MIDASURIDMENTSof the size of waves have now been made systematically for many years, but they relate chiefly to the waves of the open sea, where the depth of the water. is so great that the friction of the sea bottom exercises no modifying effect. A few months ago the North German Lloyd liner Brandenburg came into New York harbor with her crows nest, 50 feet aboye the water-line, stove in, and bearing many other marks of the damage wrought by a monster wave that broke over her bows about 1,000 miles east of Sandy Hoole The officers estimated the height of the wave at 65 feet. This height is exceptional, but not unprecedented; for it must be remembered that the breaking of a wave against an obstacle throws the water to a far greater height than the unbroken wave could attain. Unbroken waves due to the wind may, in extreme cases, reach a height from trough to crest of 40 to 50 feet. Much higher waves occasionally occur as a result of earthquakes or seaquakes. Solitary waves of this character have sometimes been encountered in otherwise tranquil weather, taking vessels by surprise and not infrequently sending them to the bottom. According to Vaughan Cornish, who has probably devoted more attention to this subject than any other contemporary man of science, the average height of the waves encountered in a severe storm at sea is 20 feet, but the ordinary maximum height of the waves in the same storm will attain 30 feet. In a storm of very exceptional violence the average height may reach 30 feet, and the maximum height 45 feet. This is regarded as about <he limit of the height of waves due to wind only. Cornish finds that in the open sea the height of a wave in feet is about one-half the velocity of the wind in miles per hour. The following table, computed by W. H. Wheeler from a great number of observations, shows the relation of the wind to the dimensions and speed of ocean waves: Wind. Waves. Speed . ' Force, Pounds Height in Feet. Length S;,eed. Feet per Second. Statute Miles per Hour. per Sq. Ft.' in }'eet. Feet per Second. Nautical Miles per Hour. 2 1A 0.02 0.66 17.4 2.32 1.4 6 4 0.08 1.98 52.2 6.96 4.2 13 9 0.36 4.29 113.1 15.08 9,0 20 14 1. 6.60 174.0 23.20 13.9 27 17 1.50 8.91 234.9 31.32 18.8 30 20 2. 9.90 261 0 34.80 2 0 . 8 35 24 3. 11.55 304.5 40.40 24.2 44 30 5. 14.55 382.8 51.00 30.0 59 40 8. 19.47 5 1 3. 3 68.44 41,0 95 67 23. 31.35 826.5 110.20 66.1 118 80 32. 147 100 50. The proportions of the!B waves are so val'ious that averages cannot be given . So much for waves on the high seas. These waves, though they may race along at the speed of an express train, do not carry the surface water far with them; each particle of water describes a local circular orbit during the transit of the wave; so that what advances is rather the form than the substance. The case is quite different when waves break upon a shore, where the shoaling water produces “waves of translation.” These waves are relatively short and steep, and break when they enter water the depth of which is equal to, or a little exceeds, their height from trough to crest. They approach the shore in a direction nearly at right angles to the general shore line, whatever the direction of the wind. This is explained by the fact that, if the wave is at first directed at an acute angle to the shore, when it reaches shallow. water the side of the wave nearest the shore is first retarded, so that the wave tends to swing around until it faces the shore. In planning harbor construction and the protection * These particular intervals of wind speed were chosen as representing the successive steps of the “ Beaufort scale,1' accol'ding to the equivalents formerly assumed for this scale. As the Iatrer are now known to have been much too great, we have omitted the Beaufort numbers from Wheeler's table. of coasts it is customary to consider the amount of exposure to which the coast is subject; i. e., the extent of open sea in a straight line at right angles to the shore. This is called technically the “fetch.” The relation of the fetch to the possible height of the waves was announced by Stevenson in 1852. According to his formula the height of waves in a gale, in feet, is one and one-half times the square root of the length of the fetch in nautical miles. The force of a great wave breaking against a sea wall or other construction is so terrific as to tax the strength of the best-planned work of the engineer. A marine dynamometer for measuring the force of impact of such waves was devised by Stevenson over half a century ago, and modifications of this instrument have since been introduced by several investigators. According to Stevenson, the maximum force of an Atlantic wave is three tons per square foot. French engineers find that the force of the waves on the breakwater at Cherbourg may attain three and a half tons per square foot. Some interesting examples of the height to whieh break'ng waves may be thrown, and the work they may do in moving heavy objects, are given by Wheeler in his “Practical Manual of Tides and Waves." Stevenson records a case in which water was thrown to a height of 106 feet at the Bell Rock Light. At the Alderney breakwater it is said that water has been thrown upward 200 feet. At Peterhead, where the “fetch” is 300 miles, waves of 30 feet in height and from 500 to 600 feet in length have been recorded, the