Transactions of the Geological Society, 1st series, vol. 2/On Vegetable remains preserved in Chalcedony

From Wikisource
Jump to navigation Jump to search

XXIII. On Vegetable remains preserved in Chalcedony.

By J. Mac Culloch, M.D. Chemist to the Ordnance, and Lecturer on Chemistry at the Royal Military Academy at Woolwich.
V. Pr. Geo. Soc.

IN examining the agates which are found on the shore at Dunglas in Scotland, in the summer of 1811, I was struck with the appearance of organized vegetable substances contained in many of them. It seeming to me impossible that any metallic or earthy matter could put on these forms, I was at some trouble to make a collection of stones exhibiting similar appearances, and to ascertain as far as circumstances permitted, the genera at least of the plants contained in them. The gradual increase and ultimate accumulation of numerous specimens, having enabled me to trace their analogies to plants actually existing, I considered that a detail of the most leading varieties would not be unacceptable to the Society. For the purposes of a more accurate and convenient illustration than the inspection of the specimens alone would convey, I also transmit some enlarged drawings, made by the aid of the microscope, exhibiting the principal varieties.

I did not know when I was first engaged in this investigation that the subject had already attracted attention, and that a detailed account of some well ascertained plants involved in chalcedony had been given by Daubenton.[1] More lately a letter from Blumenbach has been quoted in Dr. Thomson's journal, noticing the same fact, and professing the removal of his former doubts. It will be a sufficient apology for not recalling this notice after having thus discovered that the inquiry had not the merit of novelty, that so many should have overlooked, and so many others refused their assent to a fact of no uncommon occurrence.

The metallic arborizations emulating the vegetable form which occur in the fissures of many stones, and the similar well-known figures in the chalcedonies distinguished by the name of mochas, with the lively though superficial resemblance they bear to plants, have led to the hasty conclusion that all these appearances were of a metallic nature, and have probably prevented that accurate investigation of them which they deserved. The want of botanical knowledge has perhaps also assisted in concealing from most mineralogists their true origin, but I may now hope that the possessors of such specimens will hereafter by a more attentive examination confirm the frequency of an occurrence so interesting.

Another circumstance has assisted in this case in deceiving mineralogists, and that is the obscurity in which the vegetable is often involved either by the accidental mixture of metallic oxides in the same stone, or by the actual investment of the whole plant with a thick crust of carbonat of iron. Many of the specimens, and most commonly those which contain confervæ, exhibit at first sight nothing but a confused and entangled fibrous mass of oxide or carbonat of iron. It requires a patient observation to detect in these the existence of a real vegetable structure contained in the stone, and modifying the deposition of metallic matter. It will be found in fact that the whole plant is incrusted with the metallic deposit, and that it exhibits only here and there its true nature. An accurate examination will always detect in these the vegetable disposition, and the sections of the stone will, by cutting through the branches in various oblique directions, lay bare the true structure and distinguish the plant from its adventitious and metallic covering.

A third, and the most common source of deception and obscurity, will be found in the whimsical and fibrous disposition occasionally assumed by chlorite, its colour often imitating the natural hue of a plant as perfectly as its fibrous and ramified appearance does the disposition and form of one. It is by this substance particularly that appearances strongly resembling confervæ are produced, and nothing but a very accurate investigation, with considerable experience in the various forms which this substance puts on, as well as with the aspect and characters of the plants imitated, is sufficient to enable us to distinguish them. In many cases even the most scrupulous attention will fail, a fact which need excite no surprize when it is considered with what difficulty the examination must often be conducted, from a deficiency of light when transmitted through these stones, and from the impossibility of bringing into view any considerable portion of the imbedded substance. Here we must have recourse to the chemical means which I shall presently describe.

It is I believe from the deception to which those specimens have exposed mineralogists, that so much incredulity on the subject exists, every green fibre having been supposed to be chlorite, because it was apparent that chlorite sometimes assumed the form of green fibres. However difficult the distinction, it will be found real, and with sufficient care, generally practicable. The subjoined drawings contain representations of some specimens which are undoubtedly chlorite, and of others which present a very suspicious aspect. Among them however will be found some exhibiting an organization so decided, that no mode of crystallization or inorganic arrangement can be conceived capable of imitating it.

