BOTSWANA UNIVERSITY OF

AGRICULTURE AND NATURAL

RESOURCES

Comparative assessment of the effects of horse manure and urea as nitrogen

sources on seed yield, forage production and nutritional quality of Cenchrus
ciliaris post seed harvesting.

A dissertation submitted in partial fulfilment of the academic requirements of

Master of Science Degree in Animal Science (Nutrition).

By

Tinieri Maeresera (201400206)

MAIN SUPERVISOR
Dr. M. Letso

CO-SUPERVISORS
Prof. Madibela. O. R
Dr. Tshireletso. K

Department of Animal Science and Production

2020

1
 DECLARATION

I declare that all the work contained in this dissertation is my own independent investigation

for the Master of Science Degree in Animal Science which I did at Botswana University of

Agriculture and Natural Resources from January 2015 to March 2020. All the sources used

have been quoted and acknowledged by means of references. The work has not been previously

submitted and shall not be submitted to any other university for the award of any other degree

or diploma.

-------------------------------------------------------
Tinieri Maeresera
2020

2
APPROVAL

Main Supervisor’s name Date Signature

--------------------------------- ------------------------ ----------------------

Co-supervisor’s name Date Signature

---------------------------------- ------------------------- ------------------------

Co-supervisor’s name Date Signature

---------------------------------- --------------------------- ------------------------

3
ACKNOWLEDGEMENTS

I would like to express my sincere gratitude to Drs Letso. M, Tshireletso. K and Prof
Madibela. O. R for their supervision. Their support, encouragement and guidance made it
possible for me to complete this research.

I am very grateful to the Principal and Managing Director of Livingstone Kolobeng College,
Mrs Nilima Bakaya and Mr Gary Wills respectively, for allowing me to use part of the College
land for the agronomic part of my study.

I would also like to acknowledge the assistance that I received from the BUAN laboratory
technicians and my colleagues in the Animal Nutrition Department.I am indebted to my wife,
Sibongile Maeresera, my daughter Kudzai and son Aniel for encouraging me throughout the
duration of this study.

Most importantly, I would like to thank God for giving me the strength and resilience during
the duration of this study.

4
TABLE OF CONTENTS
PAGE
ACKNOWLEDGEMENTS ....................................................................................................... 4
ABSTRACT............................................................................................................................... 8
LIST OF TABLES ................................................................................................................... 10
LIST OF FIGURES ................................................................................................................. 11
ACRONYMS AND ABBREVIATIONS ................................................................................ 11
CHAPTER 1 ............................................................................................................................ 13
1.0 Introduction .................................................................................................................... 13
1.1 Livestock production on natural pastures in Botswana .................................................. 13
1.2 Characterization of desirable forage grass ..................................................................... 15
1.3 C. ciliaris (Molopo variety) as a forage grass of choice ................................................ 16
1.4 Factors limiting the use of C. ciliaris as a forage grass in Botswana at present ............ 18
1.5 Experiences of Cenchrus elsewhere from its centre of domestication ........................... 19
1.6 Opportunities of growing Cenchrus as livestock fodder crop........................................ 20
1.7 Statement of problem ..................................................................................................... 21
1.9 The objectives of the study are:...................................................................................... 24
CHAPTER 2 ............................................................................................................................ 25
2.0 Literature Review ........................................................................................................... 25
2.1 Botswana’s natural grassland ......................................................................................... 25
2.3 The use of nitrogenous fertilizers to improve seed production ...................................... 27
2.4 Importance of nitrogen in forage grasses ....................................................................... 28
2.5 Possible sources of Nitrogen for pasture fertilization .................................................... 31
2.5.1 Legumes................................................................................................................... 31
2.5.2 Inorganic fertilizers.................................................................................................. 31
2.5.3 Organic manures ...................................................................................................... 32
Figure 1: Digestive system of a horse where microbial protein synthesis and digestion
occur after the main absorptive region of the small ......................................................... 36
Figure 2: Goat digestive system. Most microbial protein digestion occurs before the
small intestines (Dutta. K.S, 2002) ................................................................................... 35
2.5.4 Urea ......................................................................................................................... 37
2.6 Effect of nitrogen fertilization on the nutritional quality of C. ciliaris at seed harvesting
stage ...................................................................................................................................... 38
CHAPTER 3: COMPARATIVE ASSESSMENT OF THE EFFECTS OF FERTILIZING C.
CILIARIS WITH HORSE MANURE VERSUS FERTILIZING WITH UREA ON THE
SEED YIELD AND FORAGE BIOMASS PRODUCTION .................................................. 41
3.0 Introduction .................................................................................................................... 41
5
 3.1 Specific objectives.......................................................................................................... 43
3.2 Hypothesis ...................................................................................................................... 43
3.3 Materials and Methods ................................................................................................... 44
3.3.1 Site description ........................................................................................................ 44
3.3.2 Soil sampling and testing ......................................................................................... 44
3.3.3 Experimental design ................................................................................................ 44
Figure 3: Arrangement of plots in the experimental garden ............................................ 45
3.3.4 Land preparation, fertilizer application, planting and management practices ......... 45
3.3.5 Data collection ......................................................................................................... 46
Figure 5: Inspecting the horse manure upon its delivery. ................................................ 47
Figure 6: Fertilizer application ........................................................................................ 48
Figure 7: Planting of the vegetative splits........................................................................ 48
Figure 8: Appearance of tussocks after one week. ........................................................... 49
Figure 9: Appearance of the grass after 22 days ............................................................ 49
Figure 10: Top dressing grass with urea after emergence of the first inflorescence ....... 50
Figure 11: Appearance of the grass after 2 months ......................................................... 50
Figure 13: Pests of grass seeds ....................................................................................... 51
Figure 14: Appearance of C. ciliaris after 3 months ....................................................... 51
Figure 15: Counting the number of tillers per crown ...................................................... 52
Figure 16: Measuring the length of the inflorescences .................................................... 52
Figure 17: Harvesting seed heads .................................................................................... 53
Figure 18: Weighing the seed heads ................................................................................ 53
Figure 19: Cutting the vegetative material for weighing. ................................................ 54
Figure 20: Measuring the forage biomass ....................................................................... 55
Figure 2.1: Air-dried seed heads ...................................................................................... 55
Figure 22: Oven drying the vegetative material .............................................................. 55
3.4 Statistical Analysis ......................................................................................................... 56
3.5 Results ................................................................................................................................ 57
Table 3.1: Soil analysis results ......................................................................................... 57
3.5.1 Length of inflorescence ........................................................................................... 57
3.5.2 Tiller density ............................................................................................................ 57
3.5.3 Seed yield ................................................................................................................ 57
3.5.4 Forage biomass yield ............................................................................................... 58
Table 3.2: Effects of fertilizer treatments on agronomic parameters of C. ciliaris. ......... 58
3.6.2 Number of tillers per crown (tiller density) ................................................................ 59

6
 3.6.3 Seed Yield ................................................................................................................... 61
3.6.4 Fresh Biomass Yield ................................................................................................... 63
3.6.5 Dry Matter Yield ......................................................................................................... 62
CHAPTER 4: Chemical composition and in vitro digestibility of C. ciliaris forage grass
post-harvest residues fertilized with horse manure or urea ............................................. 67
4.0 Introduction .................................................................................................................... 67
4.1 Specific objectives.......................................................................................................... 68
4.2 Hypothesis ...................................................................................................................... 69
4.3 Materials and methods ................................................................................................... 69
4.3.1 Determination of chemical composition of the forage grass residue ...................... 69
4.3.2 In vitro digestibility determination .......................................................................... 72
4.4 Statistical analyses.......................................................................................................... 73
4.5 Results ............................................................................................................................ 74
4.5.1 Nutritional quality of the forage residue ..................................................................... 74
Table 4.1: Effects of fertilizer treatments on chemical composition and DM digestibility
(%DM) of C. ciliaris after seed harvesting ..................................................................... 75
4.6 Pearson’s Correlation Coefficients between crude protein and ADF, NDF and IVDMD
.............................................................................................................................................. 75
Table 4.2: Correlation coefficients between ADF, NDF, IVDMD and CP. It shows a
positive correlation between CP and Dry Matter digestibility. ........................................ 76
Table 4.3: Mineral content (mg/kg DM) of C. ciliaris forage residue after seed
harvesting.......................................................................................................................... 77
4.7 Discussion ...................................................................................................................... 78
4.7.1 Crude Protein............................................................................................................... 78
4.7.2 Dry Matter Digestibility .............................................................................................. 80
4.7.3 Neutral Detergent Fibre (NDF) and Acid Detergent Fibre (ADF).............................. 81
4.7.4 Minerals ....................................................................................................................... 83
4.7.5 Conclusion ................................................................................................................... 85
Chapter 5 .................................................................................................................................. 86
General conclusions, limitations and future research........................................................... 86
REFERENCE ........................................................................................................................... 88
APPENDICES ......................................................................................................................... 98
Agronomic Parameters............................................................................................................. 98
Mineral Composition.......................................................................................................... 102

7
ABSTRACT

The objective of this research was to evaluate the comparative effects of horse manure and
urea, as sources of nitrogen, on seed and biomass yields of Cenchrus ciliaris and the nutritional
quality of its forage residues after harvesting seeds. Two studies were carried out for the
investigation. In the first study, 18 plots of C. ciliaris, grown from tussocks under irrigation,
were used in a replicated 6 x 3 completely randomized design with 4 months experimental
period. Six plots of C. ciliaris were assigned to each of the three treatments of no fertilizer,
fertilizing with horse manure and fertilizing with urea. After four months, seed yield, forage
dry matter yield, number of tillers per crown and lengths of inflorescences of the C. ciliaris
were measured. Data were analyzed using the General Linear Model Procedures in Statistical
Analysis System (SAS). Fertilizing C. ciliaris with either horse manure or urea caused a
significantly (P < 0.05) higher seed yield, forage biomass yield and number of tillers per
crown than the control treatment. There was no significant (P > 0.05) difference in the length
of inflorescence of C. ciliaris between the two fertilizer treatments and the control. Fertilizing
C. ciliaris with horse manure caused a significantly (P < 0.05) higher seed yield (18.46 ± 1.42
kg/ha) than fertilizing with urea (14.51 ± 1.92 kg/ha). There was no significant (P > 0.05)
difference in number of tillers per crown between fertilizing with horse manure and fertilizing
with urea. Fertilizing the C. ciliaris using urea caused a significantly (P < 0.05) higher forage
biomass yield (2739.62 ± 274.42 kg dry matter/ha) than fertilizing with horse manure
(2237.85 ± 118.99 kg dry matter/ha). This research, therefore, showed that fertilizing C.
ciliaris with horse manure was superior to fertilizing it with urea in terms of increasing the
seed yield but fertilizing it with urea was superior in terms of forage dry matter yield. If the
objective is to produce seed, horse manure would be the source of nitrogen to use and if the
target is to produce higher dry matter yields, urea would be preferable, all other factors held
constant. In the second study, the C. sciliaris forage residues from the 18 plots in the first study
were analyzed for nutritional quality after the seeds were harvested. It involved analyzing the
residues for Crude Protein (CP), Acid Detergent Fiber (ADF), Neutral Detergent Fiber
(NDF), Ash, minerals and In vitro Dry Matter digestibility (IVDMD). Fertilizing C. ciliaris
with either horse manure or urea caused significantly (P < 0.05) higher CP and IVDMD than
the control treatment. There was, however, no significant (P > 0.05) difference between
fertilizing with horse manure and fertilizing with urea in terms of the two parameters.
Fertilizing the grass with either horse manure or urea caused significantly (P < 0.05) lower
ADF, NDF and ash contents than the control treatment. There was no significant (P > 0.05)

8
difference between fertilizing C. ciliaris with horse manure and fertilizing with urea in terms
of these nutritional quality parameters. In terms of minerals, fertilizing with either horse
manure or urea caused no significant (P > 0.05) difference from the control treatment, except
in sulfur content where fertilizing with urea resulted in significantly (P < 0.05) higher levels
(0.435% ± 0.018 dry matter) than the control treatment (0.230% ± 0.018 dry matter). It became
evident from this study that there was no significant difference between fertilizing using horse
manure and using urea in terms of increasing the nutritional quality of the forage residues of
C. ciliaris. The two fertilizer treatments, however, increased the quality of the residues by
increasing their CP content. Both fertilizer treatments also reduced the fibrosity of the forage
residues and this could have contributed to increased digestibility of the forage residues. The
results of this research show that C. ciliaris seed production can be done in the semi-arid areas
of Botswana using horse manure and be used to establish a seed bank. Such a seed bank could
then be used as source of forage propagation material to improve pastures. As further research,
on-farm trials may need to be performed beyond the small plots experiment undertaken in this
study in order to validate the adaptability and productivity of the grass under real field
conditions across varied soils, moisture conditions and seasons.

Keywords and phrases: nutritional quality, forage residues, forage dry matter yield,
digestibility, forage biomass

9
LIST OF TABLES

Table 3.1: Soil analysis results 57

Table 3.2: Effects of fertilizer treatments on agronomic parameters of C. ciliaris. 57

Table 4.1: Effects of fertilizer treatments on chemical composition and DM digestibility

(%DM) of C. ciliaris after seed harvesting 76

Table 4.2: Correlation coefficients between ADF, NDF, IVDMD and CP 77

Table 4.3: Mineral content (mg/kg DM) of C. ciliaris forage residue after seed harvesting 78

10
LIST OF FIGURES

Figure 1: Digestive system of a horse 35

Figure 2: Arrangement of plots in the experimental garden 44

Figure 3: Inspecting the horse manure upon its delivery. 47

Figure 4: Fertilizer application 48

Figure 5: Planting of the vegetative splits 48

Figure 6: Appearance of the grass seedlings one week after planting 49

Figure 7: Appearance of the grass after 22 days 49

Figure 8: Top dressing grass with urea after emergence of the first inflorescence 50

Figure 9: Appearance of the grass after 2 months 50

Figure 10: Pests of grass seeds 51

Figure 11: Appearance of Cenchrus grass after 3 months 51

Figure 12: Counting the number of tillers per crown 52

Figure 13: Measuring the length of the inflorescences 52

Figure 14: Harvesting seed heads 53

Figure 15: Weighing the seed heads 53

Figure 16: Cutting the vegetative material for weighing. 54

Figure 17: Measuring the forage biomass 54

Figure 18: Air-dried seed heads 55

Figure 19: Oven drying the vegetative material 55

ACRONYMS AND ABBREVIATIONS

ADF acid detergent fiber

11
ANOVA analysis of variance

AOAC association of analytical chemists

ATP adenosine triphosphate

BUAN Botswana University of Agriculture and Natural Resources

Ca2+ calcium ion

CEC cation exchange capacity

CP Crude Protein

Cu2+ copper ion

DMD dry matter digestibility

DM dry matter

Fe2+ ferric ion

IVDMD in vitro dry matter digestibility

IVTDMD in vitro total dry matter digestibility

K+ potassium ion

Mg2+ magnesium ion

NDF neutral detergent fiber

Zn2+ zinc ion

PDIFF Piecewise differentiable

12
CHAPTER 1
1.0 Introduction
1.1 Livestock production on natural pastures in Botswana

Development of the livestock production enterprise as an economic sub-sector in Botswana is

hindered by the unavailability of good quality feed at critical stages of the production cycle.

The main sources of livestock feed in the country are natural pastures and crop residues which

are low in quantity and quality for sustainable and profitable livestock production during the

dry winter period for about 6 months (Madibela et al., 2002). In Botswana, goats raised on

pasture and mated at the end of the wet season (April) are exposed to adequate nutrition for

enhanced sexual activity which results in multiple ovulation and, hence, multiple births

(Madibela and Segwagwe, 2008). The rest of the pregnancy, however, goes through a period

of a shortage of nutrients. This is because most of the native grasses in the rangelands in

Botswana are so low in quality that they result in slow growth rates, poor fertility and high

mortality rates in ruminant livestock. Walker (2013) was in agreement with this sentiment as

she reported that ruminant livestock in Botswana is largely dependent on the communal range

or natural pasture and is constrained by fodder scarcity. This natural pasture which consists

mainly of grass, is high in fibre but too low in crude protein to meet animal nutritional

requirements (Walker, 2013). Most of the rangelands are just large tracts of arid and semi-arid

lands that are not suited for rain-fed crop production due to their low crop production potential.

The soil has a low humus content which makes it not to retain water long enough for crops to

absorb. It also has a low nutrient content due to excessive leaching. The leaching causes the

soil to be acidic and cannot promote growth of crops. Kolawole et al. (2013) reported some

results of laboratory analysis of 33 composite soil samples collected from 30 farmers in 3

farming communities of Makalamabedi, Nokaneng and Mohembo, situated in Ngamiland

East, Maun region, in the northern part of Botswana, which showed that most soils are low

13
in nutrients as well as in cation exchange capacity. The land has, instead, been used for

livestock production. In these areas, it is mainly extensive range management that is

practised. Animals graze and browse over large areas that are not fenced. The productivity

of the livestock in these areas has, however, remained relatively low (Kolawole et al., 2013).

This has been attributed to low quantity and poor quality livestock feed in general and forage

grasses in particular.

It has been very difficult for the range to be improved through introduction of more

valuable forage grasses due to unavailability of planting material in the form of seed. There are

no established seed banks for better performing and higher quality forage grasses and this has

been the most discouraging factor to farmers. The soils in Botswana are predominantly sandy

textured, with very low organic matter content, resulting in deficiencies in essential plant

nutrients (Kolawole et al., 2013). The grasses that grow in such soils inherently show

deficiencies in these essential nutrients as well since they can only acquire nutrients that are

available in the soil (Koech et al., 2014). Most farmers use the extensive and semi-

intensive systems of livestock production and fertilization of the pastures is not

practised. Most parts of the country receive below average and quite variable rainfall. The

low fertility of the soil and the low rainfall has resulted in the sparse and low quality forage

grasses in the natural pastures. If seed material for over-sowing in pastures can be made

available, the forage grass quantity can be improved. The seed material should be from high

quality grass, such that performance of livestock grazing on such pastures could be enhanced.

