Aiming With the Rifle*

Some of the Difficulties Encountered, and How They May Be Overcome

By Edwin Edser

So many people are now learning to shoot with the rifle that it is profitable to consider some of the dilllculties they are likely to meet with. These difficulties become greater as the age of the learner increases, and they may be minimized or accenuated by the lighting of a range at which the learner practices. A discussion of the lighting of rifle ranges, which took place at the monthly meeting of the Illuminating Engineering Society on May 18, shows very clearly that the existing conditions place artificial obstacles in the way of the learner; and it may fairly be contended that these obstacles never would have arisen, and the path of the learner would have been considerably smoothed, if certain optical principles had been recognized and utilized. Mr. A. P. Trotter, who opened the dscussiou, gave a very clear account of the difficulties encountered by a man of middle age when he attempts to shoot at one of the many indoor ranges which have recently been opened; it has appeared to me to be worth while to attempt to explain some of these difficulties, in order that those which are avoidable may be eliminated.

An experimental arrangement which can be used to illustrate the essential difficulties to be met with in aiming with the rifle, Is represented in perspective in Fig. 1. A is a rough model of the eye. It comprises

[ocr errors][graphic][graphic]

Fig. L

a tube about 1% Inches in diameter and 3 inches long, closed in front with a lens L of about 3 inches focal length; into the back of this tube fits another tube, which carries a screen of ground glass 8. B is a sheet of cardboard, with a notch in the upper edge, to represent the rear-sight of the rifle. G is a piece of card cut to a point, to represent the fore-sight of the rifle. D Is a circular opaque disk which, for convenience, may be attached to the glass of a window of the room in which the experiment is conducted; this disk represents the "bull's-eye" of the target. By sliding the screen 8 in or out, either B, C, or D may be focussed; but all cannot be focussed at the same time. If, however, the lens is covered with a piece of card provided with a circular aperture of about % Inch diameter, A, B, and C can all be focussed simultaneously; and the screen S can be moved In or out through some distance without impairing the clearness of the Image on the ground glass. The brightness of the image is, however, much diminished. This illustrates the advantage and disadvantage due to the use of "pin-hole" spectacles. If the card is arranged so that its circular aperture lies over the middle of the lens, and the images of B, C, and D are formed at the middle of the ground-glass screen, the position of the image of either B, C, or D is identical with that of the corresponding image produced, with the card removed, by adjusting the position of the ground-glass screen; but if the aperture of the card is displaced toward the edge of the lens L, the various images are displaced both relatively and absolutely.

Further, let the perforated card be removed, and let the screen S be adjusted so that the "bull's-eye" is focussed; then on covering the lens from below by a piece of unperforated card, it will be seen that as the card rises, the image of the "bull's-eye" sinks, while the Images of the sights rise. A similar effect can be observed with regard to the eye. If the model eye A is removed, and replaced by the eye of the observer, adjusted so that B, C, and D are in alignment, while D is focussed, it will be found that if the pupil of the eye is gradually covered from below by a piece of card, the "bull's-eye" appears to rise above the sights.1 To understand this result, it must be remembered that the image produced on the retina is inverted, and that an absolute depression of the image is interpreted as an apparent rise of the object viewed. The apparent motion referred to Is very marked when the light is dim

• From Nature.

1 See "Spherical Aberration of the Eye," by E. Edser (.Vuture, April 16, 1903). Also "Light for Students," by E. Edser (Macmlllnn & Co.), p. rfi5.

and the pupil is expanded; it can only be noticed with some difficulty in bright daylight

Returning to the arrangement represented In Fig. J, it will be found that when the "bull's-eye" is focussed by the unstopped lens L, raisiug the card B causes the image D to sink. Similarly, in a diu\ light, on bringing the rifle Into position so that the rear-sight intercepts light from the lower part of the pupil, the "bull's-eye" appears to rise. In a greater number of cases, when the fore-sight is brought too high, so as partially to cover the "bull's-eye," the latter appears to swell at its upper left-hand edge (at about "half-past ten"), and sometimes this swelling develops into a second "bull's-eye" detached from the first one.

The following important phenomena can also be noticed:

(1) On focussing the bull's-eye with the lens L unstopped, the image of the fore-sight V is surrounded by a narrow penumbra; a similar but wider penumbra borders the image of the rear-sight B. It the lens is now stopped down, the circular aperture of the card being over the middle of the lens, the images of B and 0 become sharp, and it will be noticed that the images of the edges of the nights note have the same positions as the edges of the corresponding penumbras produced by the unstopped lens. Thus It appears that in aiming with the rifle, when the bull's-eye is focussed, the top of the narrow penumbra surrounding the foresight should be brought level with the top of the wider penumbra bounding the shoulders of the V or U rearsight I have found that this procedure leads to consistent and good shooting. A peculiarity of the penumbra surrounding the ocular image of the fore-sight will be mentioned later.

(2) On focussing the fore-sight G with the lens L unstopped, the Image of the bull's-eye D becomes much smaller, and may even disappear. The image of the rear-sight is slightly improved. Similarly, when aiming with the rifle, the image of the bull's-eye is diminished in size when the fore-sight is focussed by the eye.

(3) On focussing the rear-sight B with the lens L unstopped, the bull's-eye D disappears, and the foresight B becomes smaller and less distinct.

Now, young people can alter the focus of their eyes without effort; they see the bull's-eye, the fore-sight, and the rear-sight in rapid succession, so that sometimes they appear to see all three at the same time. In this case sighting is easy. But with advancing age comes the necessity for effort in focussing the eye to different distances, even if this capacity Is not lost altogether. For myself, I can read print (even small print) at 10 inches from my eye, but a perceptible effort is required to alter the focus of my eyes; and from the result of my own experience, together with that of several men in a condition similar to my own, I strongly advise that the bull's-eye only should be focussed, the tip of the fore-sight being brought just below the bottom of the bull's-eye and level with the top of the penumbra which bounds the shoulders of the rear-sight.

A peculiarity of the image of the fore-sight, when the bull's-eye is focussed by the eye in a dim light, must now be mentioned. At first sight the appearance presented is that of three images' standing side by side, the central image being the darkest. On careful scrutiny, however, two overlapping images only are seen, the portion common to both being darker and giving the appearance of a third image (Fig. 2, A).

[graphic][graphic][graphic][merged small]

some distant object; a narrow penumbra will be seen round the tip of the pencil, and on observing this carefully it will become evident that there are really two overlapping images of the pencil tip standing side by side, the portion common to the two being dark (Fig. 2, B). The nearer the eye is to the pencil, the greater is the separation of the images; in daylight, separation is just visible (to me) at a distance of about 3 feet. If the right half of the pupil is now covered by a card, the left image disappears; on covering only the left half of the pupil, the right image disappears. If the pencil is placed in a horizontal position, the appearance is quite different; the pencil now appears sharply delined laterally, but its tip ends in a penumbra (Fig. 2, C).

