In chipping a block down to the size required it is best to hold it freely in the left hand, protecting the actual specimen with the palm where possible, and taking the surplus edges away by means of short sharp blows from the hammer, striking so that only small pieces come away with each blow. For delicate specimens it is wise to leave a good margin of the matrix round the specimen, and to do the final clearing with a thin-bladed penknife, taking away small flakes of the stone with delicate taps on the handle of the knife. Specimens from fine sandstones, shales, and limestones are usually thoroughly hard and resistant, and are then much better if left without treatment; by varnishing and polishing them many amateur collectors spoil their specimens, for a coat of shiny varnish often conceals the details of the fossil itself. Impressions of plants on friable shales, on the other hand, or those which have a tendency to peel off as they dry, will require some treatment. In such cases the best substance to use is a dilute solution of size, in which the specimen should soak for a short period while the liquid is warm (not hot), after which it should be slightly drained and the size allowed to dry in. The congealed substance then holds the plant film on to the rock surface and prevents the rock from crumbling away, while it is almost invisible and does not spoil the plant with any excessive glaze. 2. For specimens of _casts_ the same treatment generally applies, though they are more apt to separate completely from the matrix after one or two sharp blows, and thus save one the work of picking out the details of their structure. 3. Those blocks which contain _petrifactions_, and can therefore be made to show microscopic details, will require much more treatment. In some cases mere polishing reveals much of the structure—such, for instance, were the “Staarsteine” of the German lapidaries, where the axis and rootlets of a fossil like a treefern show their very characteristic pattern distinctly. As a rule, however, it is better, and for any detailed work it is essential, to cut thin sections transversely across and longitudinally through the axis of the specimen and to grind them down till they are so transparent that they can be studied through the microscope. The cutting can be done on a lapidary’s wheel, where a revolving metal disc set with diamond powder acts as a knife. The comparatively thin slice thus obtained is fastened on to glass by means of hard Canada balsam, and rubbed down with carborundum powder till it is thin enough.