Methods of Faecal Analysis

Christine B. de Jong, 2006. Resource Ecology Group, Department of Environmental Sciences, Wageningen University, Wageningen, Netherlands

1. Introduction

Microhistological faecal analysis is a non-destructive method of assessing botanical diet composition of wild herbivores. During faecal analysis, epidermis and/or cuticle fragments of ingested plants are identified and quantified (De Jong et al., 1995, 1997, 2004, Hooimeijer et al., 2005). This is possible as the plant cuticle, an indigestible layer covering the epidermis (fig. 1), bears a specific pattern of underlying epidermal cells and hairs along with structures of its own. This pattern is best developed in mature leaves and can be identified down to genus or species level, even after passage through the gut of a herbivore.

Figure 1
Figure 1. (A) Structure of epidermis and cuticle (after Stace, 1965); (B) Epidermis patterns (click to enlarge)

2. Materials and methods

A. Reference material

Fresh or dried parts of relevant food plants are cut in 1 cm pieces. Top and bottom 2 cm of grass leaves are omitted. They are cleared in household bleach for one night then washed out in water. Epidermis fragments are stripped off and mounted in glycerol. Slides are sealed with nail polish. Alternatively, when a blender is available grass leaves can be blended in water for one minute after bleaching and then strained over a plankton sieve. Grab samples of epidermis fragments are mounted. Photomicrographs of these slides are used to identify the cuticle fragments present in the faeces. For a direct comparison of cell and cell wall sizes in different species and in epidermis fragments in the microscope field, photographs of standard magnification and size are used.

Two different methods of collecting reference material have been used for the reference collection provided on this website. Movie 1 shows the advanced method, while movie 2 shows a simplified method which is also effective for most samples.

Movie 1. Advanced method
Movie 2. Simplified method

B. Faecal samples


Fresh dung pellets are collected in the field or from the hind gut of culled animals, sterilized on the outside with household bleach and preserved by freezing or alternatively: put in formalin or Formalin Alcohol Acetic acid (Anthony and Smith, 1974). Samples consist of a complete defecation or latrine (mice and voles), five pellets (deer, rabbits) or two pellets/tablespoons (horses, cattle, large wild herbivores) from every defecation collected at different sites and periods (Groot Bruinderink et al., 2000). All samples from the same period and population are pooled and mixed thoroughly. Mixed samples are heated to 115 °C to 125 °C in water for at least 2 hours and left to soak overnight. In order to separate inner tissue from epidermis and cuticle, a 5 g subsample is washed in a blender with tap water and strained over a plankton sieve (0.1 mm) then stored in 70% ethanol. In field conditions: Bleach dung overnight or boil in water for at least 1 h and leave to soak overnight, repeat boiling and cooling, then wash out in water and strain over a plankton sieve. Boiled dung is then stored in ethanol 70%. Bleached material can be kept in just water after washing. Bleaching and straining is used where no blender is available or when very long fibres are present.

The number of pellet groups to be sampled is at least 10-15 per population and phenological period, individual pellet groups are not repetitions as every meal changes the availability of food items for the next. Also, herbivores vary their choice of food plants between meals in order to maintain a varied but limited consumption of secondary chemicals (Freeland 1991, McArthur and others 1991). For a representative diet estimate over a given period at least 10-15 pellet groups should be sampled (Homolka 1987). Standard deviations decrease with increasing sample sizes but cannot be calculated as faeces vary in composition (Van der Voet,

Rumen or stomach samples
Rumen or stomach samples consist of 10 g fresh plant material (or 2.5 g dry) taken at random from the rumen or stomach content of every culled animal or of integral stomach contents for smaller animals. All samples from the same period and population are pooled and mixed thoroughly. Sterilize and store as for faecal samples if the material has not been sterilized previously.

C. Identification and quantification of epidermis fragments

The subsample is transferred into a Petri dish and allowed to settle. Using a Pasteur pipette, ten random grab samples of the residue are then taken (fig. 2). Each droplet is put on a glass slide, spread out evenly and covered with a 2.4 cm cover slip. On each slide, ten fragments of epidermis are identified in at least two transects using photomicrographs of epidermal material and measured using a grid of 0,01 mm2 squares in the microscope eyepiece. Cuticle fragments that cannot be identified further than ‘monocot’, ‘dicot’, ‘moss’ etc. are listed as such. If no grid is available percentage cover is estimated for every taxon in at least 5 fields per slide.