water. has struck the breakwater with such force as to be thrown upward 120 feet, and blocks of concrete weighing 40 tons have been displaced at levels of 17 to 36 feet below low water. At Wick two stones, weighing eight and ten tons each, were thrown over the parapet of the breakwater, the top of which was 21 feet above high water; while blocks of concrete weighing respectively 1,350 and 2,fOO tons were displaced, though there is some doubt whether the latter movement was due entirely to wave action. At the Bishop Rock lighthouse, which is exposed to the fuII force of the Atlantic waves, an iron column weighing over three tons was thrown up 20 feet and landed on top of a rock. At the harbor works of Bilboa, in 1894, a c:lid block of the breakwater weighing 1,700 tons was overturned from its place and dropped into the water. At Ymuiden breakwater a block of concrete weighing twenty tons, placed outside the harbor walls, was lifted by a wave to a height of 12 feet vertically and landed on top of the pier, which was 5 feet above high watet. The above cases illustrate the sheer force of the individual wave as an engine of destruction, but the imagination of mankind is more impressed by the widespread effects wrought by the great storm waves that sometimes inundate low-lying coasts. These waves are often miscalled “tidal waves,” the only justification of the latter name being the fact that their effects are most pronounced when the wave propagated outward from a storm area happens to coincide with the occurrence of flood tide on the coast affected. The precise mode of origin of the storm wave has been the subject of much discussion, and even now is not fully understood. Such waves attend every severe cyclonic storm at sea, and as they travel much faster than the storm (1. e., the storm as a whole, not the wind revolving about the storm center) they often occur on a coast when the weather is otherwise serene, and thus serve as a valuable prognostic of the storm's approach, in case the coast happens to lie in the storm track. It is well known that the barometric pressure is much lower at the center than at the periphery of a storm-the difference sometimes amounting- to two inches or more-and this difference of pressure must disturb the equilibrium of the water, causing it to become heaped up at the storm center. This bulging of the water would amount, theoretically, to about one foot for each inch of barometric depression. However, although this process doubtless contributes to the production of the wave, the violent winds at the vortex of the storm are probably a much more important factor. The mechanism of storm waves was studied by a board appointed by the Chief of the United States Weather Bureau for this purpose in 1901; and the reader is referred to the report of the board, published in the Monthly Weather Review of October, 1901, for further information on the subject. We are concerned here chiefly with the effects of storm waves (which the newspapers and the public call “tidal waves,” as persistently, and with as much reason, as they call tornadoes “cyclones"). These are most severe when the wave moves: toward a low-lying coastal region, having a converging shore-line; this convergence producing the same effect as seen in a tidal “bore." The most disastrous storm waves have occurred along tbe coast of the Bay of Bengal, on the extensive fats lying about the mouths of the Hugli, the Megna, etc. The storm wave of October 7th, 1737, is said to have risen 40 feet in the Hugli, sweeping away 300,000 souls. In May, 1787, at Coringa, near the mouth of the Godavery, such a wave is said to have taken toll of 20,000 lives. 'he Calcutta cyclone of October 5th, 1864, caused the inundation of the flats on both sides of the Hugli estuary, with a loss of about 48,000 human lives, and the destruction of 100,000 head of cattle. The greatest disaster of recent times in this much-afflicted region was the Backergunge hurricane of the night of October 31st-November 1st, 1876, which cost the lives of over 100,000 persons. In this storm the water rose from 30 to 40 feet in less than half an hour. The islands of the Pacific are also subject to vIsitations of this character on a huge scale, in connection with tropical hurricanes. The latest of these was the storm of March, 1910, which was especially remarkable for the vast area that it covered, its track extending some 2,500 miles, from Fiji to New Caledonia, Norfolk Island, and the North Island of New Zealand. Statistics of the loss of life and property in this storm are not yet availalle . Our own seaboard has repeatedly suffered from the effects of storm waves. In the Galveston hurricane of September, 19(0, a series of waves invaded the city; 6,000 lives Wlre lost, and the destruction of property amounted to $30,000,000, The damage was due to wind as well as water, but chiefly to the latter. It remains to say a word concerning quite another kind of wave, viz., that which moves. down a river in flood, often causing widespread inundations of the adjoining shores. The total volume of water discharged by a great river is so enormous that it is commonly expressed in units of a million cubic yards. The normal annual discharge of the Mississippi system is 785,190 of these units. The occurrence of a flood depends upon many factors, which fortunately rarely combine