impossible, to assess accurately the
botanical composition of herbage ingested
by the grazing herbivore. The microscopic
examination of indigestible plant cuticle
fractions in the faeces of herbivores has
been most widely used, but problems
associated with this and other procedures
have been discussed recently (Dove and
Mayes, 1996).
It has long been known that differences
in the alkane composition of different plant
species could be used in chemotaxonomic
studies. Diet composition in terms of plant
species or plant parts can now be estimated
from a knowledge of the alkane profile or
‘fingerprint’ of a representative sample of
the mixed diet, in the form of oesophageal
extrusa (Dove et al., 1993) or faeces (Salt et
al., 1994), and the alkane profiles of the
component plants or feeds making up the
diet. When faecal samples are used, correc-
tions for incomplete recovery may be
necessary in order to prevent a bias
towards dietary components with a
predominance of longer chain alkanes
(Mayes et al., 1994; Salt et al., 1994).
Simultaneous equations (Dove, 1992) or
least-squares optimization methods (Dove
et al., 1993; Salt et al., 1994; Newman et
al., 1995) have been used for estimating
diet composition. Least-squares optimiza-
tion methods tend to minimize the squared
deviations between the observed alkane
pattern in the consumed herbage or faeces
and that indicated by the predicted diet
composition, but predicted results are
otherwise similar to predictions obtained
with simultaneous equations (Dove and
Mayes, 1996). By using alkane concentra-
tions in the least-squares minimization
procedure, an estimate of whole diet
digestibility can also be obtained.
Theoretically, the number of com-
ponent species which could be determined
in a mixture is limited to the number of
alkane markers present, which may vary
from eight to 15. However, in practice, the
number may be smaller due to possible
similarities in the alkane profile of
different component species, or the possi-
bility of certain component species having
very low total alkane contents. Alkane
profiles of different plant parts such as leaf
and stem may also differ and must be taken
into account when representative samples
of ingested forage are taken. These
differences have been exploited to estimate
the intake of plant parts by the animal
(Dove et al., 1992).
In order to validate diet composition
estimates, known plant mixtures have been
analysed for alkanes and their mixed ratios
predicted by means of the alkane proce-
dure (Fig. 12.1A). The species composition
of three- and four-component mixtures has
also been estimated (Dove, 1992). Several
validation studies showed that alkane
profiles can provide accurate estimates of
the diet composition in grazing or brows-
ing animals (Fig. 12.1B) (Dove and Mayes,
1996). The use of alkanes to estimate
species selection is not restricted to
ruminants, but has also been applied to
non-ruminants such as pigs, horses and
mountain hares (Dove and Mayes, 1996).Conclusions
In recent years, major advances have been
made in the use of indigestible markers for
studying digesta kinetics, rumen protein
synthesis, digestibility, herbage intake and
species selection. Estimates of digesta flow
rates are largely influenced by the marker
and model used, and caution must be
exercised in choosing experimental proce-
dures. Due to a lack of agreement between
procedures, estimates of digesta flow para-
meters must often be regarded as relative
indices rather than absolute values.
Quantifying microbial protein synthesis is
hampered by unequal^15 N enrichment and
internal marker to N ratios of protozoal and
bacterial pools. A major advance has been
the advent of n-alkanes as internal and
external markers. Estimates of digestibility,
intake and species selection by means of
alkane markers appear to be more accurate
than with other procedures. The use of
markers has greatly enhanced our
knowledge of farm animal nutrition and in
particular the nutrition of free-ranging
animals.Use of Markers 271