUTTERFIELD et al. (1993) Vochysia guatemalensis Melia dubia VD SWAMINATHAN et al. (2012) Neolamarckia cadamba Roxb. VD, VEL ISMAIL et al. (1995) Populus×canadensis cv. “I-214" VD, VEL, VR XIAOMEI et al. (2003) Quercus garryana Dougl. VD LEI et al. (1996) Shorea acuminatissima VD ISHIGURI et al. (2012) Tecomella undulate VD RAO et al. (2003)

VD = vessels diameter, VLD = vessels lumen diameter. VEL = vessels element length and VR = vessels ratio.

5.1.3.2 Trees, forests and regions variation

The variation between trees, forests and regions was examined separately for each sample due to the significant differences observed between the inner and outer samples in section 5.1.2.1. Brief description of between trees, forests and regions variations are provided in Tables 5.18 and 5.19 (see Appendix 9.17–9.19 for Acacia seyal and Appendix 9.20–9.22 for Balanites aegyptiaca ). The variation between the two climatic zones is presented in chapter 5.4.

As shown in Table 5.18, Acacia seyal vessel dimensions showed no significant differences between trees in almost all forests. Some significant differences were found among forests and between regions. Balanites aegyptiaca showed more significant differences between trees within forests than did Acacia seyal (Table 5.19). This result is similar to that observed in fibre dimension in section 5.1.2.2. Some significant differences were found among forests and between regions in Balanites aegyptiaca (Table 5.19)

In the case of Acacia seyal , the detected non-significant differences between trees in vessel dimensions are in accordance with those found in Quercus garryana Dougl. by LEI et al. (1996), where non-significant differences in vessel diameter were detected among trees. However, these results disagreed with those of some authors who found significant differences in vessel diameter among trees, such as ISMAIL et al. (1995) in Neolamarckia cadamda , GARTNER et al (1997) in Alnus rubra and CHOWDHURY (2013) in Acacia auriculiformis . - 117 -

Results and Discussion

Table 5.18: Acacia seyal vessels characteristics variation (ANOVA/T-test) Variation sources Region Forest TN Tree Forest Region Zone SN D LD WT SN D LD WT SN D LD WT 1 S N N Al Obeid Al Ein 2 2 S N N 1 N N S 1 S S N Habila 2 2 N N S 1 S N N 1 S S S Al Al Homora 2

One 2 S N N Duwaim 1 N N N Aum Top 2 2 N N N 2 S N S 2 N N N 1 N N N Goz Fagor 2 2 N N S 1 N N S Tawla 2 1 S S S Al 2 N N S Damazien Khor 1 N N N 1 N N N 2 2 S N N Donia 2 N N N Al 1 N N S 2

Two Sheheata 2 N N N S S N 1 1 N N N Kadugli Al Homara 2 2 N N N 2 N N S 1 S S N 2 N N N Al Bardab 2 2 N N N TN = number of trees, SN = sample number (1 from 10 % and 2 from 90 %), D = diameter, LD = lumen diameter and WT = wall thickness, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

Table 5.19: Balanites aegyptiaca vessels characteristics variation (ANOVA/T-test) Variation sources Region Forest TN Tree Forest Region

Zone SN D LD WT SN D LD WT SN D LD WT 1 S S S Al Obeid Al Tainat 2 2 N N S 1 S S N 1 N N N Habila 2 2 N N S 1 S S S 1 S S S Al Al Homora 2

One 2 N N N Duwaim 1 S S S Aum Top 2 2 N N S 2 N N N 2 S S S 1 S N S Goz Fagor 2 2 N N N 1 N N N Tawla 2 1 N N S Al 2 S S S Damazien Khor 1 S S S 1 S S N 2 2 N N S Donia 2 N N S Al 1 S S N 2

Two Sheheata 2 N N N 1 S S S 1 S S S Kadugli Al Homara 2 2 S S N 2 S S S 1 N N S 2 S S S Al Bardab 2 2 S S S TN = number of trees, SN = sample number (1 from 10 % and 2 from 90 %), D = diameter, LD = lumen diameter and WT = wall thickness, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

- 118 -

Results and Discussion

As previously noted in the case of fibre dimensions, the variations between trees, forests and regions was expected due to environmental factors, geographical factors, etc.

5.1.3.3 S ummarizing description

The study species vessels characteristics values are summarized in Table 5.20. Additional information about the vessels characteristics mean values for inner (10 %) and outer samples (90 %) are provided separately in Table 5.21 in order to provide a general idea about the expected mean values for juvenile and mature wood.

As shown in Table 5.20, the vessel diameters of both species are within the normal range for hardwood species of 50–>300 µm (HOADLEY 1990). The vessels tangential diameter for both species can be classified as medium according to NORMAND (1950), METCALFE and CHALK (1979) and WAGENFÜHR (1980). Examples of tropical hardwood species with small and large vessel diameters are provided in Tables 2.8 and 2.9 in section 2.1.1.2.

Table 5.20: The study species vessels characteristics. Mean ± Std (min.-max.) Species Characteristics (µm) Acacia seyal Balanites aegyptiaca Diameter 122.40 ± 38.96 (27.13 - 246.62) 125.00 ± 40.64 (36.83 - 279.26) Lumen diameter 103.42 ± 38.10 (16.96 - 220.93) 103.09 ± 35.90 (24.72 - 233.31) Wall thickness 9.13 ± 2.36 (3.66 - 18.55) 10.68 ± 3.61 (3.66 - 24.72)

Table 5.21: The study species inner and outer wood vessels characteristics. Mean ± Std (min.-max.)

S Species Characteristics (µm) N Acacia seyal Balanites aegyptiaca 1 113.63 ± 36.69 (27.13 - 245.04) 105.05 ± 32.17 (36.83 - 213.82) Diameter 2 131.20 ± 39.21 (30.54 - 246.62) 145.37 ± 38.24 (62.99 - 279.26) 1 95.87 ± 35.16 (16.96 - 211.61) 86.83 ± 29.85 (24.72 - 184.01) Lumen diameter 2 110.96 ± 39.44 (17.44 - 220.93) 119.77 ± 33.90 (50.39 - 233.31) 1 8.57 ± 2.12 (3.99 - 17.43) 8.87 ± 2.25 (3.66 - 15.99) Wall thickness 2 9.69 ± 2.47 (3.66 - 18.55) 12.49 ± 3.81 (4.72 - 24.72) SN = sample number (1 from 10 % distance from pith to bark and 2 from 90 % distance).

The study species vessel diameter mean values obtained in the present study are within the range reported by NEUMANN et al. (2000) for the same species of 30 - 170 µm for Acacia seyal and 50 - 200 µm for Balanites aegyptiaca .

- 119 -

Results and Discussion

Balanites aegyptiaca vessel diameter value obtained by PARAMESWARAN and CONRAD (1982) of 105 µm is smaller than that of the present study but identical to the inner wood vessel diameter values found in the current study. Its wall thickness is within the range obtained by the same author of 4- 10 µm .

5.2 Density

Three different types of wood density were measured in the present study: basic density, air dry density as well as density achieved by X-ray densitometry. Information about the obtained density values for the three densities as well as the radial trend of wood basic density from pith to bark are provided in the current section.

5.2.1 Basic density

5.2.1.1 Vertical variation

To fully understand the change of wood properties with height in the tree it is necessary to study the trend in juvenile and mature woods separately as cited previously by ZOBEL and BUIJTENEN (1989). Therefore, in this part the variation between the first disc (10 % merchantable stem height) and the second disc (90 % merchantable stem height) was assessed according to the variation between each of the selected five radial portions separately.

The results showed no significant variation between the two discs for most tested trees of both species (see Appendix 9.23 and 9.24 for Acaica seyal and Balanites aegyptiaca , respectively). However, the basic density of Acacia seyal followed a declining trend with the increasing height of the tree (Figure 5.16), while that of Balanites aegyptiaca showed a more constant trend along the vertical position (Figure 5.17). Acoording to the current study results, Acacia seyal wood basic density mean values were decreased from a mean of 734.49 kg/m 3 at 10 % merchantable hight to 714.76 kg/m 3 at 90 % merchantable hight, with 2.7 % differences. In case of Balanites aegyptiaca , the wood basic density mean values were 658.91 kg/m 3 at 10 % merchantable hight and 660.05 kg/m 3 at 90 % merchantable hight, with 0.17 % differences.

- 120 -

Results and Discussion

Figure 5.16: Acacia seyal basic density vertical variation box plots (mean of 27 trees)

Figure 5.17: Balanites aegyptiaca basic density vertical variation box plots (mean of 30 trees)

ZOBEL and BUIJTENEN (1989) stated that almost all possible patterns of longitudinal variation can be found in either conifers or hardwoods, but the declining pattern is more evident in hardwood. However, most diffuse porous hardwood has little variation in wood density from the base to the top. This statement is in homogeneity with the results of the present study, where different trends and non-significant variations were observed. The results of the present study were also in agreement with those of other studies where specific gravity

- 121 -

Results and Discussion

variation with tree height was non-significant (PURKAYASTHA et al. 1982 and UNIYAL 2012 in Eucalyptus tereticornis , JAIN and ARORA 1995 in Eucalyptus camaldulensis ).

Table 5.22 provides some examples of hardwood species with a declining vertical trend in wood density regardless of the significance of the vertical variation.

Table 5.22: Some hardwood species with wood density vertical decline pattern

Species Type of measured density References Corylus colurna L. Basic density, Oven dry density ZEIDLER (2012) Eucalyptus tereticornis Sm. Specific gravity UNIYAL (2012) Gmelina arborea Roxb. Specific gravity LAMB (1968) Quercus garryana Dougl. Specific gravity LEI et al. (1996)

Contrary to the declining trend in the vertical direction found in Acacia seyal in the current study, some other authors found an increasing trend from base to the top in their studied species; for instance, GÖHRE (1960) in Populus spp , DARGAVEL (1968) and FREDERICK et al. (1982) in Eucalyptus regnans , KAMALA et al. (2000) in Grevillea robusta and SADEGH (2012) in Eucalyptus camaldulensis . In accordance with the constant vertical trend found in Balanites aegyptiaca , McELWEE and FAIRCLOTH (1966) reported no differences in wood density with stem height in Nyssa silvatica (water tupelo).

In general the decline in wood density along the vertical direction is expected in species that have a sharp radial increase in wood density. In such cases the proportion of juvenile wood increases extensively with increasing height, and as a result the wood density decreases from the base to the top (ZOBEL and BUIJTENEN 1989).

5.2.1.2 Radial variation

The basic density radial variation is discussed according to the variation among the five selected portions over the cross section (10 %, 30 %, 50 %, 70 % and 90 %). The results show significant differences among the five portions in almost all trees for Acacia seyal and in most trees for Balanites aegyptiaca (see Appendix 9.25 for Acacia seyal and 9.26 for Balanites aegyptiaca ). Generally, the wood basic density followed the radial increase pattern for both species (Figure 5.18).

- 122 -

Results and Discussion

Figure 5.18: The study species basic density radial variation box plots

The trend of wood basic density from pith to bark was determined in the current study. Figures 5.19 and 5.20 show the general radial trend for all of the studied trees (see also Appendix 9.25 and 9.26). Figures 5.21 and 5.22 show the wood density radial variation using the means of all studied trees.

- 123 -

Results and Discussion

Figure 5. 19: Acacia seyal basic density radial variation trend by tree. (Trees 6, 8 and 9 in zone one were excluded due to their abnormal density values)

- 124 -

Results and Discussion

Figure 5.20: Balanites aegyptiaca basic density radial variation trend by tree

According to Figures 5.19 and 5.20, almost all trees showed increase in basic density from pith to bark with the exception of few trees in both species where the density values were decreased from the first portion near the pith 10 % to the second portion 30 % and thin increased again until the bark. This may be explained by the possibility of the heartwood extractive, tyloses and rinse existence which concentrated near the pith and increase the density values.

- 125 -

Results and Discussion

Figure 5.21: Acacia seyal basic density radial variation trend. Mean ± Std (bars with different letters are significantly different from each other at 0.05 probability level)

Figure 5.22: Balanites aegyptiaca basic density radial variation trend. Mean ± Std (bars with different letters are significantly different from each other at 0.05 probability level)

- 126 -

Results and Discussion

According to Figure 5.21, Acacia seyal showed a sharp increase in wood basic density along the radial direction where significant differences were reported between each of the measured radial positions. Its wood density increases from 672 kg/m 3 near the pith (10 %) to 776 kg/m 3 near the bark (90 %), with 15.5 % increase. Balanites aegyptiaca showed different behaviour: the density increased slowly from the first (10 %) to the second portion (30 %), became faster in the third portion (50 %), and finally became more or less stable near the bark (see Figure 5.22). Its wood density increases from 632 kg/m 3 at 10 % portion to 681 kg/m 3 at 90 % portion, with a 7.8 % increase. This may explain the vertical decreasing trend observed in wood density of Acacia seyal vs. the constant vertical trend in Balanites aegyptiaca (see section 5.2.1.1).

The radial increasing trend in wood basic density for both species may be explained using the polynomial equations described in Figures 5.21 and 5.22. In both species, the selected equation has high R- squared values, which hence enhances their reliability.

The wood basic density radial increasing trend observed in both species may be explained by the radial increasing trend found in fibre and vessel wall thicknesses (see sections 5.1.2.1 and 5.1.3.1), particularly that observed in fibre wall thickness. These results enhance the fact provided in the literature that the wood density is mainly affected by the anatomical properties (PANSHIN and DE ZEEUW 1980, ZOBEL and VAN BUITENEN 1989 and VALENTE et al.1992). Similarly, it enhances the positive relationship found between the fibre wall thickness and wood density by SCARAMUZZI and FERRARI (1963), LEWIN and GOLDSTEIN (1991), NASROUN and ALSHAHRANI (1998), and CHOWDHURY et al. (2012). On the other hand, the radial increasing trend observed in vessel diameter may lead one to expect a radial decreasing trend in wood density, but this is not the case in the studied species. The explanation for this is that, from the literature, vessel frequency or the number of vessels per square millimetre decreases along the radial position. This fact is confirmed in several hardwood species ( Casuarina eguietifolia by CHOWDHURY et al. 2012, Eucalyptus globules by RAMÍREZ et al. 2009, Hyeronima alchorneoides and Vochysia guatemalensis by BUTTERFIELD et al. 1993, Neolamarckia cadamba Rox. by ISMAIL et al. 1995 and Shorea acuminatissima by ISHIGURI et al. 2012). The trend of vessel frequency or the number of vessels per square millimetre from pith to bark was not investigated in the present study; however, it is expected to demonstrate the same trend observed in literature (decreases along the radial position).

- 127 -

Results and Discussion

The observed radial pattern in the current study was similar to type one of PANSHIN and DEZEEUW (1980) (showing an increase from pith to bark) and in homogeneity with the conclusion of ZOBEL and BUIJTENE (1989) that the middle to high density diffuse-porous hardwoods generally follow a pattern of low specific gravity near the pith and then an increase, followed by a slower increase or levelling off towards the bark. These results are also in agreement with those of many other researchers in different hardwood species (see Table 5.23).

Table 5.23: Some hardwood species with wood density radial increase pattern

Species Type of measured density References Acacia auriculiformis Basic density CHOWDHURY et al. (2009) Air dry density (using x-ray Calycophyllum spruceanum MONTES et al. (2007) densitometer) Basic density, Oven dry Corylus colurna L. ZEIDLER (2012) density Endospermum medullosum Basic density SETTLE et al. (2012) Eucalyptus regnans Basic density FREDERICK et al. (1982) Eucalyptus tereticornis Sm. Specific gravity UNIYAL (2012) Eucalyptus urophylla Basic density HEIN et al. (2012) Gmelina arborea Roxb. Basic density LAMB (1968) Specific gravity (using x-ray Gmelina arborea Roxb. ESPINOZA (2004) densitometer) Hyeronima alchorneoides, Specific gravity BUTTERFIELD et al. (1993) Vochysia guatemalensis Shorea leprosula and S. parvifolia Specific gravity (basic) BOSMAN et al. (1994)

The current study results are similar to OSMAN’s (2001), who found a radial increasing trend in wood air dry and oven dry densities of ten hardwood species grown in Sudan including Acacia seyal as well as Balanites aegyptiaca . In disagreement with the present study results, SADEGH (2012) found a decreasing trend from pith to bark in wood basic density of Eucalyptus camaldulensis grown in Iran. HIETZ et al. (2013) found radial increasing as well as decreasing trends in basic density in their studies species. However, GARTNER et al. (1997), on six 40-year-old Alnus rubra trees, found that specific gravity remained constant along the radial direction from pith to bark.

5.2.1.3 Trees, forests and regions variation In this part the variation between trees, forests and regions is assessed for each portion separately, due to the significant differences observed among the radial samples positions in section 5.2.1.1. Tables 5.24 and 5.25 briefly present the wood basic density variation in all

- 128 -

Results and Discussion

levels (trees, forests and regions) for Acacia seyal and Balanites aegyptiaca , respectively (see also Appendix 9.27 and 9.28 for Acacia seyal and 9.29 and 9.30 for Balanites aegyptiaca ). Information about the variation between the two climatic zones is presented in chapter 5.4.

Expected variations in all levels (trees, forests and regions) were found in some cases in Acacia seyal and in most cases in Balanites aegyptiaca (Tables 9.24 and 9.25, respectively). This result is in homogeneity with those described previously in the anatomical properties (section 5.1), that Balanites aegyptiaca has more variation than Acacia seyal in almost all levels.

Table 5.24: Acacia seyal basic density variation (ANOVA/T-test)

Variation sources T Region Forest Tree Forest Region N Zone 10 30 50 70 90 10 30 50 70 90 10 30 50 70 90 % % % % % % % % % % % % % % % Al Obeid Al Ein 3 N S S S N Habila 2 N S S N N Al Al Homora S S S S N

One Duwaim S S S S N Aum Top 3 N N S S S Goz Fagor 3 N N S N N Al Tawla 3 N S N N N S N N N N Damazien Khor Donia 3 N N N N N Al Sheheata 3 S S S S S N N N N S Two Kadugli Al Homara 3 N N N S N N N N N N Al Bardab 3 S S S S S TN = number of trees, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

Table 5.25: Balanites aegyptiaca basic density variation (ANOVA/T-test)

Variation sources T Region Forest Tree Forest Region N Zone 10 30 50 70 90 10 30 50 70 90 10 30 50 70 90 % % % % % % % % % % % % % % % Al Obeid Al Tainat 3 N N S S N Habila 3 S S S S S Al Al Homora 3 S S S S N N S S S N

One Duwaim S S S S S Aum Top 2 N N N S N Goz Fagor 3 N S S S N Al Tawla 3 S S S S S N S S S S Damazien Khor Donia 3 S S S S S Al Sheheata 3 S S S S S N S S S S Two Kadugli Al Homara 3 S S S N N S S S S S Al Bardab 3 S S S S S TN = number of trees, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

- 129 -

Results and Discussion

The detected significant differences between trees are in accordance with the statement provided by ZOBEL and VAN BUIJTENEN (1989) that the tree-to-tree differences in wood properties are large. It also agrees with the significant variation found among trees in wood density of several hardwood species, such as Acacia auriculiformis by CHOUDHURY et al. (2009), Alnus rubra by GARTNER et al. (1997), Ceiba pentandra by FIMBEL and SJAASTAD (1994), Eucalyptus grandis by HANS et al. (1972) and TYLOR (1973) and in Liriodendron tulipifera by THORBJORNSEN (1961) and TYLOR (1968). However, the current study results disagreed with those of LEI et al. (1996), where no significant differences were observed among trees in specific gravity of Quercus garryana .

Variations between forests and between regions were also expected due to the variation in the geographical and environmental factors, and these are also in accordance with the literature (ZOBEL and VAN BUIJTENEN 1989). Similarly, MONTEOLIVA et al. (2005) found significant differences between clones as well as sites in wood density of Salix spp.

Despite the richness of literature about the variation between trees, forests and regions of several hardwood species, no available information was found about Acacia seyal and Balanites aegyptiaca variation.

5.2.1.4 S ummarizing description

General information about the basic density mean values, along with the values obtained from the inner wood samples (10 % and 30 %) and outer wood samples (70 % and 90 %) are provided in Table 5.26.

Table 5.26: The study species basic density. Mean ± Std (min.-max.).

Basic density (Kg/m 3) Species General inner sample outer sample 724.68 ± 62.19 685.50 ± 56.94 762.65 ± 48.15 Acacia seyal (550.00 – 960.08) (550.00 – 866.67) (643.78 – 960.08) 659.49 ± 44.23 639.54 ± 43.47 677.09 ± 38.90 Balanites aegyptiaca (495.62 – 789.79) (495.62 – 760.00) (567.94 – 789.79)

The average basic densities of both species are in the range of tropical hardwoods of 450 – 1,400 kg/m³ (KANEHIRA 1933 and TISSOT 1985). Following the classification of MELO et al. (1990) (described in chapter 2.1.2.1), Acacia seyal wood can be classified as heavy (density ≥ 720 kg/m³) while that of Balanites aegyptiaca can be classified as medium (density 500 – < 720 kg/m³).

- 130 -

Results and Discussion

Acacia seyal density obtained in the current study was higher than those obtained by other authors such as KHRISTOVA et al. (1998) and KHRISTOVA et al. (2004) (649 kg/m³ and 669–692 kg/m³, respectively). This may due to the young trees used in their studies; therefore, their values are comparable with that of the inner wood obtained in the current study (see Table 5.26). KHRISTOVA et al. (1997) found the wood basic density of Balanites aegyptiaca to be 619 kg/m³ which is more or less comparable with that of the inner wood obtained in the current study.

5.2.2 Air dry density

5.2.2.1 Vertical variation

In this part the variation between the first disc (10 % merchantable tree height) and the second disc (90 % merchantable tree height) was tested according to the variation between each of the selected three radial portions separately (10 %, 50 % and 90 %). The results showed no significant variation between the two discs for both species. The air dry density vertical trends showed a very similar pattern for those observed in basic density for both species, where a decreasing trend appeared in Acacia seyal and a constant trend appeared in Balanites aegyptiaca (Figure 5.23 and 5.24). The same trend of Acacia seyal has been observed in other hardwood species (see Table 5.18 in section 5.2.1.1. However, some other authors found the reverse trend in their studies (GÖHRE 1960, DARGAVEL 1968, FREDERICK et al. 1982, KAMALA et al. 2000 and SADEGH 2012) and some found a constant trend similar to that observed in Balanites aegyptiaca (McELWEE and FAIRCLOTH 1966).

