qualified agent has been sent to Europe to acquaint himself with the whole agricultural and manufacturing business that produces sugar.

On the present occasion, we invite the patrons of our country's industry and resources, to communicate for publication, what they know on the above interesting branch of French husbandry, &c. And we therefore request the wealthy and patriotic, to consider whether the case of oil does not resemble that of the sugar from the beet, and whether the best course would not be to adopt a plan similar to that which the friends of beet sugar have chosen.

The time will come when American parents will send their sons to Europe and to other foreign places, to learn the manufacture of beet sugar, of oil, and such other branches of the arts not possessed by us, in the same manner and with better reason that they now do to have them learn medicine and surgery:

June 4, 1836.

J. R.

Civil Engineering.

Some suggestions on the Location and Grading of Rail Roads.

By THOMAS EARLE.

In the location and grading of rail roads, it is usual to reduce the road in all parts, as near to a level as possible, and in effecting this object, to make many curvatures, some of them of small radius. Thus, a very consider. able increase of expenditure and of distance is occasioned, which appears to me inexpedient.

It is true, that if a rail road could be made perfectly level, or very nearly so, without being unreasonably curved, such a road would be better than an undulating one: because the locomotive engines would require to be transported less frequently over the ground, to convey a certain quantity of goods on such a road, although the expenditure of steam for conveying the train, independent of the locomotive, would be as great on the level road as on the undulating one. A perfectly level road, however, is impracticable in most parts of the country, except at an expense far exceeding the value of the benefit gained. Hence, it is probable that few roads will hereafter be made, without ascents and descents, in some parts, at the rate of forty to fifty feet per mile.

Such ascents being admitted in some parts of the road, the locomotives will take no greater trains than they can draw up those ascents. Hence, it will be useless to make excavations, embankments, and curvatures, to avoid other ascents of the same grade.

A locomotive will take a train up an ascent of twenty-one feet to the mile, and down a descent of the same length and grade, with precisely the same expenditure of steam, if it be constantly used, as would be required to take the same train over the same distance, on a perfect level. If the train were such as to require for drawing it on a level, a pressure of steam on the piston of thirty-six pounds to the inch, above the atmosphere, then on the ascent of twenty-one feet per mile, it would require sixty-three pounds per inch, and on the descent of the same grade, nine pounds per inch, making the average thirty-six pounds or the same as on the level. Thus, 63+7= 72+2=36 pounds.

If, however, the road were composed of alternate ascents and descents, at the rate of from thirty to forty feet per mile, with but short levels between them, the engine would transport such train as it could draw on the road,

with a less expenditure of steam than it would require to transport the same train on a level. This might be effected by shutting off the steam from the piston on descents, and suffering the train to progress by its own gravity. The saving in this case, compared with the other, would arise from dispensing, on half the distance, with the amount of steam, viz: about fifteen pounds to the inch, which is required to overcome the external resistance of the atmosphere. There would also be a further saving from the constant use of high steam, if the supposed fact be correct, that a certain volume of steam under a pressure of one hundred pounds to the inch, can be produced with less than double the fuel which is required to produce the same volume of steam under a pressure of fifty pounds to the inch.

And the result, as to the expenditure of steam, will be equally favourable on ascents and descents, as great as fifty feet to the mile, (excepting the before-mentioned inconvenience of transporting the engine a greater number of times over the ground) as with ascents and descents of a less grade, provided the inclined planes be not so long as to require the checking of the velocity of the train, by artificial means in descending: for the momentum acquired in the descent, will continue the motion on the succeeding level or ascent, until the power expended in overcoming gravity in ascending, is reimbursed.

It is further to be observed, that if there be admitted on a road, inclined planes of several miles in length, and of a certain grade, shorter planes of a higher grade may be admitted on the same road, without inconvenience, because the momentum acquired by the velocity of the train, before commencing an ascent, will give considerable assistance in overcoming it. A velocity of twenty miles per hour would give a momentum, if I have estimated it rightly, sufficient to raise the train about twelve feet in perpenticular height. Thus a road having long inclined planes, graded at forty feet per mile, will admit those not exceeding two thirds of a mile in length graded at fifty feet, or not exceeding one third of a mile in length, at sixty feet.

Hence, it is unnecessary, on long inclined planes intended for locomotives, to make them of uniform ascent, as the momentum gained where the ascent is below the average, will assist in overcoming the resistance where it is above.

The making of curves in rail roads, to avoid slight ascents and descents, is productive of several inconveniencies.

