26 



Hon. Ralph Abercromby. Relation between 



are well illustrated by the diagrams given in figs. 11, 12, 13 of a 

 typhoon which raged from September 18th to 20th, 1878, which are 

 taken from Mr. Knipping's paper " The September Taifuns, 1878," 

 in the ' Mittheilungen der Deutschen Gesellschaft fur Natur- und 

 Volkerkunde Ostasiens,' Heft 18, p. 333. 



In figs. 11 and 13 the usual oval form and pressing of the centre to 

 one side are very obvious ; and on the intervening day, fig. 12, we find 

 the typhoon in a transitional irregular form, with all the indications 

 of secondaries. The intensity of the whole is also much less than in 

 our example from Manila. 



The mean velocity of translation was 10 miles an hour, but varied 

 at times from 2"3 to 25 miles per hour. 



The incurvature of the wind is obviously much less than at 

 Manila ; and the squalls were far more pronounced in front than in 

 rear of the typhoon. 



The rain also extended much further in front than in rear ; and 

 the improvement in the weather after the passage of the trough was 

 very rapid. 



The amount of precipitation was so great that Mr. Knipping calcu- 

 lates that no less than 30,000 million tons of water fell on the 

 19th September on the portion of the earth's surface lying between 

 30° and 35° N. lat., and 120°— 130° E. long. 



Mauritius. 



The author does not propose to detail here his researches on hurri- 

 canes in the Mauritius, as they bear more on the great value of 

 Mr. Meldrum's rules for handling ships in cyclones than on the 

 subject of this paper. All that need be said here is that allowing for 

 difference of wind rotation due to the southern hemisphere, the phe- 

 nomena of a Mauritius hurricane are exactly analogous to those in 

 the Bay of Bengal or in the Philippines. 



He finds the same oval shape, with displaced centre, and the same 

 variations in the shape during the progress of any particular hurri- 

 cane. The wind is also very slightly incurved in front, and very 

 markedly so in rear of the cyclone. Cirrus extends all round the 

 storm field; a blue bull's-eye is almost constant; and any " trough 

 phenomena " are very slightly marked. But, just as at Manila, when 

 the nimbus breaks in rear the clouds are harder there than in front ; 

 and M. Bridet at Reunion has noticed that the squalls are rather 

 worse when the barometer turns to rise, i.e., along the trough of the 

 cyclone. 



The propagation of these hurricanes is usually very slow. 



