suppose scientific knowledge had sooner led us to recognize the close relation of electricity and light. Surely the mode of division would be quite different. The definition of Zoology before and after Darwin would have been different. A classification which then appeared to be scientific would now be recognized as inadequate. The very first thing we must recognize is that our scientific knowledge is imperfect and growing, and we must adopt a system capable of easy modification as our knowledge increases. Another point which Chevalier Descamps made was that we adopt different methods with regard to different sciences; in some cases the numbers are followed by symbols; in some cases the numbers are separated by a hyphen, and so on. We have gone into this question as scientific men, and, although perfectly ready to submit the result of our work to the criticism of other scientific men, we do believe that the plan that suits best one science will not suit another. Take one example. Take, for instance, Zoology. There is the question of arrangement of the subject in accordance with the species of animals, and the question of arrangement with regard to the geographical distribution. Here are two ideas to which which there is nothing similar in physics or chemistry. It would be disastrous if we attempt to force all these sciences to adopt the same method. If two things are essentially different, we do not apply the same principles to both. In the last place, Chevalier Descamps says the main object of classification is to tell us where to find a particular object with which we are dealing. I do not much believe in the average memory of scientific men being able to grasp a large number of numbers. I believe it is much easier to find the place by using symbols, which are more distinct than a large number is from a small one. Significant words which are for temporary use have their own meaning. You find them

alphabetically. I do think, on the question of general principle, that it is very desirable that the Conference should express an opinion as to whether or not they think the symbols are to be devised in such a way as to help the memory or to find the place; secondly, whether they do or do not hold the view that the plan good for one science is good for all, and whether it is desirable to attempt to plan a scheme in the belief that it will hold good for all time."

Dr. Bernoulli said that after hearing the statements in favor of the two systems he wished to add that the decimal system was in actual working order in Switzerland, and that its practical utility had been demonstrated there. He considered it superior to the system proposed by the Royal Society, although originally he had been an opponent of the decimal system.

M. Deniker replied that it was necessary to consult an alphabetic index to use the decimal-system catalogue. He favored a methodical or subject catalogue alphabetically arranged.

SMITHSONIAN INSTITUTION.

CYRUS ADLER.

(To be Concluded.)

COLOR-WEAKNESS AND COLOR-BLINDNESS. It is generally accepted as a well established fact that the traveling public is fully protected by the present tests for color-blindness to which railway employees and pilots are subjected. Yet several of the mysterious accidents that have occurred during the last two years might be explained on the supposition of color-blindness on the part of responsible lookouts. In fact, I believe myself in position to prove that persons of dangerously defective color-vision actually do pass the regular tests and obtain positions where their defects are continual dangers to public welfare.

In the first place, I have at the present time among my students one who is abso

lutely perfect at the wool-test. He can match wools with incredible precision at any distance away; he is, nevertheless, color-blind. This case is typical of a class of persons with eyes abnormally acute for differences in color, but yet with only two fundamental sensations instead of three.

In the second place, I have had among my students those who possessed perfect color-vision for near objects or bright objects, but who were practically color-blind for weakly illuminated or distant objects. These persons possess the typical three fundamental color sensations, but have one of them weaker than the normal. A person of this kind may pass the wool-test with the utmost perfection if the test is performed close by, but will fail if the wools are removed to a distance of 20 or 30 feet. This peculiar defect I take the liberty of terming' color-weakness.' The first student of this kind that I examined passed the wool-test close at hand and yet was unable to distinguish red and green lanterns a few hundred yards away. Cases similar to this have been reported by the British Marine Examiner, Edridge-Green. Among other cases he quotes a letter from an engineer containing the following statement: "I have been on the railway for thirty. years and I can tell you the card-tests and wool-tests are not a bit of good. Why, sir, Why, sir, I had a mate that passed them all, but we had to pitch into another train over it. He couldn't tell a red from a green light at night in a bit of a fog."

To eliminate both these classes of persons we must have a method of testing on quite different principles from the usual ones.

