238 F. II. Bigelow — Studies on the General Circulation 



its practical value is far greater than its apparent numerical 

 size, when rightly interpreted. The departures of tempera- 

 ture and precipitation, monthly and annual, must be interpreted 

 by the true significance of their component parts in a typi- 

 cal oscillation. The 11 -year variation may be very small, but 

 it is made up of numerous wide oscillations of short duration, 

 and the mean departure is a key to their inner meaning. 



Professor NevjcoinVs Criterion of the Mean Departure. 



Before leaving the discussion of the temperature departures 

 and residuals, it will be proper to summarize Professor New- 

 comb's criterion for the computation of a world-departure, that 

 is, a variation common to the entire earth, and due to an 

 external cause as the change in the intensity of solar radiation. 

 One may conceive the earth as hung in space and affected as 

 a whole by the intensity of the radiant heat energy that falls 

 upon it. If this outside energy produces a simultaneous rise 

 in the temperature of the atmosphere at the surface of the 

 earth, it should be possible to compute the amount by combin- 

 ing stations in different parts of the world for the same inter- 

 val of time. If this effect were a simple response to the out- 

 side cause, it would be an easy problem, but since it is complex 

 it becomes an exceedingly difficult question how to discuss the 

 observed departures and residuals. Supposing that the obser- 

 vations at each station to have been made homogeneous, it is 

 evident that they can be combined in one summation only upon 

 the following conditions : (1) The series of fluctuations depend- 

 ing upon solar action must be reduced to the same period, or 

 else by taking the same phase at different intervals of time 

 from the epoch, the summation will destroy the residuals by 

 mere dislocation. (2) Since the same outside cause produces 

 inverted effects in different parts of the atmosphere, an inver- 

 sion of departures and residuals must be admitted. (3) The 

 magnitude of the departures depends upon the place in the 

 general circulation to such an extent that the local effect of 

 oceans and continents must be considered in the method of 

 collecting the data. Since the same outside cause produces 

 opposite effects, in consequence of the general and local circula- 

 tion, unless discrimination is made in the grouping of the data, 

 a non sequitur in the argument follows. If the existing de- 

 partures are summed together indiscriminately, making no 

 allowance for periodic oscillations of irregular lengths, for in- 

 version effects between the tropics and the temperate zones, for 

 the local influence of land and water masses upon the absorp- 

 tion of solar heat, and the radiation of the heat at terrestrial 

 temperatures, the surviving sum of the departures and residuals 



