142 



SCIENCE. 



[Vol. XV. No. 369 



large, and the low retail price— threepence per dozen— shows 

 how great the take has been. 



It is clear, from the mere enumeration of the species named 

 here, that there is great variety in the fish fauna of these 

 islands; and, when the testimony of observers in all parts of 

 the colony as to their immense numbers is taken into account, 

 it is certain that from her fisheries New Zealand will yet reap 

 an immense harvest. 



BOOK-REVIEWS. 



A Popular Treatise on the Winds : Comprising the General Mo- 

 tions of the Atmosphere, Monsoons, Cyclones, Tornadoes, 

 Waterspouts, Hail-Storms, etc. By William Ferrel, M. A. , 

 Ph.D. New York, Wiley. 8°. 



Those of us who, about to reach the twoscore prime of middle 

 age, nevertheless feel a little hurt at the respect shown for our 

 advanced years by a younger generation who call us old, may 

 take comfort on realizing that the science of meteorology has 

 been made over again by a man whose labors upon it began only 

 when he had reached our measure of life. William Fen-el was 

 bom in 1817, a farmer's boy in Pennsylvania. He grew up in 

 Virginia, dividing his time between the field and the rough 

 country schoolhouse. A love for mathematics then led him into 

 teaching, and afterwards to our Nautical Almanac OfSce. In 

 1856, at the age of thirty-nine, Maury's facts made him so dis- 

 satisfied with Maury's impossible theory of the winds, that, at 

 the solicitation of a friend, he wrote an outline of what seemed 

 to him a truer conception of the general circulation of the atmos- 

 phere; and with this essay the new school of mathematical 

 meteorology began. A few years ago the appearance of Ferrel' s 

 "Recent Advances in Meteorology" gave occasion to state the 

 outline of his theory, ' in comparison with others generally in 

 vogue. Another volume now allows another reference to this 

 attractive subject. 



This ' 'Popular Treatise on the Winds' ' embodies the substance 

 of a series of forty lectures delivered by Ferrel before a class of 

 army officers of the Signal Service in February and March, 1886. 

 It is now much expanded by deliberate statement of the funda- 

 mental principles of atmospheric rest and motion, and is illus- 

 trated by abundant citation of pertinent observations and records. 

 The book is too serious, too severely argumentative, for general 

 reading ; but it will for a long time have no equal in our lan- 

 guage as a volume to which teacher and student may make safe 

 reference in the search for the solution of difficulties. The plan 

 of the book may be judged by a brief review of its contents. It 

 opens with preliminary chapters on the constitution and nature of 

 the atmosphere, and on the motions of bodies relative to the 

 earth's surface ; the latter being a subject which Ferrel has made 

 his own, and without which no safe step can be taken in the dis- 

 cussion of atmospheric movements. The thii'd chapter discusses 

 the theoretical circulation of an atmosphere lying on a rotating 

 globe, and heated around the equator, deducing therefrom certain 

 critical consequences, and confronting them with the facts as 

 ascertained by observation. He must indeed be wanting in the 

 scientific turn of mind who does not find mental entertainment 

 in the logical order of investigation here traced out, quite apart 

 from its bearing on the special science to which the book is de- 

 voted. Next follow a chapter on the climatic influences of the 

 general circulation of the winds, in the production of wet and 

 dry zones and of wet and dry mountain slopes, and in the de- 

 termination of equable and variable temperatures on the west and 

 east sides of continents, and another chapter on the monsoons, 

 littoral breezes, and mountain and valley winds, by which the 

 general terrestrial circulation is more or less broken up. Thus 

 the first half of the book is occupied. The second half discusses 

 those great travelling whirls known as cyclones, and the more 

 local tornadoes and thunder-storms, on all of which the impress 

 of Ferrel' s methods is most clearly marked. 



Through all this there runs a single theme. Some fact of 

 occurrence calls for explanation. A fit explanation is devised, 

 ■ Science, iv. 1887. 



strictly in accord with a full knowledge of physical law, and its-, 

 consequences aie then deduced as minutely as may be. Theses 

 are matched with the facts, and the validity of the theory is. 

 measured by the degree of correspondence then detected. No one^ 

 can read such a work as this without feeling a distinct intellec- 

 tual gain from the keen vigor of its methods. 



There is one feature in Ferrel' s theory of the atmospheric cir- 

 culation that does not seem to be generally appreciated. We 

 may perhaps best approach it through its misapprehension by 

 certain commentators. Professor Supan, editor of Petermann'» 

 Mitteilungen, whose extended reviews give us the best means ot 

 keeping abreast with the advance of geography in all its branches, 

 referred four years ago to Ferrel 's theory in a notice of Sprung' s. 



' 'Lehrbuch der Meteorologie. ' ' He said in effect that the dis~ 

 tribution of atmospheric pressure was the control, not the result, 

 of atmospheric motion; and that, as there is low pressure at the- 

 poles and high pressure at the ti-opics, the hypothetical return, 

 current from poles to equator cannot exist, for it would have to- 

 move against the barometric gradients.' The same question is 

 asked by M. Leon Teisserenc de Bort, one of the specialists of the. 

 Bureau Centi-al Meteorologique de France. In an essay on the- 

 general circulation of the atmosphere, ^ this author says, ' 'Mr. 

 FeiTel does not explain the causS of the gradient that is directed 

 toward the equator, and that is necessary for the return curi-ent 

 from pole to equator, which he places at a middle altitude in the- 

 atmosphere. This gradient is the more difficult to exjjlain, inas- 

 much as the pressure at sea-level decreases to-wUrds the pole, and. 

 as a similar decrease must exist aloft to determine the flow of thfc 



upper current from the equator. ' ' These criticisms appeaariieBgett- 

 able enough at first sight, but' this is because they fail toi aij^we- 

 hend one of the essential points in Ferrel' s theory. Thfi! cbis& 

 may be stated in brief as follows: — 



Given a uniform distribution of temperature in the' attfflos- 

 phere, its imaginary isobaric surfaces will stand level, ess^a- 

 tially equidistant. Given two adjacent regions, one maintaiosedL 

 at a higher temperature than the other, the isobaric surfaces, <mxk 

 no longer be level or parallel. A convectional interchaj^Stig; 

 motion will be established, as a consequence of which there- wilt 

 come to be a slight excess of pressure in the colder region, Th& 

 isobaric surfaces, not parallel, but diverging from the regiooi ef 

 cold and compressed air to the region of warm and exganjfeji 



1 Petermann's Mitteilungen, Lit. Bericht., 1886. 



• Ann. Bur. Centr. Met., 1885, part. iv. Met. Generale.. 



