UNIVERSITY OF GHANA, LEGON.

ASSESSING THE GROWTH, NODULATION 
AND NITROGEN FIXING POTENTIAL OF 
SOME MULTIPURPOSE TREES AND SHRUBS

BOAKYE EMMANUEL YAW

THESIS SUBMITTED TO THE DEPARTMENT OF 
SOIL SCIENCE OF UNIVERSITY OF GHANA IN 
PARTIAL FULFILMENT OF THE REQUIREMENTS 
FOR THE AWARD OF MASTER OF PHILOSOPHY 
DEGREE IN SOIL SCIENCE.

BY

MAY 2001



1g>', C » I



D E D I C A T I O N

I dedicate this work to Apostle Dr. Kwadwo Safo and my parents, 
Mr. and Mrs. A. K. Boakye.



DECLARATION

I hereby declare that except for references to the literature and to the 

work o f other researchers, which have been duly cited herein, this 

thesis is the result o f my own original research and that no part o f  it 

has been presented for another degree in this university or elsewhere.

BOAKYE EMMANUEL YAW 
(STUDENT)

I PROF. S. K & D A N S O  
' (PRINCIPAL SUPERVISOR)

HEAD OF DEPARTMENT 
(SOIL SCIENCE)



ACKNOWLEDGEMENT

I would first and foremost thank the Almighty God without whose 
abundant grace and providence this work would never have been 
successfully completed. To Him be all the glory.

I also wish to express my profound gratitude to Apostle Dr. Kwadwo 
Safo for giving me the needed encouragement as well as financial 
assistance in times o f need. Special thanks also go to the management 
o f ADB for awarding me ADB fellowship without which completion 
o f this work would have been extremely difficult.

I also owe a debt o f gratitude and admiration to my Principal 
Supervisor Prof. S. K. A. Danso for his diligence, guidance, 
encouragement and skilful manner in which he helped to organise and 
shape my thoughts. I particularly appreciate his strictness and 
criticisms. May God bless him exceedingly.

Special thanks also go to Ecolab and its entire staff for providing me 
the needed material support for the completion o f the project.

Finally, I would like to extend my warm thanks to all my colleagues 
and friends who in diverse ways helped for the successful completion 
o f the work, I say “Ayikoo”.



ABSTRACT

This study was conducted to assess the growth, nodulation and nitrogen fixing 

potential of some native multipurpose trees and shrubs.

A total of fourteen species of indigenous leguminous trees and shrubs were 

initially screened for nodulation in three Ghanaian soils (Toje, Hatso and Alajo 

soil series), after 8 weeks of growth in nursery bags. The experiment was repeated 

with nine species of the tree/shrub legumes, this time under two levels of 

phosphorus (0 and 60 mg P/kg soil). The final study involved four of the 

tree/shrub legume species, which were assessed for both inoculation and 

phosphorus responses, and abilities to fix nitrogen as assessed by the 1 Isotope 

dilution method.

Nodulation of the tree/shrub legumes by native rhizobia was highly variable, with 

five of the tree/shrub species, namely, Albizia lebbek, Sesbania aculiata, 

Pithecelobium spp, Tephrosia spp and Acacia farresiana being nodulated by 

native rhizobia in all three soils. In contrast, the following six tree legumes, 

Acacia adianthifolia, Albizia zygia, Acacia mangium, Senna occidentalis, Cassia 

occidentalis and Tamarindus indica did not form any nodule in any of the three 

soils. The presence of indigenous Rhizobium in all three soils, capable of 

nodulating A. lebbek, Pithecelobium spp, and Tephrosia spp was confirmed in a 

final study with four leguminous trees and shrubs; unioculated Leucaena spp 

however nodulated only in Toje soil. The number of rhizobia counted by the most 

probable number (MPN) method for the four legumes in the three soils ranged 

from 31/g soil to 1700/g soil. The populations of native Rhizobium/g soil capable 

of nodulating each of these four legumes were found to be highest in Alajo soil 

(mean 9.95 xlO2) while the least Rhizobium counts occurred in Hatso soil (mean 

1.10 xlO2).

iv



Rhizobium isolates obtained from Tephrosia spp were found to be most 

promiscuous, and except for Leucaena spp, isolates from Tephrosia spp nodulated 

the two other tree species (i. e. A. lebbek,and Pithecelobium spp ), in contrast to 

isolates from A. lebbek and Pithecelobium spp which were found to be specific 

only for their respective host plants. Phosphorus application alone resulted in 

significantly improved nodulation (on average about 63%). Sesbania specioca, S. 

Aculiata and A. farresiana did not nodulate with the indigenous rhizobia without 

phosphorus in Toje soil but did nodulate after phosphorus application. Similarly, 

S. rostrata and Leucaena spp nodulated in Hatso and Alajo soils, only after 

phosphorus application. Despite the general response by the trees to phosphorus, 

some species, like A. lebbek, S. aculiata and Tephrosia spp did not respond 

significantly to phosphorus application on Alajo soil. Phosphorus application, 

however did not result in significant increase in both %N fixed (about 38%) and 

total N (about 44 mg) fixed. Also, with the exception of Tephrosia spp, P 

application did not result in significant increase in total dry matter yield of the tree 

legumes. Although inoculation resulted in more than double nodule numbers of 

the tree species, it did not result in significant increase in both total N fixed and 

%N fixed. It also did not significantly increase total dry matter yield except in the 

case of Tephrosia spp. Leucaena responded highest to inoculation with an 

increase of over 123% total N fixed. In general however, Tephrosia spp gave the 

highest Biological Nitrogen Fixation (BNF) followed by Pithecelobium spp and A. 

lebbek with the lowest being Leucaena in terms of both percent and total N fixed. 

These studies therefore identified Tephrosia spp as having high potential for both 

dry matter yield and nitrogen fixation. Because of the high numbers of native 

rhizobia present in the soils studied, Tephrosia spp stands a good chance of being 

used for nitrogen recycling.

V



CONTENTS PAGE

Dedication............................................................................................... i

Declaration...........................................................................................  ii

Acknowledgement............................................................................... iii

CHAPTER ONE : INTRODUCTiON...............................1

CHAPTER TWO : LITERATURE R EV IEW ...........................................7

2.1 Nitrogen fixing tre e s ........................................................................................7

2.1.1 Nodulated tree legumes for agroforestry...................................................  8

2.1.1.1 Acacia spp ......................................................................................................... 8

2.1.1.2 Leucaena s p p ....................................................................................................9

2.1.1.3 Faidherbia ........................................................................................................9

2.1.1.4 Prosopis...........................................................................................................10

2 1.1.5 Caliilandra ...................................................................................................... 10

2.1.1.6 G liricidia ....................................................................................................... 11

2.1.1.7 Sesbam a .......................................................................................................... 11

2.1.2 Actinorhizal trees for Agrofbrestry.............................................................12

2.2 The role of trees in Agroforestry.................................................................. 13

2 .3 Attributes of Leucaena as a multi-purpose t r e e ........................................... 15

2.4 Nitrogen contribution by some multi-purpose trees to c ro p ..................... 18

2.5 Some nitrogen fixation measurements in nitrogen fixing tree s ................ 21



2.5.1 Leucaena s p p ................................................................................................................

2.5.2 C asuarinaspp .............................................................................................................

2.5 .3 Gliricidia s p p ............................................................................................................ 25

2.5.4 Acacia s p p .....................................................................................................................26

2.5.5 Sesbaniaspp ................................................................................................................ 27

2.5.6 Alder s p p ....................................................................................................................... 27

2.6 Factors influencing N-15 estimates of BNF in tre e s ....................................................

2.6.1 Selection of appropriate reference p lan t....................................................................30

2.6.2 N-15 labelling ...

2.6.3 Application rates

2.6.4 Sampling of plant m aterial..........................................................................................34

2.7 Influence o f phosphorus on nodulation and nitrogen fixation.....................................

2.8 Environmental factors affecting N r  fixation.................................................................

2.8.1 Physical factors............................................................................................................ 40

2.8.1.1 Temperature................................................................................................................. 40

2.8.1.2 Drought.........................................................................................................................42

2.8.1.3 Water-logging.............................................................................................................. 43

2.8.2 Chemical constraints

2.8.2.1 Toxicity..........................................................................................................................44

2.8.2.2 Nutrient deficiency.......................................................................................................46

2.8.3 Biological constraint...................................................................................................49

2.9 Some methods for estimating nitrogen fixation...........................................................



2.9.1 Total nitrogen difference (TND) method 51

2.9.2 Acetylene reduction assay (ARA)......................................................................54

2.9.3 The 15N-isotope technique................................................................................... 56

2.10 Methods of estimating soil microbial status.............................................59

2.10.1 Convention m ethods..............................................................................................59

2.10.2 Current m ethods.................................................................................................... 62

2.10.1 The Chloroform fumigation technique................................................................ 64.

2.10.2.2 Soil A T P ..........................................................................................................65

2.10.2.3 Fumigation extraction method....................................................................... 65

CHAPTER THREE : MATERIALS AND M ETHODS............................... 67

3 .1 Location of study a rea ................................................................................................. 67

3.2 Soils and site characteristics.........................................................................................67

3.2.1 Soil sampling and preparation........................................................................... 69

3.2.2 Soil analysis........................................................................................................... 69

3.2.2.1 pH( water)..............................................................................................................69

3.2.22 Total phosphorus.................................................................................................. 70

3.2.2.3 Available phosphorus...........................................................................................71

3.2.2.4 Total nitrogen.......................................................................................................72

3.3 Plant growth media......................................................................................................73

3.4 Planting materials.........................................................................................................73

3.5 Seed pretreatment........................................................................................................ 7 4

3.6 Initial screening...........................................................................................................7 4

viii



3.7

3.8

3.9

3.10

3.11

3.12

3.12.1

3.13

3.14

4.1

4.2

4.3

4.4

4.5

4.6

4.6.1

4.6.2

4.6.3

Inoculation and N-15 studies.......................................................................................75

Watering......................................................................................................................... 76

Rhizobial enumeration..................................................................................................76

Cross inoculation studies........................................................................................78

Harvesting.................................................................................................................79

Analyticla method....................................................................................................80

Total N and I5N measurement............................................................................... 80

Measurement of N2 fixation.................................................................................. 80

Statistics..................................................................................................................... 81

CHAPTER FOUR : RESULTS............................................................................ 82

Soil characteristics......................................................................................................... 82

Nodulation capabilities of 14 tree and shrub species to nodulate in three soil types
.................................................................................................................................... 83
Effect of phosphorus application on nodulation of nine leguminous trees grown in three 
soil ty p es ......................................................................................................................... 66

Most probable number counts o f native rhizobia capable o f nodulating A. lebbek, 
Pithecelobium spp. Leucaena spp and Tephrosia s p p ......................................89

Cross inoculation studies........................................................................................... 90

Nitrogen fixation response by four tree legumes to inoculation and phosphorus in three soil 
series.......................................................................................................................92

