488 Dr. H. Arctowski on Normal Anomalies oj 



The annual variation for Melbourne may therefore be 

 represented by the following diagram (fig. 1) . 



Fig. 1. 



-<N 



* N N 



a 



/ 

 t 









This diagram expresses graphically the opinion that the 

 seasonal change of temperature in Melbourne is not what it 

 should be ; that at given dates of the year (X, Y, Y', X') 

 the natural course of temperature, due to local conditions, is 

 interrupted, just as if the station had been shifted into another 

 climate, similar, but colder than it ought to be during the 

 summer, and warmer than it ought to be during the winter. 



The annual variation proceeds by steps. The steps of 

 ascent, occurring while the temperature is decreasing, corre- 

 spond to similar steps of descent observed during the other 

 half of the year. 



Each successive stage marks the passage from one phase 

 to another. 



In the preceding example the annual variation is composed 

 of three phases. 



This opinion seems never to have been advanced, and it is 

 astonishing to notice that even the fact of the existence of 

 steps during the autumn, corresponding to the steps of the 

 spring, has, so far as I know, attracted practically no 

 attention ; though the frequently recurring temperature 

 anomalies for given dates, in May and June, have been 

 very extensively studied. 



Researches of particular interest concerning these anom- 

 alies are those of Gr. Hellmann*, W. v. Bezoldf , Ch. Dufour J, 

 W. Martin §, U. Gautier||, and K. Almstedt^. 



* Ann. d. Phys. u. Chem. vol. clix. p. 36 (1876). 



t Abh. math.-phys. XL X. Bay. Akad. Wiss. vol. xiv. II. p. 69 (1883). 

 % Bull. Soc. Vaud. Sc. ser. 3, vol.xxix. p. 316 (1893). 

 § Abh. X. Preuss. Met. Inst. vol. ii. No. 3 (1902). 

 || Arch. Sc. Phys. et Nat. ser. 4, vol. xxxi. p. 497 (1911). 

 «fj Meteor. Zeit. vol. xxxi. p. 426 (1914). 



