The process of formation of a cyclone according to this theory can be most readily explained by reference to the diagrams of figure 6, where in a the broken line represents a portion of the undis turbed polar front. The actual surface separating the warm and cold air is inclined to the horizontal at an angle of the order of with the warm air above the cold. The first step is a bulge of the warm air into the cold air, as shown at b. This stage is marked by a fall of pressure at the tip of the tongue of warm air, and the bulge and the newly formed cyclone both travel with the warm current. This stage of the development is represented in greater detail in figure 7, central portion, where the arrows represent the air currents, and the main rain areas are shaded. The association of rain with the ascent of air is clearly indicated in this diagram, there being an extensive area of precipitation in advance of the warm front. Heavy continuous rain does not occur in the warm sector. The lower diagram in figure 7 repre sents a vertical section from west to east across the warm sector.
At the cold front the cold air pushes under the warm air, and this process is continued until the whole of the warm sector is lifted above the ground, as shown in figure 6c, and in the upper diagram in figure 7. The depression is then said to be occluded, and it subsequently diminishes steadily in intensity, and its motion dies away. Details of the association of weather with the polar front will be found in papers by J. Bjerknes and other Norwegian meteorologists in the Geofysiske Publikationer, Vols. i, ii, and iii, in papers by various English writers in the Q.J.R. Met. Soc. from 1923 onwards, and in a paper by J. Bjerknes, Geophysical Mem oirs, No. 5o.
A number of features of cyclones of middle latitudes can be more clearly realised by the study of fronts than has hitherto been possible. The Bjerknes cyclone starts its existence with a warm core, and ends with a cold core. Its cycle of growth is from asymmetry to symmetry, and only in the dying stage is the cyclone to be regarded as approximating to rotating fluid. The scheme of distribution of cloud and rain given by Bjerknes is in close agree ment with the scheme of Abercromby which was previously ac cepted as the standard description of a cyclone, if we restrict our attention to the occluded cyclone, figure 6c. It is moreover un deniable that many of the cyclones which reach the British Isles from the Atlantic are occluded.
A very complete summary of the views of the Bergen (Nor wegian) school, together with a detailed application of the Nor wegian methods to a particular case of a polar front by Bergeron and Swoboda, in the Proceedings of the Geophysical Institute of Leipzig (vol. iii.) should be consulted.
A point of importance which is seldom mentioned in writings on these topics is that the cold front cannot be a true wedge with its point at the ground. The effect of friction at the ground, causing slowing down of the motion of air near the ground rela tive to the air at say soo metres above the ground, is to cause the cold air to have a distinct nose raised above the ground. This has been confirmed by observations of temperature taken at dif ferent heights above the ground. Bjerknes has suggested that the polar front extends over very considerable distances, and may have a family of cyclones strung along its length. Each successive cyclone of a family passes further southward than its predecessor and finally the family of 4 or 5 cyclones passes away, and a new family starts at a fresh polar front.
The practical application of the Norwegian ideas to the analysis of synoptic charts has led to very definite advances in forecasting weather, and though the theory which presents the cyclone as a wave in the polar front is very incomplete, the practical ideas developed by J. Bjerknes have helped to focus attention upon the physical consequences of the interaction of currents of air of different origin. It is beyond question that many cyclones do form at boundaries between cold and warm currents, and that they tend to follow the motion of the warm currents. Also the phenomena associated with the "occlusion" of cyclones are readily found on the weather map.
The Austrian View of the Origin of Cyclones.—Austrian writers, notably Exner (Dynamische Meteorologie, ate Auflage, p. 337 et seq.), have regarded the development of cyclones at the surface of separation of warm and cold air as following a dif ferent course. In fig. 8 the cold air is shown shaded, the warm air unshaded. The first step is a deflection of the cold current southwards by one of the land masses of Greenland, Spitz bergen, Franz Josef Land, or Novaya Zembla, giving an out burst of cold air into the warm air, as shown in fig. 9 (a). The further development is shown in fig. 9 (b). The cold tongue C cuts off the direct supply of warm air into the region A, leading to the formation of a centre of low pressure there, with stream lines as shown in the diagram. The cold tongue is drawn along in the general direction of the warm air. Behind the cold tongue air is dammed up and forms an anticyclone.