64

discourse at Neuchatel, a sufficient answer to the claims that Guyot made a scientific examination of the Alpine glaciers two years before they w studied by Agassiz."

were

On the 5th of September, Agassiz and Guyot were present at the Réunion extraordinaire de la Société géologique de France à Porrentruy; and at the meeting of the 6th of September we read the following remarks:

"M. Agassiz présente à la société ses observations sur les glaciers, d'où il déduit d'importantes conséquences géogéniques relativement aux blocs erratiques. . M. Guyot ajoute aux observations de M. Agassiz de nouvelles considérations" (Bull. soc. geol, vol. ix. p. 407).

That is all. Guyot did not read a manuscript, but offered only a few verbal observations. He was not then a member of the society; and his remarks passed off unnoticed, although geologists were present, well prepared to discuss any point relating to glaciers, Agassiz, Jean de Charpentier, Bernard Studer, Thurmann, Max Braun, Lardy, Buckland, d'Omalius, Nicolet, and finally Renoir and Leblanc, who announced at that meeting their discoveries of old glaciers in the Vosges.

On the contrary. Agassiz's communication attracted much attention, and was the subject of many discussions and commentaries. Agassiz, strengthened and animated by the presence of de Charpentier, surpassed himself in his clear and trenchant exposition of the glacial theory.' The impression left on all those who were present at the Porrentruy meeting was such, that years after, several of them told me that Agassiz was absolutely irresistible, and won the admiration even of his strongest opponent there, Bernard Studer.

Neither Agassiz nor Guyot gave their notes to be printed; and it was almost one year later that Agassiz's memoir, Sur les glaciers,' was deposited at the 'secrétariat' of the Geological society at Paris. It was published at the end of volume ix. p. 413, as late as the spring of 1840. The same memoir appeared first in the Bibliothèque univ. de Genêve (tome xx. p. 382) in December, 1839; and it was reprinted in 1844, at the head of Excursions et séjours dans les glaciers,' etc., by E. Desor.

Many years after the death of Agassiz, and one year after the death of Desor, Professor Guyot claimed that he wrote Agassiz's memoir, and added that he was unable to finish the writing of his own memoir by an indisposition qui dura jusque tard dans l'été (1839).' Guyot returned to Neuchatel, however, in good health, in the fall of 1839; and, if his memoir remained inédit, it was because he did not think his maiden notice was of sufficient value for publication; for both the Bulletin of the geological society and the Bibliothèque universelle were open to him, and ready to accept his remarks.

James D. Forbes having claimed the discovery 'of ribboned structure' of the ice of glaciers, Agassiz took from Guyot's notes his remarks, "sur la structure lamellaire de la glace du glacier près du sommet du Gries," and published them in a pamphlet dated 11 April, 1842, Neuchatel. At the same time Agassiz begged Guyot to put his manuscript in the 'archives' of the Société des sciences naturelles de Neuchatel. This was done, and from that date the record of the existence of Guyot's notes is indisputable. Unhappily they were not published; and Guyot took them back in 1848, and carried them to America, whence,

in April, 1883, he sent them again to Neuchatel, where they were finally printed in the Bulletin Soc. sc. naturelles (tome xiii. p. 156), the 26th of April, 1883.

It is impossible not to feel an uncertainty as to the primordial communication of Professor Guyot at Porrentruy, when we think of the delays in its publication, the travelling about, and the incompleteness of the notes. This feeling is increased by a remark of his widow, who says that Guyot did not send back to Neuchatel all the original mauuscript, a part having been left in her hands (The American journal of science, May, 1886, p. 366).

But accepting the Neuchatel memoir of 1883 as correct, its scientific value is very small, and hardly justifies its publication. All that was truly of value was put in Agassiz's reply to Forbes; and even that is of small importance, considering that Rendu noticed more in detail the same phenomenon of veined structure of the ice, in his Théorie des glaciers de la Savoie,' published during the summer of 1840; and that Hugi, as far back as 1830, signalized the same phenomenon.

Accompanying his notes by a letter to M. Louis Coulon, president of the Neuchatel society, Professor Guyot claims that he has discovered not only la structure lamellaire de la glace des glaciers,' but also the different modes of progression of the glaciers, the inclination of the beds at the end of glaciers, and the disposition of crevasses en éventail.'

These facts were known before, and were discussed almost daily in the house of de Charpentier, as is proved in the book of de Charpentier on the glaciers. Besides, Grüner, Hugi. Rendu, Bischof, and others have previously signalized the same facts.

