ganic chemistry cannot be sustained, on ac- cliemistrv.
count of its evident irrationality. What is at present called organic chemistry has an essentially bastard character, half chemical, half physiological, and not, in fact, either the one or the other, as we shall have occasion to see. The division cannot even be sustained under another form, as equivalent to the general distinction between cases of dualism and of other composition. For if inorganic combinations are usually binary, there are some which are composed of three elements and even of four; while, conversely, we very often meet with a true dualism in bodies which are called organic. For a genuine division we must look to general ideas relating to Principles of composition and decomposition; and in this composition form, attending to the rule of following the ^^^'-\ ileconigradual complication of phenomena: first, P°^^ ^°"' the growing plurality of constituent principles (mediate or immediate), according as the combinations are binary, ternary, etc.; and secondly, the higher or lower degree of composition of the immediate princij^les, each of which may (as in the case of a continual dualism) be decomposable into two others, for a greater or smaller number of consecutive times. Though each of these two points of view is of high importance, the preponderance of the one or the other must be agreed upon before the rational division of chemistry can be organized. Though this is not the j^lace to discuss this new question of high chemical philosophy, it may be well to state that I regard it as solved; and that the consideration of the degree of composition is, in my eyes, evidently superior to that of the number of elements, inasmuch as it aft'ects more profoundly the aim and spirit of chemical science, as they have been characterized in this chapter. As for the rest, whatever the decision may be, we may remark that the two classifications differ from each other much less than we might at first be tempted to suppose; for they necessarily concur, whether in the preliminary or in the final case, and diverge only in the intermediate parts.
We have now reviewed the nature and spirit of chemical science; the m.eans of investigation proper to it; its true encyclopedical position; the kind and degree of perfection of which it is suscejitihle; its philosophical properties in regard to method and to doctrine; and, finally, the mode of division which would be suitable to it. We must comjilete the survey of the science by a special and direct notice of the few essential doctrines which have been disclosed by the spontaneous development of chemical philosophy. It must be remembered that the object of this work is not to present a treatise on each science, or to enlarge upon it in proportion to its proper imjDortance, or the multiplicity of its facts: but to ascertain its relative importance, as one head of positive philosophy. No one will exjDCct that chemical philosophy, in its present state, can be examined here as fully or satisfactorily as, for instance, astronomical philosophy, the perfection of which admits of a methodical analysis, clear and complete, though summary, such as befits that immutable type of natural philosophy.
INORGANIC CHEMISTRY.
WHATEVER may be the principles of ^i i f i „,• division and classification preferred ninw the stmlv. in the general system of chemical studies, it is agreed by almost all chemists that the preliminary and fundamental study should be the successive and continuous history of all the simple bodies. The plan of M. Chevreul is an exception to this, his method being to proceed at once from the study of each element to all the combinations, binary, ternary, etc., that it can form with those already examined; confining himself, however, to compounds of the first order. This plan has the advantage that simple bodies are more completely known from the beginning than by the usual method, which scatters through the different parts of the science the most important chemical properties of each of them. But, on the other hand, the history of any element remains incomplete; a factitious inequality is established among chemical researches into different elementary substances; and the didactic inconvenience which M. Chevreul proposed to escape seems to me to be unavoidable, under any method. On no plan can any chemical history be completed by a first study. The provisional information obtained by a first study must be followed by a revision which allows us to take into consideration the whole series of phenomena relative to each substance. The question is merely a didactic one, only of secondary importance in this work, though of great practical interest. On any scheme, it remains certain that the preliminary study of elementary substances is, by the nature of the science, the necessary foundation of chemical knowledge.