While on this subject it may not be useless to notice the various modifications under which chlorite when mixed with chalcedony presents itself. It is now I believe perfectly well understood (as I have already assisted in showing, in some observations on the island of Rum) that the colouring matter of heliotrope is chlorite diffused uniformly through the mass. From some specimens in my possession, I think it probable that plasma is indebted for its colour to the same substance; the different nature of the siliceous base alone constituting the difference of aspect. In these specimens (brought from Egypt) distinct grains of green matter may be observed independent of the general green stain which pervades the stone; and this probability is rendered still stronger from the colour and appearance which is occasionally assumed by the more transparent flinty matter associated with the chalcedony, the latter acquiring from the chlorite the aspect of heliotrope, while the more transparent parts put on the semblance of plasma. But of a substance so rare it is necessary to speak with hesitation. In quartz the chlorite is well known often to assume a contorted appearance resembling the intestines of animals, but in which the regular crystallization and superposition of the plates can be easily traced. Similarly contorted fibres occurring in chalcedony do not exhibit the crystallized structure, but appear to consist of very minute or of invisible particles, or even of a mere colour diffused in that particular form. At times the coloured fibre contains grains of chlorite attached to its sides or interspersed through its course, thus emulating the imbedded seeds or jointed appearance of real confervæ, and I have reason to think often mistaken for them. I am uncertain whether I should rank under the head of chlorite certain contorted fibres which occur in carnelian, and (though much more rarely) in the agate nodules of trap. These are of a colour sometimes ochry yellow, sometimes white, and sometimes brown. Their forms, so similar to the chlorite fibres, would induce us to class them together, and it is probable that they actually consist of this substance, having undergone some decomposition by which the green colour has become brown, from affections of the metallic colouring matter. It is at present impossible to account for this very singular disposition of the chlorite. In some cases it evidently forms the centre of a fine stalactite, of which the minute ramifications, afterwards preserved in further additions of chalcedony, put on the appearance well known in the green stained chalcedonies of Faroe. But that it is not necessarily stalactitical is certain, from its assuming the very same disposition when existing in quartz crystals. It is probably the result of a species of arborization, that obscure circumstance in crystallization which appears to depend on a sort of polarity between distinct crystals, or throughout the interrupted parts of the same one. Thus water on freezing, and various metallic substances on crystallizing, ramify in certain directions. Thus it will often be found that fibres of mesotype will hold their straight and radiating course across stilbite or other associated minerals, the continuity of direction in the portions of any crystal remaining unchanged, however the crystal itself may be interrupted by obstacles occurring in its course. The same appearance may frequently be observed in quartz and other crystallized substances, and it affords among the various phenomena of the obscure process of crystallization, not the least curious subject of inquiry.

It is in the transparent chalcedonies that the fibrous structures are most visible, but they are also of common occurrence in the opaque ones or agates, as they are usually called, although from the impossibility of transmitting a ray of light through these specimens, our sight of them is limited to that portion which happens to run along the fractured or polished surface. Numerous red, white and yellow agates are every where to be found among collectors and dealers, in which the appearance of fibres is such as to render it very probable that they are portions of vegetables incrusted by oxides of iron, although the obscurity of the specimens must necessarily render this uncertain until they are submitted to chemical trials. It these are vegetable fibres there is no difficulty in accounting for this concealment of the vegetable by the metallic deposit, since it is easy to understand how from a compound solution of flint and iron in water, the iron might first be precipitated on the vegetable, in the same way as we daily see it deposited in chalybeate springs, and how the subsequent deposition of flinty matter might involve and penetrate the whole. It is besides known to botanists, that many plants possess the property of attracting from their state of combination the earthy base of some of the salts which are dissolved in natural waters, a property which may be subservient to some purpose in the economy of the plant, unknown to us. Such is the case with Chara vulgaris, which is always found incrusted with a coating of chalk or calcareous subcarbonate, and such appears also to be the case with Byssus nivea.

It is worthy of remark that in almost all the specimens of chalcedony which appear to contain aquatic confervæ, not only the vegetable structure is perfectly preserved, but the plant, however light and yielding its texture, is disposed in as free a manner as if still living and floating in the water which was its native element. Together with these circumstances, the natural colour is often equally well preserved, and the various specimens of the confervæ in particular, which are the plants of most common occurrence, exhibit all the different tones of colour, from the most brilliant grass green to the darkest sap or lightest yellow green which at present characterize the different living specimens with which we are acquainted. At times however the fibre assumes a whitish aspect, or the yellowish and reddish hue which those delicate plants exhibit when dead, an appearance perhaps even more characteristic of their vegetable nature than their natural colour, since we cannot easily understand how it could be imitated either by chlorite or by metallic oxides.

