LIBRARY OF THE UNIVERSITY OF CALIFORNIA. GIFT OF .PROF. W.JB. RISING Class STANDARD WORKS ON CHEMISTRY. WOHLER'S ORGANIC CHEMISTRY Lately Issued. OUTLINES OF ORGANIC CHEMISTRY. By WOHLER and FITTIG. Translated, with Additions, from the Eighth German Edition, by IRA RKMSEN, M.D., Ph.D., Professor of Chemistry in Johns Hopkins University, Baltimore. In one handsome volume, 12mo., of 550 pp. ; cloth, $3. As a book of general reference, we do not hesitate to recommend it as the best and most satisfactory we have seen, and. in fact, it seems to have left nothing to be de- sired from such a publication. Journal of Applied Chemistry, Sept. 1873 This compact treatise is an excellent handbook of organic chemistry for the student attending chemistry or engaged in laboratory work. The notation is given in the simplest graphic form, and the relations of the various classes of bodies and of indi- vidual substances to each other are easily traced. Dr. Remsen has discharged his duty as translator in a thoroughly acceptable manner. Silliman's Journal, July, 1873. FOWNES' ELEMENTARY CHEMISTRY Lately Issued. A MANUAL OF ELEMENTARY CHEMISTRY, THEO- RETICAL AND PRACTICAL. By GEORGE FOWNES, Ph.D , late Professor of Practical Chemistry in University College, London. A new American, from the Tenth .and Revised London Edition Edited by ROBERT BRIDGES, M.D. In one large royal 12rao. volume of about 850 pp., with 197 illustrations ; cloth, $2 75 ; leather, $3 25. BOWMAN'S MEDICAL CHEMISTRY Lately Issued. A PRACTICAL HANDBOOK OF MEDICAL CHEMIS- TRY. By JOHN E. BOWMAN, M.D. Edited by C. L. BLOXAM, Pro- fessor of Practical Chemistry in King's College, London. Sixth American, from the fourth and Revised English Edition. In one neat volume, royal 12mo., pp. 351, with numerous illustrations ; cloth, $2 25. BOWMAN'S MEDICAL CHEMISTRY Lately Issued. AN INTRODUCTION TO PRACTICAL CHEMISTRY, INCLUDING ANALYSIS. By JOHN BOWMAN, M.D. Sixth Ame- rican, from the si^th and Revised London Edition. With numerous illustrations. In one neat volume, royal 12mo. ; cloth, $2 25. ODLING'S CHEMISTRY Lately Issued A COURSE OF PRACTICAL CHEMISTRY, arranged for the use of Medical Students. By WILLIAM ODLING, Lecturer on Chemistry at St. Bartholomew's Hospital, E 113 7 Fluorine ..... F 19 Gallium ..... Ga __ Gl 9.4 Gold An 197 Hydrogen . . H 1 Indium ..... In 113.7 Iodine ..... I 127 126.85 Iridium ..... Ir 198 Iron . . Fe 56 Lanthanum . La 92.5 Pb 207 Lithium . . r . . Li 7 7.022 Magnesium . . . Manganese . . Mg Mn 24 55 Mercury . . . . .. Hg 200 Molybdenum .... Mo 95.8 Nickel Ni 59 Nitrogen ..... N 14 14.044 Xli LIST OF ELEMENTS AND ATOMIC WEIGHTS. LIST OP ELEMENTS AND ATOMIC WEIGHTS. Continued. Name of Element. Symbol. Atomic weight. Atomic weight, very accnraiely determined Osmium . . . , Os 199.2 Oxygen . . ..... 16 Palladium ... Pd 106.6 Phosphorus . . ' . Platinum ..... P Pt 31 197.4 Potassium .... K 39.1 39.137 Rhodium . . . Ro 104.4 Rubidium . . . Rb 85.4 Ruthenium . ... Ru 104.4 Selenium . ... Se 79.4 Silicon . . ... Si 28 Silver ..... . , . Ag 108 107.93 Sodium ..... Na 23 23.043 Strontium .... Sr 87.5 Sulphur . . S 32 Tantalum .... Ta 182 Tellurium .... Te 128 Thallium Tl 204 Thorium Th 231 Tin Sn 118 ~ Titanium . . Ti 50 Tungsten . Uranium . . . ,. . W U 184 240' Vanadium V 51.3 Yttrium . . Y 61.7 Zinc . ... Zn 65 Zirconium ... . . Zr 89.6 PRINCIPLES OP THEORETICAL CHEMISTRY. PART FIRST. GENERAL DISCUSSION OF ATOMS AND MOLECULES. I. ATOMIC THEORY ATOMIC WEIGHTS, ETC. General Conceptions. Substances that occur in nature are, for the most part, not simple substances. They can generally be decomposed into kinds of matter that are un- like some by one means, some by another. Such sub- stances are called compound. In regard to the means required to separate these compounds into their constitu- ents, a marked difference is noticed in different cases. Sometimes only very simple mechanical processes are required to effect the separation. At other times it is found that, after the application of all purely mechanical processes has failed,, the compound yields to the influence of some of the so-called physical forces, as heat, light, electricity. This leads to the conclusion that there are at least two varieties of compound substances. Each of these varieties is more or less strongly characterized by external properties. As regards those which can be de- composed, by mechanical means alone, it is true that they possess the combined properties of their constituents; and these constituents are usually contained in the com- pound in their original forms. As regards the second class of compounds, on the other hand, it is just as true that they do not possess the properties of their constitu- 2 14 DISCUSSION OF ATOMS AND MOLECULES. ents; and these constituents are contained in the com- pounds in forms differing entirely from those originally possessed by them. Chemism. Substances which are held together by cohesion or adhesion can be separated by mechanical means. But we have here evidence of the existence of some force which holds substances together, and which cannot be overcome by mechanical means. To this force the name chemical affinity or chemism has been given. The object of the science of chemistry is the study of this force in its relations to matter; or the study of the action of matter upon matter, as far as it is influenced by this force. In regard to the position which chemistry occu- pies among the sciences, it will be seen that it is primarily a branch of physics, if to the latter science is given its broadest scope. But, further, chemism always gives rise to the formation of new bodies, and the study of these in their relations to each other becomes a legitimate part of the object of chemistry. This allies the science to that branch of study embraced under the head Natural His- tory. Owing to this twofold character, it is customary to treat the subject as forming an independent science, and, for many reasons, this is the most convenient method. We have thus far recognized the existence of the force, chemism ; and also become acquainted with one of the characteristics of its action. The knowledge of the force remained for a long time in this state, as the methods of investigation at first employed could not disclose its most important characteristics. Up to the latter part of the eighteenth century, the qualitative method of investi- gation was of necessity the principal one employed in the study of chemical phenomena ; that is to say, the quality of the substances allowed to act upon each other, and the quality of the product or products were noted, but little attention being given to the amounts of the substances employed, or of those obtained as products. As the importance of the quantitative method became more and more apparent, the means for applying it also gradually made their appearance. The balance, the sine qua non of chemistry, was improved, and, finally, in Lavoisier's hands led to tangible results. To its use is to be ascribed the correct explanation of the phenomenon ATOMIC THEORY ATOMIC WEIGHTS, ETC. 15 of combustion, a phenomenon which, considered in all the varied forms in which it is presented to us, must be looked upon as the most important of all chemical phe- nomena. But, though the explanation of combustion was correctly given, no new property of chemism was dis- covered. A new example of the kind of action which was already known to characterize the force was added to the list f the key was given to a better understanding of the chemical nature of gaseous bodies; the indestructi- bility of matter became, perhaps, more distinctly evident than it had hitherto been ; but the knowledge of chemism as such remained what it had been up to that time. That a definite result was obtained through a consideration of the quantitative relations of an experiment was the fact which, above all others, gave an impulse to the subse- quent development of the science. Investigations of chemical phenomena now took, in general, a different direction ; and soon the united work of many hands succeeded in establishing a fundamental principle of the science. The first semblance of a gene- ral law governing chemical action made its appearance when it was finally established beyond a doubt that the combination of bodies, under the influence of chemism, always takes place in fixed proportions. This principle, though perhaps tacitly acknowledged by many chemists, was not fully established until the beginning of the pre- sent century. In 1803, a strong effort was made by Berthollet, in his work entitled " Statique Chimique," to prove the incorrectness of the principle, but the oppo- sition called forth by this work, particularly from Proust, led to more and more careful examinations of the so-called chemical compounds, and thus to the firm establishment of the principle. Proust also showed that two bodies could combine with each other in more than one propor- tion, and that for each combination the relative propor- tions of the constituents were fixed. Dalton's Investigations. In the year 1804, Dalton's investigations enabled him to take another advance-step. Another general law governing chemical action was dis- covered and propounded. This was the law of multiple proportions. As this is the foundation of the science, as it is at present, let us follow, somewhat in detail, Dalton's 16 DISCUSSION OF ATOMS AND MOLECULES. reasoning. Many substances had been analyzed before his time, and the percentages of the constituents had been determined with a tolerable degree of accuracy. He examined first two gases, both of which consist of carbon and hydrogen, viz., olefiant gas and marsh-gas. He analyzed them both, and determined the percentage of the constituents contained in them. These percentages were as follows : Olefiant gas, 85.7 per cent, carbon, and 14.3 per cent, hydrogen. Marsh-gas, 75.0 " " 25.0 " On comparing these numbers he found that the ratio of carbon to hydrogen in olefiant gas was as 6 to 1 ; whereas, in marsh-gas it was 3 to 1, or 6 to 2. The amount of hydrogen combined with a given amount of carbon was exactly twice as great in the one case as in the other. For the two oxides of carbon, further, the following numbers were obtained : Carbon monoxide, 42.86 p. c. carbon, and 57.14 p. c. oxygen. Carbon dioxide, 27.27 " 72.73 But 42.86 : 57.14 : : 6 : 8, and 27.27 : 72.73 : : 6 : 16. The amount of oxygen combined with a given amount of carbon in carbon dioxide was exactty twice as great as the amount of oxygen combined with the same amount of carbon in carbon monoxide. He saw again that, in olefiant gas, one part by weight of hydrogen combined with six parts by weight of carbon ; and that in carbon monoxide eight parts by weight of oxygen combined also with six parts by weight of carbon. Water was now examined.. It contains 88.89 per cent, oxygen and 11.11 per cent, hydrogen ; and these numbers are to each other as 8 to 1. The numbers which, in the first place, repre- sented the combining proportions of oxygen and hydro- gen respectively with carbon, are also found to represent, in the second place, the combining proportions of oxygen and hydrogen with each other. All subsequent exami- nations of other compounds led to similar results, and thus Dalton had discovered the law of multiple propor- tions. This may be stated as follows : If two bodies, A and B,form several compounds with each other, and we consider any fixed amount ATOMIC THEORY ATOMIC WEIGHTS, ETC. 17 of A, then the different amounts of B which com- bine with this fixed amount of A bear a simple ratio to each other. This is a fixed law, and it was generally acknowledged as such by contemporaneous chemists. Thus another characteristic of chemism was clearly pointed out. Atomic Theory, But Dalton did not stop with the discovery of the law of multiple proportions ; he sought for its explanation. He was thus led to propose the atomic theory, as affording the simplest explanation of the facts as observed. The question as to the ultimate constitution of matter had frequently and from the earliest dates been discussed. Two views were held at different periods and by different thinkers. One of these supposed matter to be indefinitely divisible. The other supposed that there was a limit to the divisibility, and that this limit was reached when the division had been carried down to certain small particles called atoms. After the discovery of the law of multiple proportions, however, the atomic theory was further developed, and, in consequence, acquired a more definite form, as the existence of atoms was supposed to have a direct connection with chemical combinations. The results of Dalton's investigations are not fully stated in the law of multiple proportions as above given ; another fact was made clear which is also of importance. The complete results may be stated as follows : It was shown that for each element a particular number might be selected ; and that this number or a simple multiple of it would repre- sent the proportion by weight in which this element com- bined with other elements. Dalton explained this by supposing that chemical action takes place between atoms, i. e., between particles that are indivisible and have definite weights. If chemical combination takes place between one atom of one substance and one atom of another substance, or between a simple number of atoms of one substance and a simple number of atoms of another, and "these atoms have definite weights, then indeed the explanation of the laws of definite and mul- tiple proportions is given. Thus the idea of atoms became a much more tangible 2* 18 DISCUSSION OF ATOMS AND MOLECULES. one than it had been up to that time. Not only were atoms supposed to have definite weights, but a method was given by means of which their relative weights could be determined. The number assigned to an element, representing its combining proportion, would also repre- sent the relative weight of its atom. The fact that the combining proportion of an element was in some cases represented hy a multiple of the simplest number was satisfactorily accounted for by supposing that in these cases more than one atom of the element combined with one atom of another element. Determination of Atomic Weights. The determination of atomic weights became now the chief, immediate prob- lem of the science of chemistry. Dalton's atomic theory was accepted by many, though not by all. The laws governing chemical combinations could not be doubted, but the explanation could be and was. Nevertheless, the importance of determining for each element the character- izing number, call it atomic weight or combining propor- tion, was acknowledged by all ; and consequently par- ticular attention was given to this field of research during the period directly following Dalton's publication. Let us see how thoroughly the desired object could be accom- plished alone by the aid of the principles laid down by Dalton. At the time of which we are speaking, the methods for chemical analysis w r ere still far from perfect, and hence most of the determinations then made required subse- quent corrections which were gradually forthcoming as the analytical methods were improved. This fact, how- ever, has nothing to do with the subject under consider- ation. The principle alone is involved. The question is to be answered : Can we, on logical grounds, with the principles contained in Dalton's investigations, ever deter- mine the relative weights of the atoms of elements ? To decide this question we must first examine more carefully Dalton's method for determining atomic weights. In the following brief discussion the correct numbers, as given by later analyses, are employed, instead of those origi- nally found. This does not interfere with the principle, and does simplify the matter otherwise. ATOMIC THEORY ATOMIC WEIGHTS, ETC. 19 Method for the Determination of Atomic Weights de- pendent upon Analysis. As the standard the combining number of hj'drogen was first selected, and this made 1. Hydrogen combines with oxygen in the proportion of 1:8; and as water was the only known compound of hydrogen and oxygen, the conclusion was drawn that the two elements were united atom to atom, and hence the atomic weight of oxygen was 8. Further, nitrogen is combined with hydrogen in ammonia in the proportion of 1 part by weight of hydrogen to 4f parts by weight of nitrogen. Ammonia was the only compound of nitro- gen and hydrogen known ; and the same reasoning, as above employed, led to the conclusion that the atomic weight of nitrogen was 4f. Considering for a moment these two simple cases, we see that the numbers thus found, as representing the relative weights of the atoms of oxygen and nitrogen, are founded partially upon hypothesis. There is nothing to decide as to the number of atoms of hydrogen or oxygen that are contained in water, nor of nitrogen and hydrogen in ammonia, and, of course, as long as this number is unknown, it is impos- sible to draw any positive conclusion with reference to the atomic weights of nitrogen and oxygen. Any con- clusion thus drawn is dependent upon a thorough knowl- edge of the compounds of the particular element under consideration. Such a number must finally be selected as is most in accordance with the facts. This selection must remain more or less arbitrary, as may be more clearly and decidedly shown. Take again the case of oxygen. A second compound of hydrogen and oxygen is v now known containing the elements in the proportion 1 : 16. At first sight, the explanation of this may appear simple enough. In this second compound there are two atoms of oxygen com- bined with one of hydrogen, and thus the proportion is satisfied. But may we not with equal right decide that in water there are two atoms of hydrogen combined with one of oxygen ? This would give us for oxygen the atomic weight 1 6, and, in the second compound, we would have contained one atom of each of the elements. Further, if we attempt to determine the atomic weight of carbon by Dalton's method, we shall encounter diffi- culties fully as great, and our final selection among many 20 DISCUSSION OF ATOMS AND MOLECULES. numbers will be arbitrary. Taking olefiant gas, we have hydrogen combined with carbon in the proportion 1:6; in marsh-gas the proportion of the same constituents is 1 : 3 or 2 : 6. If we suppose that in olefiant gas the elements are combined atom with atom, then the atomic weight of carbon would be 6, and consequently in marsh- gas we would have two atoms of hydrogen combined with one atom of carbon. But here again we can just as well suppose that in marsh-gas we have the simplest kind of combination, and this would give us for the atomic weight of carbon 3. Then in olefiant gas two atoms of carbon would be combined with one atom of hydrogen. Finally, let us take the oxygen compound of carbon. In carbon monoxide, carbon is united with oxygen in the proportion of 6 : 8 or 3 : 4 ; whereas in carbon dioxide the corresponding proportion is 3 : 8 or 6 : 16. Now let us suppose the atomic weight of ox} ? gen to be equal to 8, then, if carbon monoxide is the simpler of the two compounds, the atomic weight of carbon is 6 ; and in carbon dioxide there are two atoms of oxygen combined witli each atom of carbon. Here, again, it is evident that we can just as well imagine carbon dioxide to be the simpler compound, in which case the atomic weight of carbon would be 3, and in carbon monoxide there would be two atoms of carbon combined with one atom of oxygen. Between these different possibilities it is impossible to draw a logical conclusion with the aid of the knowledge which can be obtained by analysis. The number of similar instances might be increased indefinitely ; the inadequacy of the method could be made more strikingly clear by examples of a more complicated kind, but the cases men- tioned are sufficient for our purpose ; we are obliged to look for other methods for the determination of atomic weights if we would free the numbers from arbitrariness. Equivalents. This necessity was recognized first and most clearly by Wollaston in 1814. As no method pre- sented itself to him which would furnish a firm founda- tion for the determination of atomic weights, he proposed to abandon the idea of atomic weights entirely, and to substitute for it that of the equivalent, thus, as he sup- posed, getting rid of all hypotheses and obtaining numbers ATOMIC THEORY ATOMIC WEIGHTS, ETC. 21 that would be the simple expressions of proved facts. The equivalent of an element was to him that quantity of the element that possessed the same chemical value* as a given quantity of another element, that quantity of an element that could play the same role as a given quantity of another element. According to the conditions of this definition, it is plain that, in order to know what portions of two elements are equivalent, we must be able to com- pare the two. Hence, primarily, only of such elements as can be compared with each other, of such as possess a certain degree of similarity, can the equivalent quantities be determined. As this direct comparison is not always, nor, indeed, in the majority of cases, possible, recourse must be had to indirect comparison. To illustrate this let us take an example. Hydrogen and chlorine combine with each other in the proportion of 1 part by weight of hydrogen to 35.5 parts by weight of chlorine, and from this fact we draw the conclusion that 35.5 parts of chlorine are equivalent to 1 part of hydrogen. We find in the same way that 8 parts of oxygen, 80 of bromine, 16 of sulphur are all equivalent to 1 part of hydrogen. Knowing that 35.5 represents the equivalent of chlorine, we determine the quantities of sodium and silver that are respectively equivalent to this quantity of chlorine. We find for sodium 23 and for silver 108. These quantities of silver and sodium are further found to be equivalent to 8 parts of oxygen, 80 parts of bromine, and 16 parts of sulphur, and hence we conclude that they are also equivalent to 1 part of hydro- gen. Thus the equivalents of sodium and silver have been determined by the method of indirect comparison. Sodium and silver do not combine with Irydrogen, yet the equivalent numbers found are intended to express the proportions in which they would combine with hydrogen, provided such combination were possible. We are amply justified in this in most simple cases, but nevertheless it must be distinctly borne in mind that such numbers as are determined by indirect comparison with the standard, whatever this may be, are not in the strictest sense expressions of proved facts ; the last step in the deter- minations, however justified we may be in taking it, requires, nevertheless, the aid of hypothesis. But if the difficult} 7 thus referred to were the only one 22 DISCUSSION OF ATOMS AND MOLECULES. to be met with in the determination of equivalent numbers, such determinations would have nearly the full value claimed for them by Wollaston. This, however, is not the case. As soon as we consider any but the* simplest forms of compounds, we are left in fully as much doubt in regard to the equivalent numbers as we were in regard to atomic weights. If it be required to determine the quantity of carbon that is equivalent to 1 part of hydro- gen, the compounds of the two elements must be examined. But there are a great many compounds of these two ele- ments. Taking but two, olefiant gas and marsh-gas, we find that in the former (see ante, p. 16) 1 part of hydrogen is combined with (equivalent to) 6 parts of carbon ; whereas, in the latter, 1 part of hydrogen is combined with (equivalent to) 3 parts of carbon. What shall here decide which is the correct number? It is evident from such instances as this that the idea of equivalent is fully as uncertain as that of the atom was at the time we are considering. That an element could be equivalent to two entirely different quantities was in itself somewhat para- doxical, if the original definition of equivalent was re- tained. These difficulties seem not to have been apparent to Wollaston. He continued his determinations of equi- valents, and during this time a fusion of the ideas of equivalent and atomic weight took place unconsciously. As neither of these ideas was then definite, as to each of a number of elements a number of atomic weights could be assigned, and almost as many equivalents, the suc- ceeding period in the history of chemistry presents a disagreeably confused condition, until it became felt on all sides that some new idea or ideas must be introduced, if a fair foundation for the science was to be reached. Determinations by Berzelius. Before the necessary new ideas were introduced, the methods at hand were employed to the full extent. All known compounds of any given element were compared with each other, and a number finally selected, that would best satisfy the facts, to represent the equivalent of the element, or its atomic weight, as it was called by others. Berzelius attacked the subject most successfully. He laid down rules, by the aid of which the number of atoms of an element con- tained in a compound could be determined, and hence ATOMIC THEORY ATOMIC WEIGHTS, ETC. 23 also its atomic weight. Then, by more careful analyses than had been previously made, the atomic weights or equivalents of all the elements were determined. A large number o'f these determinations depended for their cor- rectness upon chemical rules, similar to the following, given by Berzelius : If an element forms several oxides, and the quan- tities of oxygen contained in them, as compared with a fixed quantity of the element, bear the proportion 1 : 2, then it is to be concluded that the first com- pound consists of one atom of the element and one atom of oxygen ; the second, of one atom of the element and two atoms of oxygen (or two atoms of the element and four atoms of oxygen). If the pro- portion is 2 : 3, then the first compound consists of one atom of the element and two atoms of oxygen; the second, of one atom of the element and three atoms of oxygen, etc. This rule covers those cases in which it is required to determine the atomic weight of an element by a consider- ation of its oxides. Other rules were given in which sulphur compounds, etc., were made the basis of calcu- lation. It will be observed that, although in these rules the ox^ygen and sulphur are taken as the elements, the number of whose atoms varies, the other elements might theoreti- cally be considered in the same way, and the atomic weights obtained would then be entirely different. An example will make this clear: Mercury combines with oxygen in two proportions. In the first compound, 8 parts of oxygen are combined with 200 parts of mercury ; in the second, 16 parts of oxygen are combined with 200 parts of mercury. Adopting the rule above laid down, we would conclude that, in the first compound, 1 atom of mercury is combined with 1 atom of oxygen ; and in the second, 1 atom of mercury with 2 atoms of oxygen. If, then, 8 is the atomic weight of oxygen, 200 is the atomic weight of mercury. But if, on the other hand, we con- sider the quantity of oxygen as remaining fixed, and that of the mercury as varying, then we would have in the first compound, 8 parts of oxygen combined with 200 parts of mercuiy, and in the second, 8 parts of oxygen combined with 100 parts of mercury ; and, by a similar 24 DISCUSSION OF ATOMS AND MOLECULES. process of reasoning, we might draw the conclusion that the first compound contains 2 atoms of mercury to 1 atom of oxygen, and the second, 1 atom of mercury to 1 atom of oxygen; and thus we would obtain 100 as the atomic weight of mercury instead of 200, as found above. Ber- zelius had made certain observations on chemical bodies upon which he based his rules, but, as we shall see, these observations were not sufficient. Another difficulty presented itself in the case of those elements that only combined in one proportion with oxygen. What should decide in regard to the number of atoms of oxygen contained in a compound of such an element? Here speculation was the only aid, and it often led to false results. The Principle of Substitution employed in the Deter- mination of Atomic Weights. The researches of Ber- zelius added a vast amount to the knowledge of the com- bining weights of the elements, and it must be acknow- ledged that the determinations made by him rested upon a somewhat firmer basis than the determinations made previous^. He made the fullest and most logical use of purely chemical means that could be made at the time. Subsequently, however, a new fact was discovered in connection with chemical compounds which proved of great value in simplifying the consideration of chemical phenomena, and also aided materially in the solution of the problem of the determination of atomic weights. This is substitution. The subject will be considered more fully in the last section ; here a brief explanation will suffice for the purpose of exhibiting its connection with the problem with which we are at present dealing. It has been found that certain elements have the power of entering into compound bodies, driving out some of the constituents, and taking the place thus left vacant. For instance, water contains 2 atoms of hydrogen and one of oxygen ; if we allow potassium to act upon water, a portion of the hydrogen is given off, and a new coin- pound containing both potassium and hydrogen, in ad- dition to the oxygen, is the result. If now potassium be further allowed to act upon this new compound, the hydrogen contained in it will be driven out, and its place ATOMIC THEORY ATOMIC WEIGHTS, ETC. 25 will be taken by potassium. Thus we obtain from water, by replacing its hydrogen by potassium, a compound containing 2 atoms of potassium and 1 atom of oxygen. This kind of action is called substitution. To show how, by taking into account the transform- ations included under this head, we may draw conclu- sions of importance with reference to atomic weights, one simple example may suffice: We have seen that the chief difficulty in determining atomic weights or equivalents by chemical means consists in the lack of data for estimating the number of atoms of an element contained in any given compound. Considering marsh-gas, we find that in it 1 part of hydrogen is combined with 3 parts of carbon, and, as above stated, we might conclude from this fact that the atomic weight of carbon is 3. If, however, we can by any means prove that there are more than one atom of hydrogen contained in the gas, the conclusion would require modification. By means of the process of substitution, this can be proved. By allowing chlorine to act upon marsh-gas under proper conditions, a portion of the hydrogen will be replaced, and a compound con- taining carbon, hydrogen, and chlorine will result. This new compound treated with chlorine again gives up a portion of its hydrogen, and takes up chlorine in its place. This operation may be repeated four times, and thus finally a compound is obtained which contains only carbon and chlorine. Each time the same amount of hydrogen is given up, and is replaced by an equivalent amount of chlorine. Thus it is plain that the hydrogen originally contained in marsh-gas is divisible into four parts, and we are obliged to accept the conclusion that there are at least four atoms of hydrogen contained in marsh-gas a conclusion which we could not possibly reach by the aid of the means heretofore considered. If now we take that amount of carbon which is in combina- tion with four atoms of hydrogen as representing one atom (and, by a consideration of the whole list of carbon compounds, we are justified in this step), then the atomic weight of carbon is 12. The method thus briefly illus- trated is capable of application to a considerable extent, but not to such an extent as to render it a general method for the determination of atomic weights. 