It appears to me that these phenomena may be ascribed to the peculiar shape of the cornea. It»has long been known that the cornea is not spherical, and Sulzer has found that its form does not agree with any

[graphic]

Fig. 3.

known simple surface, and that it has no axis of symmetry. In the majority of cases the nasal side of the cornea is flatter than the temporal side, so that the section of the cornea of the right eye, when viewed from above, resembles BG, Fig. 3. The visual line OA (1. e., the line along which the most direct ray travels from the object 0 to the most sensitive portion of the retina A) passes through the flatter portion of the cornea; and the center of the pupil is also behind the flatter portion of the cornea. Thus when the light is good, and therefore the pupil is small, the rays which form the image on the retina pass through the flatter portion of the cornea; and under these conditions we obtain the best occular images.

Now, in aiming with the rifle in a dim light, the bull'seye being focussed, if the cornea were spherical, there would be a number of overlapping images of the foresight, thus giving rise to the appearance of a single dark image surrounded by a penumbra. The peculiar shape of the cornea, however, appears to cause a segregation of these images into two groups, giving rise to two overlapping images side by side. The light which enters the right eye through the left part of the cornea (i. e., the flatter portion) gives rise to the right-hand image; that which enters through the right (more strongly curved) portion of the cornea gives rise to the left image. So far as my experience goes, the right image is the darker and better defined of the two; and we might expect this to be the case, since it is formed by the rays which traverse that part of the cornea which is utilized when vision is at its best It therefore appears that the right-hand image of the fore-sight should be aligned with the middle of the notch of the rear sight, its tip being just beloto the bull's-eye at "six o'clock," and just level with the top of the penumbra that bounds the shoulders of the V or U rear-sight (Fig. 4). In a dim light it is well to allow for the fact that the bull's-eye is apparently raised, by leaving a distinct white line between the tip of the foresight and the lower side of the bull's-eye. In all cases the fore-sight should at first be aligned some distance below the bull's-eye, and raised to its final position just before firing.

When the rifle is aimed in daylight irith a bright Kky overhead, light is reflected from the upper rim of the rear-sight into the eye. When the bull's-eye is focussed, this light forms three bright linear images in the eye. The lowest bright line occupies the position of the upper boundary of the black portion of Fig. 4; the middle bright line occupies the position of the upper boundary of the penumbra shown in Fig. 4; while the upper bright line bounds a faint secondary penumbra which is scarcely visible in a dim light. Similarly, if a diaphragm with a narrow horizontal slit is placed in front of an eye focussed to see distant objects, three bright images of the slit are seen. These multiple images, which vary somewhat iu position for different observers, and even for the two eyes of a single observer, are presumably due to variations of curvature of the cornea in a vertical plane. Correct shooting can be obtained by aligning the top of the fore-sight with the central bright line which bounds the lower penumbra; as this line is clearly seen, it can be utilized as easily as the focussed image of the rear-sight. The advantage of a good overhead light thus becomes apparent.

So far as the lighting of indoor ranges is concerned, it may be inferred that we shall see best under those conditions which approximate most closely to ordinary

[graphic]

Fig. 4.

diffused daylight. The use of a small, bright illuminated target, in a room with black walls and ceiling, could be defended only if It were desired to train people to shoot at a distant searchlight. In such conditions the pupil is distended, all of the troubles discussed above are intensified, with the addition that the glare of the target tires the eyes. Similarly, the glowing filament of an incandescent lamp tires the eye more when it is viewed in a dark room than when it is viewed in daylight. I believe that the best thing to do in connection with indoor rifle ranges would be to whitewash the walls and ceiling, and have a good illumination either with electric glow lamps or incandescent gas mantles, merely taking the precaution that the lamps are shielded (say, by paper shades) from the direct view of the shooters.

So far as the utility of miniature rifle ranges is concerned, it appears to me that this may be easily overrated. It is possible, of course, at one of these

ranges to learn to hold the rifle correctly, to become accustomed to accurate sighting, and to press the trigger without moving the rifle. Difficulty, however, arises from the fact that accurate shooting entails compliance with all three of these requirements, and bad shooting may be due to a failure in one only. The position of the bullet-hole in the target gives only the net result of all the actions involved; and I have known men to ascribe their failure to get near the bull's-eye to the defective sights of the rifle, or (more rarely) to their own defective sighting, when iu reality their bad shooting was due to pulling the trigger instead of pressing it. It is clear that more rapid progress can be made if the learner can discover the particular defect to which his failures are due. Various devices have been used for this purpose.

In the sub-target, the rifle is mounted on a universal joint, and on pressing the trigger a hole is punched in a card, thus indicating the direction in which the rifle is pointed at the instant. This appliance is expensive, and since the rifle is not free, defects due to triggerpulling are not made evident

The aim-corrector is a piece of plain smoked glass mounted behind the rear-sight so that its surface is inclined at 45 degrees to the sighting line. The learner takes his sight through this glass in the usual way; the instructor watches the sights from the side, as they are seen reflected in the glass. Obviously, the instructor must possess considerable skill in order to use this appliance with advantage.

The aiming disk is a perforated metal disk which is placed in the observer's eye like a monocle. The learner aims at the perforation, and any considerable motion of the rifle during trigger-pressing can be seen by the observer. This appliance can only be used with advantage at short distances from the learner, and anyone accustomed to the use of firearms can scarcely avoid an uncomfortable feeling on watching a gun that is pointed at his eye.

I have devised a simple appliance by means of which most (if not all) of the benefits usually derived from a miniature range can be obtained without the use of ammunition. This appliance is represented diagrammatlcally in Fig. 5. A metal tube T, which can be fitted to the bayonet standards of a rifle, is provided with a

lens L at the front end, and a small electric glow-lam] > G at the. rear end. The lens L can slide In or out, so that the image of the glowing filament of the lamp can be focussed on a white screen placed near the target. The current for the lamp can be supplied by three or four Leclanche cells; or a battery of dry cells, similar to that used for an electric torch, can be fixed to the tube T, thus obviating the inconvenience of the leads from the lamp to the cells. It is best to aim at a target about 10 yards away; an observer, who need possess no qualifications other than general intelligence and quickness of perception, stands or sits by the target and watches the image of the filaments formed on the screen. I have obtained small electric glow-lamps which produce an image approximating In shape to a V. The position of the point of the V, at the instant when the trigger is pressed, can be marked on the screen; and if the rifle

[graphic][merged small]

is moved during the act of trigger-pressing, the direction of motion, and its extent, can be marked by an arrow. If the position of the iioint of the V has previously been marked when the rifle was aimed by- an expert, the correctness or otherwise of the learner's sighting is seen at a glance. I Have found that most learners aim better than they shoot; that is, they sight the rifle on the bull's-eye with some approach to correctness, and then pull it away while they are actuating the trigger. If the learner is particularly bad at sighting, the rifle may be supported on a sand-bag or tripod stand, and sighting can be practiced until a satisfactory "triangle of error" is obtained.