Botanists who are conversant with the difficulty of ascertaining the species of most of the plants contained in the troublesome class of Cryptogamia, and the uncertainty in particular, which, notwithstanding the meritorious labours of many modern authors, still hangs over the individuals of the genus Conferva, will not be surprized that I have made no attempt to ascertain the species occurring in chalcedonies. They well know the obscurity which attends this pursuit, even where access to numerous examples of the living plant, with the power of turning it in every direction, of viewing it in all its states of growth, and of dissecting it into all its ramifications, facilitate the investigation. With regard to the greater number of species indeed, it is known that no distinction can be traced, nor any accurate character laid down, even by moderate magnifying powers. It is only from those highly magnified views of the living specimen which are capable of exhibiting the peculiarities of its internal structure, and the disposition of its fructification, that characters can be formed or individuals distinguished. This sort of investigation cannot be applied to the remains in question, as the loss of light which necessarily follows the attempt to apply high powers, added to the great diminution of it when transmitted through the chalcedony, entirely deprives us of a sight of them. But in the paper of Daubenton, to which I have already referred, several species are actually described as having been well ascertained; and when we read of two plants in particular, whose characters are so strong and so decided as Lichen digitatus and Lichen rangiferinus, we cannot suppose it possible that he or any other observer could have been deceived. The execution of the plates which accompany his paper has unfortunately not rendered justice to the apparent accuracy of his observations.

It will be seen on inspection either of the specimens or of the drawings, that probably all the plants belong to the Cryptogamia class, and are limited to certain species of them. The explanation of this is sufficiently obvious. It is evident that the siliceous depositions which contain these remains must have been formed in caverns and clefts of rocks, situations only occupied by a few species of this class of plants, and these chiefly Byssi, Confervæ, Jungermanniæ, and the Mosses most commonly so called, plants which require very little light. All the specimens which I have figured will accordingly be found to belong to one or other of these families, with the exception of a few which appear to be fragments of plants, and which I have been unable from their mutilated state to compare with any known species.

It may be asked whether the plants thus preserved are specimens of existing species, or whether they are, like those found entangled in the secondary strata, the remains of a former set of organized beings. The limited number of the specimens which I possess, and the obscurity which attends not only these remains but the present living species, prevents me from attempting an answer to this question on botanical evidence, and on my own knowledge. But the observations of Daubenton appear sufficient to decide that in some instances at least they consist of existing species. Since too the formation of flinty matter by deposition from water is a daily occurrence, there is no reason to doubt that they may be specimens of plants now actually existing. But it is also necessary to consider their geological connexions before a full answer can be given to such a question. This is unfortunately attended with considerable difficulty, as the greater number of specimens are only to be found in the hands of lapidaries and jewellers, in circumstances which render it impossible even to trace the country from whence they have been brought, and much less the geological situations in which they have occurred. I have only one meagre fact to offer on this subject. I have said in the commencement of this paper, that many of the specimens found at Dunglas contain remains of vegetables. These specimens appear to have been detached from a breccia inferior to the lowest sandstone, of which a part is visible at St. Abb's Head, and which is probably a portion of the extensive breccia found in so many parts of Scotland, interposed between the primitive rocks and the whole series of flœtz strata. These then contain remains of organized substances of an epocha at least equally ancient with that in which the vegetable remains found in the flœtz strata existed. As the species ascertained by Daubenton have in all probability been preserved in recent formations of chalcedony, so it is probable that those which I am now describing have been preserved in the chalcedonies of former days.

It is said that chalcedonies of this nature are found in the Dutchy of Deux Ponts, in Siberia, and in other situations, but there is no information sufficiently accurate respecting these on which to ground any reasoning. I ought however, to add, that in many of the Siberian specimens which are known to lapidaries by the name of Moss agates, I have ascertained by chemical means that the green fibres consist of chlorite.