On account of the considerable and always increasing number of substances regarded as simple, some modern philosophers, possessed with the notion of elements ^^^® simplicity and economy of nature, have concluded a priori that most substances must be the various compounds of a much smaller number of others. But, while endeavouring to conceive of nature under the simplest aspect possible, we must do so imder the teaching of her own phenomena, not substituting for that instruction any thoughtless desires of our own. We have no right to presume beforehand that the number of simple substances must be either very small or very large. Chemical research alone should settle this; and all that we are entitled to say is that our minds are disposed to prefer the smaller number, even, if it were possible, so far as there being but two. But not the less are we bound to suppose all substances which have never in any way been decomposed to be simple, though we should not pronounce them to be for ever undeeomposable. All chemists now admit this rule as the first axiom of sound chemical philosophy, Aristotle first saw this rule, though he did not conceive of its rational grounds. His doctrine of the four elements, popularly cried down in our time, should be judged of as the first attempt of the true philosophical spirit to conceive of the composition of natural bodies, amidst the then existing deficiency of all suitable means of research. To appreciate it we must compare it with anterior notions. Now, up to that time, all the schools, however they might diifer about other things, agreed that there was only one elementary substance; and their dispute was about the choice of the principle. Aristotle, with his rational character of mind, put an end to all those barren controversies by establishing the plurality of elements. This immense progress must be considered the true orisrin of chemical science, which would be radicallv impossible on the supposition of a single element, excluding all idea of composition and decomposition. Whatever ay)pearances mav be, there is no doubt that it must be much more difficult for the human mind to pass from the absolute idea of unity of principle to the relative idea of plurality, than to rise gradually, by means of research, from the four elements of Aristotle to the fifty-six simple bodies of our PLURALITY OF ELEMENTS. 335 chemistry of this day. Our Naturists, who are all for simplicity and economy without caring much for reality, have no right to appeal to the authority of Aristotle, who had so much reverence for reality as to infringe the notion of simplicity which he found prevailing. They should go back further than Aristotle, — to Empedocles or Heraclitus, and attain the utmost simplicity at once, by admitting only a single principle.
Other philosophers, among whom was Cuvier, have objected to the simplicity of most of the elements now admitted by chemists, that some of them seem to be extremely abundant in nature, while others are scantily and partially distributed: whereas, it seems natural to presume that the different elements must be almost equally diffused throughout the globe, and that therefore chemical analysis will sooner or later prove the rare ones to be compound substances, requiring peculiar and rare influences for their formation. It would be enough to say that the presumption, though plausible, is nothing more than a presumption: but it may be added that we know nothing of our planet beyond the upper strata; and we can form no prejudgment of the composition of the whole. It would be too much to say that there should be an equality of elements on the surface, even the pi'obability being the other way; for the heaviest elements are the rarest at the surface, and the commonest are those which go to the composition of living bodies; and the probability is strong that the preponderance is reversed in the interior of the globe, to make up the mean density, which is not to be found a.mong the solids, liquids, and gases which are required for the existence of life. Thus the objection seems to be converted by chemical analysis into a sort of confirmation.
Since the time, — recent, it is true, — of the decomposition of the elements of Aristotle, there has not been a single instance of a substance having passed from the class of simple to that of compound bodies, while the inverse case has been frequent. Yet, no chemist disputes the possibility of a reduction of the elements by a more thorough analysis; for chemical simplicity, as it is to us, is a jiurely negative quality, not admitting of those irreversible demonstrations proper to positive conijDOsitioiis and recompositious. The great general example of substances called organic, the chemical theory of which is so complex, notwithstanding the small number of their elements, might lead us to suppose that such a reduction would not be, after all, so very great an advantage: but in this case, the difficulty seems to me to be referrible to the deficiency of duality. Notwithstanding this example, we cannot but think that chemistry would become more rational and more systematic, if the elements were fewer, from the closer and more general relation which must then subsist among the different classes of phenomena. But the aj^parent perfection could be only barren and illusory if we were to assume it by conjecture anticipating the real progress of chemical analysis.