- 131 -

Results and Discussion

Figure 5.23: Acacia seyal air dry density vertical variation box plots

Figure 5.24: Balanites aegyptiaca air dry density vertical variation box plots

5.2.2.2 Radial variation

In this section the radial variation was discussed according to the variation between the three selected portions from pith to bark (10 %, 50 % and 90 %). Significant differences were detected among the three selected radial positions in both species. The wood air dry density followed an increase pattern from pith to bark, with a sharper increase in Acacia seyal in comparison with Balanites aegyptiaca (Figure 5.25). This result is identical with that found in

- 132 -

Results and Discussion

wood basic density, and also explains the decline trend in vertical direction in Acacia seyal vs. the constant trend in Balanites aegyptiaca .

The radial increase in air dry density may be explained by the increase in fibre and vessel wall thickness as previously discussed in basic density. Similarly, some authors concluded that fibre wall thickness is positively correlated with air dry density, such as CHOWDHURY et al. (2012).

Figure 5.25: The study species air dry density radial variation box plots (plots with different letters within species are significantly different from each other at 0.05 probability level)

5.2.2.3 Trees, forests and regions variation

According to section 5.2.2.2, significant differences were detected among the three radial positions (10 %, 50 % and 90 %). Therefore, in this section the air dry density variation on the three levels (trees, forests and regions) was assessed separately for each sample. The results are summarised in Tables 5.27 and 5.28 (see also Appendix 9.31 and 9.32). Information about the variation between the two climatic zones is provided in chapter 5.4.

The results found non-significant differences on the three levels (trees, forests and regions) for both species with the exception of a few cases. In contrast to the results of the anatomical properties and wood basic density, Balanites aegyptiaca showed more or less the same behaviour as Acacia seyal .

The detected results of Acacia seyal variation in trees, forests and regions for air dry density are comparable with those detected for basic density (previously mentioned in Table 5.24).

- 133 -

Results and Discussion

However, Balanites aegyptiaca air dry density and basic density variation in the three levels are not in homogeneity with each other (refer to Table 5.25 for basic density results). This may be due to the lower number of samples measured in air dry density vs. basic density (see Appendix 9.32 for air dry density and Appendix 9.29, 9.30 for basic density).

Table 5.27: Acacia seyal air dry density variation (ANOVA/T-test)

Variation sources T Region Forest Tree Forest Region N Zone 10 50 90 10 50 90 10 50 90 % % % % % % % % % Al Obeid Al Ein 3 N N N Habila 2 S N N

Al Al Homora

One Duwaim S S N S N N Aum Top 3 N N S Goz Fagor 3 S N N Al Tawla 3 N N N N N N Damazien Khor Donia 3 N N N

Al Sheheata 3 N S S N N N Two Kadugli Al Homara 3 N N N N N N Al Bardab 3 S S S

TN = number of trees, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

Table 5.28: Balanites aegyptiaca air dry density variation (ANOVA/T-test)

Variation sources T Region Forest Tree Forest Region N Zone 10 50 90 10 50 90 10 50 90 % % % % % % % % % Al Obeid Al Tainat 3 N S S Habila 3 N N N Al Al Homora 3 N N N N N N

One Duwaim N N N

Aum Top 3 N N N Goz Fagor 3 S S N

Al Tawla 3 N S S N N N Damazien Khor Donia 3 N S S

Al Sheheata 3 N N N N S S

Two Kadugli Al Homara 3 N N N S N N Al Bardab 3 N S N

TN = number of trees, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

In disagreement with the current study results, BHAT and PRIYA (2004) found significant differences among the locations in air dry density in Tectona grandis .

- 134 -

Results and Discussion

5.2.2.4 Summarizing description

In this part, general information about the obtained air dry density mean values, in addition to the values obtained from the inner wood samples (10 %) and outer wood samples (90 %) were provided (see Table 5.29).

Table 5.29: The study species air dry density. Mean ± Std (min.-max.).

Air dry density (g/cm 3) Species General Inner sample Outer sample 0.893 ± 0.079 0.829 ± 0.073 0.945 ± 0.055 Acacia seyal (0.631 – 1.086) (0.631 – 0.970) (0.849 – 1.090) 0.789 ± 0.054 0.762 ± 0.054 0.811 ± 0.051 Balanites aegyptiaca (0.631 – 0.915) (0.631 – 0.876) (0.683 – 0.915)

The air dry density mean values of both species are within the range of tropical hardwood species provided previously in Table 2.12 and 2.13 in section 2.1.2.1. Based on the Malaysian grading rules for sawn timber (ANONYMOUS 1984), Acacia seyal wood can be classifies as heavy (from 0.800 to 1.120 g/cm 3), while that of Balanites aegyptiaca can be classified as medium (from 0.720 to 0.800 g/cm 3).

The obtained air dry density values of both species are much higher than those reported by OSMAN (2001) of 0.66 g/cm 3 for Acacia seyal as well as Balanites aegyptiaca .

5.2.3 Density achieved by X-ray densitometry

5.2.3.1 Correction factor

In addition to the basic and air dry densities, the density was also measured by the X-ray densitometry technique. Correction factors were calculated by calibrating each value obtained by X-ray technique by its counterpart in air dry density, and then an average was calculated. The calculated factors were 0.874 for Acacia seyal and 0.878 for Balanites aegyptiaca . The obtained factors are comparable with those calculated for some softwood of 0.886 and 0.933 for pine and spruce, respectively (SCHEINGRUBER 1988), and with those of Sessile oak (Quercus petraea ) of 0.912 (GÜNTHER 2012). The factors were used to convert the values obtained by X-ray into equivalent values.

- 135 -

Results and Discussion

5.2.3.2 Radial variation

As previously mentioned in the X-ray densitometry investigation section 4.2.2.3, the DENDRO 2003 records one value for each 20 micrometre distance, however, the measured values were divided into five portions from pith to bark which are: 10 %, 30 %, 50 %, 70 % and 90 % to be comparable with those of basic density.

In this part the within tree variation was discussed in terms of radial variation among the five representative radial positions (10 %, 30 %, 50 %, 70 % and 90 %). Non-significant differences were detected among the five selected radial positions in most of the studied trees in both species (Appendix 9.33 and 9.34). Generally, the wood density followed the increase pattern from pith to bark in Acacia seyal and the reverse trend in Balanites aegyptiaca (Figure 5.26).

Figure 5.26: The study species X-ray density radial variation box plots

The radial decline trend observed in Balanites aegyptiaca may be due to the existence of tyloses, crystals and traumatic resin ducts which are common in this species (as previously mentioned in section 5.1.1.2) and which concentrated more near the pith. The thing which prevent the penetration of light through the measured samples thereby leads to the obtaining of high density values near the pith. Also, Balanites aegyptiaca has an interlocked grain with variable fibre angles. Thus, more than one cross section was cut from each sample according to the different fibre angles. This may lead to error when moving from one sample to the other during density analysis. The above mentioned possible reasons for the decline trend were also

- 136 -

Results and Discussion

suggested by SCHEINGRUBER (1988) as possible sources of errors using the X-ray densitometry method (see section 4.2.2.3).

The radial increasing trend observed in Acacia seyal is in homogeneity with that observed in both basic and air dry densities and with those observed in the literature (Table 5.23). However, the detected non-significant differences among the radial portions in X-ray density for both species are in disagreement with those found in the case of air dry density as well as basic density for both species.

5.2.3.3 Trees, forests and regions variation

The variation in this section was assessed on the basis of each of the five radial portions separately. The results were summarised in Table 5.30 for Acacia seyal and Table 5.31 for Balanites aegyptiaca (see also Appendix 9.35, 9.36 and 9.37, 9.38 for Acacia seyal and Balanites aegyptiaca , respectively). For informations about the variation between the two climatic zones see chapter 5.4.

Significant variation among trees was detected in some cases for both species. No significant differences were observed among forests and between regions in almost all cases in Acacia seyal while some significant differences were observed in Balanites aegyptiaca for both levels (forests and regions) (see Tables 5.30 and 5.31).

Table 5.30: Acacia seyal X-ray density variation (ANOVA/T-test)

Variation sources T Region Forest Tree Forest Region N Zone 10 30 50 70 90 10 30 50 70 90 10 30 50 70 90 % % % % % % % % % % % % % % % Al Obeid Al Ein 3 N S S S S Habila 2 S S S S S Al Al Homora N N N N N

One Duwaim N N N N N Aum Top 3 S S S S S Goz Fagor 3 N N N S N Al Tawla 3 N N S S N N S N N N Damazien Khor Donia 3 S S S S S Al Sheheata 3 N N N S N N N N N N Two Kadugli Al Homara 3 S S S N N N N N N N Al Bardab 3 N S N N N TN = number of trees, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

- 137 -

Results and Discussion

Table 5.31: Balanites aegyptiaca X-ray density variation (ANOVA/T-test)

Variation sources T Region Forest Tree Forest Region N Zone 10 30 50 70 90 10 30 50 70 90 10 30 50 70 90 % % % % % % % % % % % % % % % Al Obeid Al Tainat 3 N N S S S Habila 3 N S N S S Al Al Homora 3 N N N N N N N N N N

One Duwaim S S S S S S S S S S Aum Top 3 Goz Fagor 3 N S N N S Al Tawla 3 N N N S S N N N N N Damazien Khor Donia 3 N N N S S Al Sheheata 3 N S S S S N N N S S Two Kadugli Al Homara 3 N N N N N S S N N N Al Bardab 3 N N S N N TN = number of trees, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

For Acacia seyal , the detected significant differences among trees within forest vs. no significant differences among forests and regions agreed with the statement of ZOBEL and VAN BUIJTENEN (1989) that the tree-to-tree variability within a site is generally considerably greater than it is among different site or geographic regions. This is also in agreement with CHUDNOFF (1961) how concluded that there was more variability among trees at the same site than between sites in Eucalyptus camaldulensis . Balanites aegyptiaca seemed to have more heterogeneity behaviour than Acacia seyal in case of forests and region variation.

5.2.3.4 Summarizing description

Table 5.32 provides general information about the estimated X-ray density mean values in addition to the values obtained from the inner wood samples (10 %) and outer wood samples (90 %). The X-ray density values are comparable with those of air dry density obtained in section 5.2.2.3.

Table 5.32: The study species X-ray density. Mean ± Std (min.-max.).

X-ray density (g/cm 3) Species General Inner sample Outer sample 0.901 ± 0.095 0.878 ± 0.093 0.953 ± 0.087 Acacia seyal (0.701 – 1.159) (0.701 – 1.137) (0.763 – 1.148) 0.801 ± 0.088 0.829 ± 0.100 0.788 ± 0.82 Balanites aegyptiaca (0.606 – 1.162) (0.607 – 1.162) (0.621 – 1.128)

- 138 -

Results and Discussion

5.2.4 Comparison of density values derived from X-ray technique and air dry gravimetric method

The variation between the values obtained by X-ray densitometry technique and by air dry gravimetric method were investigated using T-test depending on the mean values as well as the values for each portion separately. The results reveal no significant differences between the values obtained by the two methods in all cases, except that of the first portion (10 %) of Balanites aegyptiaca (see Figures 5.27–5.29).

In general, higher values were observed in density assessed by X-ray densitometry method for both species (Figure 5.27). This result is in accordance with GÜNTHER (2012), who observed also higher values in density achieved by X-ray than in air dry density in Sessile oak (Quercus petraea ). This may be explained by the sensitivity of the X-ray technique to any component which may exist in the wood cells that may prevent the penetration of the light through the radio graphed sample.

The existence of tyloses and crystals is common in both species, as previously mentioned in the anatomical composition descriptions provided in sections 5.1.1.1 and 5.1.1.2, in addition to the existence of gummy materials in Acacia seyal and the traumatic resin ducts in Balanites aegyptiaca . The existence of all the previously mentioned materials may lead to the obtaining of higher density values for X- ray density vs. air dry density.

Figure 5.27 : The study species air dry density vs. X-ray density box plots (means with * are significantly different from each other at 0.05 probability level)

- 139 -

Results and Discussion

Figure 5.28 : Acacia seyal air dry density vs. X-ray density box plots (means with * are significantly different from each other at 0.05 probability level)

Figure 5.29: Balanites aegyptiaca air dry density vs. X-ray density box plots (means with * are significantly different from each other at 0.05 probability level)

From the result of the present study it is obvious that the X-ray technique may be considered as a valid tool in wood density determination with some reservation in its use as a tool for the determination of the radial trend in wood density. Similarly, others authors successfully used the X-ray densitometry for hardwoods density analysis; For instance, ESPINOZA (2004) in Gmelina arborea , MONTES et al. (2007) in Calycophyllum spruceanum , RIOG et al. (2008)

- 140 -

Results and Discussion

in Poplar ( Populus x Canadensis and Populus deltoides ) as well as GÜNTHER (2012) in Sessile oak (Quercus petraea ).

In order to achieve proper results using the X-ray technique, it is recommended to extract the gummy materials or any other materials from the wood.

5.3 Hardness strength

5.3.1 Vertical variation

The vertical variation is discussed according to the values of the inner wood measured points (near the pith) and the outer wood measured points (near the bark) separately, in the case of the hardness strength in transverse section. The vertical variations in the case of hardness strength of the radial section are discussed according to the mean values of the two measured points (in the middle of the distance from pith to bark).

The results found no significant variation between the two discs for most tested trees of both species for hardness strength in transverse and radial section (see Appendix 9.39 and 9.40). These results are in harmony with the non-significant differences observed in both air dry and basic densities. The results were in agreement with CHUDNOFF (1961), who concluded that the mechanical wood properties do not vary at different height in Eucalyptus camaldulensis .

Hardness strength vertical variation showed a similar pattern as those observed in basic and air dry densities in both transverse and radial sections for both species, with a decreasing trend appearing in Acacia seyal and a constant trend in Balanites aegyptiaca (Figures 5.30 and 5.31). This result was expected, because the mechanical properties of wood are closely correlated to density, as previously cited in the literature (DAVIS 1961, BAREFOOT et al. 1970, LEWARK 1979, HAYGREEN and BOWYER 1996, BARNETT and JERONIMIDIS 2003, BOWYER et al. 2003 and KIAEI and SAMARIHA 2011).

- 141 -

Results and Discussion

Figure 5.30: The study species hardness strength vertical variation box plots in transverse section. (P1 = inner measured points and P2 = outer measured points)

Figure 5.31: The study species hardness strength vertical variation box plots in radial section

5.3.2 Radial variation

Radial variation was tested for hardness strength in transverse section according to each of the inner and outer measured points separately. No radial variation test was conducted in the case of hardness strength of the radial section, where the hardness strength was measured at two points in the middle length of the radius.

- 142 -

Results and Discussion

The results show significant differences between the values of the inner and the outer measured points in some of the studied trees for both species (see Appendix 9.41 and 9.42). Both species exhibit an increasing trend in hardness strength from the inner to the outer wood (Figure 5.32). This result is also thought to be due to the radial increasing trend observed in air dry density as well as basic density. The radial increase in hardness strength with the increase in wood density enhances the positive linear relationship between wood density and mechanical properties reported by several researchers (SHEPARD and SHOTTAFER 1992, ZHANG 1995 and IZEKOR et al. 2010).

Figure 5.32: The study species hardness strength radial variation box plots in transverse section

5.3.3 Trees, forests and regions variation Due to significant differences observed between the inner and outer measured points in the previous section, the variations in the three levels (trees, forests and regions) were assessed for each of the two measured points separately. Expected significant variations were detected in some cases for all levels of both species (see Table 5.33 and 5.34 and Appendix 9.43 and 9.44). These observed variations were thought to be due to the significant variations observed in all the measured wood densities. This confirms the conclusion reported by SCHNIEWIND (1989) that much of the variation in wood strength, both between and within species, can be attributed to differences in wood density. In similarity with the current study results, BHAT and PRIYA (2004) found significant differences in the mechanical properties of Tectona grandis according to location.

- 143 -

Results and Discussion

Table 5.33: Acacia seyal hardness strength variation (ANOVA/T-test) Transverse section Radial section

T Variation sources Variation sources Region Forest N Region Zone Tree Forest Tree Forest Region P1 P2 P1 P2 P1 P2 Al Obeid Al Ein 3 N N S Habila 2 N N N Al Al Homora N N N

One Duwaim N N S Aum Top 3 S N N Goz Fagor 3 S S S Al Tawla 3 N S N N N N Damazien Khor Donia 3 S N N S Al Sheheata 3 S N S S N Two Kadugli Al Homara 3 N N S N N N Al Bardab 3 S S N TN = number of trees, P1 = inner measured points, P2 = outer measured points, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

Table 5.34: Balanites aegyptiaca hardness strength variation (ANOVA/T-test) Transverse section Radial section

T Variation sources Variation sources Region Forest N Region Zone Tree Forest Tree Forest Region P1 P2 P1 P2 P1 P2 Al Obeid Al Tainat 3 N N N Habila 3 S S N Al Al Homora 3 N N N N S S

One Duwaim S S S N N N Aum Top 3 Goz Fagor 3 N N N Al Tawla 3 S N S N S S Damazien Khor Donia 3 N N N S Al Sheheata 3 N S S S N Two Kadugli Al Homara 3 S N S S N S Al Bardab 3 N N N TN = number of trees, P1 = inner measured points, P2 = outer measured points, S = significant difference, N = non significant difference at 0.05 probability level, X = no available statistics.

For informations about the variation between the two climatic zones see chapter 5.4.

5.3.4 Summarizing description

Table 5.35 provides information about the hardness strength mean values (calculated from all measured radial points) in addition to the mean values for the inner and outer measured points separately in the case of hardness strength in transverse section. No values are provided for the inner and outer points in the case of the radial section hardness strength, where the hardness was measured in the middle of the radius.

- 144 -

Results and Discussion

Table 5.35: The study species hardness strength (N/mm 2). Mean ± Std (min.-max.). Transverse section Species Radial section General Inner points* Outer points* 84.03 ± 13.63 78.16 ± 12.08 90.06 ± 12.93 51.18 ± 9.63 Acacia seyal (56.51 - 133.61) (56.52 - 113.64) (67.15 - 133.61) (32.44 - 76.26) Balanites 86.67 ± 13.89 78.28 ± 12.21 90.74 ± 14.10 45.53 ± 6.48 aegyptiaca (50.73 - 125.05) (50.73 - 110.18) (59.51 - 125.05) (30.67 - 68.84) * the corresponding air dry density mean values (g/cm 3) are as follows: for Acacia seyal inner and outer wood are 0.829 and 0.945, respectively, and for Balanites aegyptiaca inner and outer wood are 0.762, and 0.811, respectively.

According to the Brinell hardness strength classifications provided in Table 2.14 in section 2.1.2.2, both species’ wood can be classified as very hard.

Research has shown that higher density species tend to have stronger timber than lower density species (TSEHAYE et al. 1995, WALKER and BUTTERFIELD 1996). This is in harmony with the present study results in the case of the hardness strength in the radial section where Acacia seyal has higher values of hardness strength than Balanites aegyptiaca . However, the opposite was happened in the transverse section where Acacia seya l recorded slightly lower hardness strength values than Balanites aegyptiaca . This may be explained by the existence of banded parenchyma in Acacia seyal (previously mentioned in section 5.1.1.1), which may negatively affect the hardness in the transverse section. As a confirmation of the possible effect of the presence of parenchyma bands on hardness strength, WAGENFÜHR (2007) reported that Millettia laurentii (Wengé), with wide banded parenchyma, has a wood air dry density of 0.860 g/cm 3 and hardness strength (HB) in transverse section of 39 N/mm 2. On the other hand, the author reported an air dry density value of 0.690 g/cm 3 and a hardness strength (HB) range of 50–60 N/mm 2 for Quercus petraea (Sessile oak). Thus, in this case species with the higher density (Wengé) has the lower hardness strength. The reported density values for Wengé are comparable to that of Acacia seyal (0.893 g/cm 3) and higher than that of Balanites aegyptiaca (0.789 g/cm 3). However, its hardness strength in transverse section is much lower than those of Acacia seyal and Balanites aegyptiaca .

Figure 5.33 represents the relationship between the Brinell hardness strength and the corresponding air dry density.

- 145 -

Results and Discussion

Figure 5.33: The relationship between the study species Brinell hardness strength and air dry density

From Figure 5.33, it is obvious that the Brinell hardness strength values increase with increasing density for both species. Similarly, a compatible relationship between basic specific gravity and Janka hardness (side hardness) for softwoods and temperate and tropical hardwoods has been identified by WIEMANN and GREEN (2007) (see Figure 2.3 in section 2.1.2.2). In addition, MÖRATH (1932), also found a positive relationship between specific gravity and both end and side Brinell hardness in hardwood and softwood (see Figure 2.4 in section 2.1.2.2). YLINEN (1943) showed that for practically the whole range of densities of commercial timbers a linear relationship between Brinell hardness and oven dry density is applicable.

5.4 Variation of the wood properties due to climatic zones

This part is divided into three sections according to the tested wood properties: anatomical properties variation, density variation and hardness strength variation.

5.4.1 Anatomical properties variation

In this section the variation between zone one (with a mean of 273 mm rainfall annually) and zone two (with a mean of 701 mm rainfall annually) is discussed according to the variation in each of the two measured samples (10 % and 90 % distance from the pith) separately, with the

- 146 -

Results and Discussion

exception of fibre length where the variation is discussed according to the variation in each of the five selected portions (10 %, 30 %, 50 %, 70 % and 90 %). General information about the effect of the rainfall zones on the overall mean values was also provided in this section in order to give a general idea about the expected mean values in each zone. However, it is not discussed.

5.4.1.1 Variation of fibres characteristics

In Acacia seyal , significant differences were found in fibre length between zones in portions 50 % and 70 %, diameter and wall thickness in the inner wood samples (10 %). However, no significant differences were found in the case of fibre lumen diameter, flexibility coefficient or Runkel ratio. Regarding Balanites aegyptiaca , no significant differences were found between zones in all the studies fibre characteristics (see Table 5.36 - 5.38). This confirms the conclusion of NOSHIRO and SUZUKI (1995) that the response patterns to the same environment can differ among species.

