448 



SCIENCE 



[N. S. Vol. XLV. No. 1167 



and has revealed the fact that a few kilo- 

 meters beyond the limit of exploration 

 when he made his prediction, or at an 

 average altitude of eleven kilometers, the 

 aero-thermic gradient is interrupted and 

 succeeded by practically isothermal condi- 

 tions above. 



To-day in reputable treatises one may 

 read in round numbers the supposed tem- 

 perature at the center of the earth, and 

 based upon what? The geothermic gradi- 

 ent determined for the shell of rock imme- 

 diately beneath the earth's surface and veri- 

 fied roughly for about one four-thousandths 

 of the entire distance to the earth's center. 

 Can we assume that our yard-stick in this 

 instance is suitable for measurement 

 throughout the entire radial distance, and 

 is there no possibility of abrupt interrup- 

 tions such as occur in the temperature 

 gradient of the atmosphere ? 



William Ferrel, much the most distin- 

 guished meteorologist that America has 

 produced, and the one to whom we owe the 

 basic principle upon which modern meteor- 

 ology is founded, predicted as a corollary 

 to his theory of the winds the existence of 

 whirls about the earth's geographic poles 

 surrounding areas of calm and low atmos- 

 pheric pressure. As these polar calms and 

 whirls are an important feature of the 

 present theory of atmospheric circulation, 

 it will be profitable to examine briefly their 

 evolution as a study in the psychology of 

 theory. In the preface to his general trea- 

 tise upon the winds, Ferrel tells us how his 

 attention was first directed to this subject 

 through reading Maury's "Physical Geog- 

 raphy of the Sea," the first edition of which 

 appeared in 1855, while the first essay of 

 Ferrel was published in 1856. From 

 Maury Ferrel learned, as he has told us, 

 "that the pressure of the atmosphere is 

 less both at the poles and at the equator of 

 the earth than it is over two belts extending 



around the globe about the parallels of 30° 

 north and south of the equator." On ma- 

 king reference to Maury, we find that upon 

 the basis of recorded observations between 

 the parallels of 40° and 54° south latitude 

 the average barometer reading varies from 

 29.9 to 29.4 in passing from the lower to 

 the higher latitude. "With the gradient 

 obtained from this limited range, Maury 

 has extended the curve as a straight line to 

 the geographic pole through a range of 

 thirty-four degrees of latitude or more than 

 twice the observed distance, and obtained 

 a theoretical reading for the pole of twenty- 

 eight inches of mercury. A similar method 

 applied to the northern polar region has 

 supplied a less marked gradient and a theo- 

 retic value for the barometer reading at the 

 northern geographic pole of 29.65 inches 

 of mercury. 



Since this theory was promulgated, ex- 

 ploration has been extended to both poles 

 of the earth and has shown that but a short 

 distance beyond the latitudes which limited 

 the data employed by Ferrel, the steadily 

 lowering pressure gives place to a rising 

 barometer in the direction of the poles. 

 Studies of the free atmosphere by means of 

 balloons in the same high latitudes also 

 indicate pretty clearly that no such whirls 

 as Ferrel assumed can exist. Yet so great 

 has been the success of Ferrel's theory as 

 a whole that despite their contradiction by 

 the facts, the polar calms and whirls are 

 stiU treated in the latest text-books of 

 meteorology. 



The polar whirls of Ferrel are by no 

 means a unique example of a large con- 

 ception in science receiving general sup- 

 port because however carelessly constructed 

 it was an attachment or rider to a still 

 larger theory. The triumph of the larger 

 idea or the prestige of the author due to 

 some other achievement, has by its inertia 