Some lands, like sand dunes, which occupy large areas in Botswana, abandoned crop lands and

degraded rangelands, hardly have any forage to sustain livestock. When the forage grasses in

a pasture are sparse and of low quality, large areas of land would be required to sustain few

animals. Before even considering the quality of the forage grasses in the pastures, it may be

important to be able to increase the quantity of the available fodder through seed bank

14
establishment and making the seed available even in the remote areas where communal farmers

rear their livestock. It is important to identify grass species that can be appropriate for the

improvement of pasture for livestock in terms of its biomass production and its nutritional

quality. It would be an added advantage if the grass species can also be useful for reclamation

of wasteland and stabilization of sand dunes. In this context, the grasses would assume

importance, not only as livestock feed, but also as soil builders and binders and aid in soil

conservation (Parwani, 2013).

Growing demand for pasture grass seed and establishment material in the country and

neighbouring countries necessitates researchers to respond in order to create an opportunity for

pasture seed supply industry. In Botswana, unfortunately, the distribution of material for

propagation, whether it is the seed or vegetative, is seldom a commercial enterprise. On a local

basis, a research agency or a farmer may sell the material but usually on a small scale. The

principal issues related to pasture development in Botswana, therefore, can be

highlighted as reluctance by government to sponsor research and unavailability of seed

banks. To compound the problem, there is limited data available on the grasses response to

fertilizers. The little literature that is available is void of any data on seed production or seed

quality responses to fertilizer under conditions prevailing in Botswana.

1.2 Characterization of desirable forage grass

In order to solve the shortage of feed and increase livestock productivity, it is necessary to

introduce and cultivate high quality forages with high yielding ability and adaptability to the

biotic and abiotic environmental stresses (Hare et al., 2009).The suitable grass types would be

those that respond very quickly to any availability of moisture, no matter how slight. The

selected forage grasses should produce sufficient seed for perpetual reproduction. It should also

be of high nutritive value in terms of nutrient content, especially nitrogen and phosphorus.

According to Ashraf et al. (2013) the dry matter crude protein content should be between 6%
15
and 18% while phosphorus content should be between 0.15% and 0.65%. Ruminants require

6.9% protein for maintenance, 10% for beef production and 11.9% for milk production (Ashraf

et al., 2013). Ramirez et al. (2009) reported that growing beef requires about 0.3% dry matter

phosphorus and about 12% dry matter protein. A report from Animal Production and Research

Unit (APRU) (1977) showed that phosphorus was always below 0.03% on dry matter basis in

natural range forage while crude protein averaged 8.5% in dry seasons and 9.4% in wet seasons

in Botswana. This means that without any improvement, the ruminants would not get sufficient

crude protein for production.Its digestibility should be reasonably high if it is to be of value to

livestock. According to Donaldson and Rootman (2010), the digestibility of forage grass should

not be less than 53.8%.

1.3 C. ciliaris (Molopo variety) as a forage grass of choice

One very clearly promising forage grass in Botswana is C. ciliaris. It can play an important

role in providing a significant amount of quality forage both under smallholder farming and

intensive livestock production systems (Hassan et al., 2015). It is native to Southern Africa and

India (Koech et al., 2014). Its drought tolerance and productivity led to its uptake and by the

pastoral industry in some countries like Australia, Canada and New Zealand (Parwani, 2013).

Unfortunately, its sites in Botswana are usually not more than isolated clusters of one to a few

closely spaced individuals. Its populations are extremely fragmented. According to Akiyana et

al. (2005), C. ciliaris is always observed growing in dense monotypic stands, small clumps or

lone tussocks throughout the landscape. Unprotected, these clusters are heavily and selectively

grazed and extremely scarce as the grass is relatively more palatable than most native grasses

(Seddi et al., 2002). In some countries like Mexico (North-western Mexico), natural desert-

scrub and thorn-scrub have been converted to C. ciliaris pastures successfully (Seddi et al.,

2002). C. ciliaris is gaining attention in various fields of research, as it is quite competitive

under conditions of high temperature, solar radiation and low rainfall (Bulle et al., 2011). These

16
are the conditions that are obtaining in Botswana. It is more efficient at gathering carbon

dioxide and utilization of recycled nitrogen from the soil (Akiyama et al., 2005). The grass has

proved useful for pasture and soil retention in a wide range of environments due to its drought

tolerance, deep and clustered root system, rapid response to availability of moisture, relative

palatability and resistance to overgrazing (Arshadullah et al., 2011). The larger and deeper root

system makes it capable of providing greater strength against soil erosion than other subtropical

grasses (Bulle et al., 2011). The swollen stem base accumulates carbohydrates, allowing it to

survive drought spells and to sprout after burning and at the commencement of rain

(Arshadullah et al., 2011). Its drought hardiness makes it an ideal choice for dry communities

and for colonization of disturbed sites (Kizima et al., 2012). It produces viable seed so that

stands can be self-replacing and pastures may not need to be reseeded. According to Akiyama

et al. (2005), C. ciliaris is generally apomictic, although some which propagate sexually have

been identified. Its seed spreads easily by wind, along water courses and human and animal

traffic. It is successfully propagated by vegetative splits, rhizomes and stolons and this makes

it quite useful in stabilizing disturbed areas.

According to Bhattarai et al. (2008), the use of seed for sloping areas like railway

batters to control soil erosion is a major concern where the development of good grass cover

within a short period of time is required to minimize the risk of damage from storms. This is

where vegetative propagation capability of C. ciliaris comes into play. Its tufted nature and

having buds close to the ground makes it tolerant to heavy grazing and trampling (Arshadullah

et al., 2011). Its persistence and ease and low cost of establishment add to its competitive

advantage. Its vigorous regenerative capacity after a fire is another of its major assets (Martin

et al., 1999). Its seed germination is inherently poor, normally not exceeding 18% when the

seeds are not treated with growth hormones or by scarification and becomes unpredictable,

especially in semi-arid regions characterized by low rainfall, due to its extended dormancy

17
(Bhattarai et al., 2008). According to Brummer (2009), however, germination of Cenchrus

grass seed can be improved by treating it with GA3 growth regulator at 10 ppm at pH 7.0. It

can also be improved by soaking the seeds in 70% sulphuric acid for 5 minutes and washing in

running water (Brummer, 2009). The seeds can also be treated with Indole Acetic Acid at 10

ppm for 8 hours (Brummer, 2009).The grass’s regenerative capacity after a fire and

commencement of rain is one of its greatest assets (Marshall et al., 2012). Its protein content

was found to range from 4.0 to 7.9 % of dry matter across different saline levels of soil (Al-

Dakheel et al., 2015). This means the grass can be grown in many arid areas where vast reserves

of saline water exist. This could be important in saving fresh and good quality water sources

for other purposes.

1.4 Factors limiting the use of C. ciliaris as a forage grass in Botswana at present

Widespread use of forage depends upon the local availability of cheap seed material, which is

true to type, viable and would reliably establish good pasture when sown (Parwani, 2013).

Cenchrus pasture production has been affected by unavailability of forage seeds in Botswana.

The seed is crucial in commercial Agriculture as a basic commodity for propagation of the

forage grass and also for perpetuation of germplasm (Arshadullah et al., 2011). Most farmers

who want to improve their natural or sown pastures have no access to commercially processed

seed at a nearby retail outlet.

Research and extension services on forage grass seed production and grass management

strategies in Botswana are still at rudimentary stages such that the provision of forage seed

through supply systems is almost non-existent. The absence of an effective formal seed system

reduces the impact of publicly funded forage production. This failure of seed multiplication

and distribution translates into negative rate of returns for those progressive farmers who may

try to make a living through forage and fodder production. Although the literature relating to

its nutritive value (Albu, 2012; Parwani and Rankard, 2012; Donaldson and Rootman, 2010;
18
Mcdowell, 2003; Ramirez et al., 2009; Arshadullah et al., 2011), its involvement in feed

formulations (Ramirez et al., 2009; Mutimura and Everson, 2012; Kumar et al., 2005; Hassan

et al., 2015;During and McNaught, 2012)and its biological invasions (Ogillo, 2010; Osman et

al., 2008; Marshall et al., 2012) has grown significantly, much is lacking in terms of the

agronomical techniques needed to establish this grass for various purposes, especially for seed

production. There is little published information regarding its multiplication capabilities and

agronomical requirements (Bhattarai et al., 2008). Plant breeders and agronomists have

concentrated on increasing the forage yield and nutritive value, with little attention given to

seed production (Tacheba and Moyo, 1985). The fact that it is found in isolated clusters which

are extremely fragmented means that its use on a large scale cannot be meaningful. Farmers

are also reluctant to take up seed production as a business enterprise because having only seeds

as a source of income is not very encouraging.

1.5 Experiences of Cenchrus elsewhere from its centre of domestication

Where it has been successfully introduced like in North-Western Mexico to improve

rangelands for cattle production, it has tended to invade adjacent habitats, displacing the native

fauna and flora (Bhattaraj et al., 2008). Its spread, naturalization and invasion into native plant

communities has been recognized as a serious ecological problem threatening biological

diversity and ecological function (Gutierrez-Ozuna et al., 2009) elsewhere outside Botswana.

The grass is seen as a threat to important mesic habitats within the arid zones in Northern

Mexico (Bhattaraj et al., 2008). These mesic sites are critical parts of the landscape, providing

concentrations of water and nutrient resources and refugia for some plants and animals

(Gutierrez-Ozuna et al., 2009). It threatens the survival of rare species and alters the food

supply chain of native animals. Where it spreads into non-target areas, it has become a serious

concern to non-pastoral land managers and those responsible for conservation areas and this

has been reported in most parts of Northern Mexico where it has been introduced (Gutierrez-

19
Ozuna et al., 2009). Its invasive and allelopathic effects may also need to be investigated in

Botswana. Environmentalists in Botswana have not yet raised any concerns about

it. This could be due to the fact that the grass has not really been grown on a large scale and it

also has not yet been over-sown in natural rangelands and its effects closely monitored.

According to Marshall and Ostendorp (2012), the characteristics of C. ciliaris which make it

versatile and suited to a range of harsh conditions also make it an expert invader of non-target

environments and from an environmental point of view, it is important to prevent further spread

of the grass. When used temporarily, its eradication is very difficult. Once established, the

conversion of a Cenchrus pasture to an alternative pasture would be prohibitively expensive.

Strategic control of its spread requires knowledge of its physiological characteristics which

most farmers, especially rural ones, do not have. Lack of seed, poor seed germination, high fire

risk and its invasive and allelopathic nature also contribute to the grass not being used on a

large scale in Pakistan (Arshadullah et al., 2011). The seed germination success and

allelopathic nature of C. ciliaris in Botswana needs to be investigated.

1.6 Opportunities of growing Cenchrus as livestock fodder crop

When managed well and its spread restricted to target areas, the grass can play a very crucial

role in improving livestock production in Botswana. Large tracts of land may need to be put

under the grass so that its impact on livestock production becomes quite significant. The first

hurdle to overcome would be to produce seed and make it available on the official market.

Farmers may also be encouraged to get involved in seed production if there is more than just

seed that would be of value from the whole exercise. The forage residues need to have high

nutritional value so that farmers can benefit from them as well, post seed harvesting. The

residues can be sold to other farmers to feed their livestock. There is need, therefore, for the

use of technologies that can improve seed yield without significantly compromising the

quantity and quality of the forage residue. Use of nitrogen-based fertilizers in pasture

20
establishment is one of the technologies that help to replace and maintain soil nutrient levels

for quality seed production and increased forage biomass yield (Sahoo et al., 2015). According

to Kizima et al. (2011), the application of nitrogenous fertilizer gives a high economic response

in seed production enterprises.

In Botswana, limited data is available on C. ciliaris forage and seed production

responses to nitrogenous fertilizers. The literature that is available (Nsinamwa et al., 2005;

Koech et al., 2014) is void of any data on seed production and seed quality responses to

nitrogenous fertilizers, specifically horse manure. The prospects of using horse manure to

produce forage grass seed may not interest researchers due to the limited quantity of the horse

manure in most areas of Botswana. Seed production, however, does not need to be done on a

very large scale which would then require a lot of horse manure. On the whole, the effect of

different types of nitrogen fertilizers on C. ciliaris productivity need to be investigated. The

nutritive value of the residual forage material after harvesting seed also needs to be analysed

to find out if the fertilizer that is used can also influence forage grass properties like dry matter

digestibility and the levels of major nutrients like nitrogen, phosphorus, potassium, calcium,

sulphur and magnesium and also levels of some trace elements like zinc, iron, sodium and

copper. According to Marschner (1995), without the right balance of nutrients, the forage

residues after seed harvesting have a low feeding potential. When the forage residues after seed

harvesting are of little or no value, farmers would not be attracted to the seed production

business. The choice of fertilizer to be used when growing forage grass for seed should be such

that it is effective, available and affordable. Reliance on vegetative splits for propagation of

forage grass in natural rangelands and large pastures is almost unthinkable.

1.7 Statement of problem

At present, farmers in Botswana have no access to commercially processed forage grass seed

at their local retail outlets. The capacity of the formal sector is so limited that it is struggling to
21
meet the ever increasing national demand. However, precedence in the form of established

cereal and pulses grain seed production system has been made. Emulating this production

system for forage seed production will be the way to go. For the seed to act as a catalyst in

increasing forage biomass, the seed has to be made available to a broad base of farmers on

continuing basis. The absence of an effective formal seed system and failure of seed

multiplication and distribution locally, translates into negative perceptions about taking up seed

production as an enterprise. Plant breeders and agronomists have concentrated on increasing

the forage yield and nutritive value, with little attention given to seed production (Mganga,

2009). As such, limited information is available on pasture seed production of recommended

promising grass species under different agro-ecological zones of the country. Most literature

is on research done in other African countries like Tunisia (Seddi et al., 2002) and Sudan

(Burhan and Hago, 2000; Abedelrahman, 2007) where C. ciliaris was found to be quite prolific

in seed production.

1.8 Justification

Range animal productivity can be increased by the adoption of forage grass species that show

rapid and vigorous growth, disease and drought resistance and are digestible. Propagation of

such grass would need reliable seed material. The supply of sufficient forage grass seed is

determined by the effectiveness and efficiency of the seed production system. The increase in

forage productivity can be substantial if farmers are encouraged to be involved in seed

multiplication ventures. The seed multiplication of desired forage grass species require modern

agricultural practice. Seedbed preparation, planting depth, spacing, use of pesticides and

irrigation regimes require modern farming techniques.

The choice of the most appropriate fertilizer, its application rate and timing of

application are also technically demanding. On farm trials of forage grass introductions and

evaluations to determine seed and forage biomass yield of promising pasture species such as
22
C. ciliaris, need to be done. The data contained in Animal Production Report Unit (APRU)

reports of 1975, 1977 and 1985 is too ancient to be relied upon. According to the APRU (1985),

in normal years, only 0.04% of ruminant feed came from improved pastures and forage crops.

Just like any other new adoption, the participation of farmers in forage seed production is

important. There is, therefore, the need for a centralised and aggressive extension-based push

focusing on technological packages that combine fertilizers, improved seed production and

better forage management practices in order to improve the forage grass supply side. Once

researchers prove that seed can be produced and sold profitably, this can be liberalised, with

the emergence of private forage grass seed companies. If the forage residue that remains after

seed harvesting is of relatively high nutritive value, farmers can be convinced into taking up

seed production, knowing that income will come from both the seed and the forage residues.

Use of nitrogen sources that are relatively cheap, environmentally friendly and effective

should be done. This prevents over/under-fertilization, both of which are detrimental to forage

seed production. There is need for more published information regarding forage grass

multiplication potential using cheap, available and environmentally friendly nutrient sources

in Botswana. Developing a nitrogen fertility programme is crucial in terms of increasing forage

and seed yield without evoking the negative effects of excess nitrogen. This should be done

bearing in mind the implications of the rapidly increasing cost of commercial nitrogen

fertilizers and contamination of the environment. A source of nitrogen that promotes seed

production without significantly compromising forage yield and quality should be explored.

23
1.9 The objectives of the study are:

. To determine the effects of horse manure and urea on the length of inflorescence of C.
ciliaris

. To assess the effects of horse manure and urea on the number of tillers of C. ciliaris.

. To measure the effects of horse manure and urea on the seed yield of C. ciliaris.

. To assess the effects of horse manure and urea on forage biomass of C. ciliaris.

. To evaluate the effects of horse manure and urea on the nutritional quality of the C.
ciliaris forage residues post seed harvesting.

24
CHAPTER 2

2.0 Literature Review

2.1 Botswana’s natural grassland

Ruminant livestock production in Botswana mainly relies upon natural grassland, cut herbage

or crop residues (Nsinamwa et al., 2005). Only a few farmers depend on imported

concentrate feeds and baled legumes like Lucerne due to the costs involved. Most parts of

the country are arid or semi-arid and this is why poor quality forage grass species dominate

most grazing lands (Nsinamwa et al., 2005). The natural pastures, bare land that has been

eroded and other wastelands like sand dunes can, however, be rehabilitated through reseeding

by promising forage grass species like C. ciliaris.

2.2 The desirable characteristics of C. ciliaris

C. ciliaris is commonly associated with scattered woody legumes such as Prosopis species

and Leucaena leucocephala (Parwani, 2013). It is fast growing, shortly stoloniferous and is

quick to flower. Individual plants develop as clumps usually with limited lateral spread. Height

of flowering culms may range from 15 cm to 1.5 m. Inflorescences are from 3 cm to 15 cm

long and 1-2 cm wide (Parwani and Mankad, 2012). The appropriateness of this grass is a

result of its numerous characteristics that favour local conditions. Parwani (2013) described C.

ciliaris as a deep-rooted, summer growing and perennial tussock grass with a high herbage

yield potential. Its deep and widespread root network makes it very drought resistant and

competitive in terms of nutrient uptake. It also makes it ideal for the stabilization of sand dunes

and prevention of soil erosion in general. Yossin and Ibrahim (2013) reported that the soil

binding capacity of C. ciliaris is due to its clustered root system in the 8 – 10 cm layer of soil.

