in many summer weeks during an ordinary state of things, the steep fall of the hills favouring a rapid flowing off of the water down their precipitous sides. On the high flats of table lands, particularly if covered with an absorbing material, and exposed to the action of sun and wind, much will at first be detained and then evaporated, unless means be taken to drain it off rapidly; again, of the rain falling in such situations, and thus absorbed into the surface soil, a portion will very probably gravitate through it down to cracks, fissures, channels, and chasms, until it meets an impermeable stratum, along which it will percolate until it finds its way out in the shape of springs at the sides or feet of the hills, at a greater or less distance according to the dip of the bed, the more rapid inclination of which favours the escape of the water. The Rain-Gauge.-The instrument, by means of which the depth of rainfall is measured, is well known as the "rain-gauge," and the following is its principle of construction: if any portion of the surface of the ground on which a shower has just fallen were perfectly smooth, level, and impermeable, and also surrounded by some wall or barrier, so that no portion of the rain should escape, it would soon form a shallow pool, the depth of which could be measured with a foot-rule; this then would be the depth of rainfall at that particular spot, and such is the principle of the rain-gauge.

There are various descriptions of rain-gauges, the most simply constructed of which is the "kettle-gauge," which consists of a funnel with a square aperture of given area, from which the rain water flows into another vessel or reservoir beneath, where it is stored until measured, which is done by means of a graduated glass, which will be presently described; the aperture of this kettle-gauge is generally about 6 inches square.

Crosley's rain-gauge is a self-registering instrument; from the mouth of the gauge the water falls into a vibrating bucket; as soon as one side is full, the bucket oversets and another compartment is presented for the reception of the water, which being received is similarly discharged; during a fall of rain the bucket is thus kept in a state of vibration; attached to the instrument is a description of clock-work which carries a hand upon a dial-plate, by means of which the depth of rain is registered in inches, tenths, and hundredths.

In some gauges a float and "staff" are elevated by the water, and show by means of a scale the depth of water in the receiver; it is called a staff-gauge.*

*This gauge is defective, inasmuch as the staff, when much elevated above the funnel mouth, catches more rain than is due.

THE GAUGE GLASS.

237

The rain-gauge known as Dalton's gauge is intended, where used, to ascertain the depth of rain evaporated rather than the depth of rain fallen, and modified, we believe, has been used by Mr. Dickinson, of King's Langley, Herts. It consists of the usual funnel mouth, which conveys the water into a cylinder some three inches in diameter, about three or four feet deep, and supported on an inverted funnel answering the purpose of a stand; the cylinder is filled with two or three feet deep of earth, resting on a perforated bottom, through which the water reaching to that depth passes off into a short horizontal pipe, a little below the perforated plate, and connected with a vertical glass tube, graduated and bearing a definite proportion to the mouth of the rain-gauge. By means of a tap the water may be drawn off when required. It was by a gauge similar to this that the following results were obtained by Mr. Dickinson, at King's Langley, some years since. The entire depth of rain was measured by the usual funnel-gauge.

Although a set of gauges of this description made in a single locality cannot be looked upon as a general guide, and under certain conditions is even liable to exception, still such experiments are very interesting, as tending to throw light to a certain extent upon this important subject.

The proportion of the annual rainfall, or the mean average allowed for evaporation and absorption, is very commonly taken at one-third of the whole depth; about the south of England often more, sometimes exceeding a half; to the north generally one-third is ample, and often in excess. We shall have further remarks to make upon this subject hereafter.

Rain-gauges may be constructed of japanned tin, or better of zinc, and are mostly made circular, so that the rim of the mouth, the defined area of which it is most important to have correct, may be turned in a lathe; this rim should be stout, so as not to be easily deformed, and may be made of tinned copper; it may

be about half-an-inch deep, and should project inwardly over the body of the funnel, so as to prevent the rain in heavy showers from splashing out of the funnel. The area of the rim is generally made of 6, 9, or 12 inches in diameter, as the observer pleases.

The Gauge Glass.-As already observed, this glass is used to measure the quantity of rain fallen; it is made cylindrical, and we have ours of such form as to go into a case like that of a telescope, slung over the shoulder by means of a strap, as being the most safe and convenient mode of carrying it from gauge to gauge, accordingly as these may be disposed over the face of the district under survey; always, however, having at least one glass in store in case of accidents. Let us now suppose, as an example, that the internal diameter of the rim of the rain-gauge is nine inches; the square of any diameter of a circle multiplied by 0.7854 is equal to the area of the circle; in our example 81 x *7854 = 63.6174 square inches; and one cubic inch of distilled water at a temperature of 62° Fahrenheit is a standard of weight according to legal enactment, and has been determined to weigh 252-458 grains, of which 4375 are equal to one ounce avoirdupois. One inch in depth of rain on the gauge will give 63.6174 cubic inches.

