RECENT ADVANCES IN SCIENCE 365 



change found was most remarkable. Above the tropopause 

 the gradient was found to fall off to practically zero at a height 

 of 20 kilometres. 



The Structure of the Atmosphere when Rain is Falling (Q. J. 

 of Royal Met. Soc., April 1920, vol. xlvi, No. 194). Prof. V. 

 Bjerknes, in the summer of 19 18, initiated the experiment of 

 increasing the number of observing stations in Southern Norway 

 from eight to about ninety, in order to make some attempt at 

 reducing the distance between the points of observation to space- 

 differentials of the streams of air, whose movements and 

 physical changes cause the weather experienced in any place. 

 The experiment has had satisfactory results, and has brought 

 to light some facts that have escaped attention on the less 

 detailed type of synoptic chart in general use. The most 

 important of these is in connection with the bands of rain which 

 frequently sweep across the country in front of travelling depres- 

 sions, and which are generally replaced by showery weather 

 and a more broken cloud sheet. 



A detailed study of the surface wind in these regions makes 

 it clear that there are here two distinct converging currents. 

 Of these, one moves along within the rain-band in the direction 

 of its length, and the other moves almost normally against it. 

 The latter is the warmer of the two, but sometimes the difference 

 of temperature is very slight. We have therefore a cold 

 current lying across the track of a relatively warm current, 

 and the latter, being the most buoyant, steers above the other. 

 The dynamical cooling which takes place in the process results 

 in rain, which falls through the colder current below, and 

 under the pressure of the advancing warm air the colder 

 current yields and the rain-band moves forward. The charts 

 also showed another kind of travelling rain-band, narrower 

 than the other. Here also the Unes of flow show two distinct 

 currents, one of which is along the length of the band, and the 

 other roughly perpendicular to it. But in this case the rain is 

 behind the boundary line, and the temperature is higher in front 

 than behind, so that there is a warm current lying across the 

 path of a cold one, and the heavier cold air edges its way under 

 the warm one. The process is naturally more violent in this 

 case, and there are squalls and heavy showers, followed by 

 colder and finer weather. The two lines of separation may be 

 called the ** steering " and " squall " lines, and are found to 

 meet at the centre of the depression. The essentials of a 

 depression according to these ideas may be summed up in 

 Prof. Bjerknes' own words, as follows : 



" We have before us a struggle between a warm and cold 

 air-current. The warm is victorious to the east of the centre. 

 Here it rises up over the cold air, and approaches in this way a 