Nitrogen fixation in A. lebbek, Pithecelobium spp. Leucaena spp and Tephrosia spp and 
how this is affected inoculation.......................................................................... 92

Nitrogen fixation in A. lebbek, pithecelobium spp, Leucaena spp and Tephrosia spp and 
how this is affected by phosphorus application.................................................93

Nitrogen fixation A. lebbek, Pithecelobium spp. Leucaena spp and Tephrosia spp and how
this is affected by soil series...............................................................................96

ix



Total dry matter yield response o f four tree legumes to inoculation and phosphorus in three 
soil series...........................................................................................................98

CHAPTER FIVE : DISCUSSION 103

CHAPTER STX : SUMMARY, CONCLUSIONS AND RECOMMENDATIONS . 509

LIST OF TABLES................................................................................................................... 82

Table 1. Some chemical properties of the soils u sed ....................................................82

Table 2. Nodulation of fourteen tree legumes by indigenous rhizobia in three soil types
....................................................................................................................................................83

Table 3. Native rhizobiun/g soil in the Toje, Hatso and Alajo so ils.................................89

Table 4. Cross inoculation response by some tree legum es.............................................. 91

Table 5. Total dry matter yield of four tree legumes as influenced by inoculation.....100

Table 6. Total dry matter yield of four tree legumes as influenced by phosphorus
application................................................................................................................100

Table 7. Total dry matter yield of four tree legumes as influenced by three soil types
 101

Table 8. Total dry matter yield of four tree legumes as influenced by inoculation in three 
Soil types...................................................................................................................102

Table 9. Total dry matter yield of four tree legumes as influenced by phosphorus in three 
so ils............................................................................................................................ 102

LIST OF FIGURES..................................................................................................................85

Fig 1 A. Nodulation of nine tree legumes with or without phosphorus application to Toje 
so il................................................................................................................................85



Fig. 1B, Nodulation of nine tree legumes with or without phosphorus application to Hatso 
so il.................................................................................................................................86

Fig. 1C. Nodulation of nine tree legumes with or without phosphorus application to Alajo 
so il................................................................................................................................ 87

Fig. 2. Summary of the ability o f the three soils, Toje, Hatso and Alajo to support 
nodulation o f nine tree Legumes with or without phosphorus application

 88
Fig. 3. Nitrogen fixation response of A  lebbek, Pithecelobium spp, Leucaena

spp and Tephrosia spp to inoculation........................................................................94

Fig. 4. Nitrogen fixation response o f A  lebbek, Pithecelobium spp, Leucaena
spp and Tephrosia spp to phosphorus.......................................................................95

Fig. 5. Nitrogen fixation o f four tree legumes on three soil ty p es .....................................97

REFERENCES.......................................................................................................................112



CHAPTER ONE

11 INTRODUCTION

Ghana is richly endowed with forest and savannas which provide timber for its 

wood industries, building materials for construction, fuel wood (which is the most 

important source of energy in rural areas) and minor forest products like oils, 

medicine, chewing- sticks etc. They provide sanctuary for wildlife, protect and 

enrich the soils, maintain and regulate stream flow and preserve the genetic 

diversity of the flora and fauna. These forests and savannas have to be protected and 

used rationally if they are to continue to provide the above functions. These renewal 

natural resources are, however, being depleted at an alarming rate through certain 

farming practices like shifting cultivation and bush fallow as well as other human 

related activities like indiscriminate and sometimes unplanned commercial logging 

of timber, excessive harvesting of fuel wood, bush fires, overgrazing etc.

For instance, the dominant farming system in Ghana today is based on shifting 

cultivation and bush fallow practices where the soil had to be rested for many 

years for its fertility to be restored, after few years of cropping. These practices 

have led to increased deforestation. According to F.A.O. 1982, the tropics are

1



loosing more than 10 million hectares o f forest cover annually and shifting 

cultivation is responsible for almost 70% o f deforestation in tropical Africa. 

The situation is alarming in tropical Africa where deforestation exceeds the 

projected rate of tree planting by a ratio of 29% (FAO,1982). This problem has 

been aggravated by increase in land pressure due to increase in population 

which has let to shortening of the fallow period needed to restore soil fertility in 

the traditional farming systems. Continuous cultivation (in response to 

increasing population) without much fertilizer addition to soil has led to 

drastically reduced soil fertility and increased soil erosion in the humid tropics 

and the semi-arid areas (EL-Ashry, and Ram, 1987). This is compounded by the 

fact that most soils in humid tropical Africa are sandy, highly weathered, low in 

organic matter content and susceptible to soil erosion and compaction ( Lai,

1987). In drier areas, overgrazing and harvesting of trees for firewood have 

been major factors in reduced productivity, increased soil erosion and 

desertification (F.A.O 1982). In order to reverse the present trend, there is the 

need to reduce the rate of population growth, introduced settled agriculture, 

develop alternative sources o f energy and embark on serious tree planting 

programmes.

2



Attempts to improve the productivity of traditional farming systems in 

Ghana, by introducing inorganic fertilizers have proved futile, mainly, 

because of the inability of poor farmers to afford the prices of these 

fertilizer as a results of the removal o f subsidies by the government. The 

effective management of biological nitrogen fixing trees and other 

leguminous plants in our farming systems is a major key to restoring and 

maintaining soil fertility (Kang etal., 1981a ).

In recent years, the potential o f leguminous and actinorhizal trees for 

restoring and maintaining soil fertility has aroused considerable interest 

and international agencies (e.g. CSC, IITA, IAEA NAS,ACIAJR, IDRC 

etc) have recommended their increased use in agroforestry systems and for 

crop shade. The integration of trees, especially nitrogen fixing trees into 

agroforestry systems can make a major contribution to low input 

sustainable agriculture by restoring and maintaining soil fertility and in 

combating erosion and desertification as well as providing the needed 

firewood for domestic use. The major advantages of nitrogen fixing trees 

are their ability to establish in nitrogen deficient soils, and the benefits of 

the nitrogen fixed and extra organic matter to succeeding or associated 

crops. Sanginga et al, 1986 found that maize yields in soil in which

3



inoculated Leucaem spp had grown for 6 months, were increased from 

1.5t/ha to 2.5t/ha with prunings removed and from 2.2 to 4t/ha with 

prunings returned to the soil. Ladha et a/.(1989) indicated that, nitrogen 

fixed by Sesbania rostrata can significantly increase subsequent grain 

yield of lowland rice.

It has been assumed over the years that, all leguminous trees fix nitrogen 

and are therefore suitable for integration into agroforestry systems . 

Although nodules are important for nitrogen fixation in legumes, till now, 

many leguminous trees have not been examined for nodulation and the 

need to inoculate with rhizobia for enhanced nodule formation (Allen and 

Allen, 1981). Recent preliminary studies (Ampadu and Danso personal 

communication) have shown that there was no nodulation in “Dawadawa” 

and Tetrapleura tetraptera. With the exception of some species such as 

Leucaena leucocephela which have been reported to fix large amounts of 

nitrogen in Hawaii (Guevarra, 1976) and in Nigeria (Kang et al., 1981(a) 

our knowledge of nitrogen fixation in trees is still very limited, there is 

therefore the urgent need for more studies on nitrogen fixing trees 

especially in developing countries like Ghana. In this study, 14 species of

4



some common indigenous leguminous trees would be assessed for their 

growth, nodulation and nitrogen fixing ability on three (3) different soil 

types that occur in the coastal savanna zone.

The objectives of this study are:

To investigate the growth nodulation and nitrogen fixing 

potential of some common indigenous multipurpose trees.

To assess the need for inoculation for improving nitrogen 

Fixation in various leguminous trees of potential use in 

agroforestry and forestry.

To select leguminous tree species with capacity for use in 

agroforestry and forestry.

To isolate rhizobia likely to be of use in the preparation of 

inoculants.

To test whether nodulation of indigenous nitrogen fixing 

trees (NFT’s) can be improved by phosphorus fertilization.

1.1.2 Justification: Available literature shows that little work 

has been done on biological nitrogen fixation in trees. It is 

to fill this gab as well as towards improving and sustaining

5



soil fertility and productivity for food sufficiency in Ghana 

that this study is being proposed.

6



CHAPTER TWO

LITERATURE REVIEW

2.1 NITROGEN FIXING TREES.

Nitrogen fixing trees fall in two main groups, the nodulated legumes and 

actinorhizal trees. The family leguminosae is divided into three sub­

families, the papilionoideae (Pea-like flowers), the Mimosoideae 

(compound inflorescence with reduced petals) and the Caesalpinoideae 

(flowers usually with five petals, apparently radially symmetrical)(Polhill 

and Raven, 1981). The majority of trees used for agroforestry are 

nodulated members of the papilionoideae and the mimosoideae, but there 

are few species in the caesalpinoideae many of which do not form nodules 

but which have been used experimentally in agroforestry (Allen, et al., 

1981).

7



2.1.1 Nodulated Trees Legume For Agroforestry.

The most commonly used nodulated tree legumes in agroforestry include:

(a) Acacia spp (b) Leucaena spp (c) Faidherbia spp (d) Prosopis spp (e) 

Calliandra spp (f) Gliricidia spp and (g) Sesbania spp.

2.1.1.1 Acacia spp. (Mimosoideae)

Acacia is the largest and most divers genus of legume trees, containing 

approximately 1200 species which are distributed throughout the tropics 

and subtropics (Brewbaker, 1987). Most species are found in the semi-arid 

tropics and they are generally resistant to drought, due at least in part, to 

having very deep root systems. Species of Acacia noted for their potential 

or actual use in agroforestry include; A. Senegal, A. meamsii, A. aneura, 

A. auriculiformis, A. holoserica, A. mangium and A. tortillis. Turnbull,

(1987) reported many additional species that occur naturally in Australia 

and South-east Asia which could be of widespread use in arid regions of 

the tropics.

8



2.1.1.2 Leucaena {Mimosoideae)

This genus of leguminous trees consist of 13 species of which L. 

leucocephela, L. diversifolia and L. esculenta have excellent agroforestry 

potential (Pound and Martinez-Cairo, 1983; Anon, 1984a). The strong, 

deep root system allows Leucaena to tolerate drought and, once 

established, it can survive in areas with only 600mm annual rainfall. One 

of the major limitations of L. leucocephela is its poor growth on acid soils 

although improvement of acidity tolerance has been achieved by crossing 

with acid-tolerant accessions of L. diversifolia (Hutton, 1990). Leucaerta is 

nodulated by fast-growing rhizobia isolated for several tropical hosts but 

generally not by most temperate Rhizobium sepcies or by slow-growing 

rhizobia (Trinick, 1996; Lewin et al., 1977)

2.1.1.3 Faidherbia (Mimosoideae)

The only species in this genus is Faidherbia albida and is found 

throughout the drier regions of sub-Saharan Africa (Wickens, 1969). Its 

major characteristic is the unusual shedding of its leaves during raining 

season. F. albida forms nodules with Bradyrhizobium strains but not with

9



the fast-growing rhizobia strains tested so far (Dreyfus and Dommergues,

1981).