3 26 DISCUSSION OF ATOMS AND MOLECULES. Consideration of Chemical Decompositions for the purpose of determining Atomic Weights One more method of reasoning must be referred to as having been employed, either for the purpose of furnishing proofs of the correctness of atomic weights determined by other means, or for the direct determination of these weights. An example will best make this matter clear. We wish to know, for instance, how many atoms of hydrogen are combined with nitrogen in ammonia ; or, having by the preceding method concluded that this number is 3, we wish to verify the conclusion by other observations. By treating nitric acid (which we will suppose to contain one atom of hydrogen to every atom of nitrogen) with hydrogen, we obtain ammonia. Now, if we consider the amount of hydrogen that in nitric acid was in combina- tion with the nitrogen, we find that, in the resulting ammonia, three times as much hydrogen is combined with the same amount of nitrogen. Further, ammonia combines directly with a number of compounds, and, if we examine the amount of hydrogen contained in this ammonia, we find that it must necessarily be represented with three or some multiple of three atoms of hydrogen. Thus, if we study the various cases in which ammonia is either formed, or destroyed, or enters into combination, we find alwa\*s that the quantity of ammonia thus playing a part must contain three or some multiple of three atoms of hydrogen ; and hence we are again led to the conclu- sion that, in ammonia at least, three atoms of hydrogen are combined witli every atom of nitrogen. The methods we have thus briefly described comprise all we have at our command for the determination of atomic weights dependent upon purely chemical pro- cesses. Consider these methods as we may, we must see that they are inadequate to the accomplishment of their object. The determinations may indeed be made, but at last there must always remain a doubt concerning the result. If then we can approach the subject from an entirely different direction, we shall succeed in reducing this doubt to a minimum, if we find that the results at first obtained assert themselves as correct in the second instance. Before passing, however, to a consideration of ATOMIC THEORY ATOMIC! WEIGHTS, ETC. 27 new methods for making these determinations, it will be well to apply the knowledge we have gained in fixing more definitely than lias yet been done the ideas of ele- ments and compounds. Elements. The theoretical idea of an element has already been stated. An element, strictly speaking, is a substance that cannot by any possible means be decom- posed into kinds of matter that are unlike. This defini- tion presupposes a knowledge of all possible means for decomposing bodies. Until we are positive that we are acquainted with all these means, we cannot be positive in regard to the existence of a single element. But it is plain that to assert the possession of this amount of knowledge would be in the highest degree presumptuous. We can then never assert positively that any given sub- stance is an element ; we can only say that, the means at our command being insufficient to bring about the de- composition of a given bodj^, we consider this substance an element until such time shall arrive when, new means being given, it shall be shown to be compound. Nume- rous instances of the change of opinion concerning the elementary character of different substances might be adduced, prominent among which would be the alkaline metals, the oxides of which were for a time looked upon as elements ; chlorine, which was looked upon as a com- pound body until it had been satisfactorily shown that we were not in possession of the means for decomposing it, etc. etc. Thus the number of elements, as stated at any given time, is entirely dependent upon the state of chemical analysis at that time, and is never an expression of an absolute fact. At present, the number of elements known is 64. In other words, we can recognize the existence of 64 different kinds of matter. If we consider the atoms which make up an elementary substance, we see that they must necessarily be of the same kind, how far soever, we consider the subdivision of the substance as taking place before the atom is reached. Accepting then the existence of atoms, an element may be defined as a substance made up of atoms of the same kind ; and we shall see that the definition of an atom, which will be given further on, makes this defi- nition of an element a strict one in every respect. 28 DISCUSSION OF ATOMS AND MOLECULES. Compounds. It has been stnted that observation shows us the existence of at least two varieties of com- pound bodies. To only one of these, however, is the name compound strictly applied, and then the name signifies a chemical compound. To the other class various names are applied, according to the nature of the substance, such, for instance, as mechanical mixture, solution, alloy, etc. At one time it was thought that no strict line of division could be drawn between these two classes. The same ultimate causes were supposed to give rise to the formation of both ; and the constituents of both were supposed to be held together by the same agent. It may be shown that there is a marked difference between them, sufficient to enable us to say, in most cases, with which we have to deal. Firstly, if we examine chemical compounds, we find that one of their most prominent characteristics is the posses- sion of properties which differ entirely from those of their constituents. Hydrogen, an inflammable gas, and oxygen, a gas and energetic supporter of combustion, combine to form a liquid, water, which is not inflammable and does not support combustion. Hydrochloric acid, a gas that turns vegetable blues red, and ammonia, a gas that turns vegetable reds blue, unite to form sal-ammoniac a solid that is without influence upon vegetable colors. Chlorine, a gas, and mercury, a liquid, give a solid with none of the properties of either. The number of these examples might be increased indefinitely, and in each case a similar result would be reached. Secondly, it will be found that no purely mechanical means will suffice to separate the constituents of a chemical compound from each other ; but for this pur- pose one of the so-called physical forces, as heat, light, electricity, chemism, is necessary. Thirdly, the most important characteristic of chemical compounds is to be found in the proportion by weight in which the constituents are bound together. As regards any compound of two elements, it is a fact that the con- stituents are present in fixed proportions by weight. If we bring these elements together without reference to their quantities, and the proper conditions be brought about to induce combination, it is found that a definite quantity of one combines with a definite quantit} 7 " of the ATOMIC THEORY ATOMIC WEIGHTS, ETC. 29 other; and, if the quantity of either present is in excess of the fixed quantity necessary for the formation of the compound, this excess will remain in its original form after combination has taken place. We can only vary the proportions to a very limited extent, and then not gradually, but according to a fixed rule. This is the circumstance which above all others enables us to assert positively that a given body is or is not a chemical com- pound. Mechanical Mixtures. To compare the second class of compounds with chemical compounds proper, let us first take the so-called mechanical mixtures. If we bring oxygen and nitrogen together, a homogeneous mixture of the two is formed, and this possesses the properties of both oxygen and nitrogen ; such a mixture, for instance, is the atmosphere of the earth. Gases mix in this way by virtue of their inherent tendency to expand indefinitely and completely fill the space offered to them. It is hence unnecessary to suppose that any special force acts in this gas to hold the constituents together. Many solids may be mixed in various ways, but no matter how. finely we may divide them, nor how intimately we may mix them, provided chemical combination does not take place, we can again separate the constituents of the mixture by mechanical means ; and the mixture possesses all the original properties of its constituents. In both these cases, further, the most varied quantities of the sub- stances may be employed, and, under the same conditions, the mixtures will be formed just as readily with one pro- portion as with another. Solutions and Alloys. On the other hand, those com- pounds which are known under the names of solutions and alloys are more closely allied to chemical compounds. We may have gases, liquids, or solids in the state of solution, that is, in combination with some liquid body, and to all appearance themselves in the liquid form. The external properties of one of the constituents are no longer recognizable, and they are, indeed, in part lost. A gas loses its ordinary elasticity when dissolved in a liquid. A solid loses the cohesion which before held its particles together. Two liquids combined in this way 3* 30 DISCUSSION OF ATOMS AND MOLECULES. lose some of their original properties, and receive new ones that represent a mean between the lost ones. In all these instances some force must be imagined as acting between the particles of the dissolved bodies and the particles of the solvents, which is greater in its effect than the cohesion that originally held together the par- ticles of the solid or liquid, or the repulsion that was exerted between the particles of the gas. Further, we have the case of alloys or compounds of two or more metals. These alloys present all the appearance of per- fectly homogeneous bodies, but nevertheless possess most of the properties of the constituents. Here, too, some force must be considered as acting between the unlike particles which differs from the ordinary force of cohesion. On examining the above-mentioned cases more care- fully, we find that there is, in almost all cases, a limit to the action of the force. Substances that are soluble in water are not usually soluble to an unlimited extent ; on the contrary, for any given temperature, the proportion of the substance that can be dissolved is fixed. But, between this fixed amount and the smallest possible quantity of the substance, all proportions are equally well dissolved. Some liquids mix with each other in all proportions, a perfectly homogeneous liquid being the result. Others dissolve each other to only a limited extent, the limits being, as in the case of solids and liquids, fixed for any given temperature. Whatever the force may be that is supposed to be the essential agent in the formation of these compounds in variable proportions, it is certain that the law or laws of its action have not been discovered up to the present. Some have looked upon it as identical with chemism. yet it appears that very distinct differences between the two can be pointed out. The first feature of these compounds that indicates a radical difference in the two forces is the retaining of the chief original properties of the constituents. If we dis- solve sodium chloride in water, we can obtain all the important effects from the solution that we could from the solid substance, and added to these we would then further obtain the effects of the water. And the same holds good for all solutions ; they can produce the effects of the substances dissolved and of the solvent combined. ATOMIC THEORY. ATOMIC WEIGHTS, ETC. 31 As we have seen, tliis is not true of chemical compounds proper. Again, and most especially is the difference marked, if we consider the proportions by weight in which the sub- stances combine in the two cases. Whereas, whenever chemical compounds are formed, the constituents com- bine in fixed proportions, in the case of mixtures, solu- tions, alloys, the constituents may combine in all possible proportions up to a certain fixed limit. Whether it would be expedient then to consider chem- ism and the force that is the cause of the formation of solutions, etc., as identical, but differing in. degree, is a question that cannot be here discussed. Certain it is, from the above remarks, that there exists sufficient differ- ence between them to warrant us for the present in restrict- ing the use of the name chemism to the designation of that force which is the essential cause of the formation of chemical compounds. In this sense it will be used in the following pages. The atomic theoi 1 }* accounts for the fact that bodies combine in definite proportions by sup- posing them to combine atom to atom, and these atoms to possess definite weights. According to this, chemism, in its restricted sense, is the force which is exerted between atoms. It will be shown that these atoms ma}' be either like or unlike. If they are like, the resulting body is an element ; if they are unlike, the resulting body is a chem- ical compound. It is plain, from the foregoing, that chemical com- pounds and elements are the only substances the study of which can lead to definite conclusions concerning the action of chemism, and hence we must confine ourselves to these in our subsequent study of this force. And first, we must return to that fundamental problem of chemistry the determination of atomic weights. It having been shown that results reached by the methods already given must necessarily be uncertain, we now proceed to attack the subject from a wholly new side. II. EXAMINATION OF GASEOUS ELEMENTS AND COMPOUNDS. As bodies present themselves to us in three different states of aggregation, the solid, liquid, and gaseous, 'so our methods of investigation of bodies must take differ- ent directions. The gases possess certain properties that are not possessed by solids and liquids, and in solids and liquids we detect certain general properties that we cannot detect in gases. The study of bodies in the form of gas or vapor has led to very important results of last- ing influence upon the science, and to these let us direct our attention. Investigations of Gay Lussac. In the year 1808, Gay Lussac and Humboldt discovered the fact that when hydrogen and oxygen combined to form water, the com- bination takes place between 2 volumes of hydrogen and 1 volume of oxygen. The simplicity of this relation induced Gay Lussac to take up the study of other gaseous bodies, with the view of determining whether similar relations existed between the volumes of other combining gases. His researches permitted him soon after to deduce the following law of combination by volumes : When two or more gaseous constituents combine to form a gaseous compound, the volumes* of the individual constituents as well as their sum bear a simple relation to the value of the compound. Thus, when hydrogen and chlorine unite to form hydro- chloric acid, it was found that 1 volume of hydrogen and 1 volume of chlorine formed 2 volumes of hydrochloric * In all cases where the volumes of different gases are com- pared, the gases are, of course, supposed to be under the same conditions of pressure and temperature. GASEOUS ELEMENTS AND COMPOUNDS. 33 acid gas. Two volumes of hydrogen and 1 volume of oxygen gave 2 volumes of water-vapor; 2 volumes of nitrogen and 1 volume of oxygen gave 2 volumes of nitrous oxide. Further, 3 volumes of hydrogen and 1 volume of nitrogen gave 2 volumes of ammonia, etc. etc. On comparing this result with that already obtained by Palton, and making use of the atomic theory, accord- ing to which combination between elements takes place between their atoms, we see that some simple relation exists between the volumes of gases and the relative number of atoms contained in these volumes. This we may express in general terms as follows : The number of atoms contained in a given volume of a gaseous body forms a simple ratio with the number of atoms contained in the same volume of other gaseous bodies. As will be readily seen, this gives no foundation for the determination of atomic weights, inasmuch as we have no means of fixing the va-lue of the " simple ratio," and without this we cannot determine the relative number of atoms contained in a given volume of gas. We know that 2 volumes of hydrogen combine with 1 volume of oxygen, and we know that 2 parts by weight of hydrogen combine with 16 parts by weight of oxygen. Further, the atomic theory tells us that a certain number of atoms of hydrogen of fixed weight combine with a certain number of atoms of oxygen of fixed weight, and that these num- bers bear a simple relation to each other; hence the rela- tion between the number of atoms of hydrogen in the 2 volumes, and the number of atoms of oxygen in the 1 volume, must be a simple one, but the facts do not fur- nish us with sufficient data to enable us to state what this relation is ; without further aid either from new facts or speculations, we cannot say what the atomic weights of these elements are. Auogadro's Speculations. The numbers expressing the specific gravities of gases or vapors are those numbers which express the relative weights of like volumes of these gases or vapors. Hence it is but restating, in another form, the principle above laid down to say that the specific gravities of gaseous bodies bear a simple relation to the atomic weights of these bodies. The force 34 DISCUSSION OP ATOMS AND MOLECULES. of this statement will be readily recognized on comparing the specific gravities of some gases with the atomic weights of the same bodies determined by chemical means. The atomic weights as determined by chemical means, however, differed from each other according to the method employed in this determination; but the difference being that between one number and some multiple of that number, it is immaterial which of these numbers we employ for the purposes of the comparison. Let us then take the first of those determined. The following table hardly needs explanation. The numbers in the second column (d) represent the specific gravities of the elements in the form of gas or vapor; the fourth column contains the ratios between the atomic weights (^4) and d = , Element. d. A. A d ' Hydrogen 0.0692 I 14.45 Chlorine 2.440 35,5 14.55 Bromine 5.54 80 14.44 Iodine 8.716 127 14.57 Oxygen 1.10563 8 7.24 Sulphur 2.23 16 7.17 Selenium 5.68 39.7 6.99 Tellurium 9.08 64 7.05 Nitrogen 0.9713 14 14.41 Phosphoru 3 4.50 31 6.89 Arsenic 10.6 75 7.08 Mercury 7.03 100 14.22 Cadmium 3.94 56 14.21 We see thus that the relation between the specific gravity and the atomic weight of seven of these thirteen elements is the same, being expressed by a number vary- ing but little from 14.4. Jn the six remaining elements of the list also the relation is virtually the same, about 7.1. And, in the latter case, the ratio is expressed by a number half as great as the first. A consideration of these relations led Avogadro,* in 1811, to propose an hypothesis which, if it could be well * In 1814 Ampere proposed a similar hypothesis. GASEOUS ELEMENTS AND COMPOUNDS. 35 founded, would prove of the greatest service in simplify- ing the problem of determining the atomic weights at least of gaseous bodies. It will be seen that, if in the above schedule the atomic weights of oxygen, sulphur, selenium, tellurium, phosphorus, and arsenic be doubled, A the ratio for all the elements in the list will be the d same constant number, viz., about 14.4. But the atomic weights above given have been determined purely empiri- cally, and we are as much justified in considering these numbers doubled the true atomic weights, as we are in accepting the ones given. If we make this change, then for the above thirteen elements, the following statement would be true : The atomic weights are to each other as the specific gravities of the vapors. An examination of compound gaseous bodies showed further that a simple relation also existed between their specific gravities and the numbers expressing the sum of the atomic weights of the constituents, these sums being to each other as the specific gravities. Avogadro's hypothesis to account for these relations may be stated in the following words : All gases or vapors, without exception, contain, in the same volume, the same number of ultimate particles or molecules. The molecules were not considered to be identical with the atoms, and it is well here to draw the distinction between the two as clearly as possible. Molecules of compounds, as understood by Avogadro, and as under- stood at present, are the theoretically smallest particles of these compounds. The molecule of water is the smallest particle of water that can exist as water. As water, however, is composed of two elements, of course the smallest particle of water must necessarily still be divisible into these constituents. The component parts of molecules are called atoms, and these are indivisible. In the case of water, the molecule has the same composi- tion as the mass of the compound, but, as will be shown, this molecule of water consists of two atoms of hydrogen and one atom of oxygen. The atoms are held together by chemism, the molecules by cohesion. Now there are good reasons, which will be considered below, for believing that, in their internal structure, elementary substances are, in some respects, analogous 36 DISCUSSION OF ATOMS AND MOLECULES. to compounds, and this belief was made a fundamental condition of Avogadro's hypothesis. According to this, it is impossible by purely mechanical means to subdivide an element so far as to reach its atoms ; but if we suppose it divided as far as possible by such means, we reach, as in the case of compounds, the molecule of the element, which is the smallest particle of the element that can exist and play the part of the element. This molecule, however, usually consists of atoms which are held together by chemism, and can hence only be separated by some means, other than mechanical, that are known to have the power of overcoming the force. From these considerations we are enabled to give definitions of the terms atom and molecule : A molecule is the smallest particle of a compound or element that is capable of existence in a free state. A breaking up of the molecule necessitates the destruction of the properties of the compound, and almost always of those of the element. Atoms are the indivisible constituents of mole- cules. They are the smallest particles of elements that can take part in chemical reactions, and are, for the greater part, incapable of existence in the free state, but are always found in combination with other atoms, either of the same kind or of different . kinds. And now the justice of the definitions of elements and compounds given above (see pp. 27, 28) will be recognized, viz., an element is a substance made up of atoms of the same kind ; a compound is a substance made up of atoms of unlike kind. Recognizing thus fully the distinction between atoms and molecules, we are prepared to further follow the reasoning of Avogadro. The experiments of Gay Lussac had already proved that, under the influence of heat, all gases expand in the same proportion for the same increase of temperature, and diminished in volume to the same extent for the same decrease of temperature. Further, Mariotte, in France, and Boyle, in England, had shown that all gases conducted themselves in the same way under the influ- ence of increased or decreased pressure ; that for the same increase or decrease of pressure the consequent GASEOUS ELEMENTS AND COMPOUNDS. 37 decrease or increase of volume was for the same volume of all gases the same. These facts considered indepen- dently would lead to a suspicion that all gases possess a similar internal structure, and the simplest hypothesis to account for this is just the hypothesis of Avogadro that the same volumes of all gaseous bodies contain the same number of molecules. This subject has been considered exhaustively from a purely physical standpoint. The principles of the mechanical theory of gases being ac- cepted, it was shown that the hypothesis of Avogadro would logically follow ; and then, by a purely mathemati- cal process of reasoning, it was shown that the hypothesis was an absolute necessit}^. A discussion of the subject in the direction indicated cannot here be taken up. For those who desire to follow the discussion, and to become acquainted with the methods that hjave led to the result mentioned, the following references will be of service : Kronig, Poggendorff's Annalen, 99, 316 ; Clausius, Fogg. Ann., 100, 360; Naumann, Annalen der Chemie u. Phar- macia, 1870, Suppl. Band 7, 340 ; Pfaundler, Pogg. Ann., 144, 428; Maxwell, Phil. Mag., 1860, [4] 19,19; Phil. Trans., 1867, 1 ; Phil. Mag., 1868, [4] 35, 185 ; Thomsen, Berichte der deutsch. chem. Gesellschaft zu Berlin, 1870, 829; Lothar Meyer, ibid., 1870, 864; Thomsen, ibid., 1870, 954; Lothar Meyer, ibid., 1871, 28; Mees, ibid., 1871, 272. As a grand result of the investigations that have been made on the internal structure of gases, it may be stated that Avogadro's hypothesis has throughout asserted its correctness, and it has at last become of fundamental importance in the science of chemistry. It is at present almost universally accepted by chemists, some, indeed, going so far as to speak of it as a law. Determination of Molecular Weights. What, then, do we gain by accepting the hypothesis ? It is plain that if equal volumes of all gases contain the same num- ber of molecules, we have a means given us at once for ascertaining the relative weights of these molecules. We have merely to determine the relative weights of equal volumes of the gases, and the numbers obtained will bear the same relations to each other as the molecular weights. Then accepting the weight of some molecule as a standard, 4 38 DISCUSSION OF ATOMS AND MOLECULES. and expressing the weights of the others in terms of this standard, the molecular weights are determined. If we call the molecular weight of hydrogen 2, for instance, and find the relation between this number and the number expressing the specific gravity of hydrogen, then we have also found the number expressing the relation between the molecular weights and specific gravities of all gases and vapors, without exception. The specific gravity of hydrogen as compared with air is 0.06926, the ratio 2 rTnT-"^ 28.88 ; hence, the specific gravity multiplied by 28.88 gives the molecular weight. If d = specific gravity, M= molecular weight, then the following formula will express the relation : M = d x 28.88. As the molecule of a body consists of atoms, so the molecular weight must be the sum of the weights of those atoms. In the case of an element, the atoms being of the same kind, the molecular weight must be a multiple of the atomic weight. Now we have already seen that for every element we can, by chemical analysis, determine some number that must represent either the atomic weight itself or a multiple of this weight. Hence, we have it in our power to determine by chemical analysis a multiple of the molecular weight of an element or a compound. This determination can be made with greater accuracy than that of the specific gravity of a gas or vapor, and it must be employed as a control for the determination of molecular weights b} 7 Avogadro's rule. If we take water, for instance, we find by means of anatysis that its molecular weight is either 9 or some simple multiple of 9. The specific gravity of water-vapor is 0.623, which multiplied by 28.88 gives 1*7.99 as the molecular weight. Hence, we conclude that 18 is the true molecular weight of water. The coincidence of the numbers determined according to the two methods will be seen in the case of a few elements and compounds in the following table. The numbers under M are those found by the analytical method, that one of a series of multiples being selected that agrees most nearly with the number found according to the rule ^/=28.88'X d. GASEOUS ELEMENTS AND COMPOUNDS. 39 Name. Specific gr. = d. 28.88 X^ M. Hydrogen .... 0.06926 [ 2 2 Nitrogen .... 0.9713 28.05 28 Oxygen .... 1.1 0503 31.93 32 Sulphur .... 2.23 64.4 64 Chlorine .... 2.45 70.75 71 Cadmium .... 3.94 113.78 112 Phosphorus 4.35 125.62 124 Bromine .... 5.54 159.99 160 Selenium .... 5.68 164.03 158.8 Mercury .... 6.98 201.58 200 Water .... 0.623 17.99 18 Hydrochloric acid 1.247 36.11 36.5 Sulphur dioxide . ; 2.247 64.89 64 Ammonia . . . . 0.597 17.24 17 Phosphorus trichloride 4.88 140.93 137.5 Arsenic trichloride . .-.* 6.30- 181.94 181.5 Boron chloride . 3.942 113.84 117.5 Marsh -gas 0.557 1608 16 Methyl chloride 1.736 50.13 50.5 Chloroform 4.20 121.29 119.5 Tin chloride 9.20 265.69 260 Silicon chloride 5.94 171.55 170 Zinc-methyl 3.29 95.02 95.2 Aluminium chloride 9.35 270.03 267 Iron trichloride 11.39 328.94 325 Number of Atoms in the Molecules of Elements. Although we are thus enabled by a simple process to determine the molecular weights of some of the elements, an important part of the real problem the determination of the atomic weights remains yet to be solved. If we could know in each case how many atoms are contained in a molecule, our difficulties would be at an end, but this we plainly do not know without the introduc- tion of considerations of a different kind from those with which we have had to deal as yet. Avogadro reasoned as follows, with reference to some of the simple chemical compounds: Given hydrochloric acid, it is re- quired to know how many atoms are contained in a molecule of hydrogen and in a molecule of chlorine. If in a certain volume of hydrogen there are contained say 100 molecules, then in the same volume of chlorine there is contained the same number of molecules. Now it is 40 DISCUSSION OF ATOMS AND MOLECULES. known that 1 volume of hydrogen combines with 1 volume of chlorine. Two volumes of hydrochloric acid gas are formed, and, according to the hypothesis, these two volumes in the case under consideration contain 200 molecules. But each molecule of hydrochloric acid must contain at least one atom of hydrogen and one atom of chlorine; hence, in 100 molecules of hydrogen and 100 molecules of chlorine there must be contained at least 200 atoms of chlorine and 200 atoms of hydrogen, or a molecule of either hydrogen or chlorine must contain at least two atoms of the corresponding element. Further, as no simpler compound than hydrochloric acid of hydrogen nor of chlorine is known, any conclusions which we may draw from a consideration of this compound must be valid for all compounds of these elements. The supposition that two atoms form the molecule of hydrogen and of chlo- rine satisfies all the facts known to us, and we hence rest with this supposition. As we take the atomic weight of hydrogen as the unit of these weights, its molecular weight will then necessarily be 2, on this basis ; and for this reason the number 2 was taken in the above table as the standard of comparison for other molecular weights. It must, however, be distinctly borne in mind that no proof is here given of the absolute number of atoms contained in the molecules of hydrogen and chlorine. We cau only say that at least 2 atoms must be present in each of the molecules. There may be a much greater number, but the data permit no speculations beyond this number 2. For all similar cases a similar process of reasoning may be employed, and with the same results. Whenever 1 volume of an elementary gas or vapor combines with 1 volume of another elementary gas or vapor to form 2 volumes of a compound gas or vapor, we are justified in concluding that each molecule of these elements contains two atoms. The elements that come under this head are hydrogen, chlorine, bromine, and iodine. If we pass to oxygen, we find a material difference in the method of combination. Here 2 volumes of hydrogen combine with 1 volume of oxygen to form 2 volumes of water-vapor. Let us reason as above. If in 1 volume of oxygen there are contained 100 molecules, then in 2 volumes of hydrogen there are 200 molecules. These GASEOUS ELEMENTS AND COMPOUNDS. 41 300 molecules combine to form 200 molecules of the compound. Now, in the molecule of water, there must be contained at least one atom of oxygen and one atom of hydrogen ; hence, there must be at least 200 atoms of oxygen and 200 atoms of hydrogen. But we know that in the original 200 molecules of hydrogen there were contained 400 atoms; hence, in each molecule of water, there must be 2 atoms of hydrogen. Water is the simplest compound of oxygen known to us (i.e., it contains the smallest quantity of oxygen in the molecule), and on this account we suppose the molecule of water to contain 1 atom of oxygen. If, then, each water molecule contains 2 atoms of hydrogen and 1 atom of oxygen, in the 200 molecules of water there are 200 atoms of oxygen and 400 atoms of hydrogen, and these are obtained from 100 molecules of oxj'gen and 200 molecules of hydrogen. Therefore, each molecule of oxygen, as well as each molecule of hydrogen contains 2 atoms. For sulphur the same is true, and is proved in a similar manner. Another method of reasoning, starting from entirely different facts, also led Favre and Silbermann to suggest that the molecule of oxygen consists of two atoms. They proved that carbon, when burned in protoxide of nitrogen, evolves more heat than when burned in oxygen. The most natural interpretation of this fact consists in admit- ting that, in each experiment, a chemical combination is destroyed whilst another is formed ; and that the amount of heat actually evolved is the difference between the amount of heat disengaged by the union of carbon with oxygen and the amount of heat absorbed by the decom- position of tUe oxide of oxygen in the first instance, and of oxide of nitrogen in the second. And, if the thermic effect is less with oxygen than with protoxide of nitrogen, that is due to the circumstance that oxide of oxygen (the molecule of oxygen 0=0) absorbs more heat in decom- posing than does the molecule of protoxide of nitrogen. One volume of nitrogen combines with 3 volumes of hydrogen to form 2 volumes of ammonia. Hence in the molecule of ammonia there are 3 atoms of hydrogen, and, ammonia being the simplest compound of nitrogen, we suppose that these 3 atoms of hydrogen are combined with 1 atom of nitrogen. As each molecule of ammonia contains one atom of nitrogen, and as, further, there are 4* 42 DISCUSSION OF ATOMS AND MOLECULES. formed twice as many molecules of ammonia as there were molecules of nitrogen originally, it follows that the molecule of nitrogen contains at least 2 atoms. By this means we are enabled to determine the atomic weight of the elements mentioned, for, if in their mole- cules 2 atoms are contained, we have only to divide the molecular weight found by Avogadro's rule, and cor- rected by analytical methods by two. But, accepting the atomic weights of hydrogen, chlorine, bromine, and iodine as known, we are enabled by another process to determine the atomic weights of such elements as com- bine with these to form gaseous compounds. Take again water. We find by a comparison of the compounds of oxygen that the molecule of water, as stated above, contains the smallest quantity of this element ; and hence we suppose this quantity to repre- sent 1 atom. We first find the molecular weight from the specific gravity of the vapor. This is 18. We analyze the compound, and find that it contains 88.89 per cent, oxygen and 11.11 per cent, hydrogen, or 8 pails of oxy- gen to 1 part of hydrogen. This being the relative pro- portion of the two elements in the compound, in 18 parts by weight, which represent the molecule, there are con- tained 16 parts of oxygen and 2 parts of hydrogen. The atomic weight of oxygen is hence 16, and in water one atom of ox^ygen is combined with 2 atoms of hydrogen. In the case of nitrogen, the application of the same principle must also lead to the same number previously found, viz., 14. We come to the conclusion that the molecule contains one atom of nitrogen. The molecular weight of ammonia we find to be 17. The analysis shows us that the elements are combined in the proportion of 14 parts by weight of nitrogen to 3 parts by weight of hydrogen. Hence 14 is the atomic weight of nitrogen, and the molecule of ammonia contains 1 atom of nitrogen with 3 atoms of hydrogen. Molecules of Elements which contain more or less than two Atoms. The molecules of the elements considered contain each two atoms. This is, however, not true of the molecules of all elements. Among those compounds of phosphorus which may be looked upon as containing 1 atom of this element in the GASEOUS ELEMENTS AND COMPOUNDS. 