Nutrition of draught oxen in semi-arid west Africa. 1....
Nutrition of draught oxen in semi-arid west Africa. 1. Energy
expenditure by oxen working on soils of different consistencies
A. Fall’, R. A. Pearson’ artd P. R. Lawrence’t
‘Cer~frcfi~r Tropicnl Veferitmy Mediciw, Unizwsit!! (11 Edinbuqh, Emter Bush, Rosli,l, Midlofhim EH25 9RG
‘Iuferwafioml Licestock Reseorch Zmfifufc, Zntcrrrntmnc71 Crop Rcw~h Irzstifutefiu flw Smi-Avid Tropics, Sahelim
Centre, BP 12404, Ninnrey, Niger
Abstract
The Oxylog, CI portriblr brcntli-by-brPntlz gris nnal~~scr, 7~7s urcd 011 sczm animnls to dctrrnzitze stnndiq nlctnbolic
rate, cnergy cosf of walking on soils of different consistemies md e$ïcimcy of ruork ploz@iq nnd cnrting. Thc
merngr standing nlrfnbolic nzte of nnimnls wns 5.63 (se. 0.12) W/kg M”.7S. Thc comisfcmy of the soi1 011 ruhich
mirmls worked hnd o rnarked +ct ou fheir eucrgy cost of zunlkiq which 71x1s 1.59 (s.c. 0.069) 011 wlplo@cd soil,
2.15 (S.E. 0.084) ou plo~$cd soi1 nnd 2.0 (‘se. 0.10) ]/I?L pcr kg livc wcight 011 latcrife frncks. The eflcimcy qf
plow$ing sandy soils (i.c. rrrfio of zoork doue fo cmqy usrdfor work) 7017s 0.32 nnd ms nef sigrt$m~tly d$&rlt
frm the e mwc
J’ o
y CRYf in
f 5 wifh
A
different loads. Thc tIfficiency qf doilzg work zuas rzof irzj7rlenced by the type of
zuork perfomed, the drnugh f @ce excrted or the walkinl; speed.
Keywords: drnught anin&, rrzergy expcnditurc, zvork.
Introduction
preparation and timely planting at the onset of the
Draught animal powrr was introduced in
rainy season. In these situations draught animal
sub-Saharan Africa during the last 70 years and its
power becomes critical to supplement human energy
use is spreading (Starkey, 1994; Panin and Ellis-
SO that field operations cari be done at the right time
Jones, 1994). However, the contribution of draught
to reduce the risk of trop failure and to secure a
animals tu the power requirements for agriculture is
stable yield.
still limited. Agricultural production in this region
continues to rely primarily on human power.
Adcquate feeding to meet the nutrient requirements
Statistics in 1987 suggested that proportionately 0.89
of draught animals is a major constraint in semi-arid
of power was provided by humans while draught
areas, because food is scarce and of poor quality
animals supplied only 0.10 of the farm power input
during many months in the year. In these areas there
(Food and Agriculture Organization, 1987). There is
is a need for a rational planning of the feeding of
a need to promote draught animal power in
draught animals to supply sufficient draught animal
sub-Saharan Africa to fil1 the gap between the
power for trop production. This requires information
deteriorating level of food production and the
on the seasonal availability and the nutritive value of
increasing demand for food. This is particularly true
existing food resources, the utilization of these foods
in semi-arid areas where timeliness in cropping
by draught animals and the nutrient requirements
operations is fundamental for successful cropping
for work. Information on the energy requirements
because of the short growing season in these areas.
for work and food utilizntion of draught animals in
The low and erratic rainfall regime constrains land
semi-arid West Africa is limited. However, the recent
adaptation of portable equipment to measure oxygen
consumption
(Lawrence et nl., 1991) cari contribute
greatly to the knowledge of the energy expenditure
t Present address: Centre for Agriculture in the Tropics
of animals performing common farm operations.
and Suhtropics. Institute of Animal Production,
Department of Animal Nutrition and Aquaculture.
In the absence of direct measurement of oxygen
Universit2it Hohenheim (%Xl), D-7093, Stuttgart, Germany.
consumption,
the extra energy used to perform
---
-_.- .-
210
Fall, Pearson and Lawrence
different activities cari be estimated using the
the same diameter and turbine stator as the
factorial method developed by Lawrence and
original flowmeter but no rotor or electric
Stibbards (1990). This method integrates, additively,
connections.
They
t h e r e f o r e h a d t h e same
the energy cost of walking, carrying and pulling
resistance to air flow as the original flowmeter and
loads. The energy cost of pulling is fairly constant
by uncovering one or more of them, the range of
when expressed in relation to tractive effort and
the flowmeter could be increased from its original
distance. Therefore, this cari be accurately predicted
value of 0 to 80 l/min to 0 to 160, 0 to 270 or 0 to
if work output is known. Energy cost of walking,
3 6 0 l/min. T h e r a n g e used i n a p a r t i c u l a r
which cari account for proportionately 0.50 or more
experiment depended on the expected maximum
of the total energy expended for work (Lawrence and
respiration rate. The flowmeter was calibrated over
Becker, 1994), is more difficult to predict because it is
a11 four ranges using a reciprocating air pump as
dependent on ground surface and needs to be
described by Dijkman (1993). Expired air passed
determined directly. For instance, Dijkman (1993)
from the mask via three 25-mm diameter outlet
found in the sub-humid zone of Nigeria that the
valves into a flexible tube attached to the analysis
energy cost of walking (E,) in cattle ranged from
and display unit fixed to the animals’ back. The
8.58 J/m per kg M on ploughed waterlogged rice
Oxylog was used to take samples of inspired and
fields to 1.47 J/m per kg M on unploughed upland
expired air at every breath, and determined the
soils. The objective of this study was to investigate
difference
in oxygen concentration using two
the energy cost of walking (E,) and ploughing on
matched
polarographic electrodes. This was
sandy soils and of carting in semi-arid areas.
multipiied by the volume of air inspired to give
oxygen
consumption.
The
apparatus
made
corrections for atmospheric pressure, temperature
Material and methods
and humidity and displayed the results as oxygen
Animals and feeding
consumed a n d a i r i n s p i r e d (1) corrected t o
This experiment was conducted from October to
standard temperature and pressure. In the present
November 1994 at the International Crop Rcsearch
experiment these values were recorded every min
Institute for the Semi-Arid Tropics (ICRISAT)
from an auxiliary display pane1 connected by table
Sahelian Centre at Sadore, Niger. Seven Dinli @os
to the Oxylog which enabled data to be read more
indicus) cattle, average live weight 367 (s.e. 22) kg,
easily.
aged 5 to 7 years, were used in the study. The
animals were given food to about maintenance level
Work output, distance travelled and time spent
on natural pastures supplemented with wheat bran
working were measured using an ergometer for
and a minera1 mixture (in total about 9 to 10 g/kg
work and distance, or odometer for distance only
Mo’ 75 per day). The animals had access to water ad
(Lawrence and Pearson, 1985).
libitum during the periods when they were not
working. Mean ambient temperature and relative
The animals used in this experiment were already
humidity were 3O.P”C and 0.627, respectively when
well trained for work. Further training was necessary
animals worked in the morning and 36.5”C and 0.242
to accustom them to carrying the instruments.
when work took place in the afternoon.
Animals were trained for 3 weeks to wear the face
mask and to carry the backpack containing the
Experimcnfal tnetkods
ergometer and the Oxylog while performing
Oxygen consumption
was measured using the
common farm activities.
Oxylog, a portable breath-by-breath analyser. This
instrument (P. K. Morgan Ltd, Kent, England), was
Two trials were conducted. The first tria1 was
originally designed for use with humans and was
designed to determine the energy cost of ploughing
modified for oxen in a manner similar to that
sandy soils using a mouldboard plough. The second
described by Lawrence ef al. (1991).
tria1 measured the energy cost of carting. Light carts
with pneumatic tyres were used. In both trials the
The apparatus consists of an airtight face mask
standing metabolic rate and the energy cost of
with inlet and outlet valves and an analysis and
walking were determined. The treatment applied for
display unit. Air is sucked into the mask through a
the measurement of the energy cost of walking was
cylinder 80 mm long X 100 mm diameter mounted
the consistency of the surface: unploughed wet
on the right side. At the end of this tube near the
sandy soils, ploughed wet sandy soils and firm
mask a plate was fitted with three 25-mm diameter
laterite tracks. For the determination of the
inlet valves. At the other end there was a similar
efficiency of doing work carting, the treatment was
plate which contained the original Oxylog turbine
the load applied (300, 600 and 900 kg). The
flowmeter (Humpfrey and Wolff, 1977) and three
experimental design for both the tria1 on the energy
‘dummy’ flowmeters. The dummy flowmeters had
cost of walking and for the tria1 on the efficiency of
Nutrition of draught cattle-1
211
carting consisted of a random assignment o f
according to Lawrence and Stibbards (1990): Ej =
treatments in sequencc to each ox (block) with
(work done (kJ)/(energy expended when loaded (kJ)
repeated measures.