In the first place, the sorting of delicate shades of colors, according to likeness, must be replaced by naming certain fundamental and familiar colors. The sorting of wools is a quite unusual and perplexing task to a man brought up in a railway yard and on shipboard. It puts a nervous man at quite

a disadvantage; it furnishes the unsuccessful candidate with the excuse that the judgment required was so unlike any he had made before that he failed from nervousness; and, finally, it is not a guarantee that all who pass are not color-blind. The naming of colors should-as Donders proposed --be rigidly required. The engineer or the pilot in his daily routine is not called upon to match colors, but to decide whether a light is red, green or white; he should be tested on just this point. The color-blind student referred to above who can pass the wool-test to perfection fails at once when called upon to name the wools. The naming of delicate and perhaps unusual shades should, however, not be required; the colors to be named should be the three familiar ones red, green and white, so manipulated that every possible chance for confusion is presented.

The second necessity for eliminating danger is that of an absolutely certain test which shall detect both the color-blind and the color-weak. Acting on the basis of suggestions from the work of Donders and of Edridge-Green, I have devised a test that meets this requirement as well as the first one.

The instrument* which I have invented may be termed the 'color sight tester' or the 'color sense tester.' In general appearance it resembles an ophthalmoscope. On the side toward the person tested, Fig. 1, there are three windows of glass, numbered 1, 2 and 3, respectively. The opposite side of the tester, Fig. 2, consists of a movable disk carrying twelve glasses of different colors. As this disk is turned by the finger of the operator the various colors appear behind the three windows. At each movement of the disk the subject calls off the colors seen

*For those interested in obtaining the Color-Sight Tester I will say that I have made arrangements to have it made by the Chicago Laboratory Supply and Scale Co., Chicago.

2 is gray; No. 3 is green." The operator records from the back of the tester the letters indicating what glasses were actually used. If he finds that A, D and G were opposite the glasses Nos. 1, 2 and 3 he records: A 1, dark red; D 2, gray; G 3, green. The disk is then turned to some other position; the colors are again named, and the operator records the names used. For example, the result might be: "No. 1 is dark green; No. 2 is white; No. 3 is red;" and the record would read: G 1, dark green; J 2, white; A 3, red. Still another record might give: J 1, dark gray; A 2, red; D 3, medium gray. Similar records are made for all combinations. Of course, the person tested knows nothing concerning the records made. A comparison with a list of the true colors for each position determines whether the test has been passed or not.

The three records just cited were all obtained from the red glass, A; the gray glass, D; the green glass, G, and the ground glass, J, in combination with the dark gray, No. 1; the ground glass, No. 2, and the medium gray, No. 3. Those familiar with color

blindness will notice that these combinations place side by side the colors most confused.

The records can be taken by any one, and, on the supposition that the record has been honestly obtained and that the instrument has not been tampered with after leaving the central office, the comparison is mechanical. There is none of the skillful manipulation required in the wool-test and none of the uncertainty attaching to its results. The only instruction given to the subject is Name the colors;" the results render the decision with mechanical certainty.

One of the testers is in use on one of the English railways, another on the central division of the New York Central Railroad. From the former I have not yet heard, but

the examiner on the latter reports that since using the tester he has found men who get through the wool-test, but are caught by the tester. On the other hand, he states that "the men examined say that this test is more like the signals they are used to seeing every day on the road, and. is, therefore, fairer than to ask them to pick out a lot of delicately tinted pieces of yarn."

An experience of several years seems to justify the following claims for the color

sense tester:

1. It detects with unerring precision both the color-blind and the color-weak.

2. It is a perfectly fair test for the men concerned and injures no man by requiring an unfamiliar judgment.

3. It requires but a very small fraction of the time used on the wool-test.

4. Its decisions are self-evident and unquestionable.

E. W. SCRIPTURE.
PSYCHOLOGICAL LABORATORY,
YALE UNIVERSITY,
May 7, 1899.

AMERICAN CLIMATOLOGICAL ASSOCIATION.

THE Sixteenth annual meeting of the American Climatological Association was held in New York City on May 9th and 10th at the hall of the New York Academy of Medicine. About fifty members were in attendance from all portions of the United States. Twenty-five papers were read upon subjects pertaining to climatology, hydrology and diseases of the respiratory and circulatory organs. These papers, which will appear in the annual volume of the Transactions, were as follows:

'Presidential Address,' by Dr. Beverley Robinson, of New York.