43 molecule is phosphine. The molecular weight of phos- phine is 34. The elements are contained in it in the pro- portion of 31 parts of phosphorus to 3 parts of hydrogen. Hence 31 is the atomic weight of phosphorus. On the other hand, we find the molecular weight of phosphorus itself to be 124, which shows that at least 4 atoms are contained in the molecule. The same is true of arsenic. For reasons similar to those given aboA T e, the molecule of mercuric chloride is supposed to contain one atom of mercury. The molecular weight of this compound is found to be 270.5, and the elements are contained in it in the proportion of 200 parts of mercury to 71 parts of chlorine, which gives 200 as the atomic weight of mer- cury ; and the atom of this element is combined with 2 atoms of chlorine. The molecular weight of mercury is 200 ; hence in the molecule of mercury there is contained but one atom. The same coincidence of atomic and molecular weight is noticed in connection with cad- mium.* * An interesting experiment has recently been performed, the results of which also show that the molecule of mercury in all probability consists of a single atom. The quantity of heat con- tained in a gas is defined as the total energy of its molecules, and this energy consists solely in advancing motion, if the molecule is looked upon as a mere material point. According to this, it is a simple matter to calculate the relation between the specific neat of a gas at constant volume and the specific heat at constant pressure. It was found, however, that, in the case of the gases examined, the theoretical value of this relation was larger than the value actually obtained by observation. If c represents the specific heat at constant volume, and c' the specific heat at con- stant pressure, then G = k represents the relation above referred C to. According to the theory, &=1.67, whereas observation gives k = 1.405. In other words, it requires more heat to raise the temperature of a gas, the volume remaining unchanged, than the theory demands. The heat which thus disappears may be transformed into an inter-molecular motion ; i. e., the atoms composing the molecule may have a motion relative to some centre of gravity. This motion would not show itself as tem- perature. If the molecule of the gas consists of one atom, then the theoretical and observed value of A; should be identical. The examination of mercury gave for k the value 1.67, which is that above given as the result of calculation. It is thus shown, by a method entirely independent of chemistry, that the molecule of mercury conducts itself like a material point, and this it could only be if it consisted of one atom. 44 DISCUSSION OP ATOMS AND MOLECULES. Varying Number of Atoms in the Molecule of one and the same Element. The specific gravity of the vapor of sulphur was stated in the above table (p. 39) to be 2.23, and this led to the molecular weight 64. Now it has been found that the specific gravity of this vapor varies according to the temperature at which it is determined. The determinations which gave the number 2.23 were made at temperatures above 800 C. (860 and 1040). Other determinations, however, made below 800 gave different results. At 524 (Dumas) and 508 (Mitscher- lich) the specific gravity was found to be 6.62 and 6.90 respectively, or three times as great as at the higher temperatures. These latter determinations would give the molecular weight 1 92, and, if 32 be the atomic weight of sulphur, then in the molecule of the vapor below 800 there would be contained 6 atoms, whereas above 800 there are contained only 2 atoms in the molecule. Selenium, so similar to sulphur in all other respects, presents similar phenomena, though not in so marked a degree. Here, too, it is noticed that the specific gravity of the vapor decreases with an increase of temperature, or, what, according to Avogadro's hypothesis, is the same thing, the molecular weight decreases with an in- crease of temperature. The application of the method thus described for the determination of the molecular weights of elementary bodies is limited, as we can convert only a few of these bodies into the form of vapor. Of many elements, how- ever, we know compounds that are capable of conversion into vapor or are themselves gaseous; and, as we can determine the molecular weights of these compounds, we are in many cases thus enabled to determine the atomic weights. The following table* contains a number of such compounds, together with the densities (d); the pro : ducts of the densities into the constant 28.88 (d X 28.88) ; the molecular weights as found by analytical methods (M)] and, finally, the relative quantities of the constitu- ents of the compounds contained in the molecules as determined by analysis : * " Die motlcrnen Tbeorien der Chemie," Lothar Meyer. GASEOUS ELEMENTS AND COMPOUNDS. 45 d. g I O >- - 5 IIS s. us . bO ^ jj o ^ TH ff T i -!1 i ! I < 00 , S I i s I I g-tdo I* o GO <* e $ 1 *! 5 A w o I o I 05 tsj z; 3 P. I s 1 5 s 5--a ii | c 1! M 88U9S TH Ol CO 00 Oi O TH C1 X C1 Hydrogen. Water. Ammonia. Methane. Phosphorus chloride. Now, by imagining the constituents of these types replaced by other atoms of equal valence, we can obtain all the examples above given, and a very large number in addition. Thus, H Cl may be looked upon as de- rived from H H by the replacement of one atom of II by Cl, an atom of equal valence. So, also, Ca=0 may be looked upon as derived from H H by the replace- ment of two atoms of H by one atom of Ca, the valence of which is equal to that of two atoms of H. Or, the typical element itself may be replaced by another of equal / H valence, as, for instance, in P H , which may be looked \H / H upon as derived from the type N H by the replacement \H of the triad N by the triad P. It is thus clear what rela- tion the types bear to similar compounds. When the so-called u theory of types" was proposed, a great deal of importance was attached to it. The true secret of chemical combination was supposed to have been discovered. Efforts were made to refer every known compound to some one of the types and thus to classify the compounds. These efforts were undoubtedly valu- able. They were the necessary precursors of our present views concerning the nature of chemical compounds. Through the theory of types we arrived at our present conception of valence. As long as the theory was in vogue, it was simply necessary to refer anj r given com- GENERAL CONSIDERATIONS. 105 pound to some particular type. As soon as it was shown to which type a compound belonged, investigation ceased. The internal arrangement of the atoms was not inquired into. To conceive of the valence of atoms is to take a step beyond the theory of types, to find in the atoms themselves the reason for the types. Now that we have taken this step, it is unnecessary to retain the ideas of types at all. It has served its purpose and led to another idea, imperfect to be sure, but nevertheless more perfect than its predecessor. At present, it is not only necessary to show the resemblance between some molecule and a typical molecule, but, in every case, we have to determine which elements are in combination with each other. When, according to this latter principle, the constitution of a compound is determined, resemblances will show themselves between molecules belonging to the same type, but these resemblances will only be necessary con- sequences of the resemblances between the typical atoms of these molecules. Residues. By far the greater number of chemical compounds are more complicated than those with which we have thus far been dealing. Let us inquire into the cause of the complexity noticed in them. If we take any of the formulas above given, as, for /H H / /TI instance, H Cl, H H, N H , and C <5 , \ \ \H X H and divide them at any part, we obtain two residues of equal valence. Thus, if we divide H Cl, we obtain II and Cl, both univalent ; if we divide H H, we obtain II and OH, and these are both univalent, for, as can be readily seen, the group OH requires a univalent atom or group to saturate it ; and this is what we under- stand by a univalent group. If we divide N H , we obtain II and NH. 2 , or H 2 and Nil; by the former division there are left two univalent, b} T the latter two bivalent 106 CHEMICAL COMPOUNDS. factors. And so in the case of H /H ^H X H if we divide this formula, the following cases are possible: H and CH 3 , H 2 and CH 2 , H 3 and CH; leaving in the first case two univalent, in the second two bivalent, and in the third two trivalent, factors. This principle may be carried out further in connection with other and more complicated formulas, and thus are obtained the formulas of a great variety of these so-called residues; in most cases, however, the division made and the residues re- sulting may be compared to the simpler forms described. We speak of a water residue, OH, which, on account of the exceedingly important part it plays in the constitu- tion of chemical compounds, has received a distinct name, hydroxyl; the ammonia residue, NH 2 , is called amide; the residue NH is called imide; the methane residue, CH 3 , is called methyl; the residue CH 2 , methy- lene, etc. etc. If we operate with the groups mentioned instead of with atoms alone, we shall find that we are able to build up a large number of formulas representing known com- pounds, as follows: / H Similar to N H , ammonia, we have: /" ^o N H , Nf , \H X H Hydroxylamine. Nitrous acid. N H , p OH , \H Methylamine. Phosphorous acid. P-CH 3 , etc. Trimethylphosphine. H /H Similar to C 0 , Ba | , etc. Sulphur dioxide. Sulphur trioxide. Barium dioxide. Double and Treble Union. We have alread}^ seen that one atom may be combined with another by means of more than one affinity acting in each. In the com- pound Ca=O, for instance, two affinities of one atom are saturated by two affinities of another. This kind of union may occur in connection with atoms that have a higher valence than two. It is particularly met with in carbon compounds, as in C 2 H 4 , ethylene. In this compound, it is believed that the two carbon atoms are united by H \ / H means of two affinities each, thus : ;>C C<( H/ \H Further, in some compounds, a treble union is found between atoms of the same kind, as in acetylene, C,H. 2 , in which it is believed that the carbon atoms are united by means of three affinities each, as follows: Lastly, in some compounds, it is believed that the carbon atoms are united partially by double and par- CH tially by single union, as, for instance, in | , which is the formula usually accepted for acrylic acid. 10 110 CHEMICAL COMPOUNDS. There are also a large number of compounds in which the carbon atoms are supposed to be united alternately by double and single union, and to form a closed chain, as in benzene : H C /- \ HC CH I1C CH V H We have thus illustrated the manners in which atoms are believed to combine to form the various chemical compounds with which we have to deal. We are far from asserting that the formulas which we have given are in all cases satisfactorily proved. We entertain serious doubts in regard to many of them. ' Nevertheless, bearing in mind what is meant by the constitution of compounds, we believe that the formulas given represent, in the majority of cases, truths which are capable of proof. The proofs must, of course, in each case, remain relative, and cannot be absolute. We start witli certain assump- tions of atoms and the nature of atoms, and form a con- ception in regard to the constitution of the simplest compounds. Now, in so far as this conception is true, in just so far, in the majority of cases, we can prove the constitution of compounds. Just as soon as our ideas in regard to the constitution of hydrochloric acid change, our ideas of the constitution of all other compounds must also change. Whatever is true in our present con- ception of the constitution of hydrochloric acid, is also true in our conception of the constitution of other chem- ical compounds. The proofs of the chemical constitution of compounds are of two kinds : 1. Those which depend upon decomposing the com- pounds into simpler constituents, or the analytical proofs, and GENERAL CONSIDERATIONS. Ill 2. Those which depend upon building the compounds up from simpler constituents, or the synthetical proofs. In many cases we are able to obtain both of these proofs, and if then we reach the same results b}' both methods, these results are rendered doubly sure. In many cases, however, only one kind of proof can be given at present, but this ma}* be so strong that the results are satisfactory. Where all proof is wanting, it is sometimes possible to propose a formula which very probably repre- sents the constitution. Of this latter kind of formulas, viz., those which have not been proved, but may be shown to be extremely probable, we have a great many. The methods of proof will be fully illustrated in the following section, in which we propose to consider the formulas of most of those chemical compounds which represent classes. The proofs for the formulas commonly accepted will be given in each case as fully as is com- patible with a work of this nature. We do not necessarily undertake to give all the proofs, though we shall give enough to show clearly upon what foundations our con- stitutional formulas are based enough to show that these formulas are entirely worthy of our careful study and our respect. CLASSES OF COMPOUNDS. Chemical compounds may be most conveniently clas- sified according to their chemical properties. No system of classification which has been proposed, up to the pre- sent, can, in any sense, be called perfect, and yet the sys- tem now in most common use is convenient, and has a fair foundation in facts. If we examine the compound which has the formula HC1, hydrochloric acid, and the compound which has the formula KOH, potassium hydroxide, we find that the two compounds differ very markedly from each other. The former has a taste which we call sour, the latter has the taste of lye, or an alkaline taste. The former will turn the color of many organic substances, while the latter will undo the work done by the former, restoring the 112 CHEMICAL COMPOUNDS. original color. In whatever way we may consider these two compounds, we shall find that they have opposite or complementary properties. They are both chemically active substances, capable of producing marked changes in large numbers of other compounds. If they are brought together they neutralize each other, that is to sa} 7 , they destroy each other's active properties and give rise to the formation of a new compound, differing en- tirely from the two which gave it birth. The two com- pounds, hydrochloric acid, HC1, and potassium hydrox- ide, KOH, are representatives of two great classes of compounds known as acids and banes. Many of the members of these two classes possess just as marked properties as do the two which have been mentioned, and for these the subdivision into acids and bases is rational and simple. But there are, further, some com- pounds which appear to possess the characteristics of both classes to a certain extent, and of neither class to any very great extent. For these compounds the system is not broad enough, though their number is not great. Probabl} 7 , when the nature of the two classes of com- pounds is fully understood, no difficulty will be found in determining to which class any given compound belongs. Acids. The properties which characterize acids are the following: 1. They have an acid or sour taste. 2. They change blue litmus red. 3. They act upon metals, hydrogen being evolved, and its place being taken by the metals, as, for instance : 2(HC1) -f Zn = ZnCl + 2H Hydrochloric acid. Zinc chloride. H a SO 4 + MB == MnSO 4 + 2H Sulphuric acid. Manganese sulphate. 4. They act upon metallic hydroxides, forming neutral substances and water, as follows: HC1 -f KOH = KC1 -[- H 2 O Hydrochloric Potassium Potassium acid. hydroxide. chloride. HNO, + NaOH = NaNO, + H 2 O Nitric acid. Sodium hydroxide. Sodium uitrate. H,SO 4 + Ca(OH), = CaS0 4 + 2H. 2 Sulphuric acid. Calcium oxide. Calcium sulphate. GENERAL CONSIDERATIONS. 113 Hydrogen Acids. All acids contain hydrogen. They may consist of hydrogen and only one other element, or of hydrogen and a group of other elements of greater or less complexity. The constitution of those acids which consist of hydrogen and only one other element is, of course, very simple and readily understood. There are but few examples of this kind, some of which follow: Hydrochloric acid, HC1 ; hydrobromic acid, HBr; sulph- ydric acid, H^S, etc. According to our conceptions of the nature of chemical constitution, compounds of the above formulas can only have one constitution. It is a noticeable fact that acids of this first and sim- plest class never contain more than two atoms of hydro- gen in the molecule ; or, that no element with a higher valence than two forms these simple acids. Hydroxyl Acids. By far the greater number of acids belong to the second class mentioned. They consist of hydrogen and a group of greater or less complexity, as, for instance, H(NO 3 ), nitric acid ; H(C10 :t ), chloric acid; H 2 (SO 4 ), sulphuric acid, etc. In nearly all acids of this kind, oxygen is one of the constituents of the group with which the hydrogen is combined. The hydrogen in these compounds is the changeable constituent. It is readily given up, and metals and groups are taken up in its place. The first question that would suggest itself,.in considering the constitution of acids, would be this: In what manner is the hydrogen in them held in combination : It is believed that investi- gations thus far made, justify the answer that the hydro- gen in these acids is almost always in combination with oxygen and, in a very few cases, with that element which is so similar to oxygen, viz., sulphur. The proofs for this statement cannot always be given. In the cases of many acids, there exist no independent proofs that in these the hydrogen is combined with oxygen. On the other hand, there are so many acids in which it can be satisfactorily shown that the hydrogen is in combination with oxygen that the above answer seems to be justified. We accordingly write the formulas of acids in such a way as to indicate the fact of union between oxygen and Ii3 T drogen thus: 10* 114 CHEMICAL COMPOUNDS. (HO)NO, (HO)C10 2 (HO) a SO, Nitric acid. Chloric acid. Sulphuric acid. Or these same formulas may be made still more defi- nite by writing them as follows : H_0 X H NO., H O C10 2 , >S0 2 . H Ca ; \OH ^O/ Sulphuric acid. Calcium sulpbate. 2 OH and >Ba . NO O/ Nitric acid. Barium nitrate. Further complications are introduced when trivalent and quadrivalent elements enter into the composition of salts. From what has been said, however, the consti- tution of these salts will be readily understood. Complex Salts. Just as we have a few bases which consist of hydroxyl combined with groups of atoms, so we have salts which may be considered as derived from acids by the replacement of hydrogen by groups of atoms. Thus, a salt obtained from the acid NO, OH, and the base UO OH, has the constitution expressed by the formula N0 2 O UO. Here the hydrogen of the acid is replaced by the group UO, which is univalent. Anhydrides. The constituents of water may be ab- stracted from many acids, and thus is formed a new class of compounds called anhydrides. The most striking- characteristic of these compounds is their power to form acids with water, or to form salts by direct union with bases. The following are examples of anhydrides : Sul- phuric anhydride, SO 3 ; nitric anhydride, N 2 5 ; phos- phoric anhydride, P 2 5 ; acetic anhydride (C 2 H } 0) 2 O, etc. When an anhydride is formed from a monobasic acid, two molecules must combine to furnish the hydrogen for the water. After the abstraction of the water, the two acid residues remain united, through the instrumentality of an atom of oxygen, thus : NO.-OH) N0 2X I - H 2 = )0 ; N0 2 OH) NO/ 2 molecules Nitric acid. Nitric anbydride. 120 CHEMICAL COMPOUNDS. C 3 H 3 OH) C.,H 3 CK H,0 = >0 . C. 2 H 3 OH ) 0,H 3 O X 2 molecules Acetic acid. Acetic anhydride. When an anhydride is formed from a bibasic acid, a molecule of water may be given off' from a molecule of acid, thus: on SO/ . H^O S0=0 ; M)H Sulphuric acid. Sulphuric anhydride. /OH C0< HO C0=0 . X)H Carbonic acid (hypothetical). Carbonic anhydride. Or, two molecules of a bibasic acid may unite and give off one molecule of water, forming a compound which is, at the same time, an acid and an anhydride, thus : /OH ] /OH so/ so/ - H.O = >0 . . . S OH | . ' S '\OH 2 molecules Sulphuric acid. Pyrosulphuric acid. When an anhydride is formed from a tribasic acid, several possibilities may present themselves. 1. One molecule of the acid may lose one molecule of water, a compound being formed which is anhydride and acid, thus: OH ^O PO OH H 2 o = ro OH \ ^OH Metapho*phoric acid. Phosphoric acid. 4 2. Two molecules of the 'acid may lose one molecule of water, a compound being formed which is a tetrabasic acid, and, at the same time, an anhydride: GENERAL CONSIDERATIONS. 121 ,OH 1 X)H PO OH PO OH N> > - H ' - PO/OH \OH Pyrophosphoric acid. 2 mol. Phosphoric acid. 3. Two molecules of the acid may lose three molecules of water, a complete anhydride being formed: J&SL PO OH PO OH \3H PO OH ,0 3H.O == 2 mol. Phosphoric acid. PO PO Phosphoric anhydride. By combining a larger number of molecules of the acids and abstracting different numbers of molecules of water, a great variety of anhydrides might be produced, at least theoretically. Not many such complicated pro- ducts are positively known, however. From tetrabasic acids and acids with even higher basicity, corresponding anhydrides maybe derived. With an increase in the basicity of the acids, the complexity of the resulting anhydride is, of course, increased. Proofs of the Constitution of Anhydrides. In regard to the correctness of the formulas given for these anhy- drides, it can only be said that they are the simplest which we can conceive of. If we acknowledge that acetic anhydride, (C 2 H 3 O) y O, consists of two acid-residues combined by an ox_ygen atom, then, by analogy, we must acknowledge that the other anhydrides, mentioned above, are constituted as represented by the formulas. But, can we assume any other formula for acetic anhydride ? We know that the acid has the constitution C 2 H 3 O(OH); 11 122 CHEMICAL COMPOUNDS. we know that the anhydride has the empirical formula, C 4 H H O H , and that it is formed by the simple abstraction of water from the acid ; we know that the hydroxyl group has the power to separate from the acid with compa- rative ease. What, then, is more natural than to assume that the water which is given off from the acid is formed from the hydroxyl groups, and that the groups C 2 H H remain undecomposed ? But this would give us, besides the water, two groups, C. 2 H 3 O and an oxygen atom. These are all combined in one molecule, and, as we believe, in such a way that the oxygen atom exercises a linking power between the two groups or acid residues, C.fl.,0 giving the formula )O . C 2 H 3 (K When an anhydride is formed by the abstraction of water from one molecule of an acid, the simplest conclu- sion we can draw is that an oxygen atom fills the place before occupied by two hydroxyl groups. We have no proof of this, to be sure, but it would be gratuitous to offer any other explanation of the formation of these anhydrides at present. Oxides. Just as anhydrides may be obtained from acids by the abstraction of water, so the oxides may be regarded as anh} T drides of the bases. The consideration of the oxides is simpler than that of the anhydrides, because the bases themselves are generally simpler than the acids. The simplest oxides are .those obtained from the hydroxides of univalent elements, examples of which follow : KOH) K x - H,0 = \0 KOH) tf 2 molecules Potassium Potassium hydroxide. oxide. ) Na x H 2 >0 . Na OH ) Na/ Sodium hydroxide. Sodium oxide. GENERAL CONSIDERATIONS. 123 Of oxides obtained from the hydroxides of bivalent elements, we have, among others, the following: /OH Ca( H 2 = C=0 ; Calcium hydroxide. Calcium oxide. /OH Sr< H.O = Sr=0 . M)H Strontium hydroxide Strontium oxide. Theoretically, an intermediate anhydride mo3 r be de- rived, from either of the two proceeding oxides, analogous /OH so/ to the formation of pyrosulphuric acid, j>O , from so two molecules of sulphuric acid, thus : c/ -[ - B < - > -OK\ 2 mol. Calcium hydroxide. Intermediate anhydride. No such compounds have been obtained as yet. From the hydroxides of trivalent elements we may have more than one oxide formed. If one molecule of the hydroxide loses one molecule of water, a body is formed which is oxide and l^droxide at the same time. Reference has been made to these compounds (see ante, p. 118), under the head of bases. The compound A10 OH may be regarded as derived from the hydroxide A1(OH) 3 , by the loss of one molecule of water from one molecule of the hydroxide. It is both oxide and hydroxide. The most common method of formation of oxides from hydroxides of trivalent elements consists in the union of two molecules of the hydroxide to lose three molecules of water, thus : 124 CHEMICAL COMPOUNDS. Al OH Q OH -OH Air 3H 2 = >0 AU ^* "" Aluminium \OH 2 mol. Alum nium hydroxide. The principle of the formation of these oxides is thus seen to be the same as that of the formation of anhy- drides. What was said in regard to the constitution of the latter holds good in regard to .the constitution of the former. The view stated is the simplest which the facts permit. Analogy between Salts and Anhydrides and Oxides. As we saw above, a salt is either an acid in which the hydrogen is replaced by a base residue, or a base in which the hydrogen is replaced by an acid residue. In those salts which are derived from acids containing hydroxyl, we have a base residue and an acid residue united by means of oxygen. On the other hand, in many anhydrides, we have an acid residue and an acid residue united by means of oxygen, while in oxides, we have two base residues united by means of ox} T gen. COMPOUNDS OF CARBON. We have thus briefly considered the different classes of compounds, and have shown upon what foundations our ideas in regard to the general constitution of these classes of compounds rest. Among the compounds of carbon, we have many representatives of each of the classes above considered, and all that has been said holds good for these compounds; but owing to some peculiari- ties of carbon, which distinguish it from the other elements, certain things hold good for the carbon compounds in general, which do not hold good for the corresponding GENERAL CONSIDERATIONS. 125 compounds of other elements. In the following para- graphs, therefore, the general formulas of the different classes of carbon compounds will be briefly treated. Hydrocarbons. Of the compounds of carbon those which it forms with h} T drogen are, in general, the simplest, and, of the hydrocarbons, marsh-gas, or methane, CH 4 , is the simplest one. With oar present ideas in regard to constitution, there can be but one formula for this H compound, viz.: H C H , which represents merely H that a quadrivalent atom of carbon is saturated by means of four hydrogen atoms. This is the most rational sup- position which we can make with reference to this com- pound. The formula is certainly not proved, but it is exceedingly probable. As marsh-gas is a very important member of the group of carbon compounds, let us inquire more particularly concerning the grounds upon which the above formula is based. We first determine the empirical .formula, CH 4 , by means of analysis, and the determination of the specific gravity of the vapor of the compound. This formula is the expression of a fact and an hypothesis. The fact expressed is that methane contains 75 per cent, carbon and 25 per cent, hydrogen. The hypothesis is that the molecules of all chemical compounds, in the form of gas or vapor, have the v same volume as a mole- cule of hydrogen. This hypothesis tells us the weights of the atoms contained in the molecule of methane and the weight of the molecule of methane, and hence, further, the number of atoms of carbon and hydrogen contained in the molecule. Knowing the above, it remains for us to determine in what manner these atoms are united, or, what is the same thing, to determine which atoms are in combination with each other. From our knowledge of hydrogen, we assume that it acts in this compound, as in all its other compounds, as a univalent element. But if it does act thus, then two hydrogen atoms cannot be united in the molecule CH 4 , for, if they were, they could not remain a part of the molecule, as their affinities would 11* 126 CHEMICAL COMPOUNDS. be satisfied by their action upon each other. Conse- quentty, the hydrogen atoms must all be in combination with the carbon atom. This result we express by the H [ C H , formula H C H , which is, further, in accordance H with what we know concerning carbon, this element being in almost all cases quadrivalent. A question which would naturally suggest itself in connection with the compound CH 4 is this : Do all the hydrogen atoms play the same part in the molecule ? In regard to this point, we can only say that, as far as investigations go, an affirmative answer to this question seems to be justified. If these hydrogen atoms were dif- ferent, then, by replacing different ones by the same ele- ment or group, products should be obtained which are not identical. No such results have been reached, although the hydrogen atoms in methane have been replaced in a great variety of ways. Homologous Series. Starting with methane, we have a series of hydrocarbons of the general formula C n H2-f2- These resemble each other in many respects, and differ from each other in their formulas in a very simple way. The difference in the formulas of any two contiguous members of this series is CH 2 . Such a series is called an homologous series. A number of similar series is known. In the methane series we have: Methane, CH 4 ; ethane, C.,H 6 ; propane, C 3 H 8 ; butane, C 4 H 10 , etc. The general principle according to which the members of this series are formed is found in the combination of the carbon atoms in open chains. Thus, as we have seen above, if two carbon atoms combine in the simplest manner possible, viz., by one of their affinities each, a chain is formed having six free affinities, as follows : C C . If three carbon atoms combine in the same i i way, a chain is formed having eight free affinities, thus: C C C . In the same way, four carbon atoms I ! I GENERAL CONSIDERATIONS. 127 combining would give a chain having ten free affinities, etc. etc. By saturating these free affinities by hydrogen, we would get compounds of the formulas C. 2 H 6 , C S H 8 , C 4 H 10 , etc. etc., which are the formulas of the hydro- carbons above given. Experimental Proofs. Certain experiments have been performed which prove the correctness of the views in regard to the nature of the combination in the methane series of hydrocarbons. If methane is treated with chlorine, the following reaction takes place : H H H C H 4- Cl Cl = H C Cl + Cl H . ! I H H Methane. CLlormethane. If the product is treated with sodium, the chlorine is extracted, and a compound of the formula C a H 6 is ob- tained, according to the following equation: H H H H H-C :ci+ Cl: C H -f 2Na = H C C H + 2NaCl. MM II H H H H 2 molecules Chlormethane. Ethane With ethane similar reactions may be realized, and a product, C 3 H 8 , obtained, thus : H H H H i I II 1. H-C-C H + C1-C1 = H C-C-C1 + HC1 ; U il A Ethane. Chlurethane. H H H H H H 2. H C-C- ici + Cli C-H + 2Na = H C-C-C H -f 2NaCl . I I i I I ' I H H H H H H Chlorethane. Chlormethane. Propane. It is perfectly plain that, by continuing these reactions with the new compounds obtained, we would have it in our power to build up a whole series of hydrocarbons corresponding to the series given above. If the combi- 128 CHEMICAL COMPOUNDS. nation always took place in the manner described, we should have simple chains, in which all the carbon atoms except those at the ends, would have two free affinities, while those at the ends would have three free affinities. The hydrocarbons themselves would be respectively: H^.CH^.CH^.CH^ H 3 C.CH 2 .CH 2 .CH..CH 3 , etc. etc. These are called normal hydrocarbons, to distinguish them from others of the same general composition, but of different constitution, which will be treated of in a later paragraph. Alcohols. Running parallel to the series of hydro- carbons which we have just considered, is a series of compounds which maybe looked upon as derived from the hydrocarbons by replacing a hydrogen atom in each by the univalent group OH, or hydroxyl. These com- pounds are to organic chemistry what the hydroxides of the metals are to inorganic chemistry. They are known as alcohols. The simplest of these is derived from H methane and has the formula H C H . A Proofs. One of the hydrogen atoms of those alcohols which contain but one oxygen atom, differs from the others. It is easily replaceable by certain groups known as acid groups, which we shall consider hereafter. It is also replaceable by metals. In a compound of the for- mula CH 4 O, we must assume that one hydrogen atom is in combination with the oxygen atom, while the other three are not, in orfrer to account for its characteristic behavior. Again, if we treat the alcohol with HC1, the oxygen atom and the peculiar hydrogen atom are given off together, and their place is taken by a single atom of chlorine. This shows that the hydrogen and oxygen were present in the form of a univalent group, or as hydroxyl, which is the on'ly form that satisfies these conditions. H,C OH -f H(NO, ( ) == H a C0(N0 3 ) + H 2 O. Mtthyl alcoiiol. Nitric acid. Nitric e.her. Water. GENERAL CONSIDERATIONS. 