EVALUATION OF TWO NAPIER GRASS CULT IVARS

(PENN 1SETUM PURPUREUM) UNDER IRRIGATION

TbOf

AT DIFFERENT STAGES OF GROWTH 9

BSc. (AGRIC.) (NAIROBI)

A THESIS SUBMITTED IN PARTIAL FULFILMENT

FOR THE DEGREE OF MASTER OF SCIENCE IN ANIMAL
PRODUCTION (NUTRITION) IN THE UNIVERSITY OF NAIROBI.

1989
 DECLARATION

This is my original work and has not been presented for a

degree in any other university.

This thesis has, been submitted for examination with

our approval as University Supervisors.

1. Si gned

2. Signed

DR. AGGREY ABATE

 ii

DECLARATI ON

This is my original work and has not been presented for a

degree in any other university.

This thesis has. been submitted for examination with

our approval as University Supervisors.
 ACKNOWLEDGMENT

I am '.very grateful to my University

Supervisors Dr.. M.M. Uanyoike and Dr. A. Abate who
continually encouraged me and gave useful guidance
and constructive criticism throughout the research
project. Thanks are also due to Drs. Wahome,
Gachuiri and Baptist for their statistical advice.
I am indebted to S.P. Ng’ang’a and the entire
laboratory staff at the Department of Animal
Production, University of Nairobi for their
technical assistance.

I gratefully acknowledge the financial

assistance by the Kenya Agricultural Research
Institute (K.A.R.I.) which enabled me to complete
my studies successfully. I express my appreciation
to the Director, National Animal Husbandry
Research Station, Naivasha, for allowing me to use
the facilities of the Dairy Cattle Research
Section at the station. Sincere thanks are to John
Kutwa and Simon Otero, dear Christian friends
whose fellowship was very spiritually enriching.
Their care and concern made my stay in Naivasha
very comfortable. I am indebted to the National
Dairy Development Project (NDDP) staff, Naivasha,
and in particular Mr. B. Uouters and Mr. Van V a 1k
for their invaluable help both materially and for
making project facilities available to me. Thanks
also to NDDP technical assistants, Mr. Ayako and
Mr. Wekesa for supervising the maintenance of the
 ( iv)
experimental field. Sincere gratitude to Mr. P.
Njoroge and the entire staff of the nutrition
laboratory at N.A.H.R.S., Naivasha, for their
assistance in chemical analysis.

Last but not least I am grateful to Mr. John

Gitau for typing this thesis and to all others who
are too numerous to mention but in one way or
another contributed to the successful completion
of this study.
 TABLE OF CONTENTS

Page

Declaration ------------------------------ (ii)

Acknowledgment (iii)

Table of Contents -------------------------------- <v>

List of Tables ------------------------------ (vii)
List of Figures ------------------------------ (ix)
List of Appendices ------------------------------ <* >

ABSTRACT

1.0 INTRODUCTION -------------------------------- 1

2.0 LITERATURE REVIEW ----------------------------- 4

2.1 Origin and botanicalcharacteristics ------ 4

2.2 Cultivars 5
2.3 Ecological requirements and cultural
practices 7
2.4 Dry matter yields 10
2.5 Nutritive quality 14
2.5.1 Chemical composition ---- 14
2.5.2 Digestibility ---- 20

3.0 MATERIALS AND METHODS ----------------------- 24

3.1 Study sites 24

3.2 Site description 24
 vi

3.3 Experimental Field 28

3.4 Experimental Designand Treatments ------- 28
3.5 Harvesting and Sampling Procedures ---- 29
3.6 Analytical Procedures 30
3.6.1 Chemical Composition 30
3.6.1 Digestibility Estimates 31
3.6.2 Ruminal Dry Matter Degradability
in sacco ----- 31
3.7 Statistical Analysis 35

4.0 RESULTS AND DISCUSSION ----------------------- 36

4.1 Dry matter yields, grass height at

cutting and leaf:stem ratio ----- 36
4.1.1 Dry matter yield 36
4.1.2 Grass height atcutting 38
4.1.3 Leaf:stem ratio 39
4.2 Chemical Composition 41
4.2.1 Proximate composition 41
4.2.2 Detergent Fibres andLignin 50
4.3 In vitro Digestibility
Coefficients and Silica content ---- 52
4.4 Relationships between chemical
components and digestibility ---- 56
4.5 Dry matter degradability in sacco ----- 60

5.0 CONCLUSION 71
5.1 SCOPE FOR FURTHER WORK ---------------------- 73
6.0 REFERENCES 74
7. 0 APPENDICES 95
 vi i

List of Tables
Table Page

2.1 Average annual yields of dry matter over

a three year period in Kenya ---- 11

3.1 Changes in minimum, maximum and mean

temperature during experimental period ---- 27

4.1 Influence of cutting stage on DM yield,

grass height and leaf:stem ratio ---- 37

4.2 Influence of cutting stage on % DM content,

CP and CF 43

4.3 Influence of cutting stage on EE, NFE

and total ash 44

4.4 Influence of cutting stage on potassium,

calcium and phosphoruscontents 47

4.5 Influence of cutting stage on detergent

fibres and lignin contents 51

4.6 Influence of cutting stage on silica

content and in vitro digestibility ---- 53

4.7 Simple correlations between chemical constituents

and in vitro digestibility at different ---- 57
stages of growth
 vi i i

A.8a Relationship between fibres, lignin,

silica and in vitro digestibility
(Bana grass) 58

A.8b Relationship between fibres, lignin,

silica and in vitro digestibility
(French Cameroon) 59

A.9 DM Degradability constants used to

fit in the degradation equation 6A

A. 10 The variation in half life with changes

in stage of growth 66

A. 11 Relationship between half-life, fibres,

lignin and silica 67
 ix

List of Figures

Figure Page

3. 1 Moisture received by the Napier grass

(Rainfall + Irrigation) 26

3.2 Illustration of degradability curve

components (Sketch) 34

4. 1 Ruminal * DM loss in sacco using sheep

(Bana grass) 62

4.2 Ruminal % DM loss in sacco using sheep

(French Cameroon) 63

4.3 Relationship between half- life and

in vitro dry matter digestibility 69
 X

List of Appendices

Append ix Page

I Analyses of variance tables ----- 95

II Analysis of variance Tables for

ruminal nylon DM loss ---- 98

III Proximate composition of basal

diet of sheep on nylon bag
trial ----- 99
IV Mean values of ruminal * DM
loss of napiergrass incubated
in sheep
(a) Bana grass 100

(b) French Cameroon 100

 Two commonly used napier grass (Penniset urn
purpureum) cultivars, Bana grass and French
Cameroon were grown under irrigation to assess
the changes in nutritive value at different
stages of growth in Naivasha, Kenya. The
experimental design was a split-plot with cutting
stage (subplot) nested within variety (main plot).
There were three replications per treatment. The
response variables were evaluated at 4, 6, 8, 10,
12 and 14 weeks of age.

Dry matter yields and grass height at cutting

ranged from 1.24 to 10.01 tons/ha and 22.11 to
189.99 cm respectively and significantly (PC0.05)
increased with age. The differences between the
cultivars for the two parameters were not
significant (P>0.05). The leafrstem ratio
declined (P<0.05) with age but the differences
between the cultivars were insignificant (P>0.05).
The variations for leaf:stem ratio ranged from
5.03 at 4 weeks to 0.53 at 14 weeks of age.

Considerable changes (PC0.05) were observed in

CP, ash and carbohydrate contents as the
napiergrass matured. Crude protein decreased
(P<0.05) from 14.84 to 3.98* while CF and NFE
ranged from 23.39 to 39.01* and 33.74 to 39.86*
respectively. Ash and EE contents decreased
(P<0.05) from 26.41 to 15.75* and 3.42 to 1.49 at
 xi i

4 and 14 weeks respectively. Ash, crude fibre and

NFE increased (PC0.05) as the grass matured but
the variation between the cultivars was
insignificant (P>0.05). The dry matter percent
varied from 15.81 to 17.70% and significantly
(P<0.05) increased with age. Although French
Cameroon had consistently lower DM values,
varietal differences were insignificant (P>0.05).
Calcium and phosphorus levels decreased from 0.39
to 0.30% and 0.31 to 0.21% at 4 and 14 weeks of
age respectively. The effect of both age and
variety for these two components were however
statistically insignificant (P>0.05). Potassium
levels were high and ranged from 11.70 to 4.21%.
Although these levels decreased significantly
<P<0.05> with age, varietal differences were
insignificant (P>0.05).
The detergent fibres NDF and ADF increased
(P<0.05) with age and rose from 54.44 to 72.68%
and 25.97 to 42.00% respectively. Lignin (ADL)
increased (P<0.05) with age and ranged from 2.26
to 5.77 from 4 to 14 weeks. Varietal differences
in these constituents were not evident Silica
levels increased (PC0.05) with age from 4 to 14
weeks ranging from 4.07 to 6.67% respectively with
Bana grass variety showing correspondingly higher
<P<0.05 ) va1ues.

Dry matter and organic matter digestibility

in vitro decreased (P<0.05) with age from 72.04 to
 xiii

58.30% and 75.72 to 62.22% respectively but the

varietal differences were insignificant (P>0.05).
The nylon bag dry matter loss in sacco increased
(P<0.05) with increase in incubation time and was
higher (PC0.05) the younger the napier grass. The
ranges were from 71.09 to 27.57% at 72 and 12
hours of incubation respectively.

Significant (P<0.05) differences were observed

in grass height, ash and percent dry matter with
changes at cutting age. Silica content increased
<P<0.05) with increased maturity and differences
between the varieties were significant (PC0.05).
However, it was concluded under the conditions of
the experiment that dry matter yields, in vitro
and in sacco DM digestibility, CP, CF, EE, NFE,
detergent fibres, lignin and the minerals studied
varied with the stage of cutting although varietal
differences were insignificant (P>0.05).
 1

1 .0 INTRODUCTION

Dairy farming is an important enterprise in

Kenya and constitutes about 11% of the nation’s
cattle herd (Anon., 1981). Due to rise in human
population, farm sizes have decreased in dairying
areas with less land being allocated for pasture
and more for food crops (Jaetzold, 1983). Needless
to say, an increase in human population will mean
an increased, demand for milk. The dairy cattle,
therefore, must be adequately fed in terms of both
quantity and quality to meet the rising milk needs
and one of the options open is the use of high
yielding fodders.

Napier grass (Fennisetum purpureumJ is

already established as a popular fodder plant for
the small-scale farmers in Kenya and is the most
widely grown '(C h u 1d 1eigh, 1974; Goldson, 1976;
Potter, 1985). The main reason for its adoption is
its high yields, ease of harvesting and the fact
that it retains good quality longer (Mwakha, 1972;
Odhiambo, 1974; Thomas, 1976; Van Gastel, 1978;
Boonman, 1979 ; Karanja, 1981).

Although thirty to forty types of napier

grass have been tested in Kenya (Anon., 1985a)
studies have been restricted to a few cultivars,
Bana grass and French Cameroon being the two major
ones. Over the last 15 years, however, seemingly
contradictory data as to the superiority of Bana
grass over French Cameroon and the converse have
 2

been reported. This can be attributed partly to

inadequate information on the two varieties as
regards their nutritive value and utilization
(Anindo and Potter, 1985). In practice, farmers
feed the grass to animals at various stages of
regrowth from one month to well over three months
(Wouters, 1985 ) depending on animal needs and
availability of other feed resources with little
consideration being paid to nutritive quality.
In addition there is insufficient information
available regarding the best stage to harvest for
optimum nutritive value and yield although
maturity stage is known to affect both yield and
chemical composition. Since the nutritive value of
the fodder is greatly inf 1uenced by its stage of

growth, herbage of d ifferent ages can be

cons idered as different feed£. Therefore, to

maximise the milk output from the dairy cow it is
desirable to feed the forage at the stage of
growth when both quality and quantity are optimum.
The published work, from Kenya, relating the stage
of harvest to chemical composition and
digestibility both in vivo and in vitro is scanty.

To develop more information on the nutritive

value of the two varieties and allow more
effective utilization of napier grass this study
was initiated with the following main objectives:-

1. To determine the effect of cutting

 3

stage (regrowth stage) on chemical

composition, dry matter yield, leaf:stem
ratio and grass height at harvesting.

2. To study the in vitro digestibility and

ruminal dry matter disappearance in
sacco of the two varieties.
 4

2: . o I TERATURE REV I EW

2. 1 Origin and botanical characteristics:

Napier grass (Fennisetum purpureum) also

called elephant grass is a native of tropical
Africa (Bogdan, 1977; Karanja, 1964; Palacpac,
1985). It is widely grown in the rest of tropical
world between 10°N and 20° S and to a lesser
extent in the temperate areas. Original
identification was done by Schumacher in the late
18th century as reported by Paterson (1933). It
was first cultivated in Zimbabwe in 1890 and
recommended as a fodder by the then Rhodesian
Department of Agriculture in 1908 (Paterson,
1933).

Elephant grass is a robust, erect-stemmed

perennial growing to a height of 2-6m (Bogdan,
1977; Gohl, 1981; Vancoppeno1, 1983 land Palacpac,
1985) with upto 20 nodes per stem (Goldson, 1977).
The grass grows in clumps spreading by stem bases
and rooting from nodes or by short rhizomes. The
leaves range from 30-120cm in length, l-5cm in
width and are glabrous or hairy. Mature
napier grass forms large bamboo-like clumps.. The
panicle is only found at the end of the main stem
and the branches. Flowers are normally narrow and
13-25cm long. Although most cultivars are capable
of reproducing sexually, some cultivars are known
to be apomictic (Bogdan, 1977). The grain is 2mm
 5

long (Bogdan, 1977 and Palacpac, 1985).

2.2. C u 1t iva rs

A large number of cultivars have been tested

over a wide range of environmental conditions in
Kenya. Most of these have been developed by the
selection and propagation of local materials
with desirable agronomic and nutritive value
characteristics. There are differences in
thickness of stems, size of leaves, hairiness of
stems and leaf sheaths, general vigour, the size
of tufts, the number of tillers and height of
plant (Bogdan, 1977). These characteristics make
elephant grass varieties easy to recognize and
identify because they consist of one genotype only
(Boonman, 1979).