I have found, by the aid of the appliance just described, that different people can aim a rifle with perfect consistency according to the rules given earlier iu this article.

English Measures of Length*

The History of Their Origin and Development

Bv Colonel Sir Charles M. Watson

Although there is considerable variety in the measures of length used by the different nations of the world, there can be no doubt that they are, for the most part, derived from a common origin, and that their ancestors, if the expression may lie used, existed in times so remote that the date of their invention has been completely lost. Hut the study of what the original measures were is a matter of considerable historical importance, and the question can be investigated by an examination of the changes made in the course of generations by the people who have adopted them—changes, in some cases, apparently due to accident rather than design.

For the sake of clearness, it is convenient to divide the measures of length into four categories which are, to a certain extent, independent of one another, and may he defined as follows:

(1) The shorter measures of length, used for building and manufacturing purposes, of which the more important in ancient times were the cubit, the palm, and the digit, or finger breadth, and the English representatives are the yard, the foot, and the inch.

(2) The shorter measures of distance, such as the foot, the yard, and the paee.

(3) The longer measures of distance, including the stadium, the mile, the parasang, the schoenos, the league, the hour's march, and the day's march.

(4) Measures of length used in connection with the calculation of land areas, of which the English representatives are the perch, the chain, and the furlong.

As regards the first of these classes of measures, it is generally accepted that they were, from the earliest times, based on the proportions of the human body, so that every man had his own scale to which he could work. As, however, men are not all of the same size, there is considerable variety in the length of the different units, hut, with some exceptions, they may be included within the following limits: The digit, or finger breadth,

from 0.72 l<> 0.75 English inch.

2.SS to 3 0(1 " inches. •• 17.28 to 18.OU

5.50 to ti.OO English feet.

The pah". The cubit. The fathom,

* From the Journal of the Royal Society of Arts.

The palm is the width across the open hand at the base of the fingers; the cubit is the length of the arm from the elbow to the end of the middle finger; and the fathom the length of the outstretched arms. There is no fixed relationship between these units.

There is no record as to when an attempt was first made to combine the measures in a standard scale, but it was probably at an early period, as it must have been found inconvenient for workers on the same building, for example, to use different lengths of palms and cubits, and, when a standard was fixed, it may have been some such scale as the following:

1 digit = 0.7375 English inch.
4 digits = 1 palm =2.95 " inches.
6 palms = 1 cubit = 17.70

The cubit of this scale may be called the "cubit of a man," to distinguish it from other cubits, which will be described hereafter.

In process of time it was found desirable to have a smaller unit than the digit, and this was made by taking it as equal to six grains of barley placed side by side. In the English scale, barleycorns were also used as the smallest measure of length, but in this case they were placed end to end, three barleycorns so placed being taken as equal to one inch.

There is no evidence that the foot was included originally among the units of the hand worker given above, and it may, perhaps, more properly be regarded as belonging to the second class of measures, derived from the distance covered by a man walking, and as a subdivision of the important unit, the pace. The pace is of two kinds, the first being the single pace, or distance covered by the step of one foot, and the second, the double pace, the distance covered by both feet one after the other.

In the case of the Roman double pace, a very important measure, the pace was taken as equal to five feet, but this was an artificial connection, as there is no fixed proportion between the length of a man's foot and the length of his pace.

There is nothing to show when the fool was added to the units of the mechanic's scale, but when this was done it was assumed to be equal to four palms, or two thirds of a cubit.

The third class of measures of length is the most im portant, and the history of these is of particular interest, as they appear to have started in a state of perfection, and to have been first used by a people who possessed a high degree of astronomical and mathematical knowledge, who were acquainted with the form of the earth, and were able to carry out accurate geodetical measurements. It is also remarkable that the changes made as regards these measures in the course of time have been changes for the worse, in consequence, apparently, of the origin of the measures having been forgotten. There can be no doubt that they are based on the angular division of the circle, and on the application of this division to terrestrial measurements.

The unit of angular measurement is the angle of an equilateral triangle, and this angle was divided by the ancient geometricians, for purposes of calculation, into 60 degrees, the best number possible, as 60 = 3 X 4 X 5. Following the same principle, each degree wras divided into 60 minutes, and each minute into 60 seconds. As the circle contains six times the angle of an equilateral triangle, the circle was divided into 360 degrees. This division of the circle, although so ancient that its origin is unknown, has never been improved upon, and is still in use by all nations. An attempt on the part of certain French mathematicians to substitute a division of the circle into 400 degrees, on account of the supposed advantages of the decimal system, has proved a failure.

The manner in which the division of the circle into 360 degrees was used by the ancients to determine the unit for terrestrial measures of distance was as follows: If a circle be described cutting the equator of the earth at right angles, and passing through the north and south poles, its circumference in angular measurement is equal to 360 degrees X 60 minutes = 21,600 minutes, and the length of 1 minute, measured on t he surface of the globe, was taken as the unit, which is called a geographical mile at the present time. If the earth was a perfect sphere, every geographical mile wrould be of the same length, but, as the polar diameter is less than the equatorial diameter in the proportion of 7,900 to 7.920, the length of the geographical mile, measured on the meridian, is not the same in all latitudes, but increases in length from 0,040 English feet at the equator to 0,108 English feet at the poles. Whether the ancient astronomers were acquainted with this irregularity in the figure of the earth is not possible to say, but it is certain that the value at which they fixed it must have been close to the actual mean value as determined by modern astronomers, which may be taken as about 6,075 English feet. The Greek stadion (the same as the Roman stadium), which was one tenth of the geographical mile, was 600 Greek feet in length, and the Greek foot was about 12.15 of our present English inches.

The next step taken appears to have been with the view of assimilating the subdivisions of the geographical mile with the cubit, and it was not easy to do this, as the cubit of a man has no necessary connection with a geographical mile. The difficulty appears to have been solved by the invention of two new cubits, of which the smaller was very nearly equal to the cubit of a man and was contained 4,000 times in the geographical mile. This, for the sake of distinction, may be called tho geographical cubit. The second cubit, afterward known as the Babylonian Royal cubit, was longer, and was contained 3,600 times in the geographical milo. According to Herodotus this second cubit was three digits longer than the other cubit. On these two cubits there appear to have been based two different divisions of the geographical mile, one in accordance with a decimal, and the other with a sexagesimal system of calculation, but there is, so far as I know, no ancient record of these scales, and the following attempt to compose them is founded on inferences, drawn from the Babylonian, Greek and Roman measures, all of which, there can be little doubt, came from the same origin.