- energy expended (kJ) to walk the same distance at
the same speed but unloaded).
During the first tria1 the work routine of the six
animais included the following sequence o f
The information obtained from this experiment was
activities: standing for 15 min in the shade (SMR),
incorporated into a factorial formula (Lawrence
walking unloaded for 15 min on unploughed soils,
and Stibbards, 1990) to predict the extra net
walking unloaded for 15 min on previously
cnergy required for ploughing sandy soils for 1 to
ploughed soils and ploughing for 20 min.
6 h/days:
For each activity, measurements were taken whcn
E = A.F.M + B.F.L + W/C + 9.81 H.M/D
animals had reached a steady rate of oxygen
consumption after having worked for at least 5 min.
where: E = extra energy used for work (kJ), F =
The Oxylog was alternately attached to each animal
distance travelled (km), M = live weight (kg), L =
during each work routine. During the walking
load carried (kg), W = work done whiist pulling
sessions, the ndometer of the ergometer was wheeled
loads (kJ), II = distance moved vertically upwards
behind the animal to measure distance walked and
(km), A = energy used to move 1 kg of body weight
time spent walking. Animals were allowed to rest for
1 m horizontally (J), B = cnergy used to move 1 kg of
15 min between bouts of work.
applied load 1 m horizontally (J), C = efficiency of
doing mechanical work (work done/energy used), D
In the second tria1 the work routine involved the
= efficiency of raising body weight (work done
following sequences of activities: standing for 15
raising body weight/energy used).
min (SMR), walking unloaded around a flat laterite
circuit of 1000 m, pulling a two-wheeled cart with
The energy cost of ploughing was estimated
pneumatic tyres loaded with 300 kg around the
assuming the average drnught force (1047 N for the
1000-m circuit, pulling a cart loaded with 600 kg
team or 524 N for each animal) and the average
around the 1000-m circuit and finally pulling a cart
walking speed (0.81 m/s) found in this study. This
loaded with 900 kg around the 1000-m circuit. For
draught load would be equivalent to 0.16 of the live
both carting and ploughing oxen were paired.
weight of animais in the pair weighing 300 kg each,
During the ploughing trials, one ox in each pair
0.12 for animals weighing 400 kg and 0.10 for
walked on unploughed soi1 and one walked on
animais weighing 500 kg. Net energy requirements
ploughed soil. However, the position of oxen in
for maintenance (EM) were estimated as: EM =
the pair was changed during other ploughing
1.15(0.53[M/1.08]“‘h’)
L
_
according to Agricultural and
sessions SO that at the end of the ploughing tria1
Food Research Council (AFRC, 1993). The energy
each ox walked both on ploughed and unploughed
cost
for
maintenance
was
increased bv
soils.
proportionally 0.10 to account for the high&
metabolic rate after work as
compared t o
Heat production (H) was estimated using the
non-working days (Lawrence rt 17/., 1989a) and for
equation: H = 16.18 O? + 5.02 CO, (Brouwer, 1965)
the higher underlying resting metabolic rate during
where O2 is the volume of oxygen consumed and
work as compared to the resting metabolic rate
COZ is the volume of carbon dioxide produced.
during the same time of the day on non-working
Methane and urinary nitrogen were omitted from the
days (Lawrence ct al., 1989b).
Equation to calculate heat production proposed by
McLean and Tobin (1987) because they would have
Datn mnlysis
quantitatively little influence on the estimation of H
The following statistical mode1 was used to analyse
(see Lawrence et nl., 1991). Assuming a value of 0.9
E,,:
for the respiratory quotient (the ratio of carbon
dioxide produced : oxygen consumed),
the energy
Y,,, = Jo + 0, + Si + a(.~,,~ - Y,) + p(szi, -x2) + E,
expenditure of animals was estimated from oxygen
consumption alone, assuming 20.7 kJ/O, consumed
where: Y = kth observation of E,, for ith animal and
(Lawrence and Stibbards, 1990).
jth surface; p = mean; 0, = effect of animal, ith (i =
1...7); Si = effect of ground surface, jth, (j = 1:
The energy cost of walking (E,, J/m walked per kg
unploughed sandy soil; j = 2: ploughed sandy soil; j
M) was calculated as E,, = (energy used while
= 3: firm laterite track); a = regression coefficient of Y
walking - energy used while standing still)/
on the speed of walking (s,); fi = regression
(distance walked (m) X M (kg)). The energy cost of
coefficient of Y on the live weight of the animal (x2);
doing work was defined as an efficiency factor (E3
E,, = random error.
212
Fall, Pearson and Lawrence
Efficienci
Efficienc!
E;t,
M’alking speed
of carting
of ploughing
N O
(1 /m per kg)
(m/s)
Ground
nf
-
-
Animal Number
Mean
s.e.
Mcan
se.
surface
animais N o .
Mean s.e. Mean s.e.
1 0
0.25
0.14
0.27
0.08
Unploughed
13
0.30
0.09
0.31
0 . 0 5
sandy soils
6
21
1.50” OdhY 0.95’ 0.029
16
0.28
0.03
0.32
0.04
Ploughed
21
04
0.05
0.32
0.05
sandy soils
6
20
2.15” 0.084 04w 0.029
24
0.33
0 . 0 5
0.34
0.04
Lateritc track
6
19
I.o;J 0 . 1 0 0 1.26” 0.033
Significance
*
**
speed were not associated with changes in the
a,h Values in the samr column having differcnt supc*rscripts
energy cost of walking. Speed was higher when
are significantly differcnt (P < 005 at Ieast).
animais were walking on laterite tracks than when
they were walking on sandy field soils (Table 1).
The mode1 used to analyse E, during carting
included the main effects of oxen and load. The main
source of variation for the analysis of E, for
The average draught force required to plough sandy
ploughing was the effect of oxen. In both analyses of
soils in this experiment was 1037 (s.e. 13) N.
Ej for carting and ploughing, speed of travel and the
Ploughing was performed using a mouldboard
live weight of the oxen were included as covariates.
plough at an average depth ranging from 12.9 (s.e.
Since repeated measurements were taken on animais
0418) to 17.1 (s.e. 0.73) cm. Soi1 moisture content was
over days, animal was used as the error term to test
2.2, 2.7, 2.9 and 3.0’!4 at 0 to 5, 5 to 10, 10 to 15 and 15
the effect of ground surface on the energy cost of
to 20 cm of depth, respectively. Teams worked at an
walking and on the efficiency of doing work.
average speed of 0.81 (s.e. 0424) m/s. The efficiency
of ploughing was 0.31 (se. 0.008).
Results
The load during carting, M, draught force and
walking speed did not influence working efficiency.
Mean daily energy cost of standing was 5.63 (s.e.
The efficjency of dojng work during cartjng was only
0.12) W/kg M”“-i.
affectcd significantly (P < 0.01) by individual
animals, suggesting large variability between
animais (Table 2). The effect of individual oxen on
Ground surface affected E,,, and walking speed
the efficiency of doing work during ploughing was
(P < 0.01, Table 1). The energy cost of walking was
net however significant (Table 3).
lowest when the oxen walked on firm laterite tracks.
Energy expenditure also was lower when animais
Qm~t@don of tlw ~xtro cweqy mpir~~rt~~r~ts fat
walked on unploughed soils as compared with
ploughing
ploughed soils (Table 1). The regression of E,, on M
Table 4 shows the net energy required for
was significant. The heavier the animal, the higher
maintenance and for ploughing sandy soils for each
was the energy cost of walking. Each extra kg of M
animal in the team, one walking in the furrow and
was associated with an increase of 0.013 J/m per kg
the other walking on the unploughed soil.
in the energy cost of walking. Changes in walking
Depending on M of the anjmal and the number of
hours worked, the daily extra net energy expended
for ploughing varied between 0.10 to 0.89 times the
energy cost for maintenance.