Finally, Prof. Guyot, at the end of his letter to M. Coulon, makes statements entirely at variance with fact in regard to ‘la distribution des blocs erratiques.' For instance, he says, "The erratic map of the old glacier of the Rhone, published by de Charpentier (1840), stops it at Nyon, when by my latter observations I extended it far beyond Geneva to the Mont de Sion." Now, de Charpentier's map 'du terrain erratique de la vallée du Rhône,' accompanying his celebrated book, does not stop the glacier of the Rhone at Nyon, but elose to the city of Geneva, twenty miles farther south. As to bowlders of the Rhone valley as far as Mont de Sion, they have been described there by J. A. Deluc anterior to 1840; and R. Blanchet, in his 'Carte du glacier du Rhône' (Lausanne, 1844), extends the Rhone glacier as far as la Perte du Rhône, with a large moraine on the Mont de Sion.

From 1840 to 1847, Guyot, with great industry and perseverance, made a hypsometrical survey of the positions of the bowlders in seven of the erratic basins round the central Alps. Unhappily he only partially published his researches, in the Bulletin des sc. nat. de Neuchatel, without the map showing the distribution of those bowlders; reserving it, as he says, for an ulterior publication, in collaboration with Agassiz and Desor, which was never completed. If Guyot's map had been published then, it would have been an important contribution to the Alpine erratic phenomena. However, a great part of it- more than two-thirds at least was anticipated by the issue in 1845, at Winterthur, of an anonymous map of the old glaciers of the central Alps, showing the extent of the ancient glaciers of the Arve, Rhone, Aar, Reuss, Linth, and Rhine, with their lateral and

frontal moraines. That map is entitled 'Verbreitungsweise der Alpen-fündlinge,' and its author is the modest and very able geologist, A. Escher von der Linth.

Since 1850, Gastaldi for Piemont, Chantre and Falsan for France, and A. Favre for Switzerland, have given maps of the ancient extension of the Alpine glaciers, which render Guyot's manuscript map obsolete and valueless, except as an historical docu

ment.

To finish this already too long review of glaciers and glacialists, I will add, that, after the three original memoirs of Venetz, de Charpentier, and Agassiz, of 1833, 1834, and 1837, the other important works and landmarks in the discoveries and exposition of the glacial question are, by order of data, 1, Théorie des glaciers de la Savoie,' by the Chanoine Rendu (September, 1840); of this most important and excellent work, Tyndall said to me at the Geneva meeting of the Swiss naturalists in 1865, "If Rendu had been trained and educated as a physicist, he would have left nothing for others to do;" 2°, 'Etudes sur les glaciers,' by Louis Agassiz (October, 1840); 3°. 'Essai sur les glaciers,' by Jean de Charpentier (Oct. 31, 1840; issued in December, 1840, with the date on the titlepage of 1841); 4°, 'Travels through the Alps of Savoy,' by James D. Forbes (1843; second edition, 1845); 5°, Nouvelles études et expériences sur les glaciers actuels,' by Louis Agassiz (November, 1847); 6. The glaciers of the Alps,' by John Tyndall (1860).

Venetz was personally known to but few savants. I will add that he was a Valaisan engineer of great skill. He had the charge of rectifying and embanking the Rhone in the cantons of Valais and Vaud, from Sion and Martigny to the lake of Geneva, — works which he executed most successfully. Accustomed to observe all that relates to the freshets of mountain torrents and glaciers, a spectator of the great 'débacle de Bagnes' in 1818, he and his friend de Charpentier put a stop to the constant ravages of the Getroz glacier and the Dranse River, an affluent of the Rhone.

Venetz's modesty was extreme, and verging on great timidity, due perhaps, in part, to the infirmity so common in the Valais, and from which he was a sufferer. Not educated as a scientific man, but only as a road engineer, he did not possess the scientific method of marshalling and classifying facts and ob servations. But Venetz found in his friend de Charpentier the best man to systematize and construct a new science. In that respect de Charpentier, by his knowledge and education, was the equal and rival of his friends Alex. de Humboldt, Leopold de Buch, and Elie de Beaumont; and the association of Venetz with him was most happy and successful. Both without ambition, lovers of nature and truth, they created together what may be called now one of the most interesting branches of geology and physical geography. JULES MARCOU. Cambridge, Mass., July 7.

Barometer exposure.

It is gratifying to find that my brief letter calling in question the influence of wind on the indications of indoor barometers has elicited very satisfactory responses from Messrs. Gilbert and Clayton (Science, vol. vii. pp. 571, 572; and vol. viii. p. 14). There is one point, however, on which evidence is still wanting to fortify Mr. Clayton's induction.