Table 5.36: Results of independent sample T-test for the study species fibre length

Radial Mean (mm) Mean Std. error Species position df differences differences (%) Zone1 Zone2

10 1.07 1.08 1575 - 0.01 0.01 30 1.25 1.26 1584 - 0.01 0.01

Acacia 50 1.27* 1.32* 1657 - 0.05 0.01 seyal 70 1.32* 1.35* 1493 - 0.03 0.01 90 1.35 1.36 1258 - 0.01 0.01

Mean 1.25* 1.27* 7575 - 0.02 0.00

10 1.00 1.00 1548 - 0.00 0.01 30 1.14 1.13 1339 0.01 0.01 Balanites 50 1.20 1.19 1489 0.00 0.01 aegyptiaca 70 1.22 1.22 1333 0.00 0.01

90 1.24 1.23 980 0.01 0.01 Mean 1.15 1.15 6697 0.00 0.00 Means with * in the same row are significantly different from each other at 0.05 probability level.

- 147 -

Results and Discussion

Table 5.37: Results of independent sample T-test for Acacia seyal fibres characteristics Mean values Mean Std. error SN Property df Zone1 Zone2 differences differences Diameter (µm) 10.08* 10.64* 794 -0.56 0.17 Lumen diameter (µm) 3.72 3.88 774 -0.16 0.10 1 Wall thickness (µm) 3.11* 3.35* 785 -0.24 0.05 Flexibility Coefficient (%) 37.36 36.84 784 0.42 0.05 Runkel ratio 1.75 1.75 768 0.04 0.59 Diameter (µm) 10.94 11.16 786 -0.22 0.19 Lumen diameter (µm) 3.56 3.67 790 -0.11 0.10 2 Wall thickness (µm) 3.65 3.73 787 -0.09 0.06 Flexibility Coefficient (%) 31.62 31.94 789 -0.23 0.50 Runkel ratio 2.24 2.20 758 -0.01 0.05 Diameter (µm) 10.51* 10.90* 1582 - 0.40 0.13 Mean Lumen diameter (µm) 3.64 3.77 1566 - 0.13 0.07 of 1 Wall thickness (µm) 3.38* 3.54* 1574 - 0.16 0.04 and 2 Flexibility Coefficient (%) 34.50 34.37 1575 - 0.12 0.40 Runkel ratio 2.00 1.97 1515 - 0.03 0.04 SN = sample number (1 taken from the inner wood samples and 2 taken from the outer samples), means with * in the same row for each property are significantly different from each other at 0.05 probability level.

Table 5.38: Results of independent sample T-test for Balanites aegyptiaca fibres characteristics Mean values Mean Std. error SN Property df Zone1 Zone2 differences differences Diameter (µm) 12.39 12.67 775 - 0.29 0.18 Lumen diameter (µm) 4.83 5.00 756 - 0.17 0.11 1 Wall thickness (µm) 3.66 3.74 782 - 0.08 0.04 Flexibility Coefficient (%) 39.25 39.99 758 - 0.74 0.58 Runkel ratio 1.58 1.54 772 0.04 0.06 Diameter (µm) 13.65 13.33 767 0.31 0.18 Lumen diameter (µm) 5.72 5.56 764 0.16 0.12 2 Wall thickness (µm) 3.91 3.91 775 0.00 0.06 Flexibility Coefficient (%) 42.58 42.45 786 0.13 0.60 Runkel ratio 1.39 1.39 751 0.00 0.03 Diameter (µm) 13.02 13.00 1544 0.02 0.13 Mean Lumen diameter (µm) 5.27 5.29 1522 - 0.01 0.08 of 1 Wall thickness (µm) 3.78 3.83 1559 - 0.04 0.04 and 2 Flexibility Coefficient (%) 40.93 41.25 1546 - 0.32 0.43 Runkel ratio 1.49 1.47 1525 0.02 0.03 SN = sample number (1 taken from the inner wood samples and 2 taken from the outer wood samples), means with * in the same row for each property are significantly different from each other at 0.05 probability level.

Acacia seyal fibre length, diameter and wall thickness were decreased in the drier zone. The reduction of fibre dimensions in the drier zone may be due to the limited water, which is considered the most important factor in cell enlargement (KRAMER 1963). This result is in

- 148 -

Results and Discussion

agreement with those of OGBONNAYA et al. (1992 and 1997) who reported adversely influence of water stress in wood fibre dimensions of Kenaf ( Hibiscus cannabinus L.) and Gmelina arborea , respectively. Similarly, MUKHTAR (2008) in a study of the variation in wood anatomy of Acacia senegal grown in different rainfall zones in Sudan, found significantly lower values of fibre diameter, lumen diameter and wall thickness in the rainfall zone of < 300 mm annually than in the rainfall zone of 300–500 mm annually. AL- KHALIFAH et al. (2006) investigated the impact of water stress on the sap wood of Callgonum comosun . Their results were in accordance with Acacia seyal fibre diameter reduction in the drier zone. However, they disagreed with the current study’s results for fibre length where the authors found that water stress leads to longer fibres. Other authors such as MOYA and FO (2008) found in Gmelina arborea that precipitation has a significant effect on fibre diameter and lumen diameter: precipitation increases fibre diameter and lumen diameter, while no effect was found on fibre length and fibre wall thickness. Contrary to the reduction of Acacia seyal fibre wall thickness detected in the inner wood of the drier zone, some authors found thicker wall fibres in drier environments (FAHN et al. 1986 and ALVES and ANGYALOSSY-ALFONSO 2002). Some authors did not find any significant effect for water stress in fibre wall thickness (NKAA et al. 2007 and MOYA and FO 2008), which is in similarity with the non-significant differences detected in the outer wood of Acacia seyal and both of the inner and outer wood of Balanites aegyptiaca .

The detected significant differences in portions 50 % and 70 % in fibre length may lead also to expect differences in fibre diameter, wall thickness, flexibility coefficient and Runkel ratio in the same portions, which were not included in the present study.

From the above mentioned results it is obvious that Acacia seyal is more sensitive to the effect of rainfall zones than Balanites aegyptiaca with regard to fibre dimensions.

5.4.1.2 Variation of vessels characteristics

According to Table 5.39, significant differences were found between zones in the vessel wall thickness of Acacia seyal in both inner and outer wood samples. The higher values were detected in the drier zone. Vessel diameter and lumen diameter were not significantly affected by rainfall zones. Balanites aegyptiaca vessel dimensions’ response to the effect of rainfall zone was totally different from that of Acacia seyal . As shown in Table 5.40, significantly higher mean values of vessel diameter, lumen diameter as well as wall thickness were

- 149 -

Results and Discussion

detected in the drier zones for the inner wood, while the opposite was detected in the outer wood.

Table 5.39: Results of independent sample T-test for Acacia seyal vessels characteristics Mean values Mean Std. error SN Property (µm) df differences differences Zone1 Zone2 Diameter 113.82 113.45 589 0.37 3.02 1 Lumen diameter 96.06 95.69 594 0.37 2.88 Wall thickness 8.75* 8.40* 579 0.35 0.18 Diameter 133.58 127.93 587 5.65 3.29 2 Lumen diameter 112.18 109.75 595 2.42 3.23 Wall thickness 9.98* 9.40* 575 0.58 0.20 Mean Diameter 123.67 120.69 1178 2.98 2.29 of 1 Lumen diameter 104.13 102.72 1191 1.41 2.21 and 2 Wall thickness 9.37* 8.90* 1156 0.47 0.14 SN = sample number (1 taken from the inner wood samples and 2 taken from the outer wood samples means with * in the same row for each property are significantly different from each other at 0.05 probability level.

Table 5.40: Results of independent sample T-test for Balanites aegyptiaca vessels characteristics Mean values Mean Std. error SN Property (µm) df differences differences Zone1 Zone2 Diameter 112.05* 97.91* 586 14.14 2.59 1 Lumen diameter 92.82* 80.78* 591 12.04 2.40 Wall thickness 9.36* 8.39* 584 0.97 0.18 Diameter 139.56* 151.42* 574 -11.87 3.15 2 Lumen diameter 115.04* 124.70* 576 - 9.65 2.79 Wall thickness 12.09* 12.88* 580 - 0.80 0.31 Mean Diameter 125.73 124.25 1162 1.49 2.38 of 1 Lumen diameter 103.88 102.28 1169 1.59 2.10 and 2 Wall thickness 10.71 10.64 1166 0.07 0.21 SN = sample number (1 taken from the inner wood samples and 2 taken from the outer wood samples), means with * in the same row for each property are significantly different from each other at 0.05 probability level.

SASS and ECKSTEIN (1995), in their study of the variability of vessel size in Fagus sylvatica reported that the formation of vessels during the beginning of the cambial activity was controlled by internal factors (not specified), while adult wood formation was affected by external ones (e.g. rainfall). Following this, the higher values of vessel dimensions detected in the inner part of the drier zone in Balanites aegyptiaca may be explained by differences in the genetics of the trees collected from the drier zone in comparison with those of the more humid

- 150 -

Results and Discussion

zone. The more realistic explanation is that the trees from the drier zone may be older that those of the humid zone while having more or less the same diameters due to a slower growth rate (due to the lack of water). On the other hand, the lower vessel dimensions values detected in the outer wood of the drier zone are thought to be adaptive mechanisms to tolerate drought stress.

The decrease of the diameter of Balanites aegyptiaca outer wood vessels in the drier zone is in accordance with the findings of several studies (CARLQUIST 1975, BAAS and SCHWEINGRUBER 1987, VAN DER WALT et al. 1988, ZHANG et al. 1988, WILKINS and PAPASSOTIRIOU 1989, FEBRUARY 1993 and LENS et al. 2004, SCHUME et al. 2004, AL-KHALIFAH et al. 2006) in which decreases of vessel diameters were reported to correspond to decreasing water availability. In homogeneity with the non-significant differences detected in vessel diameter of Acacia seyal , MOYA and FO (2008) confirmed that vessel diameter is not affected by water availability in Gmelina arborea . AL-KHALIFAH et al. (2006) found that the drought makes thicker vessel walls on the sap wood of Callgonum comosun as was detected in Acacia seyal in the present study. However, the results of AL- KHALIFAH et al. (2006) as well as MOYA and FO (2008) are in disagreement with those of the present study for Balanites aeyptiaca , where the vessel dimensions were significantly decreased in the drier zone.

In similarity with the differences between the study species in their response pattern to water stress, FEBRUARY et al. (1995) reported that anatomical responses to water treatment differed between taxa. They found that vessel diameter and vessel element length in Eucalyptus grandis and E. grandis x camaldulensis were positively correlated with increases in water used. However, E. grandis x nitens vessel diameter and element length were not affected by water availability. In contrast to fibre dimension, the conductive systems of Balanites aegyptiaca (vessels) seemed to have a higher adaptive efficiency than that of Acacia seyal , where only the wall thickness was affected by rainfall zone. This also confirms that the response patterns to the same environment can differ among species, as concluded by FEBRUARY et al. (1995) and NOSHIRO and SUZUKI (1995).

Figures 34 and 35 represent the effect of rainfall on the wood anatomical properties of Acacia seyal and Balanites aegyptiaca respectively. The figures depend upon the effect of rainfall on the outer wood samples which considered as mature wood.

- 151 -

Results and Discussion

Figure 5.34: The effect of rainfall on Acacia seyal anatomical properties

Figure 5.35: The effect of rainfall on Balanites aegyptiaca anatomical properties

Depending upon figures 5.34 and 5.35, it is evident that the study species followed different mechanisms to adapt with the rainfall change between the dry and humid zones.

- 152 -

Results and Discussion

5.4.2 Basic density variation The variation in wood basic density was discussed according to the variation in each of the selected five radial portions as well as the mean of all portions. The study found significant differences between climatic zones in wood basic density in all the selected portions except the last one (90 %) for both species. Significant differences were also detected in the mean values of all portions for both species. In all cases the higher density was detected in the drier zone (Table 5.41).

Table 5.41: Results of independent sample T-test for the study species wood basic density Radial 3 Mean (kg/m ) Mean Std. error Species Position df differences differences (%) Zone1 Zone2 10 695.21* 654.68* 128 40.52 9.54 30 715.37* 680.96* 174 34.41 7.94

Acacia 50 743.27* 717.75* 183 25.52 7.29 seyal 70 763.26* 746.06* 177 17.19 6.93

90 784.23 770.54 116 13.69 8.73 Mean 740.10* 713.15* 786 26.95 4.38 10 652.12* 615.60* 136 36.52 7.45 30 658.40* 631.40* 196 27.00 5.38 Balanites 50 673.91* 650.73* 196 23.18 5.28 aegyptiaca 70 682.87* 665.58* 203 17.29 5.21 90 687.64 676.13 158 11.51 6.32 Mean 671.46* 649.21* 897 22.26 2.84 Means with * in the same row are significantly different from each other at 0.05 probability level.

Similar to the results of the current study, several authors found that water stress positively affects the wood density in their studied species; for instance, BEADLE et al. (2001) in Eucalyptus nitens , WIMMER et al. (2002) in Eucalyptus globulus Labill, ROQUE (2004) in Gmelina arborea , AL-KHALIFAH et al . (2006) in Callgonum comosun and NAIDOO et al. (2006) in Eucalyptus grandis . In contrast to the results of the current study, MACFARLANE and ADAMS (1998) found no relationship between wood density and water stress in Eucalyptus globulus .

In Balanites aegyptiaca the increase of wood density in the drier zone may be explained by the reduction of vessels dimensions in the same zone, particularly, the reduction of vessel diameter. However, in Acacia seyal , the increase in wood density in the drier zone vs. the reduction of fibre length and the absence of change in the other fibre dimension (mature wood) as well as in vessel diameter and lumen diameter in the same zone lead to expect significant differences in the cell portions according to rainfall zones. Also, the increase of

- 153 -

Results and Discussion

wood density of Acacia seyal in the drier zone may be derived from the increase of the extractives materials in the same zone. As a confirmation of this possible explanation, SEARSON et al. (2004) found a significant increase of wood density due to water stress in both Eucalyptus grandis and E. sideroxylon seedlings, while no effect was found in E. occidentalis . However, when ethanol and water-soluble extractives were removed from the wood, there was no difference in wood density between the well-watered and water-limited E. sideroxylon seedlings. This means that the higher density of E. sideroxylon detected in the drier zone (before the extraction) was due to the increase of the extractives in the same zone, while the increase of density in the drier zone detected in E. grandis was due to the significant change in wood anatomical structure detected in the same zone.

5.4.3 Hardness strength variation

In this part the variation between the rainfall zones was discussed according to the variation in each of the inner and outer measured points separately in the case of hardness strength in transverse section and according to mean values of the two measured points (in the middle of the distance from pith to bark) in the case of hardness strength in radial section.

Table 5.42: Independent sample T-test for the study species hardness strength (HB)

2 Mean (N/mm ) Mean Std. error Species SN df Zone1 Zone2 differences differences inner 81.47* 75.51* 106 5.97 2.28 Acacia Tran. outer 92.22 88.34 106 3.88 2.49 seyal Mean 86.29* 82.12* 254 4.17 1.69 Radial 53.10* 49.64* 214 3.46 1.30

inner 82.59* 73.97* 116 8.63 2.11 Balanites Tran. outer 92.42 89.07 116 3.35 2.59 aegyptiaca Mean 88.69* 84.64* 340 4.05 1.49 Radial 45. 84 44.63 236 1.23 0.84

Tran. = transverse section , Means with * in the same row are significantly different from each other at 0.05 probability level.

Significant differences were detected for hardness strength in the transverse section of the inner wood measured points for both species, as well as the hardness strength of the radial section in Acacia seyal , while non-significant differences were detected in transverse section hardness strength in the outer wood measured points for both species as well as radial section hardness strength in Balanites aegyptiaca (Table 5.42). As in wood density, the higher values were detected in the drier zone. These results were expected because the mechanical

- 154 -

Results and Discussion

properties of wood are closely correlated to density, as previously cited in literature (HAYGREEN and BOWYER 1996, BARNETT and JERONIMIDIS 2003, BOWYER et al. 2003 and KIAEI and SAMARIHA 2011). Therefore, the effect of rainfall zone is expected to be similar in wood density and hardness strength. This may explain the harmony in the effect of rainfall zone between wood density and hardness strength observed in the present study.

The non-significant differences of the outer wood measured points detected in hardness strength of the transverse section are in homogeneity with the non-significant differences detected in the outer samples (90 %) in wood density in both species (previously mentioned in section 5.4.2). Therefore, significant differences are expected in the middle portions, which are not included in the present study.

5.4.4 Summarizing description

Finally, the effect of water stress of the drier zone on all the studied wood properties is summarized in Table 5.43. The table provides information about the effect of water stress on each of the measured portions separately. However, it is recommended not to take into account the effect in the inner wood (juvenile wood), as it controlled by internal factors as reported in the literature (SASS and ECKSTEIN 1995).

Table 5.43: Summary of the effect of water stress on the studied wood properties Acacia seyal Balanites aegyptiaca Property Radial portions (%) Radial portions (%) 10 10 30 50 70 90 10 30 50 70 90 Fibre length 0 0 - - 0 0 0 0 0 0 Fibre diameter - 0 0 0 Fibre lumen diameter 0 0 0 0 Fibre wall thickness - 0 0 0 Flexibility coefficient 0 0 0 0 Runkel ratio 0 0 0 0 Vessels diameter 0 0 + - Vessels lumen diameter 0 0 + - Vessels wall thickness + + + - Hardness strength (transverse) + 0 + 0 Hardness strength (radial) + 0 Basic density + + + + 0 + + + + 0

Excluded portions, (+) positive effect, (-) negative effect and (0) no effect

According to Table 5.43, and according to the effect of water stress on the outer samples wood properties (considered as mature wood), the following conclusions can be summarized:

- Acacia seyal fibre length is negatively affected by water stress.

- 155 -

Results and Discussion

- In both species, fibre characteristics (except fibre length in Acacia seyal ) are not affected by water stress.

- Acacia seyal vessel wall thickness is positively affected by water stress.

- Balanites aegyptiaca vessel dimensions are negatively affected by water stress.

- Balanites aegyptiaca conductive system (vessels) adapt more efficiently to water stress than that of Acacia seyal .

- Vessel dimensions appeared to be more sensitive to water stress than fibre dimensions, in particular in Balanites aegyptiaca .

- Wood basic density of both species is positively affected by water stress.

- Hardness strength (HB) of both species is not affected, with the exception of hardness strength in transversal section for Acacia seyal , which is positively affected by water stress.

Generally, from results displayed in Table 5.39, it is obvious that portion 90 % is not a good detector of the possible variations due to water stress, particularly in the case of Acacia seyal , where no changes were detected in this portion in almost all the studied wood properties.

5.5 The suitability of the study species for pulp and paper, and flooring industries

This section assesses the suitability of the study species for pulp and paper, and flooring industries. For this purpose, a comparison between the study species wood properties with the acceptable values for pulp and paper making (PPM), and with hardwood species in commercial use for PPM and flooring industry was conducted. The possible effect of rainfall zones on the suitability of the study species for the PPM and flooring industries is also discussed briefly in this section.

5.5.1 Suitability for pulp and paper industry

Basic density and fibre characteristics appeared to exert considerable influence on paper properties. They are considered as a good predictor of the suitability of the wood raw material for pulp and paper making (IGARTÚA et al. 2003 and MONTEOLIVA et al. 2005). Therefore, the suitability of the study species for pulp and paper making (PPM) is assessed in this part by comparing the study species’ wood basic density and fibre characteristics with

- 156 -

Results and Discussion

those of the acceptable values for PPM and with Gmelina arborea , which is used commercially in paper production. Table 5.44 provides information about the study species wood properties values, the acceptable values for PPM as well as the wood properties values of Gmelina arborea .

Table 5.44: The study species wood properties compared with the acceptable values for PPM and with the reference species wood properties. Mean (min.- max.), Slenderness values doesn’t include min., max. because it is calculated from the mean values of fibres length and diameter. Reference Study species Acceptable species Property values for 6 Balanites Gmelina Acacia seyal PPM aegyptiaca arborea Basic density 724.68 659.49 350-650 1 510 (kg/m³) (550 - 960) (596 - 790) 1.26 1.15 Fibre length (mm) ≥ 12 1.28 (0.60 - 1.90) (0.60 - 1.60)

Flexibility 34.44 41.09 3 > 60 73 coefficient (%) (14.29 - 62.86) (19.23 - 67.66)

1.99 1.48 Runkel ratio 0.25 - 1.5 4 0.39 (0.58 - 4.83) (0.42 - 3. 33) 5 Slenderness ratio 117.76 88.39 > 33 50 1CASEY (1980) 2SOURCE 4 3PETRI (1952), OKEREKE (1962), RYDHOLM (1965) 4VALKOMER (1969) 5XU et al. (2006) 6OGUNKUNLE and OLADELE (2008)

According to Table 5.44, Acacia seyal ’s density is slightly above the range for commercial temperate pulpwood of 350–650 kg/m³ and that of Balanites aegyptiaca is almost within the range. The yield of pulp per unit volume is directly related to the wood basic density. Thus denser wood provides more volume/weight in the time unit. A high basic density is desirable economically as it means that the digester capacity required is lower. On the other hand, overly high density may inhibit impregnation and penetration with cooking liquor. Furthermore, too high density may also have a negative effect on the paper properties (KHRISTOVA et al. 2006). Fibre length mean values of both species are higher than the acceptable values for PPM of hardwood species (Table 5.44). Therefore, they are considered compatable for PPM. Moreover, several authors confirmed the suitability of hardwood species with fibre lengths equal and even shorter than the studied species for PPM. Good examples are the species studied by KHRISTOVA et al. (1997), KHRISTOVA et al. (1998), KHRISTOVA and KARAR (1999), KHRISTOVA et al. (2004) and that of KHIDER et al. (2012). For more information see Table 2.7 in section 2.1.1.1.

- 157 -

Results and Discussion

The study species’ fibres were of good slenderness ratios, much higher than the acceptable value for papermaking of > 33 (XU et al. 2006). They are much higher than almost all hardwoods used in papermaking and also bigger than some softwood species such as Pinus kesiya (56.51), as cited in DUTT and TYAGI (2011). This enhances their suitability for PPM. Slenderness ratio is an important factor in PPM having a positive effect on strength, tear, burst, breaking off, double folding resistance (AKGÜL and TOZLUĞLU 2009). According to the literature, the higher the slenderness ratio, the stronger is the resistance to tearing (RYDHOLM 1965). Therefore, the study species woods are expected to produce paper with high tear resistance.

