Thirty sediment samples were obtained from the diatom-rich unit. The disaggregated sediment was simply suspended in water on a slide and observed. This was sufficient to identify differences in radiolarian occurrences and to allow determination of the ratio of diatoms to radiolarians and the percentage of radiolarian families (Fig. F3). It must be stressed that the aim of the study was not to quantify fluxes of radiolarian and diatom tests to characterize paleoproductivity in this time interval but only to determine radiolarian maxima, typical radiolarian assemblages, and specific occurrences of taxon groups or growth forms above and below the diatom-rich unit (cf. Cortese and Bjørklund, 1999). Thus it was not necessary in this case to use sophisticated methods of radiolarian preparation (e.g., Locker, 1996) in order to obtain randomly distributed individuals. The quantitative data set was based on a 300-point grid for the diatom:radiolarian ratio and on the counting of 300 radiolarian individuals in determining the percentages of radiolarian families.