places, each pillar is divided into feveral parts or joints, that seem to be placed upon one another; and indeed it is not uncommon for crystals to be formed above one another in different layers, when the folvent has been vifibly diminished at different times; but then the upper cryftals never fit fo exactly upon the lower ones as to produce connected prifms of the fame length and depth as all the ftrata taken together, but each ftratum feparately forms its own cryftals.

How then can the Giant's Caufeway, in the county of Antrim, Fingal's Cave at Staffa, and all other affemblages of pillars of the fame kind, be confidered as cryftallizations? Precipitation, both in the wet and dry manner, requires that the particles fhould be free enough to fix themfelves in a certain order; and as this is not practicable in a large melted mafs, no cryftallizations appear in it, except on its furface, or in its cavities.

Add to this, that the bafalts in a fresh fracture do not fhew a plain smooth surface under the microfcope, but appear fometimes like grains of different magnitude, and at other times refemble fine rays running in different directions, that do not correfpond with the internal ftructure of the cryftals, which I have endeavoured to examine in another place.

From what I have hitherto mentioned, the opinion that the bafalts have been produced by cryftallization, becomes at least lefs probable, whether we admit the wet or dry method. But I must not omit that the fpars exhibit a kind of crystallization, which at firt fight relem VOL. XXIII.

bles a heap of bafalts; but, upon a clofer examination, a very great difference is obferved. The form of the fpar is every where alike, but the bafalts differ from one another in point of fize and number of fides; the former, when broken, confifts of many fmall unequal cubes, but the bafalt does not feparate in regular parts, &c.

Nature's fecond method to produce regular forms is that of crufting the outer furface of a melted mafs, by a fudden refrigeration. Nature, to effect this purpose, makes ufe of polyedrous and irregular forms. If we fuppose a confiderable bed, which is become fluid by fire, and fpread over a plain, it evidently appears that the furface must first of all lofe the degree of heat requifite for melting, and begin to congeal; but the cold requifite for this purpofe likewife Contracts the uppermost congealed ftratum into a narrower fpace, and confequently caufes it to feparate from the remaining liquid måfs, as the fide expofed to the air is already too ftiff to give way. In this manner a ftratum is produced running in a parallel direction with the whole mafs, others ftill are produced by the fame caufe, in proportion as the refrigeration penetrates deeper.

Hence we may, in my opinion, very plainly fee how a bed may be divided into frata. In the fame manner the refrigeration advances on the fides, and confequently divides the ftrata into polyedrous pieces of pillars, that can hardly ever be exactly fquare, as the ftrongest refrigeration into the inner parts of the mafs advances almoft in a diagonal line from the corners. If we add to this, that a

large mafs cannot be equal throughout its compofition, nor every where liquid in the fame degree, it will be eafy to discover the caufe of feveral irregularities. If the depth of the bed is very confiderable, in proportion to its breadth, prifmatic pillars, without crofs-divifions, are produced, at leaft lengthways from the uppermoft furface downwards.

The third way is perfectly fimilar to the preceding in refpect to the effect, but is different from it by the mafs being foaked with water, and by the bursting of it afunder, which is the effect of the contraction whilft it is drying. If we fuppofe fuch a bed to be spread over a level fpace, the drying advances in the fame manner as the refrigera

tion in the former cafe.

This feparation into ftrata properly happens when a confiderable quantity of clay enters into the whole compofition, becaufe the clay decreases more than any other kind of earth in drying.

We must now examine which of these two ways may best ferve to explain the manner in which the bafalts are produced, for it is hardly poffible that they fhould have been formed by crystallization.

However well founded the opinion may appear of deducing them from a melted fubitance, feveral very confiderable objections may nevertheless be raifed against it, that I fhall not forget to mention. It feems therefore more credible to me, that they have been produced out of their fubftance whilft it was yet foft, or at least not too hard to be foftened by exhalations. If we therefore fuppofe that a bed is fpread over a place where a volcano

begins to work, it is evident that a great quantity of the water, always prefent on thefe occafions, is driven upwards in exhalations or vapours; thefe it is well known poffefs a penetrating foftening power, by means of which they alfo produce their first effect; but when they are increafed to a fufficient quantity, they force this tough moift fub. ftance upwards, which then gra dually falls, and during this time burts in the manner described above.