Runkel ratio is one of the most important parameters in qualifying the fibre for the paper industry. Balanites aegyptiaca ’s Runkel ratio estimated in the current study as 1.48, which is at the upper end of the acceptable range of papermaking of 0.25–1.5 (VALKOMER 1969), while that of Acacia seyal (1.99) is out of the range. It is known from the literature that high Runkel ratio gives low paper strength properties (BEKTAS et al. 1999) and would produce porous paper (IWENOFU 1979). However, KHRISTOVA et al. (1997) and KHRISTOVA et al. (1998) confirmed the suitability of species with higher Runkel ratio for papermaking (see Table 2.7 in section 2.1.1.1). Mixing of the study species with other wood with a low Runkel ratio is recommended, particularly in the case of Acacia seyal .

The flexibility coefficients of both species are lower than the acceptable value for papermaking of >60 (PETRI 1952, OKEREKE 1962, RYDHOLM 1965), but are comparable to those of the species studied by KHRISTOVA et al. (1997), KHRISTOVA et al. (1998) and KHRISTOVA and KARAR (1999) (Table 2.7 in section 2.1.1.1). The study species’ fibres belong to the rigid fibre group (based on their flexibility coefficients) according to the classification of BEKTAS et al. (1999). Low flexibility coefficients negatively affect paper strength properties, such as burst, tear and tensile indexes (PETRI 1952, OKEREKE 1962, RYDHOLM 1965).

In comparison with Gmelina arborea wood properties shown in Table 5.40, the study species have comparable fibre lengths, lower flexibility coefficient, higher Runkel and slenderness ratios and higher wood density.

In some studies conducted in Sudan, KHRISTOVA et al. (1998) and KHRISTOVA et al. (2004) confirmed the suitability of using Acacia seyal var . seyal ’s wood for pulp and paper making, while KHRISTOVA et al. (1997) pointed out the suitability of using Balanites

- 158 -

Results and Discussion

aegyptiaca in PPM. However, their studies are focused on young trees of 5–10 years old collected from Blue Nile state in Sudan (see Table 2.7 in section 2.1.1.1). Therefore, the current study results may be considered more representative because the trees are collected from different states in Sudan. Also the trees are mostly older than those used in the studies of KHRISTOVA et al. (1997), KHRISTOVA et al. (1998) and KHRISTOVA et al. (2004).

Vessels also have an effect on PPM. In general, large vessels are not wanted in PPM, and may cause so-called vessel picking or/and star ground defects in the paper surface (FOELKEL 2007) (see Figure 2.2 in section 2.1.1.2). However, vessel diameters of both species are classified as medium, as previously described in section 5.1.3.3, which may reduce the probability of vessel picking or/and star ground occurrence in papers made from the study species wood.

Table 5.45 provides information about the juvenile and mature wood properties of the study species and compares it with the acceptable values for PPM. Flexibility coefficient, Runkel ratio and slenderness ratio of juvenile and mature wood were obtained from mean values of the inner wood samples (taken at 10% distance from pith to bark) and outer wood samples (taken at 90%) respectively, while basic density and fibre length values were obtained from the mean values of portions 10 % and 30 % in the case of juvenile wood and 70 % and 90 % in the case of mature wood.

Table 5.45: The study species juvenile and mature wood properties compared with the acceptable values for PPM. Mean (min.-max.), Slenderness values doesn’t include min., max. because it is calculated from the mean values of fibres length and diameter

Study species Acceptable Property Acacia seyal Balanites aegyptiaca values for

Juvenile Mature Juvenile Mature PPM Basic density 685.50 762.65 639.54 677.09 1 350-650 (kg/m³) (550-867) (644-960) (496-760) (568-790) Fibre length 1.17 1.34 1.07 1.23 ≥ 12 (mm) (0.60-1.86) (0.94-1.90) (0.60-1.54) (0.90-1.60) Flexibility 37.10 31.78 39.62 42.51 > 60 3 coefficient (%) (17.30-62.86) (14.29-57.81) (20.51-67.26) (19.23-67.66) 1.75 2.22 1.56 1.39 Runkel ratio 0.25 - 1.5 4 (0.58-4.07) (0.73-4.83) (0.49-3.33) (0.42-3.04) Slenderness 103.28 122.18 79.81 91.18 > 33 5 ratio

1CASEY (1980) 2SOURCE4 3PETRI (1952), OKEREKE (1962), RYDHOLM (1965) 4VALKOMER (1969) 5XU et al. (2006)

- 159 -

Results and Discussion

Acacia seyal juvenile wood has a lower density and Runkel ratio and higher flexibility coefficient than mature wood (Table 5.45). This enhances its suitability for PPM in comparison with mature wood. Moreover, its fibre lengths and slenderness ratio are still within the acceptable values of PPM. Therefore, Acacia seyal juvenile wood considered more suitable for PPM than mature wood.

In contrast to Acacia seyal , Balanites aegyptiaca mature wood has a lower Runkel ratio and higher flexibility coefficient than juvenile wood. Balanites aegyptiaca juvenile wood density is within the acceptable range for PPM, while that of mature wood is slightly higher than the upper end of the acceptable range for PPM. The slenderness ratios of both juvenile and mature wood of Balanites aegyptiaca are much higher than the acceptable values for PPM (Table 5.45). Therefore, in the case of Balanites aegyptiaca both juvenile and mature wood are considered suitable for PPM. However, mature wood is preferable.

Balanites aegyptiaca trees produce fruits that used for medicinal purposes (see section 2.3.2.3). Therefore, old trees are a better choice for PPM than young trees. Young trees can be used for fruit production until the trees stop producing fruit or the production becomes low, and then it can be used in PPM.

Table 5.46 provides information about the study species wood properties in each zone in comparison with the acceptable values for PPM.

Table 5.46: The study species wood properties in each climatic zone compared with the acceptable values for PPM

Study species Acceptable Property Acacia seyal Balanites aegyptiaca values for Zone1 Zone2 Zone1 Zone2 PPM 740.10 713.15 671.46 649.21 1 Basic density (kg/m³) 350 - 650 (550-960) (550-911) (561-790) (496-772) 1.25 1.27 1.15 1.15 Fibre length (mm) ≥ 12 (0.60-1.90) (0.60-1.90) (0.62-1.60) (0.60-1.50) Flexibility coefficient 34.50 34.37 40.93 41.25 > 60 3 (%) (16.44-61.21) (14.29-62.86) (19.23-67.66) (20.16-67.26) 2.00 1.97 1.49 1.47 Runkel ratio 0.25 - 1.5 4 (0.58-4.40) (0.59-4.83) (0.42-2.91) (0.49-3.33) Slenderness ratio 118.93 116.52 88.33 88.46 > 33 5

1CASEY (1980) 2SOURCE 4 3PETRI (1952), OKEREKE (1962), RYDHOLM (1965) 4VALKOMER (1969) 5XU et al. (2006)

- 160 -

Results and Discussion

As shown in Table 5.46, Acacia seyal trees growing in zone two (more humid) seemed to be more suitable for PPM purpose than those growing in zone one (drier zone) due to their lower wood density and longer fibres. As previously mentioned in sections 5.4.1.1 and 5.4.2, water stress negatively affects fibre length and positively affects that of wood basic density of Acacia seyal . In the case of Balanites aegyptiaca , the lower density has also been detected in zone two with no significant change in any other wood properties. As shown in Table 5.46, Balanites aegyptiaca wood density in zone two is within the acceptable range of PPM while that of zone one is slightly higher than the upper end of the acceptable range for PPM. Therefore, it is also preferable to use trees growing in the more humid zone for PPM.

According to all the above mentioned results, the following conclusions can be drawn:

- In general wood density and fibre characteristics of the study species are compatible for pulp and paper making.

- Young trees are considered more suitable for PPM, in particular in the case of Acacia seyal , due to their lower wood density.

- Trees growing in zone two (more humid) are considered more suitable for PPM for both species, due also to their lower density.

- Balanites aegyptiaca are considered more suitable for PPM than Acacia seyal , due to the lower density and Runkel ratio in addition to the higher flexibility coefficient detected in Balanites aegyptiaca .

- Paper made from the study species is expected to have low strength properties and good tear resistance due to their low flexibility coefficient and high Runkel and slenderness ratios.

- The mixing of those species with other species with low Runkel ratio and high flexibility coefficient is recommended in order to enhance paper strength properties.

5.5.2 Suitability for flooring industry

In the flooring industry, the most important wood properties that determine the suitability of the wood is hardness strength (HAYGREEN and BOWYER 1996). Wood density is acknowledged to affect most mechanical properties (HAYGREEN and BOWYER 1996, BARNETT and JERONIMIDIS 2003, BOWYER et al. 2003, KIAEI and SAMARIHA 2011). Therefore, this section assesses the suitability of the study species for use in the flooring

- 161 -

Results and Discussion

industry by comparing their air dry density and Brinell hardness strength mean values by those of the American’s flooring industry benchmark Quercus rubra L. (Red oak) and those of Milletia laurentii DE wild. (Wengé), which is widely used in the flooring industry in Africa, as reference species.

As in the suitability for the pulp and paper industry, the assessment for the flooring industry was based on the overall wood properties mean values (Table 5.47), juvenile and mature wood mean values separately (Table 5.48) and the wood properties mean values in each zone (Table 5.49).

Table 5.47: The study species air dry density and hardness strength (HB) values compared with the reference species. Mean (min.- max.). Study species References species*

Balanites Milletia Quercus property Acacia seyal aegyptiaca laurentii rubra 0.893 0.789 0.860 0.700 Density (g/cm³) (0.631 - 1.086) (0.631 - 0.915) (0.810 - 0.950) (0.550 - 0.980) Hardness strength (N/mm 2) 84.03 86.67 39 Transverse section 53 - 66 (56.5 - 133.6) (50.7 - 125.1) (30 - 45) 51.18 45.53 24 Radial section 29 - 36 (32.4 - 76.3) (30.7 - 68.8) (21 - 27) *WAGENFÜHR (2007)

Acacia seyal wood is considered heavy while that of Balanites aegyptiaca is considered as medium, and for both species the wood is classified as very hard, as previously mentioned in sections 5.2.2.3 and 5.3.3. The higher the density the harder the wood, which is good for flooring. However, too high density is not preferable because it may lead to a problem in shrinkage and swelling behaviour. Thus, species with too high density are not preferable in the flooring industry.

According to Table 5.47, both species have higher air dry density and hardness strength mean values than the references species. However, the density of both species falls within the range provided for the reference species. Thus, they may be considered suitable for flooring industry. In addition, Balanites aegyptiaca may be considered more suitable for the flooring industry than Acacia seyal due to its lower density in comparison with Acacia seyal .

Table 5.48 provides information about the juvenile and mature wood properties of the study species and compares them with the references species. Juvenile and mature air dry density values were obtained from the mean values of the inner wood samples (at 10 % distance from pith to bark) and outer wood samples (at 90 %), respectively. The values of juvenile and mature wood hardness strength in transverse section were obtained from the inner and outer - 162 -

Results and Discussion

measured points, respectively. No values are provided for juvenile and mature wood hardness strength in the radial section, where the hardness was measured in the middle of the radius. Therefore, the assessment for the flooring industry in this part will depend on the air dry density and hardness strength in transverse section.

Table 5.48: The study species juvenile and mature wood air dry density and hardness strength (HB) in transverse section compared with the reference species

Study species References species * property Acacia seyal Balanites aegyptiaca Milletia Quercus Juvenile Mature Juvenile Mature laurentii rubra Density 0.829 0.945 0.762 0.811 0.820 0.700 (g/cm³) (0.631-0.970) (0.849-1.09) (0.631-0.876) (0.683-0.915) (0.810 - 0.950 ) (0.550-980 ) HB 78.16 90.06 78.28 90.74 39 53 - 66 (N/mm 2) (56.5-113.6) (67.2-133.6) (50.7-110.2) (59.5 - 125.2) (30 - 45) *WAGENFÜHR (2007)

According to the values provided in Table 5.48, juvenile wood of both species has lower density than mature wood. Their density values seem to be more comparable with those of the references species. However, juvenile wood shrinkage more than mature wood (ZOBEL and VAN BUIJTENEN 1989). Thereby, mature wood may considered more suitable for flooring industry.

The possible effect of rainfall zones on the suitability of the study species for the flooring industry were discussed according to the values provided in Table 5.49.

Table 5.49: The study species air dry density and hardness strength (HB) in climatic each zone compared with the reference species Study species References species* property Acacia seyal Balanites aegyptiaca Milletia Quercu Zone1 Zone2 Zone1 Zone2 laurentii s rubra Density 0.909 0.881 0.803 0.775 0.820 0.700 (g/cm³) (0.747-1.064) (0.631-1.086) (0.691-0.915) (0.631-0.887) (0.810 -0.950 ) Hardness strength (N/mm 2) Transverse 86.29 82.12 88.69 84.64 39 53 - 66 section (62.41-129.23) (56.51-133.61) (60.94-121.07) (50.73-125.05) (30 - 45) 53.10 49.64 45.84 44.63 24 Radial section 29 - 36 (34.34-74.30) (32.44-76.26) (32.44-59.51) (30.67-68.84) (21 - 27) *WAGENFÜHR (2007)

Based on Table 5.49, trees growing in zone two (more humid zone) considered more suitable for flooring industry than those growing in zone one (drier zone). The reason is the lower wood density detected in the more humid zone for both species.

- 163 -

Results and Discussion

The following conclusions can be drawn according to all the above mentioned results:

- In general wood density and hardness strength of the study species are compatible for the flooring industry.

- For both species, trees growing in the more humid zone are preferable, due to their lower density.

- Balanites aegyptiaca is considered more suitable for the flooring industry than Acacia seyal , due to the lower density detected in Balanites aegyptiaca .

5.6 Proving hypotheses

Based on the above provided results, the proposed research hypotheses mentioned in Section 1.4 can be discussed and proven as follow:

Hypothesis 1: The study species have anatomical and physical characteristics that qualify them for pulp and paper making either alone or by mixing with other pulping species.

This hypothesis has been assessed by comparing the study species fibres length, fibres derived values and wood basic density by the acceptable values for PPM and by comparing with Gmelina arborea which is in commercial use for PPM.

According to the study finding provided in section 5.5.1 and in comparison with literature, both species have compatible fibres length, fibres derived values and wood density for PPM. However, mixing with other pulping species is preferable in order to improve the properties of paper made from their wood. Thus, the current hypothesis has been proven.

Hypothesis 2: this considers that the study species have physical and mechanical properties that qualify them for use in the flooring industry.

A comparison of the study species air dry density and Brinell hardness strength with those of two references species, namely Milletia laurentii and Quercus rubra has been conducted (see section 5.5.2) in order to assess the suitability of the study species for flooring industry. As a result of the conducted comparison, the study species considered suitable for flooring industry. Therefore, this hypothesis has been accepted.

Hypothesis 3: The different rainfall zones have a significant effect on the anatomical, mechanical and physical characteristics of the selected species.

This hypothesis has been tested by the comparison of the studied wood properties of trees grown under lower and higher precipitation, the results have been summarized previously in

- 164 -

Results and Discussion

Table 5.36. This hypothesis has been proven for some wood properties and rejected for others. Based on the study results provided in section 5.4, and depending upon each of the tested properties separately, the current hypothesis can be discussed as follows:

Anatomical properties:

The current hypothesis has been proven in case of:

- Acacia seyal fibre length, where significant differences have been observed between zones in portions 50 % and 70 %.

- Acacia seyal fibre diameter, fibre and vessel wall thickness of the inner wood (at 10 % radial distance) as well as vessel wall thickness of the outer wood (at 90 % radial distance).

- Balanites aegyptiaca vessel diameter, lumen diameter as well as wall thickness of both inner and outer wood

However, the hypothesis has been rejected in the following cases:

- Acacia seyal fibre length in portions 10 %, 30 % and 90 %.

- Acacia seyal fibre lumen diameter, flexibility coefficient, Runkel ratio, vessel diameter in both inner and outer wood samples.

- Acacia seyal fibre diameter and wall thickness of the outer wood samples.

- Balanites aegyptiaca fibres dimensions of the entire studied portions.

Basic density:

The hypothesis has been proven in both species where significant differences have been observed in all the selected radial portions with the exception of portion 90 %.

Hardness strength:

- The hypothesis has been proven in both species in case of hardness strength of the transversal section of the inner wood and rejected in the outer one.

- The hypothesis has been proven in case of Acacia seyal hardness strength in radial section and rejected in case of Balanites aegyptiaca.

- 165 -

Results and Discussion

Hypothesis 4: this hypothesis considered that expected modification of wood properties due to the selected rainfall zones may enhance the suitability of the selected species for pulp and paper production and/or for the flooring industry through its effect on their fibre dimension, hardness strength and density.

The modification of some of the wood properties of the study spices due to climatic zones, enhanced their suitability for PPM and flooring industries as follows:

- Enhancing the suitability of trees growing in zone two (more humid) for the PPM purpose by causing in producing lower density and longer fibres in the case of Acacia seyal and lower density in case of Balanites aegyptiaca.

- Enhancing the suitability of trees growing in zone two for flooring industry due to the lower wood density detected for both species.

- 166 -

Conclusions and Outlook

6 Conclusions and Outlook 6.1 Conclusions

The overall objective of the current research study was to examine the suitability of the selected species for pulp and paper (industerial utilization) and flooring industries (more workman- like utilization) by providing basic information about important anatomical, mechanical and physical wood properties of the study species, as well as to investigate the effect of rainfall (zones) on the selected wood properties.

Based on the valuable results obtained from the current study, the following conclusion can be drawn:

6.1.1 Anatomical properties

Anatomical composition

The anatomical structure of both species is typical to tree species growing in topical and sub- tropical areas. The described anatomical composition provided by the current study may be used as a base for the study species anatomical composition description depending on IAWA list.

Fibres characteristics

- Fibre length of both species is considered as medium (900 -1600 µm). However Acacia seyal has longer fibre than Balanites aegyptiaca .

- Depending upon the fibre length radial variation results, in both species the first three portions from pith (10 %, 30 % and 50 %) can be considered as juvenile wood, and the last two portions (70 % and 90 %) can be considered as mature wood.

- Fibre characteristics of both species followed the increase pattern from pith to bark with the exception of fibre lumen diameter as well as flexibility coefficient in case of Acacia seyal, and Runkel ratio in case of Balanites aegyptiaca where the decreasing trend has been observed.

- The increase patterns detected in most of the fibre characteristics of both species may be explained by the increase of cambial age.

- 167 -

Conclusions and Outlook

Vessels characteristics

- The vessel tangential diameter for both species is classified as medium (100 - < 200 µm).

- The entire vessel's characteristics of both species followed the increase pattern from the inner to the outer samples.

6.1.2 Wood density

Basic density

- Acacia seyal wood can be classified as heavy while that of Balanites aegyptiaca can be classified as medium.

- Both species follow the increasing trend from pith to bark. However, Acacia seyal has sharper increase trend than Balanites aegyptiaca , which in turn causing the vertical decreasing trend in Acacia seyal vs. constant vertical trend in Balanites aegyptiaca.

- The observed wood basic density radial increasing trend for both species may be explained by the radial increasing trend found in fibre and vessel wall thickness particularly that observed in fibre wall thickness.

Air dry density

- Acacia seyal wood is classified as heavy, while that of Balanites aegyptiaca is classified as medium.

- The air dry density radial and vertical trends showed a very similar pattern for those observed in the basic density for both species where a radial increasing trend was observed for both species (sharper in Acacia seyal) and a vertical decreasing trend in Acacia seyal vs. constant trend in Balanites aegyptiaca.

Density achieved by X-ray densitometry

- Both species showed no significant radial variation. However, Acacia seyal showed an increasing trend, while that of Balanites aegyptiaca showed the reverse trend.

- The radial decreasing trend observed in Balanites aegyptiaca may be explained by the existence of tyloses, crystals and resin canals, particularly in the inner part. The existence of such materials prevents the penetration of the light through the

- 168 -

Conclusions and Outlook

radiographed sample during X -ray analyses and thus resulting in obtaining high values.

6.1.3 Hardness strength

The wood of both species can be classified as very hard. Hardness strength in transverse section of both species showed an increase trend from the inner to the outer wood.

6.1.4 Variation of the wood properties due to climatic zones

Depending upon the effect of water stress on the outer samples wood properties (considered as mature wood) the following conclusion can be summarized:

- In Acacia seyal , fibre length and vessel wall thickness considered the only two anatomical properties which showed response to water stress.

- In Balanites aegyptiaca , vessel dimensions appeared to be more sensitive to water stress than fibre dimensions where no response has been observed.

- Balanites aegyptiaca conductive system (vessels) has more adaptive efficiency to water stress than that of Acacia seyal .

- Wood basic density of both species is considered more sensitive to water stress than their hardness strength (HB) where no response has been observed with the exception of the hardness strength in radial section for Acacia seyal.

6.1.5 Suitability of the study species for pulp and paper and flooring industries

- The overall results confirmed the suitability of the study species for PPM and flooring industry. However, for PPM aspect, mixing their wood with those of other species having a lower Runkel ratio and higher flexibility coefficient is recommended in order to improve the paper properties produced from their wood.

- Young trees are considered more suitable for PPM, in particular in the case of Acacia seyal .

- Trees growing in zone two (more humid) are considered more suitable for PPM as well as flooring industry.

- 169 -

Conclusions and Outlook

6.2 Research Outlook The current study succeed in providing valuable information about the wood properties of the study species and their variation. The provided informations help in the assessment of using the study species wood on some forest based industry. However, more researchs still needed in order to provide more proper results, such as:

- Study the wood properties of trees growing in plantation in order to know the effect of tree age in the wood properties which in turn affect the tree final uses.

- Study some chemical properties with importance to PPM such as cellulose and lignin content and others, to enhance their suitability for PPM.

- Study more physical properties with importance to flooring industry, such as shrinkage, swelling, etc, to enhance their suitability for the flooring industry.

- Silvicultural studies, to achieve sustainable yield combining with the best wood properties.

- More studies in tree ring analysis in relation to dendrochronology.

However, the overall objectives of the current study were successfully achieved. From the results of this study, it is obvious that the studied species may be put into better use than conversion to firewood and charcoal. The suitability of study species for pulp and paper and flooring industries may significantly contribute to economic development without fear of exhaustion, as they are renewable under almost all climatic conditions in Sudan.

- 170 -

Summary

7 Summary

Despite the diversity of trees found in Sudan only a few are widely used, and despite the richness of Sudan in most of the basic factors required to establish forest based industries it is still dependent on import to satisfy its needs from the products of such industries. Therefore, the current study aimed to examine the suitability of the selected species for pulp and paper and for flooring industries by providing basic information about some anatomical, mechanical and physical wood properties of the study species, as well as to investigate the effect of rainfall (zones) on the selected wood properties.