It is native to Southern Africa and India (Parwani, 2013). It survives extreme and prolonged

drought but grows vigorously when favourable conditions set in. The grass has spread

successfully from planting in areas having a dry period of 150 to 210 days, mean temperatures
25
of between 24oC and 45oC and the coldest month ranging from 5oC and 15oC. It has a low

tolerance to freezing temperatures and does not tolerate waterlogged soils (Yossin and Ibrahim,

2013). Parwani and Mankad (2012) highlighted its rapid response to moisture availability,

relative palatability and resistance to trampling and overgrazing as some of its favourable

attributes. It produces viable seed that can be used to propagate the grass in other areas. Due to

the viable seed it produces, its stands are self-replacing and pastures may not need to be re-

seeded. According to Hall (2001), C. ciliaris is a highly regarded pasture grass due to its value

as pasture for livestock and its soil protection properties. In Australia, it has brought some great

financial benefit to many individual producers and companies due to its tolerance to drought,

fire and overgrazing (Hall, 2001). It produces more biomass than many native perennial grasses

and its high seed yield and fluffy seed allow it to spread readily via wind and water (Seddi et

al., 2002).

Besides being apomictic, C. ciliaris is highly polymorphic and variable for several

traits of ecological and agronomic importance (Seddi et al., 2002). The grass, therefore, is

among the species having the greatest potential for forage production in Botswana. It has the

potential to produce high yields of high quality hay and pasture for livestock production and

ground cover for rehabilitating degraded land and stabilizing sand dunes. According to

Redfearn et al. (1990), the productive potential of this grass is limited by the inadequate soil

fertility and limited amount of rainfall. These conditions that limit C. ciliaris productivity are

generally the ones that are experienced in most parts of Botswana. This is why seed production

needs to be done under irrigation and fertilizer has to be applied as well.

Some nutrients are required by forage grasses in very small amounts and are adequately

provided from the soil, for example, zinc, cobalt, iron and manganese (Laidlaw, 2005). Others,

like nitrogen, phosphorus and potassium are required in greater amounts and are commonly

applied as fertilizer amendments (Laidlaw, 2005). . According to Sahoo et al. (2015), C. ciliaris

26
grass has a reputation as a phosphophilic grass. It has a high phosphorus content and a high

Phosphorus: Calcium ratio. Effects of fertilisation is not only observed in the forage but it is

translated into animal performance.

2.3 The use of nitrogenous fertilizers to improve seed production

Fertilizer, as a mineral nutrient source, is one of the most important technologies that can

increase forage grass and seed production. Most soils in Botswana are predominantly sandy-

textured, with very low organic matter content, resulting in lack of essential plant nutrients like

nitrates and phosphates, which are vital for pasture grass growth and seed maturation (Mudenda

and Maeresera, 2009). According to Osman (2008), nitrogen (N) fertilization has been found

to typically increase grass dry matter, forage N concentration and seed yield. A number of

studies (Hassan et al, 2015; Ihsan et al., 2014; James, 2010; Brummer, 2009; Abedelrahman,

2007 and Ashraf et al., 2013) have found that forage yields are increased by fertilization on

sandy soils. N fertilization is one of the most common practices since the nutrient is the most

limiting factors influencing yield and chemical composition of grass pastures. In terms of

forage and seed production of C. ciliaris, N is the most limiting nutrient as it does not really

stay in the soil for a long time. It is highly mobile and is easily lost from the root zone through

leaching, yet it is very crucial (Mujuni and Sibanda, 2007). It is a major factor for increasing

the pasture yield and nutritive value of the grasses, including their crude protein content and

digestibility (Hassan and Fikru, 2015). According to Donaldson and Rootman (2010), dry

matter yield in forage grasses responses primarily to nitrogen fertilizer application. However,

the rapidly increasing cost of commercial N fertilizer makes it imperative to optimize nitrogen

use efficiency.

According Marschner (1995), too little N reduces the disease fighting properties in

forage grasses and the grasses become susceptible to fungal infections. If over-fertilized with

27
N, the excess N causes the breakdown of grass tissues through sugars and amino acids, making

them susceptible to the invasion of fungal spores (Abedelrahman, 2007). In addition, Kizima

et al. (2011) reported that the higher the N fertilization rate, the greater the risk of the nitrate-

N exceeding 1000 ppm and such nitrate levels in forages may affect animal health. Nitrogen

fertilization has been found to typically increase grass dry matter, forage N concentration and

seed yield (Osman et al., 2008). Generally, phosphorus fertilization alone does not increase

forage yields sustainably but combined application of phosphorus and N does (Osman et al.,

2008).

The most rapid spread of C. ciliaris occurs in soils of good nutrient status. Kumar et

al. (2007) reported that significant increases in forage N content, digestibility and mean daily

live mass gain per sheep were found as N fertilization rates increased. Fertilization has also

shown to have increased soil water extraction by forages, and improve water use efficiency

(Sahoo et al., 2015). N application up to 60 kg/ha in two doses at 15 day interval, significantly

increased the fodder yield (Sahoo et al., 2015).However, nitrogen may be supplied by organic

sources such as manure. Application of 10 t/ha of sheep manure increased and sustained the

productivity of C. ciliaris pasture for 3 years (Sahoo et al., 2015).

2.4 Importance of N in forage grasses

N is important as a component of nucleic acids, amino acids, proteins, chlorophyll and enzymes

(Taylor et al., 1997). It affects shoot growth, shoot and root density, colour, disease resistance

and stress tolerance. It has a synergic relationship with phosphorus which is a component of

nucleic acids, membranes, ATP, co-enzymes and encourages a dense root network (Taylor et

al., 1997). When N is sufficient, forage grasses are able to significantly absorb potassium.

Potassium activates enzymes used in protein, sugar and starch synthesis. It is important in

maintaining turgor pressure in plants. It affects drought tolerance, cold hardiness and disease

resistance (Taylor et al., 1997). Soil sampling and testing are used to determine the amount of
28
nutrients that might need to be added to make up for any deficiencies. According to Ihsan et

al. (2014), a number of studies have found forage yields of C. ciliaris to be increased by

fertilization in sandy lands. N fertilization typically increases grass dry matter, forage N

concentration and seed yield (Ihsan et al., 2014). All C. ciliaris varieties respond to a good

fertility programme which supplies adequate amounts of N, phosphorus and potassium

(Redfearn et al., 1990). Actively growing C. ciliaris removes nutrients from the soil. From a

grazing management standpoint, this is of little concern as the majority of the nutrients remain

in the pasture and cow dung deposited during grazing recycle back some of the nutrients. As

for hay and seed production, the nutrients will have to be replaced as much of the harvested

hay will not be fed on the fields where it was harvested from. This causes a loss in nutrients.

N determines the forage and seed yield and is likely the driving factor in irrigated forage

production. According to Savoy (1996), for grass pasture to be productive, first priority should

be given to meeting nitrogen needs. Burhan and Hago (2000) stated that nitrogen plays an

important role in plant growth and physiological processes, as it enters all enzyme composition

and enhances vegetative growth and yield. An increase in N increases the leaf: stem ratio.

Abedelrahman (2007), however, reported that no significant effect of N fertilizer was detected

on mean plant height, inflorescence length and width. This is in agreement with the

experimental findings of Yossin and Ibrahim (2013) which showed that growth parameters

were not significantly affected by fertilizers but yield parameters. All N fertilizer treatments

were found by Yossin and Ibrahim (2013) to have a significant effect on fodder yield and seed

yield compared to the control of no N fertilizer. Meena et al. (2015) reported C. seed yield of

98 kg/ha with organic manure (sheep) compared to 89 kg/ha without any fertilizer and 110

kg/ha with NPK compound fertilizer. Kumar et al. (2005) reported C. ciliaris seed yield of 30

kg/ha with horse manure while Ashraf et al. (2013) said that the C. ciliaris seed yield when

using urea was between 21.8 kg/ha and 150 kg/ha, depending on the level of fertilizer applied

29
and the irrigation regime followed. Ashraf et al. (2013) also reported the dry matter yield of C.

ciliaris as 3.345 t/ha without any fertilizer and 3.790 t/ha for sheep manure and 4.21 t/ha for

NPK fertilizer in Rajasthan. Ihsan et al. (2014) reported Cenchrus dry matter yields of between

16.3 t/ha and 30.18 t/ha, depending on the level of the fertilizer applied. Savoy (1996) reported

that forage and seed yield response to nitrogen can be very dramatic. Donaldson and Rootman

(2010) reported that nitrogen is essential for seed production, with seed yields being raised

tenfold up to 150 kg/ha.

Unlike phosphorus and potassium, N is not retained in the soil from year to year in a

form that grasses can readily use (Taylor et al., 1997). N supplied to the soil is rapidly converted

to nitrate-nitrogen and is then often incorporated into organic materials, leached out of the

rooting zone by soil water or lost back to the atmosphere by denitrification (Savoy, 1996).

Denitrification occurs in waterlogged soils due to poor drainage which then causes the nitrate-

nitrogen to be converted back to gaseous forms and be lost to the atmosphere (Taylor et al.,

1997). Grasses respond quickly to N when other growing conditions are favourable. The rate

of application and the source of the N are factors that need to be considered when coming up

with an N supply programme for forage and seed production. According to Ihsan et al. (2014),

the rate of application and the source of the N depend upon the time of the year and weather

conditions. Based on these conditions, one source may be better than others in a specific

circumstance.

Unlike fertilization for most field crops, pasture fertilization management is more

often guided by the purpose of the grass. If the target is to increase forage biomass, higher rates

of N can be used. If the target is seed production, higher rates may increase herbage yield and

compromise seed yield. Developing an N fertility programme is an important aspect that can,

therefore, affect the seed yield and forage quality of pasture grass. The understanding of how

quickly the N is released from the source helps to determine the frequency of application and
30
appreciate the dangers therein. This is why pastures must be strategically managed through

well-planned fertilization management programmes.

2.5 Possible sources of N for pasture fertilization

The possible sources of N for fertilizing forage grass pastures are legumes, inorganic

(chemical/artificial) fertilizers from industry and organic manure from plant and animal

material.

2.5.1 Legumes

N can be supplied to forage grasses by over-sowing the pastures with legumes such as clover,

siratro and alfalfa, which can fix N directly from the atmosphere (Brummer, 2009). These

legumes use N fixing bacteria in their root nodules. Intercropping of Cenchrus pasture with a

legume crop, as a source of N, increased the dry matter yield by 5 times and the same result

was obtained when Dolichos and cowpeas were used as legumes (Sahoo et al., 2015). A

combination of 1:1 of grass and legume was found to be most appropriate for establishment of

grass-legume pasture (Sahoo et al., 2015). The maximum dry forage yield was only obtained

under normal rainfall conditions. The only drawback is that the moisture stress-tolerance of the

C. ciliaris becomes less of an asset as sufficient moisture has to be availed for the legume. The

other problem with intercropping C. ciliaris with a legume is when the grass is needed as a pure

stand for use in feed formulations or for research experiments. The use of machines to harvest

the grass becomes quite difficult if it is intercropped with a legume. In forage systems without

legumes like seed production, nitrogen has to be added as a fertilizer material to achieve the

best forage grass seed production.

2.5.2 Inorganic fertilizers

Historically, Ammonium Nitrate (33.3% N), Ammonium Sulphate (21% N) and Urea (46%

N) have been the major inorganic sources of N for grass pastures. Ammonium Nitrate is no
31
longer used on a large scale due to its explosive nature (James, 2010). Ammonium Sulphate is

very expensive on a cost per kg of N basis (James, 2010). It also has a relatively high salt index

and greater acidification potential per unit N applied than other ammonium-containing N

sources (Donaldson and Rootman, 2010). It also has a fairly low N content. Urea is one of the

generally cheaper sources of N on a cost per kg of N basis. When using irrigation, too much

water can lead to movement of N beyond the root zone since it is so mobile in the soil. Losses

through denitrification can also occur under waterlogged conditions (Brummer, 2009).

These inorganic fertilizers are quick release sources of N. Some of the N that they release

is taken up by grasses and stored as non-protein N. Forage grasses are sponges for N and

will quickly take up any available N once it moves into the soil (James, 2010).

2.5.3 Organic manures

Organic manures have a long history of use in farming systems. The nutrient supplementation

through organic sources in pasture is attributed to better availability of nutrients and improved

soil bulk density (Monroe, 1996). Organic manures also have solubilising effect on fixed forms

of other nutrients and, therefore, improve the soil fertility (Kumar et al., 2007).Chicken manure

(broiler litter) results in building of soil phosphorus and potassium levels. There is an increase

in soil pH as was observed in the Kentucky studies (Monroe, 1996) and this is attributed to the

high base content of the broiler litter. This is said to be desirable at the initial stages but has

long term deleterious effects. Hassan et al. (2005) reported a dry matter forage yield of 5.905

t/ha of C. ciliaris fertilized with urea and 5.430 t/ha due to chicken manure fertilization. Goat

and cattle manure are so much on demand in arable crop production, which is normally given

first priority over pasture fertilization. In this context, horse manure, which is normally

regarded as a heap of rubbish that needs to be discarded elsewhere, may be the solution.

32
 Horse manure is a mixture of horse dung, used bedding and urine soaked in the bedding

(Hadin et al., 2017). It is often regarded as a waste to be disposed of rather than a valuable

fertilizer resource and a necessary by-product of the livestock industry. It is the technology

involved in the treatment and use of this ‘waste’ that determines whether it becomes a valuable

resource or a costly liability that just needs to be removed from stables as it produces unhealthy

ammonia fumes as well as providing a fertile ground for moulds, bacteria and other parasites

(Yossin and Ibrahim, 2013). A horse weighing 400-600 kg excretes 19-30 kg dung and urine

per day, on average, containing 70-150 g N, 10-30 g P and 20-50 g K (Keskinen et al., 2017).

The quality of horse manure depends on the feed ration of the horses, amount of litter, bedding

or soil included on the stable floor and amount of urine concentrated with the manure. Keskinen

et al. (2017) reported that when pelleted straw is used as bedding material, the horse manure

that results has increased ability to retain nitrogen and phosphorus under rainfall. The horse

manure will have a carbon: nitrogen ratio of less than 15. In fresh horse manures, carbon:

nitrogen ratios normally exceed 30 and this leads to net nitrogen immobilization in the soil

(Keskinen et al., 2017) Drying the manure concentrates the nutrients in it on a weight basis

(Yossin and ibrahim, 2013). When properly dried, horse manure is a value added resource that

contains both major and trace elements. The drying should reduce the moisture content from

around 80% to about 10% (Yossin and Ibrahim, 2013). This shows how important storage and

handling are in quality control.

The digestive system of a horse is such that its hindgut contains an active population

of bacteria and protozoa (Jones, 1985). The microbes synthesize amino acids in the large

intestines, but essential amino acids are not absorbed in any appreciable quantities from the

hindgut (Jones, 1985). This means that, unlike in ruminants, large quantities of microbial

protein generated in the large intestines of the horse are wasted because there is no opportunity

there for significant absorption of amino acids (Donaldson and Rootman, 2010). It follows then

33
that these amino acids could be lost in the horse droppings. The microbial protein that could be

egested can be put to good use by fertilizing pasture grass. Some research indicates that the

use of horse manure may result in the build-up of phosphorus and potassium levels (Keskinen

et al., 2017; Ogren, 2013 and Ogren et al., 2014). An increase in soil pH is usually observed

and is attributed to the base content of the horse manure (Keskinen et al., 2017). For the short

term, this is very desirable as many of pasture or hay systems suffer from the effects of soil

low pH (Savoy, 1996). The N compounds in horse manure are eventually converted to the

available nitrate form (Ogren, 2013). The nitrate is soluble and is moved into the root zone

with water (Ogren, 2013). The release of this available N from horse manure during

decomposition is very gradual. This slow release of N is the horse manure’s greatest asset. The

horse droppings are more fibrous in texture compared to goat, sheep and cattle manure. This

could explain the slow nutrient release nature of the horse manure as the nutrients are tightly

embedded into the fibrous matrix (Monroe, 1996).

According to Savoy (1996), the fibrous nature of the horse manure ensures that the nitrate

nutrients are held into the fibre matrix and released slowly. This extends N availability but in

measured quantities that reduce the manure’s burn potential. The importance of this in sandy

soils cannot be overemphasized. The release of N steadily over a long time by the horse manure

requires that its application to the pasture is not frequent. This is cost effective. The nutrients

from horse manure are less likely to leach into underground water compared to nutrients from

fast release fertilizers. Horse manure contains grass and grain fibres, minerals, fat, water and

grit (Ogren, 2013). It is not as smelly as that of other non-ruminants. Most people do not find

it overly offensive. According to Mudenda and Maeresera (2009), organic manure, of which

horse manure is one, increases the water holding capacity of sandy soils. This is important in

that the sandy soils would retain the moisture long enough for the forage grasses to use.

34
 Available moisture and available soil N are closely related because both move in the soil.

N would be moved to the roots as the plants absorb water. Nutrients are retained in the humus

from decomposition of horse manure to the extent that they are not leached out of the rooting

zone of pasture grasses. This is important in that it reduces the effect of leached nutrients to the

environment through eutrophication, which can result in the death of marine organisms

(Mudenda and Maeresera, 2009). According to Brady and Weil (2002), horse manure increases

the cation exchange capacity of the soil. This is the capacity of the soil to hold onto cations

like Ca2+, Fe2+, Zn2+, Cu2+ and others. These cations are held by the negatively charged organic

matter particles in the soil through electrostatic forces and they become less susceptible to

leaching (Yossin and Ibrahim, 2013). Since the N is released steadily over a period of time,

there are less chances of N over-supply that can result in reduced seed yield (Keskinen et al.,

2017).