"Treatment of Consumption by Air and Light in Colorado,' by Dr. Charles F. Gardiner, of Colorado Springs.

'Intermediate Altitude for the Consumptive,' by Dr. B. P. Anderson, of Colorado Springs.

'The Contagiousness of Phthisis Pulmonalis,' by Dr. E. L. Shurly, of Detroit.

'Climate in Relation to Renal Disease,' by Dr. J. B. Walker, of Philadelphia.

'Climate as it affects the Skin and its Diseases,' by Dr. L. D. Bulkley, of New York.

Hygienics of the Skin,' by Dr. L. D. Judd, of Philadelphia.

'Hydrotheraphy in the Treatment of Insomnia,' by Dr. Irwin H. Hance, of Lakewood.

'Altitude and Heart Disease,' with report of cases, by Dr. R. H. Babcock, of Chicago.

'Prognosis in Chronic Valvular Affections of the Heart,' by N. S. Davis, Jr., of Chicago.

"Treatment of the Cardiac Asthenia of Pneumonia,' by Dr. H. L. Elsner, of Syracuse.

'Empyema from a Surgical Standpoint,' by Dr. Johh C. Munro, of Boston.

'Traumatic Rupture of the Heart, without Penetration of the Chest Wall,' with a case, by Dr. Richard C. Newton, of Montclair.

'Cold Wave of February, 1899,' by Dr. Guy Hinsdale, of Philadelphia.

Other papers by Drs. R. G. Curtin, C. F. McGahan, Harold Williams, F. H. Williams, E. O. Otis and V. Y. Bowditch, S. G. Bonney and H. S. Anders.

The annual dinner of the Association was held at the Manhattan Hotel, at which the President, Dr. Beverley Robinson, of New York, presided. On the following day the Association made a visit to the Loomis Sanitarium in Liberty, Sullivan County, New York. This institution was founded 1895 in memory of Dr. Alfred L. Loomis, the first President of the Association, for the treatment of tuberculosis. It has a favorable situation, 2,300 feet above tide, and is 120 miles from New York, on the Ontario and Western Railway. The remarkable success which has attended its work has been due in great measure to its physician in charge, Dr. J. E. Stubbert, liberally aided by the philanthropic support of Mr. J. Pierpont Morgan and the ladies who are associated in its management.

The scientific work of the Climatological Association tends to the better knowledge of the various American climates and health resorts and their employment in the treatment of disease.

The subject of tuberculosis is now receiving universal attention by the medical profession, and the public are being interested in measures looking to its prevention and restriction. It is encouraging to note that in all our large cities the mortality. from this disease is gradually falling, and through societies of this kind knowledge is disseminated which affords the public greater protection and prolongs life. The resources of New York and Pennsylvania for the climatic treatment of pulmonary disease are not so well known as they should be. Neither are the mineral springs of the United States fully understood and intelligently used. The Transactions of the Climatological Association, now numbering fifteen volumes, have contributed in no small degree to the better knowledge of this extensive subject.

The following officers were elected for the ensuing year: President, Dr. A. Jacobi, of New York; Vice-Presidents, Dr. R. H. Babcock, of Chicago, and Dr. John W. Brannan, of New York; Secretary, Dr. Guy Hinsdale, of Philadelphia; Representative to the Executive Committee of the Congress of American Physicians and Surgeons, Dr. F. I. Knight, of Boston.

The next meeting will be held in Washington in May, 1900. GUY HINSDALE, Secretary.

SCIENTIFIC BOOKS.

SOME SMITHSONIAN PUBLICATIONS.

Annual Report of the Board of Regents of the Smithsonian Institution, showing the operations, expenditures and conditions of the Institution to July, 1896. 8vo, lii + 728 pp., lxi pls. Washington, 1898. [Received by the Bureau of International Exchanges, January 25, 1899.] Annual Report of the Board of Regents of the Smithsonian Institution, showing the operations, expenditures and conditions of the Institution for the year ending June 30, 1896. Report of the U. S. National Museum. 8vo, xxiv +1108 pp., excviii pls. Washington, 1898. [? 1899.]