It would indeed have been unpardonable not to have used the aids with which chemistry furnishes us to distinguish these obscure substances. Chemical analysis is often the only method by which the very doubtful specimens can be ascertained, and if it be necessary to determine precisely all the specimens which bear the semblance of organization, it is the only trial which can be fully depended on, at least it is the only one on which mere chemical mineralogists will be inclined to place any reliance. A considerable experience in the several substances known by the name of Moss agates, combined with some chemical trials on the most leading varieties, and the habits of botanical investigation, may indeed produce that tact in this examination which is well known to mineralogists in other cases; a judgment founded on circumstances so evanescent and minute as to be incapable of communication by words, The inconvenience which follows chemical trials is such as necessarily to preclude its application in many instances, and to render it desirable that accurate descriptions of all the varieties could be formed. The destruction of the specimen, often rare and almost always expensive, must inevitably follow this mode of investigation. I have not therefore subjected to this fiery trial every specimen which I have examined, but have selected such a number of the principal varieties as were sufficient to confirm that evidence which had appeared to result from botanical considerations, and to define in most of the difficult cases the obscure boundary between the real vegetable fibre, and its mimic resemblance, chlorite.

The immediate object of chemical trial being to ascertain the presence of carbon in the chalcedony, the two following obvious modes were adopted. It was previously determined that all silicified wood had the property of blackening and decomposing sulphuric acid, and of giving over carbonic acid on distillation with nitre. It was also ascertained that chlorite (chlorite baldogée or green earth) did not possess the first of these properties, and that neither of these effects resulted from thus treating common chalcedony. Previously to these trials the precaution was also taken of boiling the specimens for a considerable time in a solution of pure potash, to remove in the polished ones all possibility of the adhesion of the lapidary's oil, a circumstance which would inevitably have led to fallacious results. In these experiments ample confirmation appeared of the deductions which had been made from botanical examination, and I was further enabled to detect many specimens of chlorite, where I had not suspected its existence. The same trials afforded a test which I found in many instances to be very easily applicable to the object of this distinction. This test consisted in the effervescence which is produced when boiling sulphuric acid is applied to those chalcedonies which contain chlorite, while those which contain vegetable fibres blacken the same substance without exciting effervescence. I need scarcely add that I laid no stress on the method of distillation when the stone appeared to contain carbonat of iron. It was not necessary to examine into the cause of the effervescence produced by the action of the sulphuric acid on the chlorite, a circumstance on which the very uncertain and contradictory analyses of Meyer, Höpfner, and other chemists throw no great light.

It would be a waste of time to attempt a description of the character of each individual stone which is found to contain a vegetable substance. However desirable it might be to find a specimen attached to its native place of growth, it has not, as I have before remarked, yet occurred to me, nor do I find that it has occurred to any of those who have noticed the same facts. Yet the appearance of many of the chalcedonies which contain well ascertained specimens of plants, is such as to render it likely that they do now occur in situ. Many of the specimens, and among others those which contain the hypnum figured at No. 7, have the mammillated appearance which implies that they have been formed by the stalagmitic process. And in these the plant is broken, compressed, and deranged, though it preserves its natural colour. In many others no mechanical texture or disposition can be discovered in the stone, but the whole appears one semitransparent mass of chalcedony. These are the specimens in which the confervæ preserve that freedom of arrangement and that perfection both of colour and structure which seem to imply that they had been so suddenly involved in a mass of siliceous matter as to have been preserved from all future changes. Many other specimens have that mixed aspect of jasper and chalcedony to which among other varieties of ornamental stones the lax term of agate has been, applied. In some specimens we may observe that zoned appearance which so generally characterizes the chalcedonies occupying the basaltic geodes, the zones always respecting the various parts of the plant, and forming round it parallel, angular, or curved figures. From this we may infer that the zoned disposition of those chalcedonies which go by the name of onyxes, may as well have proceeded by a successive deposition from the centre towards the circumference, as in the reverse order, an arrangement supposed by some mineralogists to have been the cause of this structure in the nodules found in trap.