In this country the most commonly grown

cultivars of n'apier grass are Bana grass, French
Cameroon and Uganda hairless (Uouters, 1986). Bana
grass and French Cameroon predominate and have
been studied most (Thairu and Tessema , 1987 )
although the information on chemical composition
is scanty (Abate et a], 1987). Most of the studies
have been on agronomic attributes rather than on
nutritive value (Potter, 1985). Two other cross­
bred types of napier grass namely, Pakistan napier
and Clone 13, have gained prominence at one time
or another. Pakistan napier (Bajra), a cross
between napier grass and Bulrush millet
 6

(P.purpureum and F. typhoides) has drought

tolerance qualities but its adoption as a fodder
has been hampered by its low dry matter yields,
low leaf :stem ratio and a very fast growth rate
which renders it stemmy too soon (Anon., 1985b;
Wouters, 1985). Clone 13 was bred from French
Cameroon and Bajra at Kitale, Kenya. Although this
cultivar is resistant to the fungal disease, snow
mould (Beniowskia sphaeroides), it has not been
popular due to its low dry matter yields, poor
establishment and reduced tolerance to drought
conditions (Kusewa et al, 1983; Anon.,1985b).

Until recently French Cameroon was known as

the " standard variety " (Goldson, 1977; Boonman,
1979). Among the reasons advanced for its
popularity are; highest total yields, highest dry
season yields, reduced hairiness, high clump
density and relatively higher palatability
compared to the other cultivars (Strange, 1959;
Boonman, 1979). The cultivar has numerous
relatively thin stems and narrow leaves (Bogdan,
1977). The leaves are almost hairless except on
the leaf sheaths (Wouters, 1986). However, the
cultivar is susceptible to snow mould (Beniowskia
sphaeroides ) particularly during the wet season
(Boonman, 1979).

Bana grass has variously been referred to as

Gold Coast, Potha Bana, and Ghana (Bogdan, 1977;
Boonman, 1979 ; Wouters, 1985). Although Bogdan
(1977) reported Bana grass as a hybrid between
 7

Bulrush millet (Penniseturn typhoides) and napier

grass (Pennisetum purpureum), tests done at
Kitale, Kenya, demonstrated that Bana grass is a
cultivar of napiergrass (Boonman, 1979). The grass
has thick stems and both the leaf sheaths and
leaves are covered with stiff hairs. The stems are
relatively soft and some investigators have
reported that it has higher dry matter intake
compared to other napier grass varieties (Anon.,
1985b). Bana grass has been valued for its high
herbage yields, competitive vigour, persistence,
palatability and good herbage quality
(Anon.,1985a).

Bana grass and French Cameroon outyield

Eajra, Uganda hairless and Clone 13 and hence
their popularity over the latter (Boonman, 1979 ;
Anon., 1985b ; Wouters , 1986).

2.3. Ecological requirements and cultural

practices:

Napier grass can be planted in virtually all

types of well drained soils but for high
production, fertile soils are essential (Bogdan,
1977 ; Palacpac, 1985). It is an ideal crop for
unirrigated lands and rolling hills (Palacpac,
1985). It grows well in a wide range of climatic
conditions from sea level to well over 2000m
altitude the upper limit being determined by its
temperature requirements. Maximum growth is
attained at between 30 and 35^C and is greatly
 8

reduced at the range of 10 and 15^C (Cooper and

Tainton, 1968). The plant ceases to grow at
0
temperatures below 10 C (Bogdan , 1977). Elephant
grass Is more drought tolerant compared to other
common pasture grasses (Bogdan, 1977).

Napier grass is mainly cross-pollinated and

the resulting seeds of low genetic stability are
few and give small and weak seedlings (Bogdan,
1977). Vegetative propagation is either by canes
(stems) or root splits . The normal spacing is
0.5-1.0m within the rows and 1.0m between the
rows (Goldson, 1977; Palacpac, 1985). The splits
are planted upright with the roots below soil
surface whereas canes are laid horizontally in
furrows or slanting at an angle (Goldson , 1977 ).
Canes in furrows are completely covered with soil
while the slanting ones are partially buried.

Planting is done at the beginning of rains in

the areas with well defined seasons or at any time
where rains are continuous or under irrigation.
Weed control is necessary to maintain high yields
(Gosnell, 1963). The growth of elephant grass is
affected by perennial grasses such as couch grass
(Digitarda scalarum), stargrass ICynodon dactylon)
and Kikuyu grass ( Pennisetum c 1andestinum ).
Gosnell (1968) corroborated general observations
that the productivity of napier grass is greatly
reduced by the presence of perennial grasses. This
is in contrast to the work of Tiley (1960) who
thought that napier grass controlled couch grass
 9

in Uganda. Despite the fact that napier grass is a

large vigorous plant, it is highly susceptible to
weed competition of a l 1 types and clean weeding is
recommended to maintain good yields.

Fodder production in the tropics closely

follows the seasonal rainfall pattern. Use of
supplemental water for sustained plant growth
during periods of water stress in the dry season
has been suggested as a means of improving animal
output. The cost of irrigation is however
prohibitive to many farmers.It has been stated
that for continuous production, at least 2-3 cm of
water as irrigation or as precipitation is needed
weekly for continuous forage production (Crowder
and Chheda, 1982) and this is enhanced when
nitrogen fertilizers are added (Rivera-Brenes et
al, 1961; Osman, 1979; Koch, 1987). Water supply
affects dry matter yields, percent crude protein,
silica content, ash and nitrogen free extractives
(Oyenuga, 1960).

In Kenya napier grass is rarely grazed.

Rather.it is cut and fed fresh or occasionally
conserved. Napier grass responds well to
fertilizers especially nitrogen application
(Bogdan, 1977; Goldson, 1977; Wouters, 1985).

2.4. Dry matter yields;

The dry matter yield of napier grass has

 10

been extensively studied in many parts of the

world although yields have varied from one region
to the other depending on varieties, climate,
prevailing weather conditions, water supply, soil
fertility cultural practices and cutting interval
(Bogdan, 1977; Crowder and Chheda, 1982; Karanja,
1984).
In East Africa, dry matter yields of 10 to 40
tons/ha are often quoted (Goldson, 1977). A dry
matter yield of 25 tons/ha/year was reported by
Thairu et a].,(1968) in Kenya. Under good weather
conditions, Vare-Austin (1963) estimated that
yields of 15-20 tons/ha/year can be expected in
Kenya and 25* of this yield is produced during the
dry season. In a 1984/85 survey covering several
districts in Kenya, the highest dry matter yield
t farm level of 26.5 tons/ha/year was recorded
in Kakamega with a rainfall of 1910 mm and cutting
intervals of 7 weeks while Kilifi with a low
annual rainfall of 612 mm recorded only 5.5
tons/ha dry matter in 542 days at cutting
intervals of 18 weeks (Uouters, 1986). These
findings show similar trends to others obtained
from a countrywide trial encompassing 9 sites ten
years earlier. The fertilizer rates applied were
50 kg P2 O 5 /ha at planting and a nitrogen dressing
of 40 kg N/ha after every harvest and the results
obtained are shown in Table 2.1.
 11

Table 2.1. Average annual yields of dry matter

(tons/ha/year) over three year period (1975 77) in
9 l o c a t i o n s in K e n ya (Boonm an, 1 9 7 9 ).

L o c a tio n DM y ie ld s no. o f cu ts R a in f a ll Mean

tons/ha/year ove r teap

Gana French 3 y e a rs (as) (°C)

Caieroon

Kakaaega 2 1 .4 25.3 14 2000 20.1

K is ii 2 0 .0 13.4 14 2180 1 9.3

La net 1 1 .2 11.2 10 850 16.8

Eabu 1 3 .3 16.9 10 1100 2 0 .7

K it a le 1 6.0 12.9 10 1186 1 8.2

N aivasha 1 0 .4 11.6 10 660 1 7 .3

01 J o ro Orok 1 0 .0 7.5 8 1023 1 3 .7

E ld o re t 1 2 .8 13.9 10 1189 15.5

Bara ton 6 .6 7.4 12 1500 17.4

 12

Lanet and Naivasha stations recorded low

yields and this could have been a result of low

rainfall and high altitude which could have led to
low temperatures. Similarly, Baraton and 01 Joro
Orok had low yields despite the impressive
rainfall data. These locations are at fairly high

a 1t itudes of over 2000m above sea level

(Jaetzo1d, 1983) and have 1ow temperatures. 01

Joro Orok for instance has absolute minimum

temperatures of between 0 and £°C in all months.

All the four stations were above 1900m in altitude
and there was no clear relationship between
rainfall and yield level. Soil factors, including
the weed status, undoubtedly had important effects
but from the soil analysis it was difficult to
draw conclusions (Boonman, 1979).

In Trinidad, (Paterson, 1938) reported dry

matter yields of 41.4 tons /ha/year. Watkins and
Lewy van Severen, (1951 ) documented a yield of
38.1 tons/ha/year in El Salvador. In Puerto Rico,
dry matter yields of 11-15 tons/ha/year have been
recorded (Rivera-Brenes, 1962; Caro-Costas et
a/,1960; and Caro-Costas and Vicente-Chand1er

(1956, 1961).

On heavily fertilized humid, lowland tropics,

yields of up to 100 tons DM/ha have been obtained
(Crowder and Chheda, 1982). In this experiment
carried out in the Carribbean, unusually high
fertilizer levels, at a rate of 1320 kg N /ha and
supplemental irrigation both of which are known to
 13

boost dry matter yields considerably, were

applied. In Australia and southern Asia, DM yields
of 11 to 43 tons/ha have been reported (Muldoon
and Pearson, 1977 and 1979). Under farm
conditions, lower herbage yields are realized and
these may range from 2 to 10 tons/ha in
unfertilized stands and from 6 to 30 tons/ha per
year in fields well fertilized with nitrogen and
given a basic dressing of phosphorus (Bogdan,
1977). The use of fertilizers therefore alters the
dry matter yields (Crowder and Chheda, 1982). An
almost linear increase of dry matter yields with
successive increments of nitrogen fertilizer has
been noted in different grasses (Oyenuga and
Hill, 1966; Vicente-Chand1er et a 7., 1974;

Crowder, 1977).

It is well established that as the cutting

interval increases so does the dry matter yield.
Vicente-Chandler et al., (1959), reported a steady
increase in both dry matter yield and dry matter
content with increase in cutting interval from 6
weeks to 12 weeks at all levels of nitrogen tested
over a period of three years. The yields ranged
from 27.2 to 46.5 tons/ ha for cutting intervals
of 6 and 12 weeks respectively while the dry
matter percent rose steadily from 14.1% at 6
weeks to 25.1% at 12 weeks. Gomide et al.(1969) in
Brazil, observed a linear increase in dry matter
percent from 12.2 at 4 weeks to 43.0% at 36 weeks.
In Nigeria an increase of dry matter yields and
percent dry matter content from 4.8 tons/ha and
 14

16.6% to 13.7 tons/ha and 25.9% respectively were

noted (Oyenuga, 1959). The cutting interval ranged
from 3 to 12 weeks. In Kenya, Van der Kamp (1987)
reported a significant increase in percent dry
matter content from 11.5 to 16.6% with increase in
cutting interval from 6 to 12 weeks while the
yields recorded varied from 3.3 tons/ha to 11.2
tons/ha per year when 100 kg N/ha and 70 kg p2°5
were applied.

2.5. Nutritive quality:

2.5.1. Chemical composition

A satisfactory appraisal of napier grass as a

fodder for livestock requires the determination
not only ol agronomic factors such as productivity
and ecological suitability under conditions of
defoliation but also an adequate assessment of the
nutritive value. It is well known that the
chemical composition of a forage is an important
criterion in deciding which forages should be
planted and fed to the ruminant animal. The
overall quality of forage is dependent on the
relative proportion of high quality fractions.
Although leaves and stems are nutritionally of
equal value in early growth,the rate of decline in
quality for the different components varies as the
plant matures. The nutritive value of napier grass
is determined by changes in chemical composition
and digestibility and these two are dependent on
stage of growth , height of cutting , season of
 15

the year and fertilizer application ( Uare-Austin,

. 1963 ; Goldson , 1977 ). Advancing forage
maturity is accompanied by increase in dry matter
yield, a rise in cell wall constituents and a
decrease in percent cell contents.

Unlike in the majority of tropical grasses ,

e 1ephant grass has a large proportion of water
during the early stages of growth and the dry
matter percent is only 12-18 (Bogdan , 1977). The
percent dry matter content is the lowest of a 11
important grasses in East Africa (Boonman, 1979).
Thi s is in contrast to the recor ded percent dry
mat ter var ia t ions in other tropica 1 grasses for
instance Setaria sphacelata. Digitaria decumbens,
Cynodon dactyl on, (Coastal Bermuda), Brachiaria
decumbens and Chi oris gayana with ranges of 14.8-
20.7, 14.8-28.6, 19.1-35.4 and 19.5-42.1 percent
respectively (Miller and Brair-Rains, 1963;
Butterworth, 1967; Bogdan, 1977). Low dry matter
content leads to inadequate dry matter intake
(Dirven and Ehrencron, 1963) especially in the wet
season and subsequently reduced milk yields
compared with sown pastures (Osborne and Allan.
1971). During the dry season however, elephant
grass is superior since it is able to retain its
nutritive quality longer (Boonman, 1979; Karanja,
1981). In addition the dry matter percent
increases as the grass matures.

It is established that the crude protein

percent declines with increase in stage of
 16

maturity. Paterson (1933) and Wilsie and Takahashi

(1934) reported a negative correlation between
crude protein and harvest interval. In their
studies a CP value of 10% was reported at four
weeks growth , and this dropped to 6% CP when
harvest was delayed to 14 week intervals. In
Puerto Rico, a gradual decline in CP content from
12.10% to 6.73% with increase in cutting interval
from 6 weeks to 13 weeks was demonstrated
(Vicente-Chandler,et al 1959). Mwakha (1972)
reported a decline in the leaf and stem CP content
in Kenya from 13.8 and 8.4% to 8.7 and 2.8%
respectively with increase in the cutting interval
from 9 weeks to 34 weeks. Crude protein content
was shown to decrease from 15.46 at 5 weeks to 3.8
% at 14 weeks with delayed cutting (Gutierrez and
Faria, 1978). With heavy fertilization of 600-1000
kg N/ha Arias (1980) reported a CP content of 9.6%
at 7 weeks and 6.3% at 10 weeks. Decrease in
napier grass CP content with increase in cutting
interval have also been reported by other workers
in Hawaii (Wilsie, 1940 ), El Salvador (Watkins
and Severen, 1951 ) , India (Ranjhan and
Talapatra, 1967 ) , Brunei (Williams, 1980 ) and
Kenya (Karanja, 1984 ;Potter, 1985 ; Wouters, 1985
Van der Kamp, 1987).