2.1.1.4 Prosopis (Mimosoideae)

All the 44 species of this genus are found in semi-arid and arid regions 

throughout the tropics (Anon, 1979). Most of the species are of potential 

use in agroforestry because of their ability to withstand extreme drought, 

and the fact that their pods are an important animal fodder in many regions 

(Felker, 1979). Prosopis spp nodulated by fast-and slow-growing rhizobia 

(Jenkinson el al., 1987).

2.1.1.6 Calliandra (Mimosoideae)

This genus has over hundred species which occur in Central and South 

America. Calliandra is widely used for fuel wood and animal fodder. It is 

best suited to growth in humid areas with annual rainfall of 2000-4000mm 

on soils with good drainage as waterlogging can rapidly kill the trees. 

Calliandra nodulates rapidly with indigenous Rhizobium strains in soils in

10



Java and is nodulated by both fast-and slow growing strains (Dreyfus and 

Dommergues, 1981).

Gliricidia spp (papilionoideae)

The widely known species of this genus of agricultural use is the Gliricidia 

sepium. The species can form nodules with both fast-and slow-growing 

rhizobia and was found to fix N2 most effectively with fast-growing strains 

originally isolated for Gliricidia nodules (Akkasaeng et al., 1986).

2.1.1.7 Sesbania spp (Papilionoideae).

The over 60 species of this genus are distributed throughout the tropics 

with representatives to all continents (Evans and Rotar, 1987; Evans, 1970) 

and vary from annual to short-lived perennial. Several species including S. 

bispinosa, S. cannabina, S. sesban, S. grandiflora, S. specioca, and S. 

rostrata have traditionally been investigated for use in agroforestry. A 

major advantage of Sesbania is that many species are both tolerant of 

waterlogging and of saline and alkaline conditions. The stem nodulating

11



Rhizobia nodulating roots of Sesbania spp are generally fast-growing 

strains, and often exhibit rapid growth, produce large (74mm) colonies in 

less than 24hours, although a few slow-growing isolates have been 

recorded (Odee, 1990).

2.1.2 Actinorhizal Trees For Agroforestry.

Species of Casuarina are the most important non-leguminous N2-fixing 

trees in the lowland tropics. Four genera are now recognised within the 

Casuarinaceae and include; Allocasuarina, Casuarina, Gymnostoma, and 

Ceuthostoma which together contain about 90 species native to Australia, 

Malaysia and Polynesia (Diem and Dommergues 1990). The most widely- 

used species is C. equisetifolia which is tolerant of saline soils and of 

strong winds, and two other species quite widely-cultivated are C. 

cunninghamiana and C. glauca (Anon, 1984b)

species S. rostrata has excited intense interest due to its ability to grow and

fix N2 in waterlogged conditions and its fast growth rate.

12



Rhizobia nodulating roots of Sesbania spp are generally fast-growing 

strains, and often exhibit rapid growth, produce large (74mm) colonies in 

less than 24hours, although a few slow-growing isolates have been 

recorded (Odee, 1990).

2.1.2 Actinorhizal Trees For Agroforestry.

Species of Casuarina are the most important non-leguminous N2-fixing 

trees in the lowland tropics. Four genera are now recognised within the 

Casuarinaceae and include; Allocasuarina, Casuarina, Gymnostoma, and 

Ceuthostoma which together contain about 90 species native to Australia, 

Malaysia and Polynesia (Diem and Dommergues 1990). The most widely- 

used species is C. equisetifolia which is tolerant of saline soils and of 

strong winds, and two other species quite widely-cultivated are C. 

cunninghamiana and C. glauca (Anon, 1984b)

species S. rostrata has excited intense interest due to its ability to grow and

fix N2 in waterlogged conditions and its fast growth rate.

12



Marked differences in effectiveness of N2 —fixation have been found both 

between different strains of Frankia (Reddell and Bowen, 1985) and 

between different accessions of Casuarina (Sanginga et al., 1990a)

2.2 THE ROLE OF TREES IN AGROFORESTRY

Alley cropping, also referred to as hedgerow inter-cropping (Torquebiau, 

1990) entails managing rows of woody plants (hedgerows) with annual 

crops planted in the alleys between the hedgerows.

The most distinctive component of an agroforestry practice is the 

multipurpose trees and shrubs (MPTS) (Nair, 1990). Multipurpose trees 

are all woody perennials that are purposefully grown to provide more than 

one significant contribution to the production and/or service functions of a 

land use system. They are so classified according to the attributes of the 

plant species as well as the plant’s functional role in the agroforestry 

practice under consideration (Huxley, 1984).

Soil improvements under alley cropping may result from the service 

function as well as the protective role of MPTS as influenced by 

management and environmental factors. Nair (1990) has enumerated some

13



points which support the soil improving capabilities of trees and shrubs 

under natural ecosystems. Even though there are reports of investigations 

conducted by Kang and Wilson (1987) Sanchez (1987), Juo (1989) and 

Avery et al. (1990) on alley cropping’s role in soil improvements, the 

results are not quite conclusive. However, Nair (1990) expresses the hope 

that, given the substantial volume of scientific information on the soil 

improving attributes of trees, it is a matter of management to realize the 

significant contribution that tree-based systems like alley cropping can 

make to soil fertility and overall soil productivity.

Characteristics that make MPT desirable for soil improvement under alley 

cropping have not been adequately elucidated. However, some guidelines 

are offered by Young (1989a) and Kang et al., (1984). These properties 

include high biomass production (below and above ground), nitrogen 

fixation and mycorrhiza association, a well-developed rooting system and 

rapid growth and ability to coppice. The other properties are fast or 

moderate rate o f litter decay to suite either nutrient release or soil erosion 

control, high nutrient content in the biomass and absences of toxic 

substances in foliage or root exudates.

14



Young (1989a) has provided a list of 32genera and 55 species which have 

been identified to be beneficial for the maintenance or improvement of soil 

fertility from several sources. Included in the lists are species such as 

Albizia lebbeck, Gliricidia sepium, Leucaena leucocephela, Cajanus cajan, 

Dactyladenia barterii, Alchomea cordiflora, among others which have 

been used in hedgerow intercropping trials in the tropics. The possible 

contributions of the biological nitrogen fixing species among the list 

should be cherished, especially in most tropical countries where inorganic 

nitrogen fertilizers are expensive.

Amara (1987a, 1987b) has provided a list of potential trees for use in 

agroforestry systems. Of importance are Albizia lebbek, C. siamea, E. 

cyclocarpum, G. sepium and L. leucocephela.

2.3 ATTRIBUTES OF LEUCAENA AS A MULTI-PURPOSE TREE

Leucaena leucocephela seems to have taken the lead among other nitrogen 

fixing trees (NFTs) in agroforestry systems. (Brewbaker, 1975; Young, 

1989a). It is considered to be the most widely used MPT in scientific

15



agroforestry (Young, 1989a). Leucaena is an arborescent legume 

belonging to the mimosaceae family. Native to Mexico, leucaena is now 

pantropic. It is said to thrive better on moderately free draining, alkaline to 

neutral and light to heary soils. (RAPA, 1987; F/FRED, 1994). There is 

no unanimity on its annual rainfall requirements and dry season tolerance. 

However, it is generally accepted that water logging conditions do not 

favour its growth (GAP A  1987; F/FRED, 1994; Young, 1989a)

The extensive root system of leucaena partly accounts for its soil 

improving properties. Dijkman (1950) associated leucaena with deeply 

penetrating tap root and lateral roots enabling them to exploit phosphorus 

and other minerals for the plants use (Brewbaker, 1975). The level of 

nitrogen in the leaves reported as 2.5 -  4.0% (Agboola, 1982; Buck, 1986) 

makes leucaena a good nitrogen and protein source for crops and livestock 

respectively. The high biomass production of leucaena is one of its 

attractive characteristics as a MPT. Young (1989a) quotes 10,000 -  25,000 

kg/ha/yr as the biomass production by leucaena. This figure compares 

favourably with the value of 20,000 kg/ha/yr which is the average estimate 

of evergreen rainforest biomass production (Nair, 1990). In savannah 

ecosystems, biomass production rates of 10,000 kg/ha/yr and 5,000

16



kg/ha/yr have been reported for moist and dry savannah, respectively 

(Young, 1989a).

Agboola (1982), working in a bimodal moist sub -  humid area in Nigeria, 

reported a leaf biomass production of 2470kgDM/ha/yr for leucaena under 

alley cropping. Figures for Cajamis cajan and Tephrosia Candida under 

the same conditions were 4100 and 3070 kg MD/ha/yr respectively 

(Agboola, 1982). While biomass production on the whole is an indicator 

of good MPT, the plant parts that go into the soil to maintain organic 

matter status, that is, leaves and roots are of paramount importance in alley 

cropping. Young (1989a) has provided some estimates of the amounts of 

above ground dry matter that need to be added to the soil to maintain soil 

organic matter content in three climatic zones of the tropics. While these 

figures did not include root biomass which can make immense contribution 

of organic inputs to the soil, they included woody components which may 

be harvested and taken out of the soil.

Another interesting attribute of leucaena is the pattern of roots distribution. 

Earlier reports on the root system of leuceana seemed to have assumed the

17



“commonality” of all trees as deep rooted. Working on a Sandy loam soil 

in Tanzania, Johnssen et al. (1988) found that, the pattern of root 

distribution in leucaena was similar to that of maize. Other workers like 

Ong (1991) and Noordwijk et al. (1990) have reported the superficial 

nature of roots in leucaena and other MPTs.

The high nutritive value of green and dry forage of leucaena is widely 

acclaimed (Brewbaker, 1975; RAPA, 1987). However, it has been found 

that when non-runinants are fed more than 5% (dry weight) of their ration, 

thyroid problems and other side effects are noticed (Brewbaker, 1975)

2.4 NITROGEN CONTRIBUTION BY SOME MULTIPURPOSE 

TREES (MPTs) TO CROPS

Prunings of leguminous hedgerows have been shown to increase yields of 

various crops. This is partly attributable to their N contribution (Guevarra, 

1976; Kang et al., 1984; Widyanatha, 1984; Yamoah, et al., 1986; Kang, et 

al., 1989; Sangakkara, 1989). For instance, trials conducted at 1ITA in 

Ibadan in Nigeria have shown that prunings of the leguminous species

18



Cassia siamea, Gliricidia sepium and Leucaena leucocephala can increase 

crop yields and improve soil properties ( Yamoah et al., 1986a; Kang and 

Wilson, 1987). The effective N contribution by the hedgerows to the 

associated crops in alley cropping can be calculated as the difference 

between N content of alley cropped plants and those grown in the control 

treatment (with no hedgerows). This method of estimation, however, 

ignores the priming effect that can occur after the application of green 

manures (Herridge and Bergersen, 1988). Most precise data on 

contribution from the hedgerow prunings to the associated crops can be 

obtained by using N-labelled plant material. A wide range of estimated N 

contribution values from hedgerow prunings to associated crops have been 

reported. Mulongoy (1986) has reported of an N contribution of 38- 

43kgN/ha with the application of 3-4t/ha of Sesbania rostrata prunings. 

