The isolated radula has to be cleaned and freed from certain membranes and it should be made as flat as possible. These operations are carried out on a microscope slide in a small drop of water (or an aqueous stain, cf. 5) or acetic acid. A binocular dissecting microscope is indispensable for all save the largest material. First of all it is necessary to find out which is the side that bears the teeth, because the radula has to be mounted the right side up, i.e. teeth uppermost. As a rule the oldest (distal) part of the radula (nearest to the jaw) bends down and the nascent end bends up (see fig. 1; the side indicated by the arrow is the right side). The handling of the radula requires some skill — which must be acquired by experience — and the proper tools. Larger ones can be cleaned with paint brushes and dissecting needles, but the smaller ones need a more subtle treatment. Matchsticks sharpened to a very fine and flexible point are very suitable, but they have to be replaced after having been used once or twice. The author uses the finest and flexible needles provided by nature in the form of cactus spines. Those of Opuntia, Echinocactus and Cereus, among other species, are excellent. They can be used many times and retain their rigidity in water whereas matchsticks become soft and useless after some time. A spine can be conveniently secured by means of sealing wax, cover glass cement, resin or Canada balsam in a holder consisting of a length of glass tubing drawn out at one end (see fig. 2). When working with needles or cactus spines it is important to hold them in an almost horizontal position and to use them as scrapers rather than as needles so as to reduce possible damage to the specimen to a minimum (see fig. 3). Start in the area indicated by the point of the arrow in fig. 1 by disrupting the membrane which keeps the radula in its bent position with one needle, holding the radula in place with the other one. Then smooth out that part and continue with the nascent end (left in fig. 1). Separate the jaw and remove the membranes connecting it with the radula. Flatten the posterior end. Check the slide without a cover glass under a microscope, using a low-power objective, to see if the teeth are up; if not, turn the radula and smooth it out again. When handling minute radulae it is sometimes difficult to see if they are the right side up. If such a radula is the wrong side up, but properly smoothed out, it is better to leave it as it is (see below). Remove membranes and other rejects from the slide. Wash the radula if necessary by taking away the supernatant liquid with a strip of blotting paper and adding a droplet of clean water (preferably distilled water should be used). Leave only a small amount of liquid on the slide and apply a cover glass by dropping it at once horizontally on the specimen. Press the cover glass firmly down without causing any slippage between it and the slide (a flat end of a small cork is excellent for the purpose) and check again under a microscope. If the mount is not satisfactory, add water at the side of the cover glass, remove the latter and start all over again. If the radula appears properly smoothed out and flat, put a weight (e.g. a 25 g or 1 oz. weight) on the cover glass and leave it on till all the water has evaporated and the radula is quite dry. This weighting is not always necessary, but it is essential if the radula shows a tendency to curl up again.