466 HEPORT— 1873. 



the electric arc, by determining the amount of heat evolved when pieces of 

 carbon, heated between the poles, are thrown into a calorimeter. When a 

 fifty- cell Bunsen's battery is employed, it is found that 1 gramme of carbon 

 evolves as a maximum 850 units of heat when cooled from the temperature 

 it acquires between the poles of the battery. This quantity of heat only 

 corresponds to a mean temperature of 2000° C. in the heated carbon when 

 the great increase in the specific heat of carbon is taken into account. In 

 the experiments made with the battery, no precaution was taken to prevent 

 the cooling of the piece of carbon between the poles from radiation, and 

 consequently the substance never attained a uniform temperature. This fact 

 is easily proved on examining the appearance of the carbon after use, when 

 the substance is only changed into graphite in a few points. That tempera- 

 ture at which carbon changes into graphite may, in experiments of this kind, 

 be used as a fixed point. 



The luminous intensity of the electric arc, according to Fizeau and Foucault, 

 is from 34 to 5G times that of the lime-light when 46 cells are employed, of 

 small or large surface. According to the empirical formula previously given, 

 this would correspond to a temperature of from 7000° C. to 8500° C. 



In the course of the experiments with the battery, several determinations 

 of the total radiation were made by the pyrheliometer. The mean of the 

 observations, which were remarkably constant, corresponds to a radiation of 

 7100 gramme-units per minute, being equivalent to a solution of 4*5 grammes 

 of zinc per minute. A concave parabolic mirror 1 yard in diameter, exposed 

 pei'pendicularly to the sun's rays in this country, concentrates as much radiant 

 energy as a 50-cell Grove's battery of large surface. 



On a Periodicity of Cyclones and Rainfall in connexion with the Sun- 

 spot Periodicity. By Charles Meldrum. 



[A communication ordered by the General Committee to be printed in exfcnso.'] 



At the Brighton Meeting (1872) it was stated that the cyclones of the Indian 

 Ocean, between the Equator and lat. 25° S., were much more freqiient in the 

 maxima than in the minima sun-spot years. 



Since that time the subject has been more fully examined, and I now beg 

 to present a Catalogue of all the cyclones known to have occurred in that 

 part of the world during the last twenty-six years. The Tables given last 

 year contained only cyclones of sufficient violence to dismast or otherwise 

 disable vessels at sea, whereas the accompanying Catalogue gives all the 

 cyclones of force 9 to 12 — that is, "strong gale" to "hurricane." 



The observations for the years 1847 to 1850 are probably not so complete 

 as those for the subsequent years, during which the Meteorological Society of 

 Mauritius made it a special duty to collect storm statistics. Still it is evident 

 that not only the years 1860 and 1872, but also the year 1848, were remark- 

 able both for the number and violence of cyclones, while the years 1856 and 

 1867 were quite the reverse. 



By taking the number of cyclones in each maximum and minimum sun-spot 

 year and in each year on cither side of it, so as to form maxima and minima 

 periods of thi-ec years each, we obtain the results given in the last column 

 of the following Table, showing that during the maxima periods 1848 to 