their maximum effects at a given moment. Thus the maximum possible discharge of the lower Mississippi is about 3,000,000 cubic feet per second, which is about one-third greater than has ever occurred, so far as known. The most destructive river floods of which history furnishes any record are those that have occurred in China; especially those of the Hwang-Ho-"China's Sorrow.” We hear of occasional changes in' the course of the Mississippi, whereby a pwperty-holder is presented with, or robbed of, a few acres of alluvial land over night, or a cut-off shifts a few square miles of land from one bank to the other; but what shall we say of a river that has bodily abandoned hundreds of miles of its channel no less than nine times, thus changing the position of its mouth to the extent of fve degrees of latitude? The havoc wrought by these migrations of the Hwang-Ho—to say nothing of the foods that have not involved a general shifting of the channel-staggers imagination. It is most conveniently stated in terms of human lives swept away; and we are told that during the fifteen years beginning with 1851 (including the last great change in the course of the · river) between forty and fifty million human beings fell a prey to the fury of the waters. The floods of the Seine, of recent memory, as well as all other similar disasters of the western world, sink into insignificance in comparison with this record. What of the means of ameliorating the regimen of our rivers-diminishing their liability to flood? For-estation used to be considered the panacea, but to-day our judgment is suspended on this question. The doctors disagree. River engineering has worked wonders. If the destructive flood comes in spite of the engineers, then we can only look to the flood forecaster to give us timely warning-and stand from under. River stage prediction has reached-above all in America-a degree of accuracy that nearly reconciles us to the floods in consideration of the pride we fee* in this particular branch of scientific achievement. This, however, is subject matter for another story. No Two Finger Prints Alike SINCE the time when the late Sir Francis Galton proposed the use of “finger-prints” as means for identifying criminals, soldiers, and people generally, there has been great hesitation on the part of many people to accept the absolute individuality of the markings on the finger tips as certain. To most of us it js very hard to conceive that there cannot be two fingers that are exactly alike. When the polic' try to fix the guilt for a crime iby showing the identity of the finger-print of the accused with that left on a window or a weapon, for example, many people consider the evidence as far from conclusive. A recent murder case in New York city involved finger-prints as evidence of identity, and the attorneys of the defendant appealed the case on the ground that the evidence left room for “reasonable doubt." A communication to the French Academy of Sciences by M. V. Balthazard is of interest in this connection. This student finds that if any finger-print be divided into a hundred squares, each square will contain some distinctive mark, rarely two marks and only in very exceptional cases three or none. These marks are either the branchings of the ridges that you can see on the tip of your finger, or terminations of ridges. Two finger-prints differ from each other either in the arrangement of the marks In tie aiffer- ent squares, or in the character of the marks in a particular square. The total number of combinations of the two kinds of marks in the hUndred squares is the one-hundredth power of 4 (4'00). This is a number that no one can possibly imagine; it is equal approximately to a number that would be represented by the figure 1 followed by sixty zeros. This means that there are possible just so many different kinds of prints, and that no particular combination will Occur more frequently than others. The chances, therefore, of any particular combination of marks occurring, may be represented by a fraction with one as the numerator and a denominator represented by 1 followed by the sixty zeros-a very tiny fraction of a chance indeed. Now in order to compare this very slim chance with human chances generally, we must consider the population of the whole world. At the present time this is estimated to be, in round numbers, 1,500,000,-000. There are about three generations in a century, or say about five thousand million different human beings. As each of these has, on the average, not more than ten fingers, that makes a possibility pf about 50,00( million different finger prints in a eent ury. In order to have enough human beings to furnish the fingers necessary for realizing all the possible combinations of marks, we should need a number of years represented by the figure 2 followed by forty-eight zeros. The chances for finding two finger-prints exactly alike would be about once in that length of time. This period is of course inconceivable, and from a practical human point of view it is as long as “eternity.” This length of time is vastly greater than the total length of the period since there was any kind of life Upon the earth at. all; it is even much longer than the length of time that astronomers calculate as necessary for the complete freezing of the sun. But if it wO'ld take that long, approximately, before