VENRICK: SMALL-SCALE DISTRIBUTIONS OF OCEANIC DIATOMS 



Table 2. — Results of sampling study at Central Pacific Station 5. 



Species 



al 



(T" 



^Vo-= 



3.08 

 1.28 



0.90 

 0.51 

 0.51 

 0.38 

 0.38 



1.54 



0.90 

 0.77 

 0.51 

 0.51 

 0.51 



0.51 

 0.13 

 0.13 



1.15 

 1.03 

 0.51 

 0.26 

 0.26 

 0.26 



0.13 

 0.13 



3.45 

 l.OC 

 3.42 

 0.45 

 0.45 

 0.21 

 0.41 



4.56 

 0.42 

 0.45 

 0.27 

 1.45 

 0.45 



0.45 

 0.09 

 0.09 



1.70 

 1.17 

 0.27 

 0.16 

 0.16 

 0.16 



0.09 

 0.09 



1 Statistics based on numbers of chains per aliquot. 

 r Nonrandomness indicated only by runs test. 



For the more abundant species, the 95% con- 

 fidence limits which can be placed around a 

 single sample are extremely broad. However, 

 without replicate samples, this interval cannot 

 be significantly reduced. For species represent- 

 ed in a sample by fewer than five cells, the con- 

 fidence interval given by the empirically derived 

 expression is smaller than that obtained from 

 the assumption of a Poisson distribution. For 

 these rarer species, it is recommended that the 



confidence interval around a single sample be 

 derived from the assumption of a Poisson dis- 

 tribution (Fisher and Yates, 1957). 



ESTIMATES OF DIVERSITY 



The variability of individual species in the 

 field determines the precision with which a 



367 