Kang (1988)) estimated effective N contribution from L. leucocephela and 

G. sepium prunings to alley cropped maize to be about 40kg N/ha. On the 

other hand, Mulongoy and Van der Meersch (1988) reported a lower N 

contribution of 4.4 — 23.8kgN/ha from L. leucocephela prunings to the 

associated maize crop. The N contribution in Mulongoy and Van der 

Meersch (1988) represents less than 30% of the N yield of the prunings.

19



This low efficiency in the crop’s use of N from prunings probably results 

from the lack of synchronization between N release from the prunings and 

plant N demand, volatilization loss of N from prunings, and leaching loss 

(Mulongoy and Akobundum , 1990). Danso et al. (1996) indicated that, 

although N content from prunings of Gliricidia sepium, Senna siamea and 

Gmelina arberea were high, their N contribution to associate crop were 

low. They indicated that, the amount of nitrogen contributed to rice by the 

prunings of Gliricidia sepium, Senna siamea and Gmelina arborea was 

20.4, 1.6 and 2.7kg N, respectively. This, on the average is equivalent to 

25%, 7% and 8% of the crop’s N met by the prunings. For cowpea grown 

in rotation with rice, Danso et al. reported that Gliricidia sepium, Senna 

siamea and Gmelina arborea contributed 2.3, 0.6 and 1.8kg N, 

respectively, corresponding to 12%, 5% and 10% of the crop’s N obtained 

from prunings.

In most alley cropping systems, more than 70% of N in prunings is 

unaccounted for (Sanginga et al., 1993). Some of the possible causes 

suggested are losses due to volatilization and leaching, denitrification, N 

retained in the soil organic matter or recovered by hedgerow trees and by

20



weeds. Research results have shown that N yield is highly correlated with 

pruning biomass yield. There are a number of management factors which 

affect the N yield of the hedgerow prunings. These factors include height 

and frequency of prunings (Duguma et al. 1988), Plant density and inter­

hedgerow spacing (Kang et al, 1989) and Timing of pruning during 

cropping season.

Duguma et al. (1988) showed that low pruning height and high pruning 

frequency of Leucaena leucocephela, Gliricidia sepium and Sesbania 

grandiflora, gave lower N yield, where as high pruning height and low 

pruning frequency gave higher N yield and also increased wood 

production.

2.5 SOME NITROGEN FIXATION MEASUREMENTS IN 

NITROGEN FIXING TREES.

Several studies done using MPTs have shown that on the average, well 

nodulated woody legumes can fix nitrogen amounting to 134-274 kg 

N/ha/yr in the field using 15N methodology. This represents an average of

21



45% of their total N content (Mulongoy, 1986; Herridge and Bergersen, 

1988; Mulongoy et al., 1988; Sanginga et al., 1989)

2.5.1 LEUCAENA spp

Sanginga et al. (1989a) reported an amount of 133kgN/ha fixed by 

Leucaena leucocephela grown for 6 months in the field in Nigeria; this 

represents only 39% of the N in the plant, indicating that more than half of 

the N in leucaena was of soil origin. A similar % Ndfa was reported 

during the first year’s growth of Leucaena in Hawaii Van et al. 1990. 

Greenhouse studies conducted at the IAEA laboratories in Seibersdorf 

using 15N gave similar results. For a 36 weeks period, two genotypes of 

leucaena derived 43 and 36% of their N from fixation representing less 

than half of their total N (Sanginga et al. 1990b and 1990c). Zaharah et al. 

(1986) on the other hand reported a higher proportion (78%) and amount 

(231kgN/ha) o fN  fixed in six months in three provenances of leucaena 

grown in the field in Malaysia, while the average % Ndfa in 11 Isolines of 

L. leucocephela in one study of Sanginga et al. (1990) was 65%. Several 

factors could account for the large differences in reported proportions and

22



amounts of N fixed in leucaena, of important is the great differences in soil 

status. Luyindala and Haque (personal communication) found that % Ndfa 

dropped from 65 to 34 by raising the N rate from 20 to 100kg /ha. Also, 

genotypic differences in ability to fix N2 could account for some of the 

differences observed. For 11 L. leucocephela Isoline studied, % Ndfa 

ranged from 37 to 74% (Sanginga et al., 1990d ) an indication of the large 

genotypic variation that can exist in the N 2 fixation abilities of leucaena 

Isolines.

2.5.2 CASUARINAS spp

Casuarinas have been identified for their high N2 -  fixing capability. 

Gauthier et al. (1985) under simulated field conditions measured only 53% 

Ndfa by isotope dilution (ID) method in C. equisetifolia grown in a very 

sandy soil low in N; by making some assumptions, the total N2 fixed was 

estimated as 40 to 60 kgN/ha/yr. Similar results were also obtained by 

Sougoufara et al. (1990), who measured a maximum contribution of 59% 

Ndfa to the first year’s growth of C. equisetifolia in this sandy soil. An 

almost equal amount of N was fixed during the second year of C.

23



equisetifolia growth as in the first year (Sougoufara et al., 1990). However, 

% Ndfa in the second year declined from 64.8 to 52.7%. Sougoufara et al. 

(1990) and Sanginga et al. (1990e), estimated similar amounts of N2 fixed 

by C. equisetifolia using either the ID or total nitrogen difference (TND) 

method, although estimates by the TND method were less precise than the 

Isotope-derived ones.

Large differences have been reported for the N2-fixing abilities of different 

Casuarina genotypes. Sougoufara et al. (1990) found an almost five fold 

difference in N2 fixed by two genotypes of C. equisetifolia, while Sanginga 

et al. (1990e) found that 18 weeks after planting, C. equisetifolia fixed 

more N2 than C. ctmninghamiana (means 63% and 43% respectively). 

Large differences however, were also found in the abilities of the different 

C. equisetifolia and C. cunninghamiana provenances to fixed N2. In the 

case of C. equisetifolia, the % Ndfa ranged from 25 to 75% (equivalent to 

4 to 29 mgN/plant), and from 14 to 76% (equivalent to 2 to 25mgN/plant) 

for C. cuninghamiana. Thus, even though C. cunninghamiana was 

generally a poorer N2 fixer some provenances were capable of achieving 

close to the maximum N2 fixed by the C. equisetifolia provenances.

24



2 .5 .3 GLIRICIDIA spp

Gliricidia spp has been reported as having great potential in Agroforestry 

after Leucaena spp. However, results obtained using the acetylene 

reduction assay suggested that Gliricidia is a poor N2 -  fixer (Duhoux et 

al., 1990). This finding is in contradiction to results of Awonaike et al. 

(1990) who found that N2 fixed in Gliricidia sepium was comparable to, or 

slightly higher than most values reported for L. leucocephela, which is 

regarded as a high N2 fixer. The average % Ndfa for five plant genotypes 

and five Rhizobium strains studies was reported as 60% or 3.95gN/plant 

(Awonaike et al., 1990). Thus on average, these five genotypes obtained 

more than half of their total N from fixation. The results indicated high 

variability in N2 fixation, a significant proportion of which was due to 

Rhizobium strain and plant genotype interaction. No one Rhizobium strain 

was ideal for all plant genotypes, and depending on the infecting 

Rhizobium strain, N2 fixed in particular plant genotype was either high or 

low. By adding small amounts of 15N at frequent intervals, Sanginga, 

Zapata, Danso and Bowen (unpublished) measured 72%Ndfa in G. sepium, 

a higher proportion than even that of Awonaike et al. (1990). Liya, Odu,

25



Agboola and Mulongoy (unpublished) using the ID method also estimated 

85% Ndfa in G. sepium whiles the TND method gave 72% Ndfa.

2.5.4 Acacia spp

The N2 fixation data available for many of the Acacia spp indicate that, 

Acacia often has a low potential for N2 fixation (Dommergues, 1987; 

Sanginga et al., 1990a). In a study using N-15 methodology to assess N2- 

fixation potential of legumes in the Sonoran desert, Shearer et al.{1983) 

concluded that A. greggii for example, did not fix any N2. Sanginga et al. 

(1990d) reported that average % Ndfa in 11 provenances of A. albida was 

20% compared to 65% in 11 isolines of L. leucocephela. In view of the 

low N2 fixation in A. albida, Sanginga et al. (1990d) found that, errors 

attributable to the reference crop were of greater significance than for 

leucaena, which is in agreement with the suggestion by Hardarson et al.

(1988) that errors involved in N2 fixation measurements are more serious 

where fixation is low.

26



2.5.5 Sesbania spp

Sesbcmia, particularly the stem nodulating species have been of great value 

as organic manure in rice farming systems, where they are capable of 

fixing virtually all their N (Pareek et al., 1990). Both the ID and TND 

methods have been used to assess N2 fixed in Various types of Sesbania. 

Ndoye and Dreyfus (1988) estimated 83 to 109kgN/ha fixed within 6 

months in the stem nodulating S. rostrata, compared to only 7 to 18kgN/ha 

in the root nodulating S. sesban. Higher values of N2 fixed than those of 

Ndoye and Dreyfus (1988) were measure by Pareek et al. (1990), who 

found that S. rostrata and and S. aculeata derived 80% of their N in one 

season, and 94% in another, with the TND approach giving significantly 

lower estimates of N2 fixed, ranging form 59 to 88%.

2.5.6 ALDER spp

Alders are noted for their large nodule sizes, which may sometimes reach 

the size of a tennis ball (Alexander, 1961), and are important component of 

many natural ecosystems in temperate climate. Cote' and Camire (1984) in

27



their study estimated % Ndfa to be 68% (equivalent to 53kgN/ha). From 

leaf samples collected from A. glutinosa, Domenach and Kurdali (1989) 

measured 87% Ndfa, after taking into account translocation of natural N 

reserves and isotopic discrimination. Domenach et al. (1987) used both 

natural 15N abundance and 15N fertilizer addition and found that both 

approaches gave similar estimates, with the % Ndfa in alders of different 

origin ranging from 40 to 80%.

2.6 FACTORS INFLUENCING N-15 ESTIMATES OF BNF IN 

TREES.

The measurement of biological Nitrogen fixation in plants (BNF) is a 

crucial step towards efforts aimed at increasing the contribution of 

atmospheric nitrogen to plant nutrition and soil fertility. Although there 

are many methods for measuring BNF in plants, none is perfect in terms of 

measuring N-fixed accurately. The advantages and limitations of each 

method have been discussed in several reviews (Bremner, 1977; Burris, 

1974; Danso, 1985; Dardy et al., 1973; and Knowles, 1981). However, of 

the several methods available the 15N methodologies are the most reliable

28



for measuring N2 fixed in plants and hence, are often the methods of choice 

(Duhoux and Dommerques, 1985; Duque et al., 1985; Hardarson et al., 

1985b; Leg and Sloger, 1975; Rennie, 1982; 1984; Sanginga et al., 1985; 

West and Wedin, 1985). These 15N methodologies are particularly useful 

as they can at a single harvest measure the integrated amounts of N2 

assimilated by both greenhouse and field-grown plants in addition to 

measuring the N contributed from soil or fertilizer sources (Danso, 1985; 

Fried et al., 1983; Vase and Victoria, 1986). There is therefore an 

increased use in the I5N methodologies for measuring N2 fixed in various 

crops and cropping systems ( Chalk, 1985; Ledgard et al., 1985; Weaver, 

1986).