129 H 3 COH + HC1 = H 3 C 01 + H,0. Metbyl alcohol. Hydrochloric Chlorine thane. Water. acid. H S C OH + Na = H :i C ONa + H. Methyl alcohol. Sodium. Sodium methylate. Hydrogen. Further, the hydroxyl group can be introduced into the hydrocarbons and the alcohols thus obtained. In order to obtain the alcohol CH 4 0, we start from chlor- methane, CH 8 C1. If this is treated with the hydroxide of silver, the following reaction is realized: CH 3 C1 + Ag(OH) = CH :i .OH +' AgCl. Chlormetbane. Methyl alcohol. The above proofs suffice to show the correctness of the H I formula H C H for the first member of the H series of alcohols. Having once recognized the presence of hydroxyl in this alcohol, we would naturally expect to find it in the other alcohols. It is found in them all, and may be detected in the manner indicated in the case just considered. Classes of Alcohols. It has been found that there are three distinct classes of alcohols, which have been called, respectively, primary, secondary, and tertiary. These differ from each other, in their properties, very markedly. This difference is particularly noticed when they are sub- jected to the influence of oxidizing agents, when they undergo change, as follows : Primary alcohols are converted first into aldehydes, and then into acids containing the saVne number of carbon atoms. Secondary alcohols are converted into acetones, which, when further oxidized, yield acids with a smaller number of carbon atoms. Tertiary alcohols are decomposed without previous formation of aldehydes or acetones, yielding acids with a smaller number of carbon atoms. Primary Alcohols. These differences in the properties are undoubtedly due to differences in constitution. In 130 CHEMICAL COMPOUNDS. all primary alcohols we find that the group CH .OH, or H C O H , is present. This we saw in methyl alcohol, H which is a compound of this group with hydrogen, thus: H C H . In i ethyl alcohol, the next member of the series, this group is also present. This follows as soon as we acknowledge the presence of hydroxyl in the H H I | alcohols ; for, in the compound H C C H , it i-i makes no difference which hydrogen atom is replaced by hydroxyl, the resulting compound will, in every case, have the same constitution and will necessarily contain the group CH...OH. In all alcohols which conduct them- selves as primary, the presence of the group CH .OH can be proved in a similar way. They are all derived from methyl alcohol by the replacement of a hydrogen atom with hydrocarbon residues of various composition and constitution. By replacing a hydrogen atom with methyl, CH 3 , ethyl alcohol, CH .CH..OH, is obtained. By replacing a hydrogen atom with ethyl, C 2 H., propyl alcohol, C,H 5 .CH. 2 .OH, is obtained. By replacing a hydrogen atom with propyl, C 3 H 8 , butyl alcohol, O t EEg.GH r OH 9 is obtained, etc. Secondary Alcohols. If we replace two hydrogen atoms of methyl alcohol with hydrocarbon residues, alcohols are obtained which do not contain the group CH^.OH, as is evident from the following examples: GENERAL CONSIDERATIONS. 131 H CH 3 C 2 H 5 I I I H C H, H 3 C C H; H 3 C C O H . H H H Methyl alcohol. Isopropyl alcohol. Secondary butyl alcohol. These bodies contain the group CH.OH, and are repre- sentatives of secondary alcohols. The simplest example of this class of bodies is iso- propyl alcohol, the formula of which is given above. Proofs of the General Formula of Secondary Alcohols. There are two alcohols of the formula, C : ,H S 6. One of these conducts itself like the primary alcohols, and is hence supposed to contain the group CH^.OH. An alcohol isomeric with the primary alcohol cannot contain the group CH 2 .OH, but must contain the group CH.OH, as may be readily shown. Both of the alcohols are H H H I I I derived from the same hydrocarbon, II C C C H . H H H In this hydrocarbon there are only two kinds of hydrogen atoms, viz., those in combination with the central carbon atom, and those in combination with the terminal carbon atoms. If we replace any one of the latter by hydroxyl, we obtain primaiy propyl alcohol containing the group CH. 7 .OH. Whereas, if we replace one of the former hydrogen atoms by hydroxyl. we obtain secondary propyl alcohol containing the group CH.OH. Only these two cases are possible. But, again, this secondary alcohol is prepared by allowing nascent hydrogen to act upon acetone. It will be shown that acetone can only have the constitution OB. C O . Now, in being converted into secondary CH 3 propyl alcohol, acetone takes up two atoms of hydrogen, and the only place where these hydrogen atoms can find 132 CHEMICAL COMPOUNDS. entrance into the above molecule is at the central carbon atom. One of the bonds of union between the oxygen and carbon is loosened, and hydroxyl is formed; thus the carbon atom becomes possessed of a free affinity, which is at once saturated with hydrogen, and we have the : c o group CH.OH or H C H, which is bivalent. Similar considerations in connection with other sec- ondary alcohols lead to similar results, and hence the conclusion is drawn that all secondary alcohols contain the group CH.OH. Tertiary Alcohols. If we replace three hydrogen atoms of methyl alcohol by hydrocarbon residues, alcohols are obtained which contain neither the group CH 2 .OH, nor the group CH.OH, as may be seen in the following examples: H CH 3 C.H 5 H C H , CH 3 C H , CH 3 C H . I I I H CH, CH 3 Methyl alcohol. Tertiary butyl alcohol. Tertiary amyl alcohol. These bodies contain the group C.OH, and are repre- sentatives of tertiary alcohols. The simplest example of this class of alcohols is ter- tiary butyl alcohol, C 4 H 10 O, the constitution of which is indicated by the formula given above. Proofs The proofs for the correctness of this formula are the following: There is a hydrocarbon, the formula of which can be OH. shown to be CH 3 C CH 3 . From this, two alco- H hols are derived, one of which conducts itself as a primary GENERAL CONSIDERATIONS. 133 alcohol, and the other of which does not. The former CH 3 must have the formula CH 3 C CH 2 OH . The J. only alcohol derived from this hydrocarbon which is not a primary alcohol must have the formula CH 3 CH 3 C CH 3 , and hence contains the group O I H OH , which is trivalent Further, similar con- i siderations of other tertiary alcohols indicate that in them also the group C.OH is contained; and consequently this is looked upon as the characteristic group of these bodies. Determination of Alcohols. With the knowledge thus gained with reference to alcohols in general, it is plain that we have it in our power to determine in any par- ticular case, 1, whether the body we are dealing with is an alcohol; and, 2, whether it is a primary, secondary, or tertiary alcohol. The first thing to be done is to determine whether the body contains hydroxyl or not. Treatment with the chlorides of phosphorus, either the terchloride or pentachlonde, is one of the best means of making this determination. If by treatment with the chloride a product is obtained containing one atom of chlorine in the molecule in the place of an oxygen atom and a hydrogen atom, we can assume that hydroxyl was present. If this hydroxyl is the alcoholic hydroxyl, then its hydrogen must be capable of replacement by the so- called acid groups. To determine this, acetyl chloride, C S H S O.C1, is very frequently employed. If this body is allowed to act upon a substance containing an alcoholic hydroxyl group, the chlorine of the chloride combines 12 134 CHEMICAL COMPOUNDS. with the hydrogen of the hydroxyl, forming hydrochloric acid, and the acid group C 2 H a C) takes the place of the hydrogen, thus: R_OH -f C 2 H 3 O.C1 = R_O G\H,0 -f HCL Alcohol. Acetyl chloride. New product. If the reaction takes place in this manner, we are justified in concluding that the hydroxyl group is alco- holic, or that the body under examination is an alcohol. It remains still to determine whether the alcohol is primary, secondary, or tertiarj 7 . This can be accom- plished by subjecting it to the influence of oxidizing agents. If it yields an aldehyde, and then an acid containing the same number of carbon atoms, it is a primary alcohol. If it yields first an acetone, and then by further oxida- tion breaks up, yielding an acid or acids containing a smaller number of carbon atoms, it is a secondary alcohol. If. without first yielding an aldehyde or an acetone, it breaks up directly with the formation of an acid contain- ing a smaller number of carbon atoms, it is a tertiary alcohol. The above tests then enable us to determine a part of the constitutional formulas of many compounds. If we have by means of these tests determined that a compound is a primary alcohol, we assume that it contains the group OH 2 .OH. If it is a secondary alcohol, it contains the group CH.O H. And, if it is a tertiary alcohol, it con- tains the group C.OH. But these groups may enter into a great variety of compounds; and frequently, after we have determined the presence of one or the other of these groups, it would still be necessary to determine the con- stitution of the group with which it is combined. These special determinations will be considered later. Mercaptans. If in place of hydroxyl, in the alcohols we have considered, we introduce the group HS, bodies are obtained which have been called mercaptans. These are in many respects analogous to alcohols, though in I' -"'r reactions they differ from them somewhat. Their constitution is the same as that of the alcohols. The principal method for their formation consists in the action of potassium sulphydrate, KSH, on the chlorides of GENERAL CONSIDERATIONS. 135 alcohol residues. These latter are obtained by replacing the hydroxyl of alcohols by chlorine, and the reaction for the formation of the mercaptans takes place as follows: RC1 + K SH = R SH -j- KC1. Alcoholic Mercaptan. chloride. It will thus be seen that the group SH occupies the place which was occupied by the group OH in the original alcohol. Theoretically, a mercaptan may be prepared corresponding to every alcohol. Thus we might have primary, secondary, and tertiary mercaptans, correspond- ing to all the known primary, secondary, and tertiary alcohols. Only such mercaptans have been prepared up to the present as correspond to the primary alcohols. Acids What has been said above concerning acids in general is true of the acids of carbon. They contain hydroxyl, and possess the properties which we have recognized as belonging to acids. In general, they are weaker than other acids, though they differ among each other in strength between comparatively wide limits. We have several series of acids of carbon, corresponding to the series of hydrocarbons and alcohols. The simplest carbon acid is derived from methane, and has the formula H It differs from the simplest alcohol in O=C O H containing an atom of oxygen in the place of two atoms of hydrogen. Just as this alcohol consists of hydrogen combined with the group CH.OH, so the acid consists of hydrogen combined with the group CO. OH. This is the characteristic group of the acids of carbon. Proofs. In the first place, the presence of hydroxyl is proved the same as in the case of ordinary acids. In the acid H s CO a , if we assume the presence of hydroxyl, we have the formula HCO OH. Further, the other hydrogen atom contained in the acid does not conduct, itself ,as if it were in combination with oxygen, but 1. ' same as hydrogen atoms which are in combination with carbon direct!}'. No changes which the acid undergoes indicate any connection between this hydrogen atom and 136 CHEMICAL COMPOUNDS. oxygen atom, so we may conclude that they are not present as hydroxyl. But, if they are not present as hydroxyl, they must be united directly with the carbon H atom, and the formula is . Now by 0=0 O H certain appropriate reactions, it is possible to replace that hydrogen atom in this acid which is in direct combination with the carbon by groups such as CH :H , C.^H 5 , etc. The compounds thus obtained must contain the group CO.O H. They possess all the properties of acids. Methods for the Formation of the Acids of Carbon. The methods of preparation of the acids of carbon enable us also to judge of their constitution. Some of these methods may be briefly described. 1. The simplest acid, above referred to, viz., H 2 CO V , is 'obtained by bringing carbon monoxide, CO, together with potassium hydroxide, KO hi. The two substances com- bine directly, yielding the potassium salt of the above acid, thus: CO + KOH HC0 2 K. From this experiment we conclude that, in the salt, one of the oxygen atoms is in direct combination with carbon, as it was in carbon monoxide, while the other oxygen atom serves the purpose of linking the carbon atom to the potassium. Hence the group COOK or O=C O K I is present in the salt, and the group O=C O H in the acid. I 2. When methane, CH 4 , is allowed to act upon carbonyl chloride, COC1 , one of the chlorine atoms is replaced by the residue CH 3 , thus: (I.) CfJ 4 -f COCI 2 == CH 3 .COC1 + HC1. When the product is treated with water, the second chlorine atom is replaced by OH, as follows: (II.) CB..COC1 -f HHO = -CH 8 .CO.OH + HC1. GENERAL CONSIDERATIONS. 137 Now carbonyl chloride is obtained by the direct addition of C1 2 to carbon monoxide, CO ; and hence must have the constitution . The simplest interpre- Ci tation of reaction (I.) above is that the residue CH 3 takes the place occupied by one of the chlorine atoms, which H 3 C C=O would give j . Lastly, the simplest inter- 01 p fetation of reaction (II.) is that the hydroxyl group enters into the place of the second chlorine atom, which H 3 C C=0 gives as the constitution of the product This product is acetic acid, a homologue of the simplest acid of carbon. It contains the group CO.OH. CN 3. The compound cyanogen, I , is converted into CN an acid by the action of water. This acid has the for- mula C.,H 2 O 4 . It is a bibasic acid, and hence contains two hydroxyl groups, which would lead to the formula C 2 2 (OH) 2 . As both the hydroxyl groups conduct them- selves in exactly the same way, it is concluded that they are combined in exactly the same way. The only for- improbable formula as, for instance, C<^ , and |\0 this would contain no Irydrogen, so that it could not be considered as characteristic of acids. Lastly, the group COH cannot lose hydrogen and gain oxygen and still remain trivalent. As then the only alcohols which can yield the group COOH by the oxidation of their own characteristic group are the primary alcohols, and as those alcohols which cannot yield this group by such oxidation do not yield corresponding acids, the con- GENERAL CONSIDERATIONS. 139 elusion is drawn that the change which takes place when primary alcohols are oxidized consists in the conversion H of the group C H into 0=C H, which is a. necessary constituent of carbon acids. Aldehydes. Aldehydes are products derived from the partial oxidation of primary alcohols, the group CH 2 OH being converted into COH. This group is not identical with the group C H of tertiary alcohols, but 1 has the constitution expressed by the formula C O . It is a univalent group, just as the group CH 2 OH, from which it is derived, is univalent; whereas, the tertiary alcohol group, COH, is trivalent. The aldehydes are intermediate products between primary alcohols and the acids which these yield. It was shown that the acids are formed from these alcohols by the extraction of hydrogen and addition of oxygen. If hydrogen is only abstracted and no oxygen added, the product is an alde- hyde, thus: R CH 2 OH, R COH, R COOH. Primary alcohol. Aldehyde. Acid. Proofs. The proofs of the general constitution of aldehydes are similar to those given for the acids. Take, for instance, the simplest aldehyde. This has the for- mula H 2 CO. The tests for the presence of hydroxyl, above considered, if applied to aldehydes, show that the group is not present. On the contrary, if the aldehydes be treated with the chloride of phosphorus, the oxygen atom is extracted and its place is taken by two chlorine atoms. This shows that the oxygen was held in combi- nation by two affinities of the carbon atom, and, conse- quently, it could not have been present in combination with hydrogen, forming hydroxyl. We are thus led to 140 CHEMICAL COMPOUNDS. H the formula 0=0 H for the above compound. It consists of a hj'drogen atom combined with the group / H C=O . Other aldehydes are derived from this sim- plest one by replacing one of the hydrogen atoms with a residue of greater or less complexity. Thus, we may intro- duce the group C H 3 or C 2 H 5 , and we would obtain the com- pounds CH 3 COH and C 2 H 5 OOH respectively, both of which are aldehydes. The methods for the preparation of aldehydes also fur- nish proof of the constitution above ascribed to them. Some of these are the following: 1. We have already seen, that, when acids are treated with the chlorides of phosphorus, their hydroxyl is replaced by an atom of chlorine, yielding chlorides of acid residues. Each such chloride, as was shown, con- tains the group 0=0 . If we could replace the chlorine atom in this group by hydrogen, we would / H plainly have the characteristic aldehyde group 0=0 . Such a replacement has been effected in the cases of some of the chlorides, and the resulting bodies were found to be the expected aldehydes. 2. When a salt of any acid of carbon is mixed with a salt of the simplest acid of carbon (formic acid), of the formula H.CO.OH, and the 'mixture distilled, an alde- hyde is obtained together with a carbonate. The carbon- ates are derived from a bibasic acid, and have the formula , 0=0 . It seems rational to suppose that the groups OM have passed directly from the compounds in which they were originally contained to the carbonate, and that the group CO also has been derived directly from one of the original acids. If these suppositions are correct, GENERAL CONSIDERATIONS. 141 then we are led to the conclusion that the aldehyde resulting from the described reaction contains the group / H C=O . For, let R CO.OM represent the formula of any salt of a carbon acid, and H.CO.OM a salt of formic acid. On bringing these two compounds together and heating them, either one of two things can take place if the above suppositions are correct. The groups forming the carbonate may be split off thus : R_[~CQHOM" H CO OM or, thus : R CO |OM H 'CO OM The remaining groups, uniting in the simplest way, will > give n s, in the first place, a compound, R C=O , and, in the second place, a compound of exactly the same constitution. 3. Aldehydes are formed only from primary alcohols, not from secondary or tertiary alcohols. If we examine the group characteristic of the alcohols, we shall find that the only one of them which is capable of transformation / H into the group C=O is that of primary alcohols, / H CJETjOH; and, further, the group C=O is the only one containing carbon, hydrogen, and oxygen in the same proportions which can be derived from the group of primary alcohols, and not from those of secondary and tertiary alcohols. These considerations make it exceedingly probable that all aldehydes contain the group C=0 , as above stated. 142 CHEMICAL COMPOUNDS. Acetones. Acetones are products of the partial oxidation of secondary alcohols, the group C OH H being converted into the group C=O . The alde- hydes, too, contain the group C^O ; but it is further characteristic of aldehydes that one of the affinities of this group is saturated with hydrogen, giving the com- plete group C=O . On the other hand, it is charac- H teristic of acetones that both of the affinities of the group C=0 are saturated with hydrocarbon residues. Thus the simplest acetone known has the formula 3 0=0 , CET 3 both the affinities of the characteristic group being satu- rated with residues of the hydrocarbon methane, CH 4 . * Proofs. As just stated, the simplest acetone has the formula C 3 H 6 O. If a chloride of phosphorus be allowed to act upon this compound, the result is similar to that obtained in the same experiment with aldehydes, viz., the atom of oxygen is abstracted, and two chlorine atoms take its place. This shows that the oxygen was not present as hydroxyl, but was combined with the carbon atom by means of two affinities, forming the group (U. GENERAL CONSIDERATIONS. 143 Again, if nascent hydrogen is allowed to act upon this acetone, secondary propyl alcohol is the product, and the CH 3 I /OH alcohol has the formula C-S . From this we |\H CH 3 conclude that in acetone, as well as in secondary propyl alcohol, the two groups CH 3 are present; and we are CH 3 thus led to the formula C=0 for the simplest ace- CH 3 tone. We recognize in this formula what we have stated to be the characteristic of all acetones, viz., it consists of two hydrocarbon residues combined by means of the bivalent group C=O . The following methods of preparation serve as proofs of the accepted constitution of acetones: 1. Just as aldehydes are obtained from acid chlorides by replacing the chlorine with hydrogen, so acetones are obtained from the same chlorides by replacing the chlo- rine with hydrocarbon residues. By treating acetyl chloride, C,H 3 O.C1, with zinc methyl, Zn(CH 3 ) 2 , ordinary acetone, CO(CH 3 ) 2 , is produced, together with zinc chlo- ride, ZnCl 2 . The formula of acetyl chloride is known to be CH 3 C=0 . Hence the simplest interpretation of the above- reaction is that a methyl group of zinc methyl takes the place of a chlorine atom in acetyl chloride, thus: /CH 3 CH-C-CH, + ZnCl.. CH 3 C CH 3 \ 144 CHEMICAL COMPOUNDS. And this leads us clearly to the formula above assumed as representing the constitution of acetone. 2. When the salts of many acids of carbon are sub- jected to dry distillation, acetones are formed, together with a carbonate or carbonates. This reaction is analo- gous to the reaction for the preparation of aldehydes, loy the distillation of a mixture of the salt of some carbon acid and a salt of formic acid. What was said in regard to the latter reaction, showing that the group C=O must be present in aldehydes, holds good in regard to the re- action under consideration, and shows just as conclusively I that the group C=0 must be present in acetones. Let I R.CO.OM represent a salt of an acid of carbon. Its decomposition by heat may be represented as follows: B.JCO.OM] R.CO.IOM The residues uniting, we have a compound, R CO R, which has the general formula of an acetone, as above assumed. Or let R.COOM represent the salt of one carbon acid and R'.COOM the salt of another carbon acid, in which R and R' are both hydrocarbon residues. The decomposition, which takes place when a mixture of these two salts is heated, is represented as follows: R.I COO M R'.CO|OM _ This gives a compound of the formula R CO R'. It will be seen that one of the first conditions for the production of an acetone by means of this reaction is that neither of the salts employed be a formate, H.COOM, as the use of the latter would give rise to the formation of an aldelryde. 3. Acetones are produced by the partial oxidation of secondary alcohols. Considerations, similar to those GENERAL CONSIDERATIONS. 145 employed in the cases of acids and aldehydes, show that the supposition, that the group CO is present in acetones, I is most in harmony with the fact of the ready transform- ation of secondary alcohols into acetones. Ethers. When acids and bases, in general terms, act upon each other, salts are formed, water being eliminated. Just so when alcohols and carbon acids act upon each other, bodies, similar to salts, are formed, water being eliminated : H.COOH + C 2 H 5 .OH = H.CO.OC,!!. + H 2 O ; Formic acid. Alcohol. New body. /OH /O.CH 3 S0 2 < + 2CH 3 .OH = S0 2 < 4- 2H a O . Sulphuric acid. Methyl alcohol. New body. NO OH + C 2 H 5 OH = N0 3 OC 2 H 5 + H 2 0. Nitric acid. Alcohol. New body. It will be seen that these bodies differ from salts in that they contain hydrocarbon residues in the place of metals. Salts were denned as acids in which the hydro- gen of the hydroxyl group is replaced by a base residue. These bodies are acids in which the hydrogen of the hydroxyl group is replaced by a hydrocarbon residue. All bodies of this kind are called ethers, and to distin- guish them from another class of bodies known as simple ethers, and which will be considered below, they are usually known as compound ethers. The analogy between com- pound ethers and salts is very close, and hence, if the nature of salts is understood, the nature of these ethers will also be readily understood. We may have compound ethers derived from monobasic, bibasic, tri basic, etc. acids, and we may have compound ethers containing uni- valent, bivalent, trivalent, etc. hydrocarbon residues ; as, for instance, Ethers from monobasic acids : NO,.O.C,H 5 , CH 3 .CO.O.CH 3 , etc. Ethyl nitrate. Methyl acetate. 13 146 CHEMICAL COMPOUNDS. Ethers from bibasic acids: /CO.O.CjHa etc. ,O.CH 3 /CO.O.C,!^ X O.CH 3 \CO.O.C a H 5 Methyl sulphate. Ethyl succiuate. Ethers from tribasic acids ,O.C 2 H 5 ,CO.O.C,H 5 PO O.C 2 H 5 , C 3 H 5 CO.O.C 2 H 5 , etc. \).C 3 H 5 \CO.O.C,H. Ethyl phosphate. Ethyl tricarballylnte. The above ethers all contain univalent hydrocarbon residues. Among those containing bivalent residues may be mentioned CH b .CO.CK CH :r CO.O X Ethylene diacetate. Proofs. The fact that compound ethers are, in many cases, formed by the direct action of acids upon alcohols, and that water is formed at the same time, taken together with our knowledge concerning the constitution of acids and alcohols, points clearly to the constitution for these ethers which has been above assumed. But another method of formation is just as decisive in its testimony. If the silver salts of acids are treated with the chlo- rides, bromides, or iodides of hydrocarbon residues, com- pound ethers are formed in which the hydrocarbon resi- dues are found in the place of the silver which was in the salts, and the silver itself is found in combination with the chlorine, bromine, or iodine which was in com- bination with the h}^drocarbon residues. This is seen in the following typical reactions : CH 3 .CO.OAg + C,H 5 I = CH 3 .CO.O.C 2 H 5 + Agl. Silver acetate. Ethyl iodide. Ethyl acetate. /CO.OAg /CO.OCH 3 CTT / !O/r i TTT\ P TT / J_ 9 A o-r 2 lA -f- A^Ha 1 ) = ^2 M 4\ H- ^ A g'- x CO.OAg \CO.OCH 3 Silver succiuate. Methyl iodide. Methyl succinate. ' Simple Ethers. Simple ethers correspond to the metallic oxides. They consist of two hydrocarbon resi- CH. >0 , CH/ Methyl ether. ' H > , CH/ Methyl-ethyl ether. CH> Ethyl ether. GENERAL CONSIDERATIONS. 147 dues united by means of an oxygen atom, just as the metallic oxides consist of two base residues united by means of an oxygen atom. Examples of these are the folio wins: : etc. Proofs. The constitution of these compounds is ren- dered clear by a consideration of one of the principal methods of their formation. When an alcohol is treated with sodium or potassium, as we have seen, the hydrogen of the hydroxyl is replaced by the metal emplo3-ed. We thus obtain compounds such as sodium ethylate C a H 5 .ONa, sodium methylate CET.ONa, etc. If these compounds are further treated with the iodides of hydrocarbon residues, the iodine com- bines with the metal and the residues unite. Thus we have the following reactions : C 3 H 5 .ONa + C 3 H 5 I = C 2 H 5 -0-C 2 H 5 + Nal, Sodium ethylate. Ethyl iodide. Ethyl ether. CH 3 .ONa + CH S I = CH :? CH 3 + Nal, Sodium methylate. Methyl iodide. Meihyl ether. CH 3 .ONa -f C 2 H.I = CH O C 2 H 5 + Nal, Sodium methylate. Ethyl iodide. Methyl-ethyl ether. C 2 H..ONa + CH 3 I = C 2 H 5 CH 3 + Nal. Sodium ethylate. Methyl iodide. Methyl-ethyl ether. From these reactions we see what the constitution of the ethers formed jnust be. We are in each case justified in assuming that the hydrocarbon residue enters into the new compound in the place occupied by the metal, and, according to our conceptions concerning alcohols, this metal is united to the rest of the molecule in which it is contained by means of an oxygen atom. Anhydrides. The anhydrides of carbon compounds are derived from carbon acids in the same way that an- hydrides in general are derived from acids ; and all the possibilities which we considered above hold good for these anhydrides. There are anhydrides derived from monobasic, bibasic, tribasic acids, etc. There are partial 148 CHEMICAL COMPOUNDS. and complete anhydrides: but, further, there are anhy- drides derived from compounds which partake of the doable character of alcohol and acid. In these com- pounds the hydroxyl which imparts the alcoholic cha- racter and that which imparts the acid character, both together furnish the elements which form the water given off. Peculiar Anhydrides. Lactic acid has the formula /OR CH .CH/ . It contains an alcoholic hydroxyl \COOH and an acid hydroxyl. When the acid is distilled it loses -0-, water, and a compound of the formula CIL.CH is formed. This is lactic anhydride. As will be seen, the anhydride is formed by the loss of the elements of water from both hydroxyl groups together. Salicylic , acid has the formula C 6 H 4 \ It forms an \COOH anhydride in a similar manner, having the formula -0- SUBSTITUTION PRODUCTS. ' We have thus far considered the various classes of chemical compounds which are known to exist, and have shown that each class is characterized by some peculiarity of constitution which we recognize in each member of the class. There is in each compound a group which deter- mines its character, making it an acid or an alcohol, an acetone or an aldehyde, etc. As long as this group re- mains unchanged, the compound belongs to the same class. If the group is changed, the compound loses its GENERAL CONSIDERATIONS. 149 characteristics, and belongs to another class. On the other hand, the residues with which the class groups are united may undergo a variety of changes without inter- fering at all with the general properties of the compounds. The most common of these changes are those which are effected by substitution. Chemical compounds act upon each other, in general, in two ways. 1st. They unite directly, forming only one product, as we see in the following reactions: + HC1 = Ammonia. Ammonium chloride. C 2 H 4 + Br 2 ** C 2 H 4 Br 2 Ethylene. Ethylene bromide. 2d. They exchange certain constituents, forming two or more new products, thus: C 2 H 6 + C1 2 = C 2 H 5 C1 + 01 Ethane. Chlorethane. CH 8 .COH + 6C1 = CCl a .COH + 3HC1 Aldehyde. Tiichloraldehyde. C 6 H 6 + H,S0 4 =* C 6 H 6 .Sq,H + H 2 O Benzene. Sulphobenzenic acid. C 6 H 6 +.-HNO, = O.H,(NOJ + H,O . , Benzene. Nitrobenzene. The latter kind of action is by far the most common. It is that which is called substitution. In the above examples, the principal products are called substitution products, though, strictly speaking, both products are substitution products. While the principle of substitution is recognized in. connection with nearly all chemical reactions, and hence nearly all chemical compounds may be considered as substitution products with reference to some other com- pounds, still it is customary to include under this head only those products which are formed by the replacement of hydrogen in carbon compounds; and the substitutions which are spoken of are only those which can be actually effected not imaginary cases. Substitution Products containing Chlorine, Bromine, or Iodine. The simplest examples of substitution pro- ducts are those which are formed by the action of any of 13* 150 CHEMICAL COMPOUNDS. the so-called haloids (01, Br, T) upon carbon compounds. The action consists in the abstraction of one or more atoms of hydrogen from the compound, and the filling of the places left vacant by a corresponding number of atoms of the substituting element. The constitution of the products is the same as that of the compounds from which they are derived. Thus we have acetic acid, CH.^.CO.OH; if chlorine acts upon it, the following reactions take place successively: CH 3 .CO.OH -f C1 2 = CH 2 C1.CO.OH -f HOI. CH 2 C1.CO.OH + C1 2 = CHC1.CO.OH -f HC1. CHC1 2 .CO.OH + C1 2 = CC1 3 .CO.OH + HC1. The constitution of these three different products is essentially the same as that of the acid from which they are derived. j\mong these simple substitution products, however, differences are possible, and are actually observed, which are not possible in the original compounds. Take the compound propane, C 3 H 8 . The constitution of this hj'rtro- H H carbon is H C C H . The hydrogen atoms cannot be arranged in any other way with reference to the carbon atoms. There is only one hydrocarbon of this composition possible. But the carbon atoms in this compound differ from each other. The two which are represented as ending the chain in the formula are alike, while the central atom differs from them. The first are in combination with carbon by means of only one affinity each, while the central atom is joined to carbon by means of two affinities. We would naturally expect then that the difference between these two kinds of carbon atoms would cause a difference between the hydrogen atoms combined with them. If such a difference exists, then, different products must be obtained according as we replace a hydrogen atom attached to one of the terminal carbon atoms, or another hydrogen atom attached to the central carbon atom. Thus, if in the following formula GENERAL CONSIDERATIONS. 