72 EIGHTH PACIFIC SCIENCE CONGRESS 



availability of fish should be made by the Research section of the Ma- 

 layan Fisheries Department. For this purpose it was decided to con- 

 centrate all observations in Singapore Straits since, besides other consi- 

 derations, one of the main gears used in this area is the fixed fish trap 

 known as the Kelong. The important features of this method of fish- 

 ing are: (1) in Singapore Straits its fishing intensity is practically con- 

 stant; (2) the fish in Singapore Straits are brought on to the long leads 

 of these traps by the tides and all fish lured into these traps are caught 

 by means of fine meshed lift nets; and (3) reliable records of the catch- 

 es of these stakes are easily obtainable. A detailed description of this 

 gear is given by Le Mare and Tham (1947). Observations on the fol- 

 lowing factors were made at regular fortnightly intervals during two 

 full years: (1) sea water temperature, (2) salinity, (3) phosphate con- 

 tent, (4) phytoplankton, and (5) copepod numbers. Full details of the 

 methods used in the measurement of these factors are given in a paper 

 by Tham Ah Kow (1953). It should be stated, however, that in the 

 case of phytoplankton the estimation of pigment content by Harvey's 

 method as well as total counts were carried out. Data on wind-force 

 and rainfall for the period of the investigations were supplied by the 

 Malayan Meteorological Service. 



In an attempt to study the effect of each of these factors on the 

 availability of fish the data so obtained were subjected to correlation 

 analysis by means of partial correlation. As the data are based on 

 fortnightly samples, they constitute time series and it is possible that 

 in each of these time series the value of the variable at one period of 

 time may influence its value in a succeeding period. In such cases the 

 use of partial correlation as an analytical tool would not be valid. How- 

 ever, according to Bartlett (1953) there should be no objection to the 

 use of partial correlation as a preliminary measure. If the partial cor- 

 relation coefficients are quite insignificant, there does not seem much 

 point in considering them any further. If the correlation coefficients 

 appear significant, then the terms of each of the series should be tested 

 for independence and if found to be independent the partial correla- 

 tion coefficients may be accepted as valid. This procedure has been 

 adopted. Where the terms of both series correlated are not indepen- 

 dent then the method of Fisher (1948) for series correlation is applied 

 to remove the effect of time. The results obtained by this correlation 

 analysis are shown in Figure I. 



The main elements of the fish population in Singapore Straits are 

 Stolephorus spp. (anchovy), Clupea spp. (sprat), Chirocentrus spp. (do- 

 rab) and Scomberomorus spp. (Spanish mackerel). They are inter- 



