west at al)out the 45th parallel; and section W running south off the 

 Grand Banks at about the 50th meridian. In 1948 sections T and U 

 were occupied once each during the June survey; and section W was 

 occupied twice, once each during the April and Ma}' surveys. Beginning 

 with 1934 there have been 81 occupations of these sections distributed 

 amongst the three sections T, U and W 27, 29 and 25 respectively; and 

 according to season within the months centered on 1 April, 1 May, 1 

 June and 1 July there have been 17, 25, 24 and 15 occupations respec- 

 tivel}'. It has been emphasized that the variations in the Labrador 

 Current from year to year are so great that the seasonal variations will 

 be difficult to derive when it is remembered that the sections seldom have 

 been occupied at regular intervals throughout a complete ice season. 

 The entire ice season represents only al^out a quarter to a third of the 

 year and so no attempt has been made to derive the annual cycle by the 

 use of Fourier series. There is reason to suspect that maxima and 

 minima in volume of flow occur near the beginning and end of the ice 

 season, and minima and maxima mean temperatures occur at similar 

 times of the year. The 27 values for section T and the 29 values for 

 section U w^ere used to develop second degree curvihnear regression 

 equations to define the time and magnitude of the normal minima in 

 volume of flow past sections T and U and the normal maximum mean 

 temperature at section T. The resulting computed times of the occur- 

 rence of the minima in volume of flow at sections T and U were 15 June 

 and 1 June respectively. These dates are surprisingly early. 



The great variations occurring from year to year, the lack of data for 

 complete ice seasons for the years represented, and the brief span of years 

 covered by the data have led to the decision to postpone further attempts 

 to arrive at curvihnear seasonal normal curves until a larger series of 

 observations is available. In the meantime straight line relationships 

 have been used to represent the seasonal normals. On the basis of the 

 81 occupations so far available these lines are defined in the following by 

 giving the value at 15 May, the middle of the 4-month season, and the 

 rate of change per month. Volume of flow has been expressed in units 

 of 1,000,000 cubic meters per second and temperatures in degrees cen- 

 tigrade : 



Referred to these normals the occupation of section W during the 

 1948 April survey (0.32) showed the Labrador Current to be 3.57 million 

 cubic meters per second below normal in volume of flow with a mean 

 temperature of 3.18° C. which was 0.7° warmer than normal. During 



77 



