For the sake of diminishing the evaporation and of measuring small quantities of rain with greater precision, the diameter of the cylinder is sometimes reduced to two inches, and the collected water is, by means of a small pipe, inserted in the bottom of the cylinder, and furnished with a cock made to pass into a glass tube whose interior diameter is half an inch. In this case, the diameter of the upper extremity of the funnel being the same as before, the area of the surface which receives the rain from the atmosphere will be to the area of a horizontal section of the glass tube as 576 to 1. Consequently a shower of rain whose depth on the ground might be one-hundredth part of an inch, would be indicated by 516 inches in the tube.
The funnel of the cylinder may be of tin or copper, and, however the instrument be constructed, it is evident that it should be placed in a vertical position in some place where no object may interfere with the free descent of the rain into the funnel. For the reasons stated in the previous article, the gauge should be placed as near the ground as possible, but the height should always be stated. At Greenwich there are several gauges, at different heights, and these record very different quantities of rain : for example, in 1846, the gauge 50 feet above the ground recorded inches of rain; that at 24 feet, 22'63 inches ; that at 2 feet, 25'86 inches. It is usual to observe the quantity of water in the vessel every morning, if rain has fallen during the pre ceding twenty-four hours ; but, as some evaporation will take place, it would be advantageous to make the observations more frequently.
The sum of all the depths observed during any period of time, as a day, month, or year, will give the whole quantity of rain which has fallen in that time at the place. It is supposed that the rain falls uniformly over the tract of ground lying within the limits of the shower, and consequently that the quantity which passes through the circular area at the upper surface of the cone is equal to that which falls upon an equal area of ground anywhere within those limits.
During windy weather rain may fall obliquely, so that a portion of it may not enter the gauge ; hence, gauges have been formed with openings inclined to each of the four principal quarters, and these are said to register more rain than ordinary gauges. The self-registering rain-gauge is described under ANEMOMETER.
A rain-gauge is a very imperfect inetrumeut, and can never serve further than to give an approximation to the quantity of rain which may have fallen, since some of the water will always adhere to the sides of the vessel, but the following method of ascertaining the allow ance to be made for the quantity thus lost has been recommended :— Let a sponge be made damp, yet so that no water can be squeezed from it, and with this collect all the water which adheres to the funnel and cylinder after as much as possible has been drawn off; then, if the spouge be squeezed, and the water from it he received in a vessel which admits of measuring its quantity, a near estimate may be made of the depth due to it ; and this being added to the depth given by the instrument would probably show very correctly the required depth of rain.
The simplest form of rain-gauge is perhaps that by Professor Fleming. It consists of a cup or receiver, with the opening flush with the surface of the soil. The water is delivered by a funnel into a graduated jar below, which can be from time to time inspected. A.
small receiving vessel is preferred by some observers to a larger one. The area of the aperture must be accurately ascertained. The circular form is the best, since it admits of being turned in a lathe. The rim may be made to project, so as to retain tho water which splashes up, especially in oblique showers. The divisions of the measuring jar should be tested by weighing the water. Some observers prefer to estimate the amount by weight, and an easy method of doing so is given in Drew's useful little volume entitled 'Practical Meteorology,' 1 S60.