Thirty trees per species were collected from eleven forests located in four states in Sudan namely: Blue Nile, North Kordofan, South Kordofan and White Nile. The study areas were divided into two precipitation zones. Zone one with 273 mm mean annual rainfall, and zone two with 701 mm mean annual rainfall. Wood samples in form of discs were obtained from two proportional heights within each tree, which are 10 % and 90 % from the tree merchantable height. Five free of defect samples (strips) were taken from each disc of 10 % height and three samples from 90 % height given a total of 8 samples for each tree, and a total of 480 samples for both species from both zones. Anatomical, physical and mechanical investigations were conducted. The studied anatomical properties were: fibres and vessel diameter, lumen diameter and wall thickness. In addition to fibre length and three fibre derived values (flexibility coefficient, Runkel ratio and slenderness ratio). The radial trend of fibre length was determined. The anatomical composition was described. Three wood density types were measured in the current study: Basic density, air dry density as well as density achieved by X-ray densitometry. The trend of wood basic density from pith to bark was determined. Brinell hardness strength was measured in transverse and radial sections.

Valuable scientific knowledge was gathered through the conducted investigations which can be summarised as follows:

7.1 Anatomical properties

Anatomical composition

- Both species have distinct tree ring, diffuse- porous, vessels in multiple or clusters, simple perforation plate.

- 171 -

Summary

- Several axial parenchyma distributions types were observed in both species. However, banded and aliform parenchyma distributions are common in case of Acacia seyal while diffuse and diffuse in aggregates are common in case of Balanites aegyptiaca .

- Fibres of both species are medium in length, with thick wall thickness and non septet.

- Rays in both species are multiseriate and composed of a single cell type.

Fibres characteristics

- Acacia seyal has a mean of 1.26 mm fibre lengths, while Balanites aegyptiaca has a mean of 1.15 mm fiber length of both species can be classified as medium (900 -1600 µm) .

- Both species have an increasing trend of fibre length from pith to bark, showing an increase in the first three portions from pith (10 %, 30 % and 50 %), and then became stable in the last two portions (70 % and 90 %).

- Acacia seyal has a fibre diameter of 10.70 µm, lumen diameter of 3.71 µm and thick wall thickness of about 3.46. µm. The flexibility coefficient is found to be 34.44 %. The Runkel ratio and slenderness ratio were 1.99 and 117.76, respectively. All the previously mentioned fibre characteristics followed the increase pattern from pith to bark with the exception of fibre lumen diameters as well as flexibility coefficient where the decrease pattern has been observed.

- Balanites aegyptiaca fibre diameter is found to be 13.01 µm, while those of fibre lumen diameters and wall thickness are found to be 5.28 and 3.81 µm, respectively. Fibre wall thickness is classified as thick. The studied fibre derived values recorded the following mean values: 41.09 % flexibility coefficient, 1.48 Runkel ratio and 88.39 slenderness ratio. An increase pattern from the inner to the outer samples has been observed for the entire fibre characteristics with the exception of Runkel ratio where the decreasing trend has been observed.

Vessels characteristics

- Acacia seyal has vessel diameter of 122.40 µm (classified as medium), lumen diameter of 103.42 µm and wall thickness of 9.13 µm.

- 172 -

Summary

- Balanites aegyptiaca vessel diameter is found to be 125 µm (classified as medium). Its lumen diameter and wall thickness are found to be 103.09 µm and 10.68 µm, respectively.

- The entire vessel's characteristics of both species followed the increase pattern from the inner to the outer samples.

7.2 Wood density

Basic density

- Acacia seyal density is found to be 724 kg/m³ while that of Balanites aegyptiaca is to 659 kg/m³.

- Both species showed significant increase in wood density along the radial position. However, Acacia seyal has more sharp increase trend.

- The vertical variation was not significant for both species. However, Acacia seyal density followed a decline trend with increasing height of the tree vs. constant trend in case of Balanites aegyptiaca.

Air dry density

- Acacia seyal air dry density is found to be 0.893 g/cm³, while that of Balanites aegyptiaca is 0.789 g/cm³.

- Both species recorded significant increasing trend from pith to bark, with a more sharp increase in Acacia seyal in comparison with Balanites aegyptiaca.

- Both species showed no significant vertical variation. However, Acacia seyal density followed a decline trend with increasing height of the tree vs. constant trend in case of Balanites aegyptiaca.

Density achieved by X-ray densitometry

- Acacia seyal density achieved by X-ray is found to be 0.901 g/cm³, while that of Balanites aegyptiaca is 0.878 g/cm³.

- No significant differences were found between the study species air dry density and X- ray density.

- 173 -

Summary

- Acacia seyal X-ray density showed an increasing trend from pith to bark, while that of Balanites aegyptiaca showed the reverse trend.

7.3 Hardness strength

- Acacia seyal has a mean of 84.03 N/mm 2 hardness strength in transverse section and a mean of 51.18 N/mm 2 in radial section. Balanites aegyptiaca has a mean of 86.67 and 45.53 N/mm 2 for hardness strength in the transverse and redial sections, respectively.

- The vertical variation was not significant in both species for hardness strength in transverse and radial sections.

- The hardness strength in transverse section for both species showed an increasing trend from the inner to the outer wood.

7.4 Variation of the wood properties due to climatic zones

Some wood properties were significantly affected by the water stress on zone one which can be summarized as follows:

- Acacia seyal fibre length (negatively), vessel wall thickness (positively), basic density as well as hardness strength of the radial section (positively).

- Balanites aegyptiaca vessel's dimensions (negatively) and basic density (positively).

However, the water stress did not affect all of:

- Acacia seyal fibre and vessel diameter and lumen diameter, fiber wall thickness, flexibility coefficient, Runkel ratio and hardness strength in transverse section.

- Balanites aegyptiaca fibre characteristics and hardness strength.

7.5 Suitability of the study species for pulp and paper and flooring industries

Suitability for pulp and paper industry:

In general, wood basic density and fibre characteristics of the study species are compatible for pulp and paper making.

- In comparison with the acceptable values for pulp and papermaking (PPM), Acacia seyal has slightly higher density mean value while Balanites aegyptiaca has comparable mean value. Both species have longer fiber, higher slenderness ratios and

- 174 -

Summary

lower felexbility coefficent. Balanites aegyptiaca ’s Runkel ratio is at the upper end of the acceptable range of PPM, while that of Acacia seyal is out of the range.

- In comparison with Gmelina arborea wood properties (the prime sources for PPM), the study species have comparable fibre lengths, lower flexibility coefficient, higher Runkel and slenderness ratios and higher wood density.

- Paper made from the study species is expected to have low strength properties and good tear resistance due to their low flexibility coefficient and high Runkel and slenderness ratios.

- Acacia seyal juvenile wood has more comparable wood properties mean values with the acceptable values for PPM than mature wood.

- The significant differences detected on some wood properties due to rainfall zones, enhanced the suitability of trees growing in zone two (the more humid) for the PPM purpose by causing in producing lower density and longer fibres in case of Acacia seyal and lower density in case of Balanites aegyptiaca.

Suitability for flooring industry:

Wood air dry density and hardness strength of the study species are compatible for the flooring industry.

- Both species have higher air dry density and hardness strength mean values than Milletia laurentii and Quercus rubra which commercially used for flooring industry. This enhance the suitable of the study species for flooring industry.

- The significant differences detected on some wood properties due to rainfall zones, enhanced the suitability of trees growing in zone two (the more humid) for flooring industry due to the lower wood density detected for both species.

- 175 -

Registers

8 Registers

8.1 Sources Ademiluyi, E. O., and Okeke, R. E. 1979. Studies on specific gravity and fibre characteristics of Gmelina arborea in some Nigerian plantations. Nigerian Journal of Science, 13: 231 - 238.

Akachuku, A. E. 1985. The effects of some extrinsic and intrinsic factors on the proportion of vessels in Gmelina arborea Roxb. Wood Science and Technology, 19: 5 - 12.

Akachuku, A. E. and Burley, J. 1979. Variation of anatomy of Gmelina arborea Roxb. in Nigerian plantations . IAWA Bulletin n. s, 4: 94 - 99.

Akgül, M., and Tozluoğlu, A. 2009. Some chemical and morphological properties of juvenille woods from beech ( Fagus orientalis L.) and pine ( Pinus nigra A.) plantations. Trends in Applied Sciences Research, 4: 116 - 125.

Al-Khalifah, N. S., Khan, P. R., Al-Abdulkader, A. M., Nasroun, T. 2006. Impact of water stress on the sapwood anatomy and functional morphology of Calligonum comosum. IAWA Journal, 27 (3): 299 - 312 .

Alves, E. S., and Angyalossy-Alfonso, V. 2000. Ecological trends in the wood anatomy of some Brazilian species. 1. Growth rings and vessels. IAWA Journal, 21: 3 - 30.

Alves, E. S., and Angyalossy-Alfonso, V. 2002. Ecological trends in the wood anatomy of some Brazilian species. 2. Axial parenchyma, rays and fibres. IAWA Journal, 23: 391 - 418.

Andrews, F. W. 1950, 1952, 1956. The Flowering Plants of the (Anglo- Egyptian) Sudan. Vol. 1, 2, and 3. The Burncle of Co., Ltd, Arbroath.

Anonymous. 1978. Pulp and Paper. West African Technical Review, 7(3): 45-47.

Anonymous.1991. DIN 52 182. Bestimmung der Rohdichte. Deutsches Institut für Normen, über Holz, Bd 31.

Anonymous. 1984. The Malaysian Grading Rules for Sawn Hardwood Timber (MGR). Malaysian Timber Industry Board, Kuala Lumpur.

Anonymous. 2000. DIN EN 1543. Bestimmung des Eindruckwiderstandes (BrinelI). Deutsches Institut für Normen, über Holz, Bd 31.

Antonova, G. F. V., and Stasova, V. V. 1997. Effects of environmental factors on wood formation in Larch ( Larix sibirica Ldb.) stems. Trees-Structure and Function, 11: 462 - 468.

Antonova, G. F. V., and Stasova, V. V. 1993. Effects of environmental factors on wood formation in Scots pine stems. Trees-Structure and Function, 7: 214 - 219.

- 176 -

Registers

Arbonnier, M. 2004. Trees, shrubs and lianas of West African dry zones. Margraf Publ, Weikersheim. 573 P.

ASTM. 2005. Annual book of standards, Vol. 04.10. Wood. American Society for Testing and Materials, West Conshohoken, PA. D 143-94. Standard methods of testing small clear specimens of timber. D 1037-99. Standard test methods for evaluating properties of wood-based fiber and particle panel materials.

Atibt. 1965. Nomenclature générale des bois tropicaux. CTFT Nogent sur Marne. In: Harzmann, L. J. 1988. Kurzer Grundriß der allgemeinen Tropenholzkunde. S. Hirzel Verlag, Leipzig. 255 P.

Auge, R. M., Duan, X. G., Croker, J. L., Witte, W. T., Green, C. D. 1998. Foliar dehydration tolerance of twelve deciduous tree species. Journal of Experimental Botany, 49: 753 - 759.

Baas, P., and Schweingruber, F. H. 1987. Ecological trends in the wood anatomy of trees, shrubs and climbers from Europe. lAWA Bulletin n.s, 8: 245 - 274.

Badi, K. H., Ahmed, A. E., Bayoumi, A. A. 1989. The forest of Sudan. The National Council for Research, Khartoum, Sudan. 184 P.

Baker, C. F. 1995. Good practice for refining the types of fiber found in modern paper furnishes. TAPPI Journal, 78: 147 - 153.

Bakhshi, R., Kiaei, M., Mosavi, S. M. S. 2012. Wood Properties Variation along Radial Position in Quercus castaneaefolia . Middle-East Journal of Scientific Research, 11 (4): 511 - 516.

Balodis, V. 1994. Quality assurance in pulpwood testing: the sample. In: Brown, A.G. (ed.) Australian Tree Species Research in China. Proceedings of an International Workshop held from 2 - 5 November 1994, Zhangzhou, Fujian Province, PR China. Australian Centre for International Agricultural Research, Proceedings No. 48, 191 - 195.

Barajas-Morales, J. 1985. Wood structural differences between trees of two tropical forests in Mexico. IAWA Bulletin n. s, 6: 355 - 364.

Barbier, E. B., Markandya, A., Pearce, D. W. 1990. Sustainable agricultural development and project appraisal. European Review of Agricultural Economics, 17: 181 - 196.

Barefoot, A. C., Hitchings, R. G., Ellwood, E. L. 1964. Wood characteristics and kraft paper properties of four selected loblolly pines. TAPPI Journal , 47(6): 343 - 356.

Barefoot, A. C., Hitching, R. G., Ellwood, E. L., Wilson, E. 1970. The relationship between loblolly pine fiber morphology and kraft paper properties. Technical Bulletin, 202. N. C. State university, Agr. Expt. Sta., Raleigh, North Carolina, 88 P.

Barnett, J. R. and Jeronimidis, G. 2003. Wood Quality and its Biological Basis. Blackwell Publishing Ltd. Oxford. 226 P.

- 177 -

Registers

Beadle, C. L., Banham, P. W., Worledge, D., Russell, S. L., Hetherington, S. J., Honeysett, J. L., White, D. A. 2001. Effect of irrigation on growth and fibre quality of Eucalyptus globulus and Eucalyptus nitens . Appita Journal, 54: 144 - 147.

Bektas, I., Tutus, A., Eroglu, H. 1999. A Study of The Suitability of Calabrian Pine ( Pinus brutia Ten.) For Pulp and Paper Manufacture. Turkish Journal of Agriculture and Forestry, 23(3): 589 - 597.

Bendtsen, B. A. 1978. Properties of wood from improved and intensively managed trees. Forest Products Journal, 28(10): 61 - 72.

Bhat, K. M. and Priya, P. B. 2004. Influence of the provenance variation on wood properties of teak from the western Ghat region in India. IAWA Journal, 25(3): 273 - 282.

Bhat, K. M., Bhat, K. V., Dhamodaran, T. K. 1989. Fiber length variation in stem and branches of 11 tropical hardwoods. IAWA Bulletin n.s, 10: 63 - 70.

Bhat, K. M., Bhatt, K. V., Dhamodaran, T. K. 1990. Wood density and fiber length of Eucalyptus grandis grown in Kerala, (India). Wood and Fiber Science, 22(1): 54 - 61.

Bhat, K.M., Priya, P.B. Rugmini, P. 2001. Characterisation of juvenile wood in teak. Wood Science and Technology , 34: 517 - 532.

Bisset, I. J. and Dadswell, H. C. 1949. The variation in fiber length within one tree of Eucalyptus regnans . Australian Forestry, 13: 86 - 96.

Blankenhorn, P.R., Norton, B.D., Allen, G. 2005. Selected quality characteristics of white ash used in professional baseball bats. Forest Products Journal, 53(3): 43 - 46.

Bosman, M. T. M., Kort., I. De., Van Genderen, M. K., Baas, P. 1994. Radial variation in wood properties of naturally and plantation grown light red meranti ( Shorea spp., Dipterocarpaceae ). IAWA Journal, 15 (2): 111 - 120.

Bowyer, L. J., Shmulsky, R., Haygreen, G. J. 2003. Forest Products and Wood Science: An Introduction, 4th Edition. Blackwell Publishing Company. Iowa. 554 p.

Brasil, M. A. and M. Ferreira. 1972. Variaçao da densidade básica e das características das fibras em Eucalyptus grandis Hill ex Maiden ao nivel do DAP. IPEF, 5: 81 - 90.

Broun, A. F. and Massey. R. E. 1929. Flora of the Sudan. Wellington House, London. 502 p.

Butterfield, R. P., Crook, R. P., Adams, R., Morris, R. 1993. Radial variation in wood specific gravity, fiber length and vessel area for two central American hardwoods: Hyeronima alchorneoides and Vochysia guatemalensis : natural and plantation-grown trees. IAWA Journal,14 (2): 153 - 161.

Carlquist, S. 1966. Wood anatomy of Compositae: a summary, with comments on factors controlling wood evolution. Aliso, 6 (2): 25 - 44.

- 178 -

Registers

Carlquist, S. 1975. Ecological strategies of xylem evolution. Univ of California Press, Berkeley. Los Angeles. London. 259 P.

Carlquist, S. 1988. Comparative Wood Anatomy: Systematic, Ecological, and Evolutionary Aspects of Dicotyledon Wood. Berlin: Springer-Verlag. 436 p.

Carlquist, S., and Hoekman, D. A. 1985. Ecological wood anatomy of the woody Southern Californian flora. IAWA Bulletin n.s, 6: 319 - 347.

Casey, J. P. 1980. Pulp and Paper Chemistry and Chemical Technology, vol. 2. Wiley Interscience, New York, USA. 576 p.

Catchpoole, S. J., Downes, G., Read, S.M. 2000. The effect of salt on wood and fibre formation in eucalypts. Report No. UM-18A. Rural Industries Research and Development Corporation, Canberra, 55 P.

Chattaway, M. M. 1932. Proposed standards for numerical values used in describing woods. Tropical Woods, 29: 20 - 28.

Chowdhury, M. Q., Ishiguri, F., Hiraiwa, T., Matsumoto, K., Takashima, Y., Iizuka, K., Yokota, S., Yoshizawa, N. 2012. Variation in anatomical properties and correlations with wood density and compressive strength in Casuarina equisetifolia growing in Bangladesh. Australian Forestry, 75: 95 - 99.

Chowdhury, M. Q., Ishiguri, F., Hiraiwa, T., Takashima, Y., Iizuka, K., Yokota, S., Yoshizawa, N. 2013. Anatomical property variation in Acacia auriculiformis growing in Bangladesh. International Wood Products Journal, 4(2): 75 - 80.

Chowdhury, M. Q., Ishiguri, F., Iizuka, K., Takashima, Y., Matsumoto, K., Hiraiwa, T., Ishido, M., Sanpe, H., Yokota, S., Yoshizawa, N. 2009. Radial variations of wood properties in Casuarina equisetifolia growing in Bangladesh. Journal of Wood Science 55, 139 - 143.

Chudnoff, M. 1961. The physical and mechanical properties of Eucalyptus camaldulensis . Isr Journal of Agriculture Research, 66: 39 pp.

Cown, D. J., Young, G. D., Burdon, R. D. 1992. Variation in wood characteristics of 20 year old half - sib families of Pinus radiata . New Zealand Journal of Forestry Science , 22(1): 63 - 76.

Creber, G.T., and Chaloner, W.G., 1984. Influences of environmental factors on the wood structure of living and fossil trees. Botanical Review , 50: 357 - 448.

Dadswell, H. E., and Nicholls, J. W. P. 1959. Assessment of wood qualities for tree breeding. CSIRO Australia, Division of Forest Products Technology. Scientific Paper No. 4, 16 p.

Dadswell, H. E., and Wardrop, A. B. 1954. Growing trees with wood properties desirable for paper manufacture. Appita Journal, 12(1): 129 - 136.

- 179 -

Registers

Dargavel, J. B. 1968. Variation in the basic density of mountain ash. Australian Forest Research, 2 (3): 25-30.

Davis, E. M. 1961. The effect of density upon wood properties. Wood Work Dig, 63(9): 52 - 53.

Dawelbait, N., Yousif, S., Branthomme, A., Elmahi, A. G., Lungo, A. D., Abdelnour, H. 2006. Global forest resources assessment 2005. Estimating forest cover and forest cover change in Sudan. FRA Working Paper 109/E. Rome: FAO.

De Guth, E. B. 1980. Relationship between wood density and tree diameter in Pinus elliottii of Missiones Argentina. IUFRO Conference Div. 5 Oxford, England. 1p (Summary).

DeBell, D. S., Singleton R., Harrrington C. A., Gartner B. L. 2002. Wood density and fiber length in young populus stems: Relation to age, growth rate, and pruning. Wood and Fiber Science, 34(4): 529 - 539.

Détienne, P. 1989. Appearance and periodicity of growth rings in some tropical woods. IAWA Bulletin n. s. 10: 123 - 132.

Dickison, W. C. and Phend, K. D. 1985. Wood anatomy of Styracaceae : evolutionary and ecological considerations. IAWA Bulletin n. s. 6: 3 - 22.

Dickmann, D. I. 1975. Plant materials appropriate for intensive culture of wood fibre in the North Central region of Iowa State. Journal of Research, 49(3) part 2: 281 - 286.

Dinwoodie, J. M. 1965. The relationship between fibre morphology and paper properties. A review of the literature. TAPPI Journal, 48(8): 440 - 447.

Dinwoodie J. M. 2000. Timber: Its nature and behaviour. 2nd ed. Taylor & Francis Group, London, UK. 272 P.

Dunham, R. A. 1999. The effect of growth rate on the strength properties of sawn beams of silver birch. Scandinavian Journal of Forest Research , 14: 18 - 26.

Dutt, D., and Tyagi, C. H. 2011. Comparison of various eucalyptus species for their morphological, chemical, pulp and paper making characteristics. Indian Journal of Chemical Technology, 18: 145 - 151.

El Amin, H. M. 1990. Tree and shrubs of the Sudan. Ithaca press, Exeter. 484 P.

El Feel, A. A. 2004. Gynecology and Conservation Status of Balanites aegyptiaca (L.) Del. in Sudan. Ph.D. Thesis. Sudan University of Science and Technology. College of post graduates studies. 150 P.

Enu-Kwesi, L., Nwalozie, M. C., Anyanwu, D. I. 1986. Effects of pre-sowing hydration- dehydration on germination, vegetative growth and fruit yield of Abelmoschus esculentus grown under two moisture regimes. Tropical Agriculture, 63: 181 - 184.

- 180 -

Registers

Esan, R. F. 1966. Study of variation in some structural features and properties of Gmelina arborea Roxb. Ph.D. Thesis, Faculty of Forestry, University of Oxford, Oxfort, England.

Espinoza, J. A. 2004. Within-tree density gradient in Gmelina arborea in Venezuela. New Forest, 28 (2/3): 309 - 317.

Fahn, A., Werker, E., Baas, P. 1986. Wood anatomy and identification of trees and shrubs from Israel and adjacent regions. The Israel Academy of Sciences and Humanities, Jerusalem. 221 P.

February, E. C. 1993. Sensitivity of xylem vessel size and frequency to rainfall and temperature: implications for palaeontology. Palaeontologia Africana, 30: 91 - 95.

