The winds near the ground are modified both in velocity and direction by friction. The free flow is often interfered with by topography. THE TRUE AIR FLOW One must rise above the ground some distance to get the true air flow, or what is known as the gradient wind, the flow which balances the gradient, i.e. a flow along the isobars. The gradient velocity is found about 300 metres above the ground, and the gradient direction a little higher. The lower clouds as a rule indicate true wind values very well; and so, it is desirable in studying winds to use cloud directions and velocities rather than surface values. In cloud work a nephoscope is essential. The unaided eye, unless properly shielded, suffers from the glare of a sunlit sky; and moreover, there are no fixed points or references. A black mirror, with suitable sighting rods and measuring devices, enables an observer to follow the cloud, estimate its height and determine with accuracy the direction from which it is moving. There is an average difference of 30 degrees between the cloud direction and the surface wind; the upper direction being more to the right. At times the directions may be opposite. [Illustration: FIG. 4. APHELIOTES--THE EAST WIND] It may seem surprising but few of us, except at sunrise and sunset, really see what is going on in cloud land. Some meteorologists hold that the circulation of air 3000 to 5000 metres above the ground controls the path and perhaps the intensity of storms. It is therefore important to know something of the flow at high levels if we would improve the forecasts. LIMITATIONS OF MAP The weather map fails to indicate what shifts of direction and changes in velocity are likely to occur. The forecaster tries to anticipate these, but he bases his conclusions chiefly upon an expected movement of the low area; using the accumulated records of the paths of past storms. But each storm is in reality a law unto itself; and while we know something of the relations between pressure and flow of the air; as yet we know very little about the relations of wind and weather. The problem is complicated by the behavior of the load of water vapor. [Illustration: FIG. 5. EUROS--THE SOUTHEAST WIND] The Chief Forecaster of one of the great national weather services recently wrote: "Despite the fact that maps have now been drawn day by day for over half a century, we may safely say that no two maps have been identical."