151 7 '? T r 1 TT p p p A 1 l_y v> \j H6 3 H H H 5 8 we replace any of the hydrogen atoms numbered 1, 2, 3, 4, 5, 6 by an element such as chlorine, the resulting com- pound would in each case be the same. If, however, we replace one .of the hydrogen atoms numered 1 or 8 by the same element, a compound of the same composition, but of different constitution, would be obtained. The formulas of the two compounds would be respectively CH 2 C1CH 2 .CH 3 and CH :J .CHC1.0H 3 . Thus we see that the position of a substituting element must be taken into consideration in studying the consti- tution of compounds. In connection with the individual compounds, which will be briefly considered in the last section of this book, the methods will be described which enable us to determine the positions of substituting ele- ments and groups. Complex Substitution Products. Under this head we include all those products which are formed by replacing the hydrogen of a carbon compound either partially or wholly by groups. In accordance with what has just been said concerning the simple substitution products, it is plain that, in studying the constitution of the complex substitution products, two things must be taken into consideration : 1st. The constitution of the substituting group itself, and, 2d. The position of the group in the molecule of the substitution product. We shall here only take up the first part of the problem. Constitution of Substituting Groups. The groups which we shall have to consider are the following : The cyanogen group ON, and an isorneric group; the sulpho group S0 3 H ; the nitro group N0. 2 ; the nitroso group 152 CHEMICAL COMPOUNDS. NO; the amid o group NH 2 ; the imido group NH ; and a few other groups intimately connected with those mentioned. Constitution of the Group CN. That acid of carbon which consists of a nitrogen atom and a hydrogen atom combined with a carbon atom, viz., hydrocyanic acid, has already been referred to. Ity appropriate reactions it is possible to transfer the group ON, contained in hydro- cyanic acid, to other compounds in such a way that it takes the place of hydrogen, forming a substitution pro- duct. It is univalent, and hence its constitution is expressed by the formula C=N. We have the follow- ing reactions: CH 2 C1.COOH -f KCN = CH 2 (CN).COOH + KC1 Monocbloracetic Potassium Cyanacetic acid, acid. cyanide. C 2 H 4 Br 2 -f 2KCN = C V H 4 (CN) 2 + 2KC1. Ethylene bromide. Potassium cyanide. Ethylene cyanide. These substitution products, which consist only of the group C~N combined with a hydrocarbon residue, are called nitriles. A few other compounds are known which have the same composition as the nitriles, but a different consti- tution. They are known as isonitriles or carbylamines. They contain the group C==N . This group is univa- lent, just as the group (feN, but the nitrogen atom contained in it plays the part of a quinquivalent element, whereas, the nitrogen atom of the group C=N is tri- valent. We maj^ have thus the two compounds C 2 H 5 -~CzEN and C=N C 2 H 5 , of the same composition, but different constitution. Both these compounds are well known. The former is called ethylcyanide or propionitrile, the latter ethylcarbylamine. As ethyl cyanide, when treated with an alkali, yields propionic"acid, we conclude that the carbon atom of the group CN is united directly with the hydrocarbon residue. For if it had not been, the removal of the nitrogen ought to have caused the formation of a product containing a smaller number of carbon atoms than the cyanide itself. The reaction which does actually take place is that which GENERAL CONSIDERATIONS. 153 we have considered above as giving rise to the formation of acids from the cyanides, viz.: C,H 5 .CN + 2H a O = C,H 5 .COOH + NH 3 Ethyl cyanide. Propiouic acid. If the group CN had been in combination with the hydrocarbon residue by means of the nitrogen atom, which would be the case if the group had the constitution Cj=iN , we should expect the nitrogen atom to remain in combination with the hydrocarbon residue, in case of decomposition, or we should expect the nitrogen atom to take with it the carbon atom with which it is most inti- mately combined. In either case, a separation of the carbon atoms would be the result, and we would obtain products containing a smaller number of carbon atoms than the original compound contained. This is exactly what takes place when the carbylamines are decomposed. When treated with hydrochloric acid they yield two pro- ducts ; one of these is formic acid, a compound contain- ing one atom of carbon ; the other consists of the hydro- carbon residue of the original. compound combined with the nitrogen atom and hydrogen. Thus, in the case of ethylcarbylamine, the decomposition may be represented as follows : C,H 5 ) CM.X^C + 2HTO = H ( N + H.COOH H\ Ethylcarbylamiue. Ethylamine. Formic acid. ( C,H 5 The fact that the compound N < II , in which the (H nitrogen atom is evidently in combination with the hydro- carbon residue, is so readily formed, leads us to the con- clusion that in the original compound the same kind of union existed. The fact, also, that the one carbon atom is given off so readily from the molecule, indicates clearly that it was held in combination in some manner different from that in which the other carbon atoms of the mole- cule are held in combination. Taking the two facts and conclusions together, we are led to the formula above assigned to the carbylamine group, viz., C^N as the correct one. 154 CHEMICAL COMPOUNDS. Constitution of the Group SO. A H. By the action of concentrated* sulphuric acid upon hydrocarbons and vari- ous other compounds containing hydrogen, derivatives are obtained which differ from the original compounds in containing the group SO S H in the place of hydrogen. The reaction consists in the formation of water and the new derivative, thus : /OH /C 6 H 5 o 6 H 9 + so / OH SO< OH + n.o . Benzene. Sulphobenzenic acid. These products all act like acids in every way, so that we are justified in assuming the presence of hydroxyl in them. As they are formed so readity from sulphuric acid, it is also fair to assume that the group S0. 2 OH is a residue of sulphuric acid. Then, if we know the con- stitution of sulphuric acid, the constitution of this group will also be known to us. The fact that this group is a residue of sulphuric acid is shown also in the following way : By replacing one of the hydroxyl groups of sul- phuric acid by an atom of chlorine, we obtain a com- /Cl pound of the formula SO./ , which, by simple treatment with water, is reconverted into sulphuric acid. There can be no doubt that the group SO 2 OH of this chloride has exactly the same constitution as the cor- responding group of the acid. But, if this chloride be allowed to act upon benzene, sulphobenzenic acid and hydrochloric acid are the products, the former having all the properties possessed by the sulphobenzenic acid ob- tained from the action of sulphuric acid upon benzene. The reaction takes place thus : Cl C 6 H 6 -f S0 2 < = C 6 H 5 .S0. 2 .OH + HC1. Here, evidently, the group SO^.OH of the chloride takes the place of an atom of hydrogen in benzene. Assuming, then, the- general formula SO, .OH for the group, it remains to decide in what manner the atoms of the sub-group S0 2 are united. If both sulphur and oxy- GENERAL CONSIDERATIONS. 155 gen act as bivalent elements, two possibilities present themselves. We may have either O J3 or S '0 . If the former expresses the constitution of the group, then it is plain that the hydrocarbon must be united with it through the instrumentality of oxygen, whereas, if the latter is the correct expression, the hydro- carbon residue may be held in combination, either through the instrumentality of oxygen or sulphur. For, we may have either 1, C 6 H 5 S OH ; or,' 2, C 6 H, S OH. Inasmuch as by reduction sulphyd rates are obtained from the sulpho acids, it is believed that, in the latter, the grouping of the atoms is similar to that in the second of the above formulas. The sulphydrates have the gene- ral formula R(SH), in which R represents a hydrocarbon residue. This residue is united directly with the sulphur, and this in turn with the hydrogen. The proof of this constitution of the sulphydrates has been considered under the head of hydroxides. Whenever then the group S OH enters into a compound in the place of a hydrogen atom, the resulting compound is a true sulpho acid. Again, when a chloride, bromide, or iodide of a hydro- carbon or other residue is treated with an aqueous solu- tion of a neutral sulphite, a derivative is obtained which has the same general composition as the sulpho acid, as may be seen in the following equation: /OK C 2 H 5 I + S0< == C a H 6 .S0 2 .OK + KI . M)K Ethyl iodide. Potassium sulphite. Potassium ethylsulphite. These compounds were supposed to be identical with the sulpho acids. Indeed, in a few cases, it seems that the compounds obtained by this latter method must have the constitution above accepted for true sulpho acids. Nevertheless, it has recently been shown that the com- pounds obtained from the sulphites differ from true sulpho acids; that the former are really ethers of sul- phurous acid, and not substitution products, in the sense in which we have here used that expression. The follow- ing experiments furnish the proofs : 156 CHEMICAL COMPOUNDS. When benzylchloride, C f H 5 .CH. 2 Cl, is treated with potassium sulphite, reaction ensues in the manner above indicated, viz.: C 6 H 5 .CH 2 C1 + K. 2 S0 3 = C 6 H 5 .CH 2 .S0 3 K -f KC1. When the new salt is treated with phosphorus chloride, the products of the reaction are phosphorus oxychloride, POC1 S , thionylchloride, SOCl z , and benzylcliloride, C 6 H 5 .CH,C1. If the salt had been derived from a true snlpho acid, a sulpho chloride of the formula C B H ft .CH,.S0 2 Cl would have been obtained. It is hence concluded that, in the salt, the sulphur does not hold the rest of the group in combination with the h3 T drocarbon residue, but that this is accomplished by means of an oxygen atom the group having the constitution ex- pressed by the formula S O H; or by S H. The formation of the above products can then readily be explained : C 6 H 6 .CH 2 .O.S.O OK yields C 6 H 5 .CH,C1 Cl S Cl C1K. The oxygen abstracted will, of course, form phosphorus ox3 7 chloride. If sulphur had occupied either of the posi- tions occupied by the oxygen atoms in the above formula which are abstracted, one of the products of the reaction would have been phosphorus sulphochloride, PSCl o , whereas not a trace of this substance could be detected. This subject requires more investigation before the conclusions above drawn can be looked upon as definitely settled. In all probability it will be found that there are, as stated, two isomeric groups, S0 3 H, only one of which can give true sulpho acids. The formula above given for the sulpho-acid group will also probably be found to be the correct one. Further, from experiments thus far made, it seems more than likely that these groups easily undergo transformation, their atoms being rearranged. Constitution of the ^Group NO., When concentrated nitric acid is allowed to act upon h} r drocarbons, etc., GENERAL CONSIDERATIONS. 157 hydrogen is frequently replaced by the group N0 2 , thus: C 6 H 6 + N0. 2 OH = CeH.-NO, -f H,O.. Benzene. Nitric acid. Nitrobenzene. The reaction, as will be noticed, is similar to that which takes place when sulphuric instead of nitric acid is used. Just as in the former case we can assume that the group S0 3 H is a residue of sulphuric acid, so in the latter case we can assume that the group NO 2 is a residue of nitric acid. The formula which we accept for nitric acid will show us the constitution of the group N0 2 . Again, nitro compounds are formed by treating a chloride, bromide, or iodide of a hydrocarbon residue with silver nitrite, AgNO a , the reaction taking place as follows: C a H 5 I + AgN0 2 = C 2 H 5 (N0 2 ) + AgL Ethyl iodide. Nitroethane. It would appear from the latter reaction that the group N0 2 has the same constitution in the nitro derivatives that it has in nitrons acid ; but this is in reality not the case, or, at least, certain facts seem to indicate clearly a dissimilarity of the groups. There are two series of compounds of the same com- position, but of different constitution, both of which con- tain the group N0 2 . The members of one of these series are ethers of nitrous acid. If nitrous acid contains hydroxyl, then the ethers would have the general con- stitution R NO, in which R represents a hydro- carbon residue. The characterizing feature in the con- stitution of these ethers is the same as that which we find in all ethers, viz., the acid group is combined with the hydrocarbon residue by means of an atom of oxygen. That this is true of the ethers of nitrous acid is shown by the fact that, when nascent hydrogen acts upon them, they yield the alcohols corresponding to the hydrocarbon residues which they contain, and at the same time ammonia. If the nitrogen atom had been directly united with the hydrocarbon residue, we would have found it in combination with this residue after the above reduction. The decomposition which actually takes place may be represented thus: 14 158 CHEMICAL COMPOUNDS. Ether, R O N=0 yields TT R_0 H and H N/" . \H Alcohol. Ammonia. With the constitution assumed for the ether, it is. evident that the formation of alcohol by the addition of hydrogen would necessitate the splitting off of the group containing nitrogen. On the other hand, the second series of compounds are not ethers of any acid, but are true substitution products. They consist of a hydrocarbon residue combined with the group N0 2 by means of the nitrogen atom. Their general constitution is expressed by the formula R N0 a . This conclusion is reached by considering the products of the reduction of nitro compounds. When treated with nascent hydrogen, they yield products known as amine bases, which are ammonia in which one hydrogen atom has been replaced by a hydrocarbon residue. The decomposition is represented as follows : R NO, + 6H = R NH 2 -f 2H 2 0. Nitro products. Aminebase. In the product obtained in this case, it is evident that the nitrogen atom is in direct combination with the hydro- carbon residue, and hence we can assume that this kind of combination also existed in the original nitro compound. Accepting the above formula for nitro compounds, it is difficult to see how they can be formed by reaction with silver nitrite. For, if the hydrocarbon residue took the place occupied by the silver in the salt, it is plain that the product would be an ether which, according to what has already been said, must have the formula R NO. The product, however, is not an ether. Consequently, some other change besides that of an inter- change of places by the silver atom and the hydrocarbon residue must be accomplished at the same time. Conse- quent! j 7 , further, the group N0 2 in nitrous acid has a con- stitution differing from that of the group N0 2 of nitro compounds. As to the respective' constitutions of these two groups we see that in nitrous acid we have in all probability one hydroxyl. This gives us one oxygen atom combined GENERAL CONSIDERATIONS. 159 by one affinity with hydrogen, and, on the other side, with nitrogen, thus : H N . The only thing that is otherwise present in the molecule is an atom of oxygen which, it is safe to suppose, is combined by both its affinities with nitrogen ; whence we have the group N=0 as the characteristic group of the acid and its derivatives. But the group of nitro compounds unites with residues by means of its nitrogen atom, as we have seen. Hence, we can conceive of two formulas /o for the group NO , viz., N/ , in which the nitro- NO ^o en is trivalent, and No, S< , S O . X X X /o As is plain, all the formulas except S<^ | and /0\ S0 , though the reasons for accepting them are not very good ones. The acids of sulphur are the following: H. 2 SO 3 , sulphurous acid. H 2 S0 4 , sulphuric acid. H a S 2 7 , pyrosulphuric acid. HaS 2 O 3 , hyposulphurous or thiosulphuric acid. H 2 S 3 O 6 , hyposulphuric or dithionic acid. H 8 S 3 6 , trithionic acid. H 2 S 4 O 6 , tetrathionic acid. H 2 S 5 6 , pentathionic acid. CONSTITUTION OF CHEMICAL COMPOUNDS. 165 Sulphurous Acid, H. 2 S0. 6 . Only derivatives of this acid are known as the salts and ethers. From a study of these derivatives, conclusions have been drawn con- cerning the constitution of the acid itself. The consti- tution of sulphurous acid is expressed by very different formulas by those who consider- sulphur as bivalent, and those who consider that it may and does 'act as quadri- valent or sexivalent. Up to the present, no conclusive evidence of either view has been furnished. We propose to consider sulphur as bivalent, and to show to what conclusions we shall then be led. This we can do without interfering with our main object, which is to show more particularly the methods of thought and experiment that lie at the foundation of constitutional formulas in general. If then sulphur is bivalent, we have as probable for- OH OH mulas for sulphurous acid SS . SH SO., so. OH \OH 2 mol. Thiosulphuric acid. Tritbioaic acid. CONSTITUTION OF CHEMICAL COMPOUNDS. 169 The latter view is rendered probable by the fact that trithionates are formed when double salts of thiosul- phuric acid are boiled with water. The reaction takes place according to the following equation : 2AgKS 8 0, Ag. 2 S + K 2 S 8 6 - /OK 1 S0 2 < /OK \3Ag I S0 2 < SO -SAg OK J When potassium trithionate is boiled with potassium sulphide, potassium thiosulphate is formed K. 2 S 3 6 + K 2 S 2K a SA- This reaction also indicates an intimate relation between thiosulphuric and trithionic acids. Tetrathionic Acid, 1T. 2 S 4 6 , is obtained b}' the action of iodine on sodium thiosulphate, thus: NaS0 2 mol. Sodiuin Sodium tefrathionate. tl)iosulohat( /ONa SO/ S0 2 <_ ,SNa 2NaI. so. SO. ONaj x ONa Hence, the formula of the acid from which the latter salt is derived is accepted for tetrathionic acid. Pentathionic Acid, H 2 S 5 O ( ., is obtained by treating barium thiosulphate with sulphur bichloride S 2 C1 2 . Tlie latter compound has the constitution Cl S S Ci ; consequently, the reaction may be interpreted most readily as follows: " 15 170 CHEMICAL COMPOUNDS. /OH 1 /OH S0 2 < SO/ '\3H N3-S V + 01 S S-C1 = I + 2HC1 . X SH I /S S SQ, S0 2 mol. Thiosul- Pentatbionic acid. phuric acid. The corresponding acids of selenium and tellurium, as far as the}' are knowji, are represented by similar formulas. Compounds of Nitrogen with Oxygen, and with Oxygen and Hydrogen. Nitrogen forms with oxygen the follow- ing compounds: N 2 0, nitrons oxide or nitrogen monoxide. NO or N 2 2 , nitrogen dioxide. N 8 8 , nitrogen trioxide. N0 2 or N 2 4 , nitrogen tetroxide. N 2 S , nitrogen pentoxide. The constitution of nitrogen monoxide is usually ex- pressed by the formula O Nitrogen dioxide is N=0, and is, therefore, unsatu- rated. The readiness with which it combines with oxygen and chlorine indicates its unsaturated condition. Nitrogen tetroxide is N=O, and is also unsatu- rated. This formula is given the compound because it is obtained so easily from the monoxide, and is converted into the latter so easily. Nitrogen trioxide is the anhydride of nitrous acid. Hence its formula depends upon that of nitrous acid (which see). Nitrogen pentoxide is the anhydride of nitric acid. Its formula depends, upon that of nitric acid (which see). CONSTITUTION OF CHEMICAL COMPOUNDS. 17i There are two acids of nitrogen, viz.: HNO.^, nitrous acid. HN0 3 , nitric acid. Nitrous Acid, HNO,. The proofs for the presence of the group N in nitrous acid have been given under the head of nitro compounds (which see). The acid has the constitution H N=0. From this formula we derive that of the anhydride, nitrogen tri- oxide, thus : I HO = \0 . 0=N H ) 0=N/ Two mol. Nitrous acid. Nitrogen trioxide. Nitric Acid, HN0 3 It has been shown above (see Nitro Compounds) that the group contained in nitro /O compounds has probably the constitution N<^ | . X O It being further probable that this group is contained in nitric acid, the constitution of the latter would be H N<^ | . The fact, that nitric acid is so volatile, X seems to indicate that the nitrogen contained in it is trivnlent rather than quinquivalent. We saw above other facts indicating the same thing, so that now the above formula is pretty generally accepted. Hydroxylamine, H.^NO. This substance is a strong base conducting itself like ammonia. By appropriate reactions it can be shown that two of the hydrogen atoms contained in the compound differ from the other one. These facts prove that hydroxylamine is a derivative of /H ammonia, and the formula is consequently N H \O H 172 CHEMICAL COMPOUNDS. Compounds of Phosphorus with Oxygen, and with Oxygen and Hydrogen. Phosphorus forms two oxides, viz.: P-jOy, phosphorus trioxide. ^fiv phosphorus pentoxide. The former is usually considered as derived from trivalent phosphorus, and to it is consequently given the formula /0\ P P . \o/ The latter, however, is considered as derived from quinquivalent phosphorus, and to it is given the formula /--\ l/0\| P P . l\0/l \-0-/ These formulas are purely hypothetical. There are several acids of phosphorus, viz.: H 3 PO 2 , hypophosphorous acid. H 3 PO S , phosphorous acid. H 3 P0 4 , phosphoric acid. H 4 P 2 O 7 , pyrophosphoric acid. HPO,, metaphosphoric acid. Hypophosphorous acid, H 3 PO V , is monobasic, and hence onty one hydroxyl group is assumed as present in its mole- H cule. This gives the formula H 2 PO.OH or P/' , I O OH in which the phosphorus atom is quinquivalent. Phosphorous Acid, H,P0 3 . In regard to the constitu- tion of this acid, two views "are held. According to the /OH first, the formula of the acid is P OH , the phos- \OH CONSTITUTION OP CHEMICAL COMPOUNDS. 173 phorus being trivalent. According to the second, the formula is P^ , the phosphorus being quin- | \OH OH quivalent. If the former formula is correct, the acid ought to be tribasic. In most of its salts, however, it is only bibasic. Still ethers are known which are evidently derived from a tribasic acid, as, for instance, P0 3 (C a H.) 3 ; and it has, further, recently been shown that a salt of the acid exists in which there are three atoms of a monova- lent metal to the molecule. These latter facts would lead to the formula P(OH) 3 . The fact, also, that phos- phorous acid is produced by simply treating phosphorus trichloride with water, is in accordance with this formula. We have Cl H Cl + H Cl H HO /OH HO = Pf OH + 3HC1 HO \OH On the other hand, the following facts speak for the O formula P<^ : | M)H OH When benzene is treated with phosphorus trichloride, under appropriate conditions, the following reaction takes place : PC1 3 + C 6 H 6 = PC1 2 (C 6 H 5 ) + HC1. When the main product, phosphenyl chloride, is treated with water, the chlorine is eliminated, and a compound of the composition P0 2 H 2 (C 6 H 5 ) is formed. The formula of this compound may be either ^ H I/* ,; .'_' ,: 1, POH ; or, 2, 6 H 5 | OH C 6 H 5 15* 174 CHEMICAL COMPOUNDS. If the latter is the formula, then we can conclude that the constitution of phosphorous acid is similar, i.e., P \ | X OH * OH If formula 1 is correct, then by the action of phos- phorus pentachloride upon the compound the following reaction ought to take place : I. P(OH) 3 C 6 H 5 + 2PC1 5 = PC1 2 (C 6 H 5 ) + 2(POC1 3 ) + 2HC1. If, however, formula 2 is correct, then under the same conditions the following reaction would take place: II. OPH(OH)C 6 H 5 + 2PC1 3 = OPC1 2 (C fi H.) -f POC1 PC1 2HC1. In the former case, phosphenyl chloride, PC1 2 (C 6 H 5 ), would be formed ; in the latter, phosphenyl oxychloride, POC1 2 (C 6 H 5 ). Direct experiments showed that phos- phenyl oxychloride, phosphorus ox3'chloride, and phos- phorus trichloride are formed, and consequently the formula OPH (OHjC 6 H 5 is correct; and phosphorous f /H acid by analogy is OPH(OH) 2 'or P< I NDH OH Phosphoric Acid, H 3 P0 4 . This acid is tribasic, and hence three hydroxyl groups are assumed to be present in it. From this follows directly the formula PO(OH) S . The question still remains whether the phosphorus is quinquivalent or trivalent in the acid. In the former -OH case, the formula would be P<( ; in the latter, | >OH OH CONSTITUTION OF CHEMICAL COMPOUNDS. 175 O OH . Phosphoric acid is obtained by treating X OH phosphorus oxychloride with water. If the oxychloride is O II /ci , then we would expect phosphoric acid to Cl have the former of the two formulas given. If the oxy- /O Cl chloride is P: Cl , then the acid has probably \C1 the latter of the two formulas given. The formulas /OH |f Cl P<( and P<( are usually accepted, | M)H I \C1 OH Cl though without positive proofs of their correctness. Pyrophosphoric Acid, H 4 P^O. This is a partial an- hydride of phosphoric acid formed by abstracting one molecule of water from two molecules of the acid, thus: /,O ^OH ^OH \OH ,OH /OH P< >O OH OH H.,0 = \0 P< OH OH \0 2 mol. Phosphoric acid. Pyrophosphoric acid. The constitution is readily understood by the aid of the general remarks on the subject of anhydrides. 176 CHEMICAL COMPOUNDS. Metaphosphoric Acid, HPO. A . This acid has a compo- sition analogous to that of nitric acid, HNO a . It is, like pyrophosphoric acid, a partial anhydride of phosphoric acid, formed by abstracting one molecule of water from one molecule of the acid, thus : H0 P-OH Phosphoric acid. Metaphosphoric acid. Accepting the formula for phosphoric acid which is employed in this equation, the constitution of metaphos- phoric acid would be that which is expressed. Nitric acid, it will be remembered, has probably the constitution /O HO N . The relations between phosphoric acid and its anhy- drides are shown by the following tables: 2(H 3 P0 4 ) H,O == H 4 P 2 7 , pyrophosphoric acid. 2(H,P0 4 ) 2H 2 O = 2HP0 3 , metaphosphoric acid. 2(H 3 POJ 3H 2 = P 2 5 , phosphoric anhydride. P 2 5 -|~ H 2 = 2HP0 3 , metaphosphoric acid. P 2 5 -f 2H 2 = H 4 P 2 O 7 , pyrophosphoric acid. P 2 5 + 3H 2 = 2H 3 i> 4j phosphoric acid. Arsenic, antimony, and bismuth form some compounds analogous to those of phosphorus here described. What has been said of the constitution of the latter holds good of the constitution of the former. Compounds of Boron with Oxygen and with Oxygen and Hydrogen. Boron forms only one oxide, viz., B 2 3 , known as boron trioxide. The acid to which it cor- responds is B(OH) b . When boric acid, 6(011)^, is CONSTITUTION OF CHEMICAL COMPOUNDS. 177 heated for some time at 100 it loses a molecule of water, a partial anhydride being formed, thus: /OH ^0 B OH H0 B OH . When this anhydride is heated to a much higher tem- perature, it is converted into B 2 3 , thus : B B OH ,0 B B/ ^O The formulas of these anhydrides here given are purely hypothetical. Compounds of Silicon with Oxygen and with Oxygen and Hydrogen. Silicon bears a close analogy to carbon in some respects. It usually acts as a quadrivalent ele- ment, as is seen in the compound SC1 4 . When this chloride is treated with water, we should expect as a product Si(OH) 4 , which may be considered as the normal acid of silicon. It appears to be possible to obtain this acid, but it is very unstable. It loses water easily, and thus yields a partial anhydride, SiO H H 2 , thus: Si(OH) 4 H 2 = SiO a H 2 = SiO(OH) a . This compound is usually called silicic acid, as from it are derived most of the silicates. If heated, this acid yields complicated polysilicic acids, which, in their turn, are partial anhydrides. They are formed by the union of two or more molecules of silicic acid and the abstraction of varying amounts of water from them. Examples of such polysilicic acids are Si 2 3 (OH) l2 , Si 3 4 (OH) 4 , Si 3 5 (OH) 2 , Si 4 7 (OH) 2 , etc., some of which are found in nature, as opal, hydrophane, etc. As final product of the action of heat on silicic acid, we have silicon dioxide, Si0 2 . 178 CHEMICAL COMPOUNDS. Salts. The constitution of the most important acids being thus understood, that of the salts, in general, requires no special consideration ; for we have seen that the salts are very simple derivatives of the acids. There are a few metals and groups, however, which have the property of yielding peculiar salts, and these require a brief consideration. Ammonium Salts. When ammonia, NH 3 , acts upon any acid, a salt is formed by direct addition, thus: NH 3 + HCl = (NHJC1 Ammonium chloride. NH 3 + HN0 3 = (NH 4 )N0 3 Ammonium nitrate. 2NH 3 + H 2 S0 4 = (NH 4 \S0 4 . Ammonium sulphate. The salts thus formed are similar to the salts of potas- sium, sodium, etc., KC1, KN0 3 K ;2 SO 4 , etc. They con- duct themselves like true metallic salts. Hence, the group NH 4 , which is contained in them, is supposed to play the part of a metal, and to it the name ammonium is given. Accordingly, the salts are called ammonium salts. These have been referred to incidentally under the head of valence. It was shown that in them the nitrogen is quinquivalent. The formulas of the above salts are, accordingly H , N H , etc. Salts of Copper and Mercury. Copper and mercury form two series of salts, of which the following are examples: Hg,Cl a , HgCl, Hg a (NO s ) a , Hg(N0 8 ), Hg,(S0 4 ), Hg(S0 4 ). If we determine tlie formulas of the two chlorides of mercury by the aid of the specific gravity of their vapors, Cu 2 Cl y , CuCl, Cu 2 (N0 3 X,Cu(NO s \ Cu,(S0 4 ), CONSTITUTION OF CHEMICAL COMPOUNDS. 179 we are led to HgCl and HgCl. 2 . According to these for- mulas, mercury is bivalent and the compound HgCl is unsaturated. It has been supposed, however, that the formula of the chloride HgCl in the solid condition is in HgCl reality Hg.CL, and that it has the constitution HgCl Hg The group is bivalent as well as the mercury atom Hg itself, and thus the above two series of salts are explained. The same explanation is given for the corresponding salts of copper, A large number of compounds are known which are derived from salts of ammonium and contain copper and mercury. They seem to consist of ammonium salts, in which a portion of the hydrogen of the ammonium groups has been replaced by copper or mercury, thus : 01 4_ 9^H s He- C1 Dimercury diamido- -- *, == -Ug 3C1 , chloride. N0 3 .NH 2 Hg^, Dimercuryamine nitrate. These formulas are purely hypothetical. Similar compounds are formed with other metals, par- ticularly with cobalt, which furnishes a very large number of interesting substances of this kind. These are too complicated and too little understood to permit the draw- ing of positive conclusions concerning their constitution. Their study promises important results. Salts of Tron and Chromium. Iron and chromium form two series of salts, as follows : FeCl a , FeiNOj., FeS0 4 , Fed,, Fe(N0 3 \, Fe.(SO A . or Fe 2 Cl,. or Fe.XN0 3 ) c . CrCl 2 , Cr(NO,),, CrS0 4 , CrCl,, Cr N0 8 ) s , Cr/S0 4 ) 8 . or Cr. 2 Cl fi . or Cr.(N0 3 ) 6 . In regard to the formula of the second chloride of iron above represented by FeCl^ or Fe a Cl , there is still doubt. 180 CHEMICAL COMPOUNDS. A determination of the specific gravity of its vapor led to the formula Fe^Cl,., whereas a determination of the specific gravity of certain other derivatives of the same constitution as the chloride led to the simpler formula. If FeCl, is correct, the iron atom in the compounds cor- responding to this chloride is trivalent, whereas, if Fe 2 CL is correct, the iron atom is probably quadrivalent, and Cl Cl the constitution of the compound is Cl Fe Fe Cl . A J. Of course, the formulas for the salts of iron will depend upon those which we accept for the two chlorides. It is most commonly considered that iron may be bivalent or quadrivalent. The salts of manganese resemble those of iron and chromium. Salts of Aluminium. The determinations which leave us in doubt concerning the formula of the higher chloride of iron also leave us in doubt concerning the formula of the chloride of aluminium. It is either A1C1 { or Al 2 Cl b , and aluminium is either trivalent or quadrivalent. Metal Acids. The four metals, iron, chromium, man- ganese, and aluminium, form hydroxides of the general formula MO. OH, which conduct themselves like weak acids, forming salts with some metals. Thus, we have A10.0K and AlO.ONa, salts of the hydroxide A10.0H.* Iron, manganese, and chromium yield acids of the general formula M0 4 H,. Thus we have FeO 4 H,, MnO 4 H 2 , and Cr0 4 H 2 . These acids are analogous to sulphuric acid H.,S0 4 , and a close resemblance is noticed between the salts of sulphuric acid and those of chromic acid, which is the best known of the three above-named acids. * Of course, if the* chloride is A1.,C1 6 , this compound is ,, and the salts have a corresponding composition, CONSTITUTION OF CHEMICAL COMPOUNDS. 