Wnlking
Draught
speed
Discussion
Efficienq
force (N)
(m/s)
Load
No. of
o f
The SMR in this study (563 (s.e. 0.12) W/kg Mo7;)
(kg)
animais
N
o
.
working Mean se. Mt,an s.e.
was higher than that recorded by Becker c’t nl. (1993)
in zebu oxen in Niger (4.75 W/kg M0.7s) but lower
300
s
18 O-32 0 . 0 3
310
8.0
1.23 0.02
600
5
17 0.32 0.03 409 84 1.23 0.02
than that calculated from oxygen uptdke in resting
900
5
16 0 . 3 3 0 . 0 3
503 8.3 1.19 0.02
Bunaji bulls in Nigeria (7.59 W/kg MO’:s, Dijkman,
1993) and resting crossbred cows in Ethiopia
300
0.16
34.75
0.15
0.13
0.29
0.25
0.44
0.38
0.59
0.50
0.74
0.63
0.89
0.75
400
0.12
43.12
0.14
0.12
0.28
0;23
0.43
0.35
057
0.47
0.70
0.59
044
0.70
5 0 0
0.10
50.99
0.12
0.10
0.24
0.19
0.36
0.30
0.48
0.40
04x1
0.50
0.72
0.59
(9.32 W/kg M(“‘s, Zerbini et a/., 1992). Differences
in
content in valley bottoms in Nigeria, (3.76 to 8.58 J/
these results may be attributed to differences
in
m per kg M; Dijkman, 1993).
breeds used, food intake, climatic conditions, altitude
and in the time of day and measuring techniques
The energy costs of walking on firm surfaces
used in the different experiments. For example in
(unploughed land and laterite tracks) found in this
this experiment, SMR was measured before work
study were similar to other measurements made in
started whereas SMR values reported by Dijkman
the field. Becker et nl. (1993) reported an E,, of 1.34 J/
(1993) and Zerbini et ai. (1992) were averages of SMR
m per kg in zebu cattle in Niger, and Clar (1991)
values before work and between bouts of work
found an E,, of 1.00 J/m per kg also for zebu cattle in
(Zerbini et a/., 1992) and during recovery periods.
Niger that was similar to the E,, measured on laterite
Lawrence et a/. (1989b) found that the rate of energy
tracks in this study. Field values recorded on the firm
expenditure of well trained oxen given food at
surfaces were usually lower than values determined
maintenance and standing still between bouts of
on treadmills such as: 2.6 J/m per kg (AFRC, 1993),
work was proportionally 0.26 higher than the
1.9 J/m per kg (Brody, 1945) and 2.1 J/m per kg
average rate during the same time of the day when
(Lawrence and Stibbards, 1990). Discrepancies
the oxen were in a respiration chamber. The high
between laboratory and field values of E,, cari be
value of 9.32 W/kg Mfl.‘5 reported by Zerbini rpf nl.
explained by the artificial conditions of the former.
(1992) may be related to the Bos tauras X Bos indicus
When oxen walk in the field they travel at their own
crossbred dairy cows they used. Those animals may
speed and are likely to be more at ease than on a
have had a higher metabolic rate than the Bos ir~lic~s
treadmill in a laboratory, where thev have to walk at
breeds used in this and the other experiments
a set speed on a moving tread&ill surface. This
(Dijkman, 1993; Becker et nl., 1993). In Ethiopia,
illustrates
the
v a l u e o f
conducting
field
crossbred oxen were found to require more energy
measurements to establish the true energy
per unit of body weight for maintenance and work
requirements of working animals.
output than local oxen (Astatke, 1983). It is important
to note first that the SMR reported in these studies is
In this experiment the energy cost of walking was
related to total M and not to empty body mass.
independent of the walking speed. The energy cost
Secondly, the energy expenditure measured while
of walking increases as speed decreases if the rest-
the animal was standing still includes heat
maintenance component
of the cost is included
increment.
(Brody, 1945). However, if the maintenance cost is
excluded from the total energy cost, as was done
Energ/ cost ofwalking
here in the calculation of the E,,, then the energy cost
The significant effect of ground surface (unploughed
of walking is independent of speed. Lawrence and
and ploughed sandy soils and laterite tracks) on the
Stibbards (1990) also found that when oxen were
energy cost of walkmg agrees with results reported
walking at a comfortable speed, i.e. either forced to
by Dijkman (1993). The E,, of 1.59 J/m per kg M on
walk very slowly or very fast, but at a speed they
unploughed sandy soils found in this experiment is
might choose naturally, then the energy cost of
close to the E,. of 1.47 J/m per kg M on unploughed
walking was no longer influenced
by speed.
upland and the E,, of 1.76 J/m per kg M on
unploughed dry valley bottom soils found by
Energy cost ofdoing mork
Dijkman (1993) in Nigeria. As might be expected, E,,,
In this experiment the efficiency of doing work was
on ploughed land was lower on the sandy soils in the
not affected either by the type of work performed
present study (2.15 J/m per kg M) than seen on
(ploughing z). carting with varying loads) or the
ploughed land on the heavier soils with a higher clay
draught force exerted. The efficiency of ploughing
.
21-l
Fall, Pearson and Lawrence
(0.31) was consistent with average efficiencies of
(500 kg) is the better one to use for ploughing in the
pulling loads reported by Lawrence and Stibbards
semi-arid areas.
(1990). These results are also in agreement with
Dijkman (1993) who found an average efficiency of
The findings of this study cari be used to estimate
0.30 to 0.31 for oxen ploughing upland and valley
accurately the energy requirements for work in
bottom soils in Nigeria. Efficiency of doing work in
semi-arid areas through the application of the
the present study was unaffected by walking speed.
factorial method (Lawrence and Stibbards, 1990)
This again agrees with the observations of Lawrence
provided work output (draught force (N) X distance
and Stibbards (1990).
(m) travelled) during the working day is known.
Functional activities such as locomotion and standing
11-1 this experiment a mouldboard plough was used
cari contribute a great deal to the daily energy budget
for the ploughing tria]. Results showed that an
in extensive livestock production systems prevailing
average draught force of 1047N would be rcquired
in semi-arid areas. The energy cost of these activities
to till at an average depth of 15 cm. The mouldboard
cari be estimated through the monitoring of the daily
plough cari also be used for direct ridging on untilled
activities of animals. This would allow the complete
sandy soils. In Zimbabwe, when ridges are already
daily energy budget of draught animals to be more
established, re-ridging moist sandy loam at the
exactly calculated for these areas.
beginning of subsequent seasons required draught
forces comparable or slightly less than those for
ploughing (Stevens, 1994). Therefore results from the
Acknowledgements
ploughing tria1 in this experiment may also be
Financial support from the Overseas Development
applicable to direct ridging of sandy soils using a
Administration (ODA) of the United Kingdom and the
International Livestock Research Institute is gratefully
mouldboard plough. Jt has been suggested that oxen
acknowledged. The authors also thank S. Fern6ndez-Ri\\Vera
cm sustain work ovcr a working day provided the
for his constructi\\rc comments, and E. A. Hunter for
draught force does net exceed about 11 kgf per
statistical ad\\ice during preparation of the manuscript.
100 kg M (Goe and McDowell, 1980). This implies
that a team totalling at least 950 kg is thc ideal M for
ploughing and ridging sandy soils in these scmi-arid
References
areas.
Agricultural and Food Research Council. 1993. EIW~~?/ aff[i
protei77 rapircrmv7ts (If r77777i77~777fs.
A
n
advisory manual
Lawrence (1985) estimated the energy expenditure of
prepared by the AFRC Technical Committee on Responses
to Nutrients. CAB International, Wallingford, UK.
cattle working in land preparation for 5.5 h/day to
be 0.42 to 067 times maintenance requirements
Astatke, A. 1983. Fc~d n?fd :(l& t1%7/i; nt Debrt3 Bcrhf, 2980-
1982. Working document, ILCA Highlands Programme,
Pearson (1989) reported estimated extra daily energy
International Livestock Centre for Africa, Addis Ababa.
requirements of 0.74 to 0.78 tiines maintenance
energy requirements for cattle pulling carts for 5 h/
Becker, K., Rometsch, M., Susenbeth, A., Roser, U. and
Lawrence, P. 1993. Chlzrïfzfcri~rlfiorl uf
day. Over a shorter working day (3 h) Mahardika rt
fllr ~~/l~~siOh@Xl
/qfi~r~~l~~~rc~~ nrid dctPr?Gatiorr of tl7r gfjirier~c~~ of h77rght ~X~II.
nl. (1994) estimated the extra energy expenditure of
Special research program (SFB 308), intermediate report
water buffaloes working at draught forces Equivalent
1991-1993, University of Hohenhcim, Stuttgart, Germany.
to proportionally 0.10 and 0.15 of M to be 0.42 and
Brody, S. 1945. Biorrqqrtic,~ & groic+lr. Hafner Press, New
0.48 t i m e s
maintenance energy requirements,
York.
respectively. Using the values for E,, and the
Brouwer, E. 1965. Report of the sub-committee on constants
efficiency of doing work obtained in this study to
and factors. Procrrrfin~~s qf tlr~ thirrl sp~~si77177
~7 fhe c~7rrg~/
estimate the extra net energy used for work
~r~tnPo/is~r $ fi7r7n nrGn~/s (ed K. L. Blaxter), EAAP
according to the method of Lawrence and Stibbards
P7rblicalh 110. I?, pp. 441-443. Academic I’ress, London.
