ALTERNATIVE SALINE SOLUTIONS TO FLOAT FORAMINIFERAL TESTS D. L. SEMENSATTO, JR.* AND D. DIAS-BRITO LAMBdA (Laborato ´rio de Ana ´ lises Micropaleontolo ´ gicas, Microbio ´ ticas e de Ambientes), UNESP, IGCE, Rio Claro, S.P., Brazil ABSTRACT Flotation has been widely used in studies of recent foraminifera in order to concentrate tests and save time during picking. In this paper, four flotation agents with different densities were compared: (1) trichloroethylene, TCE (C 2 HCl 3 ), with a density of 1.46 g mL 21 ; (2) sodium nitrate/ sodium thiosulfate solution, SNT (NaNO 3 + Na 2 O 3 S 2 ? 5H 2 O), with a density of 1.46 g mL 21 ; (3) zinc chloride solution, ZC (ZnCl 2 ), with a density of 1.70 g mL 21 ; and (4) sodium polytungstate solution, SPT (3Na 2 WO 4 ? 9WO 3 ? H 2 O), with a density of 2.50 g mL 21 . Comparison was carried out by means of qualitative and quantitative data. Results showed that ZC and SPT were the best flotation agents, recovering 91% and 96% of the total tests, respectively, whereas TCE and SNT recovered 59.1% and 72.8%, respectively. Both quantitative and qualitative results significantly improved with a higher density of the flotation liquid. Therefore, substitution of TCE with ZC or SPT solutions is strongly encouraged, because they are, addition- ally, less harmful to health and the environment. ZC is the most cost-effective, since its results were not significantly different from those of the SPT treatment. Carbon tetra- chloride (CCl 4 ) was not considered in this comparative study, because it has been banned in many countries and it is highly harmful to health and the environment. INTRODUCTION Flotation has been widely employed over many years in several types of studies of recent foraminifera to concen- trate tests and save time during picking. However, despite its wide use, there are few studies that evaluate the comparative effectiveness of various heavy liquids used in flotation (i.e., Gibson and Walker, 1967). Moreover, most heavy liquids traditionally used are organic compounds that are neither environmentally friendly nor safe to health. Carbon tetrachloride (CCl 4 ) and trichloroethylene (TCE; C 2 HCl 3 ) are organic compounds commonly used as flotation agents (i.e., Murray and Alve, 2000; Duleba and Debenay, 2003; Scott and others, 2003; Hayward and others, 2004; Duchemin and others, 2005). Nevertheless, both are carcinogenic (Orme and Kegley, 2004), and manipulation must be very careful, requiring the use of a fume hood, safety equipment and great care. In some countries, CCl 4 is currently banned. Moreover, problematic waste CCl 4 remains after flotation and requires proper storage, purification and disposal. When using TCE, an additional washing step is necessary to cleanse the sample before storing them in acrylic flasks, because the chemical attacks this type of material. There is scarce information in the literature about the use of inorganic solutions as flotation agents. