171. When the shells and corals in limestone are quoted by mineralogists, it is not always considered in what state they are found. In general, they have a sparry structure, very different from that of the original shell or coral, of which, however, they retain the figure with wonderful exactness, though probably sometimes altered in size. Though sparry, they are often foliated, and preserve their animal, in conjunction with their mineral, texture. Now, this crystallization is a mark of some operation, quite different from any that can be ascribed to the water in which these bodies had their origin, and by which they were brought into their place. They were impervious to water; and it cannot be said that their sparry structure has been derived from the percolation of that fluid, carrying new calcareous matter into their pores. We can account for the change produced in them, I think, only by supposing them to have been softened by heat, so as to permit their parts to arrange selves anew, and to assume the characteristic organization of mineral substances. All shells have not the change effected on them that is here referred to; those in chalk, for instance, retain very much their original form in all respects. This is what we might expect from the very different degree of intensity, with which the mineralizing cause has acted on chalk, and on limestone or marble. In general, it is in the hardest and most consolidated limestone, that the marine objects are most completely changed into spar. It would be exceedingly interesting to examine, whether any of the phosphoric acid remains united to shells of either of these kinds. We might most readily expect it to be united, in a certain degree, to the shells that are least mineralized. This experiment would enable us also to appreciate the force of Mr Kirwan's argument against the finer marbles, such as the Carrara, containing shells.[80] This argument proceeds on an experiment, mentioned in the _Turin Memoirs_ for 1789, from which it appears, that no phosphoric acid is found in pure limestone; and its absence, Mr Kirwan says, cannot be attributed to fusion, as phosphoric acid is indestructible by heat. [Footnote 80: Geol. Essays, p. 458.] He calls this a demonstration; but, in order to entitle it to that name, it will be necessary, first, to prove, that phosphoric acid exists in those limestones which evidently consist of shells in a mineralized state. If these are found without phosphoric acid, it is evident that the preceding argument fails entirely. If they are found to contain that acid, it will then no doubt afford a probability, though not a demonstration, that Carrara marble does not directly originate from shells.