My reafons for this opinion are thefe; firft, we do not find the internal grain of the bafalts melted or vitrified, which however foon happens by fufion, and for that purpofe a very fmall degree of fire only is requifite. It confequently is very hard to explain how this fubftance could have been fo fluid, that no traces of bubbles appear in it (at least I have not been able to difcover any on the niceft examination into the Scotch and Icelandic bafalts) and yet when broken appear dull and uneven. I know very well that lava is feldom vitrified within; but the great number of bubbles and pores which are found in the whole mafs, are more than fuficient proofs that it has not been perfectly melted to its fmallest parts, but has only been brought to be near fluid.

Secondly, the bafalts fo much refemble the more fine trapp, both in refpect to their grain and original compofition, that they can hardly be diftinguithed in fmall fragments, as will be more plainly proved in the comparifon I hereafter make. See No. 24.

But the trapp in all probability has never been melted, at least not in thofe parts where I have

had

Almoft in all the Weft Gothia ftratified mountains, the uppermost ftratum is trapp; and it must be properly obferved that it always lies upon black allum flate. Is it therefore credible that this fubftance, which in many places exceeds a hundred yards in depth, can have been perfectly melted without caufing the flate lying beneath it to lofe fome part of its blacknefs, even in thofe places where they touch one another, as this effect may be produced in a fmall culinary fire?

There is befides a more fine kind of trapp, which is generally found in veins or loads, and frequently in very antient mountains, where not the leaft traces of fubterranean fire are to be feen.

The bafalt mountains feem to be very antient, at least I do not know that the age of any one is afcertained. Should they then be fo old, that the fubftance of the trapp was not yet perfectly hardened, when were they produced? Befides, we frequently find to this day clayey fubftances at a great depth, which are fo foft that they may be fcraped by the nail, but afterwards become very hard when expofed to the air.

There have without doubt been many eruptions of fire on the inle of Staffa, as the fituation of the pillars, and their being removed out of their places, evidently

A hard fubftance, when expofed to a degree of heat infufficient to melt the whole piece, may however be attacked by it in fome parts of the furface moft liable to become fluid. The mixture of a large mafs is feldom every where so uniform, that fome parts fhould not be more liable to melt than others.

Crooked pillars may be produced as well by the drying as the refrigeration of a liquid mafs; for this purpofe it is only neceffary that the furface fhould be bent, as the ftratum always runs in a parallel direction with it.

From what I have hitherto faid, you will perceive it is my opinion, that the bafalts have been produced by the affiftance of a fubterranean fire, but that it is not yet determined whether they have been separated by the fufion, or by drying

this laft however appears more credible to me on account of the reasons I have mentioned. For to fpeak ftrictly, the fubftances inclofed in the bafalts, though they fhould even be volcanic, do not yet with certainty prove a preceding fufion, as a fubftance foftened by water may be as proper for it as one fufed by fire. I am, however, very far from being inclined to maintain my opinion any farther than it agrees with certain experiments and obfervations.

Truth will fooner or later be difcovered; and I know nothing more derogatory to the honour of a natural hiftorian, than having wilfully obftructed its knowledge.

Homo naturæ minifter & interpres, tantum facit & intelligit, quantum de naturæ ordine, re vel mente obfervaverit, nec amplius fcit aut poteft. Baco.

H 2

Natural

Natural Hiftory of the Grana Kermes, or Scarlet Grain. From Dillon's Travels through Spain.

MONGST the various and

A valuable productions with

which the beneficent hand of nature has enriched the dominions of Spain, the Grana Kermes is chiefly deferving of attention. This va luable production had been confiderably neglected in that kingdom fince the importation of cochineal from America; however, the royal Junta de Comercio, or board of trade at Madrid, having an eye to the further advantages to be drawn from this precious article, gave orders a few years ago to Don Juan Pablo Canals, director general of the madder and dyes of Spain, to report the fate of this product; and to him I am indebted for the present information on this fubject.