As clearly indicated by Mr. Gilbert, it is evident, that, according to the conditions of exposure. the influence of the wind must tend sometimes to increase, and at other times to diminish, the pressure within the building in which the barometer is placed. Now, all of Mr. Clayton's experiments seem to indicate a lowering of the barometer-readings within the building. Perhaps he may be able to verify the deductions of theory by so arranging the conditions of exposure as to secure the opposite effect, and thus obtain a complete verification of his induction. If these opposite effects can be verified by experiment, while establishing the influence of wind as a true cause of barometric fluctuations, they would render it extremely difficult to apply a correction correlated with the velocity of the wind, except under welldefined conditions of exposure.

While seeking for possible causes of fluctuations of the barometric column in relation to wind-velocity, it may be well to recall the idea first broached by Hawksbee near the beginning of the last century, and more distinctly urged by Sir John Leslie, that the barometer is depressed by wind in consequence of the centrifugal force due to the horizontal current of air (Daniell's Elements of meteorology,' vol. i. pp. 4-9, London, 1845); for although Professor Daniell's criticism of Professor Leslie's theory is quite just. in so far as it relates to the idea that the effect would be accumulated by a long series of deflections,' yet the main fact, that the tendency to rectilinear motion would give rise to a centrifugal effect, remains a vera causa tending to depress the mercurial column.

A simple calculation shows, however, that the radius of curvature is so large, or the deflection from a tangent is so small, that a horizontal wind of 60 miles per hour, or 88 feet per second (assuming the whole thickness of the atmosphere to be involved), would lower the mercury in the barometric column only about 0.00875 of a millimetre, or 0.00034 of an inch, an amount so small as to be far within the limits of observational error, and therefore quite inadequate as an explanation of the phenomenon. JOHN LECONTE. Berkeley, Cal., July 18.

Bright lines in the spectrum of ẞ Lyrae.

A short study of the spectrum of 8 Lyrae presents the following bright lines as existing in her atmosphere. A portion are probably also found in the solar atmosphere. Referred to by their numbers in Young's catalogue, they are, 2, 3, 5, 22, 36, 41, 49, (58-59), 69, 74, 86, 100. (105-106), 115, (138-139), (140-141), 181, 189, 193, 198, 208, 248, (260-261), 267, (272-273 ?). Another portion find no place, or are infrequent, in the solar atmosphere, and, referred to by their approximate wave-lengths, are 59549, 58398. 57967, 57544, 56305. 55829, 54811, 51355, 51013, 50858, 50582, 49582, 47939, 47660, 47437, 46879, 45203, 43123.

Each of these appear in at least 40 per cent of the observations; none appear in more than 70 per cent. A number more are suspected, but are not clearly separated.

At present there would seem to be a connection between the variability of the star and the lines present in the spectrum; but on this point the observations are not final. O. T. S.

New Haven, July 17.

FRIDAY, JULY 23, 1886.

over those disciples who by too close and dogmatic imitation had obscured the work of the fathers. Some of the keenest minds among students of

CAL ECONOMY.

"DURING the last thirty years," says Sidgwick in his Principles of political economy,' "political economy has risen from the state of controversy on fundamental principles and method into that of an apparently established science, and again relapsed into the state of controversy." This statement is borne out by an examination of the literature of political economy during these years. It is full of controversy. Not only do writers fail to agree on practical economic questions, such as free trade and protection, mono- or bi-metallism, direct or indirect taxation, but they quarrel over the fundamental principles which are to be taken as the basis for the solution of these problems. We have the doctrine of laissez-faire on the one side, and of social expediency on the other. To some, economics is merely a science of wealth; to others, it is eminently social; and to still others, it is, in addition, ethical. Some stick to the principle of self-interest as the only one worth regarding; others take into account all the motives which influence economic action. Some seek for principles which shall be strictly true of an abstract economic man,' and then push all practical problems into an 'art' of political economy; while others desire principles that can be directly and usefully applied to existing human society, taking into consideration time, place, and circumstance.

It would be too much to say that this controversy over principles is at all ended. The conception of pure laissez-faire has, indeed, lost its position, and will probably never be reinstated; but the advocates of new and more liberal principles have not been able to agree among themselves. Some of them are nationalist, some socialist, some ethical; while they differ infinitely in the degree to which they still cling to the old ideas and the old formulae.

In regard to method as distinct from principles, on the other hand, we are beginning to see some light through the darkness. Men can acknowledge a change in method without giving up the validity of principles which they wish to maintain. Here the triumph of the new over the old has been complete; or rather there has been a vindication of the method of the master-minds

owing to the efforts of such men as Knies, Wagner, Leslie, Jevons, and Ingram, we are reaching a substantial unanimity on the question

of method.