Tho first, based on fhe geographical cubit, which was rather longer than the average cubit of a man, is as follows: 1 digit = 0.729 English inch.

25 digits = 1 geographical

cubit = 18.225 " inches. 100" = 1 fathom = 0 .075 English feet,

1(H) fathoms = 1 stadion = 007.5 10 stadia = 1 geographical

mile = 0075 Tho second, or sexagesimal scale, based on the Babylonian Royal cubit, appears to have been as follows:

1 digit = 0.723 English inch.

28 digits = 1 Royal cubit = 20.25 "inches. GO cubits = 1 plethron =101.25 English feet. GO plethra = 1 geographical

mile = 0075 "" Some writers are of opinion tliat the Babylonian Royal cubit was composed of 30 instead of 28 digits, but this appears to be improbable, because it would make the digit too small, and, if Herodotus is correct, it would make the cubit in the decimal scale consist of 27 digits, an inconvenient number. Nor is there any evidence to prove that a cubit was ever divided into 27 digits, while Prof. Petrie has shown, in "Inductive Metrology," that the division of the cubit into 25 digits, and of the fathom into 100 digits, is very probable. There was another Babylonian measure of length called the gar, used for land measurements, which appears to havo been composed of 12 Royal cubits. It was the ancestor of the English land measure, the perch, which is 11 English cubits in length.

The ancient Egyptian measures of length, although evidently derived from the same origin as the Babylonian, differ from these in some respects. The most important smaller unit was a cubit usually known as the Egyptian Royal cubit, which was divided into seven palms, each palm of four digits. The approximate length of the Egyptian Royal cubit is well known, as a number of cubit scales havo been found which give a mean length of 20.05 inches, and an examination of the monuments of Egypt shows that this cubit was used for building purposes from ancient times.

Prof. Petrie, in "Inductive Metrology," has given a large number of samples of the Egyptian cubit derived from the measurements of buildings, which vary from 20.42 to 20.84 English inches, and yield a mean value of 20.04 English inches, or almost exactly the same as the mean length of the cubit scales.

As is generally the caso with regard to measures of length in all countries, the Egyptian cubit appears to have grown longer in course of time, and there is a good instance of this shown by a comparison of the three nilometers on tho island of Phila\ above Assuan, of which tho first gives a mean valuo for the cubit of 20.47 English inches, the second of 20.81, and tho third of 21.05 Engish inches.

The best results given by Petrie are based on his measurements of the Great Pyramid of Gizoh, the great chamber of which, having a length of 20 cubits and a width of 10 cubits, yields a cubit of 20.027 English inches, while tho height of 78 palms gives a cubit of 20.05 English inches. The length of the side of the base of this pyramid is of particular interest, as it appears to have l>een designed as one eighth of a geographical mile. This length is not easy to measure, as tho lower part of the pyramid is covered with sand and rubbish, and tho

stones which cased it have been removed. Petrie, after very careful measurement, arrived at tho conclusion that it was 755.7 English feet. There are reasons for thinking it may have been a little more than this, but less than 760 feet. The length of a geographical mile at the latitude of the pyramid is 6,060 feet, and one eighth of this is 757.5 feet, or so nearly equal to the length of the side of tho base that it is difficult to believe that tho architect had not this in view when he designed the pyramid. The side of tho base was therefore equal to 500 geographical cubits, and very nearly equal to 440 Egyptian Royal cubits—a remarkable coincidence, if it is only a coincidence. It is interesting to note that there are 440 English cubits in the English furlong, but whether this has any connection with the measure of the pyramid there is no evidence to show.

There was a good reason for making tho sido of the base 440 cubits, as the height is equal to the radius of a circle, of wliich the perimeter of the base is the circumference, so that the height was 40 X 7 cubits, and the length of the side 40 X 11 cubits. It would be interesting to know how the ancient Egyptian geometrician arrived at so close an approximation to the value of x as ty.

It is matter of controversy from whence the Greeks derived their measures of length, whether from Egypt or Babylonia; but the latter appears more probable, as their principal measure of distance, the stadion, was equal to one tenth of a geographical mile of 6,075 English feet, and this was divided into 6 plethra, each of 100 Greek feet. The Greek scale appears to have been as follows: 1 Greek foot = 12.15 English inches. IH Greek ft. = 1 cubit = 18.225

10" " = 1 reed = 10.125 English feet.

10 reeds = 1 plethron =101.25 0 plethra = 1 stadion =607.50 10 stadia = 1 geographical

mile = 6,075 There was another foot used in Greece, of which Petrie gives a number of instances, derived from old buildings, varying from 11.43 to 11.74, with a mean value of 11.00 English inches. This would appear to be a foot of 10 digits, used for building and manufactures, but, not connected with measures of distance.

The Roman system of measures was based on the Greek, but wliile adopting the stadion—called by them stadium—as tho fundamental measure of distance, they used the shorter Greek foot, and introduced another measure, the double pace. They also made the land mile to consist of 8 instead of 10 stadia, while retaining the geographical mile of 10 stadia for use at sea. As they had an affection for a duodecimal system of calculation, they also divided the foot into 12 inches in addition to the old division into 16 digits. The Roman scale, which showed considerable ingenuity in assimilating a number of differert measures which had no real relationship to one another, appears to have been as follows: 1 digit = 0.729 English inch.

1 inch = 0.972

4 digits or

3 inches = 1 palm = 2.916 " inches.

4 palms = 1 foot = 11.004 0" =1 cubit = 17.496

5 feet = 1 pace = 4.80 English feet. 125 paces = 1 stadion =607.5

8 stadia = 1 land mile = 4,800 10" =1 geographical, or

sea mile = 0,075

The land mile was probably made up of 8 stadia in order to have it exactly 1,000 paces in length, or it may have been considered that eight was a more convenient number for dividing than ten; but it was necessary to retain the mile ot 10 stadia for navigation.

The above remarks deal with the measures of distance used by the principal nations of antiquity up to and including the geographical mile, upon which they seem to have been based, but in addition to these there, are certain longer measures of distance which must be referred to, such as the parasang, the schoenos, and the league. The fundamental idea of these measures was that they represented the distance which could be marched in a given time, such as one hour, and as the rate of marching naturally varied with the nature of the country, it was not easy to have a fixed length, and when there was made a theoretical unit it did not always agree with the actual distance.