February, E. C., Stock, W. D., Bond, W. J., Le Roux, D. J. 1995. Relationships between water availability and selected vessel characteristics in Eucalyptus grandis and 2 hybrids. IAWA jounal, 16 (3): 269 - 276.

Fimbel, R. A. and Sjaastad, E. O. 1994. Wood specific gravity variability in Ceiba pentandra . Wood and Fiber Science Journal, 26(1): 91 - 96.

Fisher, J., Goldstein, G., Jones, T., Cordell, S. 2007. Wood vessel diameter is related to elevation and genotype tree in the Hawaiian Metrosideros polymorpha (Myrtaceae ). American Journal of Botany , 94: 709 - 715.

Foelkel, C. 2007. Vessel elements and eucalyptus pulps. Eucalyptus online book and newsletter. Sponsoring Organizations: BOTNIA.

Fonti, P., and Garcia-Gonzalez, I., 2004. Suitability of chestnut earlywood vessel chronologies for ecological studies. New Phytologist, 163: 77 - 86.

Fors, A. 1965. Maderas cubanas. Habana. In: Harzmann, L. J. 1988. Kurzer Grundriß der allgemeinen Tropenholzkunde. S. Hirzel Verlag, Leipzig. 255 p.

Frederick, D.J. Madgwick, H.A.I., Ologer, G.R. 1982. Wood basic density and moisture content of young Eucalyptus regnans grown in New Zealand. New Zealand Journal of Forestry Science, l2 (3): 494 - 500.

Furukawa, I. and H. Hashizume. 1987. The influence of fertilization and improvement cutting on the wood quality of mature kunugi trees. Mokuzai Gakkaishi, 33: 443 - 449 [In Japanese with English summary].

Gafaar, A. 2011. Forest Plantations and Woodlots in Sudan. Africa Forest Forum.Volume 1(15). 76 p.

Gartner, B. L., Lei, H. Milota, M. R. 1997. Variation in the anatomy and specific gravity of wood within and between trees of red alder ( Alnus rubra Bong.). Wood fiber science, 29: 10 - 20.

- 181 -

Registers

Gebrekirstos, A., Mitlöhner, R., Teketay, D. Worbes, M. 2008. Climate–growth relationships of the dominant tree species from semi-arid savanna woodland in Ethiopia. Trees- Structure and Function, 22: 631 - 641.

Gnanaharan, R. and Dhamodaran, T. K. 1992. Mechanical properties of rubberwood from a 35-year-old plantation in central Kerala, India. Journal of Tropical Forest Science, 6 (2): 136 -140.

Göhre, K. 1960. Die Verteilung von Rohwichte im Pappelstamm. Wiss Abh Deutsch Akad Landwiss Berlin, 44: 51 - 79. In: Zobel, B. J., and Van Buijtenen, J. P. 1989. Wood Variation: Its Causes and Control. Springer, New York, USA. 363 P.

Gour, V. S, and Kant, T. 2012. Balanites aegyptiaca (L.) Del.: A multipurpose and potential biodiesel tree species of the arid regions. International Journal of Science and Nature, 3(2): 472 - 475. Grosser, D. 1971. Beitrag zur Histologie und Klassifikation asiatischer Bambusarten. In: Mitteilungen der Bundesforschungsanstalt für Forst und Holzwirtschaft Reinbeck, Nr. 85. WiedebuschVerlag, Hamburg. 321 P. Günther, B. 2012. Erarbeitung einer Methode zur Erfassung von dendrochronologisch relevanten Jahrringmerkmalen der Trauben-Eiche ( Quercus petraea [Matt.] Liebl.) auf Grundlage der Röntgendensitometrie. PhD thesis, Dresden University of Technology, Dresden, Germany. 145 p.

Hall, J. B., and McAllan, A. 1993. Acacia seyal : a monograph. Bangor, UK: School of Agricultural and Forest Sciences, Univesrity of Wales.

Hannrup, B., Wilhelmsson, L., Danell, Ö. 1998. Time trends for genetic parameters of wood density and growth traits in Pinus sylvestris L. Silvae Genetica, 47(4): 214 - 219.

Hans, A. S., Burley, J. Williamson, P. 1972. Wood quality of Eucalyptus grandis (Hill) Maiden grown in Zambia. Holzforschung, 26: 138 - 141.

Hapla, F. 1993. Versuchsplanung in der Holzforschung unter Berücksichtigung der Repräsentativität der Ergebnisse sowie des Arbeitsaufwands. Holz als Roh- und Werkstoff, 51: 88 - 92.

Harris, J. M. 1965. A survey of the wood density, tracheid length and latewood characteristics of radiata pine grown in New Zealand. New Zealand Forest Service Technical Paper No. 47.

Harzmann, L. J. 1988. Kurzer Grundriß der allgemeinen Tropenholzkunde. S. Hirzel Verlag, Leipzig. 255 P.

Haygreen, J. G. and Bowyer, J. L. 1996. Forest Products and Wood Science. An Introduction 3nd Edition. IOWA State University Press, Ames, 484 P.

Hein, P. R. G., Bouvet, J. M., Mandrou, E., Vigneron, P., Clair, B., Chaix, G. 2012. Age trends of microfibril angle inheritance and their genetic and environmental

- 182 -

Registers

correlations with growth, density and chemical properties in Eucalyptus urophylla ST Blake wood. Annals of Forest Science, 69, 681 - 691.

Hietz, P., Valencia, R., Joseph Wright, S. 2013. Strong radial variation in wood density follows a uniform pattern in two neotropical rain forests. Functional Ecology, 27: 684 - 692.

Hoadley, R. 1990. Identifying wood: Accurate results with simple tools. Thunton Press, United States of America. 223 P.

Honjo, K., Furukawa, I., Sahri, M. H. 2005. Radial variation of fiber length increment in Acacia mangium . IAWA Journal, 26: 339 - 352.

Horn, R. A. 1978. Morphology of pulp fiber from hardwoods and influence on paper strength. USDA Forest Service. Research Paper FPL 312, Forest Products Laboratory, Madison, WI, USA. 12 p.

Horn, R. A., and Setterholm, V. C. 1990. Fiber morphology and new crops. p. 270 - 275. In: Janick, J., and Simon, J. E. (eds.), Advances in new crops. Timber Press, Portland, OR.

Hsaio, T. C. 1973. Plant responses to water stress. Annual Review of Plant Physiology 24: 519 - 570.

Hughes, J. F., and Esan, D. 1969. Variation in some structural features and properties of Gmelina arborea . Tropical Science, 1: 23 - 37.

Hurter, A. M. 1988. Utilization of Annual Plants and Agricultural Residues for the production of pulp and paper. Proceedings of TAPPI and Pulping Conference 1988. New Orleans, Louisiana, USA., Book 1. pp. 139 - 160.

IAWA Committee. 1989. IAWA list of microscopic features for hardwood identification. IAWA Bulletin. n.s 10 (3): 221 - 332.

Igartúa, D. V., Monteoliva, S. E., Monterubbianesi , M. G., Villegas, M. S. 2003. Basic density and fibre length at breast height of Eucalyptus globulus ssp. globulus for parameter prediction of the whole tree. IAWA Journal, 24 (2): 173 - 184.

Ilvessalo-Pfäffli, M. S. 1995. Fiber Atlas, Identification of Papermaking Fibers. Springer- Verlag, Berlin. Germany. 400 P.

Ishiguri. F., Hiraiwa, T., Iizuka, K., Yokota, S., Priadi, D., Sumiasri N., Yoshizawa N. 2009. Radial variation of anatomical characteristics in Paraserianthesis falcataria in Indonesia. IAWA Journal, 30 (3): 343 - 352.

Ishiguri, F., Takeuchi, M., Makino, K., Wahyudi, I., Takashima, Y., Iizuka, K., Yokota, S., Yoshizawa, N. 2012. Cell morphology and wood properties of Shorea acuminatissima planted in indonesia. IAWA Journal, 33: 25 - 38.

- 183 -

Registers

Ismail, J., Jusoh, M. Z., Sahri, M. H. 1995. Anatomical variation in planted kelempayan (Neolamarckia cadamba , Rubiaceae). IAWA Journal, 16: 277 - 287.

Iwenofu, V. N. 1979. The determination of some fibre characteristic and wood density of Treculia africana PRODA. Technical Report, 2: 1 - 124.

Izekor, D. N., Fuwape, J. A. 2012. Correlation between fibre length characteristics and mechanical properties of Tectona grandis wood. Journal of Applied and Natural Science, 4 (2): 258 - 263.

Izekor, D. N., Fuwape, J. A., Oluyege, A. O. 2010. Effects of density on variations in the mechanical properties of plantation grown Tectona grandis wood. Applied Science Research, 2(6): 113 - 120.

Jagiella, C., and Kürschner, H. 1987. Atlas der Hölzer Saudi Arabiens. Die Holzanatomie der wichtigsten Bäume und Sträucher mit einem holzanatomischen Bestimmungsschlüssel. Dr. Ludwig Reichert Verlag, Wiesbaden. 176 p.

Jain, V. K. and Arora. K. L. 1995. Moisture content, specific gravity and shrinkage variation with radial and axial position within a tree of Eucalyptus camaldulensis . Journal of Timber Development Association of India, 41(1): 23 - 38.

Jane, F.W. 1970. The Structure of Wood. 2nd ed. Completely revised by K. Wilson and D.J.B White. London. 478 P.

Jang, H. F., and Seth, R. S. 1998. Using confocal microscopy to characterize the collapse behaviour of fibers. TAPPI Journal, 81 (5): 167 -174.

John, 1. 1990. Variation of wood anatomy in relation to environmental factors in two southern African hardwoods. PhD thesis, Dept. Pure and Applied Biology, Imperial College London.

Jorge, F., Quilhó, T., Pereira, H. 2000. Variability of fiber length in wood and bark in Eucalyptus globules . IAWA Journal, 21: 41 - 48.

Jozsa, L. A., and Middleton, G. R. 1994. A discussion of wood quality attributes and their practical implications. Forintek Canada, Vancouver, B.C. Special Publication No. Sp- 34.

Kamala, B. S., Kothiyal, V., Sharma, S. K. 2000. Assessment of wood quality of Grevillea robusta from Banglore, Karnataka. Indian Forester, 126 (6): 625 - 633.

Kanehira, R. 1933. On light-weight woods. J. Soc. For. Japan. In : Harzmann, L. J. 1988. Kurzer Grundriß der allgemeinen Tropenholzkunde. S. Hirzel Verlag, Leipzig. 255 p.

Khider, T. O., Elzaki, O. T., Omer, S. H. 2012. Soda and soda-anthraquinone pulping of Albizia lebbeck from Sudan. Suranaree Journal of Science and Technology, 18 (4): 299 - 303.

- 184 -

Registers

Khristova, P., Gabbir, S., Bentcheva, S., Dafalla, S. 1997. Soda-AQ pulping of three Sudanese hardwoods. Tropical Science, 37: 176 - 182.

Khristova, P., Gabir, S., Bentcheva, S., Khider, T. 1998. Soda-AQ pulping of four Acacias from Sudan. Cellulose Chemistry and Technology , 32: 299 - 307.

Khristova, P., and Karar, I. 1999. Soda-anthraquinone pulp form three Acacia nilotica species. Bioresource Technology , 68: 209 - 213.

Khristova, P., Kordsachia, O., Dafalla, S. 2004. Alkaline pulping of Acacia seyal . Tropical Science, 44: 207 - 215.

Khristova, P., Kordsachia, O., Patt, R., Dafaalla, S. 2006. Alkaline pulping of some eucalypts from Sudan. Bioresource Technology, 97: 535 - 544.

Kiaei, M. 2012. The Influence of Cambial Age on Fiber Dimension in Maple Wood. Middle- East Journal of Scientific Research, 11 (8): 1009 - 1012.

Kiaei, M., and Samariha, A. 2011. Fiber dimensions, physical and mechanical properties of five important hardwood plants. Indian Journal of Science and Technology, 4 (11): 1460 - 1463.

Kilpeläinen, A., Peltola, H., Ryyppö, A., Sauvala, K., Laitinen, K. and Kellomäki, S. 2005. Scots pine responses to elevated temperature and carbon dioxide concentration: growth and wood properties. Tree Physiology , 25: 75 - 83.

Kollmann, F. 1951. Technologie des Holzes und der Holzwerkstoffe, 1.Band, Anatomie und Pathologie, Chemie, Physik, Elastizität und Festigkeit. 2. Auflage. Springer-Verlag, Berlin, Göttingen, Heidelberg. 1050 P.

Kollmann, F., and Côté, W. A. 1968. Principles of Wood Science and Technology. I Solid Wood. Springer, Berlin, Heidelberg, New York. 592 P. Kostiainen, K. 2007. Wood properties of northern forest trees grown under elevated CO2, O3 and temperature. PhD thesis, University of Oulu, Faculty of Science, Department of Biology. Finland.

Kramer, P. J. 1963. Water stress and plant growth. Agronomy Journal , 55: 31 - 35.

Lamb, A. F. A. 1968. Fast growing timber trees of the lowland tropics No. 1: Gmelina arborea . University of Oxford and Commonwealth Forestry Institute, Oxford, England. 42 P.

Lauridsen, E. B., and Kjaer, E. D. 2002. Provenance research in Gmelina arborea . A summary of results from three decades of research and a discussion of how to use them. International Forestry Review, 4(1): 1 - 15.

Leal, S., Pereira, H., Grabner M., Wimmer, R. 2003. Clonal and site variation of vessels in 7- year-old Eucalyptus globulus . IAWA Journal, 24 (2): 185 - 195.

- 185 -

Registers

Lei, H., Milota, M. R., Gartner, B. L. 1996. Between- and within-tree variation in the anatomy and specific gravity of wood in Oregon white oak ( Quercus garryana Dougl.) IAWA Journal, 17: 445 - 461.

Lei, H., Gartner B. L., Milota M. R. 1997. Effect of growth rate on the anatomy, specific gravity, and bending properties of wood from 7-year-old red alder ( Alnus rubra ). Canadian Journal of Forest Research, 27(1): 80 - 85.

Lens, F., Luteyn, J. L., Smets, E., Jansen, S. 2004. Ecological trends in the wood anatomy of Vaccinioideae (Ericaceae s.l.). Flora 199: 309 - 319.

Lenz, O., Schär, E., Schweingruber, F. H. 1976. Methodische Probleme bei der radiographisch densitometrischen Bestimmung der Dichte und der Jahrringbreiten von Holz. Holzforschung 30 (4): 114 - 123.

Levitt, J. 1972. Responses of Plants to Environmental Stresses. Academic Press, New York. 697 P.

Lewark, S. 1979. Wood characteristics in Norway spruce breeding programs. Proceeding of IUFRO Working Parties, Buchsrest, Romania, 24 P.

Lewin, M., and Goldstein, I. S. 1991. Wood Structure and Composition. Dekker, New York. 488 P.

Lindorf, H. 1994. Eco-anatomical wood features of species from a very dry tropical forest. IAWA Journal, 15: 361 - 376.

Liphschitz, N. and Waisel, Y. 1970. Effects of environment on relations between extension and cambial growth of Populus euphratica Oliv. New Phytologist, 69: 1059 - 1064.

Logan, A. F., Balodis, V. 1982. Pulping and papermaking characteristics of plantation-grown Acacia mangium from Sabah .The Malaysian Forester, 45 (2): 217 - 236.

Lohmann, U. 1991. Holz Handbuch. 4., völlig überarb. Aufl. DRW Verlag. 312 P.

Lorenz, V. M. 1909. Analytische Untersuchung des Begriffs der Holzhärte. Mitt. Forstl. Versuchswes. Österr., Wien. In: Kollmann F, and Côté, W .A. 1968. Principles of wood science and technology. I Solid wood. Springer, Berlin, Heidelberg, New York.

Lundqvist, S. 2002. Efficient wood and fiber characterization - A key factor in research and operation. Annals of Forest Science, 59: 491 - 501.

Macfarlane, C., and Adams, M. A. 1998. δ13 C of wood in growth-rings indicates cambial activity of drought-stressed trees of Eucalyptus globulus . Functional Ecology, 12 (4): 655 - 664.

Malan, F. S. 1991. Variation, association and inheritance of juvenile wood properties of Eucalyptus grandis Hille × Maiden with special reference to the effect of rate of growth. South African Forestry Journal, 157: 16 - 23.

- 186 -

Registers

Malan, F. S. 1993. The wood properties and qualities of three South African-grown Eucalypt hybrids. South African Forestry Journal, 167: 35 - 44.

Malan, F. S., and Gerischer, G. F. R. 1987. Wood properties differences in South African grown Eucalyptus grandis trees of different growth stress intensity. Holzforschung, 41: 331 - 335.

Malan, F. S., and Hoon, M. 1992. Effect of initial spacing and thinning on some wood properties of Eucalyptus grandis . South African Forestry Journal, 163: 13 - 20.

Martin, C. 1984. Tropical Timbers of the World. USDA Forest Service Ag. Handbook No. 607. 464 p.

Mbow, C., Chhin, S., Sambou, B., Skole, D. 2013. Potential of dendrochronology to assess annual rates of biomass productivity in savanna trees of West Africa. Dendrochronologia, 31: 41 - 51.

McAllan, A. 1993. Acacia seyal : a handbook for extension workers. Bangor, UK: School of Agricultural and Forest Sciences, University of Wales.

McElwee, R. L. and J. P. Faircloth. 1966. Variation in wood properties of water tupelo in the southeastern coastal plain. TAPPI Journal, 49 (12): 538 - 541.

McMillin, C. W., and Manwiller, F. G. 1980. The Wood and Bark of Hardwoods Growing on Southern Pine Sites. A pictorial Atlas, US. Government Printing Office. Washington. 58 P.

Melo, J. E., Coradin, V. T. R., Mendes, J. C. 1990. Classes de densidade para madeiras da Amazônia Brasileira. Anais do Congresso Florestal Brasileiro 6, vol. 3, Campos do Jordão, São Paulo, Sociedade Brasileira de Silvicultura, São Paulo, SP, Brazil: 695 - 699.

Metcalfe, C. R., and Chalk, L. 1979. Anatomy of Dicotyledons. Oxford 1950, 1979. In: Harz- mann, L. J. 1988. Kurzer Grundriß der allgemeinen Tropenholzkunde. S. Hirzel Verlag, Leipzig. 255 p.

Metcalfe, C. R., and Chalk, L. 1983. Anatomy of the dicotyledons. Vol 1. Clarendon press. Oxford. 724 P.

Miller, Regis B. 1999. Characteristics and availability of commercially important woods. Wood handbook: wood as an engineering material. Madison, WI: USDA Forest Service, Forest Products Laboratory. General technical report FPL; GTR-113: Pages 1.1 - 1.34.

Miranda, I., Almeida, M. H., Pereira, H. 2001. Variation of fibre biometry in different provenances of Eucalyptus globulus Labill. Appita Journal, 54: 272 - 280.

Miranda, I., Almeida, M. H., Pereira, H. 2003. The influence of spacing on wood properties for Eucalyptus globulus Labill pulpwood. Appita Journal, 56: 140 - 143.

- 187 -

Registers

Miranda, I., Pereira, H. 2002. Variation of pulpwood quality with provenances and site in Eucalyptus globulus . Annals of Forest Science, 59: 283 - 291.

Moglia, J. G., and Lopez, C. R. 2001. Radial variation trend in the wood of Aspidosperma quebracho blanco. Investigacion Agraria, Sistemasty Recursos Forestales, 10 (1): 69 - 80.

Monteoliva, S., Senisterra, G., Marlats, R. 2005. Variation of wood density and fibre length in six willow clones ( Salix spp.). IAWA Journal, 26 (2): 197 - 202.

Montes, C. S., Hernandez, R. E., Beaulieu, J. 2007. Radial variation in wood density and correlations with growth of Calycophyllum spruceanum at an early age in the Peruvian Amazon. Wood and Fiber Science, 39: 377 - 387.

Mörath, E. 1932. Studien über die hygroskopischen Eigenschaften und die Härte der Hölzer. Hannover. In Kollmann, F., Côté, W. A. 1968. Principles of Wood Science and Technology. I Solid wood. Springer, Berlin, Heidelberg, New York. 592 p. Moya, R., and Fo, M. T. 2008. Variation in the wood anatomical structure of Gmelina arborea (Verbenaceae) trees at different ecological conditions in Costa Rica. Revista Biologia Tropical, 56 (2): 689 - 704.

Mukhtar, M. A. 2008. Variations in wood anatomy of Acacia senegal (L.) Wild . in different rainfall zones in sandy soil of Kordofan region, Sudan. M.Sc. thesis. University of Kordofan, Sudan.

Muneri, A., Daido, T., Henson, M., Johnson, I. 2004. Variation in pulpwood quality of superior Eucalyptus dunnii families grown in NSW. Proceeding in the 58 th Appita Annual Conference, Canberra, Australia, pp 513 - 538.

Mustafa, A. F. 1997. Regeneration of Acacia seyal forests on the dryland of the Sudan caly plain. Tropical Forestry Report No. 15. Helsinki: Department of Forest Ecology, Helsinki University Printing House.

Naidoo, S., Zbonak, A., Ahmed, F. 2006. The effect of moisture availability on wood density and vessel characteristics of Eucalyptus grandis in the warm temperate region of South Africa. Paper presented in IUFRO symposium “Wood structure and properties ‘06”, 3-6 Sept., Sielnica, Slovakia.

Naji, H. R., Sahri, M. H., Nobuchi, T., Bakar, E.S. 2012. Clonal and planting density effects on some properties of rubber wood ( Hevea brasiliensis muell. Arg.). Bioresources, 7: 189-202.

NAS. 1980. Fire wood crops: Shrub and tree species for energy production. National Academy Press Washington, D.C., USA: National Academy of Sciences. 237 P.

NAS. 1983. Fire wood crops: Shrub and tree species for energy production volume 2. National Academy Press Washington D.C., USA: National Academy of Sciences. 92 P.

- 188 -

Registers

Nasroun, T. H., and Alshahrani, T. S. 1998. The Relationship between anatomical structure and density of wood. Journal of King Saud University, 10 (2): 235 - 251.

Neumann, K., Schoch, W., Détienne, P., Schweingruber, F. H. 2000. Woods of Sahara and the Sahel, an anatomical atlas, Birmensdorf, Eidg. Forschungsanstalt WSL. Bern, Stuttgart, Wien; Haupt. 465 P.