two human finger-prints appeardl that were exactly alike, one may still ask how near alike may two fnger prints be. Or to put the question in Hs practical form: In how many points must two fingerprints agree to make sure of identity? Dr. Balthazard shows on purely mathematical grounds that when two fyger prints agn)e in 17 of the 100 squares, it is practically certain that they were made by the same finger. His reasoning is as follows: I we look for a single coincidence, we may find it on examining four fingers. To find two finger"rints that have two coincidences, we must eXMline at least sixteen fingers. To ·fnd three coincidences between two finger-prints, we must examine at least sixty-four prints, etc. These numbers, it will be noticed, represent the powers of 4 corresponding to the number of coincidences sought. Thus, 4' = 4; 42 = 16; 43 = 64; 44 = 256, etc. The probability or “chance” of any numier of coincidences occurring is 1 divided by 4 raised to the power of the number of coincidences. In this way we must find the number of coincidences whose probability is as 1 divided by the total number of fingers on the earth at any one time. This number is slightly less than seventeen, for the seventeenth power of four (4") is 17,179,869,184. (The total population being taken at fifteen hundred millions, this number is multiplied by ten-the number of fingers for ea(h rperson-giving us 15,000,000,000 fngers at any one time.) In actual practice it may not be necessary to find as many as seventeen coincidences to mrke sure of identity. If the two finger-prints that are being compared are known to have been made by Eur:)-peans, or by Americans, or by inhabitants of a C8r-tain district, the number of coincidences necessary to establish identity is much fewer. Bertillon found two fnger-prints made by different men, having as many as thirty coincidences. This would seem to be 'tt variance with the law of probabil'ly; but on examination it appeared that the two men were twin brothers. However, the mathematical demonstrations may not be quite convincing to many persons. The doctrine of probability may be acceptable enough, but it remains to be shown whether all the combinations that are theoretically possible do actually occur in equ31 numbers ; that is, whether there is not a tendency for certain combinations to occur more frequently than others; and whether, therefore, certain coincidences are not :more likely to occu r than others. Still, the numbers involved are so enormous, that for all practical purposes we may feel quite assured that what is likely to happen only once in a long, long eternity is not likely to happen twice in a century. August 1 9 , 1 9 1 1 SCIENTIFIC AMERICAN 1 67 [The eritors are not responsible for statements made in the correspondence column. Anonymous communications cannot be considered, but the names of correspondents will be withheld when so desired.] Cosmic Speculation To the Editor of the Scientific American: To modify if not entirely to prevent dangerous thaws, intense heat-waves, extensive droughts, and seismic disturbances, should be the primary aim or mankind in the present century. SolaI rays, striking the earth at right angles, have the efect, in one region, of s u ddenly dissolving and releasing huge ice-jams; in another, of overheating and evaporating vast quantities of water. This mass, in its new form, accumulates heat and electricity. If the phenomenon were to recur indefnitely, it would probably result in the total disintegration of the terrestrial mass. However, there may be noted, as counteract.ng the infuences described: (1) the revolving motion of the earth about the sun, (2) the attmction exerted upon it by the moon, and (3) the irregular formation of its crust. Yet the proceeding does repeat itself for days and months. When the former is the case, the phenomenon brings in its trail widespread inundations and the insuferable heat that annually numbers its vIctims by the thousand-particularly in this very country. And in the latter instance, when the occurrence continues fol' a considerable length of tim 1, an exhaustion is produced of the general humidity as well as of the aqueous deposits left by previous rains in the soil and subsoil. These deposits ordinarily feed the internal streams and lakes whose function is refrigeration. In reducing these cooling infuences, the draining of of inner bodies of water heightens the internal heat, and with it the extent of evaporation and electricity; until the moment arrives when equilibrium is upset between the inner and outer electrical states, giving birth to a wave that precipitates commotion and disturbance. On occasion, the phenomenon takes on more serious proportions, because of greater intensity in the sun'a heat or through special circumstances existing in the interior regions. These contributory causes facilitate the action of “the mighty fuid” and combine with it to hurl outward the very bowels of the earth, to make room for the processes of deformation and disintegration. In view of the advanced state of the sciences and the increased means of action at the disposal of man, the phenomenon should aHow of easy modification. The remedy is to be sought in two directions: (1) By loud reports in the upper strata of the atmosphere, we may attract currents of air bringing cold from the regions that