2. Petrifactions, isolated and detached from their rock, are the most convenient in the determination of species; but when they cannot be separated from the rock, we need not hesitate to send them engaged; it is sufficient if a portion large enough for discrimination is visible.
Among shells, those are preferable which have the mouth or hinge in view; among madrepores, those on whose surface the figures (les étoiles) are distinguishable; among vegetables, those whose leaves are distinctly expanded, (expalmées.)
3. Upon the objects transmitted it is desirable to have, at least in part, the following notices:
1. The exact place from which the object comes: this is the most important circumstance, and the easiest to obtain.
2. The kind of formation in which it is found, and a specimen of the stratum, or at least of the rock, which contained it. It is desirable that this rock exhibit remains of petrifactions similar to those found in the stratum from which it has been drawn.
3. The nature of the formation of which this stratum or rock composes a part, and specimens of as many of the superior and inferior strata as can be obtained, designating the order of superposition of the strata.
4. It is important to designate, by the same mark, all the petrifactions unquestionably found in the same stratum, or at least in the same formation. The specimens ought to be almost square—about three inches or more on a side, and one and a half thick.
5. It is equally important not to mix petrifactions found in different formations, or in different strata of the same formation; or if they are packed together, to distinguish them by numbers, marks, or labels.
When the preceding notices cannot be obtained, the first will suffice.
In order to collect the petrifactions, and to render them useful, it is not necessary to know them, nor to be perplexed to find them out; nor to be afraid of sending objects already known or of little note. A part of the preceding indications, connected with the most common petrifactions, will always render them useful. The important point then is, not to mix those which are found separate, nor to separate those which are found associated in the same stratum.