The fact of the existence of entire vegetable remains in chalcedony being thus established, it may be said that it is analogous to the well-known instances of silicified wood, and of animal remains similarly situated. Yet it offers some important differences which may throw light on certain disputed points, and lead to conclusions of no small consequence, conclusions not so universally resulting from those facts. It has been maintained on one hand that the silicification of wood and of animal remains might have been the result of silex injected in a state of igneous fusion. I know not that it is necessary for the theory which offers this explanation, that this solution should be admitted, since the existence of that theory is not necessarily implicated in the universal proof of this supposition. It will scarcely be asserted that substances of so tender a structure as those I have described, substances so evidently involved in siliceous matter while freely exposed to light and air, could have undergone this change by any process of compression connected with igneous fusion. Nor could any theorist invent a scheme of this nature which should involve the remains of a land animal, so fragile as is that chrysalis of an insect figured in the plate No. 29, with so little change of structure. A watery solution of silex seems so indispensable for this purpose, that it is superfluous to insist upon it. Of such watery solutions there are abundant examples existing, examples which it is unnecessary to quote; but the instances under examination offer to our consideration views still more wide and more interesting, however difficult their explanation may be. It is plain on reviewing some of the cases above described, that a process different from the tedious one of infiltration and gradual deposition, must have produced the appearances in question. Neither the free disposition nor the forms of the delicate vegetable structures could have been preserved during so slow a process, nor could their colour have remained unaltered The loss of colour must have followed the death of the plant, and the total loss of its figure would have resulted from the gradual changes which it must needs have undergone during the continuance of a process so tedious. The remains are in fact (if I may use such an expression) embalmed alive. To produce this effect, we can only conceive a solution of silex in water, so dense as to support the weight of the substance involved, a solution capable of solidifying in a short space of time, or capable at least of suddenly gelatinizing previously to the ultimate change by which it became solidified into stone. I need not point out the extreme importance of this supposition, I had almost said of this fact, to any general theory of the earth. It is for chemistry yet to investigate experimentally the mode of imitating this unknown process. Those who know the present state of this science will not hesitate to admit its imperfections, and those who have attended to its rapid progress will not despair.

Description of Figures. Pl. 36. 37.

Transactions of the Geological Society, 1st series, vol. 2 plate page 0663.png

No. 1. I think the peculiar membranes and defined structure seen in this figure can only arise from a plant, although I am not acquainted with any analogous living vegetable. I had not enough to subject it to chemical trial.

2. This figure exhibits one of the most common forms in which chlorite is disposed in chalcedonies, a form not very likely to mislead an observer.

3. I have figured this for the purpose of shewing the remarkable imitation of a conferva, which may be produced by chlorite; that part of the drawing which is most highly magnified, exhibiting a chain which consists of distinct scales of this substance.

4. Is drawn from a specimen which appears to contain fragments of two plants; a conferva, and the leaves of some other plant, perhaps a moss.

5. That this drawing is made from a plant, I have ascertained by chemical trial, and the green spots which seem disseminated along the fibre are probably the fructification.

6. This specimen exhibits an internal structure, not very unlike that of conferva ægagropila, although its size is such as to preclude it from belonging to this species, not to notice the other important distinctions between them.

7. The seventh figure exhibits the decided character of a moss, and probably of a hypnum: I have selected two of the most entire branches, the greater number having the leaves folded and confused, as if by external force, a circumstance in itself important. Fibres, appearing to be those of a conferva, may be observed in the same stone, which is a transparent chalcedony, with the mammillated appearance that determines it to have been a stalagmite.

8. A larger specimen of a plant very like No. 6.

9. Very delicate fibres of chlorite, but much resembling the conferva tortuosa. They were ascertained to be chlorite by the chemical means already described.

Transactions of the Geological Society, 1st series, vol. 2 plate page 0665.png

10. The pale colour which surrounds the dark line in this figure, appears to arise from some metallic crust, investing a simple fibre.

11. In this figure nothing but a reticulation of minute yellow fibres can be discerned, and it is to a similar disposition of fibres that a great proportion of the red and yellow agates owe their colour. I believe that this particular description of agate has never been suspected to contain vegetable fibres, nor to owe the disposition of its colouring matter to the effect of these fibres in modifying the deposition of iron. The agates of this tribe generally exhibit a muddy uniform yellow colour; at times, though more rarely, a lively red, and are very common among lapidaries. It is only when in thin plates, and by the assistance of transmitted light, that the existence of fibres is detected in them. When subjected to the action of sulphuric acid they blacken and decompose it in the way I have before described.

12. This figure is intended to convey a notion of the mode in which oxides of iron invest the vegetable fibre.

13. Exhibits a remarkable instance of a deposit of reddish oxide of iron, modified by a white central fibre, with whose nature I am unacquainted.