There is a good example of this in the "Anabasis" of Xenophon, in which the writer recorded the distance traveled by the Greek force, day by day, on their way across Asia Minor from Ephesus to the Euphrates, and, after tho battle of C'unaxa, from the Euphrates to the Black Sea. Xenophon gives the distance from Ephesus to the battlefield as 535 parasangs, or 16,050 stadia, thus making the parasang equal to 30 stadia, or 3 geographical miles. But Col. Chesney has pointed out that the actual parasang, or hour's march, was less than this, and thai it averaged 26 stadia from Sardis to Thapsacus, and about 20 stadia from Thapsacus to the battlefield of Cunaxa. A fair average hour's march for an army would be 25 stadia, equal to 3 Roman miles, and a day's march of

eight hours to 20 geographical or 24 Roman miles. In the Antonine Itineraries the distance between important stations is, in a number of cases, given as 24 and 25 Roman miles, which looks as if the stations were fixed at distances apart suitablo for a day's march.

In Egypt the measure of distance corresponding to the parasang was the "alter," called "schoenos" by the Greeks, and stated by different writers to have been equal to 30, 32, 40, and 00 stadia in various parts of Egypt. It is evident that it was based on the geographical mile as a rule, while 32 stadia is equal to four Roman miles. There is some doubt whether the Egyptians had a fixed length for the schoenos, and a good deal has been written with regard to it, notably a paper entitled "Der Schoinos bei den Aegyptern, Griechen und Romern," by Wilhehn Schwarz, published at Berlin, 1894.

Another longer measure of distance, which was largely used in the western parts of Europe under the Roman Empire, was the Gallic league, equal to 12 stadia or one and a half Roman miles. In the Antonine Itineraries the distances in Gaul are in some cases given in leagues, and in others in both leagues and Roman miles, whilo in the "Bordeaux Pilgrim," a work dating from the early part of the fourth century, the distances in the west of France are given in leagues, and afterward in Roman miles.

An important application of measures of distance from the earliest times was for the calculation of areas of land, but there is considerable doubt as to what was the original unit, and whether this was a square, or in the form of a rectangle one stadium in length and one tenth of a stadium in width. In the latter case there would have been ten measures in a square stadium, and 1,000 measures in a square geographical mile, and such a measure would seem quite in accord with the ancient system of measures of distance. Its area would have been 40 X 400 geographical cubits (36 X 360 Babylonian Royal cubits), or 0.847 English statute acre. There is a very widely distributed type of land measures based on a rectangle of this form, of which the English acre is an instance, as it measures 44 X 440 English cubits.

The Egyptian unit of land area appears to havo been the "set," called "arura" by the Greeks, which was a square having a side of 100 Egyptian Royal cubits. A cubit of land was the one hundredth part of this, and was the area of a rectangle 1 X 100 cubits.

In the Greek system the unit of area was the square of a plethron or 100 Greek feot, equal to 0.235 English acre, of which there were 36 in a square stadion and 3,600 in a square geographical mile.

The Roman unit of land area, called the "jugerum," was a rectangle, 120 X 240 Roman feet, or 0.624 English acre, which was subdivided duodecimally, the uncia of land being the twelfth part of a jugerum, or the area of a rectangle measuring 10 X 240 Roman feet. The relative proportions of these different units of land area were as given below.

[table]

It will be seen from the above descriptions that from tho earliest times the shorter measures of length were based on the proportions of the human body, and the longer on the geographical mile, and that at some remote period an attempt was made to combine them into a continuous scale, from the digit to the geographical mile. When the digit was made the point of departure the decimal system of calculation appears to have been preferred, and when the scale was worked downward from the milo the sexagesimal system was the most convenient, while in the Roman scale the duodecimal system was introduced. But it is to be regretted that the more ancient system was not retained, by which tho geographical mile was the unit, and was divided into 10 stadia, each of 400 cubits, or 600 feet, as it is doubtful whether the changes made by succeeding generations can be regarded as improvements.

The modern measures of the civilized world are, with few exceptions, based on the ancient units, of which they may be regarded as the direct descendants. Of these exceptions the most important are the measures of the metric system, which were dosigned with the object of breaking away from tho records of the past by tho adoption of a new geographical milo, equal to 54/100 of the true geographical mile.

The English measures of length are a good example of the modern representatives of the old units, and are worthy of study from this point of view. How the measures originally came to England it is not easy to say, but there can be no doubt that they wore in use before the Roman invasion, having possibly been introduced by

Phoenician traders, and wore afterward modified by the Romans, the Saxons, the Scandinavians, and the Normans, each of whom had measures, based on the old units, but altered in course of time. It was not until the thirteenth century that they wero molded by law into one uniform system.

The English scale, as authorized by statute, may be summarized as follows:

1 inch.

12 inches = 1 foot. 3 feet = 1 yard. •*Vi yards = 1 rod, pole, or perch. 4 perches = 1 chain. 10 chains = 1 furlong. 8 furlongs = 1 English statute mile. Of these units tho inch is derived from the Roman system, being one twelfth of the foot, but the foot, on the other hand, is equal approximately to the Greek foot, while tho yard, which is simply a double cubit, comes from the Babylonian system, being approximately a double geographical cubit. The perch is the English representative of the Babylonian gar, and the furlong occupies a similar place to the stadium, while the mile is composed of eight stadia, apparently in imitation of the division of the Roman mile. For use at sea, however, the geographical mile, divided into ten stadia, or, as we call them, cable lengths, has been retained, as no other mile can be used for purposes of navigation.

In order to fully understand tho connection between the English measures and the ancient measures of length, it is necessary to write tho scale in a somewhat different manner, and to introduce some other units which are no longer used. The revised scale is as follows:

1 barleycorn. 3 barleycorns = 1 inch.

3 inches = 1 palm.

4 palms = 1 foot,
fi palms = 1 cubit.

12 palms = 1 double cubit or yard.

11 cubits = 1 perch.

405 cubits = 1 cable's length.

4 perches = 1 acre's breadth or chain.

10 chains = 1 acre's length or furlong.

8 furlongs = 1 English mile.

10 cables = geographical, or sea mile.

The English inch is equal in length to 3 barleycorns set end to end, or to the width or 8 barleycorns set side by side. The barleycorn, as a measure, is forgotten, but the inch on carpenters' rulers is still divided into eight parts, while on a shoemaker's tape the sizes of boots and shoes increase by a barleycorn or y& inch, for every size. For example: size No. 8 of a man's boot measures 11 inches; size No. 9, llj^ inches; size No. 10, 11% inches; and so on. One would havo thought that the sizes would increase by one quarter of an inch at a time, but the barleycorn has held its place to the present day.

The palm, which was originally composed of 4 digits or finger breadths, and, since the time of the Romans, of 3 inches or thumb breadths, is no longer used in England, and its place has to a certain extent been taken by a measure called the hand, composed of 4 inches and employed in measuring the height of horses. The change may have been due to the fact that the number 4 was more convenient for division than 3, and that when the digit gave way to the inch the palm of 4 digits was replaced by the hand of 4 inches.