Fertilizer application to napier grass alters

both the yields and chemical composition but the
magnitude is modified by climate, cultural
practices and other environmental factors. The
fertilizers may substantially increase the levels
 17

of protein and minerals in the grass. In the

Caribbean, Folkertsma (1981) has cited CP values
of 12, 8 and 6 weeks old napier grass ranging from
8.3 to 17.6. 6.5 *to 11.9 and 4.9 to 9.6%
respectively when fertilizer rates varied between
0 and 2250 kg N/ha/annum. From Kenya CP ranges of
4.9 to 9.5, 5.4 to 12.2 and 5.4 to 12.2% were
observed between 18 and 6 weeks of age when 0, 50
and 100 kg N/ha/year was applied (Wouters, 1985).
This observation has been reported by many other
researchers but the response to applied nitrogen
depends on stage of growth, amount and time of
N application, soil moisture and climatic
conditions (Crowder and Chheda, 1982). However, at
levels of 0, 100 and 200 kg N/ha per year, Gomide
et al. (1959) concluded that nitrogen fertilizers
had no effect on calcium, phosphorus and potassium
content.

Many soils in tropical areas respond to

modest applications of phosphorus with pronounced
increases in forage production (Cloutier, 1971;
Vicente-Chand1er et al., 1974) but in general
applied P increases the content of this nutrient
in herbage with no substantial effect on CP
content. Crowder and Chheda, (1982), have
similarly reported that raising application rates
for potassium only affects the nutrient levels in
plant tissue but not the CP content.

Increase in structural components of the

plant cells partly explains the gradual decline in
 16

the herbage quality with napier grass maturity.

Mu 1doon and Pearson (1979), reported crude fibre
values of 39 and 44% at 2 weeks and 4 weeks
respectively in Australia. The workers further
noted increases in total carbohydrates , cellulose
and lignin as the napier grass matured. This is in
conformity with the trend in most grasses where it
is well established that as the grass matures ,
the proportion of cel 1 w a 1 Is and constituent
fractions increase (Holmes, 1980). In Kenya,
though the data is limited, increase in fibre
contents including Van Soest fractions has been
reported (Karanja, 1984 ; Potter, 1985 ; Van der
Kamp,1987). Cell wall contents of 69.4% and 71.1%
for Bana grass and Pakistan napier were reported
by Brown and Chavalimu (1985) at 5 weeks of age.
Other workers have also noted rising fibre content
with increasing age as reported by Butterworth
(1967). Increasing van Soest fibre fractions have
also been demonstrated by Vicente-Chand1er et al,
(1959), Da Silva et a/, (1965), Gomide et al,
(1969), Johnson et al, (1973) and Potter and
An indo (1985).

Relatively high ash contents have been

reported for napier grass. In Venezuela, ash was
shown to decline as the grass matured from 15.4%
at 4 weeks to 9.2% at 10 weeks regrowth
(Butterworth, 1967). In Kenya, ash contents of
19.5% at 6 weeks and 15.6% at 18 weeks were
observed (Wouters, 1985). Van der Kamp (1987)
reported almost similar findings of ash contents
 19

that decreased with age from 21.7* at 6 weeks to

14.2% at 18 weeks. However, ash determination
gives no indication of the specific elements
present , or the combination in which a given
mineral occurs in the forage (Maynard et a/. ,
1979). In the work from Kenya, the napier grass
was irrigated while in the Caribbean it was
rainfed and therefore the very high ash contents
may partly be due to the marked splash from the
water sprinklers.

High potassium contents in napier grass were

noted in Kenya (Robinson and Cheney, 1958).
Moltavo et aJ (1987), has also demonstrated high
potassium contents in napier grass, the highest
concentration being in the leaves. It would
therefore be expected that as Ieaf:stem ratio
decreases percent potassium in the plant would go
down. In that experiment the leaves were also seen
to be richer in phosphorus. Under irrigation
conditions , Paretas and Gomez (1972) observed
declining calcium and phosphorus contents with
less frequent cutting of napier grass. These
results are corroborated by findings of Gutierrez
et al., (1978) who recorded a decrease in calcium
from 0.48% at 7 weeks maturity to 0.15% at 14
weeks while the phosphorus content ranged from
0.45% to 0.11% between 7 weeks and 14 weeks
respectively. For these two minerals, not only must
the forage provide adequate amounts but the ratio
between the two is important. Ratios of between
2:1 and 1:1 are optimum in animals (Maynard, et
 20

a]., 1979). However, in Kenya, little work has so

far been done to partition the ash into the
constituent elements or minerals for napier grass.

Percent ash content decline with increase in

nitrogen application has been reported (Wouters,
1986). This worker reported that for napier grass
the percent decrease was 1% for every 50 kg N
added per hectare. This may be a dilution effect
and it is expected that the ash constituents
would similarly decline. High ash content in feeds
is undesirable because it decreases the amount of
digestible energy (Holmes, 1980).

Grasses are known to contain appreciably high

amounts of silica and this may constitute 30-60*
of the ash (Jones, 1967). Van Soest and Lovelace
(1969) reported that silica is unevenly deposited
throughout the plant. Johnson (1973) observed a
highly variable growth stage effect on silica
content of napier grass. The worker noted a
rapidly increasing silica content during the first
eight weeks of growth and a tendency of levels
becoming constant in later stages. Tessema (1975),
however, reported uneven changes in silica content
with age. On the whole, reported changes in
silica content as the plant matures seem
contradictory.

2.5.2.Dieestibi1ity

Digestibility is known to be an important

 21

factor affecting the performance of ruminants fed

on forages. Study of aging and maturation has been
of particular interest since the two are known to
be inversely related to digestibility. Working
with French Cameroon cultivar, Said (1976) noted
in vivo organic matter digestion coefficients of
71.1%, 77.7% and 66.6% at 3, 5 and 7 weeks
regrowth stages respectively. Butterworth (1967)
has quoted in vivo organic matter digestibility of
napier grass to range from 64.9% at 4 weeks
regrowth to 59.8% at 10 weeks. Wouters (1986)
working with Bana grass demonstrated a declining
in vitro organic matter digestibility from 75% at
6 weeks of age to 63% at 12 weeks of age. A rather
high digestibility of organic matter in vitro of
77.4% at 6 weeks and a low one a.t 18 weeks of
54.1% (Van der Kamp, 1987) were recorded for
French Cameroon. Arias (1980) reported a decline
in in vitro digstibility from 72% at 7 weeks to
65% at 10 weeks with heavy fertilization of
of 600-1000kg N/ha. These results corroborated
previous findings by Mwakha (1972), Ogwang and
Mugerwa (1976) Reid et al (1973), Olubajo and
Oyenuga (1974), and Leon et al., (1987).

Among the factors that are known to affect

digestibility are fertilization, environmental
conditions and age of the plant. With advancing
maturity, cellulose, hemice11u 1ose and lignin
contents rise (Van Soest and Robertson, 1980).
Lignin is a primary factor causing a decline in
digestibility with maturity. Grasses also tend to
 22

accumulate silica as a secondary protective factor

for plant cell wall and it is known to depress
digestibility (Van Soest and Jones, 1968).

Chemical composition and in particular

structural components are known to markedly affect
digestibility (Jones, 1973). The in vitro
digestibility of napier grass was reported to
increase with increased level of nitrogen
application from 64.3 to 71.25% for 0 and 300 kg N
/ha fertilizer rates respectively (Ogwang and
•

Mugerwa, 1976). Environmental conditions namely

light, temperature and moisture influence
vegetative development of the forage species
(Smith, 1976) and this affects both chemical
composition and subsequently digestibility.

The nylon bag technique has been used for

many years to study the ruminal microbial
breakdown (degradability) of forages (Fina et a!.,
1958; Hopson, et al., 1963). Little, however, is
known of the relative degradabilities of a wide
range of tropical forages (0rskov, et al., 1980).
The nylon bag incubations have advantages in that
the digestion takes place in the in vivo rumen
environment. Studies made during in vivo rumen
fermentation have the added advantage that there
are no problems of temperature control and removal
of end-products which have to be coped with as
soon as rumen fluid is taken into a laboratory for
in vitro fermentation (0rskov et al., 1980). The
worker has further reported that sometimes the in
 23

vivo digestibility is quite similar to

degradation at 48 hours incubation.
 24

3.0 MATER I ALS AND METHODS

3. 1 St udv sites

The study was carried out at the following

two places:-

1. National Animal Husbandry Research Station

(N. A. H. R. S. ) , Naivasha.

2. University of Nairobi, Kabete.

The napier grass cultivars were grown at

Naivasha and the bulk of laboratory analytical
work was done at the same station. At Kabete, in
vitro digestibility and ruminal dry matter
disappearance in sacco were carried out.

3.2 Site description (Naivasha)

The station is situated at 0* 40'S and 36*

26’ E at an altitude of 1900 m. This high
altitude modifies the tropical climate. The mean
annual temperature is 16.6*C with the highest
temperature at 28.3*C in the month of March and
the lowest at 6.8*C in February. GeneralIy,
temperature and humidity fluctuations are small.

Though the average annual rainfall is

moderate at 653 mm (Jaetzold. 1983), the amount
time, and intensity of rainfall is so variable
that prediction is difficult. April, May and
 25

November are the wettest months with averages of

121 mm, 103 mm and 71 mm respectively. January ,
the driest month receives only 32 mm. The area is
in agro-eco1ogica1 zone IV and rainfed agriculture
is uneconomical (Jaetzold, 1983). In addition the
area tends to be windy and while this has a
cooling influence it has the disadvantage of
increasing the evapo-transpiration thereby
reducing the moisture available for plant growth.
The water received by the napier grass during the
experimental period can be seen in the bar chart
in Figure 3.1. This shows rainfall plus the
irrigation water supplied. Table 3.1 shows the
changes in minimum, maximum and mean temperatures
over the same experimental period.

The soils are of volcanic origin having

developed from volcanic ashes. The soil is dark
greying brown, deep and imperfectly drained. It
is sodic and of low fertility with a humic
topsoil (Jaetzold, 1983). The terrain is fairly
level and the natural vegetation consists of open
short grass highland savannah with scattered tall
acacia trees (Kimenye, 1978).
 26
(millimeters)
supplied
Water

Fig. 3.1: Moisture received by the napier grass '

(Rainfall plus irrigation)
 27

Table 3.1 Changes in minimum, maximum and mean

temperatures during the experimental
period (September, 1987 to February,
1988).

Month

Sep. Oct. Nov. Dec. Jan. Feb.

Tempe­
rature

min.
o
S-*

13. 2 12. 3 13. 9 13. 5 14.4 13. 7

o

max.
o

26.6 27.5 24. 8 28. 0 26.5 28.9

o

mean
o

19.9 19.9 19. 3 20. 7 20.5 21.3

o
 28

3.3 Experimental field

The two cultivars (Bana and French Cameroon)

were planted in April, 1987 at a spacing of 0.9m
x 0.6m. The experimental field of 97.2m x 25.2m
was subdivided into 36 plots of 2.7m x 4.2m.
A clean cut leaving a stubble height of 5cm
from the ground was done in September, 1987. This
was followed by weeding and then fertilizer
application at the rate of 50kg, 90kg and 200kg
per ha for nitrogen (N), phosphorus <P2°5> and
potassium CK2 0> respectively. Every two weeks the
field was irrigated for 12 hours continuously.
This was equivalent to 26mm water per month. The
field was kept weed free and one top dressing with
N at a rate of 50kg/ha was done at six weeks of

regrowth.

3 . 4 Experimental and Treatment Design

The design was a split-plot with cutting

stage (subplot) nested within variety (main plot).
There were two varieties and six cutting
stages. Variety was treated as a fixed effect
and stage of cutting as a random one. There
were three replications per treatment. The

stages of cutting were 4, 6, 8, 10, 12 and 14

 29

weeks. Treatment al location to plots was randomly

done.

3.5 Harvesting and Sampling Procedures

Sampling commenced in October, 1987.

The height of napier grass was measured by

means of two methods. The first method was carried
out by lowering a disc of polystyrene foam (
diameter 75 cm, thickness 1.5 cm ) on the grass
*
and reading the height with a graduated measuring
stick. This method was applied when the plant
height was 25 cm and below. The second method was
by visual observation of the average height of the
shoots and measuring with the same measuring
stick. All plants in an experimental plot were
measured and from these results, the average
height for the plot was calculated.

The experimental plots were cut manually

using sickles on the day of harvest. The grass was
cut as closely as possible to ground level but on
average a stubble height of 5 cm remained. At the
time of harvesting, the perimeter (guard row) of
each plot was cut first and discarded. The
cut herbage from the plot was weighed by means of
a weighing balance (to the nearest 0.2 kg).

The sampling of the grass for dry matter and

chemical analysis was carried out by chopping a
 30

bulk sample of about 15 kg from each plot with a

hand operated chaff-cutter into pieces of about
2.5 cm after which a sample of about 5 kg chopped
material was taken. This sample was spread on a
line ridge. Sub-sampling was done by taking two
segments of the ridge. For each treatment two
samples were taken for dry matter analysis (dried
at 105°C) and one for chemical analysis (dried at
70°). Sampling for the determination of the
leaf:stem ratio was done by picking at random a
number of shoots from the heap of cut napier
grass after weighing. The samples were separated
into fractions of leaf, stem (including leaf
sheath) and dead material. These fractions were
dried at 105°C for 48 hours in a Memert oven and
later weighed. The weights were used to calculate
the leaf :steni ratios.