(1990) gave values for the extra energy for work
Clar, U. 1991, Entwichlung ciner Feldmethode zur Messung
ranging from 0.10 to 0.89 times daily maintenance
des Energie Umsatzes bei Zugticren. PhD. thk, Ll~7i?hf!/
energy requirements, depending on the number of
ojHollerl/lrinl.
hours worked and the size of the animals used. The
Dijkman, J. T. 1993. The measurement of draught ruminant
estimated energy expenditure during 5 to 6 h of
energy expenditure in the field. Ph.0. fk$it;, lhiwrrify of
ploughing sandy soils varied between 0.50 for oxen
Eili~~bw~h.
weighing 500 kg to 0.89 X maintenance for oxen
Food and Agriculture Organization. 1987. Afik~~
weighing 300 kg. The energy expenditure during
a,qricrrlfwe: thr ~ext 1.5 1/n7rs. The Fond and Agric&re
work of the 500-kg ox was comparable to that seen in
Organization of the Unitéd Nations, Rome, Italy.
the other studies (Lawrence, 1985; Pearson, 1989;
GO~, M. R. and McDowell, R. E. 1980. A~~i.vrnl trnc‘tiofi:
Mahardika rt RI., 1994), but energy expcnditure
~[/idcli,~s jbr mfilisafion. Corne11 International Agricultural
during work by the 300-kg ox was higher than that
Development m i m c o g r a p h ( U S A ) n o . Pl. Corne11
generally found, suggesting that the larger animal
University, Ithaca, NY.
Nutrition of draught cattle-1
215
Humpfrey, S. J. E. and Wolff, H. S. 1977. The Oxylog.
Mahardika, 1. G., Sastradipradja, D., Sumadi, 1. K.,
~orn’rfd c>fPhl/SiOkJ~!/, bf7dof7 267: 12.
Sutardi, T. and Kartiaso 19% Energy expenditure of \\va&r
Lawrence, P. R. 1985. A review of the nutricnt reyuirements
buffaloes for draft power in the humid tropics. In
of
Sustnir7nblr s7t7oll-sdc rnr77h7i7f pc~d7rcfior7 in atw-arid trr7d
draught oxen. In Drn7,gl7f nr7imnl po7cw,~v prod7~fk~71 (cd.
J, W. Copland). Pmwdi77~~s $[772 ii7frlnnfio77ol workslroy, /afws
sd4777717id trqisnl mm (ed K. Becker, 1’. R. Labvrence and
C o o k U77izvrsify, ~CJilV75Zdk, A7MT7lil7, 10-16 ]7dy 198~5.
E. Ii. Lilrskov). Procwdir7~~s ~,f 1717 i77tc~rr7~7fionnl
mrkshop 077
ACIAR proceedings scries n o . 10, pp. 59-63. ACIAR,
Septcrr7brr 24 1994 at tl7r I77sfif7lfr for Al7iir7nl Pwtf7rt-tior7 ii7 f17ta
Canberra.
Tuopic’s i77u-i Sd-Tro$cs, U77 iivrsify c$ Hol7r77l7~i177,
SfuttXnrt,
Gerrrrnr7!/, pp. 127-130.
Lawrence, P. R. and Becker, K. 1994. The cffects of draft
powcr on feed utilisation and reproduction in cattle and
Panin,
A. and Ellis-Jones, J. lY94. Increasing thc
buffaloes. hi S~rtainnbk w7n/l-sa7/r
rur7rir7a77t producfio77 in
profitability of draught animal power. In I7r7pwir7g mrirrrt7l
scvri-nuid ar7d ~7717-l717777id fropicrrl ~77~7s (ed. K. Becker, 1’. R.
fnrcfio77 fd777olpp (rd. P. FI. Starkey, E. Mwenya and J .
Lawrencc and E. R. Qrskovj. Pmwdir7gs o,f 1771 ir7fcwrr7/io77~7l
Starcs). Procwd777~yL; of fhe fïrsf w~~r~ksl7q~
of tlrc A77irr7~7l Tmtio77
workd7op 077 Sej~trwibtv'
74 7994 r7f thé htitufr for Auim7l
.Nrfwod- $r Enstrr77’ (7rnl’ So7rthrrr Afii(:l7 fATNESAJ, 18-23
Pmrl7~cfio77 i77 fl7r Tropics cli7d Suh-Trqlic>. U7&ersity o
f
/nrzrrnry 7992, hsalin, Zafnbi~7,
pp. 91-l 03. Technical Ccntrc
Hoh~i7lrrir17, Stfrttgnrt, Gurr7m7y, pp. 101-121
for Agricultural Co-operation (CTA), Wageningen, The
Netherlands.
Lawrence, P. R., Buck, S. F. and Campbell, 1. lY&Ya. The
metabolic r a t e o f o x e n a f t e r w o r k . P~mmdi77~~5
c’f’ tlw
Pearson, R. A. 1989. A comparison of drnught cattle (Ho>
iN77t7+fior7 Socif’t!/ 48: 1531% (abstr.).
i77rfk.77~)
and buffaloes (BuPnl7r~ buhr7lis) tartine lords in bot
candi tiens. A77im72 Pro&tio77 49: 355-393.
”
Lawrence, P. R. and Pearson, R. A. 1985. Factors affecting
the measurement of draught force, work output and power
Starkey, P. H. 1994. A wide view of animal traction
of oxen. ~01r777nl
of A8ric7dt77n7I Scicvzcc~, Cmrbri[fp 105:
highlighting some key issues in eastern and southcrn
703-714.
Africa. In Inrprovir7g mrinrnl fmtior7 trc~777ch~~~~~ (ed. P. H.
Starkcy, E. Mwenya and J. Sbres). Prmwdi77~~s qf tlw jirst
Lawrence, P. R., Pearson, R. A. and Dijkman, J. T. .lYYl.
zcwkd7op oj fhc Ani717nl Tmfior7 Nctzrwk for Ensttw7 (777~1
‘Techniques for mcasuring whole body energy expenditure
So7dhrrrr Afiicn (ATNESA), ?R-2.3 /o7777m$ 1997, Lusab7,
of working nnimals: a critical rcview. In Isotope (777~1 nlntcd
ZnftrPin, pp. 66-81. Technical Centre for Agricultural Co-
t~d777iq7vi;
ir7 1777i771d
pvad~tiof7 m7d hrnlfh. Proceedings of an
operation (CTA), Wageningen, The Netherlnnds.
international symposium on nuclear and related techniques
in animal production and henlth, International Atomic
Steven, A . P . 1 9 9 4 . Improring animal-powerrd tillagc
Energy Agehcy and the Food ami Agriculture Organization
systems and weeding technology. In hrprovi77,y r7r7irrrnl
of thc United Nations, Vienna, 15-19 Aaril 1991,
fmcfh fcrhnolyy (ed. P. H. Starkey, E. Mwenya and J.
DD.
II
I
21 l-232. International Atomic Energy Agency, Vienna.
Stares). Procccd7r7gs of thcfirsf workshop c!ffh~ Aui777nl Tmctio77
Ncfumk f o r Enstrrrz n77rf So77tl7cwr Qricn (ATNESA), 18-23
Lawrence, P. R., Sosa, R. and Campbell, 1. 1989b. The
]munry 1992, Lnsakn, Znnrh7, pp. 168-181. Technical Centre
underlying ‘resting’ energy consumption of oxen during
for Agricultural Co-operation (CTA), Wageningen, The
work. Procrcdi77gs of trlr N77frifiu7 Society 48: 154A (abstr.)
Netherlands.
