We also, for the same purpose of proving the confidence that may be placed in an investigation by these gentlemen, shall state shortly their mode of proceeding. When a subject is decided by them to be worthy of experiment, a director of the investigation is appointed, at whose command all the means that the society, as a body, or each individual so disposed, can furnish; and to avoid distraction and confusion, and ensure effective cooperation, his instructions are implicitly followed. At the conclusion of an experiment, made under these circumstances, the whole of the members present discuss the proceeding, and suggest any omissions they may have observed; if these are important, the experiment is repeated, and so on, until every doubt of every individual is removed, and unanimity obtained *. This result cannot always be arrived at by one experiment, or in one meeting; it was not on the subject in question, but the process is repeated until it is accomplished.

It was under such a procedure that the fact described by Sir David Brewster was examined. Almost every member of the society was, at one time or other, "the load or the bearer," but particularly the heaviest and the lightest persons of the number were always lifted. As might be expected, the opinions were various in the first experiments. The differences however became less and less as the investigation went on and the proofs were multiplied; and at length they entirely vanished. The final unanimous verdict of the society being, that no such effect was produced as that described in the Letters on Natural Magic, that there was nothing whatever remarkable produced by the mode of lifting: and that the facility which was acquired in the lifting was no more than might be expected from the promptness which the bearers, by practice, acquired in acting uniformly together, upon a given signal. The feat of raising and supporting even their most minute member upon the fore-fingers of six persons, they found quite impracticable.

If this verdict should, by any accident, reach the ear of Sir David Brewster, and he should think any further trial necessary, we are authorized to say, that the "Experimental Society" wish it to be understood that they are ready to undertake it, under any modification that he may be kind enough to suggest; and, after following his instructions with the most scrupulous accuracy, to state the result to the public.

Until some such re-agitation of the question should take place, we think after the above investigation, it must be admitted that the appearances described by Sir David Brewster were illusory, and that no reasonable cause can be assigned which will produce such effects. We think, also, that scientific men should abstain from giving currency to such monstrous improbabilities, unaccompanied by refutation or explanatory remark. It would be far better to continue the inquiry into the cause, either of the fact or of the error, and abstain from publication until some satisfactory information had been obtained.

We beg to suggest the universal formation of such societies. In a wish of assisting in such an important object, we hope to be able to give the constitution, &c., of the one we refer to.

DESCRIPTION OF THE CURRENT-METER,

AS RECENTLY IMPROVED BY MR. SAXTON.

THE necessity of accurately ascertaining the velocity of rivers, &c., in the numerous cases where the rate of the current, the total volume of the passing water, &c., are required to be known, has made it extremely desirable to have a convenient means of measuring and comparing these velocities, that may be applied in every case likely to occur.

Instruments for this purpose have been designed and described by Eytelwein, Wattmann, Fontaine, and others; but it is believed that no one of them has so well satisfied the conditions which have been latterly supposed to be necessary in this kind of instrument, as that which is the subject of this communication. This meter has been in the hands of several eminent hydraulic engineers, both English and Foreign, for the last two years, and has been employed successfully by some of them in investigations of great importance; in that, for example, of estimating the sources from which it has been lately proposed to supply the metropolis with wholesome water, its use was extremely serviceable.

Among the conditions which it seems desirable that an instrument intended for such purposes should possess, the following appear important: Facility of use under all possible circumstances; portability; fewness of parts; strength and simplicity of construction, so as not to be easily deranged or broken, and in case of accident very soon put to rights or repaired. It is indispensable that it have also a means of registering the rate of the current at any point, during the whole of any given period; and this, when from immersion, or other reasons, the actual observation of the instrument is inconvenient or impossible.

These conditions, at least, are satisfied in Mr. Saxton's meter; and by it, the velocity of a current at any part of the surface or bottom of a river, and in all lines between them, can be easily and accurately ascertained; and, of course, when observations have been made in a sufficient number of lines, the mean velocity of the whole river, &c., at the place of observation, may be obtained. This, if multiplied by the sectional area of the river, would give the total volume of water passing during the time of observation. In cases where great accuracy is required, or the rate of flowing is very variable, any number of instruments may be simultaneously employed.

The instrument in question, consists of a revolving vane, a register, a tail, and a staff.

The staff is a rod of about seven-eighths diameter, generally six feet long, graduated into feet, inches, and tenths. On this staff the other parts of the instrument slide, and can be clamped at any point. The tail is a thin plate of metal, which may be from six inches to twenty inches in length, at pleasure; this is acted upon by the passing water, and preserves the axis of the revolving vane parallel to the direction of the current. The revolving vane is simply one or more arms attached to an axis; the surface of these arms is twisted, so that in parting from the axis it makes a continually increasing angle with it.