181 If the metals are quadrivalent in these acids, their con- stitution may be expressed by the general formula / M< | I \Q . If they are biyalent, their constitution OH OH would be similar to that of sulphuric acid. As far as the evidence thus far in our possession is concerned, we are as much justified in accepting one of these formulas as the other. A very important salt of chromium is that known as potassium bichromate. The formula of this salt is Cr. 2 O.K 2 . It may be regarded as the salt of an acid which is analogous to pyrosulphuric acid and derived from chromic acid by the abstraction of water, thus : CrO OH CrO a < ,OH .W H 2 = CrO OH OHj 2 mol. Chromic acid. Pyrochromic acid. Neither chromic acid itself nor pyrochromic acid can he prepared in the free condition. The group CrO, 2 does not appear to be capable of holding hydroxyl in combi- nation. So that salts of the formula Cr0 2 OM are not known. An acid of manganese furnishes salts of the general formula Mn0 4 M. No positive assertion can be made in regard to the constitution of this acid, except that it is monobasic, and hence it probably contains one hydroxyl group. This gives the formula Mn0 3 OH, but the group Mn0 3 remains unexplained. Compounds of Uranium. In connection with the sub- ject of bases it was mentioned that uranium forms a peculiar set of salts in which the monovalent group UO takes the place of the hydrogen of the acids. This group UO is, like U itself, a base residue, and hence, 16 182 CHEMICAL COMPOUNDS. according to the general definition of salts, it may take the place of hydrogen in the acids, and the resulting compounds will be just as strictly salts as those com- pounds in which the base residue consists of a metal alone. Uranium, further, forms salts of the general formula U 7 M 2 , which may be supposed to be derived from a complex hydroxide, U 4 O.H 2 = OU.O.U(OH).O.UO. This formula is, however, only hypothetical. CONSTITUTION OF CARBON COMPOUNDS. As has already been intimated, a great deal more is known concerning the constitution of carbon compounds than is known concerning the constitution of those com- pounds which do not contain carbon. Having considered the general constitution of the classes of compounds with which we meet, it only remains to study those changes which the members of the different classes can undergo without losing their main characteristics. We shall find that the compounds of carbon may be divided into a few distinct groups ; that each of these groups possesses a mother-substance from which all the other members of the group may be derived. The principal groups are : the Marsh-gas, or Methane compounds, also called Fatty Bodies ; the Benzene compounds, also called Aromatic Bodies ; the Naphthalene compounds ; and the Anthracene compounds. The first two groups comprise by far the largest number of carbon compounds. , METHANE DERIVATIVES. (FATTY BODIES.) First Group. Bodies derived from the Hydrocarbons C n H2n-\-2. The constitution of methane has been considered above (see ante, p. 125). It was also shown that by the linking of carbon atoms to each other the possibility is given for CONSTITUTION OP CHEMICAL COMPOUNDS. 183 the formation of an homologous series, the members of which differ from each other by CH >2 , or a multiple of this. The following members of this series have been particularly well studied. Methane, CH,. Pentane, C 5 H 12 . Ethane, C 2 H 6 . Hexane, C 6 H 14 . Propane, C s H b . Heptane, C 7 H 16 . Butane, C 4 H 10 . In speaking of substitution products it was stated that only one mono-substitution product of methane could exist, according to the views now held concerning consti- tution. The same thing is true of other substitution products in which more than one substituting group is present. Further, we can only conceive of one variety of methane itself, and only one variety has ever been observed. Derivatives of Ethane, C> 2 H 6 . Only one variety of this hydrocarbon can exist, and only one variety has been observed. Of its mono-substitution products also, only one variety can exist, and only one variety has been observed. Of the bi-substitution products, however, two varieties are possible, as may be seen by comparing the following formulas : H H H X I I II X C C X and H C C X . H H H H In the first, the substituting groups are in combination with different carbon atoms ; in the second, both substi- tuting groups are in combination with the same carbon atom. A number of compounds are known belonging to the classes of which these are the general formulas. X may represent any of the substituting groups with which we have had to deal; or the class groups CH 2 OH, COH, COOH, etc. 184 CHEMICAL COMPOUNDS. The simplest of these are the chlorides, one of which is CHC1 2 .CH H , and the other CH se Cl.CH 2 Cl. The first is called ethylidene chloride, the second, ethylene chlo- ride. The constitution of these compounds follows from the following facts : Ethylidene chloride is produced by the action of phos- phorus pentachloride on aldehyde. We have seen that / H an aldehyde contains the group C=0 ' Ordinary / H aldehyde is CH, C=0 ' As, in the reaction with phosphorus chloride, the oxygen is simply replaced by chlorine, the constitution CH 3 CHC1 2 follows for ethyl- idene chloride. As a consequence, the formula CH a Cl.CH 2 Cl must be that of ethylene chloride. Other compounds closely related to these two chlorides will be considered under the heads of ethylene, bibasic acids, etc. Derivatives of Propane, C. A H k . Propane may be con- sidered as a mono-substitution product of ethane, derived from the latter by replacing an atom of hydrogen with CH 3 . From what was said above, it will, hence, be seen that only one variety of propane can exist; only one variety has been observed. Under the head of substitution products, it has been .shown that there are two kinds of carbon atoms and, consequently, two kinds of hydrogen atoms in propane (which see) ; and hence, further, that two different mono- substitution products may be obtained from this hydro- carbon. These have the general formulas : H H H H X H X C C C H and H C C C H . H H H H H H The compounds represented by the first formula are known as propyl compounds; those represented by the second formula as isopropyl or pseudopropyl compounds. CONSTITUTION OF CHEMICAL COMPOUNDS. 185 The two alcohols, normal propyl alcohol CH 3 .CH.,.CH 2 OH, and pseudopropyl alcohol CPTj.CHOH.CHg, are the starting-points for the prepara- tion of the two series of isomeric propyl compounds. As the former is a primary alcohol, it follows, from what has been said concerning these alcohols, that it must con- tain the group CH 2 OH. This it can only do if the hydroxyl group is in combination with one of the ter- minal carbon atoms. Consequently, the above constitu- tion is assigned to it. By replacing the hydroxyl by chlorine, bromine, iodine, C3 r anogen, etc., corresponding derivatives are obtained. Pseudopropyl alcohol is obtained from acetone and, being a secondary alcohol, contains the group CH.OH. Consequently, its hydroxyl is in combination with the central carbon atom of propane. By replacing the hy- droxyl with chlorine, bromine, iodine, cyanogen, etc., cor- responding pseudopropyl derivatives are obtained. Derivatives of Butane, G 4 H 10 . Butane may be con- sidered as a mono-substitution product of propane, con- sequently, two varieties must be possible, one of which would have the formula H H H H I. H C C C C H ; I I I I H H H H while the other would have the formula PI H H I I I II. H C C C H H HH As a matter of fact, two varieties of butane are known to us, viz., normal butane and trimethylmethane. The former has the constitution represented by formula I. above ; the latter that represented bv formula II. 10* 186 CHEMICAL COMPOUNDS. Proofs. The proof of the formula of normal butane is the same in nature as that given for ethane (see p. 127). It is formed by the action of zinc or sodium on eth}-l iodide, according to the equation : HH HH HHHH H-C-C I + 1 C-C- H + Zn = H C-C-C C-H + ZnI 2 . HH HH HHHH lodethane. Normal butane. Of course, we here assume that we know the formula of iodethane, but we have already presented good grounds for this assumption. Starting with this formula, we are led very easily to the above formula of normal butane. Trimethylmethane is obtained from psendobutyl iodide, the constitution of which is known to be CH 3\ J>CI CH 3 . When the iodine is replaced by hy- drogen, the hydrocarbon is the product. (See Pseudo- butyl Alcohol.) Of normal butane, two kinds of simple substitution products are possible, of the general formulas: H H H H H X H III I I I I I. H C C C C X and II. H C C C C H I I I I I I I I HHHH HHHH Of trimethylformene, there are also two kinds possible, of the formulas H H H H X H III III III. X C C C H and IT. H C C C H . Ill II H C H H C /1\ / HH H HI l\ HH Representatives of all four kinds of substitution pro- ducts are known. The principal of these are the alcohols. 1. Normal butyl alcohol, CH,CH 9 .CH 2 .CH 2 OH. 2. Secondary butyl alcohol, CH 3 CH.OH.CH^CH,. CONSTITUTION OF CHEMICAL COMPOUNDS. 187 GEL 3. Isobutyl alcohol, >CH.CH 2 .OH. CH 4. Tertiary butyl alcohol, CI^.C.OH/ * From each of these alcohols the corresponding chlo- rides, bromides, etc., can easily be obtained. Proofs. Normal butyl alcohol is obtained indirectly from normal butyric acid, the constitution of which is known. Secondary butyl alcohol is converted by oxidation into ethyl-methyl-acetone. C 2 H 5 CO CH. V It is, hence, a secondary alcohol, and its constitution is that expressed above. Isolwtyl alcohol is converted into isobutyric acid by oxidation. The constitution of the acid is known, and hence also that of the alcohol. Tertiary butyl alcohol is a tertiary alcohol, and hence contains the group COH. It is prepared by treating acetyl chloride, CH 3 .COC1, with zinc methyl, Zn(CH 3 ) a , and hence contains three groups CH.. The only for- mula which is in accordance with these facts is that above assigned to the alcohol. Derivatives of Pentane, C^H^. Three varieties of pentane may exist. These have the formulas : 1. H.C.CH^.CH^.CH^CH,. CH 3 2. H 3 C.CH 2 .CH<; 3. 0- All three of these compounds are known. The first is normal pentane ; the second is ethyldimethyl methane; and the third tetramethyl methane. Proofs. Normal pentane is obtained by replacing the CN group of the cyanide of normal butane by hydrogen. 188 CHEMICAL COMPOUNDS. We have seen above how the formula of the cyanide itself is determined. Ethyldi methyl methane is derived from ordinary amyl alcohol, and hence has the same general constitution. The proofs for the constitution of this alcohol will be given below. Tetramethyl methane is derived from the iodide of tertiary butyl alcohol by the action of zinc methyl. The reaction takes place as follows : CH 3 .CI< \CH 3 /CH 3 r n/ CH 3 jll \ X CH 3 CH..CI/ Iodide of tertiary butyl alcohol. C Tetramethyl methane. A great variety of substitution products can he ob- tained from the isomeric butanes. Of the alcohols five are known, as follows: 1. Normal amyl alcohol, CH 3 .CH .CH a .CH 2 .CH a OH. 2. Amyl alcohol of fermentation, >CH.CELCH 2 OH. CH/ 3. Isoamyl alcohol, CH .CH 2 .CH 2 .CHOH.CH 3 . 4. Amylene hydrate, >CH.CH.OH.CH 3 . CH . 5. Tertiary amyl alcohol, >C.OH.CH 2 .CH 3 . CH/ These alcohols, like the others which have been con- sidered, form the starting-points for the preparation of corresponding substitution products. Proofs. Normal anv^l alcohol is obtained from normal valeric acid, and yields this acid by oxidation. The con- stitution of the acid follows from its method of prepara- tion. (See Normal Valeric Acid.) Amyl alcohol of fermentation is obtained from ordinary CONSTITUTION OF CHEMICAL COMPOUNDS. 189 valeric acid by reduction, and is converted into this acid by oxidation. Hence its constitution is similar to that of the acid (which see). Isoamyl alcohol is a secondary alcohol obtained from methyl propylketone. The constitution of the latter being given, that of the alcohol follows. Amylene hydrate is a secondary alcohol, the con- stitution of which is believed to be represented by the above formula. Good proofs are lacking. Tertiary amyl alcohol is formed from propionyl chlo- ride and zinc methyl, and, being tertiary, must contain the group COH. The only formula which is in harmony with these facts is that given above. Derivatives of Hexane, C 6 H U . Three varieties of hexane are known. These are 1. Normal hexane, H 3 C.CH 2 .CH 2 .CH 2 .CH 2 .CH 3 . ' CH. 2. Ethvl isobutane, H 3 C.CH 2 .CH 2 .CH H 3 C X CH 3 3. Diisopropane, >CH.CH<; H 3 CK \CH 3 Proofs. Proofs for the first formula are wanting. The constitution of ethyl isobutane follows from its method of preparation, which consists in treating a mixture of iodethane and iodisobutane with sodium. lodisobutane, being obtained from isobutyl alcohol, CH has the constitution ;>CH.CH I . Hence the rc- CH/ action may be represented thus: )CH.CH a I + ICH,.CH 3 -f 2Na = CH lodidobutane. Iodethane. CH 3X CH.CH 2 .CK,CH 3 -f 2N CH Etliyl isobutane. 190 CHEMICAL COMPOUNDS. Diisopropane is obtained by treating iodisopropane with sodium. This iodide, being derived from isopropyl H 3 C X alcohol, has the constitution /CHI . Hence the H i G / reaction is H 3 <\ /CH 3 >CHI -f ICH< + 2Na = H.CK X CH 3 H 3 C X /CH 3 CH HC Five alcohols are known which are derived from these varieties of hexane. They are : 1. Primary hexyl alcohol, H 3 C.CH 2 .CH 2 .CH 2 .CH 2 .CH 2 OH. 2. Secondary hexyl alcohol, H 3 C.CH 2 .CH 2 .CH 2 .CH.OH.CH 3 . H 3 (\ 3. Dimethyl propyl carbinol, )C.OH.CH 2 .CH 2 .CH 3 . TT f\^ H 3 C.CR, \, 4. Diethyl methyl carbinol, >C.OH.CH 3 . H 3 C.CH./ CH 3 CH 3 5. Dimethyl pseudopropyl carbinol, ;C.OH.CH; Proofs. Good proofs for the first formula are wanting. Secondary hexyl alcohol yields by partial oxidation methyl butyl ketone. It is consequently a secondary alcohol, and contains the groups CH 3 and CH 3 .CH 2 .CH 2 .CH 2 . This gives the above formula. Dimethyl propyl carbinol is obtained from butyryl chloride and zinc methyl. Diethyl methyl carbinol is obtained from acetyl chlo- ride and zinc ethyl. Dimethyl pseudopropyl carbinol is obtained from isobutyryl chloride and zinc metlryl. Derivatives of Heptane, C.H^. Three varieties of heptane are known, one of which is probably normal heptane. CONSTITUTION OF CHEMICAL COMPOUNDS. 191 A second varietj', ethyl-a*rnyl, /CH, H 3 C.CH 2 .CH 2 .CH 2 .CH< , X CH 3 is obtained by the action of sodium on a mixture of ethyl and amyl iodides, the latter from ordinary amyl alcohol. Ordinary amyl alcohol is >CH.CEL.CEL.OH . Consequently the iodide is CH/ . ^>CH.CH 2 .CH 2 I . The reaction is represented by / the following equation : CEU >CH.CH 2 .CH 2 I + ICH a .CH 8 + Na 2 = CH/ )CH.CH 2 .CH 2 .CH..CH 3 + 2NaI CH A third variety of hexane is dimetliyldietliylmethane, H 3 C X X CH 2 .CH 3 ^>C^ . This is obtained from zinc- H 3 CK \CH 2 .CH 3 ethyl and acetone chloride, thus : CH 3 C,H 5X _ I C 2 H 5X /CH, >|Zn + ClJC = >C< +ZnCL . C 2 H/ ^ | C 2 H/ \CH 3 CH 3 Zinc-ethyl. Acetone Dimetbyldiethyl- chloride. methane. Acetone chloride, being produced by the replacement of the oxygen of acetone with chlorine, must have the above constitution. Some of the alcohols corresponding to these hydro- carbons are known. The constitution of these is readily understood so soon as we know the methods of their formation. 192 CHEMICAL COMPOUNDS. Not much is known concerning the constitution of the remaining hydrocarbons of the methane series or their derivatives. Monobasic, Monatomic Acids, C^H^ n O v The acids of this series may be considered as substitution products of the hydrocarbons formed by replacing a hydrogen atom of the latter with carboxyl (COOH). In most cases, these acids have been prepared by converting the group CN of the cyanides of hydrocarbon residues into COOH. If we then know the constitution of the cyanide, the con- stitution of the acid is readily deduced. The principal members of the series are Formic acid, H.COOH. Acetic acid, CH 3 .COOH. Propionic acid, C 2 H 5 .COOH. Butyric acid, C 3 H r COOH. Valeric acid, C 4 H 9 .COOH. Caproic acid, C 5 H U .COOH. Of formic acid and its substitution products only one variety is known. Of acetic acid and its substitution products, also, only one variety is known. Propionic Acid With propionic acid the case is different. Of the acid itself only one variety is known, but of the mono-substitution products two varieties are known. The constitution of the acid is H H H C C COOH . Now it is plain that, in this I I H H compound, aside from the hydrogen of the carboxyl group, there are two kinds of hydrogen atoms those combined with a carbon atom which in its turn is in combination with the group CH 2 ; and those in combina- tion with a carbon a.tom which in its turn is in combina- tion with two carbon atoms. The case is similar to that of propane, of which we saw that two varieties of sub- CONSTITUTION OF CHEMICAL COMPOUNDS 193 stitution products are possible. Here we have the two possibilities expressed by the formulas H X H H II C C CO.OH and X C C COOH . ii L A Products of the first kind are designated as a-substitution products ; those of the second kind as 3-substitution products. The best representatives of these two classes of compounds are the two lactic acids (which see). Lactic acids are derived from propionic acid by replacing a hydrogen atom of the latter with hydroxyl. One of the two lactic acids is obtained indirectly from ethylene, which will be shown to have the formula CH 2 CH 2 | or || .To this body hypochlorous acid may CH 2 CH 2 be added, and the constitution of the resulting compound CH 2 .OH is | . From this, by treatment with potassium CH r Cl cyanide, KCN, is obtained the corresponding cyanide, which, when boiled with alkalies, yields lactic acid of the H II I I constitution HO C C CO.OH. Now, by replacing i the hydroxyl group of this acid with chlorine, bromine, etc., |3-mono-substitution products are formed. The isomeric compounds of the a-series are those which can- not be prepared in the manner described, and are of the same composition as those obtained from 3-lactic acid. Butyric Acids. Two acids of the formula C 3 H 7 .COOH are theoretically possible, and two are known. These are normal butyric acid, CH 3 .CH 2 .CH 2 .COOH, and CH 3X isobutyric acid, >CH.COOH . OH/ It 194 CHEMICAL COMPOUNDS. Proof*. Normal butyric acid is prepared by intro- ducing the group C 2 H 5 into acetic acid, thus : CH 2 NaCOOH -f C 2 H 5 I = C 2 H 5 .CH 2 .COOH + Nal. Further, by reduction, normal butyric acid yields one of the two possible primary butyl alcohols. It was shown (p. 18t) that the other possible primary butyl alcohol is not a derivative of normal butane; conse- quently normal butyric acid must be derived from normal butane, and it has the formula above assigned to it. Isobutyric acid is obtained from pseudopropyl c} r anide (p. 185), and this has been shown to have the constitution H CN H H C C -- C H . I I I H H H From this the above constitution follows for isobutyric acid. Valeric Acids. Two are well known, viz., normal valeric acid, CH H .CH 2 .CH 2 .CH 2 .COOH, and ordinary valeric acid, >CH.CH 2 .COOH . CH/ Proofs. The cyanide from normal butyl alcohol must have the formula CH 3 .CH 2 .CH S9 .CH 2 .CN. * By conversion of the CN group into COO H,, this cyanide yields normal valeric acid. The cyanide from isobutyl alcohol must have the CH 3X formula >CH.CH 2 .CN . This yields ordinary CH/ valeric acid. Caproic Acids __ Four of these acids are known. They are normal caproic acid, CH 3 .CH 2 .CH 2 .CH 2 .CH 2 .COOH; ordinary caproic acid. >CH.CH 2 .CH 2 .COOH CH/ CONSTITUTION OP CHEMICAL COMPOUNDS. 195 CH, CH, isocaproic acid* )>CH.CH<^ ; and CH/ \CO.OH C,H 6 \ pseudocaproic acid, /CH.COOH . Proofs. The cyanide from normal amyl alcohol yields normal caproic acid. Ordinary caproic acid is obtained from the cyanide of ordinary amyl alcohol. Isocaproic acid is prepared from the cyanide corre- sponding to amylene hydrate. Pseudocaproic acid is obtained by introducing two groups, C 2 H., into acetic acid, thus: CHNa..COOH + 2(O a HJ) = P TT 5 \CH.COOH + 2NaI . 0,H/ The other acids of this series are not very well known. By the aid of the foregoing examples, the method of determining the constitution of the known acids will be readily understood. Aldehydes. Corresponding to every primary alcohol and to every acid there is an aldehyde. The constitution of each of these aldehydes is given if we know from which acid or from which alcohol it is obtained. The aldehydes are produced from the primary alcohols by partial oxidation ; and from the acids by subjecting a mixture of a salt of the acid and a salt of formic acid to dry distillation. Acetones or Ketones. The ketones are obtained by distilling mixtures of two acids. If the constitution of the acids is known, that of the ketone obtained in each case is also known. 196 CHEMICAL COMPOUNDS. Second Group. Bodies obtained from the Hydrocarbons C n H 2n . If carbon is always quadrivalent, then the members of this series of hydrocarbons are either nnsaturated, or in them the carbon atoms are united b} 7 " more than one CH,- afflnity each. Thus ethylene, C 2 H 4 , is either , CH- CH 2 an unsaturated compound, or it is , in which the CH 2 two carbon atoms are united by means of two affinities each. Up to the present no proofs have been given for either of these formulas. In regard to these hydrocar- bons, we only know that they easily take up two atoms of monovalent elements. Ethylene and Derivatives. In connection with ethane derivatives it was stated that two chlorides are known, both of which have the formula C 2 H 4 C1 2 . One of these is obtained from aldehyde by replacing the oxygen atom with two chlorine atoms ; hence its formula was assumed to be CHCl a .CH s . The isomeric compound has the CH.C1 ' formula | CH 2 C1 This latter compound is obtained from ethylene by direct addition of chlorine, whence it is concluded that ethylene itself is symmetrical, i. e., that each carbon atom in it holds in combination two hydrogen atoms, CH 3 _ giving the constitution expressed by the formula CH 2 - CH 2 or || OH, A number of products are known corresponding to CH 2 C1 ethylene chloride, ' , among which may be men- CH 2 C1 tioned the bromide, iodide, and cyanide. By replacing the chlorine or bromine of ethylene chloride or bromide CONSTITUTION OF CHEMICAL COMPOUNDS. 197 with hydroxyl, an alcohol is obtained of the constitution CH 2 OH | , which is the simplest representative of the CH 2 OH diatomic alcohols or glycols. Propylene, etc. The remaining hydrocarbons of this series are obtained for the most part by treating the chlorides, bromides, or iodides of the hydrocarbons of the methane series with alcoholic potassa, by which means C1H, BrH orlH is abstracted from the compound. Thus, from C 3 H 7 I we obtain C 3 H 6 : from C 4 H 9 l we obtain C 4 H 8 , etc. Jn many cases the method of formation of the hydro- carbon leads us directly to its constitution. In some cases a doubt exists even after all the methods of forma- tion and the products of decomposition are taken into consideration. Alcohols. Theoretically, a series of alcohols is possible, derived from the hydrocarbons of the ethylene series by the replacement of one hydrogen atom with one hydroxyl group. Only one such alcohol is known. This is allyl alcohol, C 3 H 5 .OH, or CH 2 =CH.CH 2 .OH. Proofs. Allyl alcohol differs from propyl alcohol in containing two hydrogen atoms less. Now, by treating allyl alcohol with nascent hydrogen, it is converted into normal propyl alcohol, which, as we have seen, has the H H H I I I constitution H C C OH . Hence it is as- I I I H H H sumed that in allyl alcohol, as well as in propyl alcohol, the hydroxyl is in combination with one of the terminal carbon atoms, and, accordingly, it must be either CH 2 CH 3 II 1 CH or CH .If the second formula were CH,OH CHOH 17* 198 CHEMICAL COMPOUNDS. correct we should expect allyl alcohol to yield acetic acid by oxidation, inasmuch as it contains the group CH 3 in combination with another carbon atom. Not a trace of acetic acid is formed, however, and hence the first of the two formulas above given is usually accepted. The proofs for this formula are not positive. By replacing the hydroxyl of altyl alcohol with chlo- rine, bromine, iodine, cyanogen, etc., the corresponding chloride, bromide, iodide, and cyanide are obtained. Acids. Though allyl alcohol is primary, it cannot be directly oxidized to form a corresponding acid. But if the alcohol is first combined with bromine and then oxi- dized , a bibrompropionic acid is obtained which, when again freed of bromine, yields acrylic acid. These reactions strengthen the conclusion above drawn, viz., that allyl alcohol contains the group CH^OH. The reactions are CH 2 CH 2 Br II I 1. CH + Br 2 CHBr . CH 2 OH CH 2 OH CH 2 Br CH 2 Br 2. CHBr by oxidation yields CHBr . CH 2 OH COOH CH 2 Br CH 2 3. CHBr -f Zn = CH -f ZnBr, . COOH COOH From these changes we are led to the formula CH 3 CH for acrylic acid. This acid is the first of a series COOH each of which differs from the corresponding member of the series C n 1X2^02 by containing two hydrogen atoms less. CONSTITUTION OF CHEMICAL COMPOUNDS. 199 The member succeeding acrylic acid in this series is crotonic acid, to which the constitutional formula CH 2 =CH CH 2 CO.OH is usually assigned. The grounds for this are as follows : Allvl cyanide, as has been shown, has the formula CH 2 =CH CH 2 .CN. This cyanide, when properly treated, is converted into crotonic acid. When treated with nascent hydrogen, it yields normal butyric acid, which shows that the carboxyl group of crotonic acid is in combination with one of the terminal carbon atoms. Similar considerations lead to a knowledge of the con- stitution of the remaining members of this series. None of these, however, have been investigated as fully as acr3'lic and crotonic acids. Third Group. Bodies derived from the Hydrocarbons C n Hinz. CH 2 Br When ethylene bromide, | , is heated with al- CH 2 Br coholic potassa, two molecules of bydrobromic acid are given off and a compound of the formula C.,H 2 is pro- duced, which is the first of a series of similar hydro- carbons. Just as ethylene must be considered either as unsaturated or as having its carbon atoms combined by the action of more than one affinity of each atom, so also with the hydrocarbon C. 2 H 2 , or acetylene. In the latter case, however, if the compound is unsaturated, each carbon atom must be united by means of three affinities each. The formula of acetylene is, accordingly, =rCH CH | , or HI . The latter formula is usually ac- =CH CH cepted, though the grounds for it are weak. However, whether this treble union exists between the carbon atoms or not, we can be moderately certain .that acety- lene, like eth % ylene, is symmetrically constructed, i. e., that each carbon atom is in combination with one hydro- gen atom. This follows from the fact that acetylene is 200 CHEMICAL COMPOUNDS. formed by abstracting hydrobromic acid from ethylene bromide. For the latter compound has the formula GH,Br , and it appears most probable that the splitting CH. 2 Br off of BrH would take place as follows : CH CH HBr CH CH 2HBr = I or ||| HBr CH CH Acetylene and its homologues have this common pro- perty, they each combine with four atoms of chlorine or bromine, thus forming saturated compounds, which may be regarded as substitution products of the hydrocarbons of the marsh-gas series. Very little is known regarding the constitution of the higher members of the series. No alcohols are known corresponding to the above hydrocarbons. A few acids have been studied, the general formula of which is C w H 2n _ 4 2 , which may be considered as derived from the hydrocarbons of the acetylene series by replacing a hydrogen atom with carboxyl. Their constitution is in no case well known. Fourth Group. Diatomic Alcohols and Acids. It has been stated that when we replace the bromine of ethylene bromide with hydroxyl, an alcohol of the con- CH 2 .OH stitution | is obtained. Alcohols of this kind CH 2 .OH which contain two hydrogen groups are called diatomic alcohols or glycols. The best studied diatomic alcohol is the one the constitutional formula of which is given above. This is ethylene alcohol or ethylglycol. If the above formula is correct, we are justified in expecting that ethylene alcohol will yield two products by oxida- tion. The first would be formed if only one of the groups CH^OH were converted into CO OH. It would CONSTITUTION OF CHEMICAL COMPOUNDS. 201 CH. 2 OH have the constitution | . The second would be COOH formed if both the groups CH 2 OH were converted into COOH COOH. It would be | . Both these products COOH are actually known, and may be obtained by the oxida- tion of ethylene alcohol. They are both representatives of new classes of compounds, the nature of which may be easily understood. CH^OH The compound | , or glycolic acid, is half COOH alcohol and half acid, and it can be shown to possess the properties of both. It is acetic acid in which one hydrogen atom has been replaced by hjTiroxyl, or oxy- acetic acid. The acids of which it is the representative are known as oxyacids. COOH The compound | is an acid, and, from what COOH was said concerning acids in general, it will be recognized as a bibasic acid. It is known as oxalic acid. It is the first of a series of bibasic acids. Diatomic Alcohols, G n H ZnJrZ O v Ethylene alcohol is the only member of this series that is well known. In regard to its constitution enough has been said above to CH 3 OH show upon what grounds the formula rests. CH..OH We may obtain a great variety of derivatives from this alcohol by replacing one or both of its hydroxyl groups- with monovalent elements or groups. The products obtained by the addition of various elements or groups to ethylene may also be considered as derivatives of ethy- lene alcohol. 202 CHEMICAL COMPOUNDS. Diatomic Acids, C n H 2n O t . The acids of which gly- CH,OH colic acid, | , is the simplest known are, as COOH has been said above, half alcohols and half acids, the alcoholic character being imparted to them by the pre- sence of the group OH, in combination with unoxidized carbon, and the acid character by the presence of the group COOH. All that has been said in regard to the alcoholic group OH holds good in regard to that group in these diatomic acids ; and all that has been said in regard to the carboxyl group holds good in regard to that group in these acids. That the above formula for glycolic acid is correct, follows from the methods of its preparation. It is a pro- duct of the partial oxidation of ethylene alcohol CH^OH | , one of the primary alcohol groups being con- CH.OH verted into carboxyl. The fact that glycolic acid itself by CO.OH further oxidation is converted into oxalic acid, | CO.OH proves also that the group CH..OH is present in it. It is further obtained by treating chlor- or bromacetic acid with silver oxide, thus : CH 2 Br CH..OH | + AgOH == | + AgBr . COOH COOH Bromacetic Glycolic acid, acid. Lactic or oxypropionic acid is the succeeding homo- logue of glycolic or oxyacetic acid. As it is a monosub- stitution product of propionic acid, there must be two varieties possible corresponding to a- and /3-chlorpropi- onic acids. One of these would have the formula H OH -C C < H C C CO.OH , and the other, the formula i i CONSTITUTION OF CHEMICAL COMPOUNDS. 203 H H HO C C CO.OH . Both of these acids are known. I I H H The first is ordinary lactic acid, or ethylidenelactic acid. The second is sarcolactic acid, or ethylene lactic acid. Proofs. The proofs of these formulas are the follow- ing : Sarcolactic is obtained by boiling cyanhydrine with CH 2 .OH alkalies. Cvanhydrine, | , is obtained from ethy- CH 2 .CN lene chlorhydrine, | , by treating the latter with CH. 2 .C1 potassium cyanide. Ethylene chlorhydrine is obtained by treating ethylene with hypochlorous acid. Ethylene CH 2 is || , as has been shown. Further, sarcolactic acid CH 2 contains the group CH 2 OH, for, by oxidation, it yields an acid containing the same number of carbon atoms. If it contains the group CH 2 OH, it must have the consti- tution represented by the formula CH , which \COOH is that above given. The only other possible compound of this composition must have the formula CH(OH)<^ . Conse- X COOH quently, the latter is the formula of ordinary lactic acid. Such a compound could not, by oxidation, yield an acid containing the same number of carbon atoms. Ordinary lactic acid breaks up by oxidation, yielding both formic and acetic acids. The remaining diatomic acids of the series have not been as well studied as the few which have here been considered. 204 CHEMICAL COMPOUNDS. One peculiarity of the above acids should be noticed. They are monobasic acids, but still they are capable of forming anhydrides by losing one molecule of water from one of their own molecules. The anhydrides thus formed differ somewhat from the ordinary anhydrides of acids. /--\ Thus we have glycolic anhydride, CH./ ;> , formed X CCK by abstracting one molecule of water from one molecule of glycolic acid, CH ; lactic anhydride, \COOH /--\ CEL.CH7 > , formed from one molecule of ordi- \CO/ OH nary lactic acid, CH,.CH< \COOH Bibasic Acids, C n H 2n _ z 4 Oxalic acid, C 2 H 2 4 , or COOH , is the simplest representative of these acids COOH possible. The fact that it is bibasic,and that the number of groups COOH contained in a compound determines its basicity, leads to the formula given. The second member of this series is malonic acid, . CH 2 < . Of each of these acids only one variety \COOH is possible. The third member is succinic acid, C 2 H \COOH Of this there must be two varieties corresponding to the two lactic acids, or the two series of mono-substitution products of propionic acid. For succinic acid may plainly be considered as propionic acid in which a hydrogen atom has been replaced by a carboxyl group. The tw r o &uc- cinic acids would have the following formulas : CONSTITUTION OF CHEMICAL COMPOUNDS. 