Lawrence, P. R. and Stibbards, R. J. 1YYO. The energy costs
Zerbini, E., Gemeda, T., O’Neill, D. H., Howell, P. J. and
of walking, carrying and puliing loads on Bat surfaces by
Schroter, R. C . 1992. Relationships between cardio-
Brahman cattle and swamp buffalo. Anirrrnl Prod77ctior7 50:
respiratory parameters and draught work output in Fl
29-39.
crossbred dairy CO~S under field conditions. Anirru7l
McLean, J. A. and Tobin, G. 1987. Atzirt7nl n77d l111117~717
Pmfirctio77 55: l-10.
mlorin7rtry. Cambridge University Press, New York.
expenditure by oxen working on soils of different consistencies
A. Fall’, R. A. Pearson’ artd P. R. Lawrence’t
‘Cer~frcfi~r Tropicnl Veferitmy Mediciw, Unizwsit!! (11 Edinbuqh, Emter Bush, Rosli,l, Midlofhim EH25 9RG
‘Iuferwafioml Licestock Reseorch Zmfifufc, Zntcrrrntmnc71 Crop Rcw~h Irzstifutefiu flw Smi-Avid Tropics, Sahelim
Centre, BP 12404, Ninnrey, Niger
Abstract
The Oxylog, CI portriblr brcntli-by-brPntlz gris nnal~~scr, 7~7s urcd 011 sczm animnls to dctrrnzitze stnndiq nlctnbolic
rate, cnergy cosf of walking on soils of different consistemies md e$ïcimcy of ruork ploz@iq nnd cnrting. Thc
merngr standing nlrfnbolic nzte of nnimnls wns 5.63 (se. 0.12) W/kg M”.7S. Thc comisfcmy of the soi1 011 ruhich
mirmls worked hnd o rnarked +ct ou fheir eucrgy cost of zunlkiq which 71x1s 1.59 (s.c. 0.069) 011 wlplo@cd soil,
2.15 (S.E. 0.084) ou plo~$cd soi1 nnd 2.0 (‘se. 0.10) ]/I?L pcr kg livc wcight 011 latcrife frncks. The eflcimcy qf
plow$ing sandy soils (i.c. rrrfio of zoork doue fo cmqy usrdfor work) 7017s 0.32 nnd ms nef sigrt$m~tly d$&rlt
frm the e mwc
J’ o
y CRYf in
f 5 wifh
A
different loads. Thc tIfficiency qf doilzg work zuas rzof irzj7rlenced by the type of
zuork perfomed, the drnugh f @ce excrted or the walkinl; speed.
Keywords: drnught anin&, rrzergy expcnditurc, zvork.
Introduction
preparation and timely planting at the onset of the
Draught animal powrr was introduced in
rainy season. In these situations draught animal
sub-Saharan Africa during the last 70 years and its
power becomes critical to supplement human energy
use is spreading (Starkey, 1994; Panin and Ellis-
SO that field operations cari be done at the right time
Jones, 1994). However, the contribution of draught
to reduce the risk of trop failure and to secure a
animals tu the power requirements for agriculture is
stable yield.
still limited. Agricultural production in this region
continues to rely primarily on human power.
Adcquate feeding to meet the nutrient requirements
Statistics in 1987 suggested that proportionately 0.89
of draught animals is a major constraint in semi-arid
of power was provided by humans while draught
areas, because food is scarce and of poor quality
animals supplied only 0.10 of the farm power input
during many months in the year. In these areas there
(Food and Agriculture Organization, 1987). There is
is a need for a rational planning of the feeding of
a need to promote draught animal power in
draught animals to supply sufficient draught animal
sub-Saharan Africa to fil1 the gap between the
power for trop production. This requires information
deteriorating level of food production and the
on the seasonal availability and the nutritive value of
increasing demand for food. This is particularly true
existing food resources, the utilization of these foods
in semi-arid areas where timeliness in cropping
by draught animals and the nutrient requirements
operations is fundamental for successful cropping
for work. Information on the energy requirements
because of the short growing season in these areas.
for work and food utilizntion of draught animals in
The low and erratic rainfall regime constrains land
semi-arid West Africa is limited. However, the recent
adaptation of portable equipment to measure oxygen
consumption
(Lawrence et nl., 1991) cari contribute
greatly to the knowledge of the energy expenditure
t Present address: Centre for Agriculture in the Tropics
of animals performing common farm operations.
and Suhtropics. Institute of Animal Production,
Department of Animal Nutrition and Aquaculture.
In the absence of direct measurement of oxygen
Universit2it Hohenheim (%Xl), D-7093, Stuttgart, Germany.
consumption,
the extra energy used to perform
---
-_.- .-
210
Fall, Pearson and Lawrence
different activities cari be estimated using the
the same diameter and turbine stator as the
factorial method developed by Lawrence and
original flowmeter but no rotor or electric
Stibbards (1990). This method integrates, additively,
connections.
They
t h e r e f o r e h a d t h e same
the energy cost of walking, carrying and pulling
resistance to air flow as the original flowmeter and
loads. The energy cost of pulling is fairly constant
by uncovering one or more of them, the range of
when expressed in relation to tractive effort and
the flowmeter could be increased from its original
distance. Therefore, this cari be accurately predicted
value of 0 to 80 l/min to 0 to 160, 0 to 270 or 0 to
if work output is known. Energy cost of walking,
3 6 0 l/min. T h e r a n g e used i n a p a r t i c u l a r
which cari account for proportionately 0.50 or more
experiment depended on the expected maximum
of the total energy expended for work (Lawrence and
respiration rate. The flowmeter was calibrated over
Becker, 1994), is more difficult to predict because it is
a11 four ranges using a reciprocating air pump as
dependent on ground surface and needs to be
described by Dijkman (1993). Expired air passed
determined directly. For instance, Dijkman (1993)
from the mask via three 25-mm diameter outlet
found in the sub-humid zone of Nigeria that the
valves into a flexible tube attached to the analysis
energy cost of walking (E,) in cattle ranged from
and display unit fixed to the animals’ back. The
8.58 J/m per kg M on ploughed waterlogged rice
Oxylog was used to take samples of inspired and
fields to 1.47 J/m per kg M on unploughed upland
expired air at every breath, and determined the
soils. The objective of this study was to investigate
difference
in oxygen concentration using two
the energy cost of walking (E,) and ploughing on
matched
polarographic electrodes. This was
sandy soils and of carting in semi-arid areas.
multipiied by the volume of air inspired to give
oxygen
consumption.
The
apparatus
made
corrections for atmospheric pressure, temperature
Material and methods
and humidity and displayed the results as oxygen
Animals and feeding
consumed a n d a i r i n s p i r e d (1) corrected t o
This experiment was conducted from October to
standard temperature and pressure. In the present
November 1994 at the International Crop Rcsearch
experiment these values were recorded every min
Institute for the Semi-Arid Tropics (ICRISAT)
from an auxiliary display pane1 connected by table
Sahelian Centre at Sadore, Niger. Seven Dinli @os
to the Oxylog which enabled data to be read more
indicus) cattle, average live weight 367 (s.e. 22) kg,
easily.
aged 5 to 7 years, were used in the study. The
animals were given food to about maintenance level
Work output, distance travelled and time spent
on natural pastures supplemented with wheat bran
working were measured using an ergometer for
and a minera1 mixture (in total about 9 to 10 g/kg
work and distance, or odometer for distance only
Mo’ 75 per day). The animals had access to water ad
(Lawrence and Pearson, 1985).
libitum during the periods when they were not
working. Mean ambient temperature and relative
The animals used in this experiment were already
humidity were 3O.P”C and 0.627, respectively when
well trained for work. Further training was necessary
animals worked in the morning and 36.5”C and 0.242
to accustom them to carrying the instruments.
when work took place in the afternoon.
Animals were trained for 3 weeks to wear the face
mask and to carry the backpack containing the
Experimcnfal tnetkods
ergometer and the Oxylog while performing
Oxygen consumption
was measured using the
common farm activities.
Oxylog, a portable breath-by-breath analyser. This
instrument (P. K. Morgan Ltd, Kent, England), was
Two trials were conducted. The first tria1 was
originally designed for use with humans and was
designed to determine the energy cost of ploughing
modified for oxen in a manner similar to that
sandy soils using a mouldboard plough. The second
described by Lawrence ef al. (1991).
tria1 measured the energy cost of carting. Light carts
with pneumatic tyres were used. In both trials the
The apparatus consists of an airtight face mask
standing metabolic rate and the energy cost of
with inlet and outlet valves and an analysis and
walking were determined. The treatment applied for
display unit. Air is sucked into the mask through a
the measurement of the energy cost of walking was
cylinder 80 mm long X 100 mm diameter mounted
the consistency of the surface: unploughed wet
on the right side. At the end of this tube near the
sandy soils, ploughed wet sandy soils and firm
mask a plate was fitted with three 25-mm diameter
laterite tracks. For the determination of the
inlet valves. At the other end there was a similar
efficiency of doing work carting, the treatment was
plate which contained the original Oxylog turbine
the load applied (300, 600 and 900 kg). The
flowmeter (Humpfrey and Wolff, 1977) and three
experimental design for both the tria1 on the energy
‘dummy’ flowmeters. The dummy flowmeters had
cost of walking and for the tria1 on the efficiency of
Nutrition of draught cattle-1
211
carting consisted of a random assignment o f
according to Lawrence and Stibbards (1990): Ej =
treatments in sequencc to each ox (block) with
(work done (kJ)/(energy expended when loaded (kJ)
repeated measures.