1. It increases the cost of the road, by its greater length, and proportionably greater expenditure for land, foundation and rails.

2. By the increase of length, the time of travel and the expenditure of steam, is increased in nearly the same proportion.

3. The expenditure of steam is further increased, in overcoming the strain and friction occasioned by the operation of the wheels on the curves, the power expended not being re-imbursed, like that expended in overcoming ascents. The resistance on short curves upon a level is found to be greater than on a straight ascent of thirty feet per mile.

4. The wear of carriages and locomotives, and their liability to break or become disordered, is increased by the greater distance, and by the strain on the curves, which racks every part of the machinery to a degree much complained of by practical engineers.

5. The wear of the rails, and their liability to disorder is increased. 6. The danger of running off the road is increased.

Hence, a road should be made as straight as possible, without a great increase of expense, and without encountering ascents unreasonably great.

I will add a suggestion in relation to cars for burthen. The greater the load carried by each car, the less will be the weight and cost of cars, compared with the goods transported. Materials increase in strength in proportion to the cube of the diameter, while the weight and volume increase in proportion to the square. Hence, the cost of materials, workmanship, and transportation of cars, will all be reduced, by using as strong cars with as great loads as the road will permit. A further advantage in strong cars and heavy loads to each, will be found in shortening the train, and thus decreas ing the strain in turning curves. As locomotive engines with six wheels are used, weighing with their water and fuel, eighteen or twenty thousand pounds, I can see no serious objection to the use of burthen cars of four wheels, weighing with their load, five and a half or six tons, with a proportionate increase of weight when six or eight wheels are used.

Objections have been made to the matter contained in the forepart of this essay: 1. That on an undulating road, the steam must be blown off and wasted on descents, owing to its superabundant quantity: 2. That although none were used on descents, the pressure of steam could not be kept up sufficiently, because none would pass through the flue to aid the draught. The two objections are contradictory of each other. They can both be obviated by proper power in the boiler, with an adequate steam chamber, and by proper attention to the supplies of fire and water. They are not felt as serious inconveniencies with the best engines on the Columbia Rail Road. There is one plane on that road of upwards of ten miles, and another of seven miles, where the cars will descend by their gravity. The engineers cause a fresh supply of water to be put in the boiler at the head of the plane, and no fuel till near the foot of it, and thus they avoid the necessity of blowing off steam. If they were to add fuel, they would have to discharge steam, which shows that both objections are of little importance, in comparison with the advantages of a straight road at a diminished expense.

If it should be found that without particular attention to the addition of water and fuel on descents of moderate length, where the train progresses by gravity, there will be an inconvenient surplus of steam, the difficulty can be obviated by the use of a damper to check the draught.

Physical Science.

Proposed forms of diagrams for exhibiting to the eye the results of a register of the direction of the wind. By A. D. BACHE, Prof. Nat. Philos. and Chem., Univ. Penn.

My attention has been recently recalled to the subject of diagrams for showing the results of a register of the direction of the wind, by the first number of a meteorological publication,* received through the politeness of its author, W. R. Birt, Esq. of London. At one of the early meetings of the joint committee of the American Philosophical Society and Franklin Institute, appointed in 1834, I laid before the members several plans for the purpose above referred to. These, I propose now to make public in the

*Tabula Anemologica, or tables of the wind; exhibiting a new method of regis tering the direction of the wind, &c. &c. By W. R. Birt.

hope that one or other of them may prove acceptable to meteorologists. They exhibit to the eye the results of observations at the same or different places, thus facilitating the study of their connexion. One of the plans was considered preferable to the others by my colleagues of the committee, but as it may not be the most convenient under all circumstances, I have presented the varieties of the register as laid before the Committee. The scheme shown in fig. 4, plate 1, will be found to resemble in appearance that proposed by Mr. Birt; but the principle will be seen on examination to be entirely different from the one adopted by him.

The figures are placed in the order in which the methods suggested themselves. The first is probably the most natural form of diagram, and was the first which occurred to me, while the others are successive modifications growing out of difficulties, or objections, which appeared in studying the subject. The first was preferred by my colleagues of the committee as best accomplishing the object, while the last is adapted to the ordinary form of diagram used to represent the variations of the thermometer, barometer, &c.