Figure 2
Figure 2. Random sampling of squares (van der Steege, 1981)
The abundance of each species is represented as a percentage of the total measured fragment area (Putman, 1984, Sparks and Malechek, 1968, Stewart, 1967, Cid and Brizuela, 1990, Alipayo et al., 1992, Homolka and Heroldová, 1992). At least 100 cuticle or epidermis fragments have to be identified in each sample in order to identify all species making up 5% or more of the diet (Stewart, 1967). Fragments under 0,01 mm2 are ignored as in many species they cannot be identified (Van der Steege, 1981; Buil, 1982). Loose hairs and inner tissue fragments can be an indication for the presence of a certain species but have to be ignored for quantitative analysis.

D. Comparing diet

To compare diets Kulcynski’s Similarity Index (KSI) is used: KSI = Σ2Cx100 / Σ(A+B)
– C: the lesser percentage of a taxon found in each of the two diets.
– A and B: the percentages of this taxon found in each of the two diets (Cuartas and Garcia-Gonzalez, 1992).

In previous research using this method (Groot Bruinderink et al., 1997), the KSI was assessed of a series of 20 duplicates (subsamples of the same mixed sample) and averages and standard deviation were computed. All diets were compared both at species level and when all taxa were grouped in six major categories. A pair of diets can be considered similar when its KSI is within the standard deviation of the duplicate range (H. van der Voet, pers. comm.). Analysis of the duplicates from the earlier study indicated that a diet difference of 1.73 x the std of the duplicate-average is significant (P=0.05), 1.73 being the critical t-value of a one-sided T-test.

Table 1. Kulcynski’s Similarity Index

At taxon level 6 categories
KSI average 69 84
Standard deviation -8 -7
1,73 x standard deviation -14 -12
KSI diets similar ≥ 61 ≥ 77
KSI significant diet difference ≤ 55 ≤ 72

3. Identification


A grass epidermis has a specific pattern of long and short cells (Prat, 1932, Watson en Dalwitz, 1988, Metcalfe, 1960, Ellis, 1979). Different grass species have different patterns.

Figure 3
Figure 3. Differentiation of long and short cells in a grass epidermis (after Prat, 1932). (A) long and short cells are formed; (B) short cells divide, long cells start stretching; (C) short cells are differentiated into stomata, silica- and cork cells or prickle hairs
Figure 4
Figure 4. Example of a grass epidermis: Saccharum (Esau, 1960)
Figure 5
Figure 5. Examples of dicotyledoneous epidermis in surface view illustrating patterns formed by guard cells and surrounding cells (Esau, 1960).
Figure 6
Figure 6. Stomata in a dicot (Prunus), a grass (Oryza) and a Gymnosperm (Pinus), Esau 1960.
Identification scheme
  1. Epidermis cells not differentiated from inner tissue > BRYOPHYTA
    Epidermis cells differentiated from inner tissue > 2
  2. Epidermis cells usually rectangular or spindle-shaped, arranged parallel, cell walls straight or wavy, short cell walls more or less at a right angle to the long axis of the cells > 3
    Epidermis cells arranged in another way
  3. Guard cells of stomata in a hollow below level of other epidermis cells > GYMNOSPERMS
    Guard cells of stomata level with other epidermis cells > GRAMINOIDS, 4
  4. Between the long epidermis cells short cells are placed in a regular pattern, in pairs (cork/silica) or single (prickle)hairs (fig. 3, 4). Stomata paracytic (see fig.5), guard cells dumbbell-shaped (as in fig. 6, Oryza) > GRASSES
    No short cells between the long ones, stomata paracytic, guard cells not dumbbell-shaped > 5
    Cells thin-walled, cell walls straight, bent or wavy > CYPERACEAE, JUNCACEAE
    Cells thick-walled, square or angular, same size or smaller than stomata > ARECACEAE
  5. Walls corky or not. Cells more or less parallel, most short cell walls at a short angle to long walls. Stomata and hairs may be present > STEM OR VEIN OF FERN OR DICOTYLEDON
    Cells not arranged parallel. Epidermis cells more or less thin-walled, cell walls straight, bent or wavy. Stomata and hairs may be present > 6
  6. Plastids only present in guard cells of stomata > LEAF OF DICOTYLEDON OR SOME MONOCOTYLEDONEOUS FAMILIES.
    Plastids present in all epidermis cells > LEAF OF FERNS

