Use of n-alkanes for estimation of voluntary intake and digestibility in donkeys ( Equus asinus ) *

The aim of this work was to evaluate if the n-alkanes (C32) technique can be used to determine voluntary dry matter intake (DMI) and dry matter digestibility (DMD) in grazing donkeys. Two experiments were carried out; in trial 1 DMI and DMD were measured using n-alkanes in four caged donkeys. In trial 2, n-alkanes were used to predict DMI and DMD in six donkeys grazing a star grass sward (Cynodon nlemfl uensis). Results from both experiments suggest that the n-alkanes technique can be used to predict DMI and DMD with similar results than those obtained with other more sophisticated or expensive techniques.


INTRODUCTION
The donkeys in Mexico's rural areas are usually fed crop residues, weeds, roadside verges and natural pastures.However, little is known about voluntary dry matter intake of grazing donkeys.In the past the accurate estimates of dry matter intake (DMI) of grazing animals have been problematic because of the lack of reliable techniques.The alkanes pair technique (APT) was recently, developed and validated in a variety of ruminant species to estimate DMI in grazing animals (Mayes et al., 1988).To a limited extent this method has also been validated in horses by Ordakowski et al. (2001) and most recently in donkeys fed temperate grasses by Smith et al. (2007).The APT uses a naturally occurring odd-chain alkane (usually C31 or C33) and a dosed adjacent even chain alkane (usually C32) to estimate DMI (Mayes and Dove, 2000).There is little available information about the voluntary DMI in grazing donkeys and since DMI is the main constrain to nutrient consumption and digestibility of diets, the aim of this work was to use and validate the n-alkanes (C32) technique to determine DMI and DMD in caged animals and then test the technique in donkeys grazing star grass.

MATERIAL AND METHODS
Two 22 days trials were carried out; in each experiment the fi rst 15 days were for diet adaptation, and the last seven for sample collection.The even chain alkane was hand dosed to each donkey individually three times per day from day 1 to 21 in both trials.

Trial 1 (T1)
The alkanes pair technique (APT) was calibrated and validated under controlled conditions in this experiment before its application in grazing animals (trial 2).Four donkeys with an average liveweight (LW) of 113.4±8 kg were placed in 2.5 m 2 individual cages.LW was recorded daily during the sample collection week in order to keep the donkeys on a constant weight because they were also used to estimate digestible energy requirements (Carretero et al., 2005).A mixed diet composed of 85% chopped maize straw (5 cm particle size) and 15% lucerne hay was offered twice daily (08.00 and 16.00 h).The total daily amount of feed offered was calculated as 2% LW on a DM basis.Refusals were collected and weighted in order to calculate in vivo apparent intake.

Trial 2 (T2)
Six donkeys with an average LW 121.3±11.5 kg were kept on a star grass (Cynodon nlemfl uensis) sward for the duration of the experiment.No other forage or supplement was provided.In both experiments water was freely available.

External marker preparation
Five grams of n-alkane C32 was diluted in 4.0 l of n-heptane at 60°C and kept under agitation and constant heat.Then 4 ml of this solution were added to a 10 g biscuits which served of medium for administration.After the heptane had evaporated each biscuit contained 50 mg of alkanes.A biscuit was offered three times per day so 150 mg n-alkanes were dosed per donkey.

Sample collection
In T1 150 g of forage samples were taken daily during the collection week, dried at 60 o C to constant weight and kept for analyses.Faecal samples per donkey were collected for the last 5 days.Gravimetric measurement of the total daily faecal output (DFO) of each donkey was carried out for the fi nal 5 days of the experiment.The DFO of each donkeys was thoroughly mixed at the end of each collection day and a small sample (100 g) taken and frozen until the end of the experiment.Once the study had fi nished, all faecal samples were defrosted and 50 g sub-samples of the daily faecal samples from each donkey were pooled to make a single pooled faecal sample per donkey.These pooled samples were then dried at 60 o C and retained for laboratory analysis.For T2 grass samples collection was carried out in the last seven days using a 0.25×0.25 m 2 .Quadrants were thrown at random three times daily before and after grazing.The herbage was then cut within the quadrant using scissors approximately 1cm from the ground (as it would be grazed by the donkeys).The herbage was then dried at 60 o C. In the last fi ve days 100 g of faeces were taken directly from rectum of each animal and the process above described for T1 for faeces processing was repeated.

Alkane analysis
Alkane concentrations were determined according to Ali et al. (2004) technique.All analyses were carried out in duplicates in 25 m × 0.32 mm capillary column, type BP1 (SGE) which was fi tted in a Pye Unicam 4400 gas chromatograph.Faecal recovery of each odd chain alkanes was calculated as in Equation 1 (Ordakowski et al., 2001): where: R -faecal recovery (%); O -mean faecal output (kg DM /d); Co -oddchain alkane concentration in faeces, mg/kg DM; I -intake, (kg DM/d); Ci -alkane concentration in the diet, mg/kg DM.
The means of in vivo and alkane values for DMI, DMD and FO of T1 were analysed by t test in order to check the accuracy of APT predictions in relation to in vivo observations.

RESULTS
Alkane concentrations in the diet and faeces are shown in Table 1.During T1 the concentration of C31 alkane within the forage was highest (73.9 mg/kg DM) followed by C29 and C33.In T2, C33 had the highest concentration in forage and faeces.

DISCUSSION
The results of the present work are in line with those of Ordakowski et al. (2001) where the faecal recovery of alkanes increased as the length of the alkane chain increased.In the same line, Brosh et al. (2003) showed a positive relationship between faecal recovery and the length of alkane chain.The DMI and DMD measured in vivo and predicted by the APT (Table 2) showed no signifi cant differences, suggesting that this technique can be used to measure DMI and DMD in donkeys.Dry matter intakes observed in T2 in donkeys under grazing conditions are within the range reported for these animals by Colunga et al. (2005).In the same way the DMD of the diet observed in T2 is similar to the values reported by Carretero et al. (2005) for a similar mixed forage diet fed to Mexican donkeys too.Unfortunately there is little information on DMI and DMD published on donkeys making diffi cult to further compare our results, however the APT has been widely used in horses with good Odd alkanes results.For example, DMD from 44 to 48% were reported in horses fed with Alta fescue and Caucasian bluestem in Virginia (USA) by Crozier et al. (1997), which are similar to the DMD values observed in this work.

CONCLUSIONS
The alkanes pair technique can be used to predict dry matter intake and digestibility in grazing donkeys, with levels of accuracy and confi dence similar to those observed in other herbivores species.