How important this change is, and how fruitful of result it is going to be, will appear if we consider for a moment the difference between the old method and the new. Without going into the finer questions, and without being too exact in our definition, we may call the old method the deductive, and the new method the inductive. These terms will cover the other designations, such as a priori,' 'abstract, 'philosophical,' sometimes applied to the old method; and similar terms, such as 'realistic,' 'historical,' and 'practical,' applied to the new.

6

The old method is essentially deductive. It finds certain premises which are true, and reasons from these premises to the solution of specific problems. These premises, as laid down by Cairnes, the most brilliant expounder of this view, and summarized by Cossa ('Guide to political economy,' p. 38), are as follows:

"1. In the economic order of things the principal motive of human actions is individual self-interest. This induces man (a) to avoid pain (fatigue, work); (b) to desire pleasure (wealth); (c) hence to aim at obtaining the greatest amount of wealth with the least amount of labor, or, in more general terms, the greatest result with the least effort, which is, as it is now expressed, the law of least resistance.

"2. The earth, indispensable to man as a place in which to live and work, and as the source whence he may extract food and raw materials, is naturally limited (a) in the products which it contains; (b) in its actual extent; (c) in its relative fertility (different qualities of soil); (d) in its successive fertility (decreasing productiveness at a certain point with every new application of capital and labor).

"3. The physical and psychological tendencies of man lead him to multiply his own species with a rapidity which, if it met with no obstacles, would bring about an unlimited increase of population."

From these premises are deduced the three great theories of value, rent, and population; and by means of these theories concrete problems, such as free trade and protection, are solved.

It is not necessary here to describe how this deductive method of political economy has been overthrown. These assumed premises, although containing an element of truth, were in themselves incomplete and sometimes inapplicable. For instance, it is a matter of experience that men are actuated by other motives than self-interest, such

as patriotism, charity, and custom. Again, common sense revolted against the assumption that these theories were universal and perpetual; that is, true everywhere and at all times. Experience showed that at different epochs in civilization, and among differently situated nations at the present time, the premises would require very great modifications.

The new method in political economy is inductive; that is, it proceeds from observation of facts to general rules and principles. It carefully observes the limits of time and place, and abstains from asserting its principles to be either universal or perpetual. It makes use of what knowledge we have of man and nature; but it uses this knowledge for the purpose of guiding and helping its investigations, not as a priori premises. It studies history for the purpose of discovering what blunders men and nations have made in their economic experience, and how those blunders may be avoided in the future. The inductive method is also comparative; that is, it compares economic institutions performing the same function among different nations of the same degree of civilization, in order to discover which is the best. The method is, finally, statistical; that is, it collects statistical data as a basis for its knowledge, in order to measure economic forces and gauge the results of economic action. The present method of political economy as recognized by the greatest modern economists, such as Wagner, Schmoller, Leslie, Jevons, Marshall, etc., is historical, comparative, and statistical.

I do not propose to defend this new method against the old, much less to vindicate it. Neither do I deny that the old method has had able representatives, and that in its time it has done good service. All I assert is, that it is now practically abandoned as a method by itself, and that the future of political economy depends upon the scientific application of the new method to the complex phenomena of modern civilization.

It will be useful, however, to describe more fully how the new method is actually applied, what sort of results it is able to give us, and some of the advantages which flow from its use. I propose, therefore, to discuss, 1°, how to investigate particular economic problems; 2°, how to reach general principles of economic life; 3°, what are the collateral advantages of this method; and, 4°, how to make method and results useful in the study of other social sciences and in guiding state action in economic affairs.

How to investigate particular economic problems. Every reader of John Stuart Mill will remember the opening paragraph of his Principles of

political economy:' "In every department of human affairs, practice long precedes science; systematic inquiry into the modes of action of the powers of nature is the tardy product of a long course of efforts to use those powers for practical ends. The conception, accordingly, of political economy as a branch of science, is extremely modern; but the subject with which its inquiries are conversant has in all ages necessarily constituted one of the chief practical interests of mankind, and, in some, a most unduly engrossing one."

In the same way it might be said that the solution of economic problems precedes the formulation of an economic science. Mankind has always had its economic problems, and philosophic heads have ever busied themselves trying to solve them. The method of doing this is both of very great importance in itself, and indicative of the character of the science which will by and by be formulated on the basis of this method. It will be of interest, therefore, to show how the inductive method of political economy attacks practical economic problems, and to see what sort of a science results from this method. In choosing my illustrations, I have purposely selected modern economic questions, and American and English authors, in order to escape the common slur that this method is fitted only for the antiquarian, and used only by learned but unpractical and idealistic German professors.

