In other specimens, there is first a lining of colourless siliceous crystals, then another lining of amethystine crystals, and sometimes within that, fuliginous crystals.  Upon these fuliginous and amethystine crystals are many sphericles or hemispheres of red compact iron-ore, like haematites.

In others, again, the primary crystals are siliceous, and the secondary calcareous.  Of this kind, I have one which has, upon the calcareous crystals, beautiful transparent siliceous crystals, and iron sphericles both upon all these crystals, and within them.

Lastly, I have an agate formed of various red and white coats, and beautifully figured.  The cavity within the coated part of the pebble is filled up without vacuity, first, with colourless siliceous crystals; secondly, with fuliginous crystals; and, lastly, with white or colourless calcareous spar.  But between the spar and crystals there are many sphericles, seemingly of iron, half sunk into each of these two different substances.

From these facts, I may now be allowed to draw the following conclusions: 

1_st_, That concretion had proceeded from the surface of the agate body inwards.  This necessarily follows from the nature of those figured bodies, the figures of the external coats always determining the shape of those within, and never, contrarily, those within affecting those without.

2_dly_, That when the agate was formed, the cavity then contained every thing which now is found within it, and nothing more.

3_dly_, That the contained substances must have been in a fluid state, in order to their crystallizing.

Lastly, That as this fluid state had not been the effect of solution in a menstruum, it must have been fluidity from heat and fusion.

Let us now make one general observation and argument with regard to the formation of those various coated, concreted, crystallized, and configured bodies.  Were the crystallization and configuration found to proceed from a central body, and to be directed from that centre outwards, then, without inquiring into collateral appearances, and other proofs with regard to the natural concretion of those substances, we might suppose that these concretions might have proceeded from that central body gradually by accretion, and that the concreting and crystallizing substances might have been supplied from a fluid which had before retained the concreting substance in solution; in like manner as the crystallizations of sugar, which are formed in the solution of that saccharine substance, and are termed candies, are formed upon the threads which are extended in the crystallizing vessel for that purpose.  But if, on the contrary, we are to consider those mineral bodies as spheres of alternate coats, composed of agate, crystal, spars, etc.; and if all those crystallizations have their bases upon the uncrystallized coat which is immediately external to it, and their apices turned inwards into the next internal solid coat, it is not possible to conceive that a structure of this kind could have been formed in any manner from a solution.  But this last manner is the way without exception in which those mineral bodies are found; therefore we are to conclude, that the concretion of those bodies had proceeded immediately from a state of fusion or simple fluidity.