The dry matter yield per plot was calculated

as foilows:-

Dry matter Fresh wt. Percent

yield tons/ = (kg/plot) X 10000 X dry matter
ha 1000 Area of
plot (m^)

3.6 Analytical Procedures

3.6.1 Chemical Composition

Analysis for proximate components was done

by methods of A0AC (1984). Acid detergent fibre
 31

(ADF), Neutral detergent fibre (NDF) and acid

detergent 1ignin (ADL) values were determined
following the procedure of Goering and Van Soest
(1970). The silica values were obtained as the
ash after ADF ash was leached with 46% hydrobroiaic

acid (HBr) for one hour.

For mineral analysis, the dry ash was

dissolved in 20% H C 1 and necessary dilutions made
with deionized water. Calcium (Ca) and Potassium
(K) were estimated with a flame photometer
(Beckmann, Klina flame). Ultra-violet
electrophotometer was used to estimate phosphorus

content of the samp.les.

3.6.2 Di ge s tib il ity Estimates

The in vitro digestibility of dry matter

(1VDMD) and in vitro organic matter digestibility
(1V0MD) were estimated using the two-stage Tilley
and Terry technique ((Tilley and Terry, 1963).

3.6.3 Ruminal Dry matter degradability— ln sacco

Three Red Maasai sheep were used in this

investigation. The animals ranged in age from 48
to 56 months and liveweights varied from
approximately 45 to 50 kg. The sheep were kept in

metabolism cages.

The animals were fistulated and had a basal

 32

diet of Rhodes grass hay (Chloris gayana) offered

ad libitum. A dairy meal concentrate was given at
the rate of 250 g per sheep per day. Chemical
composition of these feeds on offer is given in
the appendix (Appendix III).

Sub-samples of napiergrass at different

stages of growth were dried at 70*C and ground
with a Wiley hammer mill through a 2 mm screen.
For each growth stage six sub-samples of 3 g each
were weighed and put into nylon bags (7 cm x 14
cm). The pore density was 1600/cm2 . Samples in
nylon bags were marked and inserted into the rumen
at 8.00 a.m and 8.00 p.m. Every 12 hours a new set
of bags were introduced and at the end of 72
hours, all the bags were removed. Therefore
samples incubated for 12, 24, 36 48, 60 and 72
hours were obtained to determine the rate of dry
matter disappearance. At the end of incubation,
the bags were rinsed in tap water for 30 minutes,
suspended in the air for water to drip and then
oven-dried at 70*C for 48 hours. They were again
weighed to give the dry matter loss in the rumen.
To estimate disappearance at zero-time, another
set of six samples for each growth stage were
I
washed in tap-water for 5 minutes. These were
then dried in the oven in a manner similar to the
incubated samples.

Three replications of the experiment were

conducted. Data were blocked by sheep with a
randomized block design in which time and stage of
growth of the forage were the factors considered.
 33

The data obtained was described by equation

of j/rskov and McDonald (1979). The percentage
of napiergrass degraded (p), was described as
fo 1 lows:
p = a + b ( l - e C ^) (1)

where a, b and c are constants.

Thus we have several terms:

p = the actual degradation after time ’t*

a = degradation at zero time
b = proportion of napiergrass which will be
degraded given sufficient time
c = the rate constant for the degradation

of ’b ’.
t = time of incubation in hours.

The determination of c is particularly

important because it provides information on
factors affecting intake of roughages, namely
digestibility and rate of digestion (Orskov,
1986). The values of (b) can be obtained by
extrapolation. The (a ♦ b) value gives some
expression for the potential digestibility for the
fibrous feed. The sketch in Figure 3.2 attempts to

show some of these components.

The half-life <T1/ 2 > is the time taken to
reduce the incubated sample by one half both by
dissolving in the rumen liquor and through
microbial break-down. To obtain the degradability^
rate at a given time Ct), the components are
 34
disapppearance (p) of feed
Percent

Fig. 3.2: Illustration of degradation curve component

(Sketch).
 35

fitted in equation (2) and thus:-

e
-ct = a + b -p

b (2)
and by taking the natural logarithm on both sides
of the equation, it is found that:
-C t = In (a + b - p )

The average dry matter degradability rate

1/2
(c), was used for computing the half life <T )
within the rumen. These rate constants (c) were
calculated from the slope of the regression of the
semi-log of undegraded napiergrass against time
(Mapoon, 1980).They were subsequently used to
obtain the half-life T 1/2 from equation (3) shown

be 1ow: -

T 1 /2 = -l n
c <3>

3.7 Statistical analysis.

Means, standard deviations, standard errors,

regression coefficients and correlations were
obtained using PANACEA Programme of Pan Livestock

Services Ltd. (1986).

The analyses of variance were obtained using

the procedures described by Steel and Torrie
(1980) and mean separation was accordingly done by
use of least significant difference (LSD) at

P<0.05.
 36

^ - O RESULTS AND DISCUSSION

4. 1 Dry matter yields, grass height at cutting

and leaf;stem ratio

Table A.1 shows the means for dry matter

yield, plant height at harvest and leaf:stem
ratio.

4.1.1 Dry matter yield

Dry matter yields increased significantly

(P<0.05) from 1.37 to 9.00 tons/ha and from 1.24
to 10.01 tons/ha for Bana grass and French
Cameroon respectively as napier grass matured.
Varietal difference in dry matter yield was
insignificant <P>0.05). Increase in dry matter
yields as the grass matures are consequences of
additional tiller and leaf formation, leaf
elongation and stem development (Akinola et al,
1971; Dovrat et al., 1971; Fisher, 1973; Michelin
et al., 1968; Robertson, 1976). Assuming
unchanging climatic and cultural practices,
the yields recorded in the experiment would
range from 16.12 to 37.18 tons/ha/year.
This yield falls within the range reported in
East Africa of 10-40 tons/ha/year (Thairu et al.
1968; Mwakha, 1972; Goldson, 1977; Karanja, 1984;
Wouters, 1985 ). However, much higher yields
have been reported elsewhere for instance Colombia
 37

Tatle 4.1: Influence of cutting stage on dry

matter yields, grass height at
harvesting and leaf:stem ratio

Cutting Dry matter H e i g h t at L e a f : S tem

stage ( tons/ha) harvesting ratio

( weeks) (cm)

B F B F B F

4 1.37a 1 1.24a 22.1 l a 27.51a 3.77a 5 .03a

6 2.71a 3 .06b 5 6 .86b 66.61b 2.53b 2.40b

8 4.14b 4.96c 58.52b 92.16c 2.23b 1 .33c

o
o
10 5 - 2 4 bc 5 .14c 75.52c 108.48c 1 .50c

o
1•20cd 0.87d
CD
m

12 7.68d . 92.54d 151.88d

O

14 9.00d 10.01e 124.88e 189.99e 0.93d 0. 53d

B indicates the columns for Bana grass while F

indicates the columns for French Cameroon.

1 Means with different subindices within the same

column differ significantly (P < 0.05).
 38

(Crowder and Chheda, 1982), Puerto Rico (Vicente-

Chandler et al. 1959), and El Salvador (Uatkins
and Lewy-Van Severen, 1951), a possible effect of
the comparatively higher fertilizer application
rates in the Caribbean and ample water supply.
This is an illustration that differences in dry
matter yields exist in napier grass grown under
different environmental and agronomic conditions.

4.1.2 Grass height at cutting

At four weeks the mean heights were 22.11 cm

and 27.51 cm for Bana grass and French Cameroon
respectively. This increased significantly
(PC0.05) with increase in napiergrass maturity up
to 124.88 and 189.99 cm at 14 weeks respectively.
Cultivar French Cameroon showed insignificantly
higher (P>0.05) heights at all stages of growth
studied. A number of previous experiments have
been conducted with the aim of using height as a
criterion for optimum cutting stage. It is
generally agreed that with delayed cutting there
is a corresponding increase in total dry matter
yield and that height is a reflection of dry
matter yields (Boonman, 1979). The results
obtained in the experiment supported this
view since for both varieties the taller the
grass, the higher the dry matter yields. Tiley
(1970) as reported by Boonman (1979) observed that
napier of 100cm height corresponds with a crude
 39

protein percent of about 10%, an organic matter

digestibility of 57% and a leaf:stem ratio of 1.0.
At the same height Wouters (1986) reported a CP
percent of 6.8 and organic matter digestibility of
64.9% for Bana grass which had received 50 kg N/ha
and cut at 12 weeks of age. In the current study
1m height was achieved at about 12 weeks for both
cultivars. The corresponding CP content and in
vitro OM digestibilities were 5.89 and 64.54% for
Bana grass and 5.68 and 64.64% for French Cameroon
respectively. However, nutritive value parameters
derived from grass height should be treated with
caution since height is a function of climate and
cultural practices. For instance higher nitrogen
and water application will considerably increase
the rate of growth and at 1 metre cutting height
crude protein content, dry matter yield and
digestibility is expected to be much higher.

4.1.3 Leaf:Stem ratio

As the napier grass matured, leaf:stem ratio

significantly declined (P<0.05). This ranged from
3.77 to 0.93 and 5.03 to 0.53 for Bana grass and
French Cameroon at 4 and 14 weeks respectively.
After 6 weeks the decrease was more gradual.
However, differences between the varieties were
insignificant (P>0.05). The ratios recorded in
this experiment compare with the declining ranges
of between 2.10 and 0.28 reported by Mwakha (1972)
between 2 and 8 weeks. Although the ratios for
Mwakha (1972) are for different ranges of growth
 40

stages, a trend similar to the one observed in the

current experiment is seen. It should be noted
however that it is important to include leaf
sheath to avoid giving a misleading leaf:stem
ratio.
Elephant grass is capable of producing high
yields but half or more of this is composed of
inedible stem, which may be wasted under a cutting
feeding system (Boonman, 1979). Strange (1959)
and Boonman (1979) reported that Bana grass
produces more leaf than any other variety of
napier grass. Except for the 4 weeks regrowth,
which demonstrated particularly high leafistem
ratio of 3.77 and 5.03 for Bana and French
Cameroons respectiveIy, the findings in this study
are comparable to the results of other
investigators. The decrease in 1eaf:stem ratio as
the grass matured was more drastic in French
Cameroon than was the case with Bana grass over
the same period. Crowder and Chheda (1982)
reported that leaves have a higher digestibility
than stems in mature herbage. Leaf:Stem ratio is
known to affect digestibility and chemical
composition of forages (Van Es, 1982, Crowder and
Chheda, 1982). Calcium and iron contents are
known to be higher in leaves than in stems
(Mo 1tavo et a]., 1987). As the plant matures, the
percentage of leaf declines due to stem elongation
and death of older leaves. High leafrstem ratio is
desirable since it may improve forage intake.
Whereas this may be true in unchopped napiergrass
due to animal selection for leaves, the increase
 41

in intake may be partially masked by chopping.

This is a common practice with many farmers.

4•2 Chemical composition

4-2.1 Proximate composition

Data in Table 4.2 indicate that the dry

matter content increased significantly (PC0.05)
with increase in cutting age. The values varied
from 15.81% to 17.67% and 14.52% to 17.70% for
Bana grass and French Cameroon respectively at 4
weeks to 14 weeks. From 10 weeks to 14 weeks the
increases in dry matter percent were however
gradual for both cultivars. Differences between
the cultivars were not significant (P>0.05).

The observed dry matter percent is low and

confirms the observations reported by Butterworth
(1967), Gomide et at., (1969), Vicente-Chand1er et
al., (1974), Wouters (1985) and Van der Kamp
(1987). The percent dry matter content of elephant
grass is the lowest of all important grasses in
East Africa. The thick fleshy stems are
responsible for the high water content (Boonman,
1979). The high moisture content is likely to be
a limiting factor to dry matter intake in animals
particularly in the early stages of growth.
Spoilage in ensiled napier grass has also been
attributed to this high moisture content and this
almost rules out hay making from napier grass
(Boonman, 1979).
 42

The proximate analysis data presented in

Table 4.2 and 4.3 showed considerable variations
in the crude protein, ash and carbohydrate
contents. The percentage values of ether extract
indicated relatively narrow ranges. Varietal
differences were evident (P<0.05) for crude
protein, crude fibre, ether extract and ash. The
rise in crude fibre from 23.39 to 36.30% for Bana
grass and 23.41 to 39.01% for French Cameroon at 4
and 14 weeks respectively shows a significantly
(P<0.05) rapid increase during the early regrowth
stages (4-8 weeks) while the increase became

gradual in later stages.

Crude protein content.1varied from 14.84 to

3.98% and 13.85 to 5.01% for French Cameroon and
Bana grass respectively. Similar trends have
previously been reported by other workers
(Paterson, 1933; Wilsie and Takahashi, 1934;
Vicente-Chand1er et al ., 1959; Mwakha 19/2,
Karanja, 1984; Potter, 1985; Wouters, 1985 and Van
der Kamp, 1987). However, napier grass at the same
stage of growth may have completely different
crude protein content. Many researchers have
shown that nitrogen application increases crude
protein content of napier grass (Thairu et a!.,
1968; Vicente-Chand1er et al., 1974; Wouters,
1985). Other studies have shown this to be true
for Bermuda grass (Burton and . Jackson, 1962)
Pangola grass (Whitney and Green, 1969) and mixed
natural pasture (Olsen and Santos, 1975).
 43

Table 4.2: Influence of cutting stage on * dry

matter, crude protein and crude fibre.

Cutting Dry Crude Crude

stage matter Protein fibre
(weeks) percent
B F B F B F

4 15.81a 14.52a 13.85a 14.84a 23.39a 23.41a

6 15.72a 14.96ab 10.87b 10.23b 26.86b 29.59b

8 16.43^ 15.12bc 8.61c 7.54c 32.21c 34.02c

10 17.09c 15. 72cd 7.24d 5.68d 34.93d 37.14d

12 17.24c 16.09d 5.89e 4.07e 34* 56de 38.45d

14 17.67c 17.70e 5.01f 3.98e 36.30e 39.01d

B indicates the columns for Bana grass while

indicates the columns for French Cameroon.

1 Means with different subindices within the same

column differ significantly (P < 0.05).
Table 4.3: Influence of cutting stage on ether
extract, nitrogen free extract and
total ash.