Prior to the thirteenth century, the length of the foot in England was uncertain, and there appear to have been several measures in use, varying from the Roman foot of 11.66 English inches to tho Belgic toot of 13.12 English inches; but, by the ordinance known as tho Statute for Measuring Land, enacted in the reign of King Henry III., the relations of the inch, the foot, and the cubit to one another wero definitely fixed, and havo never since been altered. The cubit of this statute is the double cubit, afterward (railed tho yard. A translation of the Latin words of the statute, describing the different measures, is as follows:

"It is ordained that 3 grains of barloy, dry and round, make an inch; 12 inches make a foot; 3 feet make a cubit; 5^2 cubits make a perch; 40 perches in length and 4 perches in breadth make an acre.

"And it is to be remembered that tho iron cubit of our Lord the King contains 3 feet and no more; and the foot must contain 12 inches, measured by tho correct measure of this kind of cubit; that is to say, ono thirty-sixth part of the said cubit makes one inch, neither more nor less. And 5H cubits, or 16 feet, make ono perch, in accordance with tho above-described iron cubit of our Lord tho King."

It is interesting that, in this statute, tho doublo cubit, thus accurately described, should have been called tho cubit of the King, just as the longer cubits of Babylon and of Egypt were called Royal cubits to distinguish them from the shorter cubits of those countries. In the Latin original of the ordinance the word used is "ulna," the usual word for cubit. The word "yard," to signify the English double cubit, occurs for the first timo in the

laws of England in a statute of 1483, written in French.

The perch, equal to 11 single or 5J^ double cubits, is a very ancient measure, but I cannot find at what period it was first used in England. It was employed principally in connection with the measurement of land, and I have already pointed out its likeness to the Babylonian measure the gar, composed of 12 Babylonian cubits.

The two measures, the acre's breadth, afterward called the chain, and the acre's length or furlong, have also been used from a very early period. The former is equal to 44 single cubits, 22 yards, or 66 English feet, while the latter is exactly ton times this, 440 cubits, 220 yards, or 660 feet. The furlong is the modern representative in our system of the ancient stadium, which had a length of 600 Greek feet, or 607.5 Erglish feet, but the reason for its being longer than the stadium has so far as I know not been satisfactorily explained. But the change may have been due to the fact that other measures of distance were in use in England prior to the present statute mile, which varied in different parts of the country, and the mean of these was approximately equal to the Gallic league of 12 stadia or 7,290 English feet. One-eleventh of this, 663 English feet, is approximately equal to the English furlong, and eight of these measures, following the Roman system, were combined to form the English statute mile.

But whether this is the origin or not, there appears little doubt that the mile, furlong, and chain, or acre's breadth, were in use in England in Anglo-Saxon times, as there is a law of King Athelstane, who reigned A. D. 925-940, in which it is enacted:

"Thus far shall be the King's grith from his burgh gate where he is dwelling, on its four sides; that is three miles, and three furlongs, and three acre's breadth, and nine feet, and nine palms, and nine barleycorns.

The length of the measure called the King's grith, or King's peace, was the distance from his house within which peace was to be maintained, and it is evident that in this law an attempt was made to express the distance in terms of ordinary measures. Converting these terms into feet we have:

3 miles = 3 X 5,280 feet = 15,840 foet.

3 furlongs = 3 X 660 " = 1,980"

3 acre's breadth = 3 X 66 " = 198" 9 feet = 9 " - 9"

9 palms = 2M " ="

9 barleycorns = l/i foot = lA foot.

Total 18,029}^ feet. 18,029}^ = 601 X 30 very nearly, so that it would appear that the length of the King's grith was 30 stadia, the same measure as that known in the East as the parasang, and in Egypt as the schoenos. It is remarkable that this measure should thus appear to have found its way to England, and there be regarded as a Royal measure.

There was another measure of distance used in England, known as the leuga, composed of 12 furlongs, which corresponded to the Gallic league of 12 stadia already described. In the Chronicles of Battle Abbey, which extend over the period A. D. 1066-1176, in the account of the lands belonging to the abbey, the following statement occurs: "The English leuga contains 12 roods (furlongs), and 40 perches make one rood; the perch is 16 feet in length; the acre is 40 perches in length and 4 in breadth; but if it is 20 perches in length, it shall bo 8 in breadth." The acre's length, here called rood, and the acre's breadth appear to have been the same as in the time of King Athlestane, and the foot is the Saxon foot, equal to 12.375 of our present inches. The measurement of the acre, 4 X 40 perches, is the same as that given in the Statute for Measuring Land enacted in the reign of King Henry III., which has already been referred to.

The terms acre's length and rood are no longer used, and this measure is now known as the furlong, while the acre's breadth has been called the chain since the beginning of the seventeenth century, when it was divided into 100 links instead of 66 feet. The chain, which was the invention of Prof. Gunter, has proved vory convenient for the measurement of land acres, and is always used.

Since the introduction of the chain, the perch or rod has been less employed in connection with land measures, but is still used by builders for the measurement of brickwork. Tho common English stock brick is half a cubit in length, one quarter of a cubit in width, and one sixth of a oubit in thickness, or rather less than these dimensions, to allow for the thickness of the mortar joints, while a rod of brickwork, which is the unit for builders' work, is a mass of brickwork, one rod or 22 bricks in length, one rod or 66 bricks in height, and tlireo bricks in thickness. Tho perch or rod of brickwork containo 4,356 bricks.

The English sea mile is exactly the same as the geographical milo of tho Babylonian system, and its tenth part, the cable length, is identical with the stadium. In these measures there has been no change, and the only difference is that the cable length is 40.5 English cubits, whereas the stadium was 400 original cubits. This is due to the fact that tho English cubit is a little shorter than the latter in consequence of the English foot, as

fixed by law, being rather less than 1/6,000 part of tho geographical mile.

The Medical Needs of Modern Armies

Interesting side lights on the need for a sufficient supply of medical officers In war are shed by correspondence to the London Lancet of November 20, 1915. The secretaries of the Harvelan Society called a meeting of that society to discuss the organization of the British medical profession for war service, and in their official announcement said: "This topic is assuming very great. proiM)rtions, for the actual personnel of the Army Medical Service already be approaching 10,000 in place of the peace establishment of 1,000." And on the same page another communication says: "The authorities at the War Office are very uneasy about the supply of doctors to the Army. At their request a War Emergency Committee in connection with the British Medical Association has been established, and a committee is working with all its power to see if they can find by the middle of January some 2,000 odd medical men which the War Office deem necessary."

It appears, then, that the British army will shortly include no less than 12,000 physicians with the colors, in order to satisfy present immediate needs in the medical service. This despite the fact that the main bulk of the fighting is in Flanders, where the conditions of trench warfare and short haul permit of great economies in medical personnel through the ability to eliminate many of the intermediate sanitary formations which are ordinarily required to bridge the gap between the firing line and the base.