„.. „. This profusion of elements has naturally of elements ^^^ ^^ endeavours to classify them. The high importance of the question has become manifest through the deep persuasion that the rational classification of simple bodies must determine that of compound substances, and therefore that of the whole chemical system. The first principle to be laid down is that the hierarchy of elementary substances is not to be determined only by their proper essential characters, but by the less direct consideration of the principal phenomena of the compounds which they form. Without this requisition, the classification would have little use or interest; for it would be of small consequence in what conventional order we studied fifty-six bodies all independent of each other: whereas, with its proper condition, this question is as important as any that chemical philosophy can present.
The old division of the elements into the comburent and combustible, (those which burn in the active and in the neuter sense,) — and the subdivision of these into metallics and non-metallics, are evidently too artificial to be maintained, except provisionally. For many years, endeavours have been made to supersede it; but no irreversible classification has been yet obtained. M. Ampere seems to have been the first who pointed out the necessity; and he proposed a system in 1816; but it was not one which induced the chemists to abaudon their ancient distribution, the binary structure of which made it easy of aj^plication, whatever might otherwise be its defects. A few years after, Berzelius offered, in a simj^le and almost incidental form and manner, a far superior system of classification. He first understood the necessity of rising finally to a unique series, constituting, by a uniform and preponderant character, a true hierarchy; whereas, M. Ampere saw only the importance of natural groups, which might be arbitrarily co-ordinated. Both conditions are imposed by the general theory of classifications; but that which Berzelius had chiefly in view is unquestionably suj^erior to the other; and especially in the present case, when the small number of objects to be classified renders the formation of groujjs a matter of secondary im2:)ortance, provided the series be naturally ordained.
M. Berzelius's conception is grounded on the consideration of electro-chemical pheno- J;jassification mena. Its simple and lucid principle is that the elements are to be so disj^osed as that each shall be electro-negative to those which precede it, and electropositive to those which follow it. The series thus derived, api^ears, thus far, to be in conformity with the whole of the known properties of both the elements themselves and their principal compounds. It is too soon however to speak decisively of this: and, on the other hand, the chemical j)reponderance of electric characters is by no means so logically established as to compel us to seek the bases of a natural classification in that order of phenomena. It must, it seems to me, be clearly proved, at the outset, that the point of departure is a real one, — that is, that a constant order of electrization exists among the different elements, which is maintained under all conditions of exterior circumstances, of aggregation and decomposition: but, not only has this never been adequately undertaken, but there is some reason to apprehend that its result would be opposite to the proposed principle. Whatever may be the issue of future labours, Berzelius has secured the eternal honour of having first exhibited the true nature of the problem, and the aggregate of its principal conditions, and perhaps the order of ideas in which its solution is to be I. z I. z sought. Whenever this solution is obtained, chemistry will have made a great stride towards a truly rational state: for, under a hierarchy of the elements, the systematic nomenclatui-e of compound substances will almost suffice to give a first indication of the general issue proper to each chemical event; or, at least, to restrict the uncertainty within narrow limits. Tet, through this very connection of such a research with the whole of chemical studies, I do not think it can be efficaciously pursued while we separate it, as has hitherto been done, from the question about the establishment of a complete system of chemical classification for all bodies, simple and compound. Now, this great question seems tome at present premature.
The preliminary conditions, both of method liematuie ^^^ ^^ doctrine, are, as we have seen, far from being completed. As such a general svstem of classification must constitute both the summing lip and the fundamental view of the whole of chemical philosophy, I shall further expand my idea about it in this place.