Horse manure loosens any heavy clay soils and increase their drainage and aeration. This

prevents the conversion of nitrate-nitrogen to gaseous forms which get lost to the atmosphere

by denitrification (Mudenda and Maeresera, 2009). Horse manure, like most organic manures,

moderates soil temperature in the pasture (Mudenda and Maeresera, 2009). The soil

temperature should not fluctuate widely between very hot and very cold times of the day. This

would not be conducive for microbes involved in processes like N fixation, root nodulation in

legumes, ammonification, nitrification and for the process of root respiration (Donaldson and

Rootman, 2010). Horse manure result in the proliferation of soil organisms that are involved

in decomposition of any new organic matter from leaf fall and the more these organisms in the

soil, the richer the soil for forage production (Donaldson and Rootman, 2010). Horse manure

promotes the formation of a crumb structure which enhances seed germination and growth of

pasture grasses (Abedelrahman, 2007). The larger bulk that must be handled and spread is a

definite disadvantage of horse manure. For just a hectare of pastureland, a large quantity of

35
horse manure is required as the nutrient concentration is low. According to Eriksson and

Hennessy (2015), the amount nutrients in horse manure depends on whether the horse is

sedentary or exercising. On average, horse manure that is from a sedentary horse and managed

by composting contains 4.3% K, 3% Mg, 0.3-1.2% Na and 3.8% N (Eriksson and Hennessy,

2015). Decomposing microbes in unmanaged horse manure absorb released N to satisfy their

growth requirements, resulting in a high C: N ratio (Trottier et al., 2016). Due to N

immobilization, horse manure is not a desired fertilizer for forage production but for seed

production (Trottier et al., 2016). Its fibrous nature makes it attractive to termites that may end

up attacking the forage grass seedlings.

Figure 1: Digestive system of a horse where microbial protein synthesis and digestion
occur after the main absorptive region of the small

The digestive system of a horse is typical of a non-ruminant. Microbial fermentation of

cellulose occurs in the caecum and colon (Cubitt, 2010). The microbes that are in the caecum

are able to use non-protein N to build their bodies. This microbial protein synthesis takes place

in the hindgut, as shown in Figure 1. Most of this microbial protein is most likely to be flushed

out since there is limited absorption of proteins in the hindgut (Cubitt, 2010). Unlike in horses,
36
in ruminant, microbial protein synthesis occurs in the rumen. When the microbes are killed by

the acid in the abomasum they are digested and absorbed in the small intestines. This means

that more microbial protein would be utilized by a ruminant than that which would be lost in

the faeces. Notwithstanding the differences in diets and selectivity, horse manure is expected

to contain more nitrogen than ruminant faeces. However, some ruminants like goats are

selective browsers which normally feed on highly nutritious and digestible parts of browse

shrubs and grasses and this causes an increase in the nitrogen of their faecal matter (Cubitt,

2010).

2.5.4 Urea

Urea (46-0-0) is usually a cheaper source of N on a net N basis compared to other single

element N fertilizers (James, 2010. It is non-combustible and non-explosive which make it easy

to store. Its high analysis of 46% nitrogen helps reduce handling, storage and transportation

costs over other dry N forms. According to Marinari et al. (2000), urea can be applied as a

solid or solution to certain crops as foliar spray. When applied as a foliar spray, its availability

to the forage grass is not affected by the soil conditions like pH. During manufacturing of urea,

there are less pollutants that are released to the environment (Marinari et al., 2010). Urea is a

quick release N source which is very soluble, fast acting and gives a rapid green-up response

in forage grasses (Marinari et al., 2000).

James (2010) reported that when field applied, urea changes to ammonium hydrogen

carbonate due to the activity of the enzyme urease. The process releases hydroxide ions which

raise the soil pH to as high as 8.5. At this alkaline pH, the ammonium ions (NH 4)+ tend to

convert to ammonia gas ( NH3). This is why when urea is applied on the soil surface and not

incorporated into the soil, it is subject to volatilization loss as ammonia. Savoy (1996) reported

some research findings which showed that the potential for nitrogen loss as ammonia gas

increases as temperature, soil pH and moisture increase and as rate of application increases. In
37
fact, Redfearn et al. (1990) said that N volatilization losses from urea may be as high as 50%

under a combination of high humidity, hot temperatures and windy conditions. To reduce

volatilization losses from urea, it is best to apply the fertilizer later in the day when the dew

has dried. It is also important to apply it no more than seven days to an anticipated precipitation

event (Redfearn et al., 1990).The urea needs to be incorporated into the soil. James (2010)

explained that if the urea – NH4+ reaction takes place in the soil, the nitrogen will be captured

as exchangeable ammonium on the soil exchange complex and little, if any, ammonia gas

would be lost to the air. This, however, could be a challenge when establishing pasture grass

like C. ciliaris. During pasture grass seedling establishment, a rapid pH increase after urea

application caused by hydrolysis of urea can result in ammonia release that can damage

seedlings (Redfearn et al., 1990). After applying urea, thorough watering is needed as the urea

compounds, if not dissolved, remain in the root zone of the grass seedlings, causing burning.

Another shortfall of urea is that it does not supply any other nutrients apart from nitrogen. If

soil analysis shows that a number of nutrient elements are deficient, the use of urea becomes

counter-productive.

2.6 Effect of N fertilization on the nutritional quality of C. ciliaris at seed harvesting

stage

As far as the chemical composition and digestibility of the grass after seed harvesting are

concerned, Hassan et al. (2015) reported that when N fertilizer was applied to C. ciliaris, the

hay after seed harvesting produced a range of 8.5-10.6% ash, 0.9-2.5% N, 38.5-45.4% crude

fibre and 53.8-70.5% In Vitro digestibility. Unfertilized C. ciliaris produced 8.5-10.1% Ash,

0.6-0.9% N and 42.9-44.7% fibre. According to Al-Dakheel (2015), at the seed harvesting

time, C. ciliaris has a CP ranging from 4-6%, ADF of between 36.6% and 47.7%, NDF ranging

from 66.5% and 77.6% and Ash of between 9.4 and 16.7% dry matter. Ashraf et al. (2013)

reported that C. ciliaris fertilized with horse manure had 37.34% crude fibre, 0.3% sodium,

38
4.7% potassium and 13-17.5% protein at seed harvesting stage. Donaldson and Rootman

(2010) reported results that showed C. ciliaris with protein content ranging from 6-16%, crude

fibre of 38.5-45.4%, In Vitro digestibility of 53.8-70.5% and NDF of 65% when fertilized with

urea. Brady and Weil (2002) said that fertilizing grasses with nitrogen often substantially

increases crude protein levels in the forage but has little or no effect on digestibility. In fact,

fertilization with phosphorus, potassium or other nutrients that increase yield was said to

actually slightly reduce forage quality (Brady and Weil, 2002). Excessive levels of some

elements like potassium in some cases were found to decrease the availability of other elements

such as magnesium and calcium (Keskinen et al., 2017; Ogren et al., 2014; Eriksson and

Hennessy, 2015).

Maturity at harvest of seeds has the greatest influence on NDF digestibility. The dry

matter digestibility of forage grasses decrease with maturity. According to During and

McNaught (2012), as forage matures, NDF digestibility can decline significantly. Tailor et al.

(1997) explained that with advancing maturity, plants develop xylem tissue for water transport,

accumulate cellulose and other complex carbohydrates which become bound together by a

process called lignification. Lignin is more difficult to digest than either cellulose or

hemicellulose. Again, as maturity proceeds, leaf- to- stem ratio declines and this causes NDF

digestibility to decline. Parwani et al. (2009) explained that during the forage maturation

process, accumulation of the stem mass exceeds leaf mass addition. The stems contain a higher

proportion of thick-walled tissues of sclerenchyma, xylem fibre and xylem vessels and less

photosynthetic tissues of mesophyll and chlorenchyma than found in leaves (Tailor et al.,

1997). Lignin concentration in forages has been reported to be negatively correlated with

digestibility of forages (Parwani et al., 2009). It would, therefore, be interesting to find out the

level of lignification that would have taken place in the forage by the seed harvesting time and

also the influence of applying either urea or horse manure as inorganic and organic fertilizers

39
respectively. The amount of NDF and ADF can be used to predict the digestibility of the forage

and its general utilization potential (During and McNaught, 2012).

40
CHAPTER 3: COMPARATIVE ASSESSMENT OF THE EFFECTS OF FERTILIZING C.
CILIARIS WITH HORSE MANURE VERSUS FERTILIZING WITH UREA ON THE
SEED YIELD AND FORAGE BIOMASS PRODUCTION

3.0 Introduction

Livestock production in rural Botswana is hindered by low quantities of high quality forage

grasses. Most farmers in these rural areas depend on natural rangelands for the grazing of their

livestock. For more than 6 months in a year, there is shortage of pasture during the dry period.

Strategy by farmers to sustain their livestock during these difficult times is to supplement

animals with grass, crop residues, legume forages or commercial energy and protein

supplements or concentrates. Most of the time, these supplements are imported from South

Africa, especially grass, legume forage and concentrates. Due to the bulkiness of grass, its

importation from South African over long distance may not make economic sense. Ideally,

farmers should be buying bulky roughage feeds locally or grow them to reduce both transport

and environmental costs. However, challenges of establishing pastures on a commercial scale

are numerous.

According to Nsinamwa et al., (2005), it has been very difficult to improve the range

through introduction of more valuable and productive forage grasses due to unavailability of

planting material in the form of seed. C. ciliaris is gaining attention in various fields of research

(Akiyana et al., 2005; Al-dakheel et al., 2015; Arshadullah et al., 2011; Bulle et al., 2011;

Ihsan et al., 2014) as it is quite thriving and productive under conditions of high

temperatures, solar radiation and low moisture. The grass is more efficient at absorbing,

storing and utilizing carbon from the atmosphere and recycled nitrogen from the soil

(Monroe, 1996). The isolated clusters of valuable forage grass like C. ciliaris cannot provide

sufficient tussocks for vegetative propagation (Monroe, 1996). In natural rangeland, vegetative

propagation of the grass would almost be impossible due to the amount of work that would

41
be involved in terms of land preparation and the whole planting process. There are no seed

banks for such better performing forage grasses. There is, therefore, the need to research into

the possibility of producing forage grass seeds cheaply and abundantly to improve the quality

of the grazing areas, even in rural Botswana. These can be achieved by seeding natural pasture

or planting the grass on farms. The growing demand for pasture grass seeds locally and in

neighbouring countries necessitates researchers to respond by trying to bring the pasture seed

industry to a functional level. After having managed to produce seed, the next challenge would

be to use it for over-sowing in the rangelands in order to raise the quantity of the available

forage. With growing professionalism and experience in seed production by researchers,

farmers can then be incorporated.

The use of N fertilizers has been shown to increase the seed and forage biomass yield

of C. ciliaris (Afzal and Ullah, 2007). N has been found to be the most limiting factor to seed

production and biomass yield of C. ciliaris (Koech et al., 2014). Most soils in Botswana lack

N mainly due to low levels of organic matter and leaching from the mostly sandy soils (Tacheba

and Moyo, 1985). C. ciliaris has, however shown to respond well to improved N supply.

According to Kizima et al., (2012), there is need for a constant supply of moderate amounts of

N for significant increase in seed production to be realized. This is because Cenchrus grass

stands become unproductive with time as N is tied up in the root system of the forage grass

(Bulle et al., 2011). According to Burhan and Hago (2000), N enhances the vegetative growth

of C. ciliaris. It increases the stem: leaf ratio. The increase in vegetative material is, however,

detrimental when the aim is to produce seed. When a lot of energy is channelled towards

vegetative growth, less is used in seed production. On the other hand, if a plant does not develop

sufficient leaves, it would not produce enough food to store in the seed and this compromises

the quality of the seed. Establishment of pastures for seed production, therefore, requires

enough agronomic information and technical support so that N is not under or over-supplied.

42
The present research seeks to make comparisons between horse manure and urea as sources of

N, with reference to their effect on the quantitative parameters of C. ciliaris. This is important

in that an increase in the yield of forage grass goes a long way in supporting the livestock sector

in the country.

3.1 Specific objectives

To evaluate the effect of the type of N source (horse manure versus urea) on the length
(cm) of the inflorescences, the number of above ground tillers per crown (tiller density),
seed yield and biomass production of C. ciliaris.

3.2 Hypothesis

𝐻0: There is no significant difference in the inflorescence length, tiller density, seed yield and

forage dry matter yield of C. ciliaris produced when using horse manure and urea as fertilizer

treatments and no fertilizer as a control.

𝑯𝟎: µ 𝒉𝒐𝒓𝒔𝒆 𝒎𝒂𝒏𝒖𝒓𝒆/𝒖𝒓𝒆𝒂 = 𝝁 𝒄𝒐𝒏𝒕𝒓𝒐𝒍

𝐻𝑎: There is a significant difference in the inflorescence length, tiller density, seed yield and

forage dry matter yield of C. ciliaris produced when using horse manure and urea as fertilizer

treatments and no fertilizer as a control.

𝑯𝒂: µ 𝒉𝒐𝒓𝒔𝒆 𝒎𝒂𝒏𝒖𝒓𝒆/𝒖𝒓𝒆𝒂 ≠ 𝝁 𝒄𝒐𝒏𝒕𝒓𝒐𝒍

43
3.3 Materials and Methods

3.3.1 Site description

This study was carried out at Livingstone Kolobeng College in Block 8, Gaborone. It is located

on 24.6020o S and 25.9067o E. The site of the experiment was a school garden that measured

60m by 40m.The average annual rainfall in Gaborone and surrounding areas is 550 mm

of which normally falls between November and March (Burgess, 2006). The rainfall is,

however, extremely variable from year to year. The mean maximum temperature is

28.60C in summer and the minimum is 12.80C in winter (Burgess, 2006). The soils in the

site showed a dominance of the sandy fraction and it was light in colour, showing low levels

of organic matter. The study area was a previously cultivated land that just needed to be turned

over and levelled.

3.3.2 Soil sampling and testing

At the beginning of the investigation, 8 soil samples were taken randomly from the

experimental site at a depth of 10 cm using an auger. The top soil is the one that is critically

important during the root establishing stage of the cuttings. The collected soils were mixed

together to make a composite sample. A 300 g subsample of the composite soil was taken to

Botswana University of Agriculture and Natural Resources (BUAN) for analysis using

procedures of the AOAC (1996). Total N, phosphorus, and cation exchange capacity (CEC)

were determined. Soil pH was determined using the Universal indicator method and was found

to be around 6. The soil was disaggregated using a wood pestle and sieved at 2 mm and was

ready for digestion and testing.

3.3.3 Experimental design

A completely randomized design (CRD) with two fertilizer treatments (horse manure and urea)

and a control treatment (no fertilizer) was used for this research. Each treatment and the

44
control were replicated 6 times, giving a total of 18 plots. Each plot was 3 m long and 1.2

m wide, giving an area of 3.6 m2. The inter-plot distance was 0.6 m. For assigning plots

to treatments, 18 pieces of paper were used, with each of the three treatments inscribed

on 6 of the pieces of paper. The pieces of paper were put in one box and shuffled around.

The plots were numbered 1 to 18. For every plot number, a piece of paper was picked

from the box and the treatment inscribed on that paper was the treatment for that plot.

T2 T3 T2

T1 T2 T1

T2 T1 T3

T3 T2 T1

T1 T3 T2

T3 T1 T3

Figure 2: Arrangement of plots in the experimental garden

3.3.4 Land preparation, fertilizer application, planting and management practices

Plots were marked using pegs and a garden line. Ridges were made using a spade and a rake

and digging was done to an average depth of 20 cm using a spade. A pick was used for breaking

up hard ground. Horse manure was applied to six (6) of the plots at a rate of 12 t/ha and was

incorporated into the soil using a digging fork, before planting. Urea was applied to 6 more

plots at a rate of 100 kg/ha and was worked into the soil. The fertilizer application rates used
45
in this study for horse manure were as recommended for light textured soils by Mujuni and

Sibanda (2007). The remaining six plots did not receive any fertilizer. Thirty (30) well

developed tussocks (vegetative splits/cuttings) were planted in each plot in three (3) rows, with

ten (10) tussocks per row. The seedling tussocks were obtained by separating C. ciliaris clumps

into individual splits ready for planting. The seedling tussocks were obtained in the bushy area

behind Choppies supermarket in Block 8 in Gaborone. The upper parts of the tussocks were

cut off to leave them 25 cm in length. No rooting chemicals were used. The planting depth was

10 cm, with 3 nodes set into the soil, leaving 15 cm above the ground. A transplanting trowel

was used for planting. The inter-row spacing was 40 cm and the intra-row spacing was 30 cm.

Routine operations of irrigation, weed control and pest control were carried out. Irrigation

was done at 3 day intervals. Weeds were controlled mainly by hand pulling. Malasol was

the main pesticide used for pest control.

3.3.5 Data collection

Response variables were number of tillers per crown (tiller density), seed yield, length of

inflorescence, forage fresh biomass yield and forage dry matter yield. After four months, the

number of tillers per crown were counted. The lengths of inflorescences were measured using

a ruler. Seed heads were harvested manually, air-dried, put in bags per plot and weighed. All

these variables were measured for the middle row and the results were multiplied by three since

there were three rows in each plot, to give the results per 3.6 m2. This was used to extrapolate

the seed head yield per hectare. The middle row was used in order to reduce the edge effect.

The seeds were extracted from crushed seed heads through a combination of sieving and air-

blowing until most of the chuff was removed. To estimate the mass of seeds per unit weight of

seed heads, five samples of seed heads weighing 250 g were crushed by putting them in a sack

and thrash them with a log before sieving the chuff off and weighing the seed. The average

weight of the five samples was then used as the weight of seed per 250 g of seed heads. This

46
was then used to extrapolate the seed yield in kilograms per hectare. Soon after seed

harvesting, the above ground plant material was clipped at the base of the plant, cut into shorter

pieces, put into pre-weighed khaki bags and weighed to obtain the forage yield per plot. The

result was used to extrapolate the forage yield in kilograms per hectare. Forage material from

different plots under the same treatment was mixed together and a sub-sample was then taken

from the composite sample for laboratory tests. The sub-samples were placed in weighed and

labelled khaki bags. After weighing them, they were oven-dried at 600C for 48 hours at BUAN

Animal Nutrition Laboratory. The oven-dry weights were used to calculate dry matter (DM)

yields which were extrapolated to kg/ha.