- energy expended (kJ) to walk the same distance at
the same speed but unloaded).
During the first tria1 the work routine of the six
animais included the following sequence o f
The information obtained from this experiment was
activities: standing for 15 min in the shade (SMR),
incorporated into a factorial formula (Lawrence
walking unloaded for 15 min on unploughed soils,
and Stibbards, 1990) to predict the extra net
walking unloaded for 15 min on previously
cnergy required for ploughing sandy soils for 1 to
ploughed soils and ploughing for 20 min.
6 h/days:
For each activity, measurements were taken whcn
E = A.F.M + B.F.L + W/C + 9.81 H.M/D
animals had reached a steady rate of oxygen
consumption after having worked for at least 5 min.
where: E = extra energy used for work (kJ), F =
The Oxylog was alternately attached to each animal
distance travelled (km), M = live weight (kg), L =
during each work routine. During the walking
load carried (kg), W = work done whiist pulling
sessions, the ndometer of the ergometer was wheeled
loads (kJ), II = distance moved vertically upwards
behind the animal to measure distance walked and
(km), A = energy used to move 1 kg of body weight
time spent walking. Animals were allowed to rest for
1 m horizontally (J), B = cnergy used to move 1 kg of
15 min between bouts of work.
applied load 1 m horizontally (J), C = efficiency of
doing mechanical work (work done/energy used), D
In the second tria1 the work routine involved the
= efficiency of raising body weight (work done
following sequences of activities: standing for 15
raising body weight/energy used).
min (SMR), walking unloaded around a flat laterite
circuit of 1000 m, pulling a two-wheeled cart with
The energy cost of ploughing was estimated
pneumatic tyres loaded with 300 kg around the
assuming the average drnught force (1047 N for the
1000-m circuit, pulling a cart loaded with 600 kg
team or 524 N for each animal) and the average
around the 1000-m circuit and finally pulling a cart
walking speed (0.81 m/s) found in this study. This
loaded with 900 kg around the 1000-m circuit. For
draught load would be equivalent to 0.16 of the live
both carting and ploughing oxen were paired.
weight of animais in the pair weighing 300 kg each,
During the ploughing trials, one ox in each pair
0.12 for animals weighing 400 kg and 0.10 for
walked on unploughed soi1 and one walked on
animais weighing 500 kg. Net energy requirements
ploughed soil. However, the position of oxen in
for maintenance (EM) were estimated as: EM =
the pair was changed during other ploughing
1.15(0.53[M/1.08]“‘h’)
L
_
according to Agricultural and
sessions SO that at the end of the ploughing tria1
Food Research Council (AFRC, 1993). The energy
each ox walked both on ploughed and unploughed
cost
for
maintenance
was
increased bv
soils.
proportionally 0.10 to account for the high&
metabolic rate after work as
compared t o
Heat production (H) was estimated using the
non-working days (Lawrence rt 17/., 1989a) and for
equation: H = 16.18 O? + 5.02 CO, (Brouwer, 1965)
the higher underlying resting metabolic rate during
where O2 is the volume of oxygen consumed and
work as compared to the resting metabolic rate
COZ is the volume of carbon dioxide produced.
during the same time of the day on non-working
Methane and urinary nitrogen were omitted from the
days (Lawrence ct al., 1989b).
Equation to calculate heat production proposed by
McLean and Tobin (1987) because they would have
Datn mnlysis
quantitatively little influence on the estimation of H
The following statistical mode1 was used to analyse
(see Lawrence et nl., 1991). Assuming a value of 0.9
E,,:
for the respiratory quotient (the ratio of carbon
dioxide produced : oxygen consumed),
the energy
Y,,, = Jo + 0, + Si + a(.~,,~ - Y,) + p(szi, -x2) + E,
expenditure of animals was estimated from oxygen
consumption alone, assuming 20.7 kJ/O, consumed
where: Y = kth observation of E,, for ith animal and
(Lawrence and Stibbards, 1990).
jth surface; p = mean; 0, = effect of animal, ith (i =
1...7); Si = effect of ground surface, jth, (j = 1:
The energy cost of walking (E,, J/m walked per kg
unploughed sandy soil; j = 2: ploughed sandy soil; j
M) was calculated as E,, = (energy used while
= 3: firm laterite track); a = regression coefficient of Y
walking - energy used while standing still)/
on the speed of walking (s,); fi = regression
(distance walked (m) X M (kg)). The energy cost of
coefficient of Y on the live weight of the animal (x2);
doing work was defined as an efficiency factor (E3
E,, = random error.
212
Fall, Pearson and Lawrence
Efficienci
Efficienc!
E;t,
M’alking speed
of carting
of ploughing
N O
(1 /m per kg)
(m/s)
Ground
nf
-
-
Animal Number
Mean
s.e.
Mcan
se.
surface
animais N o .
Mean s.e. Mean s.e.
1 0
0.25
0.14
0.27
0.08
Unploughed
13
0.30
0.09
0.31
0 . 0 5
sandy soils
6
21
1.50” OdhY 0.95’ 0.029
16
0.28
0.03
0.32
0.04
Ploughed
21
04
0.05
0.32
0.05
sandy soils
6
20
2.15” 0.084 04w 0.029
24
0.33
0 . 0 5
0.34
0.04
Lateritc track
6
19
I.o;J 0 . 1 0 0 1.26” 0.033
Significance
*
**
speed were not associated with changes in the
a,h Values in the samr column having differcnt supc*rscripts
energy cost of walking. Speed was higher when
are significantly differcnt (P < 005 at Ieast).
animais were walking on laterite tracks than when
they were walking on sandy field soils (Table 1).
The mode1 used to analyse E, during carting
included the main effects of oxen and load. The main
source of variation for the analysis of E, for
The average draught force required to plough sandy
ploughing was the effect of oxen. In both analyses of
soils in this experiment was 1037 (s.e. 13) N.
Ej for carting and ploughing, speed of travel and the
Ploughing was performed using a mouldboard
live weight of the oxen were included as covariates.
plough at an average depth ranging from 12.9 (s.e.
Since repeated measurements were taken on animais
0418) to 17.1 (s.e. 0.73) cm. Soi1 moisture content was
over days, animal was used as the error term to test
2.2, 2.7, 2.9 and 3.0’!4 at 0 to 5, 5 to 10, 10 to 15 and 15
the effect of ground surface on the energy cost of
to 20 cm of depth, respectively. Teams worked at an
walking and on the efficiency of doing work.
average speed of 0.81 (s.e. 0424) m/s. The efficiency
of ploughing was 0.31 (se. 0.008).
Results
The load during carting, M, draught force and
walking speed did not influence working efficiency.
Mean daily energy cost of standing was 5.63 (s.e.
The efficjency of dojng work during cartjng was only
0.12) W/kg M”“-i.
affectcd significantly (P < 0.01) by individual
animals, suggesting large variability between
animais (Table 2). The effect of individual oxen on
Ground surface affected E,,, and walking speed
the efficiency of doing work during ploughing was
(P < 0.01, Table 1). The energy cost of walking was
net however significant (Table 3).
lowest when the oxen walked on firm laterite tracks.
Energy expenditure also was lower when animais
Qm~t@don of tlw ~xtro cweqy mpir~~rt~~r~ts fat
walked on unploughed soils as compared with
ploughing
ploughed soils (Table 1). The regression of E,, on M
Table 4 shows the net energy required for
was significant. The heavier the animal, the higher
maintenance and for ploughing sandy soils for each
was the energy cost of walking. Each extra kg of M
animal in the team, one walking in the furrow and
was associated with an increase of 0.013 J/m per kg
the other walking on the unploughed soil.
in the energy cost of walking. Changes in walking
Depending on M of the anjmal and the number of
hours worked, the daily extra net energy expended
for ploughing varied between 0.10 to 0.89 times the
energy cost for maintenance.
