540 



8Ciujsrcu. 



[Vol. IX., No. 226 



statement, but only carried it before a limited 

 circle of readers, from 1858 to 1860. A briefer 

 and more popular account appeared in the Ameri- 

 can journal of science in 1861, when school-mas- 

 ters might have seen it more generally, had not 

 their attention been distracted by the war. Other 

 brief articles have appeared in the same journal 

 and in Nature. About ten years ago, an extended 

 memoir, entitled 'Meteorological researches,' ap- 

 peared as appendices to the coast-survey reports 

 for 1877 and 1878, where they were said to be ' for 

 the use of the coast pilot.' Like the earlier 

 articles, these researches were too advanced and 

 too little known to reach the school-master di- 

 rectly ; but a review of them in Nature by Archi- 

 bald has brought them before British meteorolo- 

 gists, where they were truly as much needed as 

 with us. Still, it is only in Germany that they 

 have had much effect on recent text-books, and it 

 is to be feared that even the present work may 

 not reach the readers who ought to have it : hence 

 the hope expressed already, that Professor Ferrel 

 may write a more popular book. We may hope, 

 further, that it may find a way into our schools 

 through some regular book-publishers rather than 

 through dealers in second-hand government re- 

 ports. Reflecting on this, how different was the 

 immediate conquest of popular interest by Maury's 

 famous ' Physical geography of the sea ' from the 

 long obscurity of Ferrel's ' Essay on the winds,' 

 and how different the brief life of Maury's theories 

 from the continually increasing vitality of Fer- 

 rel's ! Perhaps, after all, the Nashville journal is 

 a good medium of publication for the young scien- 

 tist. 



Ferrel showed in his first article, that, in con- 

 sequence of the earth's rotation, ail the atmos- 

 phere outside of the tropical belts of high press- 

 ure must have a general motion from west to 

 east, and this disposed of Dove's north-east, ' po- 

 lar ' wind as a meoiber of the planetary circula- 

 tion. He also showed, that, as a consequence of 

 the general eastward motion, the atmosphere 

 would be drawn from the poles and thrown 

 toward the tropics, thus causing the tropical belts 

 of high pressure, and reversing the polar high 

 pressure, that would be caused by the cold of the 

 frigid zone, into polar low pressures. But, in 

 order to explain the oblique recession of the sur- 

 face winds in temperate latitudes from the tropics 

 towards the poles, Ferrel then reversed the whole 

 circulation of the winds at the tropics, placing the 

 return current from the pole at the top, while it is 

 at the bottom in the torrid zone. 



The correction of this inversion appeared un- 

 consciously and independently in a brief paper 

 'On the grand currents of atmospheric circula- 



tion,' by Prof. James Thomson, read before the 

 British association at Dublin in 1857. It occupies 

 but little more than a page in the report of the 

 meeting, and has never been expanded in the more 

 complete form that it fully deserved. Thomson, 

 like Ferrel, saw that the general motion of the 

 atmosphere must be eastward, except in the trade- 

 wind belt, and that the centrifugal force of the 

 great polar whirls thus generated would decrease 

 the pressure at the vortices or poles ; but Thom- 

 son perceived also that the lowest part of the 

 oblique return current, losing velocity by friction 

 with the earth, tends to flow towards the pole, to 

 supply the partial void in the central parts of the 

 vortex. He states explicitly, "that, in temperate 

 latitudes, there are three currents at different 

 heights ; that the uppermost moves towards the 

 pole, and is part of a grand primary circulation 

 between equatorial and polar regions ; that the 

 lowermost moves also towards the pole, but is 

 only a thin stratum forming part of a secondary 

 circulation ; that the middle current moves from 

 the pole, and constitutes the return current for 

 both the preceding ; and that all these three cur- 

 rents have a prevailing motion from west to east 

 in advance of the earth." This was a most sig- 

 nificant addition to Ferrel's first paper, but it 

 lacked quantitative completeness. Ferrel's second 

 paper modifies his first statements and diagrams, 

 introducing the three-current system, and refer- 

 ring to Thomson's paper in a final paragraph, 

 from which we may infer that the correction had 

 occurred independently to our author. Be this as 

 it may, Thomson's suggestion deserves more 

 recognition than it has generally received. A 

 second modification of the plan of general circula- 

 tion appeared in the * Researches ' of 1877, in which 

 the north-east winds of the arctic regions are 

 omitted from the scheme of winds that would ap- 

 pear on a homogeneous earth, and thus by impli- 

 cation referred to a class that may be called con- 

 tinental, as depending directly and indirectly on 

 the diversity of land and water surface on the 

 globe : they are not known to occur around the 

 south pole, where the surface is so largely water. 

 Thus simplified, the scheme appears in the present 

 work, where it demands the closest attention. 



Another great generalization is that which con- 

 nects cyclonic storms with the general circula- 

 tion. Itmay be summarized as follows : a cyclone, 

 or revolving storm, that appears in most sym- 

 metrical development in the tropical regions, has 

 a centi-e of decidedly low pressure, surrounded by 

 a ring of slightly higher pressure than the nor- 

 mal ; outside of the ring, the surface winds move 

 away from the centre and turn to the right of a 

 radius ; inside of the ring, the surface winds cir- 