Clearly the situation as to medical officers in which Great Britain finds itself has many morals for us. We must appreciate that in the United States, too, the day of little things is over and that no question of national preparedness Is complete, without the inclusion of the medical profession in civil life in terms of many thousands. Such immense numbers of troops will be necessary that any regular medical personnel which could be maintained in peace will scarcely suffice to leaven the mass in war. Efficient administrative machinery must be created and maintained to secure and instruct in the elements of their medico-military duties the vast mass of medical men whose services will be required. This additional work will chiefly fall upon the Medical Corps of the army, now both actually and relatively too small to do its routine work in time of peace. Efficiency demands that not only must the Medical Corps be given enough officers in the coming defense plans to do the peace work of the standing army, but it must share proportionately in the large extra and unassigned list of officer instructors which all schemes for defense agree upon as being absolutely necessary for the education and training of the second line forces.

An army is a many-sided and very complex structure, every part of which has a definite relation, usefulness and proportion to the rest. This is a fact of which many of the civilians upon which we have to depend for service legislation are ignorant, and one which a certain few line officers who are better informed seem to choose to ignore. To attempt to pile up fighting men without sufficient medical personnel to maintain physical efficiency defeats any expectation of securing maximum fighting strength. The Medical Department wants only what is necessary. It must insist on having enough officers in peace time to make it feel a reasonable competence to perform satisfactorily the tasks which devolve ui>on It in peace. It will expect in time of war a i>crsonnel adequate to iierform war duties. With anything less than this reasonable provision it will not be satisfied.—From the Military Surgem.

[See "The Medical Reserve Corps, U. S. Army," in Scientific American Supplement No. 2075, October 9, 1915.]

Planting Trees With Dynamite

So much is now being written about the use of high explosives in war that many are apt to lose sight of the fact that this same agent is an exceedingly useful servant for peaceful operations, and that new and ingenious methods of applying are constantly being devised. A recent novel application of the powers of dynamite was for planting trees. There was an apple orchard of four thousand trees to be planted, and as winter was approaching, no time could be lost, lest a sudden turn in temperature should freeze the ground. Tho man who undertook the work first mounted a 2% horse-power gasoline engine on the running-gear of a light farin wagon, and arranged it to operate a soilauger, and with this outfit two men were able to put down as ninny holes in a day as thirty men could have punched with a bar and sledge. In these holes light charges of dynamite were exploded to form an excavation in which to plant the trees, a number of holes being tired at a time. By this method the entire orchard was planted in less than 15 days of 9 hours each.

Battery Versus Magneto Ignition*
By Frank Conrad1

Thk mechanism for igniting the charge in the cylinder of an explosive engine is one of the foremost problems of design with which the automobile engineer has to contend. Its general development is one of evolution, a condition which is true of all new problems in engineering.

The ignition mechanism on the early automobile engines was copied from the system as used on stationary engines of that day. This was the so-called make-andbreak system in which an electric circuit was mechanically closed and opened through contact points placed inside the cylinder. The electric circuit usually consisted of a simple reactance coil in series with a few cells of primary battery. The inductive discliarge from the coil produced a spark on separation of tbe contacts which ignited the charge in the cylinder. Under limited conditions this system leaves little to be desired. The electrical apparatus is simple and easily understood. The ignition spark is very hot and if the mechanical part is well made its operation is practically perfect.

ADVANTAGE OF JUMP-SPARK SYSTEM.

The next- scheme to be proposed was the jump-spark or high-tension system in which the spark electrodes in tbe cylinder are fixed with a small separation between which the spark is produced at the proper instant by applying a voltage sufficiently high to jump the gap. This arrangement has the advantage that there are no working parts in the combustion space and therefore the parts are mechanically much more simple than the lowtension system. This system had the apparent disadvantage of giving a much weaker spark, and owing to the high voltage this spark might be further weakened by leakage; l>etween the parts forming the terminal supports and the connecting wires. These problems were more readily solved than the ones connected with the makeand-break mechanism. In fact, it can be stated that the modern high-speed gasoline engine would be impracticable with make-and-break ignition. For the slow-speed stationary engine, especially where, due to a low-grade fuel being used, compression is carried to a high value, the low-tension system has, however, held its own.

In the high-tension system the mechanism connected with the sparking points has been simple and tho problem of insulation of these points has been solved to a satisfactory degree. It is, therefore, upon tbe mechanism which furnishes the high-tension current at these points that our attention is now centered. The first and most obvious solution of this problem was to use an ordinary vibrating contact induction coil, the primary circuit of which was supplied from a primary battery, and in the secondary circuit of which was induced a voltage sufficient to jump the space between the electrodes in the cylinder, these electrodes and their mounting forming the now well-known spark plug. The primary circuit of the induction coil was controlled by a contact device which served to close this circuit at the proper instant. The shortcomings of this system were mainly in the battery which supplied the electric energy to the coil, and in order to obtain a reasonable life from the type of batteries available for this service, it Whs necessary to reduce the energy consumption to the minimum that would give satisfactory ignition. To overcome this defect, designers turned their attention to the matter ot supplying this energy from a mechanical generator which would obtain its power from the gasoline engine. As the only object of this generator was to supply energy to the induction coil, it was evident that this coil and its contact mechanism would be combined in one piece of apparatus with the generator.

IGNITION BY SINGLE SPARK.

As a simple alternating current generator which gave a comparatively low trequency was used, it was found necessary to abandon the magnetically operated vibrator and substitute a mechanically operated one which would open and close tie primary circuit of the induction coil at definite points on the voltage wave induced in the windings of the generator armature. This, of course, gave but one spark in the cylinder instead of the series of sparks induced by the vibrating coil, but as it is necessary to obtain proper timing of the explosion that the first spark ignite the mixture, the succeeding sparks are superfluois, although it is possible that should the first spark, owing to unfavorable conditions, fail to ignite the explosive mixture, a succeeding spark may do so. This condition can be practically overcome by supplying more energy to the single initial spark, which is permissible when this energy is generated by a mechanical device.

As a generator with permanent magnet field gives the simplest arrangement and has the advantage of overcoming the time element necessary for any electromagnet field to be built up, the permanent type is universally used and the name magneto has by usage in automobile engineering circles been taken to cover generating and spark producing mechanism as a whole.

* Heart before the Society of Automobile Kngineers. 1 Electrical Engineer, Wostinghousi Klictrir and Manufacturing Company.