4. References

  • Alipayo, D., Valdez, R., Holecheck, J.L. and Cardenas, M., 1992. Evaluation of microhistological analysis for determining ruminant diet botanical composition. J. Range Manage., 45; 148-152.
  • Anthony, R.G., Smith, N.S., 1974. Comparison of rumen and fecal analysis to describe deer diets. Journal of Wildlife Management 38: 535-540.
  • Buil, M., 1982. Een vergelijking tussen rumeninhoud- en faecesanalyse m.b.t. de dieetsamenstelling van één paardantiloop (Hippotragus equinus koba, Gray 1872). Vakgroep Natuurbeheer. Verslag no. 659, Landbouwho- geschool, Wageningen.
  • Cid, M.S., Brizuela, M.A., 1990. Grass blade and sheath quantification by microhistological analysis. J. Wildl. Manage. 54: 349-352.
  • Cuartas, P. and Garcia-Gonzalez, R., 1992. Quercus ilex browse utilization by Caprini in Sierra de Cazorla and Segura (Spain). Vegetatio 99-100: 317-330.
  • De Jong, C.B., Gill, R.M.A., Van Wieren, S.E., Burlton, F.W.E., 1995. Diet selection in Kielder Forest by roe deer Capreolus capreolus in relation to plant cover. For. Ecol. Manage. 79, 91-97.
  • De Jong, C.B., Cornelissen, P. en Vulink, J.T., 1997. Grote grazers in de Oostvaardersplassen. Dieetsamenstelling op basis van faecesanalyse. RIZA werkdocument 79.11-4X.
  • De Jong, C.B., Van Wieren, S.E., Gill, R.M.A., Munro, R., 2004. Relationship between diet and liver carcinomas in roe deer in Kielder Forest and Galloway Forest. Veterinary Record, 155:197-200.
  • Ellis, R.P. 1979. A procedure for standardizing comparative leaf anatomy in the Poaceae. II. The epidermis as seen in surface view. Bothalia 12: 641-671.
  • Esau, K., 1960. Anatomy of Seed Plants. 1st ed. New York: Wiley.
  • Groot Bruinderink, G.W.T.A., Lammertsma, D.R., Kuiters, A.T., Wösten, G., Den Hollander, W., De Jong, C.B., 2000. Terreingebruik en menusamenstelling van reeën, damherten en runderen in de Mantelingvan Walcheren. Tussenreportage Alterra.
  • Groot Bruinderink, G.W.T.A., Van Wieren, s.e., Hazebroek, e., Den Boer, M.H., Maaskamp, F.I.M., Lamers, W., Slim, P.A., De Jong, C.B. (1997). De ecologie van hoefdieren. In Hoefdieren in het boslandschap. Eds Van Wieren, S.E., Groot Bruinderink, G.W.T.A., Jorritsma, I.T.M. & Kuiters, A.T. Backhuys Publishers Leiden.
  • Hooimeijer, J.F., Jansen, F.A., De Boer, W.F., Wessels, D., Van der Waal, C., De Jong, C.B., Otto, N.D., Knoop, L. , 2005. The diet of kudus in a mopane dominated area, South Africa. Koedoe 48: 93- 103. Homolka, M., 1987. Problems associated with investigations into the diet of the European hare. Folia Zoologica 36, 193-202.
  • Homolka, M. and Heroldová, M., 1992. Similarity of the results of stomach and faecal contents analyses in studies of the ungulate diet. Folia Zool., 41: 193-208.
  • McArthur, C., Hagerman, A. & Robbins, C. T., 1991. Physiological strategies of mammalian herbivores against plant defenses. In Plant Defenses Against Mammalian Herbivory. Eds R. T. Palo, C. T. Robbins. Boca Raton, CRC Press. pp 103-114.
  • Metcalfe, C.R., 1960. Anatomy of the Monocotyledons. I Gramineae. Clarendon Press, Oxford. 731 pp.
  • Prat, H, 1932. L’épiderme des Graminées. Étude anatomique et systematique. Ann. Sci. Nat. Bot. Sér. 10,14: 117-325.
  • Putman, R.J. 1984. Facts from faeces. Mammal Rev., 14: 79-97.
  • Sparks, D.R., Malechek, J.C. 1968. Estimating percentage dry weight in diets using a microscopic technique. J. Range Manage., 21: 264-265.
  • Stace, C.A., 1965. Cuticular studies as an aid to plant taxonomy. Bull. Brit. Mus. (Nat. Hist.) Bot. 4: 1-78.
  • Stewart, D.R.M., 1967. Analysis of plant epidermis in faeces: a technique for studying the food preferences of grazing herbivores. J. Appl. Ecol., 4: 83-111.
  • Van der Steege, J.G., 1982. Voedselkeuze van een aantal herbivoren in West-Afrika, bepaald d.m.v. faecesanalyse: techniek en resultaten. Vakgroep Natuurbeheer, Verslag nr. 728. Landbouwhogeschool Wageningen.
  • Watson, L., and Dallwitz, M. J. (1992 onwards). ‘Grass Genera of the World: Descriptions, Illustrations, Identification, and Information Retrieval; including Synonyms, Morphology, Anatomy, Physiology, Phytochemistry, Cytology, Classification, Pathogens, World and Local Distribution, and References.’ Version: 28th November 2005.