1991). It is noteworthy that provided a pair
of alkanes of similar recoveries are used
for intake predictions, their incomplete
recoveries in the faeces would not matter,
since these values would cancel out (see
Equation 12.9).
It is important to ensure that the
herbage sample analysed is representative
of the herbage actually consumed by the
experimental animal in terms of alkane
composition. Hand-plucked samples may
be adequate in the case of uniform,
cultivated pastures, but may be inadequate
for mixed pastures of complex composi-
tion. Representative extrusa samples may
be obtained from oesophageal fistulated
animals fitted with bags or remote-
controlled oesophageal valves (Raats and
Clarke, 1992; Raats et al., 1993).
Faecal concentrations of dosed alkanes
reach equilibrium after 5 or 6 days.
However, diurnal variation in faecal
marker concentration may occur, depend-
ing on the time and frequency of marker
dosing and faecal collection. The diurnal
variation in the faecal alkane concentration
of sheep dosed once daily with alkane-
impregnated paper or twice daily with
alkane-coated cellulose powder is small,
but may be higher with cattle (Dove and
Mayes, 1996). With markers such as Cr 2 O 3
and Ru-phe, the diurnal variation in
absolute marker concentration is important,
while in the case of alkanes an error will
only be introduced if diurnal variation in
the ratio of faecal concentrations of herbage
and dosed alkanes exists.
Alkanes have many advantages over
most other indigestible markers used for
intake estimations in free-ranging animals.
Intake estimates are not dependent on in
vitrodigestibility determinations. The actual
feed digestibility of individual animals is
calculated and the method is therefore well
suited for studying genetic differences in
animals regarding intake, digestibility and
feed conversion. The feeding of supple-
ments is accommodated by the method and
has been validated by several studies
(Mayes et al., 1986; Dove et al., 1995).
Because alkanes are simple straight-chain
hydrocarbons, their extraction, purification
and quantitative separation by gas
chromatography are relatively simple
procedures and involve less work than the
chromium oxide technique.
In ten validation studies reported by
Dove and Mayes (1996), the largest dis-
crepancy between known and estimated
intake amounted to 2.6%. Piasentier et al.
(1995) showed that the external alkane
marker (C 32 ) in his study gave a higher
faecal recovery than the natural odd-chain
alkanes, C 31 and C 33 , which led to an
underestimation of intake by 3% on
average. In comparison, the chromium
oxide method overestimated intake by 5%
due to the incomplete faecal recovery of
the marker.
Supplement intake in the field situa-
tion can be estimated by mixing a marker
such as ytterbium with methyl cellulose
and spraying it onto a grain supplement.
After total faecal collection, supplement
intake can be estimated from the recovery
of marker in the faeces (Curtis et al., 1994).
Intake of supplement by individual
animals has been determined by labelling
the supplement with Cr 2 O 3 and measuring
total faecal excretion (Dove and Coombe,
1992). A disadvantage of this method is the
need to measure total faecal excretion. By
using chromium oxide and tritiated
gypsum as markers and measuring the
accumulation of tritium in the body water
pool of animals, the need to measure total
faecal excretion is eliminated. Both
methods gave similar results, although the
chromium method had smaller standard
errors, while the tritiated gypsum method
gave more accurate results under field
conditions at low levels of feed intake
(Dove, 1984; Dove and Coombe, 1992).
Because of its near absence in feed-
stuffs and its high faecal recovery (0.95),
hexatriacontane (C 36 ) could be used as a
supplement marker by spraying it onto the
supplement.Herbage species selectionBefore the advent of the alkane marker
technique, it was extremely difficult, if not270 J.P. Marais