Figure 3: Inspecting the horse manure upon its delivery.

47
Figure 4: Fertilizer application

Figure 5: Planting of the vegetative splits

48
Figure 6: Appearance of tussocks after one week.

Figure 7: Appearance of the grass after 22 days

49
Figure 8: Top dressing grass with urea after emergence of the first inflorescence

Figure 9: Appearance of the grass after 2 months

50
Figure 10: Pests of grass seeds

Figure 11: Appearance of C. ciliaris after 3 months

51
Figure 12: Counting the number of tillers per crown

Figure 13: Measuring the length of the inflorescences

52
Figure 14: Harvesting seed heads

Figure 15: Weighing the seed heads

53
Figure 16: Cutting the vegetative material for weighing.

54
Figure 17: Measuring the forage biomass

Figure 18: Air-dried seed heads

Figure 19: Oven drying the vegetative material

55
3.4 Statistical Analysis

To compare the significant differences in response variables, ANOVA analysis was done using

Procedures for the Generalized Linear Models (PROC GLM) of Statistical Analysis Systems

of 2004. Data was analysed as a completely randomized design with three treatments and in a

completely randomized design with 18 plots, 6 plots for each of the 3 treatments. The effects

of treatments were considered significant at P < 0.05. When means were significantly (P <

0.05) different, multiple comparison of means was conducted using the least squares means

separation which was performed using the PDIFF option (SAS 2004) to evaluate the

significance and magnitude of the fixed effects at P ≤ 0.05.For statistical analysis, the following

model was used:

𝒀𝒊𝒋 = µ + Ƭ¡ + 𝜺ij

𝑤ℎ𝑒𝑟𝑒:

𝒀𝒊𝒋 = 𝑅𝑒𝑠𝑝𝑜𝑛𝑠𝑒 𝑣𝑎𝑟𝑖𝑎𝑏𝑙𝑒

µ = 𝑃𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑚𝑒𝑎𝑛

𝑻¡ = ¡ 𝒕𝒉 𝑇𝑟𝑒𝑎𝑡𝑚𝑒𝑛𝑡 𝑒𝑓𝑓𝑒𝑐𝑡

𝜺ij = 𝑅𝑎𝑛𝑑𝑜𝑚 𝑒𝑟𝑟𝑜𝑟 ~ 𝑵 (𝟎, Ơ𝟐) (𝑝𝑜𝑝𝑢𝑙𝑎𝑡𝑖𝑜𝑛 𝑚𝑒𝑎𝑛, 𝑣𝑎𝑟𝑖𝑎𝑛𝑐𝑒)

56
3.5 Results

Table 3.1: Soil analysis results

Soil analysis results are presented in Table 3.1 below. The results showed that the soil N
was lower than the requirements for most forage grasses.

Parameter
%

Nitrogen
3.92

Phosphorus 0.36

8
Cation Exchange Capacity

Table 3.1 above shows the results from analyzing the soil from the experimental site.

3.5.1 Length of inflorescence

There was no significant (P > 0.05) difference in the average length of inflorescence between

horse manure or urea and the control treatment

3.5.2 Tiller density

The results of this study showed that tiller density did not significantly (P > 0.05) differ

between the two fertilizer treatments, but both fertilizer sources resulted in significantly (P <

0.05) higher tiller density than the control treatment

3.5.3 Seed yield

Both horse manure and urea resulted in a significantly (P < 0.05) higher seed yield than the

control treatment. Generally, horse manure yielded a significantly (P < 0.05) higher seed

yield than urea fertilizer.

57
3.5.4 Forage biomass yield

Urea produced a significantly (P < 0.05) higher fresh biomass and dry matter yield than both

horse manure and the control. Horse manure produced a significantly (P< 0.05) higher dry

matter yield than the control.

Table 3.2: Effects of fertilizer treatments on agronomic parameters of C. ciliaris.

Parameter

Fresh Forage Dry

Length of
Treatment Tillers per Seed Yield Biomass Matter
Inflorescence
crown (n) (kg/ha) Yield Yield
(cm)
(kg/ha) (kg/ha)

No fertilizer 23.67b 11.07c 4346.14c 1627.82c 10.96a

(Control)
Horse Manure 30.00a 18.46a 5447.44b 2237.85b 11.23a

Urea 31.67a 14.51b 6398.53a 2739.62a 11.02a

P-value 0.0013 0.0001 0.0002 < 0.0001 0.65

Treatment (column) means with different superscripts significantly (P < 0.05) differ.

58
3.6 Discussion

3.6.1 Length of inflorescence

There was no significant (P < 0.05) difference between either of the two fertilizer treatments

and the control treatment in average length of inflorescence. There was also no significant (P

< 0.05) difference between horse manure and urea. This could be explained in terms of the

genotypic effect. Since one variety (Molopo) was used, the genetic expression of inflorescence

length could not show significant differences between fertilizer treatments and the control and

between the two treatments themselves. Even the nutrients that were inherently present in the

soil (control treatment) were sufficient for the inflorescences to fully grow. Although the

differences were insignificant, the average length of inflorescence from the use of horse manure

was slightly longer than the use of urea and the control. The inflorescence length has an

influence on seed yield. This could have contributed to the differences that were observed in

this study in seed yield where fertilizing with horse manure produced a higher average seed

yield than urea.

3.6.2 Number of tillers per crown (tiller density)

There was a significant (P < 0.05) difference between horse manure and the two fertilizer

treatments and the control in terms of number of tillers per crown of C. ciliaris. This could

have been due to the high N concentration in urea (46% N) and the sustained supply of N from

the slow releasing horse manure. This concurred with the findings of Laidlaw (2005) where

tiller densities varied with fertilizer application. High levels of nitrogen cause vigorous

sprouting of vegetative material (Mganga, 2009). A sufficient supply of N makes the grass to

recruit more tillers. In addition to adding N to the soil, urea can also increase the vegetative

sprouting of grasses through its effect on soil pH. When urea dissolves in soil water, it raises

the pH of the soil and this is crucial in sandy soils which are usually acidic. According to

59
Donaldson and Rootman (2015), soil pH values of between 5.5 and 6.5 enhance tillering of

forage grasses. This is important because tillering is an important attribute of forage grasses as

it increases the chances of survival and the amount of available forage for livestock (Laidlaw,

2005). When forage grass produces a large number of tillers, such grass normally attains

maximum growth at an earlier age and recovers fast after defoliation (Laidlaw, 2005). Tillering

also has an influence on leaf-area production and dry matter yield (Kumar et al., 2005; Tailor

et al., 1993; Kizima et al., 2012). In this regard, it determines photosynthetic rates and act as

food reserves. According to Mganga (2009), a high rate of tillering complements both forage

yield and resilience of a grass under defoliation. It also contributes to the effectiveness of the

grass towards soil conservation through increased ground cover. This may be important for

forage grasses that have to be grazed in-situ and those used for soil conservation and land

rehabilitation but not so important for grasses that are being produced for seed production.

Ordinarily, the more the tillers the more the inflorescences and hence the expected seed yield

as tillers produce inflorescences which bear the seed.

This study, however, produced results to the contrary. Fertilizing with horse manure,

which produced less tillers than urea, yielded more seed by weight, though the weights did not

differ significantly. It would seem as if too much vegetative material lowered the seed yield.

This is in agreement with the findings of Hassan and Fikru (2015) who reported a highly

negative correlation between seed yield and tiller density. Such a correlation agrees with the

classical trade-off between reproductive and vegetative allocation of resources. The overall

treatment mean of 28.44 tillers per crown in this study was higher than 13.30 tillers per crown

reported by Meena (2015) for C. ciliaris and this could have been due to the fact that the results

reported by Meena (2015) were for a research carried out under rain-fed treatments while this

study used irrigation. Lower tiller numbers under rain-fed treatments could be attributed to the

response of the grass to some water stress during the growing season. Moisture stress reduces

60
tillering as a way of reducing transpiration losses (Hassan, 2015). Tillering also has an

influence on forage digestibility. According to Kumar et al. (2005), tillers contain leaves which

have more easily digestible nutrients and less structural components, especially from newly

developed tillers with young leaves. Typically, however, a tiller will have both old and new

leaves. The old leaves would be lower in quality but would also contribute to biomass yield.

There was no significant (P > 0.05) difference between the use of horse manure and the use of

urea in terms of tillering. This could mean that fertilizing with either horse manure or urea

provided sufficient levels of N in the soil for the tillering of the grass. This means that where

tillering is the positive attribute being sought from the forage grasses, a farmer can use either

of the nitrogen sources without significantly affecting the tillering. The issue of cost may then

need to be considered.

3.6.3 Seed Yield

The future survival and general propagation of forage grass are determined by the ability of the

forage grass to produce seed (Ogillo, 2010). This study recorded a significant (P < 0.05)

difference in the mean seed yields between fertilizing with either horse manure or urea and the

control treatment and between the fertilizer treatments themselves. The amount of N supplied

by the two fertilizers could have been sufficient for causing a significant increase from the

control treatment. N is a nutrient that is highly mobile and does not stay long in the soil. The

length of time from the previous application of a N fertilizer and the carrying out of this

research could have been long enough to result in insufficient amounts of N in the soil. This

experiment was carried out 5 months after the last N fertilizer application in the area. Loss of

nitrogen from the soil could have been due to leaching, denitrification, absorption by the

previous crop or volatilization (Donaldson and Rootman, 2010). When the fertilizers were

added, they raised the N content to levels that caused a significant difference in seed yield. The

significant difference in seed yield between fertilizing with horse manure and using urea could

61
be attributed to the slow and steady release of nitrogen from horse manure. Nitrogen, a very

important nutrient for seed formation, would be available to the grass throughout its growing

and seed formation stages and in measured amounts (Koech et al., 2014). Due to its fibrous

nature, horse manure decomposes slowly and releases its nutrients steadily over a long time.

Sufficient N in grass enhances the process of photosynthesis which builds carbohydrates that

would then be stored in seeds (Tailor et al., 1997).

The slow release means that there is no N over-supply during the seed formation stage.

Such an over-supply of N at this stage would result in grass having a more vigorous vegetative

sprouting at the expense of seed formation (Hassan and Fikru, 2015). When there is a lot of

vegetative growth, there is often a corresponding vigorous growth and development of roots so

that absorption of water and nutrients can sustain the above ground material. This means that

a lot of energy would be expended on root growth rather than on seed formation. Well

established root systems do not guarantee high seed yields but vigorous sprouting during

regeneration (Bulle et al., 2009).

Urea, on the other hand, supplies N in surges and the N quickly disappears from the soil

through absorption, volatilization or leaching. Such losses may cause N deficiencies at critical

times in the formation of seeds, resulting in reduced seed yields. Fertilizing with horse manure

avails other nutrients like potassium which facilitate formation of starch and its storage in the

seeds, while urea only avails N. As an organic fertilizer, horse manure improves the structure

of the soil, raising the water holding capacity of the characteristically sandy soils as were the

soils in this study. This could have ensured the availability of moisture for a reasonable amount

of time after irrigation. Seed formation requires a lot of moisture at the initial stages as the seed

itself would be about 90% water by composition at that stage. A similar study by Kizima et al.

(2012) recorded an average seed yield of 77.5 kg/ha, which was way higher than the average

of 14.68 kg recorded in this study. The difference could be explained in terms of higher levels

62
of N that were used than in this study. The different agro-ecological zones in which the

experiments were carried out could also be a factor. The results of this research were, however,

close to the findings of Ashraf et al. (2013) of 21.8 kg/ha. Kumar et al. (2005) reported seed

yields of 75 kg/ha at wider spacing of grasses. This could have been because of reduced

competition for moisture, nutrients and sunlight. Wider spacing results in better transmission

of light to the lower canopy, which results in greater tillering of grass. The seed yield in this

study could have been lowered by wild birds. Despite erecting scare-crows and using cassette

tape to scare wild birds away, the birds were a constant bother after seed setting.

3.6.4 Forage Biomass Yield

Fertilizing C. ciliaris with either horse manure or urea caused a significantly (P < 0.05) higher

forage fresh biomass yield than the control treatment. This could have been due to the fact that

nitrogen makes plants to be hydrophilic (Tailor et al., 1993). According to Tailor et al. (1993),

high levels of N in a plant make the plant to absorb a lot of potassium ions from the soil. This

lowers the osmotic potential in the plant root hair cells, causing the soil to get into the root hair

cells from the soil by osmosis. Due to increased leaf growth, absorption of water from the soil

is increased to cater for losses due to transpiration. This results in plant cells to be turgid. The

forage fresh biomass may be important in areas where water for livestock consumption is scarce

and the grass is fed fresh to the animals.

Ruminant livestock production is directly and overly influenced by the dry matter

productivity of forage plants (Hare et al., 2009). Fertilizer regimes that yield the highest dry

matter should be used when forage production is for livestock feed but a compromise may need

to be reached when the aim is to produce seed. This study showed a significant (P < 0.05)

difference in forage dry matter yield between the two fertilizer treatments (horse manure and

urea) and the control treatment. This could have been because of the N added to the soil through

the fertilizers. Application of N to the soil increases vegetative (above ground) growth of forage
63
grasses. According to Guiot and Melendez (2003), high N content in the soil results in large

sized leaves and thick but soft stems in forage grasses. This causes an increase in forage dry

matter. When there is sufficient N, chlorophyll formation is enhanced and more photosynthesis

takes place. The presence of carbohydrates (from photosynthesis) and N (absorbed from the

soil) in a plant results in the formation of amino acids which are precursors for protein synthesis

which contribute to the increase in the dry matter yield. Some of the proteins formed would be

growth hormones such as auxin. Auxin causes an increase in the above ground growth (shoots),

hence contributing to an increase in forage dry matter yield. The leafy nature of forage grasses

might, however, be retrogressive in dry areas where water supply is limited, as it speeds up

water loss through transpiration (Guiot and Melendez, 2003).

This study recorded a significantly (P < 0.05) higher forage dry matter yield when

fertilizing with urea than using horse manure. This could be attributed to the high concentration

of N in urea (46%). The N in the horse manure could have been released slowly and steadily

over a long time and as such, may not have been sufficient at certain crucial times. Fertilizing

with horse manure produced a lower forage dry matter yield probably because of weed

pressure. There were always a problem of weeds on plots that were treated with horse manure.

This could be due to weed seeds that were ingested by the horses, passed through the digestive

system intact and were present in the horse droppings. As far as the results of this study are

concerned, fertilizing with urea would be more preferable as a source of N for forage grasses

if dry matter yield is the most desirable attribute and not the seed yield. A very high N

concentration in the soil due to urea application can, however, cause a greater risk of nitrate-

nitrogen concentrations exceeding 1000 ppm. Such high nitrate levels may affect animal health

(Donaldson and Rootman, 2010). As such, a steady supply of moderate levels of nitrogen, such

as from horse manure, would be preferable. The mean forage dry matter yields of 2237.85

kg/ha when fertilized with horse manure and 2739.62 kg/ha for urea in this study were higher

64
than the 1913.30 kg/ha reported by Sawal et al. (2009) when the C. ciliaris was fertilized with

75 kg N/ha. The DM yield of this study was, however, less than an average of 3790 kg/ha

when the grass was fertilized with sheep manure and 4210 kg/ha for NPK fertilizer as reported

by Meena (2011). This could have been due to differences in soil, climatic and management

differences. The control treatment in the results in Meena’s report averaged 3345 kg/ha while

in this research, it averaged 1627.82 kg/ha. This points to the inherent soil and climatic

conditions. The DM yield of this study was, however, quite comparable with the 2420 kg/ha

that was reported by Donaldson and Rootman (2010), albeit, without any nitrogen applied. The

soil that was used could have been rich in nitrogen from the fertilizers applied for the previous

crops.

65
3.7 Conclusion and recommendations

The findings of this study show that fertilizing with horse manure or using urea as N sources,

significantly increases the seed yield, tiller density and forage dry matter yield of C. ciliaris.

This is important as farmers who may want to venture into seed production can have income

from both seed production and forage production from residue after seed harvesting. The

results also show that fertilizing with horse manure, which is a cheap resource, causes a

significantly higher seed yield than fertilizing with urea. The use of horse manure is, therefore,

of greater importance when the objective is to produce seed for the establishment of seed banks.

Fertilizing with urea causes the production of a significantly higher forage dry matter

yield than using horse manure. This means that the fertilizer can be of higher value when

harvesting seed is not part of the objectives. This is because urea encourages more vegetative

growth which compromises seed production. The slow N release by horse manure ensures its

measured availability which moderates vegetative growth and enhances seed production.

Considering the cost implications and how easily the fertilizer can be obtained, horse manure

would be more appropriate for seed production, especially for large scale and rural livestock

farmers who may want to propagate C. ciliaris in their pastures or establish planted pastures

through seed. Propagation of the forage grass using vegetative splits would be quite costly in

terms of labour and time. The slow nutrient release by horse manure may also mean that there

would be reduced leaching and, hence, less chances of eutrophication in water bodies.

66
CHAPTER 4: STUDY TWO

Chemical composition and in vitro digestibility of C. ciliaris var. Molopo forage grass post-

harvest residues fertilized with horse manure or urea.

4.0 Introduction

Botswana usually has short rain seasons which make the country experience prolonged

dry winters. This poses a constraint to livestock farmers in their management of fodder flow.

This can be overcome by using forage grasses that are resilient to dry conditions and do not

lose their quality drastically with advanced maturity. When allowed to grow and set seeds, the

nutritional value of the grass should not deteriorate to levels that cannot, at least, provide

sufficient nutrients for maintenance. Soils in Botswana are predominantly sandy-textured, with

very low organic matter content, which results in deficiencies of essential nutrients. According

to Joshua (1988), most of the soils in the southern parts of Botswana are predominantly sub-

desert soils with low levels of carbon (< 0.5%) and low clay proportion (< 40%). Van Waveren

(1988) also reported that the soils in Botswana were generally light textured with low

phosphorus content. Grasses that grow in such soils lack the essential nutrients as they can only

contain what they absorb from the soil, in terms of mineral nutrients. The low fertility of the

soil results in forage grasses of low quality.

