In any region two zones of flow may be distinguished. The _upper zone of flow_ extends from the ground-water surface downward through the waste mantle and any permeable rocks on which the mantle rests, as far as the first impermeable layer, where the descending movement of the water is stopped. The =deep zones of flow= occupy any pervious rocks which may be found below the impervious layer which lies nearest to the surface. The upper zone is a vast sheet of water saturating the soil and rocks and slowly seeping downward through their pores and interstices along the slopes to the valleys, where in part it discharges in springs and often unites also in a wide underflowing stream which supports and feeds the river (Fig. 24). [Illustration: Fig. 27. Diagram of Well which goes dry in Drought, _a_, and of of Unfailing Well, _b_ Redraw the diagram, showing by dotted line the normal ground-water surface and by broken line the ground-water surface at times of drought] [Illustration: Fig. 28. Diagram of Wet Weather Stream, _a_, and of Permanent Stream, _b_ Redraw the diagram, showing ground-water surface by dotted line] A city in a region of copious rains, built on the narrow flood plain of a river, overlooked by hills, depends for its water supply on driven wells, within the city limits, sunk in the sand a few yards from the edge of the stream. Are these wells fed by water from the river percolating through the sand, or by ground water on its way to the stream and possibly contaminated with the sewage of the town? At what height does underground water stand in the wells of your region? Does it vary with the season? Have you ever known wells to go dry? It may be possible to get data from different wells and to draw a diagram showing the ground-water surface as compared with the surface of the ground.