are far distant from the scene of overheating. (2) The natural course of events can further be modifed by collecting, simultaneously with the explosions on high, the electricity stored in the atmospher and in the bosom of the earth. Small acts determine large efects. Deprive a huge boulder of a litHe stone that contributes to its equilibrium, and the mighty mass will overturn. !et a man shout, a gun give forth its report or a tiger roar, at a given place and at a given moment: immediately there will be unchained a down-pour, a tempest lasting hours. A decade or more must el8;pse Ibefore we can determine the precise spot and occasion for such action. F'or perhaps a pair of decades our plans must hold fre; until we can reach into the depths of the earth with a receiver. Yet both ends sha].! certainly be attained. In the meanwhile, advantage can be taken of the art of aviatIon and of craters and their data to study the laws of the atmosphere and general geological chemistry. The task set before us is not one for the individual. Rather does it confront the nations. World conventions, competitive contests for generous prizes, observatories and laboratories already in existence and yet to be created, all to be united into one association better to se and better to J8bor-suoh is the path to be followed. New York city. R. Febre s Cordero. The Bridgeport Railroad Wreck To the Editor of the Scientific American: In your article entitled “An Imminent Railroad Danger” in the issue of July 22nd, you speak of the “perfection of the track and equipment of our leading railroads.". I feel, however, that, as far as the Federal Express is concerned, the description “perfection of equipment” does not apply. Being interested in such matters, I hav frequently noticed particularly the equipmeDt of this train as It stood in the South Station, Boston. The train usually consists of a large Pacifc type locomotive followed by a baggage car (sometimes a Pennsylvania steel car), a New Haven wooden vestibuled day-coach (not of the most modern tYlpe) and then several Pullman sleepers, some of which are, sometimes, of the all-steel type built for use on the Pennsylvania Railroad. According to your opinion repeatedly expressed in your paper, such a sandwiching of a weak wooden day-coach between a heavy locomotive and heavy Pullman sleepers is about the most dangerous arrangement that could be devised. The fact that nearly all the deaths occurred in the day-coach shows the validity of your contention, which, however, is not difcult to understand in the abstract. As for the matter of human control, i seems that it ought to be possible and not seriously inconvenient to so arrange the controlling levers in the cab that it would be necessary for the engineer to exert some little pressure on them to keep the throttle open and the brakes of. Then, if, through his unconsciousness or sudden death, the pressure were released, the train would automatically be brought to a stop. Braintree, Mass. Aubrey D. Beidelman. Possible Size of Battleships To the Editor of the Scientific Ame r ican: Apropos of the limits of si2le of battleships, at least for the American navy, it would seem the Panama Ganal sets very certain restrictions. The llocks have a length of 1,000 feet, . width of 100, and a depth of 40. Supposing the maximum draft of a battleship to be 30 feet, its length to breadth 6: 1, and its beam the maximum 100 feet, we haye a limit (using the coefcient of fneness 0.600) 30;000 tons. If we extend the length and breadth to 7: 1, we have with the same draft -f 30 feet 36,000 tons. Suppose the draft to be enlarged to 35 feet, for the frst set of proportions we have again 36,000 tons, and the Becond set (7: 1) 42,000 tons. It is hardly possible that the :length of a battleg,hip with regard 'o its beam will exceed 7:1 with due regard to stability as a gun platform. Coinci-dently, 35 feet in all probability will be the maximum possible draft, having taken into consideration the depth of onr harbors, and the possibly increased draft of a disabled warship. As it is, with the maximum dimensions of 100 feet beMn, 700 feet length, and 35 feet draft, there is no dock in the United States as far as I know that will take such a ship. Hence.it appears that these are the outer limits of dimensions for our battleships; for after having constructed the canal, it would be superlative madness to neutralize it by 'building ships which could not use it. By increasing the coefcient ,of fneness, a somewhat larger tonnage might be obtained. Hence it would seem that the utter limit in size of our battleships is roughly 40,000 to 45,000 tons. From 1907 t 1911 the tonnage has jumped froll 20,000 to 30,000-50 per cent. At the same rate of -increase, in 1915 the limit will have 'Leen reached, and the battle of displaeement will have given place to the battle of calibrs. Pittsburg, Pa. Wiliam BeBgen ChalFant. Photography of the Invisible Ray To the Editor of the Scientific American: I was much interested in an article appearing in the Scientific AmeBican under date of June 10th upon “Photography by InvisL'le Rays,” with photo illustrations of the theory. If you will look in a “Science Record” of 1874, published by Munn&Co., under the heading of “Photometry of Colors,” you will fnd a brief outlining of some crude experiments I made to determine so far as possible