N fertilization influences the nutritional value of forage grasses (Hassan et al., 2015).

According to Ramirez et al. (2009), the crude protein concentration in forage grasses is

influenced mainly by the supply of available N in the soil and the state of maturity of the grass.

Nitrogen fertilization has also been shown to increase the dry matter digestibility of C. ciliaris.

However, there are different sources of nitrogen which may elicit different responses

depending on the supply or release of nitrogen to the soil and plants. Information regarding the

comparative effects of different sources of nitrogen on the nutritional quality of forage grasses

67
is important. Minerals are required to meet the needs of livestock for optimum development,

health and productivity (Monroe, 1996). According to Albu (2012), calcium and phosphorus

play a very important role in the growth and development of animals. These two macro-

elements should be analysed in combination because the dietary levels of calcium and

phosphorus should be balanced to increase their availability and utilization Ogren et al. (2014).

Ogren et al. (2014) explained that if forage supplies more phosphorus than calcium to a

livestock, calcium absorption can be impaired and skeletal malformations, poor growth and

muscle disorders can occur. Even if the total diet contains adequate calcium, excessive

phosphorus intake may cause abnormalities (Ogren et al., 2014).

Even when the objective is seed production, forage residues that remain after

harvesting seeds can be a useful feed resource for ruminants and hence the mineral composition

in these residues needs to be known. The source of N that results in the increased nutritional

quality of C. ciliaris after seed harvesting, needs to be found. With the concept of sustainable

Agriculture taking traction in Botswana, organic fertilizer will become more important.

However, a possible source of organic fertilizer; manure from horses, have not been researched

adequately except a few studies in the 1980s by Department of Agriculture Research (APRU,

1985). The two possible sources of nitrogen that this study seeks to compare are horse manure

and urea.

4.1Specific objectives

-To determine the chemical composition of C. ciliaris forage residues after seed
harvesting.

-To determine the in vitro dry matter digestibility of C. ciliaris forage residues after
seed harvesting.

68
4.2 Hypothesis

𝐻0: There is no significant difference in chemical composition and digestibility (forage

quality) of C. ciliaris forage grass residue after the use of horse manure or urea as

fertilizer treatments and no fertilizer (control treatment).

𝑯𝟎: µ 𝒉𝒐𝒓𝒔𝒆 𝒎𝒂𝒏𝒖𝒓𝒆/𝒖𝒓𝒆𝒂 = 𝝁 𝒄𝒐𝒏𝒕𝒓𝒐𝒍

𝐻𝑎: There is a significant difference in chemical composition and digestibility (forage

quality) of C. ciliaris forage grass residue after the use of horse manure or urea as

fertilizer treatments and no fertilizer (control treatment).

𝑯𝒂: µ 𝒉𝒐𝒓𝒔𝒆 𝒎𝒂𝒏𝒖𝒓𝒆/𝒖𝒓𝒆𝒂 ≠ 𝝁 𝒄𝒐𝒏𝒕𝒓𝒐𝒍

4.3 Materials and methods

For details of the methods, see sections 4.3.1.1 up to 4.3.2.2

4.3.1 Determination of chemical composition of the forage grass residue

After determination of the dry matter yield for the control, horse manure treated and urea

treated, the dried C. ciliaris forage biomass samples from the same treatment were mixed

together and a subsample extracted from each composite sample for chemical analysis. The

subsamples were ground through a 1mm sieve Wiley mill and stored in air-tight plastic bottles

in preparation for the analysis.

4.3.1.1 Ash determination

Samples were analyzed for Ash by combustion in the muffle furnace. Ash content was

determined using the formula:

(𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑐𝑟𝑢𝑐𝑖𝑏𝑙𝑒 + 𝑎𝑠ℎ − 𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑐𝑟𝑢𝑐𝑖𝑏𝑙𝑒) ∗ 100

% 𝐴𝑠ℎ =
𝑤𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑜𝑣𝑒𝑛 𝑑𝑟𝑦 𝑠𝑎𝑚𝑝𝑙𝑒

69
4.3.1.2 Determination of phosphorus content and mineral cations

For the determination of phosphorus and mineral cations, the 6 samples for each treatment were

mixed together and subsamples taken for testing, in duplicates. The phosphorus content was

determined using a spectro-photometer using the Molybdenum Blue Method of Dickman and

Bray (1940). Plant sample solutions were prepared by digesting 1.25 g of each of the C. ciliaris

samples. The blank did not contain any Cenchrus material and was used as a correction factor.

One ml of plant digest solution was put into a 50 ml glass beaker using a pipette. Thirty ml of

chloromolybdic acid working solution were added and gently mixed. One ml of stannous

chloride working solution was also added and mixed gently. The spectrophotometer was

calibrated and the wavelength set to 670 nm. The absorbance of the blanks and samples were

measured at 10 minutes intervals. A possible source of error is the presence of contaminants in

water, reagents or glassware. For this reason, a blank was run at the same time as the samples.

This meant that if the samples contained the same concentrations of contaminants as the blank

then the concentration of the mineral in the blank was subtracted from the value determined

for the samples. Mineral cations (Ca2+, Mg2+, K+, Na+, Cu2+, Zn2+ and Fe2+) were determined

as per the standard methods described in AOAC (2005) using a Perkin Elmer ICP-Optical

Emission Spectrometer Optima 7300 DV Series.

4.3.1.3 Determination of nitrogen content

Digestion, neutralization and titration were all done manually.

4.3.1.3.1 Digestion

The nitrogen of protein was transformed into ammonium sulfate by acid digestion with boiling

sulfuric acid. 1.25 g of ground sample material was weighed onto a piece of lens tissue. The

lens tissue was folded carefully and dropped into digestion tube. This was done for all the

samples. Two blanks of only lens tissue were also put into their own digestion tubes. 20ml of

70
98% sulfuric acid and some selenium solution were added to the digestion tubes using

dispensation. The tubes were inserted into the digestion blocks and rubbers inserted. The block

was switched on and temperature was controlled as follows; 150 oC for 1 hour, 250oC for the

following hour and 330oC for the last two hours. After a total of four hours of digestion, the

block was switched off and was left to cool for 60 minutes. 4 ml of hydrogen peroxide were

added to each tube. The tubes were inserted onto the digestion block, the scrubber unit put and

the digestion block switched on for 2 hours at 330oC. After 2 hours, the tubes were removed

from the block and allowed to cool in the fume cupboard for 50 minutes. The contents of each

tube were transferred into a volumetric flask of 250 ml capacity and 100 ml of distilled water

were added. After 24 hours the contents were then made up to volume with distilled water.

4.3.1.3.2 Neutralization and titration

Twenty-five ml of each solution was added into a distillation tube using a pipette and 40ml of

3% sodium hydroxide was automatically added by the distillation unit. The distillation was

carried out for 4 minutes. The distillate was collected in 1% boric acid solution and then titrated

with standard sulfuric acid. The blank samples were treated in the same way as samples. The

nitrogen percentage in the sample was determined using the formula

%𝑁 =

% [𝑣𝑜𝑙𝑢𝑚𝑒 𝑜𝑓 𝑠𝑡𝑎𝑛𝑑𝑎𝑟𝑑 𝑎𝑐𝑖𝑑 (𝑐𝑚3 ) − 𝑣𝑜𝑙𝑢𝑚𝑒 𝑜𝑓 𝑏𝑙𝑎𝑛𝑘(𝑐𝑚3 )] 𝑥 𝑀𝑜𝑙𝑎𝑟𝑖𝑡𝑦 𝑜𝑓 𝑎𝑐𝑖𝑑 (𝑁) 𝑥 1.4
𝑊𝑒𝑖𝑔ℎ𝑡 𝑜𝑓 𝑠𝑎𝑚𝑝𝑙𝑒(𝑔)
The conversion of nitrogen percentage to protein percentage was done by multiplying the

percentage nitrogen by 6.25 as follows:

𝑃𝑟𝑜𝑡𝑒𝑖𝑛 % = 𝑁𝑖𝑡𝑟𝑜𝑔𝑒𝑛 % 𝑥 6.25

71
4.3.1.4 Determination of Neutral Detergent Fiber (NDF) and Acid Detergent Fiber
(ADF)

The NDF and ADF determinations were done in duplicates. Twelve empty fiber bags were

weighed. Two fiber bags were used for each of the three treatments. A sample amount of 0.5 g

was placed in each of the two bags per treatment and weighed. This gave a total of 12 sample

bags. Two empty fiber bags were included to determine the NDF and ADF in the blank bags

for correction. The bags were heat sealed. The NDF and ADF were then determined according

to the AOAC procedures of 2009 and Van Soest’s proximate analysis (1994)

4.3.2 In vitro digestibility determination

4.3.2.1 Animal care and ethics

Animals that donated rumen fluid and those which nylon bags were incubated in were cared

according to international guidelines for biomedical research involving animals (Council for

International Organization of Medical Science-CIOMS, 1985).

4.3.2.2 Animal handling, feeding, collection and processing of rumen liquor for in vitro
dry matter digestibility

A rumen cannulated ox that was kept at the Department of Agricultural Research kraals was

used as the source of rumen liquor for in vitro digestibility determination. The rumen liquor

was collected and kept in thermos flasks pre-heated to 390C with hot water and flashed with

carbon dioxide in order to maintain anaerobic conditions inside. This is important as oxygen is

toxic to rumen bacteria.The ground forage samples were put in fiber bags which were heat

sealed and placed in individual flasks and incubated with rumen liquor containing rumen

microbes. The flasks also contained buffers, macro-minerals, trace-minerals, nitrogen sources

and reducing agents to maintain pH and provide nutrients required for growth of rumen

bacteria. The bags were incubated in the flasks for 96 hours. At the end of the incubation period,

72
the bags were rinsed four times. Buffer solution consisted of two solutions prepared according

to Ankom Daisy Incubator digestibility procedure (solution A and solution B; Ankom

Technology Corporation, Fairport, NY, USA). Solution A was made of 20 g of KH2PO4, 1.0 g

MgSO4.7H2, 1.0 g NaCl, 0.2 g CaCl2.H2O and 1.0 g Urea (reagent grade) in 2 litres of distilled

water while buffer Solution B consisted of 15 g Na2CO3 and 1 g Na2S.9H2O in 1 litter of

distilled water. Both buffer solutions (A and B) were pre-warmed to 39oC by placing them in

a water bath set at 39oC before use. In separate containers, 266 ml of solution B was added to

1330 ml of solution A (1:5 ratio). The exact amount of solution A to B was adjusted to obtain

a final pH of 6.8 at 39°C. Four hundred millilitres (400 ml) of rumen inoculum in a graduated

cylinder was added to the mixture of solution A and B (1:4 ratio) with distilled water, dried,

weighed and placed in an ANKOM fiber analyzer using reagents specified in AOAC (2005)

for NDF and refluxed for 60 minutes in neutral detergent solution. In vitro total dry matter

digestibility (IVTDMD) was computed as the difference between dry matter incubated and the

residue after the neutral detergent analysis. The following formula was used:

%𝐼𝑉𝑇𝐷𝑀𝐷 =[100– (𝑓𝑖𝑛𝑎𝑙 𝑤𝑒𝑖𝑔ℎ𝑡 𝑎𝑓𝑡𝑒𝑟 𝑁𝐷𝐹 𝑒𝑥𝑡𝑟𝑎𝑐𝑡𝑖𝑜𝑛 − 𝑏𝑙𝑎𝑛𝑘) /

𝑖𝑛𝑖𝑡𝑖𝑎𝑙 𝑤𝑒𝑖𝑔ℎ𝑡 𝑥100]

4.4 Statistical analyses

Data were analysed using the General Linear Model in Statistical Analysis Systems (SAS,

2004) and simple correlation using Pearson’s correlation analysis (SAS, 2004). Data was

considered as a completely randomized design with two treatments and a control, using 18

plots, 6 plots for each treatment. The following model was used:

The treatment effects were considered significant at P < 0.05. When means were significantly

different at P < 0.05, multiple comparison of means was conducted using the least squares

means separation which was performed using the PDIFF option (SAS, 2004) to evaluate the

significance and magnitude of the fixed effects at P ≤ 0.05

73
 𝒀ij = µ + Ƭ¡ +𝜷𝒋 + 𝜺ij

𝐰𝐡𝐞𝐫𝐞:

𝐘ij = 𝐑𝐞𝐬𝐩𝐨𝐧𝐬𝐞 𝐯𝐚𝐫𝐢𝐚𝐛𝐥𝐞

µ = 𝐩𝐨𝐩𝐮𝐥𝐚𝐭𝐢𝐨𝐧 𝐦𝐞𝐚𝐧

Ƭ¡ = 𝐢𝐭𝐡 𝐭𝐫𝐞𝐚𝐭𝐦𝐞𝐧𝐭 𝐞𝐟𝐟𝐞𝐜𝐭

𝛃𝐣 = 𝐣𝐭𝐡 𝐛𝐥𝐨𝐜𝐤𝐢𝐧𝐠 𝐞𝐟𝐟𝐞𝐜𝐭

𝜺ij = 𝐑𝐚𝐧𝐝𝐨𝐦 𝐞𝐫𝐫𝐨𝐫 ~ 𝑵(𝟎, Ơ𝟐)(𝐩𝐨𝐩𝐮𝐥𝐚𝐭𝐢𝐨𝐧 𝐦𝐞𝐚𝐧, 𝐯𝐚𝐫𝐢𝐚𝐧𝐜𝐞).

4.5 Results

The post-harvest nutritional quality of the C. ciliaris forage residues determines how useful the

residues can be as animal feed sources and their contribution to the overall income to the

farmer.

4.5.1 Nutritional quality of the forage residue

The results in Table 4.1 show that fertilizing C. ciliaris with either horse manure or urea caused

a significantly (P < 0.05) higher crude protein content in forage residues than the control

treatment. The control treatment caused a significantly (P <0.05) higher ADF than the two

fertilizer treatments. The NDF was also significantly (P < 0.05) higher for the control

treatment than for either horse manure or urea. The findings of this study also show that C.

ciliaris from the control treatment had the highest Ash content (12.86%), which was

significantly (P < 0.05) higher than when fertilized with horse manure (11.74%) or urea

(12.22%). The IVDMD of the forage when the C. ciliaris was fertilized with horse manure or

urea was significantly (P <0.05) higher than the control treatment.

74
Table 4.1: Effects of fertilizer treatments on chemical composition and DM digestibility

(%DM) of C. ciliaris after seed harvesting

Treatment CP ADF NDF Ash

IVDMD

No fertilizer (control) 10.11a 43.82b 37.45b 12.86b 33.73a

Horse manure 12.29b 39.53a 34.60a 11.74a 44.14b

Urea 12.43b 39.62a 35.94a 12.22a

37.81b

P – value 0.0001 0.0113 0.0290 0.0438

0.0001
Treatment (column) means with similar superscripts do not significantly (P < 0.05) differ
CP= Crude Protein; ADF= Acid Detergent fiber; NDF=neutral; IVDMD= In Vitro Dry Matter
Digestibility.

4.6 Pearson’s Correlation Coefficients between crude protein and ADF, NDF and
IVDMD

This study showed that there was a generally moderate negative linear association (r = -

0.52146) between ADF and crude protein as shown in Table 4.2. As the ADF increases, the CP

generally decreases but this relationship between the two variables was, however, statistically

insignificant (P > 0.05). There is no probability of less than 0.05 that r = -0.52146, with n=6,

could occur if there is no relationship between the two variables. Table 4.2 also shows a

generally negative association (r = -0.57744) between NDF and CP. This relationship was

significant (P < 0.05). There was a relatively large positive relationship (r = 0.70639) between

crude protein and dry matter digestibility. For larger values of CP, the values of IVDMD are

75
also larger and there is conclusive evidence about the significance of the association between

these two variables (P < 0.05).

Table 4.2: Correlation coefficients between ADF, NDF, IVDMD and CP. It shows a

positive correlation between CP and Dry Matter digestibility.

Crude Protein

ADF -0.52146
0.2887

NDF -0.57744
0.0121*

IVDMD 0.70639
0.001*
Values with a * denote significant levels

The mineral content analysis results for the forage residues are shown in Table 4.3. They
showed that, except for sulphur, there was no significant (P > 0.05) difference in mineral
content between fertilizer treatments and the control treatment. There was no significant (P >
0.05) difference between fertilizing with urea and using horse manure. For sulphur, fertilizing
with horse manure or urea caused a significantly (P < 0.05) higher content of the mineral than
the control treatment.

76
 Table 4.3: Mineral content (mg/kg DM) of C. ciliaris forage residue after seed

harvesting

Treatment P Mg Ca K Na Cu Zn Fe
S

No fertilizer (control) 0.49a 0.40a 0.49a 1.55a 0.35a 0.95a 1.49a 1.11a
0.23a

Horse manure 0.52a 0.43a 0.56a 1.72a 0.37a 0.97a 1.36a 1.30a
0.40b

Urea 0.55a 0.46a 0.57a 1.89a 0.37a 0.91a 1.41a 1.32a

0.44b

P – value 0.33 0.73 0.47 0.06 0.95 0.45 0.24 0.42

0.01
Treatment means with different superscripts differ significantly (P < 0.05)

77
4.7 Discussion

4.7.1 Crude Protein

Crude protein and digestible dry matter are the most important components of a feed (Afzal

and Ullah, 2007). In the present study, fertilizing with horse manure or urea resulted in

significantly higher crude protein levels in the forage residues than the control treatment. The

N applied to the soil through horse manure or urea was absorbed and may have been rapidly

converted to nitrate-nitrogen and was then incorporated into organic materials (Abedelrahman,

2007). The crude protein contents of C. ciliaris in the present study (12.59% when fertilized

with horse manure and 12.43% with urea) were significantly higher than the 7.9% reported by

Al- Dakheel (2015) when no fertilizer was used. Walker (2013) reported a protein content of

C. ciliaris of 10.2% in the wet season and 4.2% in the dry season, which were lower than the

findings of the current study. Yossin and Yassin (2012) reported that the crude protein content

of grass depends on soil nitrogen availability and that fertilization increases it.