Cutting Ether Nitrogen Total

stage extract free ash

(weeks) extract

B F B F B F

4 2.74a 3.42a 33.74a 35.07a 26.41a 23.27a

6 2.62a 2.85b 34.46ab 36.28b 25.07ab 20.93b

8 2.08b 2.20c 35.46b 36.41b 23.32b 19.97c

10 1•?8C 1.96c 36.10bc 37.29b 20.84cd 18.60cd

12 l-60cd 1-91c 36.80cd 36.52c 20.60de 17.75d

14 1.49d 1.64d 36.22d 39.86d 18.96e 15.75e

B indicates the columns for Bana grass while

indicates the columns for French Cameroon.

* Means with different subindices within the same

column differ significantly (P < 0.05).
 45

In this experiment, the water supplied was far

above the weekly minimum requirement of 2-3cm
for growth (Crowder and Chheda,1982). The low
protein content observed especially after 6 weeks
may therefore be due to dilution effect where
foliage growth is rapid and the demand is thus
higher than the nitrogen absorption rate from the
soil (Crowder and Chheda, 1982). Bana grass had a
crude protein drop of 48% from 4 to 10 weeks and a
drop of 30% from 10 to 14 weeks. The corresponding
percentage decline for French Cameroon were 55%
and 40%. Therefore the drop in CP in French
Cameroon appeared to be more drastic and this is
positively correlated to the drop in leafzstem
ratio. The critical protein levels for optimum
microbial activity of 6-7% (Milford and Minson,
1966) were not met after 10 weeks. These low
values are likely to affect digestibility and may
reduce dry matter intake. The low crude protein
content after 8 weeks tend to suggest the need for
protein supplementation if napiergrass is the only

source of feed.

Ash was high in the dry matter but it

significantly decreased (P<0.05) with age from
26.41% to 18.96% for Bana grass and
from 23.27 to 15.75% for French Cameroon
at 4 and 14 weeks of age respectively.
High ash content for napier grass has been
reported by other investigators and may be
attributed to the fact that napier grass is a
heavy mineral feeder (Boonman, 1979). In
 46

Venezuela, Butterworth (1967) reported a range of

15.4 to 9.2% between 4 and 10 weeks of regrowth.
In Kenya, ash contents of 19.5% at 6 weeks and
15.6% at 18 weeks have been observed (Wouters,
1985). Similar findings were reported by Van der
Kamp (1987) when he recorded 21.7% ash at 6 weeks
and 14.2% at 18 weeks. The latter two workers
used irrigated napier grass. Soil contamination
due to water splashing during irrigation and rain
showers are likely to have caused these high
figures. Thefindings of Van der Kamp (1987) and
Wouters (1985) obtained under similar cultural
practices and climate to the current study gave
comparable results. The declining trend of ash
content with age may be a result of reduction of
soil splash by increased ground coverage by
foliage. Conversely, the lower plant heights
observed for Bana grass may have contributed to
the recorded insignificantly (P>0.05) higher ash
contents. In addition, mineral absorption rate may
have been unable to cope with the increased growth
rate in biomass due to continued depletion of the
soil reserves.
Variations in potassium (K), phosphorus (P)
and calcium content are presented in Table 4.4. At
4 weeks of growth, percent calcium and phosphorus
were 0.36 and 0.29 ; 0.39 and 0.31% for Bana
grass and French Cameroon respectively. These
decreased insignificantly (P>0.05) with age to
0.30 and 0.22 ; 0.32 and 0.21% at 14 weeks for
Bana grass and French Cameroon respectively.
Potassium constituted about 50% of total ash and
Table 4.4: Influence of cutting stage on
potassium, calcium and phosphorus.

Cutting
stage
(weeks) Potassium Calcium Phosphorus

B F B F B F

A 10.05a 1 11•?0a .36 .39 .29 .31

.35 .39 .29 .31

CM
CD
•H

6 9’98ab
(tl

8 9.A5b 9.70b .35 .37 .26 .27

10 6.87c 9.33b .32 .35 .2A .25

->l

.22 .25
cn

A.97c .31 .32

CL

12

1A 4.5Ad A.21d .30 .32 .22 .21

B indicates the co 1umns for Bana grass while F

indicates the columns for French Cameroon.

Means with different subindices within the same

column differ significantly (P < 0.05).
 48

significantly (PC0.05) declined with age. There

were no significant (P>0.05) cultivar differences
in calcium, potassium and phosphorus contents.
Crowder and Chheda (1982) have reported that
advancing plant maturity depresses potassium
content in herbage due to translocation to the

roots.

High potassium contents have been reported

not only in napier grass but also in other
grasses. Luxury consumption of K is known in
elephant grass (Robinson and Cheney, 1958;
Annenkov, 1982; Vicente-Chand1er et al., 1974; and
Moltavo et al., 1987). Increased nitrogen and
phosphorus application was observed to induce
increased K consumption in P. pur pureum, P.
maximum and Brachiaria mutica (Vicente-Chand1er et
al., 1959). However, the levels of K decreased
(P<0.05) with increased cutting age and this was
also true for calcium and phosphorus. The
declining calcium and phosphorus content trends
observed here corroborate with the findings of
Gomide et al., (1969); Paretas and Gomez (1972);
and Vicente-Chand1er et al., (1974). Gutierrez et
al., (1978) noted a decline from 0.48 to 0.15
percent and 0.45 to 0.11 percent for calcium, and
phoshorous respectively at 5 and 14 weeks of
growth. Gomide et al. (1969) recorded a calcium
content decrease from 0.61 at 4 weeks to 0.30 X at
36 weeks. Over the same period, phosphorus
content decreased from 0.33 to 0.08%. Compared to
the current experiment these levels showed a wide
 49

range. The trend may be explained by the dilution

effect and soil depletion since Gomide et
aV.,(l969) found no nitrogen fertilizer effect on
phosphorus and calcium levels at the rates of
between 0 and 200 kg N/ha per year.

Mineral requirements in cattle is largely

determined by the age of the animal, its
physiological condition, and level of productivity
(Annenkov, 1982). At the same time insufficient
intake of minerals is known to impair
acceptability and digestibility of a feed as well
as reducing milk yield. A 400 kg cow with a yield
10 kg milk per day requires 51 and 41 g calcium
and phosphorus respectively per day (Annekov,
1982). Given the observed levels of calcium and
phosphorus and assuming a similar production and
dry matter intake levels of 3* body weight, it may
be necessary to supplement for these minerals if
napiergrass is the sole feed. High potassium
levels, ranging from 11.70% to 4.21% were recorded
in the study and such levels may interfere with
magnesium absorption (Newton et aJ., 1962;
Fontenot et al., 1973). However, the ratio of
calcium and phosphorus is within the optimum
ranges of 1:1 and 2:1 (Maynard, et a!., 1979),
thus interference with absorption of other
mineral elements would not be expected.
 50

4*2.2 Detergent Fibres and Lignin

Table 4.5 shows the mean values of neutral

detergent fibre (NDF), acid detergent fibre (ADF),
and acid detergent lignin (ADL) as the cutting
stage was delayed. Neutral detergent fibre rose
steadily from 52.18 to 72.68 for Bana grass and
from 54.44 to 71.81% for French Cameroon at 4 and
14 weeks respectively. Acid detergent fibre (ADF)
and Lignin (ADL) showed similar trends. For these
three components the increase tends to level off
after 12 weeks regrowth, although this is not
very distinct in cultivar French Cameroon. No
varietal differences (P<0.05) were observed for
the constituents.

The above observation concur with the

findings of Van der Kamp (1987), who reported
ranges of 46.21 to 68.9%, 25.0 to 39.3% and 2.4 to
5.5% and for NDF, ADF and ADL respectively for
French Cameroon between 6 and 18 weeks of age.
Over a period of one year at Muguga, Kenya, ADL
was recorded to vary from 3.3 to 5.7% with a
cutting interval of 4 weeks for Bana grass (Potter,
1985). The variations in reported values of these
constituents may be due to seasonal and climatic
changes in the locality. Similar trends were
observed by Mwakha (1972), Vicente-Chand1er et
a 1., 1974), Karanja (1984) and Wouters (1985). The
same trend is recorded for other grasses for
Table 4.5: Influence of cutting stage on
detergent fibres, and lignin •

Cutting

s t age

(weeks) NDF ADF ADL

B F B F B F

4 5 2 . 18 a 1 54.44a 27.10- 2 5.97a 2.31a 2.26a

6 5 6 . 4 8 ab 59.77a 31.01b 31.01b 2.81a 2.73a

8 61.44b 65.37b 35.42c 36.66c 3.69b 3. 4 8 b(.

10 69.29c 6 8 . 5 5 bc 36.92c 3 8 . 10 c 3 . 7 4 bc 3.51c

12 72.22c 6 9 . 7 4 bc 40.08d 3 9 . 0 5 cd 4 - 2 8 cd 4.50d

14 72.68c 71 •8 1 c 41 • 45d 42.0 0 d 4.95d 5 . 77e

B indicates columns for Bana grass while F

indicates the columns for French Cameroon.

1 Means with different subindices within the same

column differ significantly (P < 0.05).
 52

instance as reported for Melinis minuti f1ora

(Gomide et a/., 1969), Cynodon dactylon
(Wilkinson et al.f 1970) and Chloris gayana
(Abate, 1978).

4.3 In vitro Digestibility coefficients and

silica content

Table 4.6 illustrates the silica percent

content in the dry matter. The levels were high
and constituted about 25% of the ash. The levels
increased gradually as the cutting stage
was delayed. This ranged from 4.07 to 6.67% and
4.57 to 6.35% for Bana grass and French Cameroon
respectively at 4 and 14 weeks of age. French
Cameroon had lower (P<0.05) silica contents than
Bana grass. Probably the splashing effect was more
on Bana grass since it had correspondingly lower
recorded plant heights and possibly greater soil
contamination of leaves.

Johnson (1973) and Singh, (1975) have

reported silica levels that did not indicate a
decreasing or increasing pattern as napier grass
matured but rather fluctuated inconsistently with
age. The current results however, concur with
those of Jones (1967) and Jackson (1977). Grasses
in general tend to accumulate silica as a
secondary protective factor for plant cell wall,
and this may explain the observed silica trends
with napier grass. The contradictory results, from
different investigators may be mainly due to the
 53

Table A.6: Influence of cutting stage on silica

(percent in dry matter) and in vitro
digestibility coefficients.

Cutting DM OM
staa= digest. digest.
(weeks) Silica

B F B F B F

4 4.07a 4.57a 69.00a 72.04a 74.65a 75.72a

6 5.01b 5.12b 66.64b 69.81b 72.73a 74.09a

8 6.12c 5.32b 64.00bc 63.50c 69.15b 68.67b

10 5.73c 63.30c 60.99d 65.57c 64.64c

12 6-46cd 6.06cd 63.18c 59.43d 63.54c 63.30c

14 6.67d 6. 35d 58.39d 60.62d 63.42c 62.22c

B indicates columns for Bana grass while F

indicates the columns for French Cameroon.

Means with different subindices within the same

column differ significantly (P < 0.05).
 54

fact that silica arising from soil contamination

cannot be distinguished from plant metabolic forms
(Van Soest and Jones, 1968).

Cellullose and hemicel1u 11ose, the major

components of plant cell walls, though resistant
to enzymatic breakdown are broken down by rumen
microbes to simpler carbohydrates. However, their
utilization is dependent on the silica and lignin
levels of the forage (Crowder and Chheda, 1982).
Lignin and silica have high resistance to chemical
degradation and complex with carbohydrates making
them unavailable for digestion. Lignin levels rose
(P<0.05) as the grass matured and so did the
silica levels (P<0.05). Silica is absorbed by
forages along with water and reduces digestibility
much in the same way as lignin (Van Soest, 1968).
Therefore the higher the silica and lignin
contents the lower the digestibility.

In vitro dry matter and organic matter

digestibility (IVDMD and IVOMD) values are shown
in Table 4.6. Dry matter digestibilities decreased
(P<0.05) with age from 69.00 to 58.39% and 72.04
to 60.62% at 4 and 14 weeks for Bana grass and
French Cameroon respectively. Organic matter
digestibility coefficients at the same maturity
stages decreased (P<0.05) from 74.65 to 63.42% and
75.72 to 62.22% for Bana grass and French Cameroon
respectively. After 8 weeks the depression in
digestibility is however gradual. Varietal
differences for both dry matter and organic matter
 55

in vitro digestibility were insignificant (P>0.05)

although the values for Bana grass seemed to be
numerically higher after 8 weeks.

Reid et al., (1973) observed in vitro dry

matter digestibility values for napiergrass that
decreased similarly from 70.2% at 4 weeks to 58.1%
at 14 weeks. The author reported digestibilities
of 63.2, 63.1 and 63.8% at 8, 10 and 12 weeks
respectively. In this study Bana showed values
close to those reported by Reid et a)., (1973) but
French Cameroon had slightly lower values over the
same range of growth stage. Van der Kamp (1987)
has also reported digestibility coefficients that
are similar to the findings of the present work.
At 6 weeks the in vitro organic matter
digestibility was 76.9% and this dropped to a low
value of 54.15% at 18 weeks. Similar results have
been obtained by Gomide et al. (1969), Mwakha
(1972) Vicente-Chand1er et al., (1974), Said
(1976), Karanja, (1984), Uouters (1985), and

Potter, (1985).

These observed changes of digestibility with

age can be attributed to changes in chemical
composition particularly in fibre, lignin and
silica content. The variation in digestibility
between the cultivars may be a result of
differences in genetic constitution among
varieties since environmental conditions and
management practices were uniform in the current

exper iment.
 56

The Ieaf:stem ratio was noted to decrease

with stage of growth, and a positive correlation
was observed between this ratio and digestibility.
This can be explained by the fact that the
rate of 1ignification is much slower in leaves
with increase in age than stems. This results in
a rapid decline in digestibility of the stem
fraction compared to the leaf fraction with
advancing maturity (Raymond, 1969).

