Nov. 19, 1885] 



NAl ORE 



51 



tions on the Atlantic— there being observed no rigorous 

 correspondence between the average direction of the 

 movement of storm-centres and the prevailing wind ; 

 but that in some regions the average course of storm- 

 centres is more northerly than that of the wind, and in 

 some regions more southerl)'. 



While in middle latitudes the generally progressive 

 movement of cyclones is in an easterly direction, cyclonic 

 areas are occasionally observed, both in Europe and 

 America, advancing to westward. After a careful in- 

 vestigation of forty-one of the most decided cases whicli 

 have occurred of these westerly movements of cyclones, 

 it is considered that the following conclusions are 

 warranted — viz. that the westerly movement of low- 

 pressure centres is due to a fall of rain or snow, in most 

 cases unusually great, in the region towards which the 

 low centre advances ; and the influence of one low- 

 pressure area acting apparently as an attractive force 

 upon another adjacent low-pressure area ; to the influ- 

 ence exerted by two areas of high pressure, not far apart, 

 by which a new movement is imparted to the air included 

 between them, a new low centre being sometimes de- 

 veloped ; or to the influence of a high pressure on the 

 north-east side of a low-pressure area, when the gradients 

 on the south-west side of the low area are slight, in 

 which case the centre of the low-pressure area may be 

 crowded towards the south-west. 



Rate of Progress of Cyclones. — The rate of progress of 

 the United States storms' for thirteen years has been 

 calculated, and the results arranged according to the 

 months, and expressed in miles per hour. The average 

 rate of progress for the year is 2S'4 miles, rising to the 

 maximum, 34'2 miles in February, and falling to the 

 minimum, 22'6 miles, in August. As regards different 

 years, the variation is also much greater in the winter 

 than in the summer months. Thus, in November, 1S78, 

 the rate was 2r2 miles per hour, but in the same month 

 of the following year it was 407 miles ; and, on the other 

 hand, in July, 1882, the rate was iQ'S miles, but in July, 

 1881, it was 26'6 miles — the difference between the 

 extremes of November being thus i9'5 miles, and in July 

 only 6'8 miles. 



In Europe during the five years ending iSSo the mean 

 annual rate of progress was 167 miles, rising to the 

 maximum, ig'o miles, in October, and falling to the 

 minimum, i4'o miles, in August. Hence the onward 

 movement of storms in the United States is two-thirds 

 greater than in Europe, the rate of excess for the United 

 States over Europe being i '9 in winter, and i'5 in summer. 

 On the mean of the year the average onward movement 

 of storms is, in miles per hour, 28'4 for the United States, 

 i8-o for the middle latitudes of the Atlantic, 167 for 

 Europe, 147 for the West Indies, and 8-5 for the Bay of 

 Bengal and the China Sea. 



Prof Loomis is led to conclude that the general system 

 of atmospheric circulation, consisting of the trades of 

 equatorial regions and the westerly winds of the middle 

 latitudes, is the primary cause which determines both the 

 direction and velocity of the movement of storm centres ; 

 but as respects each individual storm, the determining 

 cause is not so much the average system of atmospheric 

 circulation, as the general movement of the atmosphere 

 going on at the time, and in the vicinity of that particular 



storm. The influence of this general movement is, more- 

 over, materially modified by a variety of causes, among 

 which may be enumerated the rainfall, and the position of 

 the region over ivhich it falls with respect to the centre of 

 the storm ; the size and position of neighbouring areas of 

 high and low pressure, the distribution of temperature, 

 and the physical configuration and character of the 

 surface. 



In further prosecuting this important discussion, the 

 time has perhaps now come for meteorologists to give 

 more consideration and weight to the physical conditions 

 of the cyclone, more particularly to the method of distri- 

 bution of temperature and aqueous vapour within and in 

 the more immediate neighbourhood of the cyclone. This 

 point, which was so strongly dwelt on and urged by Dove, 

 has for some time past been allowed to fall too much into 

 the background. A cyclone is not merely a system of 

 low pressure with winds all around blowing vorticosely 

 inwards upon the centre ; but it is further distinguished 

 by this, that the atmosphere in front of its path is 

 relatively warm and moist, and in the rear cold and dry. 

 These features are seldom kept so steadily in view by 

 meteorologists as they ought to be in the discussion of 

 such questions as Prof Loomis has here brought under 

 review. 



One outstanding difference of the storms of America 

 and those of Europe is that nearly all of the American 

 storms originate on the continent, not far from the Rocky 

 iVIountains, whereas the storms of Europe originate mostly 

 on the ocean. It is not improbable that more than one 

 of the important points of difference between these two 

 classes of storms sliown by Prof Loomis have their ex- 

 planation in the different conditions under which they 

 have their origin. 



OUR BOOK SHELF 



A Treatise on the Calculus of Variations. By L. li. 



Carll, A.M. (London: Macmillan, 1885.) 

 A Text-Book on tlie Method of Least Squares. Bv 



Mansfield Merriman. (London: Macmillan, 1S85.) 

 Both these works by American mathematicians have 

 been, we believe, printed in America, and are now intro- 

 duced to the attention of English students by Messrs. 

 Macmillan. They are first-class representatives of the 

 good work now being done in this field : we have 

 (Nature, vol. xvi. p. 21, vol. x.xvi. p. 59) already given 

 account of other .American mathematical publications. 



Mr. Carll, on his title page, states that his treatise is 

 "arranged with the purpose of introducing, as well as 

 illustrating, its principles to the reader by means of 

 problems, and designed to present in all important parti- 

 culars, a complete view of the present state of the science." 

 The subject is one which certainly has not engaged the 

 time of our book-compilers, for w^hich good and sufficient 

 reasons might be assigned. In iSio, as Todhunter 

 writes, a work was published by Woodhouse, which has 

 obtained a high reputation for accuracy and clearness. 

 That work was not followed by any systematic treatise in 

 English until the year 1850, when Mr. Jellett brought out 

 his valuable "Elementary Treatise on the Calculus of 

 Variations," an octavo volume of 377 pages, with an 

 introduction of 20 pages. In the later editions of Mr. 

 Todhunter's integral calculus are gi\-en such portions of 

 the subject as are generally read by students. The same 

 writer's "History "and "Researches" should be in the 

 hands of all who desire to get up this branch thoroughly. 

 After the lapse of so long an int-^rval as thirty-five years 