the relative focal length of the colors of the solar spectrum, concluding that the shorter the focal length of a color, the greater its vibratory rate. WHh this theory as a working hypothesis, and the idea that light travels 186,000 miles per second of time, I made an estimate of the vibratory rate of each color motion in its action upon the photographic plate, thus disclosing the descre-parcies between high light vibrations and those which promoted indistinctness in the negative. By interposing a thin pale straw-colored glass between the lens and the object photographed, all colors of a high vibratory rate were reduced to greater harmony. In indoor photography, I recommended fooding the sitter with some suitable color. Still another way is to attach a pasteboard tube to the camera lens, and food the inside with some selective color, through which the color vibrations have to pass. I have employed every shade and tint of colored glass I could get. I obtained some remarkable results with fuids for a screen. A glass receptacle made of thin white fint glass an eighth of an inch for the fuid; the receptacle wide enou"h and equaly long to cover the lens. Transparent fuids of diferent chemical density give astonishingly varied results. I was the lecturer for the National Photographic Association at the time I refer to, and I think the Scientific American had a reporter in attendance. At that time I made the statement that “if the principles of molecular radiation were true, the time was not far distant when we could be able to photograph' the internal human anatomy.” I also said that “disease as a mode of motion in some of its phases could be photographed, in consequence of its action upon the epidermiS.” This has been done, and is done. The striking similarity between your reproduction of Prof. Wood's experiments and several out-door photos made by my experiments, elicited my attention; and although 1 have no copies of my work in that line at hand, I am still deeply interested in any discovery that extends the lines of scientifc research. I do not know how recently Prof. Wood made these investigations, but I want you to know that an American practically covered this ground, and much more, thirty-fve years ago. You may recall me as one of your contributors in an essay upon “The Fourth Dimension of Space." Bufalo, N. Y. Dr. w. M. Lockwood. Nitrogen and the Soil To the Editor of the Scientific American: In the issue of your paper dated July 1st, there appears on page 9 an article entitled “An Important Contribution to the Nitrogen Problem.” This article refers to a bulletin published by Lyon and Bizzell of CoreU University, and entitled: “A Heretofore Unnoted Beneft from the Growth of Legumes.” I am writing now to correct an error, since the priority of this discovery belongs to the New Jersey Experiment StJtion. A preliminary statement of these experiments is made in the annual report of the Ofce of Experiment Stations, United States Department of Agriculture, for the year 1909, page 150. A copy of this report may be secured from the director of the ofce at Washington. A fuller statement of these experiments was submitted to the Journal of Agricultural Science, published at Cambridge, England, in the summer of 1909. After much delay this paper W8;S published in the fall of 1910. The paper in question is abstracted in the Experiment Station record for April, 1911, page 423. The inclosed copy of this paper will show you that the problem was solved in a very exact manner by a method devised at the New Jersey Experiment Station. I should feel indebted to you for calling attention to this fact in the columns of your paper. Jacob G. Lipman, Acting Director. New Brunswick, N. J. The Moving Picture and the Psychology of the Chaufeur To the Editor of the Scientific Ame r ican: I like your comment on the article in your paper, “Psychological Apparatus for Testing Chaufeurs,” and would like to suggest that the complicated signaling system could be done away with by substituting a' moving-picure machine. It ought to be possible to make suitable flms for the purpose, with means for Ithrowing the recording apparatus in motion, at the instant the person to be tested sees the picture on the screen. Jewell, Ohio. H. G. Panning. Testing Chaufeurs with the Moving Picture Machine To the Editor of the Scientific American: I was interested in reading Automobile Novelties. It doesn't seem to me that the signals for danger and diferent degrees of danger is a very fair test to the chaufeur, as he would have to b very familiar with the diferent signals before he could control his machine as well and as quick as he could under the conditions the signals represent or indicate. Now the moving-picture machine is perfected to such an extent, that the actual reproduction of danger can be produced in a most realistic manner; and it seems to me that by this means the test would be a far fairer one, 8 the chaufeur would have only his machine ItO t.hink of, and in some emergencies this is a great plenty. It also seems that to put a real machine instead of the dummy would be an improvement, as it is well known thlat all men familiar with moving machinery of all kinds have a sense of feel that is higMy developed, and would tend to put the chaufeur more at ease, as it would be more like the real thing. Orane. Texa. F. T. RnEB.