The increased use of I5N techniques in most N2-fixation measurement has 

revealed some practical problems. As suggested by Rose and Victoria 

(1986), a major problem with the use of 15N techniques has been a general 

lack of understanding of some of the underlying concepts. It is therefore 

likely that with better understanding of the concepts, and adequate 

precautions to reduce some of the avoidable errors, many of the limitations 

associated with the 15N techniques could be reduced. N2-fixation 

measurements in trees are more problematic than for annual legumes, and

29



Danso et al., (1992) have suggested some reasons for it. Among the factors 

likely to affect and limit the use of I5N methodologies in measuring BNF in 

trees include the following: (a) Selection of appropriate reference plant

(b) 15N Labelling (c) 15N application rates.

2.6.1 Selection of appropriate reference plant:

The selection of appropriate reference plant for a particular N2 fixing plant 

appears to be the greatest problem encountered with the I5N methodology. 

Fried et al. (1983) suggested the criteria for selecting reference crops for 

estimating the amounts and proportions of N2 fixed by the 15N 

methodology. It is important that, serious attempts are made to ensure that, 

the reference plant is a non-N2 fixer, with similar and N uptake pattern as 

the nitrogen fixing tree (NFT), and that, both are obtaining their N from a 

similar soil horizon. However, the higher the N2 fixed, the less stringent 

are the requirements for all the criteria to be met rigidly (Hardarson et al.,

1988). According to Danso et al. (1986), Philips et al. (1986), the need for 

the precise quantification of N2 fixation may not be compelling in many 

agronomic trials, e.g., in comparing some treatment effects, or ranking

30



varieties or Rhizobium strains for N2 fixing effectiveness. To achieve such 

objectives, a reference crop is often not necessary, as the results can be 

obtained using the relative 15N enrichments of the different trees; the lower 

the i5N enrichment, the higher the N2 fixation.

The problem of selecting a suitable reference crop is more serious in trees 

than in seasonal crops, as the 15N uptake patterns need to be matched over 

the different seasons and not over one or only a few seasons.

Where the 15N is added once, only at the beginning, it is likely that because 

the 15N / I4N ratio in soil declines less rapidly with time (Pareek et al., 

1990), the errors due to mismatched reference and fixing plants are likely 

to become smaller with time (Fried et al., 1983; Witty, 1983). Sanginga et 

al. (1990) and Pareek et al. (1990) have emphasised on the criteria for the 

selection of reference plants for NFTs. The results of Sanginga et al. 

(1990) clearly show that highly erroneous values can be obtained unless 

efforts are made to select suitable reference crops.

31



2.6.2 1SN Labelling.

The problem associated with the 15N labelling include lack of knowledge 

on what N rate to apply to the fixing and reference plants, the time and 

frequency of application. Although for most N2 fixation studies, 15N 

labelled fertilizers have been used mainly as tracers, the N they contain can 

affect growth and N2 fixation and hence the N rate to be used in fixation 

studies must not be over looked. In many studies, N rates may have been 

used without prior knowledge of how they would affect N2 fixation under 

the local conditions. For the use of the 15N techniques to estimate N2 

fixation, a zero 15N fertilizer rate which would have been ideal is only 

possible with the 15N natural abundance method. As a rule, the 15N applied 

must be high enough to ensure satisfactory 15N detection in plants, without 

reducing N2 fixation. Low N rates will therefore necessitate the use of 

highly enriched 15N fertilizers (West and Wedin, 1985). However, cost 

consideration may not always make this approach the best (rennie, 1986; 

Danso et al., 1987). Increasing the quantity of 15N fertilizer (if it does not 

significantly inhibit N2 fixation) with a corresponding decrease in 15N 

enrichment is thus economical and more advisable (Danso et al., 1988). It

32



is generally advisable to apply high versus low rates of N to the reference 

and fixing crops respectively, to estimate N2 fixed. This approach 

overcomes the limitation of poor growth of the reference crop when soil N 

is inadequate, while avoiding the suppression of N2 fixation by high 

inorganic N (Fried and Broeshart, 1975).

2.6.3 15N Application Rates:

A single application of 1SN labelled fertilizer satisfactory for N2 fixation 

measurements over more than one season, for instance, in perennial crops, 

may be very expensive and or inhibitory to N2 fixation. Besides, the 

greatest problem with this approach, even with annuals is the sharp decline 

in 15N /14N ratio of soil N, often necessitating the rigorous selection of 

reference crop. To overcome these problems, equal-sized, multiple 

additions of small amounts of 15N fertilizer during various times of the 

growth cycle were first used by Vallis et al. (1967), to estimate N2 fixed. 

Since then, several studies using this approach have been reported (Boddey 

et al., 1983a and b; Danso et al., 1988; Edmeades and Goh, 1987; Vallis et 

at ., 1977). Boddey et al. (1984) found that frequent additions of small

33



doses of ,5N fertilizer ensured high 15N fertilizer enrichments in soil, and 

resulted in small declines in 15N /,4N ratio, compared to the single 

application. Frequent 15N additions may therefore assist in minimizing 

errors due to mismatch between reference and fixing crop. Rennie 

(1985;1986) has, however, criticised the use of repeated 15N additions on 

plots. However, the arguments and evidence he adduced in support of his 

claim appear unconvincing.

2.6.4 Sampling of plant material:

The 15N enrichments in various plants parts (e.g. seed, herbage, crowns and 

roots) frequently differ (Butler, 1987; Butler and Ladd, 1985; Fried et al., 

1983; Jensin, 1986; Ladd, 1981; Rennie etal., 1978). Measurements of 

nitrogen fixed based on only one plant part e.g. seed or leaf (Ruschel et al.,

1982) may therefore not adequately represent N2 fixed in the whole plant 

(Ladd, 1981, Phillips et al., 1983a). The sampling of plant material would 

not pose much problem in the case of annuals, but may be extremely 

difficult in trees, because of massive size o f trees, which makes it difficult 

to completely sample the whole tree. There is also the difficulty in

34



harvesting roots which in the case of NFTs may contain more than half of 

the N in the whole plant (Sanginga et al., 1990b). Significant N reserves 

occur in the roots of trees, and in NFTs, N reserves may account for a 

significant proportion of the N used for regrowth. Domenach and Kurdali

(1989) reported that, the reserves ofN  in roots of A lm s glutinosa 

accounted for 10% of the N in leaves at the end of the growing period.

Part of the sampling problems is also due to the heterogeneity in 1SN 

enrichments in different plant parts (Sanginga et al., 1990b; Shearer et al.,

1983). Leaves generally gives estimates of % Ndfa closest to that for the 

whole tree, and thus where %Ndfa is the most desired estimate, it seems 

appropriate to sample the leaves if the whole tree cannot be harvested.

Nitrogen cycling from litter decomposition influences the 15N enrichment 

in the rhizosphere (Jordan, 1985). With trees, litter fall and decomposition 

can be significant, and can consequently cause large changes in the !5N 

/14N ratios of soil under the fixing and non-fixing crops. With the 15N 

enrichment in a fixing crop being usually lower than that in a reference

35



crop, the expected net effect would be a lower 1SN /14N ratio in soil under 

the fixing crop, and therefore a higher than real estimate of N2 fixed.

2.7 INFLUENCE OF PHOSPHORUS ON NODULATION AND 

NITROGEN FIXATION

Phosphorus is a macronutrient element and it is an essential constituent of 

all living organisms. Apart from nitrogen, phosphorus has been reported as 

the most important limiting nutrient in Ghanaian soils (Nye, 1972). Most 

soils in Northern Ghana where most of Ghanaian cereals are produced are 

highly deficient in phosphorus (Nyamekye, 1987). Phosphorus fertilizers 

are therefore used to increase yield. This inherent low P level, according to 

Acquaye and Oteng (1982), is due to the nature of parent materials which 

are mainly sandstone and/or sesquioxide.

Several workers (Gate and Wilson, 1974; Jakobsen, 1985;01ofintoye, 

1986; Pereira and Bliss, 1987) have reported that a high phosphorus level 

is needed for maximum noduiation and nitrogen fixation in legumes, but 

the amount required for optimum nodulation and N2 fixation differs widely

36



between genotypes (Pereira and Bliss, 1987; Saleem and Kaufmann, 1986). 

Danso et al. (1992) have also reported the effect of phosphorus on 

nodulation, root proliferation and its overall beneficial effect on legumes. 

Sanginga, (1985) reported that, low level of phosphorus is among the main 

chemical constraints for establishing legume trees on tropical soils and 

that, the availability of phosphorus is closely associated with high 

nodulation and nitrogen fixation in tree legumes (Andrew, 1982; Sanginga, 

1985). Benge (1992) also reported that, acute phosphorus deficiency in 

most tropical soils is among the main nutritional constraints to the 

successful establishment of Leucaena spp on some selected soils of 

Nigeria. Sanginga et al., (1985) further indicated that leucaena plants 

require 80kgp/ha for good establishment even when they are inoculated 

with effective rhizobial strains.

Diem and Gauthier (1982), obtained increased nitrogen fixation of 

Casuarina equisetifolia by adding phosphate fertilizer to a phosphorus -  

deficient soil. Reddel et al. (1988) similarly found a 245% increase in 

wood volume of inoculated C. cunninghamians (field studies at Gypie, 

Australia) with the addition of phosphate to soil. Based on the results of

37



another study conducted in Australia, Reddel et al. (1986) concluded that 

low soil phosphorus status was a frequent limitation to nodulation of 

naturally occurring Casuarina and Allocasuarina species. Singleton et al. 

(1985) reported an increase in nodule dry weight and nitrogenous activity 

with increasing phosphorus supply. It has been suggested that the 

requirements of phosphorus for nodulation and maximum nodule activity 

are much greater than for host plant growth (de Mooy and Pesek, 1966). 

Independent of the source of phosphorus used, leaf area and seed 

production were found to be highest at 400mgP/kg (Bataglia and 

Mascarenhas, 1977). OfFei (1990), applying P at varied rates of 0, 40,80 

and 120kgP/ha in L leucocephela observed a significant increase in nodule 

and total P with increasing levels of P.