The result of the current study, with an average of 11.71% crude protein was

comparable to the 12% reported by Ramirez et al. (2009). Donaldson and Rootman (2010)

reported a 15.63% crude protein content in C. ciliaris while Mcdowell (2003) reported a 17.5%

and the differences between these two sets of results and those of the current study could be

because of the differences in the supply of available N in the soil and the state of maturity of

the grass. According to Ramirez et al. (2009), the crude protein content of forage grass

markedly declines with maturity, possibly because of the relative increase in cell wall and

decrease in cytoplasm. The crude protein content of the C. ciliaris across treatments, in the

current research, was higher than the minimum requirements for ruminants (Mcdowell, 2003;

6.9% for maintenance, 10% for beef production and 11.9% for milk production). It was the

control treatment (10.11%) that resulted in the grass having less crude protein than the

minimum required for milk production. This means that if the forage grass residues after seed

78
harvesting are meant to be fed to lactating cows, grass needs to be fertilized. It also means that

soon after harvesting seed, the forage residues need to be quickly harvested before further

decline in crude protein content. In the context of this study, very high levels of nitrogen would

not be recommended due to the negative correlation between seed yield and vegetative biomass

yield.

Significant increase in forage N content has shown to increase the soil water extraction

by forages and improve water use efficiency (Yassin and Yossin, 2013). There was no

difference in crude protein content between fertilizing with horse manure and fertilizing with

urea. Since there is no difference between fertilizing using the two, the source of nitrogen that

would be economically beneficial and provide a profit would be ideal. All other factors held

constant, horse manure would be preferable in that regard since its costs are likely to be less.

Despite being insignificantly a difference, fertilizing with horse manure produced a higher

crude protein content in the forage than urea. This could be because when using irrigation as

was the case in the current study, the water can lead to movement of nitrates beyond the root

zone since it is very mobile in the soil. This would be more pronounced when urea is used than

when horse manure is used as the horse manure increases the water retention capacity of the

soil (Mudenda and Maeresera, 2009).

Apart from losses through leaching, N in urea can also be lost through volatilization

(Yossin and Yassin, 2013). The level of phosphorus in the soil also affects the efficiency with

which forages utilize N. According to Mcdowell (2003), low levels of phosphorus in the soil

reduces the efficiency of nitrogen utilization. This could explain the slightly higher crude

protein content in forage due to application of horse manure (12.59%) than due to urea

(12.43%) as horse manure supplies phosphorus and other nutrients while urea only supplies N

(Albu, 2012; Erikssen et al., 2009; Keskinen et al., 2017). Despite the fact that for forage grass

to be productive, the first priority should be given to nitrogen, the grass responds to nitrogen

79
rapidly and vigorously when phosphorus and potassium are adequate (Arshadullah et al.,

2011). The fact that the difference between the effect of horse manure and urea is not significant

in this study, however, shows that the soil used in this research had relatively sufficient

phosphorus.

4.7.2 Dry Matter Digestibility

Fertilizing C. ciliaris with horse manure or urea resulted in significantly higher dry matter

digestibility than the control treatment in the current study. This could have been because the

N supplied by these two fertilizers stimulated the growth of new tillers, shoots and leaves and

accelerated the rate of stem development and accumulation of dead materials which were low

in cell wall and lignin content, leading to higher digestibility (Ros Barcelo, 1997). This is

contrary to what was reported by During and McNaught (2012) that fertilization has usually

little or no effect on forage digestibility. There was no difference in dry matter digestibility

between fertilizing with horse manure and with urea indicating that the two fertilizers similarly

enhance digestibility of the forage residues by effectively reducing the proportion of ADF and

increasing the protein content as shown in the current study.

Fertilization with phosphorus, potassium or other nutrients that increase yield may

slightly reduce forage quality when growth is rapid (During and McNaught, 2012), followed

immediately by maturation of tissues. According to During and McNaught (2012), maturity at

harvest of seeds has the greatest influence on forage digestibility. As forages mature, their

digestibility declines significantly. It is at the flowering stage that accumulation of stem mass

starts to exceed leaf mass addition and stems contain a higher proportion of thick walled xylem

tissue and less photosynthetic mesophyll tissue (During and McNaught, 2012). Overall herbage

cell wall concentration increases as the leaf: stem ratio shifts towards a greater proportion of

stem. Ashraf et al. (2013) reported that if a tissue begins to deposit lignin in its cell wall due to

advanced maturity, its digestibility rapidly declines. Lignin in plant cells is more difficult for
80
rumen bacteria to digest than cellulose and hemicellulose. The in vitro dry matter digestibility

of 37.81% when fertilized with urea and 44.14% when fertilized with horse manure found in

the current study, were below the 50% that Hassan et al. (2015) reported as the critical threshold

level for ruminant efficient digestion in the rumen. Values lower than that would limit intake

through delayed dry matter degradation resulting from inefficient microbial environment.

The digestibility of the forage grass after seed harvesting found in the current study

was also lower than the 53.8% - 70.5% range for C. ciliaris that was reported by Donaldson

and Rootman (2010). It was also lower than the 67.5% reported by Waramit and Moore (2006).

This could be attributed to the differences in the stage of maturity at seed harvesting. According

to Hassan et al. (2015), if C. ciliaris is grown in a season with a lower environmental

temperature, its quality could be higher. During the first two months of the current study

(December and January), the temperatures were quite high. Higher environmental temperatures

encourage lignification, rapid physiological development and metabolic activity, resulting in

the decline of forage quality (Hassan et al., 2015). The findings of this research could therefore,

be made more meaningful if it is carried out across seasons. The grinding of the forage grass

to 1mm before the in vitro dry matter digestibility determination increases the proportion of

the cell wall which is immediately accessible to the microbes in the rumen liquor. This may

result in higher digestibility indication than what actually happens in reality.

4.7.3 Neutral Detergent Fibre (NDF) and Acid Detergent Fibre (ADF)

Fibrosity indicates the extent to which the grass can be degraded by rumen micro-organisms

(Topps, 1996). The NDF and ADF content of C. ciliaris with no fertilizer was higher than when

the grass was fertilized with either horse manure or urea. There was, however, no difference

between the effect of horse manure and urea on both ADF and NDF content of the C. ciliaris.

Without addition of nutrients, especially nitrogen to the soil, forage grass tends to mature,

flower and set seed early. This could be in a bid to successfully reproduce in an environment
81
that is not quite conducive and would, otherwise, shorten the life span of the grass. Since the

forage grasses were harvested at the same time in the current study, the grass from the control

may have been more mature, more lignified and hence more fibrous than the grass fertilized

with horse manure and urea. According to McNaught (2012), the flowering stage heralds the

start of the accumulation of stem mass exceeding leaf mass addition. This increases the

fibrosity of the grass.

The ADF findings of the current study (40.99%) was within the range of36.60% -

47.70% reported by Al-Dakheel (2015) and higher than the 38.50% reported by Donaldson and

Rootman (2010). NRC (1989) recommended a minimum dietary ADF content of 19-21% DM

for lactating cows. Lu et al. (2005) recommended dietary ADF of 18-20% DM for goats. The

ADF in the current study could have been higher due to the fact that the grass was analysed

after seed harvesting and this shows that their maturity was advanced. Nsinamwa et al. (2005)

concurred as they reported that the fibre content of forage increases with age. The NDF of 36%

found in this study was lower than the 66.5% reported by Donaldson and Rootman (2010) and

the 70.2% reported by Al-Dakheel (2015). It was lower than the NDF content of 25-28% DM

recommended by NRC (1989) for lactating cows and 41% for lactating goats (Lu et al, 2005).

The fibre content of the C. ciliaris in the current study was comparable to the 37.34% reported

by Ashraf et al. (2013) for C. ciliaris grown in Cholistan desert, Pakistan. Koech et al. (2014)

qualified the forage grass with crude fibre of less than 50% to be of high quality and that with

more than 60% to be low quality. This means that the forage residues were still of high quality

even after seed harvesting. This can be attributed to the fact that the grass was irrigated and

hence did not get moisture stressed. According to Topps (1996), moisture stress causes grass

to mature early and become more fibrous.

82
4.7.4 Minerals

Minerals are vital for normal growth, reproduction, health and proper functioning of the animal

body. They protect and maintain the structural components of the body, organs and tissues and

are constituents of body fluids and tissues as electrolytes (Taylor, 1997). They catalyse several

enzymatic processes and hormone systems and maintain the acid-base balance, water balance

and osmotic pressure in the blood and cerebral spinal fluid (Soni et al., 2014). Both deficiencies

and excessiveness of minerals have adverse effects on the animal body and production

(Mcdonald et al., 2011). It is, therefore, important to know the mineral composition of forage

feed if it is to be used to feed livestock. The results of this study show that there was no

difference in concentrations of most minerals between fertilizer treatments and the control. The

concentrations of these minerals were within the sufficient range requirements for livestock

(Soni et al., 2014). This means that livestock that could be fed on the forage residues after seed

harvesting would not have deficiencies.

The average phosphorus content of the C. ciliaris of 0.518% in this study was in the

same range as the 0.53% reported by Sawal et al. (2009) and the 0.15% to 0.65% range reported

by Ramirez et al. (2009). It was above 0.25% DM critical level for maintenance of a dry cow

as reported by Soni et al. (2014). Phosphorus is critical because forage grasses respond to

nitrogen rapidly when they also absorb sufficient phosphorus from the soil. The phosphorus

content of C. ciliaris recorded in the current study across treatments, was lower than the 0.8%

reported by Mutimura and Everson (2012). It was higher than the 0.26% reported by Juma et

al. (2006) for most tropical grasses at their prime harvesting time. This could be because C.

ciliaris has a reputation of being a phosphophilic grass. It efficiently absorbs phosphorus from

the soil (Donaldson and Rootman, 2010). The calcium content of an average of 0.54% of dry

matter was higher than the 0.3% for C. ciliaris reported by Soni et al. (2014). This could have

been due to the liming that could have been done for the previous crop since the sandy soils in

83
the area were generally acidic. The calcium: phosphorus ratio of 1:1 could result in some

mineral imbalances that could lower their bioavailability (Soni et al., 2014). Excessiveness of

one mineral may cause antagonistic effects for other elements and causing mineral imbalances

(Soni et al., 2014).

In the current study, there were higher potassium and sulphur contents caused by

fertilizer treatments than the control. When forage grasses have sufficient N supply, they absorb

more sulphur in order to make sulphur containing amino acids. The N from urea and horse

manure may have caused an increase in sulphur uptake from the soil. As the forage produces

more seed, as was the case with forage grass fertilized with both horse manure and urea, the

demand for potassium increases. This is because potassium is essential for flower and seed

formation and it is also required for the formation of starch.

84
4.7.5 Conclusion

The results of the current study reveal that fertilizing C. ciliaris with either horse manure or

urea cause a significantly higher crude protein content of the grass at seed harvesting stage of

maturity than when no fertilizer is applied. The fact that there was no significant difference

between fertilizing the C. ciliaris with horse manure and fertilizing it with urea in terms of

protein content means that they can be interchangeably used. Cost implications and availability

become the real factors to consider. The results show that the protein content of C. ciliaris after

seed harvesting is way above the maintenance level of livestock. This means that the forage

residues after seed harvesting can actually be used to feed livestock for production. The forage

residues can be baled and be sold by seed producing farmers. This would widen the sources of

income. Seeds can be produced without significantly compromising the quality of feed

produced from a piece of land.

The results of this study also show that the ADF and the NDF contents of the C. ciliaris

forage residues are significantly reduced by application of nitrogen fertilizers. There was a

negative correlation between the crude protein content and the crude fibre of the forage residue.

Fertilization C. ciliaris with nitrogen increases the quality of the forage residue as it becomes

less fibrous and hence easier to digest even at seed harvesting stage. Fertilizing with either

horse manure or urea significantly increased the dry matter digestibility of C. ciliaris forage

residue. Since there was no significant difference between fertilizing C. ciliaris with horse

manure and fertilizing with urea in terms of forage dry matter digestibility, it would be

advisable to use horse manure if it is available as it is applied once in three years (Mujuni and

Sibanda, 2007). This reduces production costs and also limits the effect of excess nutrients on

the surrounding open water bodies as well as underground water.

85
Chapter 5

General conclusions, limitations and future research

From the findings of the current research, it can be concluded that horse manure is a superior

source of nitrogen to urea in terms of seed yield and quality of the forage residue. This was

shown by the horse manure causing significantly higher seed yield, crude protein content and

dry matter digestibility. Urea is superior to horse manure in terms of the forage dry matter

yield. This means that when producing seed, fertilizing with horse manure adds more value

than using urea but when the aim is just to produce high herbage yields, fertilizing with urea

would be a more suitable choice. The limitations to the reliability of the results of this study

include the fact that it has not been replicated both in space and in time. This means that it

cannot represent what is expected from all areas of Botswana. The experiment may need to be

carried out on various soil types, as opposed to on station experiments as was in this study. The

experiment needs to be done across seasons as well. Since the current study was conducted

under irrigation, the results may not be directly applicable to rain-fed conditions, which are

prevalent under normal farming conditions in Botswana. N use efficiency is limited by water

stress, a condition normally encountered under rain-fed situation.

The quality of the forage residue reported in the current study was only from laboratory

analysis. Studies on feeding trials by animals would be necessary to find out if the quality is

mirrored by the animal performance in terms of body weight gain or milk production. This

would help to authenticate the forage residue’s nutritional value to livestock. There is little

published information specifically relating to its management in situations where it is not

wanted. This is pertinent as seeds are bound to be dispersed from seed production fields to

other places. It is not enough to just label it as one of the most notorious weeds without finding

out how best to control its propagation. A similar research can be carried out with varying

amounts of urea. It could be possible that a lower urea application rate could increase seed

86
yield by reducing vegetative growth. The grass is reported to be fairly high on oxalate levels

that may cause poisoning of young sheep and big head condition in horses. A research could

be carried out to find out how the oxalate levels vary with stage of maturity. It may be possible

that the forage residues obtained after harvesting seed could be very safe for feeding the sheep

and horses.

87
REFERENCE

Abedelrahman, F.I. (2007). Effect of seed rate and NPK fertilization on growth and yield of
forage quality of Rhodes grass. MSc Thesis: Faculty of Agriculture, University of Khartoum,
Sudan.

Afzal, J and Ullah, M. A. (2007). Assessment of productive potential and utilization of

rangelands and sown pastures in Pothowar Plateau. Annual Report 2006-07. National
Agriculture Research Centre, Islamabad: pp10

Ahmad, M.S.A., Ashrafin, M. and Ali, Q. (2010). Soil salinity as a selection pressure and a key
determinant for the evaluation of salt tolerance in blue panigrass. Journal of Flora 205:37-45:

Akiyana, Y., Hanna, W. and Ozias-Akins, P. (2005). High-resolution physical mapping

revealing the apospory-specific genomic regions of Cenchrus ciliaris. Theoretical and Applied
Genetics 111: 1042-1051

Albu, .A. (2012). Calcium and Phosphorus concentration in dairy cow feed. Lucrari Stiintifice-
Seria Zootehnie 57: 70-74

Al-dakheel, A.J., Hussain, M. I. and .Rahman, A.O.M. (2015). Impact of irrigation water
salinity on Agronomical and quality attributes of Cenchrus ciliaris. Journal of Agricultural
Water Management 159: 148-154

Ankom Technology Corporation. Method for determining acid detergent fibre, neutral
detergent fibre and crude fibre, using the Ankom Fibre Analyser. Ankom Technology
Corporation, 14Turk Hill Park, Fairprt New York 14450, USA

88
AOAC. (2009). Official Methods of Analysis (16th) edition. Association of Official Analytical
Chemists: Washington D.C, USA

AOAC. (1996). Official Methods of Analysis (16th) edition. Association of Official Analytical
Chemists, Arlington, VA

APRU. (1985). Livestock and Range Research in Botswana. Ministry of Agriculture,

Gaborone.

Arshadullah, M., Malik, M.A., Rasheed, M., Jilani, G., Zahoor, F. and Kaleem, S. (2011).
Seasonal and genotypic variations influence on biomass and nutritional ingredients of
Cenchrus ciliaris grass forage. International Journal of Agricultural Biology 13:120-124

Ashraf, M.A., Mahmood, K., Yusoff, I and Qureshi, A.K. (2013). Chemical constituents of
Cenchrus ciliaris. L. from Cholistan desert, Pakistan. Archives of Biological Sciences 65 (4):
1473-1478

Bhattarai, S.P., Fox, J., Gyasi-Agyei, Y. (2008). Enhancing buffel grass seed germination by
acid treatment for rapid vegetation establishment on railway batters. Journal of Arid
Environments 72:255-262

Brady. N.C. and Weil, R.R. (2002). Soil colloids: Seat of soil chemical and physical activity.
In the nature of properties of soil, third edition. Prince Hall, N.J, USA

Brummer, J. E. (2009). Does it pay to fertilize grass hayfields with nitrogen? Colorado State
University, Department of Soil and Crop Sciences, 1170 Fort Collins, Co. USA

89
Bulle, H., Mamo, M. and Geikuku, P. (2011). Comparative evaluation of Chloris gayana,
Eragrostis superba and Cenchrus ciliaris for pasture production in Marsabit Central District,
Northern Kenya. Proceedings of the 12th KARI Biennial Scientific Conference, 8th-12th
November, Nairobi, Kenya, pp826-830

Burgess, J. (2006). Country pasture/forage profile;

Botswana.www.org/ag/FAOINFO/AGRICULT/AGP/AGPC/doc/counprof/Botswana/botswana. htm
(accessed;10/06/17).

Burhan, H. and Hago, T. E. (2000). Principle of crop production. University of Khartoum

Printing Press, Sudan

Cubitt, T. (2010). The horse’ digestive system. Hygain Services

Donaldson, C.H. and Rootman, G.T. (2010). Effects of nitrogen level and cutting frequency on
digestibility and voluntary intake. Journal of Proceedings of the Annual Congresses of the
Grassland Society of Southern Africa 12 (1):91-93

Dutta, K. S. (2002). Goat Digestive System. Animal Feed Production Line, Lochampchina
International.