Requisite pre- A-S for the method, it requires perfecting paration as to in two ways, for which chemists must resort method. ^o physiology. They must understand the fundamental theory of natural classifications, which can be obtained nowhere else: and they must, for the same reason, study in the same school the general spirit of the compai'ative method, of which chemists have very little idea, and without which they can never proceed properly in search of a rational classification. These two improvements must be derived from biological philosophy; the one to lay down the problem of chemical classification, and the other to undertake its solution. It will be by perceiving these harmonies and mutual applications among the sciences commonly treated as isolated and independent, that philosophers in all departments will at length become aware of the reality and utility of the fundamental conception of this work; the cultivation of the different branches of natural philosophy under the impulsion and direction of a general system of positive ])hik)sophy. as a common basis and uniform connection of all scientific labours. We have little idea what we lose by the narrow and irrational spirit SYSTEM OF CLASSIFICATION. 339 in which the different sciences ai'e ci^ltivated, and especially with regard to method. When the great scientific relations of the futiii-e shall be regularly organized, men will scarcely he able to imagine, otherwise than historically, that the study of nature could ever have been conceived and directed in any other way.
As to the doctrine, we have seen that the a + ] <- • desired classification cannot take place till we have settled the jireponderance of the one or the other consideration,— the order of composition of the immediate principles, or their degree of plurality. Now, such a problem has not yet been rationally proposed. If we suppose it resolved, adopting the rule which I think almost incontestable, as I explained before, of treating the first point of view as necessarily superior to the second, we must still attend to two special conditions, before we can proceed to the rational construction of the system of chemical substances.
Bv the first of these conditions, we must t-.- ^ t^--,. •'..-,. ■• 1 T.•,• J? 1 rirst condition, dismiss the irrational distinction ot substances into organic and inorganic. We shall see hereafter that organic chemistry must soon dissolve, parting with some of its questions to chemistry proper, and others to physiology. Wlien any combination is susceptible of a chemical examination, it must be subjected to a fixed order of homogeneous considerations, whatever may have been its origin and mode of concrete existence, with which chemistry has nothing to do, unless as a source of information. As long as any classification must be adapted to the strange conception of a sort of double chemistry, established upon a false division of substances, it must be precarious and artificial in its details, because it is vitiated in principle. The evil is felt, as is shown more and more by the tendency to refer organic combinations to the general laws of inorganic combinations: but it would not be enough, as might be supposed, that a distinguished chemist should take the initiative, in a large and direct manner, to accomplish this important reform. Such a work demands a special and difficult operation, requiring a delicate combination of the chemical and physiological point of view, in order to make a true division of what should remain with chemistry, and what should return to physiology.
The second condition is closely connected condition with the first. It requires that all combinations should, if possible, be submitted to the law of dualism, erected into a constant and necessary principle of chemical pliilosophy. Great as would be such an improvement in the v/ay of simplification of chemical conceptions, it must however be admitted that it is not so indispensable to classification as the preceding. Without the first condition, rational classification would be impossible: whereas, it might take place, with imperfection and difiiculty, without the second. As for the prospects of the case, the tendency to improvement is as real and marked in the one case as the other; as any one may observe for himself.
,,,,,. It is of the more importance to set the analvsis consideration or the order of composition oi immediate principles above that of their degree of pkirality, as before proposed, because the first is,,,,,. - by its 1 nature, clear and incontestable, while duaHsni *"*^ other is always more or less obscure and dubious. The one is, in fact, the simple appreciation of an analytical or synthetical fact: the second has always a certain hypothetical character, since we then pronounce upon the mode of agglomeration of elementary particles; which is a thing radically inaccessible to us. Thus, for example, a chemist may establish with certainty that such or such a salt is a comjjound of the second order, and that certain acids and alkalies are, on the contrary, of the first order; for analysis and synthesis can demonstrate that each of the last bodies is composed of two elementary substances, and that, on the contrary, the immediate principles of the salt are decomposable into two elements. But, in another view, when the analysis of any substance has established the existence in it of three or four elements, as in the case of vegetable or animal matters, we cannot, without resort to hypothesis, pronounce that this combination is really ternary or quaternary, instead of being simply binary: for we can never assert that we could not, by a preliminary analysis CHEMICAL DUALISM. 341 less violent than this final one, resolve the proposed substance into two immediate principles of the first order, each of which should be further susceptible of a new binarydecomposition.