Then,

63.6174 x 252.458
437.5

1=3

36·710 ounces avoirdupois.

Therefore any cylindrical glass holding this quantity of water marked on it in depth, and such depth divided into tenths, and these subdivided each into ten other equal parts, will show one inch, or tenths and hundredths of an inch of rain on the area of the gauge of nine inches diameter supposed circular.

Then, as a general rule, let A be the area of the mouth of the gauge in inches, and D any depth in inches also of rain falling on it, and we shall have,

A x D x 252.458

437.5

weight of water in ounces avoirdupois.

It is well to remember this rule, because circumstances may arise when we have to attend to the making of these divisions on the glass ourselves, and as a truly cylindrical glass is not always to be obtained, we may have to pour in small quantities at a time and graduate the glass accordingly, for it is only on the assumption that the glass is truly cylindrical that equal quantities will occupy equal depths; in large towns we have only to give the total quantity of water, and the divisions we require, to

POSITION OF RAIN-GAUGES.

239

a maker of mathematical instruments, or a manufacturer of chemical apparatus, and we get the thing done to our hand.

If we wish to define the diameter of the cylindrical glass, and to ascertain the depth that shall hold one inch of rain over the gauge, the rule will be: multiply the area of the glass by 252-458, and divide by 4375; then the weight in ounces, of one inch depth of rain over the mouth of the funnel, divided by the quotient last found, will give the depth of the glass of the given diameter.

These calculations may be shortened by making use of the constant 0.577, instead of 252-458-437.5.

The above rules will now be as follow:

1. With a given diameter of rain gauge, and any given depth of rain over its area, to find the weight of the rain water: square the diameter in inches, and multiply by 0·7854 to obtain the area A; multiply the area by D, the depth in inches also, and multiply the product by 0·577, which will give the weight required in avoirdupois ounces.

2. With a given diameter of cylindrical glass, to find the depth that shall hold a given weight of water. Multiply the area of the glass in square inches by 0.577, and divide the weight of water in ounces avoirdupois by the product; the quotient thus found will give the depth required in inches. This depth in the glass being then divided into tenths and hundredths as above, will give the means of ascertaining the depth of rainfall on the gauge, and thousandths may be read off by approximation.

The following tablet will in most cases save the trouble of making the necessary calculations.

A heavy rainfall may be measured in two or three times.

Position of Rain Gauges.-We have already observed that they should be placed in the most exposed, that is, the least sheltered places, and at different altitudes, as well as at the different principal spots where circumstances are such as to be likely to

interfere with the rainfall, as on or near the summits, and also the lower parts of hills, on table-land, and in valleys, &c. If the reader will study the table we have given of the rainfall in different parts of Great Britain, and the appended notes, with some of the Ordnance maps before him, he will be able to form a very good idea of how circumstances may interfere as regards quantities of rain in different parts of the same locality. In all cases rain-gauges should, if possible, be placed on the ground, and where not likely to be interfered with; they should be watched to prevent undue interference; every day they should be examined, or the rainfall at least should be registered; particularly after a heavy fall this should be done with as little delay as possible, for if dry windy weather should ensue directly after, it is surprising, under some circumstances, how great the effect of evaporation may be. It is very desirable to note also the direction of the wind every day, and keep a systematic register.

However valuable a series of observations made by the above means may be when obtained with care, it will not fail to be remarked, from a slight study of the table already referred to, that the greatest uncertainty prevails as to the quantity of rain that falls from year to year, and this more particularly in hilly districts; therefore observations made during a series of years are very desirable, and should be obtained when possible.

We have already said enough, however, to show that, setting aside evaporation and absorption, very great difference exists as to the proportion of rainfall over the drainage area which may be discharged by the streams and rivers by which valleys are drained. The following very valuable remarks and tables by Mr. Beardmore, will further tend to elucidate this portion of our subject.

"Districts may greatly vary in their general slope and geological character; grauwacke, granite, and the volcanic districts generally throw water in great rapidity, and are equally liable to great drought in summer time, unless they are capped by moss beds, which act as sponges not always the most pure; some of the newer rocks, on the other hand, such as the old and new red sandstones, have great power of storing water; the latter rocks from their flatness, generally holding it, as indicated by the wells, which are always plentiful in this formation; the former, on the other hand, generally give out the purest spring-water when occurring on mountain slopes, rising above the plains occupied by our numerous coal-fields.

* Hydraulic Tables, by Nathaniel Beardmore.