A register of the wind should not only show its direction at the time of observation, but the direction through which it may have passed when changing from one point of the compass to another. A diagram illustrating such a register must admit of an easy mode of expressing the results, and the less artificial this method the better will it answer the purpose of addressing the eye. In his valuable meteorological essays Professor Daniell has adopted a method of representation first used, I believe by Mr. Howard. A horizontal line is drawn and points assumed upon it at convenient, equal distances, to represent the times of observation. Above this line points are assumed at regular intervals, to denote the points of the horizon between west and east, by the north. Supposing the cardinal and ordinal points only to be marked; the north west point will be on the left hand, and the east on the extreme right. The positions assumed for the points of the horizon will of course depend upon the degree of nicety to which it is intended to note the direction of the wind. From any one of these points to one of those in the horizontal line first assumed, representing the times of observation, a line being drawn represents the direction of the wind. A similar arrangement is made below the first horizontal line or at the foot of the diagram, if it is also to exhibit the state of the barometer, thermometer, &c., for the points from west to east, passing through the south. This method does not admit easily of expressing the direction through which a wind has changed, and the lines of direction of the wind, sometimes cross each other at such acute angles as to render it difficult to trace them. For example, when at the close of a month the wind is north westerly for several successive days, the lines expressing this fact cross the whole figure. They meet other lines sometimes quite obliquely, and being but slightly inclined to each other, the eye does not readily follow them. These remarks are not offered in the spirit of criticism, but merely to point out why I thought it advisable to obtain a different scheme of registry.

The ordinary method of representing the rise and fall of the thermometer or barometer, is a natural one; equidistant points on a horizontal line being taken to represent the times of observation and the perpendicular lines drawn through these points, or corresponding ordinates, being made proportional to the height of the column of mercury. In like manner the wind being registered with reference to the points of the horizon, the natural system of representing it is to assume a system of concentric circles the in

tervals between which shall correspond to equal intervals between the times of observation, and the angular divisions upon which shall correspond to the rhumbs. Such a scheme is represented in figure 1, plate 1. In order to bring the figure within the compass of the page, it has been necessary to make it so small that it does not fully show the advantages of the plan. A diagram in which the outer circle is seven or eight inches in diameter admits of entire distinctness, when observations are not more frequent than four times during twenty-four hours, even at the season when the wind is most variable. Where observations are frequent, the interval between the concentric circles may conveniently represent a day, the first circle corresponding to twelve o'clock at night, on the last day of the preceding month, and the second to midnight of the first of the month, and so on. The observations at intermediate hours will be placed in their appropriate positions between the two circles just referred to, and the registry will be carried on in a similar manner throughout the month. In the case actually represented, fig. 1, plate 1, the regular observations were at S P. M. of each day, and I have drawn the concentric circles, represented by the finer lines, to correspond to this time. The intermediate observations when a change of wind required their use, have been placed within or without the several circles, according as they were made before or after 3 o'clock of the particular day to which the circle corresponds. The variable month of April has been selected for representation, as putting the diagram to a severe test. The circle having been divided as shown in the figure, so as to point out the cardinal or ordinal points, a dot is placed on that radius cor responding to the point from which the wind blows. Thus, on the 1st of April, 1836, at 3 P. M. the wind at Philadelphia was S. S. W. the dot numbered 1, is on the first circle recking from the centre, at the intersection of a radius, which would bisect the angle S-W. c. S. On the 2d at 3 P. M. the wind was N. W. a dot is therefore placed to denote this on the second circle, at the intersection of the radius N-W, c. The table from whence the direction of the wind was taken shows that the wind passed from the S. W. to the N. W. through the west as is expressed by the curve 1, 2, passing through the west point. On the Sd the wind was S. W. as denoted by the dot at 3, having passed back by the west, which direction therefore the curve 2, 3, is made to intersect. The wind passed the S. W. to S. S. W. between the 3d and 4th, as is shown by the curve 3, 4. It remained at this point until the morning of the 5th, as indicated by the straight line from 4, when it changed by the west to N. N. W. Passing this point to the north in the evening of the 5th, it returned to the N. W. on the 6th. It is hardly necessary to trace the courses further to show how the diagram represents the results of the table, but it is probably worth while to refer to one of the cases, when the change of the direction of the wind is not through the smaller angle between its two directions. On the 10th of April the wind was N. E. as indicated by the dot 10; on the 11th, it was N. W. having changed round by the South as fully represented in the curve 10, 11, which sweeps through the angle of 270°.

One objection occurs to the curves between the times of observation, namely, that they represent the wind as gradually changing, whereas, in fact it frequently ceases, an entire calm preceding the wind from the new direction. This false impression is entirely avoided in such a case, by a stem of numbers, or symbols, representing the force of the wind. These

*Kindly loaned to me for this purpose, by James P. Espy, Esq.