Genetic variation in the ability of the host legume as well as the rhizobia to 

grow in soils with marginal P content has been reported. By examining 23 

provenances of Gliricidia sepium and 11 isoline of L. leucocephela for 

growth in low P soils, 2.3-fold and 2.1-fold differences in growth, 

respectively, occurred between the L. leucocephela and G. sepium 

genotypes (Beck and Munns, 1984). Also, by assessing P nutrition of 23

38



strains of Rhizobium and 17 Strains of Bradyrhizobium, Beck and Munns, 

(1984) observed that, most strains tested were able to grow at P levels as 

low as 0.06micomolar, but significant strain and species variation was 

found in rhizobial growth response to low phosphorus. Smart et al., (1984) 

however, attributed this genotypic differences in low phosphorus tolerance 

of (brady)rhizohia to their ability to switch on efficient P uptake systems. 

The ability of Bradyrhizobium strains to acquire, store and utilize 

phosphorus under varying levels of soil P (luxury, sufficience and 

insufficiency), and their transfer to subsequent generations have also been 

examined (Smart et al., 1984; Cassman et al., 1981). Many workers have 

reported a close interaction between phosphorus and nitrogen. For 

instance, Hamissah et al. (1980) have reported that a combination 

107kgN/ha and 36kgP/ha produced the greatest economic yield in most 

legumes. In another experiment, nitrogen content in the plant was found to 

increase with increasing levels of nitrogen and phosphorus together 

(Sampet, 1978).

39



2.8 ENVIRONMENTAL FACTORS AFFECTING N2- FIXATION

The main environmental factors affecting nitrogen fixation can be grouped 

into (a) Physical (b) Chemical and (c) Biological factors.

2.8.1 Physical Factors.

Those physical factors that significantly influence N2-fixation include: 

temperature, drought and waterlogging.

2.8.1.1 TEMPERATURE

Temperature appears to be the most important physical factors affecting N2 

fixation. Although some Rhizobium strains have been reported to survive at 

temperature around 70°C in dry soil (Marshall, 1964), excessive high 

temperatures in general, can kill the majority of the bacteria in the surface 

layers of soil. In certain parts of the tropics, the surface soil temperature 

can occasionally reach 65-70°C and temperatures above 50°C can be found 

at a depth of 5cm (Dudeja and Khrana, 1989), sufficiently high enough to

40



inhibit germination of seeds and kill many bacteria. High temperatures can 

prevent nodulation, or if nodulation does occur, can inhibit the activity of 

N2-fixation in legumes (Day et al., 1978) even though root nodules may be 

insulated from the highest temperatures by the soil. Conversely, cool 

temperatures lead to delayed development of plants including delays in the 

formation of nodules, and so decrease rates of N2-fixation. Unlike many 

species of Cyanobacteria which can form spores which are highly resistant 

to desiccation, most heterotrophic free-living N2-fixers and rhizobia do not 

possess this capability. Survival of bacteria in soils at high temperatures 

appears to be improved by the presence of clay particles and soil organic 

matter, but many of the soils where temperatures are high are sandy. Day 

et al. (1978) reported rhizobial population of only 4-40 cells/g soil at a 

surface layer of 5cm soil and reported a population of up to 104 cells/g soil 

at a depth of 20-25cm below the surface. The optimum temperatures for 

growth and N2-fixation vary widely between legume species and reflect 

their environmental adaptation. Eaglesham and 

Ayanaba(1984),demonstrated the differences in environmental adaptation 

to high temperatures between rhizobia isolated from different climatic 

zones.

41



2.8.1.2 DROUGHT

Drought as a physical environmental stress has been noted to have drastic 

effects on nitrogen fixation in legumes. Rates of N2-fixation are more 

sensitive to reductions in soil water content than other processes such as 

photosynthesis, transpiration, or leaf growth rates (Sinclair et al., 1987). 

Even in species considered to be drought resistant e.g. Acacia spp, slight 

changes in the plant water potential caused a marked reduction in both the 

rate ofN2-fixation and in the translocation of the products of N2-fixation to 

the shoot (Rao and Venkateswarlu, 1987;Venkateswarlu and Rao, 1987). 

In general, it has been noted that, the numbers of rhizobia in soil decline 

drastically as soil dries. Bushby and Marshal (1977), in their comparative 

study of the survival of Rhizobium and Bradyrhizobium in drying soil, 

indicated that, Bradyrhizobium strains are more tolerant of desiccation than 

strains of Rhizobium over short periods. Other workers found no simple 

relationship between the desiccation tolerance of fast-or slow-growing 

rhizobia but they did find that specific strains of each survived in much 

greater numbers than others (Mahler and Wollum, 1981). Rhizobia

42



generally survive poorly on drying in soils which contain only small 

amounts of clay or organic matter (Chao and Alexander, 1982). Strains 

which survive under greater water stress are those which retain less water 

within the cells (Bushby and Marshall, 1977; Al-Rashidi et al., 1982).

2.8.1.3 WATER-LOGGING

Water-logging is also a major environment stress affecting ^-fixation. 

Whilst rhizobia are normally aerobic organisms, some strains of 

Bradyrhizobium and i t  meliloti possess a dissimilatory nitrate reductase 

which can function as an electron acceptor, and thus enable the bacteria to 

survive under anaerobic conditions (zablotowicz et al., 1978; Daniel et a l, 

1982). Because of the aerobic characteristic of rhizobia, the survival of 

rhizobia during long periods of flooding is of particular importance in 

cropping systems in which legumes are grown in rotation with rice. 

Toomson (1990), indicated that, the size of the population of rhizobia 

sampled from the field was generally larger ( 1 0 2- 1 0 4 cells g/soil) when the 

soil was moist or folly waterlogged as compared to when the soil was dried 

(<10-10’3 cells g/soil). In contrast, large reductions in the numbers of

43



rhizobia nodulating chicken Pea which are generally fast-growing rhizobia 

and which may not possess a dissimilatory nitrate reductase have been 

reported when grown after paddy rice (Toomson et al., 1982).

Lack of oxygen is also a major problem for root respiration and can rapidly 

result in loss of nitrogenase activity (Sprent and Gallacher, 1976; Witty et 

al., 1986). Waterlogging can result in the rapid release into the soil of 

certain heavy metals like iron and manganese which are highly toxic to 

both rhizobia and plants in high concentration.

2.8.2 CHEMICAL CONSTRAINTS

Among the chemical factors that significantly affect nitrogen fixation 

include toxicity and nutrient deficiency.

2.8.2.1 TOXICITY

The bacterial symbionts may be directly affected by the acidity of their 

environment. Bacteria with a greater capacity to regulate their internal pH

44



show increased survival rates at low pH (O’Hara et al., 1989). Aluminium 

toxicity, probably the most severe component of stress in acid soils, also 

has a major impact on rhizobial survival, and strains of rhizobia which 

were able tolerate a pH of 4.5 were not necessary tolerant of aluminium 

toxicity (Keyser and Munns, 1979b). The toxicity of aluminium to 

rhizobia may be due to aluminium binding to DNA and thus inhibiting 

DNA replication (Johnson and Wood, 1990). Strains of Rhizobium tolerant 

to aluminium had different DNA repair mechanisms to those found in 

strains susceptible to aluminium toxicity.

Other problems of acid soils such as manganese toxicity and calcium or 

phosphorus deficiency appear to have a lesser effect on Rhizobium survival 

(Keyser and Munns, 1979b). Low pH per se is not the principal cause of 

toxicity to plants. Rather, the problems of plant growth in acid soil result 

from aluminium and manganese toxicity. Franco and Munns, (1982), 

using solution cultures indicated that acidity and not aluminium was the 

principal constraint to nodulation.

45



Large differences in sensitivity to the toxic effects of acid soils have been 

found between different species of tropical pasture legumes (Andrew et al., 

1975; Andrew, 1976; de Carvalho et al., 1981). Several species of 

Stylosanthes are tolerant of concentrations of aluminium in solution which 

severely depress nodulation and plant growth of other species de Carvalho 

et al., (1981). de Carvalho et al., (1981), even reported differences in 

aluminium tolerance among different species of Stylosanthes. Stylosanthes 

is nodulated by direct infection mechanism and the lowered toxic effects of 

aluminium on nodulation may be related to a reduction in lateral root 

formation reducing the number of possible infection sites (de Carvalho et 

al., 1982). Clarkson, (1965) indicated that, the reduction in nodule 

initiation and development commonly seen both in Stylosanthes and in 

other legumes may be due to the reduction in meristematic activity of roots 

in solution high in aluminium.

2.S.2.2 NUTRIENT DEFICIENCY

Several of the nutrients essential for growth of plants or bacteria play 

specific roles in nodulation and /or N2 -  fixation. Thus, deficiencies in

46



these nutrients, or other nutrients essential for the growth of bacteria or 

plants can cause reduction in the numbers and size of nodules as well as 

the amount of N2 fixed (Giller and Wilson, 1991). For instance, Beck and 

Munns (1984), indicated that there is marked variation in the ability of 

rhizobia to grow in low concentrations of phosphorus which appears to be 

due to variation in the efficiency of phosphorus uptake systems (Smart et 

al., 1984). Strains of Rhizobium also differ in their ability to store 

phosphorus, but even in the most efficient case the amount of phosphorus 

stored can only support growth for 3-4 generations after removal of all 

phosphorus from the medium (Cassman et al., 1981; Beck and Munns, 

1984). In low concentrations of phosphorus, Beck and Munns (1985) 

indicated that, high concentrations of calcium are required for growth of 

rhizobia. It has been noted that, acute deficiency of phosphorus can 

prevent nodulation by legumes. Phosphorus and sulphur are required for 

nodule metabolism and tend to be concentrated in the nodules when the 

plant is deficient in these nutrients (O’Hara et al., 1988a). It has been 

suggested that, as nodulated plants often have less well-developed root 

system than unnodulated plants, the ability of nodulated plants to capture 

nutrients, particularly phosphorus, is decreased (Cassman et al., 1980).

47



Most legumes therefore depend heavily on mycorrhizas for efficient uptake 

of phosphorus (Hayman, 1986).

Mycorrhizas assist in uptake of phosphorus and other nutrients which are 

not mobile in the soil by increasing the volume of soil effectively explored 

by the plant. The degree of dependence on mycorrhiza for capture and 

uptake of phosphorus is partly determined by the root geometry of the 

legume; legumes with poorly-branched root system with few root hairs e.g. 

Leucaena leucocephala (Munns and Mosse, 1980), and Centrosema 

pubescens (Crush, 1974) tend to be more dependent on mycorrhizas than 

those plants with many long root hairs e.g. Lotus pedunculatus (Crush, 

1974). Many researchers have indicated that, nodulation and growth of 

legumes growing on phosphorus -deficient soil are often stimulated by 

mycorrhizas inoculation (Howeler et al., 1987; Arias et al., 1991).

Apart from the chemical nutrient deficiencies that affect nodulation and 

subsequent nitrogen fixation, there is increased evidence that pesticides 

used in agriculture can have adverse effects on the survival of rhizobia or 

on nodulation of legumes (Edwards, 1989; Roberts, 1991). For instance,

48



pollution of soils by heavy metals has been shown to completely suppress 

N2-fixation in white clover (Trifolium repens) due to the toxicity of the 

heavy metals to Rhizobium (McGrath et al., 1988; Giller et al., 1989).