205 H CO.OH H H I I II 1. H CC CO.OH and 2. CO.Ofl C C CO.OH II II H H H H The second formula is that of ordinary succinic acid, and the first that of isosuccinic acid. Proofs. Ordinary snccinie acid is obtained from /3- eyanpropionic acid, the constitution of which we know H H be CN C C CO.OH ; and from ethylene cya- to CH 2 .CN nide, | , which Is obtained by treating ethylene CH,.CN bromide with potassium cyanide. Isosuccinic acid is obtained from a-cyanpropionic acid, H CN which is H C C COOH . A Fifth Group. Triatomic Alcohols and Acids. Glycerin. Only one alcohol is well known which con- tains three hydroxyl groups. This is glycerin. Such alcohols are known as triatomic alcohols. The formula CH,OH of glycerin is CHOH . This formula is very pro- OET,pH bable, because, as a result of a large number of observa- tions of carbon compounds, it seems to be a general fact that one carbon atom cannot hold in combination more than one hydroxyl group. If this is true, the above for- 18 206 CHEMICAL COMPOUNDS. mula is the only one possible for glycerin. But, again, by oxidation, glycerin yields a monobasic acid containing the same number of carbon atoms; and, by further oxi- dation, apparently a bibasic acid also containing the same number of carbon atoms. These facts would show that the group CH.,OH occurs twice in glycerin. But if there are two groups CH 2 OH present in glycerin, then the for- mula above accepted must be correct. Glyceric Acid is obtained by partially oxidizing gly- cerin. As the acid contains the same number of carbon atoms as glycerin contains, it is assumed that the oxi- dation consists in a transformation of the primary alco- hol group CH.OH into COOH; hence, the formula of CH 8 OH I glyceric acid is CHOH COOH COOH According to this, a bibasit; acid , CHOH , ought to be obtained by oxidizing COOH glyceric acid, just as this bibasic acid is obtained by oxi- dizing glycerin. This transformation has not yet been effected. Sixth Group. Tetr atomic Compounds. The best known members of this group are tartaric acid and citric acid. The former is a bibasic acid,, con- taining, in addition to the two carboxyl groups, two alcoholic hydroxyl groups. Jt is hence a bibasic tetratomic acid. It is dioxysuccinic acid, and must CH.OH.COOH have the formula | It is obtained CH.OH.COOH from dibromsuccinic acid by treating the latter with water, thus : CONSTITUTION OF CHEMICAL COMPOUNDS. 207 CHBr.COOH CH.OH.COOH | -f 2H = | -f- 2HBr . CHBr.COOH CH.OH.COOH Dibromsucciuic acid. Tavtaiic acid Citric acid is tribasic, containing, in addition to its three carboxyl groups, one alcoholic hydroxyl. It is hence a tribasic tetratomic acid. In addition to the groups above referred to, there are pentatomic and hexatomic compounds. Of the former, there is only one representative known. Of the latter, however, a large number of members are known. Among these are the different varieties of sugars, cellulose, and starch; and the acids which are derived from them. All that is positively known of these compounds is that they contain a certain number of hydroxyl groups, or of hydroxyl and carboxyl groups. The presence of the carboxyl groups is detected through the acid properties of the substance. If the substance is a monobasic acid, one carboxyl group is assumed as being present in it ; if it is a bibasic acid, two carboxyl groups are assumed as being present in it, etc. The number of hydroxyl groups present is determined by allowing acetyl chloride or acetic anhydride to act upon the compound. If the latter contains only one hydroxyl, it will take up only one acetyl group, C 2 H S ; if it contains two hydroxyl groups, it will take up two acetyl groups, etc. Seventh Group. Cyanogen Compounds. In speaking of the group CN as a substituting group, the proofs for the formula C^N for this group were given (ante, p. 152). Now this same group is obtained from the compounds known as cyanides, and hence the cyanides have an analogous constitution. Cyanogen C~N itself has the formula C 2 N 2 or | . The simplest C=N compound of cyanogen is hydrocyanic acid, which con- 208 CHEMICAL COMPOUNDS. sists of the group C N combined with hydrogen, viz., H C=N. The Itydrog^n atom of this acid may be replaced by a variety of groups or other elements, as, for instance, OH, SH, NH. 2 , etc. Then a large number of derivatives are obtained which have a constitution similar to that of the acid. Thus we have cyanic acid, HO C:^N ; sul- phocyanic acid, HS C N; cyanamide, H 2 N C^N, etc. It has already been shown that there are compounds containing the group CE=N called carbylamines, which are isomeric with the cyanides of hydrocarbon residues, and the proofs for the formula Ci^N have also been given (see ante, p. 153). Mustard, Oils. Sulphocyanic acid, HS C N, like other acids, yields salts and ethers by exchanging its hydrogen for metals or hydrocarbon residues. We have potassium sulphocynnate, KS C=N; methyl sulpho- cyanate, CH 3 S C==N, etc. Running parallel to the ethers of sulphocyanic acid is a series of compounds known as mustard oils. These have the same composi- tion as the above ethers, but entirely different properties and constitution. The simplest representative of this series is methyl mustard oil, which has the constitution expressed by the formula S=C=N CH 3 . A number of corresponding compounds are known, one of which is allyl mustard oil, S=C=N C 3 H 5 . This is the oil obtained from black mustard seed. The proofs of the constitution assigned to the mustard oils are as follows : Ethyl mustard oil is formed by a somewhat circuitous method. When carbon bisulphide, CS^, is brought in /C S H 5 contact with ethylamine, N H , the ethylarnine \H salt of ethyl sulphocarbamic acid is formed, thus: CS 2 + 2(NH,.C 2 H 5 ) = CS< CONSTITUTION OF CHEMICAL COMPOUNDS. 209 I5y appropriate reactions, this salt is split up into ethylamine, hydrogen sulphide, and ethyl mustard oil. The decomposition can be best interpreted as follows: --/NJHJ.C.H. ^J O , , ei rr I TVT TT H|.|NH..C,H 5 ' Hence the resulting mustard oil retains an atom of sulphur, combined by means of two affinities with carbon, and the residue of ethylamine, =N.C. J H 5 , being bivalent, would naturally be held b}' the two remaining affinities of the carbon. But, if we examine the products of de- composition of ethyl mustard oil, we are also led to the formula above given. With water or hydrochloric acid it yields ethylamine, carbon dioxide, and hydrogen sulphide; with nascent hydrogen it yields ethylamine formylsulphaldehyde and hydrogen sulphide. The pro- duction of ethylamine indicates clearly that, in the mus- tard oil, the ethyl group is in combination with the nitrogen atom; and the production of formylsulphalde- hyde, which differs from formic aldehyde, H.COH, only in containing sulphur in the place of oxygen, also indi- cates that in ethyl mustard oil the sulphur atom is in combination with carbon. These results are embodied in the formula accepted for the mustard oil. The ether of sulphocyanic acid, which is isomeric with ethyl mustard oil, conducts itself towards reagents in an entirely different manner. It never yields ethylamine, but always yields a compound in which the ethyl group is in combination with sulphur, as ethylsulphide or ethyl- sulphurous acid ; while the nitrogen' is split off in com- bination with hydrogen alone, or with carbon, hydrogen, and oxygen. Eighth Group. Derivatives of Carbonic Acid. The salts of carbonic acid have the general formula M 2 CO S . They are derived from a bibasic acid, H 2 C(X. This acid being bibasic contains two hydroxyl groups, OH and hence we are led to the formula C0( for 18* 210 CHEMICAL COMPOUNDS. carbonic acid. No such acid is known, however, If we attempt to prepare it from its salts, we always get the compound C0 2 , which may justly be considered as the anhydride of the true carbonic acid. It has already been stated, that it appears to be a general truth, that one carbon atom cannot hold in combination more than one hydroxyl group. This breaking up of carbonic acid into water and the anhydride is in harmony with the general truth. Whether the acid is formed or not when the anhydride is conducted into water is not yet decided. Though we are not acquainted with carbonic acid, we are acquainted with a very large number of its derivatives. These are obtained, 1, by replacing the hydrogen of the acid by elements or groups; 2, by replacing one or both of the hydroxyl groups by groups or elements; 3, by replacing the oxygen by sulphur. Thus we obtain first a series of salts and ethers ; then compounds, such as carbonyl chloride, C=O , carbon sulphoxide, \C1 ^s ,s C^ , carbon bisulphide, CZ ; and finally snch ^O ^S SH com pounds as sulphocarbonic acid, CS<^ , xan- ihogenic acid, CS<^ , etc. Among the most important derivatives of carbonic acid is the amide urea or carbamide, which has the con- NH, stitution expressed by the formula CO The proofs of this formula are as follows : It is formed by the action of carbonyl chloride upon ammonia, thus : COC1 2 -f 2NH 3 = C0< + 2HC1 . CONSTITUTION OF CHEMICAL COMPOUNDS. 211 Also by the action of ammonia upon ethylcarbonate, thus: OC 2 H 6 C0< + 2NH 3 = C0 N)C 2 H 5 The latter is a general reaction employed for the pro- duction of acid amides from the ethers. Urea has the power of combining with bases, acids, and salts, and of forming with them crystallizing com- pounds. Instead of the ammonia residue NH.^, further, it may contain residues of the amine bases, as NH.CH S , NH.C 2 H 5 , etc. Or, again, one or more of the hydrogen atoms of urea may be replaced by acid residues, such as C 2 H 3 0, C 7 H 5 O, etc. A large number of compounds are allied to and derived from uric acid. They have frequently been the subjects of exhaustive investigations, but, up to the present, no formula has been proposed for uric acid which is in every respect satisfactory. It is a weak bibasic acid, but it does not contain two carboxyl groups, for its formula is C 5 N 4 H 4 3 , while a compound which contains two carb- oxyl groups must contain four atoms of oxygen. The presence of the group C^rN seems to be pretty clearly indicated in the acid, for it yields, with great ease, pro- ducts which certainly contain this group. The amide group NH. 2 is also probably present in it, for, when heated with hydriodic acid, it vields, among other products, gly- /NH, cocol or amido-acetic acid, CH { X COOH BENZENE DERIVATIVES. (AROMATIC BODIES.) A large class of compounds exists which possess the property in common that, when decomposed in a number of ways, they yield benzene as one of the products. Benzene itseli' has the formula C 6 H 6 . Just as" the mem- bers of this class of compounds yield benzene as a de- composition product, so, also, they may all be built up 212 CHEMICAL COMPOUNDS. from benzene by the introduction of a variety of groups or elements in the place of hydrogen. All these com- pounds bear a similar relation to benzene to that which the fatty bodies bear to marsh-gas. In studying the aromatic bodies, then, it is plainly our first duty to de- termine the constitution of benzene itself, as the consti- tution of the derivatives cannot be understood until this determination is made. Constitution of Benzene. The great stability of ben- zene indicates that it is a saturated compound. Now, if the carbon atoms contained in it are quadrivalent, the simplest hypothesis which can be formed concerning its constitution would be indicated in the following formula : H C CH HC CH V '; ' H According to this, the molecule of benzene consists of a closed chain of carbon atoms, each united, on the one hand, by two affinities with another carbon atom ; on the other hand, by one affinity with a second carbon atom. This formula, which was originally proposed by Kekule, accounts satisfactorily for nearly all the facts known concerning aromatic bodies. These facts are mainly the following: 1. Of the substitution products of benzene which con- tain one substituting group, only one variety is known. 2. Of the substitution products of benzene which con- tain two substituting groups, three varieties have been observed, and only three. 3. Of the substitution products of benzene which con- tain three substituting groups, more than three varieties have been, observed. If all the hydrogen atoms in benzene play exactly the same parts, then the first fact mentioned would follow as CONSTITUTION OF CHEMICAL COMPOUNDS. 213 a matter of course. In the above formula, all the hydro- fen atoms are represented as playing the same parts. ]ach one is situated exactly like all the others with refe- rence to the whole molecule. A great many efforts have been made to obtain iso- meric mono-substitution products of benzene, but they have all been unsuccessful. Again, if we examine the above formula carefully, we find that there are three and only three pairs of hydrogen atoms in it which differ from each other in the positions of their individual atoms. Numbering these hydrogen atoms as follows : 1 H C /\ 6 HC CH 2 II I 5 HC CHS \> C H 4 we can distinguish the following pairs: 1.2, 1.3, 1.4, 1.5, 1.6 ; or, beginning with 2, we would also have five pairs ; but, as all the hydrogen atoms of benzene play exactly the same parts, it is plainly immaterial with which one we begin, the resulting pairs will be identical. Thus, 1.2 is identical with 2.3, 3.4, 4.5, and 5.6 ; 1.3 is identical with 2.4, 3.5, 4.6, and 5.1; 1.4 is identical with 2.5, 3.6,4.1, and 5.2; etc. But, further, 1.2 is also identical with 1.6, and 1.3 with 1.5. Hence, of the five original pairs we have only three left. These are 1.2, 1.3, and 1.4. They are the only ones that differ from each other essentially in the benzene formula of Kekule. If, then, substitution products, containing two substituting groups, are obtained from benzene, they have one of the three following for- .mulas, in which X represents a monovalent substituting group or element : 214 CHEMICAL COMPOUNDS. ceo / \ / \ / ,\ HO CX HC CH HC CH 1- II I ; 2. || | ; 3. || | . HC CH HC CX HC CH \S \ ^ V> c c c H H X As was staterl above, only three varieties of bisubsti- tution products of benzene have ever been observed. So that here, again, we have perfect harmony between facts and the hypothesis. No one claims that the benzene formula of Kekuld represents the actual arrangement of the atoms in space. It undoubtedly represents certain truths, however. It represents that in the molecule of benzene, the hydrogen atoms are arranged symmetrically, and that all the parts of the molecule are symmetrically arranged. We do not know positively that there is such symmetry in the ben- zene molecule, for we know nothing of molecules them- selves, but, from all the facts known to us, it seems fair to conclude that this symmetry of the different parts is characteristic of the benzene molecule. Substitution Products of Benzene. Of mono-substi- tution products we have only one variety. We have only one monochlorbenzene, C 6 H 5 C1 ; only one oxybenzene, or phenole, C fi H..OH; only one benzole acid, C 6 H 5 .COOH; only one toluene, C 6 H 5 .CH 3 , etc. etc. The constitution of most of these derivatives is very simple. There is a peculiarity, however, connected with those which are formed by replacing one hydrogen atom of benzene with a hydrocarbon residue. The simplest compound formed in this way is toluene, which consists of benzene in which a hydrogen atom has been replaced by the methane resi- due CH 3 ; if, instead of the residue CH,, we introduce C 2 H., we obtain ethylbenzene, C 6 H 5 .C 2 H 5 , which is plainly an homologue of toluene; so, also, the residues C 3 EJ., C 4 H,, C.H n , etc., may e employed, and thus we obtain an homologous series of aromatic hydrocarbons, all of which are mono-substitution products of benzene. These may, CONSTITUTION OF CHEMICAL COMPOUNDS. 215 further, all be regarded as substitution products of the hydrocarbons of the methane series. Accordingly, of toluene and ethylbenzene, which are mono-substitution products of methane and ethane respectively, only one variety each is possible ; while of the next homologue, or propylbenzene, C a H 5 .C 3 H., two varieties are possible, cor- responding to the a- and |3-monosubstitution products of pro pyl, or to the prop} 7 ! and isopropyl compounds (which see). The main members of the series of hydrocarbons thus referred to are : Benzene, C 6 H 6 . Toluene or methylbenzene, C 7 H 8 or C (i H 5 .CH 3 . Ethylbenzene, C 8 H 10 or C 6 H 5 .C 2 H 5 . Propylbenzene, C 9 H ]a or CfLCJBL. Butylbenzene, C 10 H I4 or C 6 H 5 .C 4 H 9 . Amylbenzene, C n H 15 or C 6 H 5 .C 5 H n . Of these hydrocarbons, two kinds of mono-substitution products are possible, viz., those in which the substi- tuting group or element is situated in the benzene nucleus, and those in which the substituting group or element is situated in the other residue. These other residues, however they may be constituted, are known as lateral chains. It is plain that substitution products of the latter kind correspond closely to those of the hydrocarbons of the methane series, and hence they need no special con- sideration here. If a substituting group or element enter into the benzene nucleus of any of these hydro- carbons, of course we have no longer to deal with mono- substitution products of benzene. Bisnbstitution Products. The three classes of bi- derivatives of benzene which we have above recognized as possible, have been designated respectively as ortho, meta, and para compounds, or, by others, as 1.2. 1.3, and 1.4 compounds. The former expressions are to be pre- ferred, for the} 7 are independent of any hypothesis con- cerning the positions of the substituting groups. It is usual to consider the expressions ortho and 1.2, meta and 1.3, para and 1.4, as identical, but this implies that the following formulas have been proved, while they have not been : 216 CHEMICAL COMPOUNDS. XXX c c c /\ /\ /\ HC CX HC CH . HC CH HC I ; II i ; II I - CH HC CX HC CH c c c H H X Ortho-compound. Meta-compound. Para-compound. What we really know is that there are three classes of these bi-substitution products, and that the members of any one of these classes can be converted into each other, thus showing that they are allied. There are three com- pounds, each representing one of the three classes to which all other bi-substitution products are referred, if possible. If the constitution of any such product is unknown, it is only necessary to convert it into one of the three compounds, when the series to which it belongs is assumed to be known. The three compounds are the /COOH isomeric, bicarbonic acids of benzene, C 6 H / , \COOH viz., phthalic, isophthalic, and terephthalic acids. All bi- substitution products which can be converted into phthalic acid are known as ortho-compounds ; all that can be con- verted into isophthalic acid are known as meta-compounds ; and all that can be converted into terephthalic acid are known as para-coin pounds. The conversion into these acids need not always be direct. If it be possible to convert a compound into another which, in its turn, can be converted into one of the above acids, the same conclusion is drawn as in the case of a direct conversion. Of course, the accuracy of the conclusions drawn with reference to the constitution of bi-substitution products depends upon the trustworthi- ness of the reactions employed in effecting the conver- sions. Some reactions employed for this purpose have been found to give inaccurate results ; jthat is to say, the products resulting from an application of these reactions belong to different series from those to which the original compounds belonged. It is very probable that some CONSTITUTION OP CHEMICAL COMPOUNDS. 217 compounds now classified with one series in consequence of some conversion, may be found, by future investiga- tions, to belong to a different series. The formulas given above as representing the relative positions of the substituting groups in ortho-, meta-, and para-compounds are based upon the following facts : It will be shown that naphthalene (which see; probably has the formula H H C C HC CH \ / c=e / \ HC1. 2.CH \ / C C H H By oxidation, naphthalene yields phthalic acid. It seems probable, therefore, that the carboxyl groups in the acid have the same relative position as that of the groups numbered 1 and 2 in this formula. Consequently, ortho- compounds, or those which can be converted into phthalic acid, have their substituting groups in the positions 1.2 in the benzene nucleus ; or, what is the same thing, the substituting groups in ortho compounds are combined with adjacent carbon atoms. It will also be shown that mesitylene (which see) prob- ably has the formula HC CH II I CH 3 .C C.CH 3 \S ' C H By partially oxidizing this hydrocarbon, an acid is obtained of the formula 19 218 CHEMICAL COMPOUNDS. COOH C /\ HC CH II I CH.C C.CH 3 V H When this acid is heated under proper conditions carbon dioxide is given off and a hydrocarbon is obtained of the formula H C x\ HC CH !! I CH.C C.CH, H Lastly, when this hydrocarbon is oxidized, both the groups CH 3 are converted into COOH, and the resulting acid is isophthalic. Hence, if the formula of mesitylene is correct, that of isophthalic acid is also correct. By exclusion, terephthalic acid becomes a 1.4 com- pound, and, consequently, all para-compounds are 1.4 compounds. It must be confessed that these proofs are not strong enough to command universal respect among chemists. As the expressions 1.2, 1.3, and 1.4 are in common usage, it is well, however, to know the grounds upon which their use is based. Some of the principal bi-substitution products of benzene are given in the following table, which shows also to which series the compounds belong: Ortho. Meta. Para Phthalic acid, . Isophthalic acid, Terephthalic acid, Orthoxylene, Isoxylene, Xylene, Salicylic acid, Oxybenzoic acid, Paroxybenzoic acid, Pyrocatechin, Resorcin, Hydroqninone, Orthodinitrobenzene, Metadinitrobenzen'e, Paradinitrobenzene, Orthobibrotii benzene. Mctabibrombenzene. Parabibrombenzene. CONSTITUTION OF CHEMICAL COMPOUNDS. 219 It is well to note here that whether both substituting groups be the same or not, only three kinds of products can be obtained. Tri-substitution Products. The most important of the tri-substitution products of benzene is rnesitylene. The formula of this hydrocarbon is CyH^. By oxidation it yields, according to the energy of the reaction, three different products. The first, mesitylenic acid C 8 H 9 .COOH, is monobasic; the second, uvitic acid C 7 H g .(COOH) 11 is bibasic ; and the third, trimesinic acid, C H S .(OOOH) 3 , is tribasic. All of these acids, when heated with lime, yield either benzene itself or derivatives of benzene. Hence, it is concluded that mesitylene is benzene in which three hydrogen atoms are replaced by the residue CH 3 thus, Cf 6 H 3 (C H 3 ) 3 . By oxidation each one of these groups in turn is converted into carboxyl, yielding thus the three acids above mentioned. It still remains, however, to decide what the positions of these three substituting groups in the benzene nucleus are. The following method of consideration leads to the formula for mesitylene given on page 2 It : When acetone "is treated with concentrated sulphuric acid, water is abstracted and the residues of three mole- cules unite to form mesitylene. It seems to be fair to assume that the three residues are constituted exactly the same, as they are formed under exactly the same conditions, from the same compound. If they are the same, they must each be C 3 H 4 . Three such residues might be formed from acetone, thus : Acetone is CH 3 CO CH 3 ; three molecules may be arranged CH 3 H CH o c ^ |5 ~ X CHffi n I 13 1,-c o i C CII 3 / H/ 220 CHEMICAL COMPOUNDS. If water is abstracted in the manner indicated by the lines, we have left three residues, 8 H 4 , and, if these unite, they would form a compound of the constitution represented by the following formula: CH, HC CH CH, C C CH 3 ' V H This is the formula accepted for mcsitylene; and from this we conclude, as above seen, that meta-compounds have their substituting groups in the positions 1.3. If this formula is carefully examined, it will be seen that each one of the three hydrogen atoms remaining in the benzene-nucleus occupies a similar position to that occupied by the other two. Accordingly, if this formula is correct, we should expect to find that, by the intro- duction of one substituting group into mesitylene, only one product would be formed. This has actually been found to be true. Besides mesitylene, there are many tri-substitution products of benzene known, containing such elements as 01, Br, I, and such groups as N0 2 , NH.,, SO.,OH, etc. The principle, according to which the position of the substituting groups in these compounds is determined, is this : One of the groups is split off, and the constitution of the resulting bi-substitution product is determined as above; then from the original compound some other group is split off, and the constitution of the bi-substitu- tion product resulting in this case also determined. We are thus able to judge of the positions of the three groups with reference to each other. There are not many com- pounds, however, which can be subjected to this kind of examination with satisfactory results, so that the con- stitution of these tri-clerivatives is not really as well known as that of the bi-derivatives. Those derivatives of benzene which contain four or CONSTITUTION OF CHEMICAL COMPOUNDS. 221 five substituting elements or groups have not been very full} 7 investigated. Also only a few hexa-derivatives are known, the most important of which is mellitic acid, C (i (OOOH) 6 , or benzene, in which all six hydrogen atoms are replaced by carboxyl groups. Peculiar Benzene Derivatives. Among benzene derivatives we find three classes which are not represented among the fatty bodies, and hence they require some attention here. These are the jjhenoles, quinones, and azo-bodies. Phenoles. Phenoles are the oxy-derivatives of ben- zene and its homologues, formed by the introduction of hydroxyl into the place of hydrogen in the benzene nucleus. The corresponding compounds of the hydro- carbons of the methane series are all alcohols, either primary, secondary, or tertiary. The phenoles are, however, not alcohols in the sense in which that term has been used up to the present. By oxidation they yield neither aldehydes, acids, nor ketones. The presence of hydroxyl in phenoles can be proved in the same way that it was proved for other bodies containing hydroxyl. There are monatomic phenoles, containing only one hydroxyl; biatomic phenoles, containing two hydroxyls; triatomic phenoles, containing three hydroxyls, etc. Quinones. The quinones are derived from benzene and its homologues by the introduction of two atoms of oxygen in the place of two hydrogen atoms in the ben- zene-nucleus. Thus the simplest quinone has the formula C 6 H 4 O k . The two oxygen atoms are supposed to form a bivalent group, O , by combining with each other by means of one of their affinities each. Most quinones are derived from para-compounds by oxidation, as from hydroquinone, and hence it is concluded that the hydrogen atoms replaced by the bivalent group O in the formation of quinones usually occupy the para-position with reference to each other. Accordingly, if the para- position is 1.4, the formula of ordinary quinone is 19* 222 CHEMICAL COMPOUNDS. C HC CH | IS nUin = C H <" Y ' ;f|p All quinones are supposed to be similarly constituted, though it is still a question whether all quinones are para-compounds. Certain experiments seem to indicate that there are quinones which belong to the meta-series. Azo- and Diazo-Bodies. These bodies, as their names imply, are nitrogen derivatives. They are derived from benzene and its homologues by the replacement of hydro- gen by nitrogen. We shall consider those which are derived from benzene, as the others are very closely related to these, and will be understood if these are. The diazo-derivatives of benzene are obtained from the salts of anilin or amidobenzene, C 6 H 5 .NH V , by the action of nitrous acid. Thus anilin nitrate, C 6 H 5 .NH 2 .HN0 3 , yields diazobenzene nitrate ; anilin sulphate, (C 6 H..NH 2 ) a .H 2 S0 4 , yields diazobenzene sulphate, etc. If we consider simply the empirical formulas of the salts of diazobenzene thus obtained, we shall find that they differ from the anilin salts in containing C 6 H 4 N 2 in the place of C 6 H.NH 2 . The salts consist of the acids plus this group. Thus the nitrate is C 6 H 4 N 2 .HNO 3 ; the sulphate is C 6 H 4 N 2 .H 2 S0 4 , etc. These formulas are not supposed, however, to represent the constitution of the salts. If the group C 6 H 4 N 2 actually existed in these diazo-bodies, it is plain that they would be bi-substitution products, that is to say, two hydrogen atoms of benzene would be replaced by two nitrogen atoms. It was at first supposed that each of these nitrogen atoms played the part of a monovalent element, and the diazo-com- pounds were looked upon as analogous to bichlorbenzene, binitrobenzene, etc., thus: CONSTITUTION OF CHEMICAL COMPOUNDS. 223 N C HC CN Cl C HC CH HC =1 CCl I CH C H Diazobenzene. H Bichlorbenzene. It was soon found, however, that, when the cliazo- bodies were decomposed, they almost always yielded derivatives of benzene in which the group C 6 H 5 was undoubtedly present. Thus the following decomposi- tions of diazobenzene sulphate yield, in each case, a derivative containing C 6 H 5 : When boiled with alcohol, the products are benzene, nitrogen, and sulphuric acid C 6 H 5 .X 2 .HS0 4 H H yield H H When boiled with water, the products are phenole, nitrogen, and sulphuric acid C 6 H,.X 2 .HS0 4 OH H yield OH HS0 4 H When treated with hydriodic acid, the products are iodbenzene, nitrogen, and sulphuric acid C 6 H 5 .X 2 .HS0 4 I H yield X. HS0 4 H Other reactions indicate as well that the group C 6 H. is present in the diazo-compounds. But, if this group is present, the two nitrogen atoms must form a monovalent group, or, at all events, they must be so combined that they can take the place of one hydrogen atom. Xow, if two nitrogen atoms which have the same valence be combined, they must either form a neutral group with all its affinities satisfied, or a- group which is at least bivalent. Such a bivalent group would be formed, for 224 CHEMICAL COMPOUNDS. instance, if two nitrogen atoms were to be united by means of two affinities each, thus, N=N . If this group should replace one hydrogen atom of benzene, the constitution of the resulting compound would be C 6 H 5 N N . iSuch a compound would be unsaturated. No compound of the formula C S H 5 N 2 has been obtained, but all the derivatives of diazobenzene can be explained on the supposition that they are derived from the com- pound C 6 H 5 N=N . Griess, who discovered the diazo- and azo-compounds, and has given a great deal of time to their study, has described a body of the formula C 6 H 4 N 2 , which he calls diazobenzene. If this body really exists, the conclusion above drawn concerning the pre- sence of the group C H H 5 in diazo-bodies would be weak- ened ; but there seems to be just cause to doubt its existence. Accepting the group C 6 H 5 N=N as the foundation of the diazo-compounds, these may be formulated as follows : C 6 H 5 N=N Br, diazobenzene bromide, C 6 H 5 N=N NO., diazobenzene nitrate, C (i H 5 N N HS0 4 , diazobenzene sulphate, C 6 H. N=N OK, diazobenzene potassa, [zene. C 6 H 3 N=N NH(C 6 H 5 ) diazobenzene diamicloben- Azobenzene is formed by the reduction of nitrobenzene. Its formula is C la H 10 N 2 . As nitrobenzene contains the group C 6 H 5 combined with N, we can assume that azo- benzene consists of two such groups C 6 H 5 N=. If these combine in the simplest manner, we would have C.H.-N the formula || , expressing the constitution of C 6 H-N azobenzene. This is the formula which is now generally adopted. According to this, the azo-compounds are very closely related to the diazo-compounds. Both contain the group N=N in combination with C (1 H 5 . In' reality, the azo-compounds differ very much in their chemical con- duct from the diazo-compounds. The decompositions which they undergo' take place in a manner entirely dif- ferent from that already -noticed as characterizing the decomposition of diazo-compounds. CONSTITUTION OF CHEMICAL COMPOUNDS. 225 This difference has led some chemists to abandon the formulas above given for the diazo-compounds, proposing in their place others. The compounds are supposed to be ammonium compounds of the general formula R N R/ They contain one quinquivalent and one I N0 3 ; N NO, trivalent nitrogen atom. The relation between anilin nitrate and diazobenzene nitrate is shown thus : C 6 H 6 C U H 5 J I' H 3 N Anilin nitrate. Diazobenzene nitrate. It remains to be determined, by future experiments, which of the formulas for diazo-compounds is correct. Up to the present there exist no good proofs of either. NAPHTHALENE DERIVATIVES. The hydrocarbon naphthalene has the formula C 10 H 8 . It is considered as being formed by the union of two benzene nuclei, and as having the constitution expressed by the formula H H C C S \ HC CH r\ r\ HC CH C C H H This formula is deduced from the following facts: There is a derivative of naphthalene known as dichlor- naphthoquinone, which has the formula C 10 H 4 C1 2 2 . When this substance is oxidized, it yields phthalic acid, 226 CHEMICAL COMPOUNDS. which is a bi-substitution product of benzene. We see thus that those carbon atoms in dichlornaphthoquinone which are not in combination with chlorine form a ben- zene nucleus, so that we might write the formula of the compound C 6 H 4 .C 4 C1 2 2 . This formula does not tell us in what manner the atoms C 4 C1. 2 2 are united, but, by the aid of another experiment, this can be determined. When dichlornaphthoquinone (the same substance used in the preceding experiment) is treated with phosphorus pentachloride, it is converted into pentachlornaphthalene, the formula of which, according to what was said above, is C 6 H 3 C1.C 4 C1 + . By analogy, we would expect this com- pound by oxidation to yield monochlorphthalic acid ; it, however, yields tetrachlorphthalic acid. This shows that the four carbon atoms which are in combination with chlorine form part of a benzene nucleus, as well as the other carbon atoms of naphthalene. It is thus proved that in naphthalene there are two benzene nuclei. The only formula which agrees with this fact is the one above given. The derivatives of naphthalene resemble those of ben- zene, and much that has been said concerning this latter would hold good in regard to the former. All the hydro- gen atoms of naphthalene may be replaced by substi- tuting groups or elements, and thus, as will be readily seen, a large number of substitution products may be obtained. The possibilities for instances of isomerisrn are greater in the case of naphthalene than in the case of benzene, but the principles governing the matter of iso- merism are essentially the same as those which we have already considered in connection with the isomeric sub- stitution products of benzene. Anthracene Derivatives. Anthracene, like naphtha- lene and benzene, is the mother-substance of a large group of compounds. Its formula is 14 H 10 . In regard to its constitution, the view is now commonly held that it consists of two benzene nuclei, C 6 H 4 , held together by i i means of the group HC CH , each carbon atom of i i which is united with both benzene nuclei, thus: CONSTITUTION OF CHEMICAL COMPOUNDS. 