The instruments called Rain Gauges or Pluviometers are, as their name implies, constructed to measure the amount of rain falling in any given locality, and those in most general use have this principle in common: that the graduated glass always bears a definite relation to the area of the receiving surface. A very extraordinary and hitherto unexplained fact in connection with the fall of rain, and which justifies the opinion that its formation is not limited to the region of visible cloud, is that a series of rain gauges placed at different elevations above the soil are found to collect very different quantities of rain, the amount being _greater_ at the _lower_ level. Thus, twelve months’ observations by Dr. Heberden determined that the amount of rain on the top of Westminster Abbey was only twelve inches, that on a house close by but much lower eighteen inches, and on the ground during the same interval of time twenty-two inches. Accordingly, ten inches is the height at which meteorologists have agreed the edge of the rain gauge should be placed from the ground. The spot chosen should be perfectly level, and at least as far distant from any building or tree as the building or tree is high, and, if the gauge cannot be equally exposed to all points, a south-west aspect is preferable. It is also important that the rain gauge should be well supported, in order to avoid its being blown over by the wind; and, should frost follow a fall of rain, the instrument should be conveyed to a warm room to thaw before measuring the collected contents. The graduated glass furnished with each instrument should stand quite level when measuring the rain, and the reading be taken midway between the two apparent surfaces of the water. The best form of rain gauge is that in use in the Meteorological Office. [Illustration: 48. Howard’s Rain Gauge. Scale about 1/5.] Howard’s Rain Gauge consists of a vertical glass receiver, or bottle, through the neck of which the long terminal tube of a circular funnel, five inches in diameter, is inserted. A metal collar or tube fits over the outside of the neck of the receiver, and aids in keeping the funnel level, while the tube extends to within half an inch of the bottom, thus ensuring the retention of every drop of rain which falls within the area of the funnel. The glass vessel furnished with the instrument is graduated to 100ths of an inch. A modification of this instrument is made with a glass tube at the side graduated to inches, 10ths, and 100ths, showing the amount of rainfall by direct observation, thus dispensing with the use of a supplementary graduated measure.