4.4 Relationships between chemical components and

digestibi1itv

The fibrous constituents, neutral detergent

fibre, acid detergent fibre plus lignin and silica
varied significantly (P<0.05) with age at cutting
but they were negatively correlated to the
estimated digestibility values. The correlation
coefficients are shown in Table 4.7. Acid
detergent fibre was significantly correlated to
neutral detergent fibre and acid detergent
lignin. The regression equations are presented in
Table 4.8 (a) and (b). All the correlations were
significant (PC0.05). Neutral detergent fibre was
highly positively correlated (P<0.05) with silica
and negatively correlated to crude protein. Crude
protein showed negative correlations (P<0.05) with
ADF, ADL and silica. These relationships
concurred with those reported by Da Silva et al. (1965),
i
 57

T a b le 4 . 7 : S i a p l e c o r r e la t io n c o e f f i c ie n t s fo r th e c h e t ic a l components and d i g e s t i b i l i t y

in v itro tor d if f e r e n t growth sta g e s (u n d e rlin e d c o e f f i c ie n t s represent

Bana g r a s s w h ile the o th e rs show F rench Caaeroon).

D.H. O.N. CP CF NDF ADF ADL S IL IC A LEAF:STEH RATIO

D IC E S. DIGES

J9 -.8 5 -.7 8 -.8 7 -.8 8 -.8 8 JO

DH D IG E S 1 J4

CM
CO
-.9 0 -.8 8 .87

1
T IB IL IT Y 1 .96 .9 6 - .9 4 -.8 9

J4 -.9 4 -.91 - .8 6 -.8 6 -.9 1 J9

O . H . d ig e s 1.

.9 4 - .9 3 -.8 9 -.8 6 -.8 4 - .9 3 .87

t ib ilit y 1

1 -.9 4 -.8 7 - .9 4 -.9 0 -.9 5 J5

.94
CD
CO

- .8 3 -.9 3
1

CP 1 -.9 2 -.9 1

JO .92 .82 J l. -. 9 2
t

1 .95 .94 .80 .93 -.9 4

CF

1 J2 J9 JO -.8 6

l .91 .81 .95 - .8 8

HDF

1 J4 J2 - .9 2

1 .84 .94 -.8 9

ADF

1 J4 -.8 6

1 .80 -.7 6
ADL

1 -.9 2

Leaf:stem ratio 1
 58

Table 4.8: Relationship between detergent fibres,

lignin, crude fibre, silica and in vitro
dry matter (DMD) and (OMD) digestibility.

Ca) Bana grass

Y X Regression
2
equation R

DMD NDF Y = 87.76 - 0. 37x 0.79

DMD ADF Y = 85.88 - 0. 62x 0.88

DMD ADL Y = 77.34 - 3 . 65x 0.95

DMD CF Y = 84.21 - 0 . 64x 0.84

DMD CP Y r 55.39 + l.Olx 0.87

DMD Silica Y = 82.52 - 3 . 19x 0.8 2

OMD NDF Y = 1 0 3 . 5 6 - 0. 55x 1.00

OMD ADF Y = 98.57 - 0 . 86x 0.96

OMD ADL Y = 85.40 - 4 . 74x 0.90

OMD CP Y = 56.04 + 1.41 x 0.96

OMD CF Y = 96.46 - 0 . 90x 0.94

OMD Silica Y = 93.89 - 4 . 45x 0.90

 b9

Table A.8: Relationship b e t w e e n

d » t e r * e nt
fibres.
1 1 gnin , crude fibre * 1 1 ics a nd
in vitro
dry matter (DMD) and (OMD) dl
gestlbi1

(b) French C a m e r o o n

Y X Regression
equation R2

DMD NDF Y = 11 A.70- 0.77* 0.95

DMD ADF Y = 9A.83 - 0.86* 0.92
DMD ADL Y = 76.97 - 3. 39* 0.66

DMD CP Y = 55.03 ♦ 1.21* 0.9A

DMD CF Y = 92.77 - 0.8A* 0.95

Si 1ica Y = 105.39- 7. A2* 0.64

DMD
Y = 123.10- 0. 85* 0.97
OMD NDF
Y = 101.36- 0.94* 0.93
OMD ADF
Y = 83.02 - A. 02* 0.79
OMD ADL
0.92
CP Y = 58.01 ♦ 1.31*
OMD
0.94
CF Y = 98.69 - 0.91*
OMD
0.92
Silica Y = 11A.73- 8.AA*
OMD
 60

Gomide et al. (1969) and Pezo (1971). The

correlation between lignin and silica contradicts
the findings of Van Soest (1970) while they concur
with those of Da Silva et al., (1965), and
J ohnson (1973).

These results are in agreement with

observations from not only napier grass but also
other tropical grasses, (Smith et al., 1971;
Coward-Lord et al., 1974; Kayongo-Male et al.,
1976; Abate et al., 1981 and Onyango, 1986).

The fibre components, silica and lignin were

highly negatively correlated with both dry matter
and organic matter digestibility in vitro. The
two napier grass varieties had similar trends.
Silica was positively correlated to NDF, ADF, ADL
and CF while it had high negative correlations
with in vitro digestibility, crude protein and
total ash. Van Soest and Jones (1968) reported
that cell wall silica functions much in the
same manner as lignin by complexing with the cell
wall components and thus reducing digestibility.
The findings tally with those of other
investigators where silica content has been known
to be an important determinant of digestibility
(Van Soest and Jones, 1967; Butterworth, 1967).
4 •5 Dry Matter degradability in sacco

The ruminal dry matter disappearance in sacco

or degradability refers to the proportion of dry
matter that is broken down by rumen microbes after
a given time of incubation. When dealing with
fibrous feeds the dry matter or organic matter
degradation is of great value. As far as napier
grass is concerned there is little published data
available about its degradation characteristics.

The graphs in Figures 4.1 and 4.2 show the

percent ruminal dry matter loss with time for Bana
grass and French Cameroon respectively at
different regrowth stages. The curves were fitted
using t h e ‘means of measurements shown in Appendix
IV. The degradability curves for the two cultivars
showed similar patterns and dry matter loss with
time was higher <P<0.05) the younger the napier

grass was.

The values for the dry matter degradability

equation (Orskov and McDonald, 1979) are presented
in Table 4.9 (a) and (b). These are; DM loss in
the rumen at zero time (a), potential napiergrass
degradability with sufficient time (b) and the
degradability rate (c) of 'b’. It was observed
that as the napier grass matured the dry matter
disappearance rate decreased and this was
reflected by the decline in the ruminal dry matter
loss. In the present study a significant
 62

50-

Incubation time (hrs.)

Fig. 4.1: Ruminal % DM loss in sacco using sheep (Bana grass)

 63

Fig. 4.2: Ruminal % DM loss in sacco using sheep

(French Cameroon).
 64

Table 4.9: DM degradability constants a, b and c

used to fit in the degradation equation

(a) Bana grass

Age
<weeks) a b c

4 18.62 72. 83 0.0144

6 18.50 68. 50 0.0125

8 17.50 67. 33 0.0123

10 15.05 66. 21 0.0124

12 13.00 63. 84 0. 0121

14 10.50 61.42 0.0117

(b) French Cameroon

Age
(weeks) a b c

4 18. 50 71.85 0. 0136

6 16.50 68.85 0. 0129

8 15.00 66.97 0.0120

10 12. 50 64.81 0.0119

12 10.31 62.69 0. 0109

14 10. 00 61.07 0.0102

 65

variation (P<0.05) in dry matter loss with

incubation time and stage of growth was observed
for both cultivars. Degradability was rapid
between 0 and 36 hours after which a steady
leveling off was observed.

The half-life (T y /2* *s shown in Table 4.10.

The difference between half-life of the youngest
napier (4 weeks) and the oldest is over ten hours.
The time taken for half the dry matter incubated
to be broken down was a reflection of the quality
of the grass. Within half life it may be safe to
assume that most of the material useful to the
animal was degraded. The trend observed compares
favourably with the changes in nutritive value as
the napier grass matured. Cell wall components are
known to increase with age and in this experiment
a positive relationship was evident between crude
fibre, detergent fibres, silica and lignin and the
grass age. The remarkable decrease in percent dry
matter loss and the increase in half-life was also
probably due to the decrease in the crude protein
content of the napier grass. Values of crude
protein lower than 6-7% (Milford and Minson, 1966)
are known to impair microbial activity and hence
reduce digestibility and as observed from the
proximate analysis data, the CP content after 8
weeks was lower than these critical limits for
both varieties.

The relationship between the half-life (T^/2*

of dry matter degradation measured in the rumen
 66

Table A. 10: The variation in half life (T j /2 > with

changes in stage of growth.

Ha 1f- life T 1/2 (hours)

Age
(weeks) Bana French

grass Cameroon

4 48. 16 51.03

6 55. AA 53. 72

8 56. 57 57.75

10 57. 32 58.09

12 58. 18 63. 35

1A 59. 18 67.88
 67

Table 4.11: The relationships between half-life

(Tl / 2 ) and the detergent fibres, crude

fibre, lignin and silica.

(a) Bana grass

2
Y X Regression R

NDF Y = 30.45 + 0. 40X 0. 75

T l/2
ADF Y = 32.35 + 0.66X 0. 84
t 1/2
ADL Y = 42.69 + 3.61X 0. 76
T 1/2
Silica Y = 34.54 + 3.68X 0. 89
T 1/2
CF Y = 33.71 + 0.71X 0. 83
t 1/2

(b) French Cameroon

2
Y X Regress ion R

NDF Y = 2.93 + 0.86X 0. 85

t 1/2
ADF Y = 24. 19 + 0.97X 0. 86
T l/2
ADL Y = 40.65 + 4.85X 0. 99
T 1/2
Silica Y = 8.05 ♦ 9. 16X 0. 93
T 1/2
CF Y = 28.23 ♦ 0.91X 0. 80
t 1/2
 68

and the crude fibre, detergent fibres lignin and

silica content at different cutting stages are
shown in Table 4.11 (a) and (b). As the CF, NDF,
ADF, lignin and silica of the forage increased,
the T^/2 f°r degradation in the rumen increased
indicating that the higher the level of fibres,
lignin and silica in the forage, the longer it
took to be degraded. This reflected the extent of
microbial activity which seems to have been
substantially curtailed by the increased levels of
lignin and silica. The trend observed compares
favourably with the results obtained for in vitro
digestibility.

It was observed from the experiment that in

vitro dry matter digestibility and the half life
of the incubated napiergrass were highly
correlated (r = -0.86). The relationship between
half life (Tl/2) and in vitro dry matter
digestibility is illustrated graphically in Figure
4.3. It can be observed that when Tl/2 is less
than 56 hours, the digestibilty is greater than
64%. The linear regressions of T^/2 against the
digestibility coefficients are given in equations
(4) and (5).
 70

Y = 107.55 - 0.78X R2 =0.74 (4)

(r 0 . 86)

for Bana grass and,

Y = 107.56 - 0.74X R2 =0.74 (5)

(r 0 . 86)

for French Cameroon, where,

Y the dependent variable is the in vitro dry
matter digestibility coefficient and X is the half
life (Ti/2^

The greatest deviations in the relationships

occur when fibres, lignin and silica contents are
high and this explains the importance of these
components in defining the nutritive value of
napier grass.
 71

s - O CONCLUS I ON

The following inferences were drawn from the

s t u d y :-

1. The dry matter matter yield, plant height,

ash and percent dry matter were
not significantly different between the
cultivars Bana grass and French Cameroon.
The components increased significantly
with age for the two cultivars under study.

2. Crude protein, ether extract and potassium

decreased with age. The numerical decline
in phosphorus, potassium and calcium
levels were not significant as the grass
matured. The observed levels for calcium
and phosphorus suggested supplementation
for animals feeding on napier grass. In
these, the two varieties were also not
significantly different.

3. The detergent fibres (NDF and ADF), crude

fibre and lignin increased as the grass
matured. No significant differences were
noted between the cultivars for
detergent fibres and lignin. Silica
declined with age while varietal
differences were also significant.

A. Both digestibility in vitro and ruminal

dry matter loss in sacco decreased with
age. A high correlation between in vitro
 72

digestibility and half-life, defined as

the time taken for half the dry matter
incubated to be broken down was observed.

5. Between 4 and 8 weeks cutting stages the

crude protein levels were above the
critical values (6-7%) and digestibility
was high. However at 4 weeks both the dry
matter yield and percent dry matter might
be limiting. Thus 6 to 8 weeks is probably
the appropriate stages to harvest since
beyond 8 weeks crude protein becomes
limiting although digestibility is

reasonably high.
 73

5. 1 SCOPE FOR FURTHER WORK

1. Animal performance studies with napier grass

alone or in combination with other feed

resources (e.g. farm by-products) in terms of

milk yield and liveweight gains.

2. Napier grass conservation methods to cater for

feed deficit periods.

 74

Abate, A., 1978. Dry matter and nutrient intake by

grazing dairy heifers from a
predominantly Chloris gayana pasture.
MSc. Thesis, University of Nairobi,

Kenya.

Abate, A., H. Kayongo-Ma1e , and C.N. Karue, 1981.

Dry matter, protein, energy and fibre
intake by Dairy heifers grazing a Rhodes
(Chloris gayana) pasture. Anim. Feed.
Sci. Tech. 6: 15-26.

Abate, A., H. Kayongo-Male and M. Wanyoike, 1987.

Fodder for h igh-po tent ia 1 areas in

Kenya. J.A. Kateg i 1e, A. N. Sa id and

B. H. Dzowe1a (Eds. ) J_n An im a 1 Feed

Resources for Sma 1 1-Scale Livestock

Producers. IDRC, Nairobi, Kenya. 1985.

Akinola, J.O., J.A. Mackenzie,, and H.R. Chheda,

1971. Effects of cutting frequency and
level of applied nitrogen on
productivity, chemical composition,
growth components and regrowth potential
of three cynodon strains. W. Afr. J.
Biol. Appl. Chem. 14 : 7-12.

Anindo, D.O. and Potter, H.L. .(1985). Milk

production from napier grass in a zero-
grazing feeding system. Kenya
 75

Agricultural Research institute Muguga,

Kenya.

Annenkov B.N.,1982. Mineral feeding in cattle.

1n Mineral nutrition of animals by V. 1.
Georgeovskii, B.N. Annenkov and V . I.
Samokhin, Butterworth, London.