Nicolini, G., Tarchiani, V., Saurer, M., Cherubini, P. 2010. Wood-growth zones in Acacia seyal Delile in the keita valley, Niger: Is there any climatic signal? Journal of Arid Environments, 74: 355 - 359.

Nkaa, F. A., Ogbonnaya, C. I., Onyike, N. B. 2007. Effect of differential irrigation on physical and histochemical properties of kenaf ( Hibiscus cannabinus L. ) grown in the field in Eastern Nigeria. African Journal of Agricultural Research, 2: 252 - 260.

Normand, P. 1950. Atlas des bois de la Côte d`Ivoire. Nogent sur Marne. In : Harzmann L J. 1988. Kurzer Grundriß der allgemeinen Tropenholzkunde. S. Hirzel Verlag, Leipzig. 255 P.

Noshiro, S. and Baas, P. 2000. Latitudinal trends in wood anatomy within species and genera: case in Cornus S.L. ( Cornaceae ). American Journal of Botany, 87: 1495 - 1506.

Noshiro, S. and Suzuki. M. 1995. Ecological wood anatomy of Nepalese Rhododendron (Ericaceae). 2. Intraspecific variation. Journal of Plant Research, 108: 217 - 233.

Ogbonnaya, C. I. 1993. Effects of nitrogen sources on the wood properties of Gmelina arborea relevant to pulp and paper production. Forest Ecology and Management, 56 (1-4): 211 - 223.

Ogbonnaya, C. I., Nwalozie, M. C., Nwaigbo, L. C. 1992. Growth and wood properties of Gmelina arborea (Verbenaceae ) seedlings grown under five soil moisture regimes. American Journal of Botany, 79 (2): 128 - 132.

Ogbonnaya, C. I., Roy-Macauley, H., Nwalozie, M. C., Annerose, D. J. M. 1997. The Physical and histochemical properties of kenaf ( Hibiscus cannabinus L.) grown under water deficit on a sandy soil. Industrial Crops and Products, 7 (1): 9 - 18.

Ogunkunle, A. T. J., and Oladele, F. A. 2008. Structural Dimensions and Paper Making Potentials of Wood in Some Nigerian Species of Ficus L. (Moraceae). Advances in Natural and Applied Sciences, 2(3): 103 - 111.

Ohbayashi, H. and Shiokura, T. 1990. Wood anatomical characteristics and specific gravity of fast growing tropical tree species in relation to growth rates. Mokuzai Gakkaishi 36: 889 - 893.

Ohbayashi, H., and Shiokura, T. 1989. Anatomical structure of fast-growing tropical tree species with differing growth rates. IAWA Bulletin n.s. 10: 342 - 343.

Ohsawa, J. 1988. Vessel picking in printing papers. Paper presented at Tropical Wood Pulp

- 189 -

Registers

Symposium, Singapore, June 21 - 23, pp. 220 - 233.

Ohtani, J., Jing, W., Fukazawa, K. Shao Qun, X. 1989. Multiple perforation plates in Gmelina arborea Roxb. ( Verbenaceae ). IAWA Bulletin n.s. 10: 35 - 41.

Okereke, O. O. 1962. Studies on the fibre dimensions of some Nigerian timbers and raw materials, Part 1. Research Report N 16. Fed. Ministry of Commerce and Industry, Lagos, Nigeria.

Oluwafemi, O. A., and Tunde, E. Z. 2008. Wood quality studies in plantation-grown Sterculia Sterculia setigera Del. in the Guinea Savanna, Nigeria. Research Journal of Forestry 2(1): 22 - 33.

Orwa, C., Mutua, A., Kindt, R., Jamnadass, R., Simons, A. 2009. Agroforestree Database: a tree reference and selection guide version 4.0. Accessed on January 2011, from http://www.worldagroforestry.org/ .

Osman, M. E. 2001. Variation in density and fiber dimensions and the dimarcation of juvenile and mature wood of ten trees species grown in Sudan. M.Sc. Thesis, University of Khartoum, Sudan. 123 p.

Overdieck, D., Ziche, D., Böttcher-Jungclaus, K. 2007. Temperature responses of growth and wood anatomy in European beech saplings grown in different carbon dioxide concentration. Tree Physiology, 27: 261 - 268.

Pande, P. K., Singh, M. 2009. Individual tree, intra- and inter-clonal variations in wood properties of the clonal ramets of Eucalyptus tereticornis Sm. The Indian Forester, 135 (5): 629 - 646.

Pande, P. K., and Singh, M. 2009. Individual tree, intra- and inter-clonal variations in wood properties of the clonal ramets of Eucalyptus tereticornis Sm. Indian Forester, 135(5): 629 - 646.

Pande, P. K., Naithani, S., Kothiyal, V., Mohanta, S. S., Juyal, N., Rawat, R. 2008. Intra and inter tree variations in physico-chemical and wood anatomical properties of Leucaena leucocephala (Lam.) de wit. Indian Forester, 134(5): 622 - 632.

Pande, P. K., and Singh, M. 2005. Intraclonal, inter-clonal and single tree variations of wood anatomical properties and specific gravity of clonal ramets of Dalbergia sissoo Roxb. Wood Science Technology, 39 (5): 351 - 366.

Panshin, A. J., and de Zeeuw, C. 1980. Textbook of Wood Technology 4 th ed. Mc Graw- Hill Book Company. New York. 722 p.

Parameswaran, N., and Conrad, H. 1982. Wood and bark anatomy of Balanites aegyptiaca in relation to ecology and taxonomy. IAWA Bulletin n. s, 3(2): 75 - 88.

Peszlen, I. 1994. Influence of age on selected anatomical properties of Populus clones. IAWA Journal, 15: 311 - 321.

- 190 -

Registers

Petri, R. 1952. Pulping studies with African tropical woods. TAPPI Journal, 35: 157 - 160.

Polge, H. 1963. A new method for determining the texture of wood. The densitometric analysis of radiographics. Annales de l’Ecole Nationale des Eaux et Foréts et de la station de recherches et experiences, 20 (4): 533 - 581.

Purkayastha, S. K., Agrawal, S. P., Tandon, R. D. 1982. Studies on the variation in wood quality of Eucalyptus tereticornis Sm. Trees grown in different plantations. Part I. Wood specific gravity. Indian forest records new series Wood anatomy, 3(1): 1 - 28.

Quilho, T., Miranda, I., Pereira, H. 2006. Within-tree variation in wood fibre biometry and basic density of the urograndis eucalypt hybrid ( Eucalyptus grandis x E.urophylla ). IAWA Journal, 27: 243 - 254.

Ramirez, M., Rodriguez, J., Peredo, M., Valenzuela, S., Mendonca, R. 2009. Wood anatomy and biometric parameters variation of Eucalyptus globulus clones. Wood Science and Technology, 43: 131 - 141.

Rao, R. V., Sujatha, M., Hemawathi T. R. 2003. Radial variation in anatomical properties of plantation grown Tecomella undulata . Journal of Tropical Forest Products, 9 (1,2): 68 - 76.

Roderick, M. L., and Berry, S. L. 2001. Linking wood density with tree growth and environment: a theoretical analysis based on the motion of water. New Phytologist. 149: 473 - 485.

Roig, F. A., Calderón, A., Naves, N., Somoza, A., Lisi, C. S., Fo, M. T. 2008. Poplar Wood Density Assessed by X-Ray Densitometry: New Insights for Inferring Wood Quality. Proceedings of the 51 st International Convention of Society of Wood Science and Technology. November 10-12. Concepción, Chile. pp 181 - 188.

Roque, R. M. 2004. Effect of management treatment and growing regions on wood properties of Gmelina arborea in Costa Rica. New Forests, 28: 325 - 330.

Roque, R. M., and Fo, M. T. 2007. Wood density and fiber dimensions of Gmelina arborea in fast growth trees in Costa Rica: relation to the growth rate. Investigación Agraria: Sistemas y Recursos Forestales, 16: 267 - 276.

Runkel, R. 1942. Über die Herstellung von Zellstoff aus Holz der Gattung Eucalyptus und Versuche mit zwei unterschiedlichen Eucalyptusarten. Das Papier, 3: 476 - 490.

Rydholm, S. A. 1965. Pulping process. In: Wiley and Sons (Eds.), New York, 1270 p.

Saborowski, J., and Hapla, F. 1985. Retrospektive Diskussion der Versuchsplanung für eine Holzqualitätsuntersuchung. Forstarchiv, 56: 118 - 121.

Sadegh, A. N. 2012. Variation of basic density in Eucalyptus camaldulensis dehnh wood grown in Iran. Middle-East Journal of Scientific Research, 11 (10): 1472 - 1474.

Sahni, K. C. 1968. Important trees of the northern Sudan. Khartoum University Press.138 P.

- 191 -

Registers

Sahri, M. H., Ibrahim, F. H., Nor Aini, A. S. 1993. Anatomy of Acacia mangium grown in Malaysia. IAWA Journal, 14 (3): 245 - 251.

Saikia, S. N., Goswami, T., Ali, F. 1997. Evaluation of pulp and paper making characteristics of certain fast growing plants. Wood Science and Technolology , 31 (6): 467 - 475.

Sands, M. J. S. 2001. The Desert Date and its relatives: A revision of the genus Balanites. Kew Bulletin 56: 1 - 128.

Sass, U., and Eckstein, D. 1995. The variability of vessel size in beech ( Fagus sylvatica L.) and its ecophysiological interpretation. Trees, 9: 247 - 252.

Scaramuzzi, G., and Ferrari, G. 1963. Relation of specific gravity to structure in poplar wood. IUFRO sect 41 Melbourne, Australia, 6 PP. In: Zobel, B. J., and Van Buijtenen, J. P. 1989. Wood Variation: Its Causes and Control. Springer, New York, USA. 363 P.

Schniewind, A. P. 1989. Concise encyclopaedia of wood and wood-based materials. Pergamon Press. 248 P.

Schultz, H. R., and Matthew, M. A. 1993. Growth, osmotic adjustment, and cell-wall mechanics of expanding grape leaves during water deficits. Crop Science, 33: 287 - 294.

Schultz, J. 1995. The ecozones of the world: the ecological divisions of the geosphere. Berlin: Springer-Verlag. 449 P.

Schume, H., Grabner, M., Eckmullner, O. 2004. The influence of an altered groundwater regime on vessel properties of hybrid poplar. Trees, 18: 184 - 194.

Schweingruber, F. H. 1983. Der Jahrring: Standort, Methodik, Zeit und Klima in der Dendrochronologie. Bern, Haupt. 234 P.

Schweingruber, F. H. 1988. Tree rings: Basics and Applications of Dendrochronology. Kluwer Academic Publishers, Dordrecht, Netherlands; Boston, Massachusetts, USA. 276 P.

Searson, M. J., Thomas, D.S., Montagu, K.D., Conroy, J. P. 2004. Wood density and anatomy of water-limited eucalypts. Tree Physiology, 24: 1295 - 1302.

Settle, D. J., Page, T., Bush, D., Doran, J., Sethy, M., Viji, I. 2012. Basic Density, Diameter and Radial Variation of Vanuatu Whitewood ( Endospermum medullosum ): Potential for Breeding in a Low Density, Tropical Hardwood. International Forestry Review, 14 (4): 463 - 475.

Sharma, S. K., Rao, R. V., Shukla, S. R., Kumar, P., Sudheendra, R., Sujatha, M., Dubey, Y. M. 2005. Wood quality of coppiced Eucalyptus tereticornis for value addition. IAWA Journal, 26(1): 137 - 147.

- 192 -

Registers

Shashikala, S. and Rao, R. V. 2005. Variation in clonal wood quality of Eucalyptus tereticornis and its role in paper and pulp production. Journal of Indian Academy of Wood Science, 2 (2): 14 - 30.

Shawgi, H. M. 2007. Variation in wood fiber characteristics among hardwood species growing in low-rainfall woodland savanna (Sudan). M.Sc. Thesis, Faculty of Forestry, University of Khartoum, Sudan. 95 p.

Shepard, R. K., and Shottafer, J. E. 1992. Specific-gravity and mechanical property-age relationships in red pine. Forest Products Journal, 42: 60 - 66.

Simpson, W., and TenWolde, A. 1999. Physical properties and moisture relations of wood. In: Wood handbook: wood as an engineering material. Madison, WI: USDA Forest Service, Forest Products Laboratory, 1999. General technical report FPL; GTR-113: Pages 3.1 - 3.24.

Suliman, A., M., and Jackson, J., K. 1959. The Hegleig Tree. Sudan Silva 9 (1): 1 - 4.

Swaminathan C., Vijendra Rao, R., Shashikala, S. 2012. Preliminary evaluation of variations in anatomical properties of Melia dubia Cav. wood. International Research Journal of Biological Sciences, 1(4): 1 - 6.

Tarhule, A., Hughes, M., 2002. Tree-ring research in semi-arid West Africa: need and potential. Tree-Ring Research, 58: 31- 46.

Tavares, F., Quilhó, T., Pereira, H. 2011. Wood and bark fiber characteristics of Acacia melanoxylon and comparison to Eucalyptus globules . Cerne, Lavras, 17(1): 61 - 68.

Taylor, F. W. 1968. Specific gravity differences within and among yellow-poplar trees. Forest Products Journal, 18: 75 - 81.

Taylor, F. W. 1969. Variation of wood properties in pecan. For Prod Utiliz Lab Mississippi State Univ Res Rep 7, 18 PP. In: Zobel, B. J., and Van Buijtenen, J. P. 1989. Wood Variation: Its Causes and Control. Springer, New York, USA. 363 P.

Taylor, F. W. 1973. Variation in the anatomical properties of South African grown Eucalyptus grandis . Appita Journal, 27(3): 171 - 184.

Thirakul, S. 1984. Manual of Dendrology, Bahr el Ghazal and Central Regions. Forest Inventory Project Areas. CIDA, Groupe Poulin, Theriault Ltee Consultants, Quebec, Canada. 472 p.

Thomas, D. S., Montagu, K. D., Conroy, J. P. 2004. Changes in wood density of Eucalyptus camaldulensis due to temperature - the physiological link between water viscosity and wood anatomy. Forest Ecology and Management, 193: 157 - 165.

Thomas, D. S., Montagu, K. D., Conroy, J. P. 2007. Temperature effects on wood anatomy, wood density, photosynthesis and biomass partitioning of Eucalyptus grandis seedlings. Tree Physiology, 27: 251 - 260.

- 193 -

Registers

Thorbjornsen, E. 1961. Variation in density and fiber length in wood of yellow poplar. TAPPI Journal, 44(3): 192 - 95.

Tissot, M. 1985. Papeterie Tropicale. Nogent-sur-Marne: CTFT.

TROPIX 7. 2013. The main technological characteristics of 245 tropical wood species. Developed by CIRAD. Available at: http://tropix.cirad.fr/temperate/CHENE.pdf . (Accessed on December 2013).

Tsehaye, A., Buchanan A. H.,Walker, J. C. F. 1995. A comparison of density and stiffness for predicting wood quality or density: The lazy man's guide to wood quality. Journal of the Institute of Wood Science, 13(6): 539 - 543.

Tsoumis, G. 1991. Science and technology of wood: Structure, properties, utilization. Van Nostrand Reinhold, New York. 494 P.

Tsuchiya, R., and Furukawa, I. 2009. Radial variation in vessel lumen diameter in relation to stem increment in 30 hardwood species. IAWA Journal, 30 (3): 331 - 342.

Uniyal, S. 2012. Variation in wood anatomical properties and specific gravity in Eucalyptus tereticornis Sm. PhD. thesis. Forest Research Institute University Dehradun, Uttarakhand. India. 143 p.

Valente, C.A., Sousa, A. P. M., Furtado, F. P., Carvalho, A. P. de. 1992. Improvement program for Eucalyptus globulus at PORTUCEL: Technological component. Appita Journal, 45 (6): 403 - 407.

Valkomer, P. J. 1969. Wood raw materials for pulp paper in tropical countries. In: FAO International Review of Forestry and Forest Products; Unasylva 23 (3).

Van den Oever, L., Baas, P. and Zandee, M. 1981. Comparative wood anatomy of Symplocos and latitude and altitude of provenance . IAWA Bulletin. n. s . 2: 3 - 24.

Van der Walt, J. J. A., Werker, E., Fahn, A. 1988. Wood anatomy of the Pelargouium (Geraniaceae ). IAWA Bulletin n.s. 10: 201 - 207.

Veenin, T., Fujita, M., Nobuchi, T., Siripatanadilok, S. 2005. Radial variations of anatomical characteristics and specific gravity in Eucalyptus camaldulensis clones. IAWA Journal, 26: 353 - 361.

Ververis, C., Georghiou, K., Christodoulakis, N., Santas, P. Santas, R. 2004. Fiber dimensions, lignin and cellulose content of various plant materials and their suitability for paper production. Industrial Crops and Products, 19: 245 - 254.

Vogt, K. 1995. A Field Guide to the Identification, Propagation and Uses of the Common Trees and Shrubs of Dryland Sudan. SOS Sahel International (UK), London. 309 p.

Von Maydell, H. 1990. Trees and shrubs of the Sahel, their characteristics and uses. Weikersheim, Germany: GTZ. Verlag Josef Margraf Scientific Books.

- 194 -

Registers

Wagenführ, R. 1980. Anatomie des Holzes. Fachbuchverlag Leipzig: 377 p.

Wagenführ, R. 1989. Anatomie des Holzes unter besonderer Berücksichtigung der Holztechnik. VEB, Fachbuchverlag Leipzig. 334 P.

Wagenführ R. 2007. Holzatlas. München: Fachbuchverl. Leipzig im Carl Hanser Verlag. 816 p.

Walker, J. C. F., and Butterfield, B. G. 1996. The importance of the microfibril angle for the processing industry. New Zealand Journal of Forestry, 40 (4): 34 - 40.

Wangaard, F. F. 1962. Contributions of hardwood fibers to the properties of kraft pulps. TAPPI Journal, 45: 548 - 556.

Watson, A. J., and Dadswell, H. E. 1961. Influence of fiber morphology on paper properties. 1. Fiber length . Appita Journal, 14: 168 - 78.

Wickens, G. E., Seif El Din, A. G., Nahal, I. 1995. Role of acacia species in the rural economy of dry Africa and the Near East. FAO Conservation Guide 27. Rome: FAO. 137 P.

Wiedenhoeft, A. 2010. Structure and Function of Wood, Chapter 3, Wood handbook - Wood as an engineering material, General Technical Report FPL-GTR-190, Madison, WI: U.S., Department of Agriculture, Forest Service, Forest Products Laboratory, 508 p.

Wiemann, M. C, and Green, D. W. 2007. Estimation Janka hardness from specific gravity for tropical and temperate species. United States Department of Agriculture, Forest Service, Forest Products Laboratory, research paper FPL-RP-643. 21 p.

Wiemann, M., and Willianson, G. B. 2002. Geographic variation in wood specific gravity: effects of latitude, temperature and precipitation. Wood Fiber Science, 34: 96 - 107.

Wilkes J., and Abbott D. 1983. Influence of the rate of tree growth on anatomy of eucalypt species. Appita Journal, 37(3): 231 - 323.

Wilkes, J. 1987. Review of the significance of variations in wood structure in the utilisation of Pinus radiata. Australian forest research, 17: 215 - 232.

Wilkins, A. P., and Papassotiriou, S., 1989. Wood anatomical variation of Acacia melanoxylon in relation to latitude. IAWA Bulletin n.s, 10: 201 - 207.

Wimmer, R., Downes, G., Evans, R. 2002. High-resolution analysis of radial growth and wood density in Eucalyptus nitens , grown under different irrigation regimes. Annals of Forest Science, 59: 519 - 524.

Worbes, M .1995. How to measure growth dynamics in tropical trees - a review. IAWA Journal, 16: 337 - 351.

- 195 -

Registers

Worbes, M. 1999. Annual growth rings, rainfall dependent growth and long-term growth patterns of tropical trees from the Forest Reserve Caparo in Venezuela. Journal of Ecology, 87: 391 - 403.

Worbes, M. 2002. One hundred years of tree-ring research in the tropics-a brief history and an outlook to future challenges. Dendrochronologia, 20: 217 - 231.

Xiaomei, J., Yafang, Y., Urakami, H. 2003. Variation within tree of wood anatomical properties and basic density of 1-214 poplar in beijing area and their relationship modelling equations. Scientia Silvae Sinicae, 39 (6): 115 - 121.

Xu, F., Zhong, X. C., Sun, R. C., Lu, Q. 2006. Anatomy, ultra structure, and lignin distribution in cell wall of Caragana korshinskii . Industrial Crops and Products, 24: 186 - 193.

Yahya, R., Sugiyama, J., Silsia, D., Gril, J. 2010. Some anatomical features of an Acacia hybrid, A.mangium and A. auriculiformis grown in Indonesia with regard to pulp yield and paper strength. Journal of Tropical Forest Science, 22(3): 343 - 351.

Yang, J. L. and Evans, R. 2003. Prediction of MOE of eucalypt wood from microfibril angle and density. Holz als Roh- und Werkstoff, 61: 449 - 452.

Ylinen, A. 1943. Über den Einfluß der Rohwichte und des Spätholzanteils auf die Brinellhärte des Holzes. Holz als Roh- und Werkstoff, 6: 125 - 127.

Yousif, S. M. O. 2000. Comparative wood anatomy of some Acacia Species in Blue Nile region, Central Sudan. M.Sc. Thesis, University of Khartoum, Sudan. 161 P.

Zeidler, A. 2012. Variation of wood density in Turkish hazel ( Corylus colurna L.) grown in the Czech Republic. Journal of Forest Science, 58(4): 145 - 151.

Zhang, S. Y., and Morgenstern, E. K. 1995. Genetic variation and inheritance of wood density in black spruce ( Picea mariana ) and its relationship with growth; implications for tree breeding. Wood Science and Technology , 30: 63 - 75.

Zhang, S.Y. 1995. Effect of growth-rate on wood specific-gravity and selected mechanical- properties in individual- species from distinct wood categories. Wood Science and Technology, 29: 451 - 465.

Zhang, X., Deng, L., Baas, P. 1988. The ecological wood anatomy of the Lilacs ( Syringa oblata var. giraldii ) on Mount Taibei in North-western China. IAWA Bulletin. n. s. 9: 24 - 30.

Zimmermann, M. H. 1983. Xylem structure and the ascent of sap. Springer-Verlag, Berlin. 284 p.

Zobel, B. J. and Talbert, J. T. 1984. Applied Forest Tree Improvement. John Wiley & Sons, New York. 505 P.