If an unskilled chemist should at this day apply unduly strong means to the analysis of saltpetre, the results might authorize him, following our present erroneous procedure, to conceive of this substance as a ternary combination of oxygen, azote, and potassium: and yet we know that such a conclusion woitM be false, as the substance may be easily reconstructed by a direct combination between nitric acid and potash, which might have been separated by a less disturbing analysis, without occasioning their decomposition. How do we know tbat it may not be so with every combination habitually classed as ternary or quaternary?
Immediate analysis being as yet so imperfect in comparison with elementary analysis, especially with regard to these substances, would it be rational to proclaim, for the time to come, its necessary and eternal impotence with regard to them? Such judgments seem to be founded on a confusion between these two kinds of analysis, so really different in themselves, and so characterized in their operations by delicacy in the one case and energy in the other. — One important consideration, relating to the synthetical point of view, is evidence of this confusion between the two analyses: and that is, the extreme difficulty, if not impossibility, of verifying by synthesis the analytical results proper to these substances. We have seen that immediate synthesis is usually very easy, while elementary synthesis is scarcely practicable. Thus, reciprocally, it seems to me rational to suppose that when the recomposition cannot be effected, the analysis has not been immediate,— there being no other objection to such a conclusion. For example, we exhibit the impossibility of reproducing by synthesis vegetable and animal substances: and this has been even set vip as a sort of empirical principle. But is not this impossibility owing to our persisting in an elementary synthesis when we ought to proceed by an immediate synthesis, the materials of which might in many cases be discovered beforehand? This remark is true with regard to a multitude of combinations the dualism of which is, however, very certain, with the sole difference that the immediate principles are better known. If we tried to recompose saltpetre by directly combining oxygen, azote, and potassium, we should succeed no better than in reproducing organic substances by throwing together their three or four elements: the obstacles which we admit in the last case apply equally to the first. The most striking achievement is that of M. Wcehler, in producing the animal substance urea. He could not have done this if he had tried, according to the common prejudice, to combine directly oxygen, hydrogen, carbon, and azote, which concur in the elementary constitution of this substance, instead of uniting only its two immediate principles, till then unknown in this quality. Is there any reason to suppose that it is otherwise in any other case? — It appears then that chemists will be safe in attributing an entire generality to the fundamental principle of the dualism of all combinations, under the one easy condition of regarding as still very imperfect the analysis of substances exceeding the binary composition; and esj)ecially the substances called organic, the true immediate prmciples of which would thus remain to be disco veiled. These principles can be conceived of only by imagining a considerable number of new binary combinations, of the first and second orders, between oxygen, hydrogen, carbon, and azote: and the realization of this may seem, in the present state of our knowledge, almost impossible. But we have no right to conclude it to be so, while otu- analytical procedures are what they are; and there is no scientific objection to our supposing that there may be many more direct and binary combinations among the elements of ternary or quaternary substances than chemistry has yet established.