227 H H C C / \ HC CH HC CH HC CH \ / C C H H This formula shows the relation between anthracene and anthraquinone, which latter compound appears to be CCK C 6 H/ /C 6 H 4 . The latter formula best explains \C(K the formation of anthraquinone from benzoic acid, and the formation of benzoic acid from anthraquinone. The former transformation is represented thus : C.H.J1J CO |oH| " ' . = C 6 H,< >C.H. + 2H,0. C.H 4 jH| CO OHj 2 molecules Benzoic acid. Anthraquinone. The formation of anthraquinone from anthracene would be then represented thus: CH C0 According to these interpretations, anthraquinone is a double acetone. It has been suggested, further, that all quinones are nothing but double acetones. (See Qui- nones, p. 221.) The derivatives of anthracene resemble in some re- 228 CHEMICAL COMPOUNDS. spects those of naphthalene and of benzene. It will be seen, however, that some differences must exist between them. Other hydrocarbons allied to naphthalene are pyrene, chrysene, and phenanthrene. These have not been in- vestigated very fully as compared with naphthalene and anthracene themselves. All of these three undoubtedly contain benzene nuclei as essential parts of their mole- cules, but there is still some doubt in regard to the manner in which these nuclei are united. INDEX. ACETONES, 142, 195 Acetylene, 199 Acids, 112 Acids : acrylic, 198 boric, 176 butyric, normal, 193 caproic, normal, 194 caproic ordinary, 194 carbonic, 209 of chlorine, 163 chromic, 181 citric, 207 cyanic, 208 dithionic, 168 gly eerie, 206 glycolic, 202 hypophosphorous, 172 hyposulphurous, 167 isobutyric, 194 isocaproic, 195 isophthalic, 216 isosuccinic, 205 lactic (oxypropionic), 202 metaphosphoric, 176 nitric, 171 nitrous, 171 oxalic, 204 pentathionic, 169 phosphoric, 174 phosphorous, 172 phthalic, 216 polysilicic, 177 propionic, 192 pseudocaproic, 195 pyrochromic, 181 pyrophosphoric, 175 pyrosulphuric, 167 sarcolactic, 203 silicic. 177 20 Acids, succinic, 205 sulphuric, 166 sulphurous, 165 tartaric, 206 terephthalic, 216 tetrathionic, 169 thiosulphuric, 167 trithionic, 168 uric, 211 valeric, normal, 194 valeric, ordinary, 194 Acids of carbon, 135 Acids, hydrogen, 113 Acids, hydroxyl, 113 Acids, subdivision of, 116 Alcohols, 128 Alcohols: allyl, 197 amylene hydrate, 188 amyl of fermentation, 188 amyl, normal, 188 amyl, tertiary, 188 butyl, normal, 187 butyl, secondary, 187 butyl, tertiary, 187 diethylmethylcarbinol, 190 dimethylcarbinol, 190 dimethylpseudopropylcarbi- nol, 190 ethylene (glycol), 201 hexyl, primary, 190 hexyl, secondary, 190 isoamyl, 188 isobutyl, 187 propyl, normal, 185 pseudopropyl, 185 Alcohols, primary, 129 Alcohols, secondary, 130 Alcohols, tertiary, 132 Aldehydes, 139, 195 230 INDEX. Alloys, 29 Aluminium, salts of, 180 Amido-group, 100 Ammonium salts, 100 Ampere, 34 Anhydrides, 119 Anhydrides of acids of carbon, 147 Anhydrides, peculiar, 148 Anthracene, 226 Anthraquinone, 227 Artiads, 92 Atomic compounds, 85 Atomicity, 7 9^. Atomic weights, determination of, 18 method by analysis, 19 method of Berzelius, 22 method by substitution, 24 method by chemical decom- position, 26 by Avogadro's method, 32 ff. Atomic theory, 17 Atoms, 17, 36 Avogadro, 33 BASES, 117 Benzene, bisubstitution pro- ducts, 215 Benzene, constitution of, 212 Benzene-nucleus, 215 Berthelot, 15 Berzelius, 22 Boron, an exception to the law of Duloug and Petit, 64 Boron, the element unknown, 67 Boron trioxide, 176 Butane, derivatives of, 185 CARBON, compounds of, 124 Carbon, an exception to the law of Dulongand Petit, 64 Carbylamines, 152 Chemism, defined, 14 Chains, 108 Chains, closed, 109 Chains, open, 108 Chromium, salts of, 17,9 Clausius, 37 Combination, forms of, 101 Compounds, defined, 28 Constitution, 99 Copper, salts of, 178 Cyannmide, 208 Cyanogen compounds, 207 Cyanogen group, 152 D ALTON, investigations, 15 Diazo-bodies, 222 Double union, 91 Dulong, 58 Dumas, 44 T^LEMENTS, defined, 27 LJ Equivalents, 20 Ethane, derivatives of, 183 Ethers, 145 Ethers, simple, 146 Ethylene chloride, 184 Ethylene derivatives, 196 Ethylidene chloride, 184 FAVRE, 41 Formulas, molecular, 52 GASES, 32 Gay Lussac, 32 Glycerin, 205 Glycols, 200 HAMPE, 67 Heptane, derivatives of, 190 Hexane, derivatives of, 189 Homologous series, 126 Humboldt, 32 Hydrocarbons, 125 Hydroxylamine, 171 TMIDE GROUP, 100 1 Iron, salts of, 179 Isomerism, 162 Isomorphism, 68 Isonitriles, 152 J7-RONIG, 37 INDEX. 231 T AVOISIER, 14 1J Law of Avogndro, 37 of Dulong and Petit, 58 of multiple proportions, 16 of Neumann and Regnault, 59 periodic, 76 MARRIOTTE, 36 Maxwell, 37 Mendelejeff, scheme of, 71 Mercaptuns, 134 Mercury, salts of, 178 Mesitylene, 219 Metal acids, 180 Meyer, L., 37, 44 scheme of, 77 Mitscherlich, 44 Mixtures, mechanical, 29 Molecular compounds, 85 Molecular weights, 37 Molecule, 36 Mustard oils, 208 ATAPHTHALENE, 225 11 Nascent state, 50 Naumann, 37 Neumann, 58 Nitriles, 152 Nitrogen, oxides of, 170 Nitrogen, quinquivalent, 89 Nitro group. 156 Nitroso group, 159 XIDES, 123 Ozone, 51 pENTANE, derivatives of, 187 JL Periods in groups of ele- ments, 72 Perissads, 92 Petit, 58 Pfaundler, 37 Phenoles, 221 Phosphorus, oxides of, 172 Profane, derivatives of, 184 Propylene, 197 Proust, 15 QUALITATIVE METHOD, 14 Quantitative method, 14 Quinones, 221 REGNAULT, 58 Residues, 105 OALTS, 118, 178 kJ Saturated compounds, 90 Silbermann, 41 Silicon, an exception to law of Dulong and Petit, 64 Solutions, 29 Specific heat, 56 Substituting groups, constitution of, 151 Substitution, 24 Substitution products, 148 Sulpho group, 154 rpHOMSEN, 37 1 Treble union, 109 Trimethylmethane, 186 Types, 103 Types, theory of, 104 TTNSATURATEDCOMPOUNDS, U 90 Uranium, compounds of, 181 Urea, 210 VALENCE, 79 /. Valence, apparent, 97 Valence, true, 95 Valence, variable, 91 Volume of gases, 32, 33 w OLLASTON, 20 Wxirtz, 93 ERRATA. viv i Page 84, line 2 from bottom, omit " nitric acid, N0 2 (OH)." Page 84, bottom line, omit "nitrates and." Page 191, middle of page, read "A third variety of heptane" for "A third variety of hexane." Page 217, middle of page, after the sentence ending " 1 and 2 in this formula," insert "these groups being considered in their relations to the upper benzene nucleus." Page 227, the two middle carbon atoms in the formula for anthracene should be connected by a line. CATALOGUE OF BOOKS PUBLISHED BY o. (LATE LEA A; BLANCHARD.) The books in the annexed list will be sent by mail, post-paid, to any Post Office in the United States, on receipt of the printed prices. No risks of the mail, however, are assumed, either on money or books. Gen- tlemen will therefore, in most cases, find it more convenient to deal with the nearest bookseller. Detailed catalogues furnished or sent free by mail on application. An illustrated catalogue of 64 octavo pages, handsomely printed, mailed on receipt of 10 cents. Address, HENRY C. LEA, Nos. 706 and 708 Sansom Street, Philadelphia. Free of Postage. A MERICAN JOURNAL OF THE MEDICAL SCIENCES. " Edited by Isaac Hays, M.D., published quarterly, about 1100 large 8vo. pages per annum, MEDICAL NEWS AND LIBRARY, monthly, 384 large 8vo. pages per annum, OR, A MERICAN JOURNAL OF THE MEDICAL SCIENCES, ". Quarterly, MEDICAL NEWS AND LIBRARY, monthly, MONTHLY ABSTRACT OF MEDICAL SCIENCE, 48 pages per month, or nearly 600 pages per annum. In all, about 2100 large 8vo. pages per annum, For five Dollars per annum, in advance For six Dollars per annum, in advance. M M EDICAL NEWS AND LIBRARY, monthly, in advance, $1 00. ONTHLY ABSTRACT OF MEDICAL SCIENCE, in advance, $2 50. OBSTETRICAL JOURNAL. With an American Supplement, edited by J. V. INGHAM, M. D. $5 00 per annum, in advance. Single Numbers, 50 cents. Is published monthly, each number containing ninety- six octavo pages. Commencing with- April, 1873. HENRY C. LEA'S PUBLICATIONS. A SHTON (T. J.) ON THE DISEASES, INJURIES, AND MALFOR- -0. MATIONS OF THE RECTUM AND ANUS. With remarks on Habitual Constipation. Second American from the fourth London edition, with illustrations. 1 vol. 8vo. of about 300 pp. Cloth, $3 25. ASHWELL (SAMUEL). A PRACTICAL TREATISE ON THE DIS- & EASES OF WOMEN. Third American from the third London edi- tion. In one 8vo. vol. of 528 pnges. Cloth, $3 50. A SHHURST (JOHN, Jr.) THE PRINCIPLES AND PRACTICE OF -ti SURGERY. FOR THE USE OF STUDENTS AND PRACTI- TIONERS. In 1 large 8vo. vol. of over 1000 pages, containing 533 wood-cuts. Cloth, $6 50; leather, $7 50. ATTFIELD (JOHN). CHEMISTRY; GENERAL, MEDICAL, AND L*- PHARMACEUTICAL. Seventh edition, revised by the author. In 1 vol. 12mo. Cloth, $2 75; leather, $3 25. BLOXAM (C. L) CHEMISTRY, INORGANIC AND ORGANIC. With Experiments. In one handsome octavo volume of 700 pages, . with 300 illustrations. Cloth, $4 00 ; leather, $5 00. BBINTON (WILLIAM). LECTURES ON THE DISEASES OF THE STOMACH. From the second London ed. 1 vol. 8vo. Cloth, $3 25. (T. LAUDER). A MANUAL OF MATERIA MEDIC A AND THERAPEUTICS. In one 8vo. volume. (Preparing.) BIGELOW (HENRY J) ON DISLOCATION AND FRACTURE OF THE HIP, with the Reduction of the Dislocations by the Flexion Me- thod. In one 8vo. vol. of 150 pp., with illustrations. Cloth, $2 50. B&SHAM (W. E.) RENAL DISEASES; A CLINICAL GUIDE TO THEIR DIAGNOSIS AND TREATMENT. With illustrations. 1 vol. 12mo. Cloth, $2 00. BUMSTEAD (F. J.) THE PATHOLOGY AND TREATMENT OF VENEREAL DISEASES. Third edition, revised and enlarged, with illustrations. 1 vol. 8vo., of over 700 pages. Cloth, $5 ; leather, $6. . AND CTJLLERIEK'S ATLAS OF VENEREAL. See"CuLLERiER." BARLOW (GEOKGE H.) A MANUAL OF THE PRACTICE OF MEDICINE. 1 vol. 8vo., of over 600 pnges. CJoth, $2 50. QAIRD (ROBERT). IMPRESSIONS AND EXPERIENCES OF THE -D WEST INDIES. 1 vol. royal 12mo. Cloth, 75 cents. BARNES (ROBERT). A PRACTICAL TREATISE ON THE DIS- EASES OF WOMEN. In one handsome 8vo. vol. of about 800 pages, with 169 illustrations. Cloth, $5 ; leather, $6. BUYANT (THOMAS). THE PRACTICE OF SURGERY. In one handsome octavo volume, of over 1000 pages, with many illustra- tions. Cloth, $6 25 ; leather, $7 25. TYRISTOWE (JOHN SYEE). A MANUAL OF THE PRACTICE OF -D MEDICINE. A new work, edited with additions by James II. Hutchinson, M.D. In one handsome 8vo. volume of over 1100 pages. Cloth, $5 50 ; leather, $6 50. (Just ready) DOWMAN (JOHN E.) A PRACTICAL HAND-BOOK OF MEDICAL JJ CHEMISTRY. Sixth American, from the fourth London edition. With numerous illustrations. 1 vol. 12mo. of 350 pp. Cloth, $2 25. INTRODUCTION TO PRACTICAL CHEMISTRY, INCLUD- ING ANALYSIS. Sixth American, from the sixth London edition, with numerous illustrations. 1 vol. 12mo. of 350 pp. Cloth, $2 25. MANUAL' OF SURGICAL ANATOMY, with numerous illustrations. In one royal 12mo. vol. Cloth, $2 25. (Lately issued.) HENRY C. LEA'S PUBLICATIONS. BLANDFORD (9. FIELDING). INSANITY AND ITS TREATMENT. With an Appendix of the laws in force in the United States on the Confinement of the Insane, by Dr. Isaac Ray. In one handsome 8vo vol., of 471 pages. Cloth, $3 25. pARTER (R BRUDENELL). A PRACTICAL TREATISE ON DIS- U EASES OF THE EYE. With additions and test-types, by John Green, M.D. In one handsome 8vo. vol. of about 500 pages, with 124 illustrations. Cloth, $3 75. pKAMBERS (T. K.) A MANUAL OF DIET IN HEALTH AND ^ DISEASE. In one handsome octavo volume of 310 pages. Cloth, $275. (Just issued.) RESTORATIVE MEDICINE. An Harveian Annual Oration delivered at the Royal College of Physicians, London, June 21, 1871. In one small 12rno. volume. Cloth, $1 00. pOOPER (3. B.) LECTURES ON THE PRINCIPLES AND PRACTICE U OF SURGERY. In one large 8vo. vol. of 750 pages. Cloth, $2 00. pARPEHTER (WM. B.) PRINCIPLES OF HUMAN PHYSIOLOGY. ^ From the Eighth English Edition. In one large vol. 8vo., of 1083 pages. With 373 illustrations. Cloth, $5 50 ; leather, raised bands, $650. (Just issued) PRIZE ESSAY ON THE USE OF ALCOHOLIC LIQUORS IN HEALTH AND DISEASE. New edition, with a Preface by D. F. Condie, M.D. 1 vol. ]2rno. of 178 pages. Cloth, 60 cents. PLELA.ND (JOHN) A DIRECTORY FOR THE DISSECTION OF U THE HUMAN BODY. In one small royal 12mo. vol. (In press.) pENTURY OF AMERICAN MEDICINE. A HISTORY OP MEDICINE IN V AMERICA, 1776-1876. By E. H. Clarke, M.D., Late Prof, of Materia Medica in Harvard Univ. ; Henry J. BSgelow, M.D., Prof, of Surgery in Harvard Univ. ; Samuel D. Gross, M.D., D.C.L. Oxon., Prof, of Surgery in Jefferson Med. Coll., Philadn. ; T. Gaillard Thomas, Prof, of Obstetrics, etc., in Coll. of Phys. and Surgeons, N. Y. ; J. S. Billings, M.D., U.S.A., Librarian of National Medical Library, Washington. In one handsome royal 12mo. volume of 366 pages. Cloth, $2 25. PHRISTISON (ROBERT). DISPENSATORY OR COMMENTARY ON V THE PHARMACOPOEIAS OF GREAT BRITAIN AND THE UNITED STATES. With a Supplement by R. E. Griffith. In one 8vo. vol. of over 1000 pages, containing 213 illustrations. Cloth, $4. pHURCHILL (FLEETWOOB). ON THE THEORY AND PRACTICE U OF MIDWIFERY. With notes and additions by D. Francis Condie, M.D. With about 200 illustrations. In one handsome 8vo. vol. of nearly 700 pages. Cloth, $4 ; leather, $5. ESSAYS ON THE PUERPERAL FEVER, AND OTHER DIS- EASES PECULIAR TO WOMEN. In one neat octavo vol. of about 450 pages. Cloth, $2 50. pDNDIE (0. FRANCIS). A PRACTICAL TREATISE ON THE DIS- ^ EASES OF CHILDREN. Sixth edition, revised and enlarged. In one large 8vo. vol. of 800 pages. Cloth, $5 25 ; leather, $6 25. pULLERIER (1.) AN ATLAS OF VENEREAL DISEASES. Trans- ^ lated and edited by FREEMAN J. BUMSTE An, M.D. A large imperial quarto volume, with 26 plates containing about 150 figures, beauti- fully colored, many of them the size of life. In one vol., strongly bound in cloth, $17. Same work, in five parts, paper covers, for mailing, $3 per part. HENRY C. LEA'S PUBLICATIONS. CYCLOPEDIA OF PRACTICAL MEDICINE. By Dunglison, Forbes, *J Tweedie, and Conolly. In four large super-royal octavo volumes, of 3254 double-columned pages, leather, raised bands, $15. Cloth, $11. p.\MPBELL'S LIVES OF LORDS KENYON. ELLENBOROUGH, AND W TENTERDEN. Being the third volume of " Campbell's Lives of the Chief Justices of England." In one crown octavo vol. Cloth, $2, DA.LTON (J. C.) A TREATISE ON HUMAN PHYSIOLOGY. Sixth edition, thoroughly revised, and greatly enlarged and improved, with 316 illustrations. In one very handsome 8vo. vol. of 830 pp. Cloth, $5 50 ; leather, $6 50. (Just issued.) DAVIS (F. H.) LECTURES ON CLINICAL MEDICINE. Second edition, revised and enlarged. In one 12mo. vol. Cloth, $1 75. DON QUIXOTE DE LA MANCHA. Illustrated edition. In two hand- some vols. crown 8vo. Cloth, $2 50 ; half morocco. $3 70. DEWEES (W. P.) A TREATISE ON THE DISEASES OF FEMALES. With illustrations. In one 8vo. vol. of 536 pages. Cloth, $3. A TREATISE ON THE PHYSICAL AND MEDICAL TREAT- MENT OF CHILDREN. In one 8vo. vol. of 548 pages. Cloth, $2 80. DUTJITT (ROBERT). THE PRINCIPLES AND PRACTICE OF MO DERN SURGERY. A revised American, from the -eighth London edition. Illustrated with 432 wood engravings. In one 8vo. vol. of nearly 700 pages. Cloth, $4 ; leather, $5. HTINGLISON (UOBLEY) MEDICAL LEXICON; a Dictionary of -LJ Medical Science. Containing a concise explanation of the various subjects and terms of Anatomy, Physiology, Pathology, Hygiene, Therapeutics, Pharmacology, Pharmacy, Surgery, Obstetrics, Medical Jurisprudence, and Dentistry. Notices of Climate and of Mineral Waters; Formulae for Officinal, Empirical, and Dietetic Preparations, with the accentuation and Etymology of the Terms, and the French and other Synonymes. In one very large royal 8vo. vol. New edi- tion. Cloth, $6 50; leather, $7 50. (Just issued.) HUMAN PHYSIOLOGY. Eighth edition, thoroughly revised. In two large 8vo. vols. of about 1500 pp., with 532 illus. Cloth, $7. NEW REMEDIES, WITH FORMULAE FOR THEIR PREPARA- TION AND ADMINISTRATION. Seventh edition. In one very large 8vo. vol. of 770 pages. Cloth, $4. DE LA BECHE'S GEOLOGICAL OBSERVER. In one large 8vo. vol. of 700 pages, with 300 illustrations. Cloth, $4. DANA (JAMES D.) THE STRUCTURE AND CLASSIFICATION OF ZOOPHYTES. With illustrations on wood. In one imperial 4to. vol. Cloth, $4 00. PLLIS (BENJAMIN). THE MEDICAL FORMULARY. Being a " collection of prescriptions derived from the writings and practice of the most eminent physicians of America and Europe. Twelfth edi- tion, carefully revised by A. H. Smith, M. D. In one 8vo. volume of 374 pages. Cloth, $3. ERICHSEN (JOHN). THE SCIENCE AND ART OF SURGERY. A new and improved American, from the sixth enlarged and re- vised London edition. Illustrated with 630 engravings on wood. In two large 8vo. vols. Cloth, $9 00; leather, raised bands, $11 00. ENCYCLOPAEDIA OF GEOGRAPHY. In three large 8vo. vols. Illus- trated with 83 maps -and about 1100 wood-cuts. Cloth, $5. pOTHERGILL'S PRACTITIONER'S HANDBOOK OF TREATMENT. * In one handsome octavo volume. ( in press.) HENRY C. LEA'S PUBLICATIONS. 5 pENWICK (SAMUEL). THE STUDENTS' GUIDE TO MEDICAL J- DIAGNOSIS. From the Third Revised and Enlarged London Edi- tion. In one vol. royal 12rao. Cloth, $2 25. FLETCHER'S NOTES FROM NINEVEH, AND TRAVELS IN MESO. POTAMIA, ASSYRIA, AND SYRIA. In one 12mo. vol. Cloth, 75 cts. FOX ON DISEASES OF THE STOMACH. From the third London edi- tion. In one octavo vol. Cloth, $2. (Jiist issued.) pOX (TILBURY). EPITOME OF SKIN DISEASES, with Formula? * for Students and Practitioners. In one small 12mo. vol. Cloth, $1. PLINT (AUSTIN). A TREATISE ON THE PRINCIPLES AND L PRACTICE OF MEDICINE. Fourth edition, thoroughly revised and enlarged. In one large 8vo. volume of 1070 pages. Cloth, $6 ; leather, raised hands, $7. (Just issued.) A PRACTICAL TREATISE ON THE PHYSICAL EXPLORA- TION OF THE CHEST, AND THE DIAGNOSIS OF DISEASES AFFECTING THE RESPIRATORY ORGANS. Second and revised edition. One 8vo. vol. of 595 pages. Cloth, $4 50. A PRACTICAL TREATISE ON THE DIAGNOSIS AND TREAT- MENT OF DISEASES OF THE HEART. Second edition, enlarged. In one neat 8vo. vol. of over 500 pages, $4 00. ON PHTHISIS : ITS MORBID ANATOMY, ETIOLOGY, ETC., in a series of Clinical Lectures. A new work. In one handsome 8vo. volume. Cloth, $3 50. (Just issued.) A MANUAL OF PERCUSSION AND AUSCULTATION ; of the Physical Diagnosis of Diseases of the Lungs and Heart, and of Tho- racic Aneurism. In one handsome royal 12mo. volume. Cloth, $1 75. (Now ready.) MEDICAL ESSAYS. In one neat 12mo. volume. Cloth, $1 38. pOWNES (GEORGE). A MANUAL OF ELEMENTARY CHEMISTRY. *- From the tenth enlarged English edition. In one royal 12mo. vol. of 857 pages, with 197 illustrations. Cloth, $2 75 ; leather, $3 25. PULLER (HENRY). ON DISEASES OF THE LUNGS AND AIR -L PASSAGES. Their Pathology, Physical Diagnosis, Symptoms, and Treatment. From the second English edition. In one 8vo. vol. of about 500 pages. Cloth, $3 50. GALLOWAY (ROBERT). A MANUAL OF QUALITATIVE AN- ALYSIS. From the fifth English edition. In one 12mo. vol. Cloth, $2 50. (Lately published.) GLUGE (GOTTLIEB). ATLAS OF PATHOLOGICAL HISTOLOGY. Translated by Joseph Leidy, M.D., Professor of Anatomy in the University of Pennsylvania, &c. In one vol. imperial quarto, with 320 copperplate figures, plain and colored. Cloth, $4. GREEN (I. HENRY). AN INTRODUCTION TO PATHOLOGY AND MORBID ANATOMY. Second Arner., from the third Lond. Ed. In one handsome 8vo. vol., with numerous illustrations. Cloth, $2 75. (Just issued.) GRAY (HENRY). ANATOMY, DESCRIPTIVE AND SURGICAL. A new American, from the fifth and enlarged London edition. In one large imperial 8vo. vol. of about 900 pages, with 462 large and elabo- rate engravings on wood. Cloth, $6; leather, $7. (Lately issued.) GRIFFITH (ROBERT E.) A UNIVERSAL FORMULARY, CON- TAINING THE METHODS OF PREPARING AND ADMINISTER- ING OFFICINAL AND OTHER MEDICINES. Third and Enlarged edition. Edited by John M. Maisch. In one large 8vo. vol. of 800 pages, double columns. Cloth, $4 50 ; leather, $5 50. # HENRY C. LEA'S PUBLICATIONS. GROSS (SAMUEL D.) A SYSTEM OF SURGERY, PATHOLOGICAL, DIAGNOSTIC, THERAPEUTIC, AND OPERATIVE. Illustrated by 1403 engravings. Fifth edition, revised and improved. In two large imperial 8vo. vols. of over 2200 pages, strongly bound in leather, raised bands, $15. GROSS (SAMUEL D.) A PRACTICAL TREATISE ON THE DIS- eases, Injuries, and Malformations of the Urinary Bladder, the Pros- tate Gland, and the Urethra. Third Edition, 'thoroughly Revised and Condensed, by Samuel W. Gross, M.D., Surgeon to the Phila- delphia Hospital. In one handsome octavo volume, with about two hundred illustrations. Cloth, $4 50. (Now ready.) A PRACTICAL TREATISE ON FOREIGN BODIES IN THE AIR PASSAGES. In one 8vo. vol. of 468 pages. Cloth, $2 75. pIBSON'S INSTITUTES AND PRACTICE OF SURGERY. In two 8vo. vJ vols. of about 1000 pages, leather, $6 50. GOSFELIN(L) CLINICAL LECTURES ON SURGERY, Delivered at the Hospital of La Charite Translated from the French by Lewis A. Stirmon, M.D., Surgeon to the Presbyterian Hospital, New York. With illustrations. (Publishing in the Medical News and Library for 1876-7.) HUDSON (A..) LECTURES ON THE STUDY OF FEVER. 1 vol. 8vo., 316 pages. Cloth, $2 50. HEATH (CHRISTOPHER). PRACTICAL ANATOMY ; A MANUAL OF DISSECTIONS. With additions, by W. W. Keen, M. D. In 1 volume ; with 247 illustrations. Cloth, $3 50 ; leather, $4. HARTSHORNE (HENRY). ESSENTIALS OF THE PRINCIPLES AND PRACTICE OF MEDICINE. Fourth and revised edition. In one 12ino. vol. Cloth, $2 63; half bound, $2 88. (Lately issued ) CONSPECTUS OF THE MEDICAL SCIENCES. Comprising Manuals of Anatomy, Physiology, Chemistry, Materia Medica, Prac- tice of Medicine, Surgery, and Obstetrics. Second Edition. In one royal 12mo. volume of over 1000 pages, with 477 illustrations. Strongly bound in leather, $5 00 ; cloth, $4 25. (Lately issued.) A HANDBOOK OF ANATOMY AND PHYSIOLOGY. In one neat royal 12mo. volume, with many illustrations. Cloth, $1 75. HAMILTON (FRANK H.) A PRACTICAL TREATISE ON FRAC- TURES AND DISLOCATIONS. Fifth edition, carefully revised. In one handsome 8vo. vol. of 830 pages, with 344 illustrations. Cloth, $5 75 ; leather, $5 75. (Just issued.) HOLMES (TIMOTHY). SURGERY, ITS PRINCIPLES AND PRAC- TICE. In one handsome 8vo. volume of 1000 pages, with 411 illus- trations. Cloth, $6; leather, with raised bands, $7. ( Jit st ready.) HOBLYN (RICHARD D.) A DICTIONARY OF THE TERMS USED IN MEDICINE AND THE COLLATERAL SCIENCES. In one 12mo. volume, of over 500 double-columned pages. Cloth, $1 50 leather, $2. HODGE (HUGH L.) ON DISEASES PECULIAR TO WOMEN, IN CLUDING DISPLACEMENTS OF THE UTERUS. Second and revised edition. In one 8vo. volume. Cloth, $4 50. THE PRINCIPLES AND PRACTICE OF OBSTETRICS. Illus trated with large lithographic plates containing 159 figures from original photographs, and with numerous wood-cuts. In one large quarto vol. of 550 double-columned pages. Strongly bound in cloth $14. HENRY C. LEA'S PUBLICATIONS. H HOLLAND (SIR HENRY). MEDICAL NOTES AND REFLECTIONS. From the third English edition. In one 8vo. vol. of about 500 pages. Cloth, $3 50. HODGES (RICHARD M.) PRACTICAL DISSECTIONS. Second edi- tion. In one neat royal 12ino. vol., half bound, $2. HUGHES. SCRIPTURE GEOGRAPHY AND HISTORY, with 12 colored maps. In 1 vol. 12mo. Cloth, $1. TT3RNER (WILLIAM E.) SPECIAL ANATOMY AND HISTOLOGY. -*-L Eighth edition, revised and modified. In two large 8vo. vols. of over 1000 pages, containing 300 wood-cuts. Cloth, $6. HILL (BERKELEY). SYPHILIS AND LOCAL CONTAGIOUS DIS- ORDERS. In one 8vo. volume of 467 pages. Cloth, $3 25. SILLIER (THOMAS). HAND-BOOK OF SKIN DISEASES. Second Edition. In one neat royal 12mo. volume of about 300 pp. , with two plates. Cloth, $2 25 ALL (MRS. M.) LIVES OF THE QUEENS OF ENGLAND BEFORE THE NORMAN CONQUEST. In one handsome 8vo. vol. Cloth, $2 25; crimson cloth, $2 50; half morocco, $3. JONES (C. HANDFIELD). CLINICAL OBSERVATIONS ON FUNC- TIONAL NERVOUS DISORDERS. Second American Edition. In one 8vo. vol. of 348 pages. Cloth, $3 25. EIRKES (WILLIAM SENHOUSE). A MANUAL OF PHYSIOLOGY. A new American, from the eighth London edition. One vol., with many illus., 12mo. Cloth, $3 25; leather, $3 75. KNAPP (F.) TECHNOLOGY ; OR CHEMISTRY, APPLIED TO THE ARTS AND TO MANUFACTURES, with American additions, by Prof. Walter R. Johnson. In two 8vo. vols., with 500 ill. Cloth, $6. KENNEDY'S MEMOIRS OF THE LIFE OF WILLIAM WIRT. In two vols. 12mo. Cloth, $2. T EA (HENRY C.) SUPERSTITION AND FORCE ; ESSAYS ON THE -Ll WAGER OF LAW, THE WAGER OF BATTLE, THE ORDEAL, AND TORTURE. Second edition, revised. In one handsome royal 12mo. vol., $2 75. STUDIES IN CHURCH HISTORY. The Rise of the Temporal Power Benefit of Clergy Excommunication. In one handsome 12mo. vol. of 515 pp. Cloth, $2 75. AN HISTORICAL SKETCH OF SACERDOTAL CELIBACY IN THE CHRISTIAN CHURCH. In one handsome octavo volume of 602 pages. Cloth, $3 75. LA ROCHE (R.) YELLOW FEVER. In two 8vo. vols. of nearly 1500 pages. Cloth, $7. PNEUMONIA. In one 8vo. vol. of 500 pages. Cloth, $3. T INCOLN (D. F.) ELECTRO-THERAPEUTICS. A Condensed Man- -1 ual of Medical Electricity. In one neat royal 12mo. volume, with illustrations. Cloth, $1 50. (Just issued.) TEISHMAN (WILLIAM). A SYSTEM OF MIDWIFERY. Includ- Ll ing the Diseases of Pregnancy and the Puerperal State. Second American, from the Second English Edition. With ndditions, by J. S. Parry, M.D. In one very handsome 8vo. vol. of 800 pages and 200 illustrations. Cloth, $5 ; leather, $6. (Just issued.) r AURENCE (J. Z.) AND MOON (ROBERT C.) A HANDY-BOOK LI OF OPHTHALMIC SURGERY. Second edition, revised by Mr. Laurence. With numerous illus. In one 8vo. vol. Cloth, $2 75. HENRY C. LEA'S PUBLICATIONS. T EHMANN (C. G.) PHYSIOLOGICAL CHEMISTRY. Translated by LI George F. Day, M. D. With plates, and nearly 200 illustrations. In two large 8vo. vols., containing 1200 pages. Cloth, $6. A MANUAL OF CHEMICAL PHYSIOLOGY. In one very handsome 8vo. vol. of 336 pages. Cloth, $2 25. T AWSON (GEORGE). INJURIES OF THE EYE, ORBIT, AND EYE- Ll LIDS, with about 100 illustrations. From the last English edition. In one handsome 8vo. vol. Cloth, $3 50. T UDLOW (J. I.) A MANUAL OF EXAMINATIONS UPON ANA- -Ll TOMY, PHYSIOLOGY, SURGERY, PRACTICE OF MEDICINE, OBSTETRICS, MATERIA MEDICA, CHEMISTRY, PHARMACY, AND THERAPEUTICS. To which is added a Medical Formulary. Third edition. In one royal 12mo. vol. of over 800 pages. Cloth $3 25 ; leather, $3 75. TAYCOCK (THOMAS). LECTURES ON THE PRINCIPLES AND LI METHODS OF MEDICAL OBSERVATION AND RESEARCH. In one 12mo. vol. Cloth, $1. T YNCH (W. F.) A NARRATIVE OF THE UNITED STATES EX- J-l PEDITION TO THE DEAD SEA AND RIVER JORDAN. In one large octavo vol., with 28 beautiful plates and two maps. Cloth, $3. Same Work, condensed edition. One vol. royal 12mo. Cloth, $1. T EE (HENRY) ON SYPHILIS. In one 8vo. vol. Cloth, $2 25. T YONS (ROBERT D.) A TREATISE ON FEVER. In one neat 8vo. -LI vol. of 362 pages. Cloth, $2 25. MARSHALL (JOHN). OUTLINES OF PHYSIOLOGY, HUMAN AND COMPARATIVE. With Additions by FRANCIS G. SMITH, M. D., Professor of the Institutes of Medicine in the University of Pennsylvania. In one 8vo. volume of 1026 pages, with 122 illustra- tions. Strongly bound in leather, raised bands, $7 50. Cloth, $6 50. MACLISE (JOSEPH). SURGICAL ANATOMY. In one large im- perial quarto vol., with 68 splendid plates, beautifully colored; con- taining 190 figures, many of them life size. Cloth, $14. MfiIGS (CHAS. D.). ON THE NATURE, SIGNS, AND TREATMENT OF CHILDBED FEVER. In one 8vo. vol. of 365 pages. Cloth, $2. TV/TILLER (JAMES). PRINCIPLES OF SURGERY. Fourth American, 1YL from the third Edinburgh edition. In one large 8vo. vol. of 700 pages, with 240 illustrations. Cloth, $3 75. THE PRACTICE OF SURGERY. Fourth American, from the last Edinburgh edition. In one large 8vo. vol. of 700 pages, with 364 illustrations. Cloth, $3 75. MONTGOMERY (W. F.) AN EXPOSITION OF THE SIGNS AND SYMPTOMS OF PREGNANCY. From the second English edition. In one handsome 8vo. vol. of nearly 600 pages. Cloth, $3 75. MULLER (J.) PRINCIPLES OF PHYSICS AND METEOROLOGY. In one large 8vo. vol. with 550 wood-cuts, and two colored plates. Cloth, $4 50. TUTIRA.BEAU ; A LIFE HISTORY. In one 12mo. vol. Cloth, 75 cts. MACFARLAND'S TURKEY AND ITS DESTINY. In 2 rols. royal 12mo. Cloth, $2. MARSH (MRS.) A HISTORY OF THE PROTESTANT REFORMA- TION IN FRANCE. In 2 vols. royal 12mo. Cloth, $2. M-ELIGAN (J. MOORE) . AN ATLAS OF CUTANEOUS DISEASES. In J-' one quarto volume, with beautifully colored plates, &c. Cloth, $5 50. HENRY C. LEA'S PUBLICATIONS. 9 TVTEILL (JOHN) AND SMITH (FRANCIS G.) COMPENDIUM OF IN THE VARIOUS BRANCHES OF MEDICAL SCIENCE. In one handsome 12mo. vol. of about 1000 pages, with 374 wood-cuts. Cloth, $4; leather, raised bands, $4 75. NIEBTJHR (B. G.) LECTURES ON ANCIENT HISTORY; com- prising the history of the Asiatic Nations, the Egyptians, Greeks, Macedonians, and Carthagenians. Translated by Dr. L. Schmitz. In three neat volumes, crown octavo. Cloth, $500. ODLING (WILLIAM). A COURSE OF PRACTICAL CHEMISTRY FOR THE USE OF MEDICAL STUDENTS. In one 12mo. vol. of 261 pp., with 75 illustrations. Cloth, $2. PLAYFAIS (W. S ) A TREATISE ON THE SCIENCE AND PRAC- TICE OF MIDWIFERY. In one handsome octavo vol. of 576 pp., with 166 illustrations, and two plates. Cloth, $4; leather, $5. (Jttst issued.) PAVY (F. W.) A TREATISE ON THE FUNCTION OF DIGESTION, ITS DISORDERS AND THEIR TREATMENT. From the second London ed. In one 8vo. vol. of 246 pp. ^Cloth, $2. A TREATISE ON FOOD AND DIETETICS, PHYSIOLOGI- CALLY AND THERAPEUTICALLY CONSIDERED. In one neat octavo volume of about 500 pages. Cloth, $4 75. (Just issued.) p.\RRISH (EDWARD). A TREATISE ON PHARMACY. With many L Formulae and Prescriptions. Fourth edition. Enlarged and thoroughly revised by Thomas S. Wiegand. In one handsome 8vo. vol. of 977 pages, with 280 illus. Cloth, $5 50 ; leather, $6 50. pIRRIE (WILLIAM) THE PRINCIPLES AND PRACTICE OF SUR- -L GERY. In one handsome octavo volume of 780 pages, with 316 illustrations. Cloth, $3 75. pEREIRA (JONATHAN). MATERIA MEDIC A AND THERAPEU- L TICS. An abridged edition. With numerous additions and refe- rences to the United States Pharmacopoeia. By Horatio C. Wood, M. D. In one large octavo volume, of 1040 pages, with 236 illustra- tions. Cloth, $7 00; leather, raised bands, $8 00. PTJLSZKY'S MEMOIRS OF AN HUNGARIAN LADY. In one neat royal 12mo. vol. Cloth, $1. PAGET'S HUNGARY AND TRANSYLVANIA. In two royal 12mo. -L vols. Cloth, $2. TDEMSEN (IRA). THE PRINCIPLES OF CHEMISTRY. An Intro- t ductior- to Modern Chemistry, for the Use of Students. In one 12mo. vol., cloth. (In press.) ROBERTS (WILLIAM). A PRACTICAL TREATISE ON URINARY AND RENAL DISEASES. A second American, from the second London edition. With numerous illustrations and a colored plate. In one very handsome 8vo. vol. of 616 pages. Cloth, $4 51. RA.MSBOTHAM (FRANCIS H.) THE PRINCIPLES AND PRAC- Iw TICE OF OBSTETRIC MEDICINE AND SURGERY. In one im- perial 8vo. vol. of 650 pages, with 64 plates, besides numerous wood- cuts in the text. Strongly bound in leather, $7. RIGBY (EDWARD). A SYSTEM OF MIDWIFERY. Second Ameri. can edition. In one handsome 8vo. vol. of 422 pages. Cloth, $2 50. RANKE'S HISTORY OF THE TURKISH AND SPANISH EMPIRES in the 16th and beginning of 17th Century. In one 8vo. volume, paper, 25 cts. HISTORY OF THE REFORMATION IN GERMANY. Parts I., II., III. In one vol. Cloth, $1. 10 HENRY C. LEA'S PUBLICATIONS. qCHAFER (EDWAKD ALBERT) A COURSE OF PRACTICAL HIS- O TOLOGF : A Manual of the Microscope for Medical Students. In one handsome octavo vol. With numerous illustrations. (In press.) SMITH (EUSTACE). ON THE WASTING DISEASES OF CHILDREN Second American edition, enlarged. In one 8vo. vol. Cloth, $2 50. OARGENT (F. W.) ON BANDAGING AND OTHER OPERATIONS O OF MINOR SURGERY. New edition, with an additional chapter on Military Surgery. In one handsome royal 12mo. vol. of nearly 400 pages, with 184 wood-cuts. Cloth, $1 75. QMITH (J. LEWIS.) A TREATISE ON THE DISEASES OF IN- W FANCY AND CHILDHOOD. Third Edition, revised and enlarged. In one large 8vo. volume of 724 pages, with illustrations. Cloth, $5 ; leather, $6. (Just issued.) QHARPEY (WILLIAM) AND QTJAIN (JONES AND RICHARD) . O HUMAN ANATOMY. With notes and additions by Jos. Leidy, M.D., Prof, of Anatomy in the University of Pennsylvania. In two large Svo.vols. of about J 300 pages, with 51 1 illustrations. Cloth, $6. SKEY (FREDERIC C.) OPERATIVE SURGERY. In one 8vo. vol. of over 650 pages, with about 100 wood-cuts. Cloth, $3 25. SLADE (D. D.) DIPHTHERIA ; ITS NATURE AND TREATMENT. Second edition. In one neat royal 12mo. vol. Cloth, $1 25. SMITH (HENRY H.) AND HORNER (WILLIAM E.) ANATOMICAL ATLAS. Illustrative of the structure of the Human Body. In one large imperial 8vo. vol., with about 650 beautiful figures. Cloth, $4 50. SMITH (EDWARD). CONSUMPTION; ITS EARLY AND REME- DIABLE STAGES. In one 8vo. vol. of 254 pp. Cloth, $2 25. STILLE (ALFRED). THERAPEUTICS AND MATERIA MEDIC A. Fourth edition, revised and enlarged. In two large and handsome volumes 8vo. Cloth, $10 ; leather, $12. (Jitst issued.) SCHMITZ AND ZUMPT'S CLASSICAL SERIES. In royal 18mo. CORNELII NEPOTIS LIBER DE EXCELLENTIBUS DUCIBUS EXTERARUM GENTIUM, CUM VITIS CATONIS ET ATTICI. With notes, Ac. Price in cloth, 60 cents; half bound, 70 cts. C. I. CJESARIS COMMENTARII DE BELLO GALLICO. With notes, map, and other illustrations. Cloth, 60 cents; half bound, 70 cents. C. C. SALLUSTII DE BELLO CATILINARIO ET JUGURTHINO. With notes, map, Ac. Price in cloth, 60 cents ; half bound, 70 cents. Q. CURTII RUFII DE GESTIS ALEXANDRI MAGNI LIBRI VIII. With notes, map, Ac. Price in cloth, 80 cents ; half bound, 90 cents. P. VIRGILII MARONIS CARMINA OMNIA. Price in cloth, 85 cents; half bound, $1. M. T. CICERONIS ORATIONES SELECTJE XII. With notes, Ac. Price in cloth, 70 cents; half bound, 80 cents. ECLOGJE EX Q. HORATII FLACCI POEMATIBUS. With notes,