- 196 -

Registers

Zobel, B. J., and Van Buijtenen, J. P. 1989. Wood Variation: Its Causes and Control. Springer, New York, USA. 363 p.

Zziwa, A., Kaboggoza, J. R. S., Mwakali, J. A., Banana, A. Y., Kyeyune, R. K. 2006. Physical and mechanical properties of some less utilised tropical timber tree species growing in Uganda. Uganda Journal of Agricultural Sciences, 12 (1): 29 - 37.

Internet sources:

SOURCE 1: http://www.atozmapsdata.com/zoomify.asp?name=Country/Modern/Z_Sudan_Precip Accessed on February 2012. SOURCE 2: http://upload.wikimedia.org/wikipedia/commons/c/c6/Sudan_location_map_Topographic.png Accessed on February 2012.

SOURCE 3: http://www.google.de/imgres?imgurl=http://upload.wikimedia.org/wikipedia/commons/4/42/ Sudan_sat.jpg&imgrefurl=http://commons.wikimedia.org/wiki/Atlas_of_Sudan&h=1787&w= 2058&sz=978&tbnid=biTGjbQRWONFWM:&tbnh=90&tbnw=104&zoom=1&usg=__1dHH y0X7LxLvqt8yhqYw8WkG_A=&docid=aACgXRlGvfVt7M&sa=X&ei=gGSjUsm7Gs7EtAa zqYGADg&ved=0CFcQ9QEwBg&dur=455ource: modified from ??? http://upload.wikimedia.org/wikipedia/commons/c/c6/Sudan_location_map_Topographic.png Accessed on March 2012. SOURCE 4: http://www.paperonline.org/paper-making/paper-production/pulping/pulping-properties-of- hardwoods-and-softwood Accessed on January 2011.

Total number of sources: 305

- 197 -

Registers

8.2 Figures

Figure 1.1: Sudan forest cover. Adapted by the author from DAWELBAIT et al. (2006) .… 1 Figure 1.2: Sudan precipitation map. Adapted by the author from SOURCE1……………... 2

------

Figure 2.3: Relationship between Janka side hardness and basic specific gravity for greenwood (WIEMANN and GREEN 2007) ……....…...... ………………...... 22

Figure 2.4: Relationship between Brinell hardness (end and side) and specific gravity. Measurement by MÖRATH (1932). Source: KOLLMAN and CÔTÉ (1968)...... 22

Figure 2.5: General features of Acacia seyal var. seyal. 1. Adult tree; 2. Powdery red bark; 3.Leaves, fruits and flowers; and 4. Wood (cross section) ...... …...... …….…...... 42

Figure 2.6: Acacia seyal var. seyal’s distribution over the world (after ORWA et al . 2009). 45

Figure 2.7: Acacia seyal var. seyal’s distribution in Sudan. Adapted by the author from SOURCE 2 ...... 46

Figure 2.8: General features of Balanites aegyptiaca. 1. Adult tree; 2. Fissured bark; 3. Leaves and flowers; 4. Green immature fruit and yellow mature fruit, the photo shows also the thorns; 5. Wood (cross section showing the fluted stem)...... 48

Figure 2.9: Balan ites aegyptiaca ’s distribution over the world (after ORWA et al. 2009)... 52

Figure 2.10: Balanites aegyptiaca ’s distribution in Sudan. Adapted by the author from SOURCE 2………….…...... …………………………………………………...... 53

------

- 198 -

Registers

Figure 3.2: Climatic diagram (2000-2009) for Al Obeid locality, Northe Kordofan ……... 56

Figure 3.3: Climatic diagram (2000-2009) for Al Duwaim locality, White Nile ……….… 56

Figure 3.4: Different forests in zone one .1, 2. Al Tainat Forest, 3. Al Ein Forest in Al Obeid, North Kordofan State, and 4. AL Homora Forest, 5. Goz Fagor Forest, 6. Aum Top Forest in Al Duwaim, White Nile State. ……...... …….....…………….... 57

Figure 3.5: Climatic diagram (2000-2009) for Kadugli Locality, South Kordofan ...... 58

Figure 3.6: Climatic diagram (2000-2009) for Al Damazien locality, Blue Nile ...... 58

Figure 3.7: Different forests in zone two. 1. Al Bardab Forest; 2. Al Sheheata Forest; 3. Al Homara Forest in Kadugli, South Kordofan State, and 4. Tawla Forest; 5. Khor Donia Forest in Al Damazien, Blue Nile State...... 60

------

Figure 4.1: Acacia seyal var. seyal samples’s debarking and storage. 1, 2. Samples debarking, 3, 4. Samples wrap by plastic bags, 5.Samples storage and covering by Azadirachta indica leaves ...... 64

Figure 4.2: Samples processing. 1, 2, cutting the discs edges using circular saw 3. reduction of samples thickness using planer; 4,5 final samples obtained samples (strips) …...... 65

Figure 4.3: The study species logs processing flowchart (Acacia = Acacia seyal ; Balanites = Balanites aegyptiaca ; C = City and F = Forest) ...... 66

Figure 4.4: Tree sampling ...... 70

Figure 4.5: Fibre length measurement using measuring scale ………………………….….. 72

Figure 4.6: The used microtomes: 1.GSL Microtome; 2. Freezing Microtome (REICHERT-JUNG 1206) …………………..…………………………………….... 73

Figure 4.7: Microscopic slides including cross sections divided into four equal quarters .....74

Figure 4.8: Fibres and vessel’s diameter and lumen diameter measurements using ImageJ software ………………………………….………………………...... ………….….. 74

Figure 4.10: Correction factor calculation .……………………………………...... 80

- 199 -

Registers

------

Figure 5.5: The study species fibre length box plots ……….……………………………… 97

Figure 5.6: The study species fibre diameter and lumen diameter box plots. Values for 10 % and 90 % radial distance (D = diameter and LD = lumen diameter) ……... 97

Figure 5.7: The study species fibre wall thickness box plots. Values for 10 % and 90 % radial distance ……………………………………………..………………….. 98

Figure 5.8: The study species Runkel ratio box plots. Values for 10 % and 90 % radial distance …………………...…………………………………………………………. 97

Figure 5.9: The study species flexibility coefficient box plots. Values for 10 % and 90 % radial distance …………….………………………………………………………… 98

Figure 5.10: Acacia seyal fibre length radial variation trend by tree ……………………. . 101

Figure 5.11: Balanites aegyptiaca fibre length radial variation trend by tree …………… 102

- 200 -

Registers

Figure 5.12: Acacia seyal fibre length radial variation trend. mean ± Std (bars with different letters are significantly different from each other at 0.05 probability level) ...... 104

Figure 5.13: Balanites aegyptiaca fibre length radial variation trend. mean ± Std (bars with different letters are significantly different from each other at 0.05 probability level) …………………………………………………….………...... 104

Figure 5.14: The study species vessels diameter and lumen diameter box plots. Values for 10 % and 90 % radial distance (D = diameter and LD = lumen diameter) …….…………………………………………………………………..... 116

Figure 5.15: The study species vessels wall thickness box plots. Values for 10 % and 90 % radial distance …………………..……………………………………..…….. 116

Figure 5.16: Acacia seyal basic density vertical variation pox plots (mean of 27 trees) … 121

Figure 5.17: Balanites aegyptiaca basic density vertical variation pox plot (mean of 30 trees) …………………………….…………...……………………… 121

Figure 5.18: The study species basic density radial variation box plots…………………... 123

Figure 5.19: Acacia seyal basic density radial variation trend by tree. (Trees 6, 8 and 9 in zone one were excluded due to their abnormal density values) ...... 124

Figure 5.20: Balanites aegyptiaca basic density radial variation trend by tree ……….….. 125

Figure 5.21: Acacia seyal basic density radial variation trend. Mean ± Std (bars with different letters are significantly different from each other at 0.05 probability level) ……………………………………………………………………………….126

Figure 5.22: Balanites aegyptiaca basic density radial variation trend. Mean ± Std (bars with different letters are significantly different from each other at 0.05 probability level) …………….…………………………………………………..... 126

Figure 5.23: Acacia seyal air dry density vertical variation box plots ………………….... 132

Figure 5.24: Balanites aegyptiaca air dry density vertical variation box plots…………… 132

Figure 5.25: The study species air dry density radial variation box plots (plots with different letters within species are significantly different from each other at 0.05 probability level)….……..………………………………………….……….... 133

Figure 5.26: The study species X-ray density radial variation box plots ……………….… 136

Figure 5.27: The study species air dry density vs. X-ray density box plots (means with* are significantly different from each other at 0.05 probability level)…...... ……… 139

Figure 5.28: Acacia seyal air dry density vs. X-ray density box plots means with * are significantly different from each other at 0.05 probability level)………………… 140

- 201 -

Registers

Figure 5.29: Balanites aegyptiaca air dry density vs. X-ray density box plots (means with * are significantly different from each other at 0.05 probability level) ..…...... 140

Figure 5.30: The study species hardness strength vertical variation box plots in transverse section, P1 = inner measured points and P2 = outer measured points …. 142

Figure 5.31: The study species hardness strength vertical variation box plots in radial section ……………..…………………………………………………………….. 142

Figure 5.32: The study species hardness strength radial variation box plots in transverse section ……………………...……………………………………………………… 143

Figure 5.33: The relationship between the study species Brinell hardness strength and air dry density ...... 146

Figure 5.34: The effect of rainfall on Acacia seyal anatomical properties ...... 152

Figure 5.35: The effect of rainfall on Balanites aegyptiaca anatomical properties ...... 152

Total number of figures: 66

- 202 -

Registers

8.3 Tables

Table 2.1: Fibre length (µm) classifications ………………………………………………… 7

Table 2.2: Examples for tropical hardwoods with short fibres (HARZMANN 1988) ……… 7

Table 2.3: Examples for tropical hardwoods with long fibres (HARZMANN 1988) ………. 8

Table 2.4: Examples for tropical hardwoods with thin fibre walls (HARZMANN 1988) ….. 8

Table 2.5: Examples for tropical hardwoods with thick fibre walls (HARZMANN 1988)..... 9

Table 2.6: Gmelina arborea wood density and fibres characteristics (OGUNKUNLE and OLADELE 2008) ...... 10

Table 2.7: Wood fibre characteristics of some species grown in Sudan ………………….... 10

Table 2.8: Examples for tropical hardwoods with small vessel diameters (<100 µm) (HARZMANN 1988) ………………………………………………………………. 14

Table 2.9: Examples for tropical hardwoods with large vessel diameters (>200 µm) (HARZMANN 1988) ...... 15

Table 2.10: Definitions of wood density ...... 16

Table 2.11: Some tropical hardwoods with low wood density (KANEHIRA 1933) ...... 17

Table 2.12: Some tropical hardwoods with high wood density (after ATIBT 1965 and FORS 1965) ……………………………..………………………………….….. 18

Table 2.13: Definitions of hardness tests applied for wood ...... 20

Table 2.14: Hardness classifications ...... 21

Table 2.15: Air dry density and Brinell hardness strength of Milletia laurentii and Quercus rubra …………………………………………………….……………...... 21

Table 2.16: Frequency of cell types in some tropical hardwoods (HARZMANN 1988) ….. 24

Table 2.17: The effect of water stress on some vessel characteristics ……………………... 36

Table 2.18: Fibre characteristics of Acacia seyal var . seyal ...... 44

Table 2.19: Fibre characteristics of Balanites aegyptiaca ...... 51

------

Table 3.1: Mean Annual Rainfall (mm) and Temperature (ºC) for ten years (2000-2009) and their average in the study locations R = Mean annual rainfall (mm); T =

- 203 -

Registers

Mean annual temperature (ºC); N.K = North Kordofan State; W.N = White Nile State; B.N = Blue Nile State; S.K = South Kordofan State and Loc. = Location...... 55

Table 3.2: Description of tree samples of Acacia seyal var . seyal and its growing conditions ...... 62

Table 3.3: Description of tree samples of Balanites aegyptiaca and its growing conditions. 63

------

Table 4.1: Acacia seyal selected trees for anatomical properties investigations…………… 67

Table 4.2: Balanites aegyptiaca selected trees for anatomical properties investigations ….. 68 Table 4.3: The samples position of each wood property along the radial and vertical axis…71

Table 4.4: Possible sources of error in X-ray densitometry analysis ……………….……… 81

------

Table 5.1: Anatomical composition description of Acacia seyal in comparison with literature. ………………………………………...…………………………………. 89

Table 5.2: Anatomical composition description of Balanites aegyptiaca in comparison with literature ...………………………………………………………..…………… 93

Table 5.3: Some hardwood species with fibres characteristics radial increase pattern ...…. 100

Table 5.4: Some hardwood species with fibre length radial increase pattern …………….. 103

Table 5.5: Acacia seyal fibres characteristics variation (without fibre length) (ANOVA/ T-test)…………………………………………………………..……….………….. 106

Table 5.6: Acacia seyal fibres length variation (ANOVA/T-test) ………………………... 107

Table 5.7: Balanites aegyptiaca fibres characteristics variation (without fibre length) (ANOVA/T-test)……...………………………………………………………….... 108

Table 5.8: Balanites aegyptiaca fibres length variation (ANOVA/T-test) ……………….. 108

Table 5.9: The study species fibres characteristics. Mean ± Std (min.-max.), *Slenderness values doesn’t include Std., Min., and Max.values because it is calculated from the mean values of fibre length and diameter ..……….…...…….. 109

Table 5.11: Acacia seyel fibre characteristics in comparison with literature ………….….. 111

- 204 -

Registers

Table 5.12: Balanites aegyptiaca fibre characteristics in comparison with literature ….… 112

Table 5.13: Acacia seyel fibres characteristics of tree growing in Blue Nile state in comparison with literature ……….…………….………………………….………. 113

Table 5.14: Balanites aegyptiaca fibres characteristics of tree growing in Blue Nile state in comparison with literature …….…………………………………………….…. 114

Table 5.15: The study species fibres characteristics of tree growing in White Nile state in comparison with literature …………..……...………………………………..….. 114

Table 5.16: Balanites aegyptiaca fibres characteristics of tree growing in South Kordofan state in comparison with literature .……………………………………. 115

Table 5.17: Some hardwood species with vessels dimensions radial increase pattern ….... 117

Table 5.18: Acacia seyal vessels characteristics variation (ANOVA/T-test) …………….. 118

Table 5.19: Balanites aegyptiaca vessels characteristics variation (ANOVA/T-test) …… 118

Table 5.20: The study species vessels characteristics. Mean ± Std (min.-max.) …………. 119

Table 5.21: The study species inner and outer wood vessels characteristics. Mean ± Std (min.-max.) …………...…………………………………………………………..... 119

Table 5.22: Some hardwood species with wood density vertical decline pattern …….…... 122

Table 5.23: Some hardwood species with wood density radial increase pattern …………. 128

Table 5.24: Acacia seyal basic density variation (ANOVA/T-test) ….……………....……129

Table 5.25: Balanites aegyptiaca basic density variation (ANOVA/T-test) …..…………. 129

Table 5.26: The study species basic density. Mean ± Std (min.-max.) …………………... 130

Table 5.27: Acacia seyal air dry density variation (ANOVA/T-test) ……………….……. 134

Table 5.28: Balanites aegyptiaca air dry density variation (ANOVA/T-test) ……………. 134

Table 5.29: The study species air dry density. Mean ± Std (min.-max.) …………………. 134

Table 5.30: Acacia seyal X-ray density variation (ANOVA/T-test) ……………………… 135

Table 5.31: Balanites aegyptiaca X-ray density variation (ANOVA/T-test) …………….. 138

Table 5.32: The study species X-ray density. Mean ± Std (min.-max.) ………………….. 138

Table 5.33: Acacia seyal hardness strength variation (ANOVA/T-test) ………….…...…. 144

Table 5.34: Balanites aegyptiaca hardness strength variation (ANOVA/T-test) ………… 144

Table 5.35: The study species hardness strength (N/mm 2). Mean ± Std (min.-max.) ……. 145

- 205 -

Registers

Table 5.36: Results of independent sample T-test for the study species fibre length …….. 147

Table 5.37: Results of independent sample T-test for Acacia seyal fibres characteristics…148

Table 5.38: Results of independent sample T-test for Balanites aegyptiaca fibres characteristics …………….………………………………………………………... 148

Table 5.39: Results of independent sample T-test for Acacia seyal vessels characteristics..150

Table 5.40: Results of independent sample T-test for Balanites aegyptiaca vessels characteristics …………………………………………………………………….... 150

Table 5.41: Results of independent sample T-test for the study species wood basic density …………..………………………………………………………………..... 153

Table 5.42: Independent sample T-test for the study species hardness strength (HB) …… 154

Table 5.43: Summary of the effect of water stress on the studied wood properties ……… 155

Table 5.45: The study species juvenile and mature wood properties compared with the acceptable values for PPM. Mean (min.-max.), Slenderness values doesn’t include min., max. Because it is calculated from the mean values of fibres length and diameter ………………...... 159

Table 5.46: The study species wood properties in each climatic zone compared with the acceptable values for PPM ………………………………………………………... 160

Table 5.47: The study species air dry density and hardness strength (HB) values compared with the reference species. Mean (min.- max.) …..…………..……….. 162

Table 5.48: The study species juvenile (JUV.) and mature (MAT.) wood air dry density and hardness strength (HB) in transverse section compared with the reference species …………………………………...……………………………... 163

Table 5.49: The study species air dry density and hardness strength (HB) in climatic each zone compared with the reference species .………………………...………... 163

Total number of tables: 75

- 206 -

Registers

8.4 Definitions of selected terms in wood science

Air-dried: Wood dried by exposure to air, without artificial heat; 12 % moisture content. Bark: Collected tissues of the cylindrical tree axis outside the vascular cambium. Bounded water: Moisture contained within cell walls and held by hygroscopic forces. Cell wall ultrastructure: Sub-microscopic organization of wood cell walls (super-molecular morphology). Cellulose: The principal polysaccharide (C6H10O5)n of higher plant cell walls; yielding only glucose on decomposition (1-4-ß-D glucosan). Cross secton: Section cut at right angles to the major tree axis (transverse section, end- grain). Earlywood: That portion of growth zones produced at the beginning of the growing season (springwood). Fibre: An elongated (1.0-1.5 mm) pored-wood cell with pointed ends, thickened walls and with/without functional pits; as "fibre tracheid" or "libriform fibre". Fibre tracheids: Cells generally long with pointed extremities. Bordered pits with oval to slit-shaped apertures. Bordered pits most frequent in tangential section. Hardwood: Wood produced by broad-leaved trees (angiosperm dicotyledons) such as oak, elm, ash; same as pored woods. Heartwood: Inner dead core of the woody tree stem. Hemicellulose(S): Non-cellulosic cell wall polysaccharides; easily decomposed yielding several different simple sugars. Increment zone: Xylary sheath or shell over the entire branched, cylindrical axis resulting from periodic growth (growth zone, growth ring, annual increment) Juvenile wood: Physiologically young wood close to the tree pith (core wood, crown wood, pith wood). Latewood: That portion of growth zones produced during the latter part of the growing season (summerwood). Libriform fibres: Similar to fibre tracheids, however, simple pits with narrow, slit-like, oblique orientated apertures. Pits mostly absent in tangential section. Lignin(s): Major non-carbohydrate fraction of wood; an irregular polymer of substituted propylphenol groups. Longitudinal parenchyma: Parenchyma cell derived from fusiform cambial initial; main axis along the grain. Lumen: The cell cavity.

- 207 -

Registers

Mature wood: Wood produced by cambial initials which have reached maximal dimension (outer wood, adult wood). Microfibril angle (micelle angle, θθθ): Measure of crystallite orientation with the wood cell axis. Modulus of Elasticity (MOE): Rate change of wood strain ( ε) as a function of stress ( σ) within the proportional limit; indicates material stiffness. Modulus of Rupture (MOR): Maximum fibre stress at failure in bending. Moisture content: Amount of water (%) contained in wood expressed as percentage of oven- dry wood weight. Oven-dry: Wood moisture condition attained by prolonged heating at 100°C ± 3°C. Parenchyma: Relatively thin-walled cells retaining contents and participating in wood storage and protective functions. Pit: Recess in wood secondary cell wall, together with external closing membrane and opening internally to the cell lumen; in contrast to (pored wood) perforation between two vessel members, that has no closing membrane. Pores: Cross-section of vessel; a vessel as it appears on a transverse surface or in a transverse section of wood. Radial section: Section cut along the grain parallel to wood rays and perpendicular to growth zones. Ray parenchyma: Parenchyma cell derived from ray cambial initial; main axis across the grain (coniferous woods) or across/along the grain (pored-woods). Sapwood: Outer partly live portion of the woody tree stem. Softwood: Wood produced by coniferous trees (gymnosperms) such as pine, spruce, redwood; same as non-pored wood. Tangential section: Section cut along the grain perpendicular to wood rays and tangent to growth zones. Vessel: Tubelike structure in pored-woods formed through fusion of often large diameter vessel elements and with perforation plates (openings) in common end walls; the cross-section of a vessel is a pore.

- 208 -

Registers

8.5 Abbreviations

ANOVA Analysis of Variance

ASTM American Standard Testing Method

C.-America Central-America cm Centimetre

DBH Diameter at breast height (cm) df Degree of freedom

E-Afica East-Africa g/cm 3 Grams per cubic centimetre

Kg/m 3 kilograms per cubic meter kV Kilovolt

L.-America Latin America m Meter mAs Milli-Ampere-Sekunden mm Millimeter

MOE Modulus of Elasticity

MOR Modulus of Rupture N/mm 2 Newton per square millimeter

PPM Pulp and paper making

S-America South America

SE-Asia South-East-Asia

Sig. Significance

SPSS Statistical Package for the Social Sciences

TU Dresden Technische Universität Dresden

W-Africa West Africa

µm Micrometer ( 0,001 mm)

- 209 -

Appendices

9 Appendices

The attached CD containing all the Appendices mentioned in results and discussion chapter in PDF file namely Appendices. It inculeds a total of 44 Appendices.

- 210 -

Declaration

I hereby certify that this thesis entitled “ Investigation of selected wood properties and the suitability for industrial utilization of Acacia seyal var. seyal Del. and Balanites aegyptiaca (L.) Delile grown in different climatic zones of Sudan ” is my own work and that it has not been submitted anywhere for the award of any other academic degree. Where other sources of information have been used, they have been duly acknowledged in the text.

Tharandt, Germany. Signature: Date: 07.02.2014 Name: Hanadi Mohamed Shawgi Gamal

- 211 -