Wnlking
Draught
speed
Discussion
Efficienq
force (N)
(m/s)
Load
No. of
o f
The SMR in this study (563 (s.e. 0.12) W/kg Mo7;)
(kg)
animais
N
o
.
working Mean se. Mt,an s.e.
was higher than that recorded by Becker c’t nl. (1993)
in zebu oxen in Niger (4.75 W/kg M0.7s) but lower
300
s
18 O-32 0 . 0 3
310
8.0
1.23 0.02
600
5
17 0.32 0.03 409 84 1.23 0.02
than that calculated from oxygen uptdke in resting
900
5
16 0 . 3 3 0 . 0 3
503 8.3 1.19 0.02
Bunaji bulls in Nigeria (7.59 W/kg MO’:s, Dijkman,
1993) and resting crossbred cows in Ethiopia
300
0.16
34.75
0.15
0.13
0.29
0.25
0.44
0.38
0.59
0.50
0.74
0.63
0.89
0.75
400
0.12
43.12
0.14
0.12
0.28
0;23
0.43
0.35
057
0.47
0.70
0.59
044
0.70
5 0 0
0.10
50.99
0.12
0.10
0.24
0.19
0.36
0.30
0.48
0.40
04x1
0.50
0.72
0.59
(9.32 W/kg M(“‘s, Zerbini et a/., 1992). Differences
in
content in valley bottoms in Nigeria, (3.76 to 8.58 J/
these results may be attributed to differences
in
m per kg M; Dijkman, 1993).
breeds used, food intake, climatic conditions, altitude
and in the time of day and measuring techniques
The energy costs of walking on firm surfaces
used in the different experiments. For example in
(unploughed land and laterite tracks) found in this
this experiment, SMR was measured before work
study were similar to other measurements made in
started whereas SMR values reported by Dijkman
the field. Becker et nl. (1993) reported an E,, of 1.34 J/
(1993) and Zerbini et ai. (1992) were averages of SMR
m per kg in zebu cattle in Niger, and Clar (1991)
values before work and between bouts of work
found an E,, of 1.00 J/m per kg also for zebu cattle in
(Zerbini et a/., 1992) and during recovery periods.
Niger that was similar to the E,, measured on laterite
Lawrence et a/. (1989b) found that the rate of energy
tracks in this study. Field values recorded on the firm
expenditure of well trained oxen given food at
surfaces were usually lower than values determined
maintenance and standing still between bouts of
on treadmills such as: 2.6 J/m per kg (AFRC, 1993),
work was proportionally 0.26 higher than the
1.9 J/m per kg (Brody, 1945) and 2.1 J/m per kg
average rate during the same time of the day when
(Lawrence and Stibbards, 1990). Discrepancies
the oxen were in a respiration chamber. The high
between laboratory and field values of E,, cari be
value of 9.32 W/kg Mfl.‘5 reported by Zerbini rpf nl.
explained by the artificial conditions of the former.
(1992) may be related to the Bos tauras X Bos indicus
When oxen walk in the field they travel at their own
crossbred dairy cows they used. Those animals may
speed and are likely to be more at ease than on a
have had a higher metabolic rate than the Bos ir~lic~s
treadmill in a laboratory, where thev have to walk at
breeds used in this and the other experiments
a set speed on a moving tread&ill surface. This
(Dijkman, 1993; Becker et nl., 1993). In Ethiopia,
illustrates
the
v a l u e o f
conducting
field
crossbred oxen were found to require more energy
measurements to establish the true energy
per unit of body weight for maintenance and work
requirements of working animals.
output than local oxen (Astatke, 1983). It is important
to note first that the SMR reported in these studies is
In this experiment the energy cost of walking was
related to total M and not to empty body mass.
independent of the walking speed. The energy cost
Secondly, the energy expenditure measured while
of walking increases as speed decreases if the rest-
the animal was standing still includes heat
maintenance component
of the cost is included
increment.
(Brody, 1945). However, if the maintenance cost is
excluded from the total energy cost, as was done
Energ/ cost ofwalking
here in the calculation of the E,,, then the energy cost
The significant effect of ground surface (unploughed
of walking is independent of speed. Lawrence and
and ploughed sandy soils and laterite tracks) on the
Stibbards (1990) also found that when oxen were
energy cost of walkmg agrees with results reported
walking at a comfortable speed, i.e. either forced to
by Dijkman (1993). The E,, of 1.59 J/m per kg M on
walk very slowly or very fast, but at a speed they
unploughed sandy soils found in this experiment is
might choose naturally, then the energy cost of
close to the E,. of 1.47 J/m per kg M on unploughed
walking was no longer influenced
by speed.
upland and the E,, of 1.76 J/m per kg M on
unploughed dry valley bottom soils found by
Energy cost ofdoing mork
Dijkman (1993) in Nigeria. As might be expected, E,,,
In this experiment the efficiency of doing work was
on ploughed land was lower on the sandy soils in the
not affected either by the type of work performed
present study (2.15 J/m per kg M) than seen on
(ploughing z). carting with varying loads) or the
ploughed land on the heavier soils with a higher clay
draught force exerted. The efficiency of ploughing
.
21-l
Fall, Pearson and Lawrence
(0.31) was consistent with average efficiencies of
(500 kg) is the better one to use for ploughing in the
pulling loads reported by Lawrence and Stibbards
semi-arid areas.
(1990). These results are also in agreement with
Dijkman (1993) who found an average efficiency of
The findings of this study cari be used to estimate
0.30 to 0.31 for oxen ploughing upland and valley
accurately the energy requirements for work in
bottom soils in Nigeria. Efficiency of doing work in
semi-arid areas through the application of the
the present study was unaffected by walking speed.
factorial method (Lawrence and Stibbards, 1990)
This again agrees with the observations of Lawrence
provided work output (draught force (N) X distance
and Stibbards (1990).
(m) travelled) during the working day is known.
Functional activities such as locomotion and standing
11-1 this experiment a mouldboard plough was used
cari contribute a great deal to the daily energy budget
for the ploughing tria]. Results showed that an
in extensive livestock production systems prevailing
average draught force of 1047N would be rcquired
in semi-arid areas. The energy cost of these activities
to till at an average depth of 15 cm. The mouldboard
cari be estimated through the monitoring of the daily
plough cari also be used for direct ridging on untilled
activities of animals. This would allow the complete
sandy soils. In Zimbabwe, when ridges are already
daily energy budget of draught animals to be more
established, re-ridging moist sandy loam at the
exactly calculated for these areas.
beginning of subsequent seasons required draught
forces comparable or slightly less than those for
ploughing (Stevens, 1994). Therefore results from the
Acknowledgements
ploughing tria1 in this experiment may also be
Financial support from the Overseas Development
applicable to direct ridging of sandy soils using a
Administration (ODA) of the United Kingdom and the
International Livestock Research Institute is gratefully
mouldboard plough. Jt has been suggested that oxen
acknowledged. The authors also thank S. Fern6ndez-Ri\\Vera
cm sustain work ovcr a working day provided the
for his constructi\\rc comments, and E. A. Hunter for
draught force does net exceed about 11 kgf per
statistical ad\\ice during preparation of the manuscript.
100 kg M (Goe and McDowell, 1980). This implies
that a team totalling at least 950 kg is thc ideal M for
ploughing and ridging sandy soils in these scmi-arid
References
areas.
Agricultural and Food Research Council. 1993. EIW~~?/ aff[i
protei77 rapircrmv7ts (If r77777i77~777fs.
A
n
advisory manual
Lawrence (1985) estimated the energy expenditure of
prepared by the AFRC Technical Committee on Responses
to Nutrients. CAB International, Wallingford, UK.
cattle working in land preparation for 5.5 h/day to
be 0.42 to 067 times maintenance requirements
Astatke, A. 1983. Fc~d n?fd :(l& t1%7/i; nt Debrt3 Bcrhf, 2980-
1982. Working document, ILCA Highlands Programme,
Pearson (1989) reported estimated extra daily energy
International Livestock Centre for Africa, Addis Ababa.
requirements of 0.74 to 0.78 tiines maintenance
energy requirements for cattle pulling carts for 5 h/
Becker, K., Rometsch, M., Susenbeth, A., Roser, U. and
Lawrence, P. 1993. Chlzrïfzfcri~rlfiorl uf
day. Over a shorter working day (3 h) Mahardika rt
fllr ~~/l~~siOh@Xl
/qfi~r~~l~~~rc~~ nrid dctPr?Gatiorr of tl7r gfjirier~c~~ of h77rght ~X~II.
nl. (1994) estimated the extra energy expenditure of
Special research program (SFB 308), intermediate report
water buffaloes working at draught forces Equivalent
1991-1993, University of Hohenhcim, Stuttgart, Germany.
to proportionally 0.10 and 0.15 of M to be 0.42 and
Brody, S. 1945. Biorrqqrtic,~ & groic+lr. Hafner Press, New
0.48 t i m e s
maintenance energy requirements,
York.
respectively. Using the values for E,, and the
Brouwer, E. 1965. Report of the sub-committee on constants
efficiency of doing work obtained in this study to
and factors. Procrrrfin~~s qf tlr~ thirrl sp~~si77177
~7 fhe c~7rrg~/
estimate the extra net energy used for work
~r~tnPo/is~r $ fi7r7n nrGn~/s (ed K. L. Blaxter), EAAP
according to the method of Lawrence and Stibbards
P7rblicalh 110. I?, pp. 441-443. Academic I’ress, London.
