702 SCIENCE PROGRESS 



out at about three-halfpence each. The author is to be congratulated on the fact 

 that he has in every case restricted his maps and diagrams within the limits of the 

 format of the book, thus avoiding the inconvenience of folded sheets. 



Though showing occasional traces of haste, the text is as a rule clear and 

 readable, besides containing much that is suggestive and original. Perhaps the 

 least satisfactory portion of the book is that which deals with the regional distribution 

 of pressure in high latitudes, the explanations that are offered being hardly calculated 

 to carry conviction to the mind of the student. It is true that much remains to be 

 done in working out the details of the subject ; but the main principles admit of 

 very simple expression. Briefly, the variations in pressure on the earth's surface 

 are dependent on differences of temperature, moisture, and centrifugal force — 

 to use a useful term by no means free from objection but sufficiently well 

 understood. The tendency of heat alone would be to make the pressure increase 

 continuously from the equator to the poles. That of centrifugal force alone has 

 exactly the contrary effect, for air moving towards the poles has a greater rotational 

 velocity than that proper to its new latitude, and therefore tends to rise, while air 

 moving towards the equator tends to sink. The magnitude of this effect is 

 dependent on the rate of change in the length of the radius of rotation, a rate 

 which increases rapidly as the pole is approached. Thus it comes about that 

 while the influence of heat difference is greater in low latitudes that of difference 

 of centrifugal force is more powerful in high latitudes. If then the earth were 

 a uniform solid spheroid, there would be a belt of low pressure round the equator 

 and two areas of low pressure at the poles with high-pressure belts between. In 

 the colder regions, however, the air over the open sea is as a rule warmer as well 

 as moister than that over ice or land masses. Tracts of open sea in high latitudes 

 are therefore characterised by abnormally low pressure. Accordingly we find 

 in the North Atlantic and Pacific two areas of low pressure imposed upon the 

 regular decrease of pressure from the horse latitudes to the North Pole itself. 

 Similarly in the Southern Hemisphere the open circum-antarctic sea gives rise 

 to a circle of abnormally low pressure. Beyond, in the region of Antarctic land 

 and ice, the pressure is probably somewhat higher, representing the normal low 

 pressure of the pole. 



There are some minor points on which a word of explanation might have 

 been added for the benefit of English readers. For instance, on page 509 we find 

 the word " billion " used in the popular American sense of a thousand million 

 instead of the square of a million, according to the usage of mathematicians, 

 though, curiously enough, in a quotation from Sir John Herschell, on the very next 

 page, a trillion is employed correctly, without comment, for the cube of a million. 

 On page 613 we are told that the " English " unit of pressure gradient is a hundredth 

 of an inch in seventeen miles, which appears a very unreasonable convention in the 

 absence of the explanation that it is equivalent to one millimetre in a degree. 



These however are minor matters. The book as a whole is excellent, and the 

 best testimony to its value is the keen interest it has awakened in every student 

 in whose hands it has been placed. 



John W. Evans. 