Anon., 1981. National Livestock Development

Policy. Ministry of Livestock
Development, Nairobi, Kenya, 1981.

Anon., 1985(a). Technical Note No. 1 on Animal

Nutrition. Kenya Agricultural Research
Institute (K.A.R.I.), Kenya, 1985.

Anon., 1985(b). Forage plant development and seed

production Kenya. Project Findings and
Recommendations, F.A.O., Rome, 1985.

Arias, P.J., 1980. Some results with elephant

grass (Pennisetum purpureum) in the North
*
Central region of Venezuela. Trop. Anim.
Prod. 5: 281-282.

Association of Official Analytical Chemists, 1984.

Official Methods of Analysis. (16th
edn.) Association of Official Analytical
Chemists, Washington, D.C.
 76

Bogdan, A.V. 1977. Tropical Pasture and Fodder

plants, Longman, London.

Boon m a n , J.G., 1977. Fodder farming in East

Africa. A preliminary report on the
status of the development of tropical
pastures in East Africa (Part II).
Ministry of Agriculture, Nairobi, Kenya.

Brown, D.L. and E. Chavalimu, 1985. Effects of

ensiling or drying on five forage
species in Western Kenya. Zea mays
(Maize stover), Pennisetum purpureum
(Pakistan Napier grass), Pennisetum spp.
(Bana grass), /pomea batata (sweet
potato vines)) and Cajanus cajan (Pigeon
Pea leaves). Anim. Feed Sci. Tech. .13.:

1- 6 .

Burton, G.W. and J.E. Jackson, 1962. Effect of

rate and frequency of applying six
nitrogen sources on coastal Bermuda
grass. Agron. J. 54: 40-43.

Butterworth, M.H. 1965. Some aspects of the

utilization of tropical forages. 1.Green
elephant grass at various stages ot
regrowth. J. Agric. Sci. (Camb.) 65.:

233-239.
 77

Buttervorth M.H. 1967. The digestibility of

tropical grasses. Nutr. Abstr. Rev.
3 7 : 349-368.

Caro-Costas, R. and J. V icente-Chand 1e r 1956.

Comparative production of meur1ler grass
and of Kudzu-mueller grass mixtures as
affected by season and height of
cutting. J. Agric. Univ. Puerto Rico,
4 0 : 144-149.

Caro-Costas, R. , J. Vicente-Chand1er and J.

Figarella 1960. The yields and
composition of five grasses growing in
the humid mountains of Puerto Rice as
affected by nitrogen fertilizers, season
and harvest procedures. J. Agric. Univ.
of Puerto Rico. 4 4 : 107-120.

Caro-costas, R. and J. Vicente-Chandler 1961.

Effects of two cutting heights on yields
of five tropical grasses. J. Agric.
Univ. Puerto Rico, 45: 46-49.

Chuldleigh, P.D. 1974. A description of small­

holder ruminant production systems in
the high potential areas of Kenya.
Pasture Research Project No. 4.
National Agric. Res. Station, Kitale,

Kenya.

Cloutier, J.E. 1971. Agronomic factors in pasture

 78

and forage crops management in tropical

Australia. Trop. Grassl. 5: 245-254

Cooper, J.P. and N.M. Tainton, 1968. Light and

temperature requirements for the growth
of tropical and temperate grasses. Herb.
Abstr. 38: 167-176.

Coward-Lord, J., J.A. Arroyo-Agui11u and 0.

Garcia, Molinavi 1974. Proximate
nutrient composition of ten tropical
forage grasses. J. Agric. Univ. Puerto
Rico, 58: 305-311.

Crowder, L.V. (1977). Potential of tropical zone

cultivated forages. In Potential of the
worlds Forages for Ruminant Animal
Production, Winrock Livestock Centre,
Arkansas, pp. 49-79.

Crowder and L.V. and H.R. Chheda 1982. Tropical

grassland husbandry. Longman, New York.

Da Silva D.J.,J .H.Conrad and J.Campos (1965). In

vitro digestibilities of some tropical
forage species. Proc. IX International
Grassland Congress, Sao Paulo, 307-312

(Vo 1. 1) .

Di rven, J.G.P. and V.K.R. Ehrencron, 1963. Dry

matter percentages of grasses in the

hum id tropics. Surin. Land Bull. 11 : 88-

 79

93

Dovrat, A.J.G.P. Dirven and B. Deinum (1971). The

influence of defoliation and nitrogen on
the regrowth of Rhodes grass iChJoris
gayana Knuth). Dry matter Production
and Tillering. Neth. J. Agric. Sci. 19:
74-101.

Fina, L.R., G.W., Teresa, and E.E. Bartley, 1958.

An artificial rumen technique for
studying rumen digestion in vivo. J.
An im. S c i . 17: 667-674.

Fisher, M.J. (1973). Effect of times, height

and frequency of defoliation on growth
and development of stylo in pure ungrazed
swards at Katherine, N.T. Aust. J.
Explt. Agric. Anim. Husb. 1_3: 389-397

Folkertsma, R. 1981. The nutritive value of

* Pennisetum purpureum (Elephant grass).
Hogeschoo1, Wageningen, Netherlands.

Fontenot, J.p., M, . B 1ui se and K.E. Webb, J r.

1973. Interrelationship of potassium,

nitrogen and magnesium. Fed. , Proc. 32:

 80

1925-1928.

Fren ch, M.H. 1957. Nutritional value of tropical

grasses and fodders. Herb. Abstr. 27:

1-9.

Georgievskii, V.I. 1982. Biochemical regions.

Mineral composition of feeds. J_n Mineral

n u t r it i on of animals. V.I. Georgievskii

B.N. Annenkov and V.I. Samokhin.

Butterworth, London.

Goering, H.K. and P.J. Van Soest, 1970. Forage

Fibre Analysis (Apparatus, Reagents,

Pr ocedures and Some Applications).

Agricultural Research Service, U.S.

Department of Agriculture. Agricultural

Handbook No. 379: 1-12, Washington, D.C.

Glover, J. and H.W. Dougall i960. The apparent

digestibility of the non-nitrogenous
components of ruminant feeds. J. Agric.

Sci. (Camb.). 55.: 391-394.

G o h 1, B. (1981). Tropical Feeds. Food and

Agricultural organization of the United

Nations, Rome, Italy.

Go 1dson, J.R. 1976. Pasture, Fodder and Animal

Husbandry Problems on the small-holder
(arm In seleoted high potential
Pasture Research
districts of Kenya.
 81

project Rept. No. 4. National

Agricultural Research Station, Kitale,
Kenya.

Goldson, J.R. 1977. Napier grass in East Africa

A review, Pasture, Research Project
(Technical Report No. 24), Ministry of
Agriculture, Kenya.

Gomide, J.A., C.H. Noller, G. 0. Mott, J.H. Conrad

and D.L. Hill (1969). Effect of plant
age and nitrogen fertilization on
chemical composition and in vivo
cellulose digestibility of tropical
grasses. Agron. J. 61_: 116-120.

Gosnell, J.M. 1963). Vleiland development in West

Kenya. 1. Pasture introduction. E.
Afric. Agric. For. J. 29.: 99-105.

Guerrero J.and D. Pezo, 1973. Cell wall

constituents in in vivo digestibility of
napier grass, (Pennisetum purpureum),
J. Anim. Sci. 37: 1262-1267.

Gutierrez, L.E., V. Faria and P. De (1978).

Effect of maturity on the mineral (Ca
and P) and protein content of 4
cultivars of elephant grass (P
purpureum). Herb. Abstr. (1979). 49:
2632-2632.
 62

Holmes, U. I960. Grass: Its production and

utilization. Blackwell. London.

J ackson, M.G., 1977. Review Article: The alkali

treatment of straws. Anim. Feed. Sci.
Tech. 2: 105.

J a e t z o 1d, R, and Schmidt, H. 1982. Farm

management Handbook of Kenya Vol. II.
Part B, Central Kenya, Ministry of
Agriculture, Kenya.

Johnson, W.L., J. Guerrero, J. and D. (1973).

Cell wall constituents and in vitro
digestibility of napier grass
(Pennisetum purpureum). J. Anim. Sci.

37: 1255-1261.

Jones, L.H.P., and K.A. Handreck, 1967. Silica in

soils, plants and animals. Adv. Agron.
19: 107-112.

Jones, D.l.H. (1971). The chemistry of grass for

animal production. Report of the welsh
Plant Breeding Station, 1971.

Karanja, G.M. 1981. Bana grass (Pennisetum

purpureum) c.v. Bana. Kenya Farmer, No.

3. 50-52.

Karanja, G.M. (1984). The effects of frequency

and height of cutting on tillering,
 63

herbage yield and nutritive value of

four varieties of napier grass. M.Sc.
Thesis, University of Nairobi.

Kayongo-Male , H., J.U. Thomas, D.E. Ullrey, R.J.

Deans and J.A. Arroyo-Agui1u (1976).
Chemical composition and digestibility
of tropical grasses. J. Agric. Univ.
Puerto Rico, 6JD: 186-200.

Kimenye, D. 1978. A genetic study of the Kenya

Sahiwal Breed. Ph.D. Thesis, University
of Nairobi.

Koch, D.U. 1987. Animal Evaluation of forages

following several methods of field
renovation. Agron. J. 7J3: 1044-1048.

Kusewa, P.K.; A .B . Orodho; A.G.O. Okech; J.E.E.

Kevelenge; K.R.G. Irungu and C.E.M.
Ndirangu, 1983. Napiergrass management,
Production and Utilization in Kenya. A
report of a review mission, National
Agricultural Research Station, Kitale,

Kenya.

Leon, J. , G.l. Barra, N .A . Costa, and M. Flores

(1987). Nutritive value of King grass.1.
Bromato1ogica1 composition and in vitro
digestibility. Herb. Abstr. 5 7 : 11-15.

Macleaod, and Minson 1969. Sources of variation

 84

in the in vitro digestibility of

tropical grasses. J. Brit. Grass 1. Soc.

24: 244-249.

Ma poon, •K., 1980. Degradability of some high

protein forages in the rumen. Trop.

Anim. Prod. 5: 53-56.

Maynard, L.A.; J.K. Loosli; H.F. Hintz and R.G.

Warner, 1979. Animal Nutrition (7th

edn. ). McGraw-Hill, New Delhi.

M i nson, D.J. J.H. Stobbs, M.P. Hegarty and M.J.

Playne, 1976. Measuring the nutritive
value of pasture plants. In N.H. Shaw
and U.U. Bryan eds), Tropical Pasture
Research Principles and Methods.

Hurley, England.

Miche1i n, A.A., J. Ramirez, Lotero and H. Chaverra

(1968). Effect of nitrogen application
on coastal bermuda and pangola grass.
Agric. Trop. (Colombia) 24: 282-285.

Mi 1 ford, R. and D.J. Minson, 1966. The feeding

value of tropical pastures. J_n Tropical
Pastures by W. Davis and C.L. Skidmore

(eds), Faber and Faber, London.

Miller , T.B. and Blair-Rains 1963. The nutritive

value and agronomic aspects of some
fodders in northern Nigeria 1. Fresh
 65

herbage. J. Brit. Grassld. Soc. 18:

158-167.

1 tavo, I•# J.V. Viega, L.R. McDowell, W.R.

Ocumpaugh and G.O. Mott. Mineral
content of dwarf Pennisetum purpureum
under grazing conditions Nutr. Reports
Int., 35: 157-169

^ l-J 1 d o o n D.K. and C.J. Pearson (1977). Hybrid

Pennisetuw in a warm temperate climate:
Productivity span and effect of nitrogen
fertilizer and irrigation on summer
production. Aust. J. Expt. Agric. Anim.
Husb. 17: 982-990.

M u 1doon. D.K. and Pearson (1979). Morphology and

physiology of regrowth of the tropical
tall grass hybdrid pennisetum. Ann. Bot.
43: 719-728.

Mwakha, (1972). Effect of cutting frequency on

productivity of napier and Guatemala
grasses in Western Kenya. E. Afr. Agric.
For. J., 37: 206-210.

Nat iona1 Research Council. 1978. Nutrient

requirements of dairy cattle. 5th
Revised edition, 1978. Nat. Acad. Sci.
Washington D.C.

Newton, G.L., J.P. Fontenot, R.E. Tucker and C. E.

 86

P o 1an. 1972. Effects of high dietary

potassium intake on metabolism of
potassium in sheep.
J. Anim. Sci. 35.: 440-447.

QcJ iambo, J.F. (1974). The nutritive value of

various grwoth stages of Pennisetum
purpureum. E, . Afr. Agric. For. J. 39.:
325-329.

O gwang, B.H. and J.S. Mugerwa, (1976). Yield

response to nitrogen application and in
vitro dry matter digestibility of
elephant grass and bulrush millet
hybrids. E. Afr. Agric. For. J., 4_L:
231-242.

01 sen, F.J. and G.L. Santos (1975). Effects of

lime and fertilizers on natural pastures
in Brazil. Expt 1. Agric. 1_1_: 173-176.

0 1 uba jo , F.O. and V.A. Oyenuga (1974). The yield,

intake and animal production of four
tropical grass species grown at Ibadan.
Niger. J. Anim. Prod. 1_: 217-224.

Orskov .R.and 1. McDona1d ,1979. The estimation of

protein degradability in the rumen from
incubation measurements weighted
according to rate of passage. J. Agric.
Sci., (Camb.), 92.: 499-503.
 87

° r s k o v E. R - , D.F.D. Hovel 1 and F. Mould, 1980. The

use of the nylon bag technique for the
evaluation of feedstuffs. Trop. Anim.
Prod. 5_1_: 195-213.

° r- s kov E.R., 1986. Analytical methods for

characterising feed resources for
ruminants. ]_n Better utilization of by­
products in animal feeding: Research
guidelines- 2. A practical manual for
research workers. FAO, Rome.

Os man, i.E . (1979). Productivity of irrigated

tropical grasses under different
clipping frequencies in the semi-desert
region of the Sudan. J. Range Mgmt, 32.:
182-185.

Onyango, T.A. (1986). Feeding value of forage

sorghum "E6518" silage at different

stages of growth. M.Sc. Thes is

University of Nairobi.

