For assessing the diet it is necessary to identify parts that are retained in the
droppings but for which there are only a few (ideally one) per individual.
Mouthparts are usually hard and are often used for this purpose with legs as an
alternative. A particular pair of legs is most characteristic for some insect taxa and
are therefore best to count. For example, the front legs of carabid beetles have
characteristic notch-like indentations. In some cases it is not possible to identify
such rare and distinctive features and instead it is necessary to count the number
of broken pieces of a large body part (e.g. pieces of snail shell or millepede
exoskeleton) or body parts of which each prey individual has many (e.g chaetae
of earthworms, wing scales of moths).
For the microscopic examination of droppings samples it is best to spread the
sample in a sufficient volume of ethanol that pieces do not obscure one another,
but it is inconvenient to handle a large volume of liquid by making temporary
mounts on microscope slides or attempting to scan the whole of a petri dish.
Figure 10.1 shows one method of overcoming these problems. The dish can be
inscribed with radial marks (say at 10 degree intervals) to provide sub-units that
can be used for sampling. This helps in counting a mixture of fragments of rare
and common prey types. For example, Stone Curlew Burhinus oedicnemus
droppings often contain hundreds or thousands of earthworm chaetae, each of
Diet composition| 237Fig. 10.1A modified petri dish for counting fecal samples. An internal ring is glued
in, so that the channel between the internal ring and the edge of the petri dish is the
same width as the microscope field of view, which makes it much easier to search
the entire sample without double counting or missing any (From Green and Tyler
1989).