During, G. and McNaught, K.J. (2012). Effects of cow urine on growth of pasture and uptake
of nutrients. New Zealand Journal of Agricultural Research 4 (6) 591-605

Eriksson, O. and Hennessy, J. (2015). Energy and nutrients from horse manure: Life cycle data
inventory of horse manure management systems in Gavleborg. Gavle University Press,
Sweden.

90
Gutierrez-Ozuna, R.,Eguiarte, L.E. and Molina-Freaner, F. (2009). Genotypic diversity among
pasture and roadside populations of the invasive buffelgrass in north-western Mexico. Journal
of Arid Environment 73: 26-32

Hadin, A., Hillman, K. and Eriksson, O. (2017). Prospects for increased energy recovery from
horse manure- A case study of management practices, environmental impact and costs.
University of Gavle, Sweden.

Hare, M.D., Tatsapong, P. and Phengphet, S. (2009). Herbage yield and quality of Brachiaria
cultivars, Paspalumatratum and Panicum maximum in North-East Thailand. Journal of
Tropical Grasses 43: 65-70

Hassan, A., Zewdu, T., Urge, M. and Fikru, S. (2015). Effect of nitrogen application fertilizer
on in sacco rumen digestibility and in vitro dry matter digestibility of Cenchrus ciliaris and
Panicum maximum grown under irrigation. Global Journal of Medical Research: Nutrition and
Food Science 15 (1): 2249-4618

Ihsan, A.A., Aran, A., Hamad, O. and Awaga, A. (2014). Yield and quality of Cenchrus ciliaris
affected by nitrogen and phosphorus fertilizers. Journal of Food, Agriculture and Environment
12 (4): 139-142

James, D.W. (2010). Urea: A low cost nitrogen fertilizer with special management
requirements. Utah State University Cooperative Extension, USA.

Jones, S. (1985). Digestive system of the horse and feeding management. University of
Arkansas Cooperative Extension Service, USA, FSA 3038

Joshua, W.D. (1988). Methods for the measurements of physical properties of soils. Project
AG/BOT/85/011. Field Document 19
91
Keskinen, R., Nikama, J., Sarkijarvi, S., Myllymaki, M., Salo, T and Uusi-Kamppa, J. (2017).
Recycling nutrients from horse manure: effects of bedding type and its compostibility. Journal
of Agricultural and Food Science 26: 68-79

Kizima, J.B., Mtengeti, E. J. and Nchimbi-Msolla, S.N. (2012). Seed yield and vegetation
characteristics of Cenchrus ciliaris as influenced by fertilizer levels, row spacing, cutting
height and season. Journal of Forage Sciences 73 (2): 445-461

Koech, O. K., Kinuthia, R. N., Mureithi, S. M., Karuku, G.N. and Wanjogu, R.N. (2014).
Effects of varied soil content on seed yield of six range grasses in the rangelands of Kenya.
Universal of Agricultural Research 2 (5): 174-179

Kolawole,O. G (2013). Soil Science, indigenous knowledge and sustainable intensification:

Implications for smallholder farming systems. Okavango Research Institute, University of
Botswana. P. Bag 285 Maun

Kumar, A., Tripathi, H. and Yadav, D.S. (2007). Correcting nutrients for sustainable crop
production. Indian Journal of fertilizers 2 (11): 37-44

Kumar, D., Dwivedi, G.K. and Singh, S.N. (2005). Seed yield and quality of buffel grass
(Cenchrus ciliaris) as influenced by row spacing and fertilizer level. Journal of Tropical
Grasslands 39: 107-111

Kungwang, T., Worapong, S., Sereethai, S. and Jutamal, S. (2010). Energy and protein
requirement of yearling Thai native cattle. KHON KAEN Agriculture Journal Supplement:
144-152

92
Laidlaw, A.S. (2005). The relationship between tiller appearance in spring and contribution of
dry matter yield in perennial rye-grass cultivars differing in heading date. Grass and Forage
Science 60: 200-209

Linn, J.G. and Martin, N.P. (1999). Forage quality tests and interpretations. MAG (Ministrio
de Agricultura Ganaderia, 2004). Officina National de Semillas. San Jose, Costa Rica.

Lu, C.D., Kawas, J.R. and Mahgoub, O. G. (2005). Fibre digestion and utilization in goats.
Small Ruminant Research 1-2: 45-52

Madibela, O.R, Boitumelo, W.S, Manthe, C, Raditedu, I (2002). Chemical composition

and in vitro dry matter digestibility of local landraces of sweet sorghum in Botswana.
Journal of Livestock Research for Rural Development 14 (4)

Madibela, O. R and Segwagwe, B.V.E (2008). Nutritional effects of supplementary

feeding of maternal metabolites, cortisol, thyroid hormone levels and on outcome of
pregnancy of dry season kidding Tswana goats. Journal of Livestock Research for
Rural Development 20 (4)

Marinari, S., Masciandaro, G., Ceccanti, B. and Grego, S. (2000). Influence of organic and
mineral fertilizers on soil biological and physical properties. Bioresource Technology 72 (1):
318-323

Marschner, H. (1995). Mineral nutrition of higher plants. Academic Press, New York

Mcdowell, L.R. (2003). Mineral, Animal and Human Nutrition 2nd edition. Elsevier,The
Netherlands: 256-315

Marshall, V.M., Lewis, M.M. and Ostendorp, B. (2012). Buffel grass as an invader and threat
to biodiversity in arid environments: A review. Journal of Arid Environments 78: 1-12
93
Martin, R.J.R., Cox, F., Ibarra-F, D.G., Alston, R.E. and Banner, J.C. (1999). Spittlebug and
buffelgrass responses to summer fires in Mexico. Journal of Range Management 52: 621-625

McDonald, C.J. and McPherson, G.R. (2011). Fire behaviour characteristics of Buffel grass-
fuelled fires and native plant community composition in invaded patches. Journal of Arid
Environments 75: 1147-1154

Meena, L.R., Mann, J.S. and Chand, R. (2011). Effects of inter-cropping, row ratios and
organic manures on productivity and quality of forage under semi-arid regions of Rajasthan.
Annals of Arid Zone 50 (1): 41-45

Mganga, K.Z. (2009). Impact of grass reseeding technology on rehabilitation of degraded

rangelands. A case study of Kibwezi district, Kenya. MSc Thesis, University of Nairobi,
Kenya.

Monroe, R. (1996). Broiler litter production in Kentucky and its potential use as a nutrient
resource. Cooperative Extension services, University of Kentucky Publication AGR-168.
Lexington KY

Mujuni, G. and Sibanda, C. (2007). BGCSE Agriculture: A student companion. Pula Press

Mudenda, V. and Maeresera, T. (2009). Agriculture in Action Book 1. Longman Botswana

Mutimura, M. and Everson, T.M. (2012). On-farm evaluation of improved Brachiaria grasses
in low rainfall and aluminium toxicity prone areas of Rwanda. International Journal of
Biodiversity and Conservation 4 (3): 137-150

94
Nsinamwa, M., Moleele, N.M. and Sebego, R.J. (2005). Vegetation patterns and nutrients in
relation to grazing pressure and soils in the sandveld and hardveld communal grazing areas of
Botswana. African Journal of Range and Forage Science 22 (1): 17-26

NRC. (1989). Nutrient Requirements of Dairy Cattle. National Academy Press. Washington
D.C.

Ogillo, B.P. (2010). Evaluation of range grasses under different micro catchments and financial
returns from reseeding in Southern Kenya. Unpublished MSc Thesis, University of Nairobi,
Kenya.

Ogren, G. (2013). Phosphorus to horses and cows. Licenciate Thesis, Swedish University of
Agricultural Sciences.

Ogren, G., Holtenius, K. and Jansson, A. (2014). Phosphorus balance and faecal losses in
growing standard bred horses in training fed forage only diets. Journal of Animal Sciences 91:
2749-2755

Osman, A. E., Makawi, M. and Ahmed, R. (2008). Potential of indigenous desert grasses of
the Arabian Peninsula for forage production in a water scarce region. Grass and Forage
Sciences 63: 495-503.

Parwani, R. (2013). Environmental impact on biomass and allelopathic effect of Cenchrus

ciliaris. European Academic research 1 (3): 275-285

Parwani, R. and Mankad, A. (2012). Seasonal impact on Biomass production of Cenchrus

ciliaris grass. Government Science College, Gandhinagar Gujarat University, Ahmedabad,
India

95
Ramirez, R.G., Gonzalez- Rodriguez, H., Morales-Rodriguez, R., Cerrilo-Soto, A., Juarez-
Reyez, A., Garcia-Dessomes, G.J. and Guerrero-Cervantes, M. (2009). Chemical composition
and dry matter digestibility of some native and cultivated grasses in Mexico. Czech Journal of
Animal Science 54 (4): 150-162

Ros Barcelo, A. (1997). Lignification in plant cell walls. International Review of Cytology
176: 87-132

Redfearn, D., Arnall, B., Zhang, H. and Rice, C. (1990). Fertilizing Bermuda grass Hay and
Pasture. Oklahoma Cooperative Extension service P55-2263

Sahoo, A., Sankhyan, S. K. and Sharma, S. C. (2015). Pasture in small ruminant production:
Issue and solutions. 1 CAR-Central Sheep and Wool Research Institute, Avikanagar-304 501,
India.

Savoy, H. (1996). Nitrogen Sources for Pasture and Hay. University of Tennessee. Agricultural
Extension service Publication 163. Knoxville,TN

Seddi, K.M., Visser, M., Neffati, M., Reheul, D. and Chaieb, M. (2002). Seed and spike traits
from remnant populations of Cenchrus ciliaris in South Tunisia: high distinctiveness, no
ecotypes. Journal of Arid environments 50:309-324

Tacheba, G. and Moyo, A. (1985). A Review of Range Pasture Plants in Botswana. Ministry
of Agriculture, Gaborone, Botswana

Tailor, G., McDonald, A.J.S., Stadenberg, I. and Smith, P.H. (1993). Nitrate supply and the
biophysics of leaf growth in Salix viminalis. Journal of Experimental Botany 44: 155-165

96
Taylor, D.J., Green, N.P.O. and Soper, R. (1997). Biological Science. Cambridge University
Press, UK

Topps, J.H. (1996). Assessment of forage legumes as protein-rich supplement in ruminant

production systems in Zimbabwe. In: Ndikumana and de Leeuw (Eds). Sustainable feed
production and utilization for small holder livestock enterprises in Sub-Saharan Africa.
Proceedings of the second African Feed Resources Network (AFRNET). 6-10th December
1993, Harare, Zimbabwe, pp69-72

Trottier, N.L., Bott, R.C., Woodward, A., Green, E.A., Williams, C.A., Westendorf, M.L.,
Swinker, A.M., Mastellar, S.L. and Martinson, K. (2016). Gastro-intestinal nitrogen
metabolism of equids and impact on protein requirement. Journal of Equine Veterinary
Sciences 45: 79-86

Van Soest, P.J. (1994). Nutritional Ecology of Ruminants, 2nd ed. University Press. Ithaca, NY

Van Waveren, E. J. (1988). Botswana Soil Database User Manual. FAO/UNDP/Government

of Botswana Project BOT/85/011

Walker, K. P. (2013). Performance of Dorper lambs fed buffel grass and browse as basal diet
and sorghum bran as supplement. International Journal of Plant and Animal Sciences 1 (2):
037-041

Yossin, A.M. and Ibrahim, Y.M. (2013). Effects of fertilizers (urea, farmyard and chicken
manure) on growth and yield of Rhodes grass. Universal Journal of Plant Science 1 (3): 85-90

97
APPENDICES

Agronomic Parameters

Sample Seed yield Number of Inflorescence Forage dry Fresh mass

(kg/ha) tillers length (cm) matter yield yield (kg/ha)
/crown (kg/ha)
C1 10.25 29 11.6 1525.92 4833.32
C2 12.18 19 10.98 1748.33 4583.51
C3 11.73 23 11.2 1498.76 4166.25
C4 10.87 25 11.5 1387.39 3333.76
C5 12.17 20 10.66 1581.84 3750.43
C6 9.22 26 9.8 2024.67 5409.57
Mean 11.07 23.67 10.96 1627.81 4346.14

H1 16.33 32 10.7 2409.44 6111.11

H2 18.95 28 11.6 2265.41 5000.12
H3 17.55 27 11.2 2123.67 4986.76
H4 18.12 31 11.7 2256.64 5334.85
H5 20.22 30 10.8 2081.52 5416.67
H6 19.56 32 11.4 2290.43 5835.13
Mean 18.46 30 11.2 2237.85 5447.44

U1 16.27 28 11.2 2918.11 6250.06

U2 13.46 37 10.2 2940.25 6283.33
U3 16.22 32 11.6 2415.81 5833.82
U4 14.12 34 11.2 3081.67 7500.34
98
 U5 11.36 29 10.6 2583.52 6666.73
U6 15.62 30 11.3 2498.33 5856.92
Mean 14.51 31.67 11 2739.59 6398.53

Effects of fertilizer treatments on agronomic parameters of C. ciliaris.

Data analysis results
Parameter General F value Pr > F
Treatment Means ± s.e
Mean
Number of Control* 23.67 ± 3.78a
tillers per
Horse
crown 30.00 ± 2.10b 28.44 10.64 0.0013
manure
Urea 31.67 ± 3.39b
Seed yield Control 11.07 ± 1.18a
(kg/ha)
Horse
18.46 ± 1.42b 14.68 34.65 <0.0001
manure

Urea 14.51 ± 1.92c

Fresh biomass Control 4346.14 ± 753.22a

yield (kg/ha)
Horse 5397.37 16.41 0.0002
5447.44 ± 450.93b
manure
Urea 6398.53 ± 621.85c
Forage Dry Control 1627.82 ± 227.58a
Matter Yield
(kg/ha) Horse 2201.76 39.50 <0.0001
2237.85 ± 118.99b
manure
Urea 2739.62 ± 274.42c

Length of
Control 10.96 ± 0.66a
inflorescence
(cm)
Horse
11.23 ± 0.41a
manure
11.07 0.44 0.65

Urea 11.02 ± 0.52a

99
ab
Means in the same column within a parameter with different superscripts differ significantly;
P <0.05; means ± s.e;*No fertilizer.

Chemical Composition and Dry Matter Digestibility Raw data

Sample ADF (%) NDF (%) Ash (%) In vitro Dry Protein (%)
Matter
Digestibility
(%)
C1 46.2 35.43 12.67 34.46 10.4
C2 43.4 37.9 12.54 37.4 9.46
C3 43.63 38.35 15.17 36.6 10.61
C4 43.2 38.7 12.13 31.94 9.82
C5 43.51 36.03 12.44 33 10.15
C6 43 38.3 12.18 29 10.21
Mean 43.82 37.45 12.86 33.73 10.11

H1 37.13 37.35 11.64 41.4 12.71

H2 37.6 32.9 11.49 41 12.38
H3 47.5 38.1 11.88 43 12.32
H4 38.52 32.9 11.4 45.4 13.14
H5 37.4 33.1 12.09 45.8 12.58
H6 39 33.23 11.93 48.21 12.42
Mean 39.53 34.6 11.74 44.14 12.59

U1 40.2 36.1 12.09 37 12.44

U2 38.72 34.89 11.98 40.2 12.48
U3 39.08 36.23 12.19 38.54 13.12

100
 U4 40.6 35.9 12.05 35.8 12.2
U5 39.92 36.43 12.59 38.2 12.02
U6 39.17 36.1 12.39 37.1 12.34
Mean 39.62 35.94 12.22 37.81 12.43

Data analysis results for chemical composition and digestibility

Parameter General F value Pr > F
Treatment %DM Means ± s.e
Mean

10.11±0.41a
Control*
Crude Protein Horse 11.71 86.00 <0.0001
12.59±0.30b
manure
Urea 12.43±0.38b
Control 43.82±1.19a
Horse
ADF 39.53±3.97b 40.99 6.13 0.0113
manure
a
Urea 39.62±0.73b
Control 37.45±1.37a
Horse
NDF 34.60±2.44b 36.00 4.52 0.0290
manure
Urea 35.94±0.23b

Control 12.86±1.15a

Horse
Ash 11.74±0.27b 12.27 3.88 0.0438
manure

Urea 12.22±0.23b

Control 33.73±3.11a
Horse
44.14±2.81b
IVDMD manure 38.56 24.84 <0.0001

Urea 37.81±1.52b

101
ab
Means in the same column within a parameter with different superscripts differ significantly;
Pr<0.05; means ± s.e;*No fertilizer.

Mineral Composition

Sample Phosphorus Magnesium Calcium Potassium Sodium Copper Zinc Iron Sulphur

C1 0.46 0.38 0.55 1.53 0.33 0.97 1.54 0.96 0.24

C2 0.51 0.42 0.43 1.57 0.37 0.92 1.43 1.25 0.22

H1 0.55 0.45 0.6 1.81 0.31 0.95 1.33 1.36 0.41

H2 0.49 0.41 0.52 1.62 0.42 0.99 1.39 1.23 0.38

U1 0.53 0.53 0.55 1.92 0.39 0.95 1.44 1.41 0.46

U2 0.57 0.38 0.58 1.86 0.34 0.87 1.37 1.22 0.41

Data analysis results for minerals

Mineral content (mg/kg)

Treatment P Mg Ca K Na Cu Zn Fe S

No fertilizer 0.485 0.400 0.490 1.550 0.350 0.945 1.485 1.105 0.230
(control)

Horse manure 0.520 0.430 0.560 1.715 0.365 0.970 1.360 1.295 0.395

Urea 0.550 0.455 0.565 1.890 0.365 0.910 1.405 1.315 0.435

General mean 0.518 0.428 0.538 1.718 0.360 0.942 1.417 1.238 0.3533

P-value 0.3287 0.7277 0.4726 0.0590 0.9469 0.4543 0.2446 0.4197 0.0076

102
103