2.8.3 BIOLOGICAL CONSTRAINT

The growth and survival of rhizobia in soil have been shown to be 

influenced by biological factors. Competition and antagonism by other 

micro-organisms are two biological factors influencing rhizobia survival 

and growth. Danso et al., (1975) indicated that, grazing of rhizobia in soil 

by protozoa is responsible for the reduction in the populations of rhizobia 

in soil and susceptibility of strains to bacteriophages may result in their 

poor survival (Barnet, 1980). Also, Danso et a/.(1974) indicated that 

sterilised soils support far larger population of inoculated rhizobia 

compared to unsterilised soils, implicating biological agents.

49



2.9. SOME METHODS FOR ESTIMATING NITROGEN FIXATION

An accurate method of measuring Nj-fixation is essential for any 

evaluation of the usefulness of different technologies, yet such a method 

has remained remarkably elusive. Despite active research on this subject 

since the discovery of ^-fixation, many measurements of the amount of 

N2 fixed in the field remain little better than informed guesses. Difficulties 

encounted in measuring N2-fixation in grain and pasture legumes are

magnified enormously when we consider trees. Of all the N2-fixing

symbiosis trees are the most problematic for measurement of the amount of 

N2-fixed, hence most research work have been concentrated on annual 

herbaceous plants than trees. Danso et al., (1992) have identified tow 

major reasons for this situation;

(a) The perennial nature of trees/shrubs

(b) The relatively large size of trees.

Despite the problems associated with nitrogen fixation measurements, 

several methods for estimating N2-fixation in plants are available although, 

all have some degrees of error. Most of these methods have been reviewed

50



by various researchers (e.g. Danso, 1985; Knowles, 1980; Rennie and 

Rennie, 1983). The methods that have been commonly used for estimating 

nitrogen fixation in trees are:

(1) The total N difference method (TND)

(2) Acetylene reduction assay (ARA)

(3) 15N isotope dilution method (ID).

2.9.1 TOTAL NITROGEN DIFFERENCE METHOD (TND).

This method is the most primitive and simplest of all the methods (Danso 

et al., 1992). This method involves quantification of total nitrogen in both 

the fixing plant and the reference crop plant (Danso and Hardarson, 1993). 

The technique operates upon the assumption that two genotypically similar 

plants of the same isolines that differ only in the ability to nodulate and fix 

nitrogen should absorb the same amount of soil N whether or not the 

nitrogen fixing species is actively fixing under uniform environmental 

conditions. Thus by analysing for the total N in the two plants, the amount

51



of N2 fixed could simply be estimated as the difference between the two 

(Danso and Herridge,1987)

i.e.: TN(fi) - TN(nfi) =Ndfa 

where:

TN(fi) = Total Nitrogen in the N2-fixing isoline.

TN(nfi) = Total nitrogen in the non-fixing isoline.

Ndfa = Amount of N2-derived from atmosphere (fixed N2)

One major problem with this technique is the difficulty in getting a non- 

fixing isoline for the reference. Indications are that non-fixing isolines of 

N2-fixing trees seem not to exist (Danso and Herridge, 1987). Hence, N2- 

fixing trees in non-fixing mode (e.g. uninoculated; inoculated, with killed 

or ineffective rhizobia; in soils devoid of homologous Rhizobium strains) 

have frequently been used as NFT references (Gautherir et al., 1985).

In some cases however, unrelated species have been used as NFT controls 

(Pareek et al., 1990) though it has been suggested that this could introduce

52



undesirable error in the N determination and consequently in the accuracy 

of the BNF estimates (Danso, 1985 Rennie and Rennie, 1983).

Furthermore, as a limitation to the assumption upon which the technique 

operates, it is probable that the fixing and non-fixing reference plants may 

be taking up different amount of soil N, particularly in N rich soils, which 

could be a major source of error (Danso, 1985; Rennie and Rennie, 1983). 

To minimize this error source therefore, it is recommended that soils with 

marginal N content should be used when applying the TND method 

(Danso, 1986). Thus serious problems or errors occur when this method is 

employed under conditions thought to be free of mineral N but which are 

in fact contaminated with N. For example, vermiculite was used as an N- 

free growth medium for the study of associative N2-fixation (Rennie and 

Larson, 1979), but later research showed that significant quantities of 

mineral N can be released from vermiculite when it is incubated under 

warm moist conditions (Giller et al., 1986).

Another limitation to this method is the fact that, the method becomes 

progressively inefficient and impractical as trees mature. Apart from the

53



difficulty in the recovery of all parts of the mature plant for analysis, 

including roots and litter, it is also difficult to account for the re-cycled N 

(Danso et al., 1992). According to Danso et al. (1992), this might be one 

reason for the observed gross differences between the TND approach and 

other techniques in BNF estimation.

Despite the above limitations to the TND approach, the method has the 

advantage of giving a measure of the total amount of N fixed over the 

length of the experiment and is indispensable for many laboratory based 

studies (Giller and Wilson, 1991).

2.9.2 ACETYLENE REDUCTION ASSAY (ARA).

This method has been used widely for estimating BNF in leguminous trees 

(Danso et al., 1992). In this method, the amount of N2 fixed is measure 

indirectly by analysing for the amount of ethylene produced (nitrogenase 

activity/acetylene reduction activity) by detached nodules or whole plants 

in a gas-tight container filled with acetylene gas over a given period of 

time.

54



Based on the theoretical electron flow relationship, it was established that 3 

moles of acetylene reduced are equivalent to 1 mole of N2 reduced or fixed, 

and therefore Hardy et al., (1973) proposed that the amount of acetylene 

reduced to ethylene be multiplied by a conversion factor of 3 to get the 

amount of reduced N.

There are, however, some serious limitations associated with this 

technique;

(a) The conversion factor of 3 does not conveniently apply in all cases 

(Hansen et al., 1987). For example, Turner and Gibson (1980) 

cautioned that environment and other plant effects acting independently 

on the nitrogenase reduction of acetylene and nitrogen can sufficiently 

alter the 3:1 ratio.

(b) Also the technique is said to be instantaneous and may 

not truly reflect BNF over a long duration (Fried et al., 1983).

Application of acetylene reduction assays for measurement of N2-fixation 

in soils is complicated by the effect of acetylene on other microbial 

processes. For example, acetylene has been found to block the last step of 

denitrification (Balderston et al., 1976) and autotrotrophic nitrificataion 

(Hynes and Knowles, 1978). In addition, acetylene blocks bacterial

55



oxidation of ethylene in soil so that “endogenous” ethylene accumulates 

(de Bonte, 1976; Nohrstedt, 1976). Thus, control treatments used to 

estimate background concentrations of ethylene in soil in which ethylene 

accumulation is measured in the absence of acetylene, greatly 

underestimate the accumulation of endogenous ethylene that occurs in the 

presence of acetylene. Witty (1979) demonstrated this clearly by using 14C 

labelled acetylene for ARA measurements of N2-fixation in soil cores. 

Only half of the ethylene that accumulated was from the labelled 14C and 

the remaining unlabelled ethylene must have come from the soil.

Other problems in the application of acetylene reduction assays for 

measurement of low rates of nitrogenase activity in soil result from the 

differences in solubility and rates o f diffusion of acetylene and ethylene in 

water (Van Berkum and Bohlool, 1980).

2.9.3 THE lsN-ISOTOPE TECHNIQUE.

This method holds promise as an effective and straight forward approach 

for measuring nitrogen fixation in trees. It is based on the differential

56



dilution of the 15N /14N isotope ratios in NFTs with varying N2-fixing 

potentials (Fried and Middelboe, 1977). Nitrogen in the atmosphere is 

virtually all 14N2 (99.633%) plus a small amount (0.367%) of the natural 

15N (Marrioti, 1983). Any substance which has an atom % 15N greater than 

that of the atmosphere (0.367%) is said to be enriched with 15N and the 15N 

enrichment is expressed as atom %15N excess. Similarly, a material 

depleted in 15N has an atom %15N below that of the atmosphere and it is 

said to be depleted 15N material. If a plant is grown in conditions where its 

sole source of N is fertiliser, which is entirely composed of 15N (100 atom 

% 15N), then all the N in the plant (apart from the amount that is originally 

present in the bacterial inoculum) will be 15N.

If the plant is able to fix dinitrogen (,5N2) from the atmosphere then the 

plant will have an atom %13N which is less than that of the fertiliser (i.e., 

less than 100 atom %I5N). This difference can be used to calculate the 

proportion of N derived from N2-fixation and is the underlying principle 

behind the isotope dilution method for measurement of N2-fixation (Giller 

and Wilson, 1991).

57



Fried and Middelboe (1977) established the following working formula for 

the calculation of %Ndfa in fixing plants;

%Ndfa = (1- %Ndfff fixer) x 100

%Ndff(Non fixer)

where;

%Ndfa = percent N derived from the atmosphere

by the fixing plant.

%NdfRTixer) = percent N derived from the 15N labelled

fertilizer by the fixing plant.

%Ndff(Non-fixer) = percent N derived from the 15N labelled 

fertilizer by the non-fixing plant.

Some basic assumptions in using this technique include:

i) That the 13N fertilizer enrichment is even with depth in the soils under 

text.

ii) That the N-fixing and the non-fixing reference plants have similar N- 

uptake patterns (Witty, 1984).

58



iii) That the 15N soil enrichments are the same under both 

the N2-fixing and non-fixing plant.

2.10 METHODS OF ESTIMATING SOIL MICROBIAL BIOMASS

Increasing awareness that the microbial biomass of soil constitutes a major 

nutrient sink has underlined the need to quantify soil biomass and to 

understand the dynamics of soil population. Several methods have evolved 

over the years for estimating soil biomass. Unfortunately technical 

limitations associated with each method obstruct the realization of these 

objectives. The various methods which have been used can be grouped 

under the following headings (i) Conventional methods (ii) Current 

methods.

2.10.1 CONVENTIONAL METHODS

The conventional methods include the following :

1. Direct counting

2. Plate count

59



3. Most probable Number method (MPN)

The direct counting method involves the direct observation of cells on agar 

plates. The cells are first stained and then subjected to microscopic 

examination. The main disadvantage of this technique is that it 

overestimates the microbial biomass due to : (1 ) the inability to distinguish 

cells from stained non-microbial organic particles and (2 ) the failure to 

distinguish living from dead cells (Skinner et. al., 1952).

The plate count method assumes that cells are fully dispersed into units 

after agitation such that every colony that developed arises from a single 

cell. The plate count method therefore underestimates the microbial 

biomass because after agitation some cells clump together with soil 

particles and are counted as one. Secondly, some cells are killed in the 

dilution medium and thirdly, there is the failure of certain spores to 

germinate. Further, the walls of the pipette often absorb some of the 

microbial cells (Gray and Williams, 1971).