(1990) gave values for the extra energy for work
Clar, U. 1991, Entwichlung ciner Feldmethode zur Messung
ranging from 0.10 to 0.89 times daily maintenance
des Energie Umsatzes bei Zugticren. PhD. thk, Ll~7i?hf!/
energy requirements, depending on the number of
ojHollerl/lrinl.
hours worked and the size of the animals used. The
Dijkman, J. T. 1993. The measurement of draught ruminant
estimated energy expenditure during 5 to 6 h of
energy expenditure in the field. Ph.0. fk$it;, lhiwrrify of
ploughing sandy soils varied between 0.50 for oxen
Eili~~bw~h.
weighing 500 kg to 0.89 X maintenance for oxen
Food and Agriculture Organization. 1987. Afik~~
weighing 300 kg. The energy expenditure during
a,qricrrlfwe: thr ~ext 1.5 1/n7rs. The Fond and Agric&re
work of the 500-kg ox was comparable to that seen in
Organization of the Unitéd Nations, Rome, Italy.
the other studies (Lawrence, 1985; Pearson, 1989;
GO~, M. R. and McDowell, R. E. 1980. A~~i.vrnl trnc‘tiofi:
Mahardika rt RI., 1994), but energy expcnditure
~[/idcli,~s jbr mfilisafion. Corne11 International Agricultural
during work by the 300-kg ox was higher than that
Development m i m c o g r a p h ( U S A ) n o . Pl. Corne11
generally found, suggesting that the larger animal
University, Ithaca, NY.
Nutrition of draught cattle-1
215
Humpfrey, S. J. E. and Wolff, H. S. 1977. The Oxylog.
Mahardika, 1. G., Sastradipradja, D., Sumadi, 1. K.,
~orn’rfd c>fPhl/SiOkJ~!/, bf7dof7 267: 12.
Sutardi, T. and Kartiaso 19% Energy expenditure of \\va&r
Lawrence, P. R. 1985. A review of the nutricnt reyuirements
buffaloes for draft power in the humid tropics. In
of
Sustnir7nblr s7t7oll-sdc rnr77h7i7f pc~d7rcfior7 in atw-arid trr7d
draught oxen. In Drn7,gl7f nr7imnl po7cw,~v prod7~fk~71 (cd.
J, W. Copland). Pmwdi77~~s $[772 ii7frlnnfio77ol workslroy, /afws
sd4777717id trqisnl mm (ed K. Becker, 1’. R. Labvrence and
C o o k U77izvrsify, ~CJilV75Zdk, A7MT7lil7, 10-16 ]7dy 198~5.
E. Ii. Lilrskov). Procwdir7~~s ~,f 1717 i77tc~rr7~7fionnl
mrkshop 077
ACIAR proceedings scries n o . 10, pp. 59-63. ACIAR,
Septcrr7brr 24 1994 at tl7r I77sfif7lfr for Al7iir7nl Pwtf7rt-tior7 ii7 f17ta
Canberra.
Tuopic’s i77u-i Sd-Tro$cs, U77 iivrsify c$ Hol7r77l7~i177,
SfuttXnrt,
Gerrrrnr7!/, pp. 127-130.
Lawrence, P. R. and Becker, K. 1994. The cffects of draft
powcr on feed utilisation and reproduction in cattle and
Panin,
A. and Ellis-Jones, J. lY94. Increasing thc
buffaloes. hi S~rtainnbk w7n/l-sa7/r
rur7rir7a77t producfio77 in
profitability of draught animal power. In I7r7pwir7g mrirrrt7l
scvri-nuid ar7d ~7717-l717777id fropicrrl ~77~7s (ed. K. Becker, 1’. R.
fnrcfio77 fd777olpp (rd. P. FI. Starkey, E. Mwenya and J .
Lawrencc and E. R. Qrskovj. Pmwdir7gs o,f 1771 ir7fcwrr7/io77~7l
Starcs). Procwd777~yL; of fhe fïrsf w~~r~ksl7q~
of tlrc A77irr7~7l Tmtio77
workd7op 077 Sej~trwibtv'
74 7994 r7f thé htitufr for Auim7l
.Nrfwod- $r Enstrr77’ (7rnl’ So7rthrrr Afii(:l7 fATNESAJ, 18-23
Pmrl7~cfio77 i77 fl7r Tropics cli7d Suh-Trqlic>. U7&ersity o
f
/nrzrrnry 7992, hsalin, Zafnbi~7,
pp. 91-l 03. Technical Ccntrc
Hoh~i7lrrir17, Stfrttgnrt, Gurr7m7y, pp. 101-121
for Agricultural Co-operation (CTA), Wageningen, The
Netherlands.
Lawrence, P. R., Buck, S. F. and Campbell, 1. lY&Ya. The
metabolic r a t e o f o x e n a f t e r w o r k . P~mmdi77~~5
c’f’ tlw
Pearson, R. A. 1989. A comparison of drnught cattle (Ho>
iN77t7+fior7 Socif’t!/ 48: 1531% (abstr.).
i77rfk.77~)
and buffaloes (BuPnl7r~ buhr7lis) tartine lords in bot
candi tiens. A77im72 Pro&tio77 49: 355-393.
”
Lawrence, P. R. and Pearson, R. A. 1985. Factors affecting
the measurement of draught force, work output and power
Starkey, P. H. 1994. A wide view of animal traction
of oxen. ~01r777nl
of A8ric7dt77n7I Scicvzcc~, Cmrbri[fp 105:
highlighting some key issues in eastern and southcrn
703-714.
Africa. In Inrprovir7g mrinrnl fmtior7 trc~777ch~~~~~ (ed. P. H.
Starkcy, E. Mwenya and J. Sbres). Prmwdi77~~s qf tlw jirst
Lawrence, P. R., Pearson, R. A. and Dijkman, J. T. .lYYl.
zcwkd7op oj fhc Ani717nl Tmfior7 Nctzrwk for Ensttw7 (777~1
‘Techniques for mcasuring whole body energy expenditure
So7dhrrrr Afiicn (ATNESA), ?R-2.3 /o7777m$ 1997, Lusab7,
of working nnimals: a critical rcview. In Isotope (777~1 nlntcd
ZnftrPin, pp. 66-81. Technical Centre for Agricultural Co-
t~d777iq7vi;
ir7 1777i771d
pvad~tiof7 m7d hrnlfh. Proceedings of an
operation (CTA), Wageningen, The Netherlnnds.
international symposium on nuclear and related techniques
in animal production and henlth, International Atomic
Steven, A . P . 1 9 9 4 . Improring animal-powerrd tillagc
Energy Agehcy and the Food ami Agriculture Organization
systems and weeding technology. In hrprovi77,y r7r7irrrnl
of thc United Nations, Vienna, 15-19 Aaril 1991,
fmcfh fcrhnolyy (ed. P. H. Starkey, E. Mwenya and J.
DD.
II
I
21 l-232. International Atomic Energy Agency, Vienna.
Stares). Procccd7r7gs of thcfirsf workshop c!ffh~ Aui777nl Tmctio77
Ncfumk f o r Enstrrrz n77rf So77tl7cwr Qricn (ATNESA), 18-23
Lawrence, P. R., Sosa, R. and Campbell, 1. 1989b. The
]munry 1992, Lnsakn, Znnrh7, pp. 168-181. Technical Centre
underlying ‘resting’ energy consumption of oxen during
for Agricultural Co-operation (CTA), Wageningen, The
work. Procrcdi77gs of trlr N77frifiu7 Society 48: 154A (abstr.)
Netherlands.
Lawrence, P. R. and Stibbards, R. J. 1YYO. The energy costs
Zerbini, E., Gemeda, T., O’Neill, D. H., Howell, P. J. and
of walking, carrying and puliing loads on Bat surfaces by
Schroter, R. C . 1992. Relationships between cardio-
Brahman cattle and swamp buffalo. Anirrrnl Prod77ctior7 50:
respiratory parameters and draught work output in Fl
29-39.
crossbred dairy CO~S under field conditions. Anirru7l
McLean, J. A. and Tobin, G. 1987. Atzirt7nl n77d l111117~717
Pmfirctio77 55: l-10.
mlorin7rtry. Cambridge University Press, New York.