Oyenuga, V.A. (1959) Effect of stage of growth

and frequency of cutting on the yield
and chemical composition of some
Nigerian fodder grasses 1. Pennisetuw
pur pureum Schum. J. Agric. Sci. 53^:2-

33.

O y e n u g a , V.A. (1960). Effect of stage of growth

and frequency of cutting on the yield
 88

and chemical composition of some

Nigerian fodder grasses - Panicum
maximum Jacq. J. Agric. Sci. (Camb.), 55_
: 339-50.

Palacpac, O.S. (1985). Phi 1ippine-Japanese

venture finds market for elephant grass
is booming Agri-Business Worldwide,
Magazine, Phi11ippines.

Pa retas, J.J., Gomez, L. (1972). Influence- of

cutting frequency and irrigation on the
calcium and phosphorus content of
elephant grass (P. purpureum). Herb.
Abstr. (1979), 49:795.

Paterson, D.D. (1933). The influence of time

cutting on the growth, yield and
composition of Tropical fodder grasses.
I. Elephant grass (Pennisetum purpureum)

J. Agric. Sci., 23: 615.

Pa terson. D.D. (1938). Further experiments with

cultivated tropical fodder crops. Emp.
J. Exp. Agric. 6: 323-340.

Potter , H.L. (1985). Bana grass for the small-

scale farmer in Central Kenya. Kenya,
Agricultural Research Institute, 1985.

Ranjhan, S.K. and S.K. Talapatra (1967). Tield

chemical composition and nutritive value
 89

of giant napier grass. Indian J. Dairy

Sci . , 20: 134-138.

Raymond, U.F. (1969). The nutritive value of

forage crops. Adv. Agron. 21_ : 100-108.

Reid, R.L., A.J. Post, F.J. Olsen and J.S. Mugerwa

(1973). Studies on the nutritipnal
quality of grasses and legumes in
Uganda. I. Application of in vitro
digestibility techniques to species and
stage of growth effects. Trop. Agric.
(Trin.) 50: 1-15

Rivera-Brenes, L. , J. Torres and J. Arroyo

(1961). Response of guinea, pangola and
coastal bermuda grasses to different
nitrogen fertilization levels under
irrigation in Lajas Valley of Puerto
Rico. J. Agric. Univ. P.R. 45.: 123-46.

Robertson, A.D., L.R. Humphreys and D.G. Edwards

(1976). Influence of cutting frequency
and phosphorus supply on the production
of stylosanthes pusilla at Khon Kaen,
Northern east Thailand. Trop. Grass 1.,
10: 33-39.

Said, A .N . (1976). Evaluation of feeds. Some

results from the study of the nutritive
value of Kenya grasses and fodder crops.
Conference, National Agricultural
 90

Research Station, (Ed. C.T. Thomas),

Kitale, Kenya.

Said, A.N. and M.M. Wanyoike (1987). The

prospects of utilizing urea-treated
maize stover by smallholders in Kenya.
1n Utilization of Agricultural By­
products as livestock feeds in Africa.
J.A. Kategile and A.N. Sa-id (Eds.)
1LCA, (ARNAB), Addis Ababa, Ethiopia.

Singh R., (1975). Chemical composition of cell

wall and in vitro DM and cell wall
digestibility of some important grasses.
Indian J. Vet. Sci. and Anim. Husb.
46.: 339-344.

Strange, R. (1959). Perennial late season

pastures. File report National
Agriculture Research Station, Kitale,
Kenya.

Tessema, S., 1975. Nutritional value of some

tropical species compared to some
temperate grass species. Herb. Abstr.
7: 218-218.

Thairu, D.M., S. Tessema (1987). Research on

animal feed resources. Medium-Potential
areas of Kenya. J.A. Kategile, A.N.
Said and B.H. Dzowela (Eds.) 1n Animal
Feed Resources for Small-Scale Livestock
 91

Producers. IDRC, (PANESA), Nairobi,

Kenya.
Thairu, D.M., J. Morrison and N.C.O. Keya (1968).
Annual Report. National Agricultural
Research Station, Kitale, Kenya. Vol.
11. page 21.

Thornton, R.F. and D.J. Minson, 1973. Effect of

soils, fertilizers and stocking rates on
pasture and beef production on W a 1lum of
South east Queensland. Aust. J. Exptl.
Agri. Anim. Husb. , 1^3: 537-43.

Tiley, G.E.D. 1960. Elephant grass in Uganda,

Unpublished mimeogragh, Department of

Agriculture, Uganda.

Tilley, J.M.A., and R.A. Terry, (1963). A two

stage technique for the in vitro

digestion of forage crops,. J Brit.

Grass 1. Soc. 1J3: 106-11, 1963

Van Coppeno11e R.(1983). Fiches Phytotechniques.1.

Penn is e turn purpureum 2. Tripsacum laxum,
3. Setaria splendida. Burundi Publication
No. 23, Bujumbura, Burundi.

Van der Kamp, A. (1987). Napier grass in Kenya.

A review, Dairy Development Project,
Ministry of Agriculture and Livestock
Development, Naivasha, Kenya.
 92

Van Es A.J.H. (1982). In recent advances In

Nutrition, William Haresign <ed.).

Buttervorth, London.

Van Gas t e 1, A.J.G. (1978). Napier grass breed ing

at National Agricultural Research

Station, Kitale, Kenya. Technica1

Report, Grass Breeding Project.

Van Soest, P.J. and L.H.P. Jones, 1968. Effect of

silica in forages upon digestibility.

J. Dairy Sci. 51: 1644-8.

Van Soest, P.J. (1965). Use of detergents in

analysis of fibrous feeds. IN. Studies
of effects of heating on yield of fibre
and lignin in forages. J.Ass. Off. Anal.

Chem. , 48.: 785-90.

Van Soest, P.J. and F.E. Lovelace, (1969;.

Solubility of silica in forages. Abstr.

J. Anim. Sci,. 29.: 182.

Vicente-Chandler, J.. S. Silva and J. Figarella

(1959). The effect of nitrogen and
frequency of cutting on the yield and
chemical composition of three tropical

grasses. Agron. J. §1: 457-63.

Vicente-Chandler, J., F. Abruna, R.

Caro-Costas,
J. Figarella, S. Silva and R.W. Peason,
 93

(1974). Intensive grassland management

in the humid tropics of Puerto Rico.
Bulletin 233 University of Puerto Rico.

^ a t k ins, J.M. and M. L. Lewy-Van Severen, 1951.

Effect of frequency of cutting on the
yield, stand and protein content of some
forages in El. Salvador. Agron. 4J3:

291-296.

Ware-Austin, W.D. (1963). Napier grass for milk

production in the Trans Nzoia. E. Afr.
Agric. For. J. 28: 223-227.

Whitney, A.S. and R.E. Green, (1969). Fangola

grass performance under different levels
of fertilization in Hawaii. Agron. J .6_1_

: 577-81.

U i 1k inson, S.R. and G.W. Langdale, (1974).

Fertility needs of warm-season grasses.
In D.A. Mays (ed.). Forage
Fertilization, Madison Wisconsin.

Williams, C.N. (1980). Fertilizer response of

napier grass under different soil

cond it ions in Brunei. Exp. Agric. 16.:

415-423.

Wilsie, C.P. and M. Takahashi, (1934). Napier

grass (Pennisetum purpureum) a pasture
and fodder crop for Hawaii. Bull. Hawaii
 94

Agric. Exp. Stn., No. 72.

Wi l si e, C.P., E.K. Akamine and M. Takahashi

(1940). Effect of frequency of cutting
on the growth, yield and composition of
napier grass. J. Amer. Soc. Agron.

32: 266-273.

Uouters,
A.P. (1985). Fodder production for zero
grazing, recent research findings. Anim
Prod. Soc. Kenya, Proc. Kabete, Kenya.

A.P. (1966). The productivity

Vlouters,
quality of napier grass <var. Bans
grass) during the dry season at three
levels of nitrogen application.
Ministry of livestock Development, Dairy
Development Project. Naivasha, Kenya.
 95

5 . O
APPENDICES
Appendix l: ANALYSES OF VARIANCE TABLES

1 .1

Mean Squares
Sources of E rro r
v a r ia tio n df lin e DND OMD CP

V a rlety 1 a .90 .50 6 .5 7

'•ar i e t y : T re a t m e n t (E rro r l) 10 b 61.0 7 * 81.00* 4 2 .9 8 *

Replication 2 b .58 4.19 .53

Residual ( E r r o r 2) 22 1.58 1.71 .1 7

Appendix 1.2

Mean Squares
S o u rc e s of E rro r
v a r ia t io n df lin e NDF ADF ADL

V a r ie t y 1 a 7 .41 .35 .0 6

V a r ie t y :T r e a t a e n t ( E r r o r 1) 10 b 1 7 8 .0 6 * 9 3 .5 4 * 3 .7 9 *

R e p lic a t io n 2 b 1 .3 0 3 .1 8 .2 2

R e s i d u a l ( E r r o r 2) 22 1 5 .0 8 3 .7 8 .1 6

Key:
DHD = In vitro Dry N a tte r D i g e s t i b i l i t y
OMD - I n vitro O rg an ic M a tte r D i g e s t i b i l i t y
CP = Crude P ro t e in
NDF = N e utra l D etergent F ib r e
ADF = A c id D etergent F ib r e
ADL * A cid D etergent L i g n i n
a = E rro r 1
b = E rro r 2
• = P<0.05
 96

*g p g n d ir 1 .3

Mean Squares
Sources of E rro r
ASH EE
v a r ia t io n df lin e CF

38.23 89.59 .70

V a rie ty 1 a

22.74* 1.10 *
b 9 2 .5 5 *
V a r ie t y r T r e a t ie n t (E rro r 1) 10

7 .5 0 * 4.79* .10*
R eplication 2 b

1.55 1.13 .02

Residual ( E r r o r 2) 22

A p p e n d ii 1.4

Mean Squares

E r ro r
So u rce s of p
lin e K Ca
v a r ia t io n df

0 .00 0 .0 0
3 .1 5
V a r ie t y 1
| 1
0.01 0 .0 0
b 2 5 .3 0 *
V a r i e t y •.‘fre a ta e n t ( E r r o r 1) 10

0 .0 0 0 .0 0
h 0 .0 0
R e p lic a t io n L
0 .00 0 .0 0
.11
R e s id u a l (E rro r 2) 22

Key:
CF 1 Crude F ib r e
E E 1 E th e r E x t ra c t
K = P o t a s s iu i
Ca * C a lc iu n
P = Phosphorus
a : E rro r 1
b = E rro r 2
i = P<0.05
 97

(lean Squares
urces of
E rro r
Elation df lin e Si DHY L :S

u r ie t y .25
1 a 14.03* 2 .9 6

'* r t e t y : T r e a t a e n t ( E r r o r 1) 10 b 2.19 * 2 5.74* 5 .8 8 *

R e p lic a t io n 2 b .02 1.00 .04

R e s id u a l (E rro r 2) 22 .05 .67 .12

*?T>endi» 1.6

Mean Squares

S o u rc e s of E r ro r
lin e CH HFE DHP
v a r ia t io n df

10837.16 0 .00 8.38

V a r ie t y 1 a

6913.01* 0.00 * 2.85*

V a r i e t y : T r e a t i e n t ( E r r o r 1) 10 b

4.79 0 .00 .01

R e p lic a t io n 2 b

100.38 0.00 .15

f e e s i d u a l ( E r r o r 2) 22

Key:
Si = S ilic a
D flY = D ry (la tte r Y ie ld
L :S = L e a f : S t e i r a t io
C H = G r a s s H e igh t a t C u t t in g
HFE = N it r o g e n Free E x t ra c t
D H P = P e rc e n t D ry Hatter co n te n t
a = E rro r 1
b = E rro r 2
• = P<0.05
 98

Analysis of variance tables for

ruminal dry matter loss
*> B,

=au r C e
df ss ms Fca 1
:oti
107 15066.52
ll ( she e p ) 2 511.00 255.50 13. 50
r e a tments 35 13230.68 378.02 19.97
CV

C u b a t i o n time 5 11222.87 2244.57 118.57

o f cutting 5 1314.68 262.94 13. 89
x S 25 693. 13 27. 73 1.46
=> r o r 70 1324.84 18.93

^ * F r e n c h Cameroon

Source df ss ms Fca l

Total 107 16063.07

3 1 o c k s ( sheep) 2 74.81 37.41 1.60
T r eatments 35 14332.00 409.49 17.31

I n c u b a t i o n time 5 11812.69 2362.54 99.85

S t a g e of cutting 5 2074.81 414.96 17.54
I x S 25 444.29 17. 77 0. 75
E r r or 70 1656.26 23.66
 99

Appendix III; Proximate composition of the feed

offered to sheep under the nylon
bag degradability trial

1. Rhodes grass (.Chloris gayana)

Chemica1 consti tuent

XDM CP CF EE NFE ASH

86.2 7. 3 36.0 1.8 43. 1 11.8

2. Dairy mea 1

Chem ic a 1 const ituent

%DM CP CF EE NFE ASH

91.5 15.6 17. 4 5.9 51.9 9.2

 100

Appendix IV: Mean values of percent ruminal dry

matter loss of napier grass at
different stages of growth and
varying incubation times.

(a) Bana grass

Age in weeks
Time of 4 6 8 10 12 14
Incub­
ation
(hrs)

0 18.83 18.50 17.50 15.05 13.00 10.50

12 38.50 36.27 35.24 33.08 30.04 27.57
24 54.66 52.31 48.21 44.40 43.10 40.56
36 63.42 57.35 54.94 52.98 51.63 50.13
48 66.93 63.40 62.20 61.05 59.03 55.74
60 69.79 65.47 63.09 62.05 61.52 60.28
72 71.09 67.05 66.22 65.33 62.84 60.28

(b ) French Cameroon

0 18.50 16.50 15.00 12.50 10.31 10.00

12 43.32 38.44 36.92 33.81 32.54 28.91
24 57.56 50.00 48.26 47.44 46.46 42.87
36 63.87 59.55 57.32 54.49 52.50 49.50
48 67.50 64.31 60.79 58.11 55.20 52.68
60 69.85 66.20 63.80 62.12 59.52 56.24
72 70.35 67.14 64.83 64.02 60.69 59.07
/