Mr. Sidgwick has remarked, that, in that portion of political economy dealing with the production of wealth, the inductive and analytical method has been much more used than in those portions dealing with exchange and distribution. Take, for instance, the question of land-tenure, one which has interested political economy for a long time, and which is to-day one of the burning political questions in England. It is apparent at a glance that the method of holding land must have a great influence on its productiveness. We can even reason a priori that where there is absolute proprietorship on the part of the cultivator, or at least a long leasehold which will secure to him the reward of his labor, he will be apt to work harder, and that the gross produce will thereby be increased. But the English economists, even Mill, Thornton, and Fawcett, have approached the subject in a different way. They have studied the condition of the French and Belgian peasants where absolute ownership exists, and have pointed out the prosperous condition of these countries as the proof that peasant proprietorship is the best system. This is the pure comparative method in political economy.

Let us take a more specific question. The issue

of bank-notes is a useful and at the same time dangerous function to intrust to a bank. Shall the issue of bank-notes be free, or shall it be regulated by government? How shall we answer such a question? If we examine the history of banking in the United States, as President Walker does in his book on money, or as Comptroller Knox did in his report for 1876, we shall find that freedom of issue has always been abused, and has always led to disaster, and that the only good bank money we have ever had in this country has been the national bank-notes secured by United States bonds. Study of the experience of England, Germany, and France will show that the liberty to issue bank-notes has everywhere been restricted, and is now exercised only by institutions under the direct or indirect control of the state. It can therefore be accepted as a rule that the privilege of issuing bank-notes should be carefully regulated by the state. This is the pure historical method in political economy.

Let us take a question which has not yet been solved, or where, at any rate, no practical solution has been reached by the legislature. Let us take, for example, the present silver question in the United States. Should the United States try to re-establish the silver dollar as a standard? There are two questions here. One is the question of the single or the double standard; the other is whether we can dispense with either one of the precious metals as money. The first, which is commonly known as bimetallism, although it is more properly the question of the single or the double standard, is already settled in the opinion of the best economists. One has only to read Professor Laughlin's book on the history of bimetallism to see that the double standard has been thoroughly tried in the United States from 1790 to 1873, and that it has signally failed. always results in the presence of one metal and the absence of the other. At first, with a ratio of one to fifteen, we could keep no gold in the country afterwards, with the ratio of one to sixteen, we could keep no silver. The history of France proves exactly the same thing, so that even professed bimetallists acknowledge that the double standard cannot be maintained except by international agreement. This, again, is the historical method.

It

The second part of the problem - viz., is there sufficient gold in the world to supply the demand for money, so that it is safe to demonetize silver? --is much more difficult to answer, and is, I venture to say, as yet unanswered. It can be solved only by the statistical method; viz., by showing that prices are declining, while at the same time the supply of gold is decreasing, and that

the latter is the only adequate cause discoverable for the former phenomenon. As an example of an attempt to prove this connection, I may cite Mr. Giffen's well-known 'Essays in finance.' An even more noted example of the same style of applying the statistical method to economic problems may be found in the essay of Jevons, and also those of Cliffe Leslie on the effect of the gold discoveries in California and Australia on prices in Europe.

Finally, we may ask, what can the inductive method do when it faces some great economic problem which affects the whole community and civilization itself? Such a problem is the laborproblem. What is the condition of the laboring class? Has that condition deteriorated or improved? The inductive method has not shrunk from attempting to find an answer to even such questions as these. Thorold Rogers has laboriously traced the condition of the English laborer during the last six centuries, for the purpose of answering this question historically. Giffen has attempted, by statistics, to show that the condition of the laboring class has materially improved during the last fifty years.

These are examples of the historical, comparative, and statistical method applied to modern economic problems. In some cases the method has only confirmed what was known or at least surmised before; in most cases it has added directly to our knowledge; in a few cases it has given us results which could have been obtained in no other way. Such is the value of the method in these isolated cases. Can it be so utilized as to enable us to formulate a body of truth worthy to be called a science? This brings us to our second point,

How to reach principles of economic life.

It is often said, that, although the inductive method may aid us in solving economic problems, it falls far short of what is required by a true science, because it does not enable us to formulate a body of principles which shall at the same time embody the highest truth, serve as a guide in future economic action, and be an explanation of all economic life. Nothing was more characteristic of the old school than the perfect confidence that they had the key to all knowledge on this subject. They were accustomed to speak of 'immutable laws' and 'eternal principles.' Selfinterest, demand and supply, the law of diminishing returns from land, Malthus' law of population, Gresham's law, the wage-fund, equality of profits, these were the touch-stones the application of which settled every problem. Is it a question whether strikes are able to raise wages?