It must be observed, however, that universal and indefinite dualism cannot be maintained unless chemists will scientifically determine the sense of the word substance; that is, restrict it to mean real cotnhhiation: for it would be easy to cite, and especially in physiological chemistry, very marked cases of the defect of dualism. But we cannot regard as a true chemical substance an accidental assemblage of heterogeneous substances, whose agglomera- MEANING OF SUBSTANCE. 343 tion is evideutly mechanical, such as sap, blood, a biliary calculus, etc., uuless we confound the notion of dissolution, and even of mixture, with that of combination. If we extend in this way the use of the term substance, so valuable in chemistry, we might as well treat, as so many chemical substances, the waters of different seas, different mineral waters, soils, etc.: and even more, artificial mixtures of a variety of salts dissolved together in water or alcohol. We shall see hereafter that all difficulties in this subject may be disposed of by our learning that they proceed from our not having clearly and rigorously separated the chemical from the physiological point of view. We may be assured that the most elementary notions of chemical philosophy cannot be rationally established, in their due clearness, generality, and stability, without being founded on a full comparison with biology; a comparison which can be organized only under a complete system of Meantime, there is a marked tendency in the present movement of chemical ideas, towards a complete dualism. The increasing assimilation attempted between organic and inorganic substances is an indirect advance in that road: but much more striking, in this view, are experiments like those of M. Woehler, which I'efer the most refractory compounds to dualism, either by analysis or synthesis. A binary formula is adopted, too, to represent the proportion of elements proper to the most complex substances: and, though this is not a true dualism, it helps to prepare minds for the establishment of a real and general one. The sum of what has been said on this important subject of chemical dualism is this: — the real mode of agglomeration of elementary particles is, and ever must be, unknown to us, and therefore no proper object of our study: — our positive researches being thus circumscribed, we may rationally conceive of the immediate composition of any substance as binary; but so as to represent all the phenomena that chemistry can offer to us, in any future state of perfection. Thus, I do not ]>ropose universal dualism as a law of nature; for this we could never establish: but I declare it to lie a fundamental artifice of true chemical philosophy, destined to simplify our elementary concep- 344. POSITIVE PHILOSOPHY.
tious, by using our optional intellectual liberty in accordance with the true end and aim of positive chemistry.
These are the conditions necessary to the institution of a system of natural classification, answering in chemistry to tlie universal hierarchy of living bodies in biology, if the complication of phenomena would admit of our obtaining such a system. Up to this time, perhaps no one has formed an adequate idea of the nature and spirit of siTch an operation: but, in my view, chemical classification, thus conceived of, is the science itself, condensed into the most substantial summary. All I claim to have done is to have introduced into chemical science the special kind of philosophical spirit which is naturally developed by biological science, as it has been conceived of by all its great masters, from Aristotle downwards.
It is because I have high expectations of what Chemistry will become, that I attach so much importance to the preceding discussion. The science is now weak and desultory, notwithstanding its rich collection of facts: but, extended and complex as it is, there is no fundamental science, except astronomy, whose phenomena are so homogeneoTis, and therefore so fit for a true systematization, in the positive spirit. Now, this future constitution of chemical science miist, it seems to me, consist in a complete system of natural classification, which cannot be obtained till all combinations, Avhatever their origin, are subjected to a fixed order of homogeneous considerations, and, on the other hand, constantly referred to a fundamental dualism.
We cannot form any certain expectation of the fiitvire condition of Chemistry from its present state: but, before proceeding to examine the two doctrines which at this day approach nearest to positive rationality, — that of definite proportions and the electro-chemical theory, — I will indicate two points of doctrine which seem, by their nature, to indicate with precision the true dogmatic formation towards which the science, as a whole, must tend. Law of double First, there is the great law of double saline decom- saline decompositions, discovered by Berpositions. thollet, and completed by M Dulong's investigations on the reciprocal action of soluble and insoluble salts. The case of double solubility, considered by Ber- CHEMICAL THEORY OF AIR AND WATER. 345 thollet, is this: two soluble salts, of any kind, mutually decompose each other whenever their reaction may produce an insoluble salt, or one less soluble than either of the two. This theorem stands first among general propositions in chemistry, and is the only one which can as yet give an exact idea of what, in chemistry, constitutes a true law. It has all the characters of a law: it relates to the proper subject of chemical science; it establishes a relation between two classes of phenomena before independent; and, above all, it admits of prevision of phenomena according to their positive relations. In establishing this law, Berthollet escaped some metaphysical snares, rejecting hypotheses of affinities; but he fell into one when he attempted to explain the law which he had just discovered. No law can be exjDlained otherwise than by showing that it enters into another, more general than itself: but this law of Berthollet's is alone of its kind; and it therefore admits of no exjilanation. It may hereafter be attached to a fundamental theory of the reciprocal action of all compounds of the second order; and such a relation will truly explain it: but at present it is simply a general fact which, inexplicable itself, serves to explain each of the particular facts which it comprehends.