The grana kermes is the corcos baphica of the Greeks; the vermiculus, or coccum infectorium of the Romans; and the kermes, alkermes, of the Arabs; being the ingredient with which the antients ufed to dye their garments of that beautiful grain colour, called coccinus, coccineus, or cocceus, different from the purpura of the Phoenicians, which at firit had been obtained from that teftaceous

fish, called the murex*. But in courfe of time the purple colour and other tints having been more eafily effected by means of the kermes, the murex was neglected on account of the expence, and the kermes we are now fpeaking of, was introduced; which giving a ftronger and brighter colour, was univerfally adopted, and fupported its reputation for ages, till the difcovery of America; as is evident from the many old tapestries, damasks, and velvet hangings, ftill preferved in cathedrals, which feem yet to retain their primitive luftre and brightness +.

In the reign of Lewis the fourteenth, Giles and John Gobelin, in the year 15667, under the patronage of Colbert, introduced the fecret into France of dying woollen of that beautiful fcarlet called after their name, which was done with the kermes that had been long in ufe in Flanders, where many old pieces of tapestry, though above two hundred years old, had fcarcely loft any thing of their bloom. But cochineal, being now introduced into the dyehoule, fo called from the Latin word coccinella, as a diminutive of coccum, and giving that brightnefs to fearlet, at firit called Dutch, and afterwards Paris fcarlet, the invention of which, according to Kun

kel,

kel, is owing to Kufter, a German, by means of a folution of tin in aqua regia; the kermes then began to decline, and yield in its turn as the murex had done before, of which Colbert makes a particular complaint, in his general inftructions to the dyers of France, in the year 1671*. Infenfibly, the kermes was totally laid afide, and cochineal made ufe of, not only in yarns, but also in filk; this new method being every where in fashion, except at Venice, and in Perfia, for fcarlet, and in other parts of the east for crim

fon.

The antients thought the kermes was a gall-nut on account of its figure and fize, not being larger than a juniper berry, round, fmooth, gloffy, and rather black, with a cinereous down. It is found ticking to the branches, or tender leaves of the oak called in Spain cofcoxa, a derivative of the Latin word cufculium, the coccus illicis of Linneus, likewife called carrasca in Spanish, from the Arabic word yxquerlat, foftened afterwards to e/carlata; being the falleft fpecies of oak, the fame which Cafpar Bauhine and other botanilts call ilex aculeata cocci-glandifera.

This tree, whofe height is about two or three feet, grows in Spain, Provence, Languedoc, and along the Mediterranean coaft; alfo in Galatia, Armenia, Syria, and

Perfia, where it was firft made. use of

Jofeph Moya, a Catalan writer of the last century, published a treatife entitled Ramillet de Tinturas, dedicated to the city of Barcelona, under the feigned name of Phefio Mayo. He fays, the kermes is common all over Spain, principally in that part of Aragon bordering on Catalonia, in Valencia, and in the bishoprick of Badajoz in Eftremadura, as likewife in Setimbre of Portugal, where it is the beft, and equal to the kermes of Galatia and Armez nia. Mr. Hellot of the French academy of fciences, in his Art of Dying, chap. 12. fays it is found in the woods of Vauvert, Vendeman, and Narbonne; but more abundantly in Spain, towards Alicant, and Valencia. It not only abounds in Valencia, but also in Murcia, Jaen, Cordova, Seville, Eftremadura, la Mancha, Serranias de Cuenca, and other places.

In Xixona and Tierra de Relleu, there is a diftrict, called De la Grana, where the people of Valencia firft began to gather it, whofe example was followed all over Spain. It has, fome years, produced thirty thousand dollars (5000l.) to the inhabitants of Xixona. In the year 1758, there went out of that town, Relleu, Buffot, Caftilla, Ibi, Tibi, Unil, Santa faz, Muchiamel, and San Juan de la Huerta de Alicante,