DYNAMIC METEOROLOGY. 877 



Lowell Institute in Boston. Tlic author is [nol'e«8or of y;eology and 

 meteorology in Harvard College. The work is written in a very popu- 

 lar and easy style, and deserves to be read and used in elementary 

 schools. 



(17) Abercromhy. — "Weather," by Hod. Kalpli Abercrom by, London, 

 1887. As the name implies, this book deals with the weather, its 

 causes and changes, as distinguished from the statistics ot climate. 

 Neither formula nor numerical tables interrupt the reader, as the ob- 

 ject of the author is to sketch the general principles of the science and 

 give a picture of the methods ordinarily adopted in predicting the 

 weather. 



The present book reads as though it had its origin in the thought, 

 " Farmers and sailors liave always foretold the weather by the sky and 

 cloud; why may not the meteorologist do the same?" 



He therefore devotes one-fourth of the book to clouds aud cloud prog- 

 nostics, and, again, fifty pages to types of weather. The last eighty 

 pages are given to the rules and methods of weather predictions or 

 (as tlie English call them) forecasts. His whole system, as he says (pp. 

 430), " depends neither on theory nor calculation, but solely on obser- 

 vation and experience, aud success depends on natural ai)titu(le and 

 the experience of many years' study." 



To this we add, that these same statements apply to work in astron- 

 omy, chemistry, etc , and that he who makes forecasts merely by means 

 of generalization relative to types of weather and clouds, is sure to 

 suffer from the neglect of a deeper study into causes and laws of opera- 

 tion, but to those who have mastered these mechanical laws the addi- 

 tional information given in Abercromby's book would be of value. 



(18) Eliot. — The report by John Eliot in 187C on the Backergunge 

 cyclone, and his report of 1879 on the Madras cyclone, marked an im- 

 ])ortant step in the progress made, by one of the ablest meteorologists. 

 He tells us he began his inquiry with a strong bias towards the hypoth- 

 esis that cyclones originate between belts of parallel opposing winds, 

 but that tliis proved so unsatisfactory that he rejected it and adopted 

 the condensation tiieory, which accounts satisfactorily for the entire 

 range of atmospheric action that constitute an extensive and intense 

 cyclonic storm in the Bay of Bengal. At the close of his work on the 

 ]Madras cyclom^, and after refuting some views as to cyclone formation 

 that Dr. Hann had advanced, Eliot sums up his own conclusions, from 

 which we make the following extract: 



''The invariable antecedents of powerful cyclones at the two transi- 

 tion periods, April and October, are {a) approximate uniformity of 

 l)ressure over and around the coast of the Bay of Bengal ; [h) light and 

 variable winds or calms over a considerable portion of the bay ; (c) little 

 or no rain-fall over the coast region as w^ell as in the bay; {d) hence 

 the weather is sultry; (e) hence the sea is smooth; (/) the amount of 

 aqueous vapor accumulates and finally gives rise to peculiar sky effects} 