60



The most-probable-number (MPN) method permits estimation of 

population density without an actual count of single cells of colonies. It is 

sometimes called the method of ultimate or extinction dilution or, less 

descriptively, simply the dilution method. The method is based on a 

determination of the presence or absence of micro-organisms in several 

individual aliquots of each of several consecutive dilutions of soil or other 

material. Informative discussions of the MPN method have been prepared 

by Halvorson and Ziegler (1933) and by Cochran (1950). The procedure 

involves a long period of incubation and is therefore time consuming. It is 

also less accurate as compared to the plate count since it is based on 

statistical assumption.

Generally, the conventional methods have the limitations of being 

inappropriate for comparing the relative contributions of different groups 

of soil microbes to the total soil microbial biomass and or activity. This is 

because the methods estimate numbers of cells and hyphae. However, 

fungal hyphae have higher protoplasmic weight than bacterial cell. Thus, 

in terms of numbers, bacteria have been estimated to exceed fungi in the 

ratio of 100:1 but in terms of weight, it is less than 1:1 (Gray and

61



Williams, 1971). Consequently, techniques have been developed to 

convert microbial numbers to biovolume and then to biomass giving a 

better index of soil microbial biomass and activity. However such 

conversions are time consuming and tedious.

2.10.2 CURRENT METHODS

Current methods for estimating microbial biomass are based on either 

direct microscopical measurement (Joones and Mollison, 1948; Paul and 

Johnson, 1977, Soderstrom, 1977), the fomigation-reinoculation principle 

(Jenkinson and Powlson, 1975) or the measurement of a cell component 

such as ATP (Verstraete et. al., 1983).

Anderson and Domsch (1978) proposed a physiological respiratory method 

based on C0 2 production, but caliberated on the basis of the fiimigation- 

reinoculation principle.

Sparling (1981) proposed a microcalorimetric method while Van de Werf 

and Verstraete proposed a method for estimating the active soil biomass, 

base on Oxygen consumption (physiological principle), but caliberated by

62



means of theoretical values that conform to microbial physiology and 

growth kinetics. These methods, were developed in an attempt to evolve 

simpler and more objective techniques for determining the soil microbial 

biomass (Anderson and Domsch, 1978).

The method based on the extraction of cell components often fails to fully 

extract the desired component and the relative recovery often depends on 

substrate characteristics or the type of extraction procedure used. Also, the 

quantity of a particular cell component can vary considerably with growth 

conditions and within different members of the microbial population. The 

ratio of specific cell component to the actual biomass is therefore not 

necessarily constant (Anderson and Domsch, 1978).

Although the respiration method gives biomass data within six hour, it is 

limited by the fact that a sensitive C 0 2 monitoring system is necessary to 

estimate hourly what is called the ‘maximum initial response’ of the soil. 

The physiological methods are also limited by the fact that a factor that 

links respiratory rate with biomass has not been fully derived (Anderson 

and Domsch, 1978).

63



2.10.2.1 THE CHLOROFFORM FUMIGATION TECHNIQUE

This method releases the carbon bound is microbial cells by minerlization 

and provides a means of calculating its weight (Jenkinson and Powlson, 

1976; Anderson and Domsch, 1978).

The technique needs no special equipment and requires only the titration of 

C02 absorbed in alkali ( IN NaOH). However, it requires a relatively long 

period of incubation (at least 1 0  days) before analysing for the carbon 

dioxide released. It offers a better means for estimating soil microbial 

biomass and also estimate the relative contribution of the different 

physiological groups to the total microbial biomass (Lynch and Panting, 

1980; Anderson and Domsch, 1978).

The technique, however, breaks down in strongly acid soils (pH<4.5) and 

further, the choice of an appropriate ‘K’ factor which is used in the formula 

for calculating biomass is still a matter of controversy and experimental 

research.

64



2.10.2.2 SOIL ATP

Soil ATP concentration provides another satisfactory way of obtaining 

microbial biomass in strongly acid soils. The soil ATP concentration is 

determined by the luciferin-luciferase assay, using a TCA-phosphate- 

paraquat extractant (Jenkinson and Oades, 1979) as modified by Tate and 

Jenkinson (1987).

The extractant: soil ratio is 10:1 and soil is ultrasonified for 2 min with a 

20 kHz, HOW, MSE sonifier with a 12.5mm dia probe operating at full 

power.

2.10.2.3 FUMIGATION EXTRACTION METHOD

A new, and probably the best, way of measuring microbial biomass in 

strongly acid soils is by fumigation extraction (Vance et al., 1987). It was 

proposed that the organic C rendered extractable to 0.5 M K2SO4 after a 24 

hour chloroform-fumigation (Ec) comes from the cells of the microbial

65



biomass and can be used to estimate soil microbial biomass C in both 

neutral and acid soils.

The Ec (the difference between organic C extracted by 0.5 M K2SO4 from 

fumigated and non-fumigated soil) is closely related to the microbial 

biomass C measured by fumigation-incubation according to the equation:

Biomass C = (2.64 + 0.60) Ec.

However, since this method has only been tested on 10 soil samples, of 

which two were taken at different times from the same sites, its validity, 

particularly on soils of high clay content, and on a wider range of soils, is 

yet to be ascertained.

This new method may prove useful in acid soils, in freshly sampled soils 

and in soils recently amended with substrates, where the fumigation- 

incubation technique breaks down.

6 6



CHAPTER THREE

MATERIALS AND METHODS.

3.1 LOCATION OF STUDY AREA.

Most of the experiments were carried out in small nursery bags set up on 

raised benches covered with polythene canopies situated in an open space 

in front of the Department of Soil Science, University of Ghana, Legon.

3.2 SOILS AND SITE CHARACTERISTICS:

Three soil types taken from the Accra plains were used for the study. The 

soils belong to the Toje, Hatso and Alajo series (local Names) (Brammer, 

1967). The three soils occur on the same soil catena with Toje series being 

at the top and Hatcho and Alajo series being at the middle and bottom 

respectively.

67



Toje soils consist of more than 30 inches of red sandy loam to light clay, 

may overlie red iron-concretionary clay or iron pan at depth or indirectly 

overlie ferruginized weathered rock, mainly Togo quartzite schist 

(Brammer, 1967). The soils are well drained and absorb water readily 

except when the surface is left bare.

Hatcho soils consist of several feet of pale brown sand increasing to sandy 

clay with depth, humus-stained near the surface and slightly mottled 

orange in the subsoil, but occasionally found with seepage iron pan at a 

depth exceeding 2ft (Brammer, 1967). The soils ‘pact’ on exposure and 

the ground-surface becomes hard and impermeable. The soils have little 

retentive power and dry out rather deeply during the main dry season.

Alajo series consist of soils developed from alluvium in a marine terrace, 

are grey-brown, heavy plastic swelling clays containing lime concretions in 

the subsoil (Brammer, 1967). The soils occur on valley flats along a 

tributary of the Odaw stream which drains the area.

The three soil series occur under dry climatic condition with mean annual 

rainfall of about 1 1 0 0 mm which is distributed bi-modally with the long

68



raining peak in May/June and short peak season in September/October. 

The mean annual temperature is about 31°C.

The three soils were all collected under native grassland vegetation,

3.2.1 SOIL SAMPLING AND PREPARATION:

The soil samples were taken as cores from a depth of 0-15cm after clearing 

the vegetation cover after which the soil samples were bulked. The soil 

were air-dried, pulverised and sieved through a 2 mm diameter sieve.

3.2.2 SOIL ANALYSIS.

The soils were analysed for the following parameters; pH (in water), total 

phosphorus, available phosphorus, and total nitrogen.

3.2.2.1 pH (Water):

69



Ten grammes of sieved soils were weighed into a beaker and 20ml of 

distilled water added to give a ratio of 1 soil: 2 water. The mixture was 

vigorously stirred continuously for 30minutes after which it was left to 

stand for about 1 hour until most of the clay particles were settled and also 

to allow the mixture to attain room temperature. The glass electrode -pH 

meter was standardised using two aqueous solutions of pHs 4 and 7 and 

then later used to measure the pH of the prepared soil sample by carefully 

immersing the glass electrodes in the supematent. The samples were 

replicated four times and the average pH recorded.

3.2.2.2 TOTAL PHOSPHORUS:

Five grammes of the sieved soil samples were weighed into 25ml conical 

flask. Ten millilitres of concentrated HN03 (aq) was added to the content 

of each flask, followed by 15ml of 60% HC104 The contents of the flasks 

were digested in a digestion cupboard until the solution became clear and 

the dense white fiimes of HCIO4 had ceased. The digests; were then 

cooled, diluted with distilled water and filtered into 250ml flasks. The 

filtrates were then made up to volume and 5ml aliquot each taken for

70



phosphorus content determination using the molybdate ascorbic acid 

method of Watanabe and Olsen ( (1965) involving calorimetry.

The formula below was used to calculate the phosphorus content

P(ppm) = (R-B) X Ve 

W XVa

Where: R = Spectrometer reading of sample

B = Spectrometer reading for blast 

Ve= Volume of extractant 

W = Weight of soil used 

Va = Volume of aliguote.

3.2.2.3 AVAILABLE PHOSPHORUS ( BRAY 1 METHOD)

Five grammes of the sieved soil samples were weighed into extraction 

bottles followed by 30ml of the extractant (O.3 HN4F + 0.1HC1). The 

bottles were then shaken thoroughly for 2 minutes on a mechanical shaker.

71



The soils suspension were then filtered and 5ml aliquot of the filtrates 

taken for phosphorus analysis. The Phosphorus determination was done 

colorimetrically using the Watanabe and Olsen (1965) modified technique.

3.2.2.4 TOTAL NITROGEN:

Ten grammes of the sieved soil samples were weighed into 300ml kjedhal 

flasks and moistened with distilled water. Tablets of selenium were added 

into the content of each flask, followed by 2 0 mls of concentrated 

H2S04(aq).

The mixtures were then cooled and transferred with distilled water into 

250ml volumetric flasks and made up to volume. Five millilitres aliquot 

was taken into a Markham distillation apparatus followed by 5ml of 40% 

NaOH (aq). The mixture was distilled and the distillate collected into 50ml 

flask containing 5ml of 2% H3BO4 (aq) indicator to which three drops of 

methyl red-methyl blue indicator had been added. The green solution 

formed was titrated against 0.01NHC1 to a purplish-red end point. The 

percent total nitrogen was calculated using the following formula:

72



Weight of soil X aliquot of digest used.

%N = 0.01 X 50 X 0.014 (titre value)

3.3 PLANT GROWTH MEDIA:

Each of the polythene bag used had a length of 15cm with an interior 

opening of diameter 7cm, and posterior sealed end but with three 

perforated holes.

Each nursery bag was filled with one kilogramme of a type of soil and each 

nursery bag with its content were placed in a plastic saucer after which the 

soil were kept semi-moist by distilled water prior to planting.

3.4 PLANTING MATERIALS:

Seeds of fourteen perennial tree and shrub legume species were collected 

from the Botanical Gardens, University of Ghana, Legon and the Kwadaso 

Agricultural Research Station. The tree species used for the initial 

screening exercise included: Albizia lebbek, Sesbania specioca. Sesbania

73



aculiata, Sesbania rostrat