This angle is such that the action of the current upon an arm may, in all velocities, produces one revolution in some given length, as, for

instance, in a yard, a foot, &c. Where this is accurately obtained, two consequences follow; namely, a particle of water, impinging on whatever part of the curved surface it may, will constantly exert the same effort to turn the axis, and it will pass across the whole breadth of the vane in the same time, whether its transit be near the axis, where the surface is little inclined and narrower, or near the extremity, where it is broader and the inclination is greatest.

The register consists of a wheel graduated upon its side, for the purpose of exhibiting the number of revolutions which may be made by the vane, when it is geared with it. This gearing takes place and terminates instantaneously, at the discretion of the observer. At all other times the register is ungeared and immoveable, whatever may be the rotation of the vane, or the position or motion of the instrument. Fig. 1. Side elevation of the register, vane, and part of the staff and tail.

The axis (E) of the vane (A) is carried by a frame (F), and has upon it an endless screw (G). In this frame hangs the wheel (H) of the register, finely toothed on its edge, and graduated on its face; its axis rests on two carrying-bars (1J), the front one (1) only of which can be seen in this figure. From the lower part of the frame a detent (K) projects, and holds the wheel immoveable so long as it is engaged with it. The carrying-bars are attached to the frame at one end only, and they there move on a pivot (1); their angular motion is limited by a stop (M) at the other end, which moves in a slot, shown by dotted lines, in the frame: this slot just permits the wheel-teeth to enter the endless screw (G), high enough for gearing without jamming. The moveable ends of the carrying-bars are raised by a cord (N) attached to an eye (o), and thence going up into the hand of the observer; when this cord is not acting, the bars remain depressed, the wheel is freed from the endless screw, and forced upon the detent (K) by a spring (P), which lies between the front carrying-bar and the wheel. A joint (a) at the connexion of the register with an arm (R), extending from the staff-socket (s), permits the axis of the vane to be inclined to the staff (c). This is sometimes useful in using the meter on the surface of currents. staff-socket (s) has its lower part cut into a conical screw (T), with slits (u); a milled-nut (v) works on this, and clamps the socket on the staff (c). Another milled-nut (x) on the vane-axis (E) permits it to be taken out to be wiped or cleaned. The tail removes for packing up, by unscrewing the milled-screw (T), which clamps it when in use.

The

Fig. 2. Plan of the vane, register, staff, and part of the tail.-In this figure the rear carrying-bar (1) may be seen.

Fig. 3.

E

Fig. 3. Front elevation of the vane.

In using the meter, the zero end of the staff is generally set upon the bottom of the river, &c., and the vane is clamped upon the staff at the height of the line above the bottom in which the velocity of the current is desired to be known. When the staff rests upon the proposed point, and is held upright and free, the tail immediately swings down the current, and presents the vane full to it. The current sets the vane in action, but as the apparatus is now in the state represented in fig. 1., the vane is unconnected with the register, and the latter is therefore fixed and at rest. When it is certain that the vane has acquired a uniform velocity, and the observers are ready, time may be called, the string is at the same instant pulled up, the wheel of the register is geared with the vane-axis, and the vane and register now both move together. At the expiration of the time, the string is let go: the spring attached to the carrying-bar then draws it downward, and disengaging it from the vane-axis, gears it into the detent. This holds it steady, and though the vane still continues to run, it has no further effect upon the register. The latter may therefore be conveniently and leisurely examined, and by this means the number of revolutions made by the vane, (and, consequently, the velocity and the passing volume of the river,) may be ascertained with great precision.

When it is wished that the register should commence at zero, it may be done by depressing the spring (x), which carries the detent; this detaches it from the register-wheel (H), and leaves it quite free. The zero upon it may then be set to the detent point as an index.

As by neglect after exposure, &c., the amount of the friction in the axis, and of the instrument may be different from that which existed at the time of the graduation, the meter should, previous to experiments where great accuracy is required, be carefully examined, and the effect of the actual friction be ascertained, by drawing the instrument with any convenient velocity through some known length of still water, and observing whether the register give this length exactly or not*. If there be any alteration, the difference will be discovered by this means, and should be allowed for in any immediate subsequent experiments.

To obtain portability, the tail and the vane are made to be easily separated from the staff-socket and register. The whole of the apparatus

• The longitudinal canal in the Gallery of Practical Science has been used for this purpose. A length of sixty feet was marked off, and the meter drawn through the water, care being taken that the latter was always in a quiescent state. This was done many times with very varied velocities, and, whatever these happened to be, the length passed through by the meter was always indicated on the register, with almost mathematical precision.