14. I suspect that this consists of grains of chlorite, become brown from some of the changes to which I alluded in the paper, and surrounded with an additional metallic covering.

15. The ramified structure of this is too decided to admit of its being any thing but a vegetable substance; possibly it is the root of some moss.

16. In the chalcedonies, which go by the name of Mochas, brown arborizations are known to be very common, and often to assume appearances which render these stones much sought after for ornaments. I have here figured one which bears so strong a resemblance to the imbricated Jungermanniæ, that it is difficult at first sight to distinguish them. The detached scales render the deception still stronger, their appearance being that of leaves which have been broken off. But it will I believe be found that all these are metallic. How much soever, the bases or middle parts of the pretended plants may put on an appearance difficult to discriminate, it will almost always be found that the extremities of the branches are undefined and shapeless, while in the Jungermanniæ the regular imbrication of the scales is continued to the minutest extremity; The eye of a botanist will also with care discover an irregularity in the superposition of the scales, which never occurs in the plants themselves, and a decision in the setting on of the branches in the real plants, which is ill imitated by the clumsy way in which the ramifications are disposed in the metallic arborization. I do not however mean to deny that these stones may not contain organized bodies, as I have below given a figure of such an one.

17. Fibres, very common in chalcedonies, and bearing a considerable resemblance to Byssus nigra,—conferva ebenea of Dillwyn.

18. Fibres of a brown colour, with whose, nature I am unacquainted. They are possibly vegetable, but I had not enough to apply the chemical test to the specimen.

Transactions of the Geological Society, 1st series, vol. 2 plate page 0667.png

19, 20, 21. Were drawn for the purpose of shewing various coloured fibres which are among the most common of those found in chalcedonies. The generally great decision and character of their ramifications seem to establish them as vegetable fibres, and which, as I observed in the paper, have lost their colour by the process of death going hand in hand with their lapidification.

22. Appears to consist of fragments of a lichen of the foliaceous imbricated kind, akin to centrifugus and saxatilis, and bearing a considerable resemblance in colour and general appearance to Parmelia Borreri, as figured in the Linnean Transactions. The fragments are too insufficient in extent, and too deeply bedded in the stone to admit of any very accurate judgment respecting their affinities.

23. Is another example of a moss, apparently of the same family with that figured at No. 7. I have drawn it precisely as it appears. It differs from the former in the more orbicular and obtuse form of the leaves. It is contained in a large nodule of chalcedony, which exhibits much colour and the zonular disposition.

Transactions of the Geological Society, 1st series, vol. 2 plate page 0669.png

24, 25, and 26. Brown ramified fibres of the same apparent nature as 19, 20, and 21. For these drawings I am indebted to Mr. Blore, and they are from specimens in his possession which I have not had an opportunity of examining very particularly.

27. These are contained in an oriental chalcedony or mocha. They are evidently hollow tubes, of which various aspects are exhibited by the chance section of the stone.

28. Similar tubular bodies which appear equally to belong to the same class of beings. I have figured these because they had been supposed by some of my friends to be examples of the fact under review. They appear to belong to the tribe of Zoophytes.

29. The singularity of this occurrence has induced me to give a figure of it. I know not indeed that any other example than the present one has been produced of an animal substance of this nature preserved in chalcedony. The stone which contains it is a striped onyx agate, and is part of a ring in the possession of the Earl of Powis. Its fortuitous section by the lapidary has exposed the internal as well as the external side of the chrysalis, from which the fly had escaped previously to its lapidification. Its structure points it out to be the pupa of a lepidopterous insect, probably of a moth.

30. I have figured this, although I conceive it to be another example from the Zoophytes, because I suspect that it has been mistaken for the fructification of a moss.

It is proper to remark that the figure attached to each specimen is intended to express the number of times it is magnified in the drawing.

Independently of the figures which I have now given, I possess numerous other varieties, which I thought it superfluous to represent, as they afforded no striking peculiarities. I have chosen the figures among the most perfect, as well as among the most obscure, for the purpose of illustrating the one by the other; and I have also added the mimic resemblances to guide, if possible, the investigations of others, and to prevent them from attributing too much to any bias which might be supposed to have warped my own observations.

  1. Mémoires de l'Académie Royale des Sciences, 1782, p. 668.