The advent of electric lighting, and later of electric starting of the gasoline engine, produced demands for electric energy on the automobile which could only bo mot by the use of a much more efficient type of generator than that which served for the supply of ignition energy. As it is necessary to have power when the engine is not running, a storage battery is required. This condition is being met by the variable-speed battery-charging generators now on the market. The presence on the automobile of a supply of electrical energy much greater than required for simple ignition seems, therefore, to render superfluous the use of a separate generator for the ignition system. Naturally, in view ot this, the first impulse would be to go back to the original vibrating coil system with its primary battery energy being supplied from the lighting and starting system instead. This system, however, has certain disadvantages which have been in large measure corrected in the development of the magneto generator, the principal ones being the time which elapses between the closing of the primary circuit and the production of the spark at the plug and the limitation which the vibrating contact mechanism places on the amount of energy it is possible to deliver to the spark plug. As this time is constant and independent of engine speed, it is obvious that the spark would occur progressively later as the engine speed increased. To overcome these defects it has been necessary to design a mechanism in which the primary contact points are operated mechanically the same as in the case of the magneto. As the spark is produced at the opening of these contacts, and this opening occurs at a definite angular position of the engine crankshaft, the spark is not regarded as the speed increases, thereby necessitating only the amount of angular advance required for the complete combustion of the exploding charge, and the type of contact device also permits of increasing the amount of energy which may be deliverod to the spark plugs. The remaining distinction between the operation or a magneto and a bat tery system lies in a difference in the action which takes place in the induction coil.

ENERGY CONDITIONS IN BATTERY SYSTEM.

In the battery system the total energy to be supplied to the spark plug is delivered through the primary^ winding of the induction coil and stored as magnetic, energy in the magnetic circuit of this coil. On opening the primary circuit this energy, minus the incidental losses, is delivered through the secondary winding to the spark plug. An oscillograph curve of the current in the spark plug circuit show's that this current rises instantaneously to a definite value from which value it gradually falls off to zero. The amount to which the circuit rises is determined by the value of the primary current and ratio of primary to secondary turns; while the time required tor this current to fall to zero is determined by the energystored in the magnetic circuit of the induction coil and by the resistance of the secondary' or spark plug circuit.

In the case of the magneto the inital current value through the spark plug is, as in the case of the coil operated from a battery, determined by the primary current and ratio of turns. Its subsequent value, however, may be modified by the fact that energy can l>c delivered directly to this secondary through the mechanical motion of the generator armature itself. The time required fol the current in the spark-plug circuit to reach zero may, therefore, be prolonged over that which would be possible in the case of the simple induction coil system. So far as performing its function of igniting the explosion charge is concerned, this prolonged spark can lx> of no value as, in order to properly develop the power in the cylinder, the combustion of the charge which is started at the spark plug points must take place through the whole volume of charge in an extremely limited time as compared to one complete engine oycle. No increase of power could, therefore, be obtained byr maintaining the spark at the spark-plug points after the explosive mixtun; in the vicinity of these points has been burned. This effect of prolonging tho duration of spark may be very deceptive in comparing the relative intensities of the spark produced by different systems by observations. Thus, a greatry prolonged spark of comparatively low current value may appear much hotter than one of short duration but higher current value, although this latter would be the more efficient igniting spark.

LIMITATIONS IN MAGNETO VOLTAGE.

A further distinction between magneto and battery ignition lies in the limitations on possible secondary' voltage it is practical to generate in the magneto equipment, especially at the lower engine speeds. In tho batterysystem it is possible to wind the secondary coil for any required voltage. In the case of the magneto there are limitations of design, due particularly to t he limited space in which it is possible to place a secondary winding and the difficulty of obtaining clearances necessary for effective insulation. This condition can be observed by noting the size of the magneto required to give efficient ignition when the jump-spark system is used on the comparatively slow-speed high-compression stationary engine.

I have not gone into any details of the construction or operation of either the magneto or the battery system,

as it can be seen from the foregoing that there is no fundamental difference between the two systems, and the method used to work out the problem of any particular design would have no bearing on the question under discussion. That the ignition system in which the primary source of electrical energy is the generator, which can supply in common all electrical demands of the automobile, is the most rational solution, is evidenced by a study of changes in the type of equipment used within the last few years.

SCIENTIFIC AMERICAN SUPPLEMENT

Founded 1876

NEW YORK, SATURDAY, FEBRUARY 19, 1916.

Published weekly by Munn & Company, Incorporated Charles Allen Munn, President; Frederick Converse Beach, Secretary; Orson D. Munn, Treasurer; all at 233 Broadway, New York

Entered at Post Office of New York, N. Y., as Second Class Matter Copyright 1916 by Munn & Co., Inc.

The Scientific American Publications

Scientific American (established 1845) ...." 3.00 Scientific American Supplement (established 1876) per year $5.00 The combined subscription rates and rates to foreign countries, including Canada, will he furnished upon application Remit by postal or express money order, bank draft or check

Munn & Co., Inc., 233 Broadway, New York

The purpose of the Supplement is to publish the more important announcements of distinguished technologists, to digest significant articles that appear in European publications, and altogether to reflect the most advanced thought in science and industry throughout the world.

Back Numbers of the Scientific American
Supplement

Supplements bearing a date earlier than January 3rd, 1914, can be supplied by the H. W. Wilson Company, 30 Mamaroneck Avenue, White Plains, N. Y. Please order such back numbers from the Wilson Company. Supplements for January 3rd, 1914, and subsequent issues can be supplied at 10 cents each by Munn c% Co., Inc., 233 Broadway, New York.

We wish to call attention to the fact that we are In a position to render competent services In every branch of patent or trade-mark work. Our staff Is composed of mechauical, electrical and chemical experts, thoroughly trained to prepare and prosecute all patent applications, irrespective of the complex nature of the subject matter involved, or of the specialized, technical, or scientific knowledge required therefor.

We also have associates throughout the world, who assist in the prosecution of patent and trade-mark applications filed in all countries foreign to the United States.

Munn & Co.,
Patent Solicitors,
233 Broadway,
New York, N. Y.

I!ranch Office:

625 F Street, N. W., Washington, D. C.

Table of Contents

PAOE

Mysteries of Matter.—By John Candee Dean 11-1

The Distortion of Iron Castings 110

(■olden ltod 115

Modern Science and War Surgery.—6 illustrations 111!

Latent Heat of Fusion of Ice 110

Stellite lit!

Engineering Education Kaults 117

Co-Operation in Foreign Trade.—By Hon. Joseph E.

Davles 118

Emulsions and Emulslncations 11:'

Protective Coatings for Small Articles 119

A Novel Construction for (ias Holders 11S»

Kapok—A New Textile liber.—By Jacques Buyer.—8

illustrations 120

Imitation as Pioneer of the Genuine 121

Correspondence- The Simplex Calendar 122

The Locomotive and the Revolutionist 123

Aiming With tbe Klfle. By Edwin Edser.—5 Illustrations 124

English Measures of Length.- By t'ol. Sir Charles M.

Watson 125

Tlie Medical Needs of Modern Armies 127

Planting Trees Witli Dynamite 127

Battery Versus Magneto Ignition.--By Prank Conrad .... 128

« PreviousContinue »