VALIDITY OF ANALYSES 



37 



i.e. six times by each method. The numbers estimated as being in the sample were as 

 follows, according to the sub-samples taken: 



50 per cent of the sub-samples show a range of variation from the actual number in 

 the sample of from — 16 to + 36 per cent. 



75 per cent of the sub-samples show a range of variation from the actual number in 

 the sample of from — 59 to + 63 per cent. 



90 per cent of the sub-samples show a range of variation from the actual number in 

 the sample of from — 59 to + 96 per cent. 



The total of the sub-samples show a range of variation from the actual number in 

 the sample of from — 59 to + 226 per cent. 



As in the case of Rhincalonus the pipette tends to take more than a fair sample. The 

 mean of the estimated numbers is 246, whereas the actual number present was 184, an 

 error of 33 per cent. The mean variation was 87, or 47 per cent of the actual number 

 present. Again in actual practice, when it was necessary to use the pipette for those 

 forms, particular care was taken to see that a fair proportion was taken up so that the 

 error would not be as great as represented in this test. We must assume, however, that 

 the error might be as high as 70 per cent. In considering the horizontal distribution 

 of the Chaetognatha more attention is paid to the N 100 H results where the error due 

 to the stempel pipette method did not arise. 



Taking into account the errors just outlined together with those greater little-known 

 errors in the field it is clear that the actual numbers estimated in the survey have a value 

 which is no more than the most general approximation to the actual conditions if by 

 " conditions " we mean, as we must, the simultaneous conditions existing at the different 

 stations compared (unless, of course, we are considering changes in conditions due to 

 definite changes in time). But this does not mean that they are of no value. Let us take 

 some examples. Let us compare the estimated numbers of three species of Copepod, 

 Oithona frigida, Drepanopus pectinatus and Calanus propinquns, at the stations on three 

 different lines in the area: say the A line and the C line of the South Georgia survey, 

 and the line between South Georgia and the Falkland Islands, which we will call the 

 SG/FI line. The numbers at each station are expressed as numbers per 50 m. haul for 

 the top 250 m. 



