DAILY VARIATION— d" APE AD ARE. 



73 



Daily Variation of Temperature at Cape Adare. — Observations of temperature were taken at 

 Cape Adare every two hours in June and July ; during otber months observations were not 

 taken during the night hours. As it is impossible to correct the latter for non-periodic change 

 they will not be dLscussed here, but the data will be found in the book of tables. 



Table 31. 



No excess of temperatures at 4 a.m. is shown in either of these months. The day hours 

 were warmer than the night hours in June, but in July the period 16 hours to 24 hours 

 was above the average. 



Non-periodic Changes of Temperature. 



In the previous section the object was to determine the true daily variation of tempera- 

 ture unaffected by irregular changes. With this end in view, data from as large a number 

 of days as possible, having a similar periodic variation, were combined in order that irregular 

 changes which are independent of the time of day would cancel one another, the residual 

 variation was assumed to be mainly due to true periodic causes. The irregular variations 

 which were removed in this process are, however, of considerable meteorological importance 

 and \vill be studied in this section. We have now therefore to employ a process which will 

 eliminate the true daily variation and leave the non-periodic changes. 



Many methods have been used by other writers for this purpose, but most of them lead 

 only to statistical results. As the essential purpose of the investigation is to find the physical 

 laws underiying the meteorological observations, I have chosen two methods which appear to 

 me to be the most suitable for the available data. The first method which depends on daily 

 maximum and minimum temperatures will be used in discussing the non-periodic temperature 

 variations in McMurdo Sound, because maximum and minimum temperatures are available for a 

 longer period than temperatures recorded in any other way. On the other hand, no maximum 

 and minimum temperatures were recorded at Cape Adare and Framheim, hence to compare the 

 irregular temperature changes at these stations with those at McMurdo Sound another method 

 will be adopted. 



Maximum ami Minimum Temperatures in Relation to N on- periodic Temperature Changes.— U 

 there were no irregular temperature changes the maximum and minimum temperature of the 

 day would be determined by the daily periodic temperature variatioiis. In this case the daily 

 temperature range would be the same as the daily temperature amphtude.* This is practically 

 the case in tropical countries. When, however, there are irregular temperature changes, these 

 increase the difference between the range and the amplitude. In the Antarctic where irregular 

 changes are very large compared with the regular changes, the mean range is considerably 

 greater than the amplitude, and the difference, range - amplitude, affords a qualitative measure 

 of the non-periodic changes. The difference between the range and the amplitude which we 

 shall now study will for convenience be called the ' reduced range.' 



* See the definition of these terms, page 53. 



10 



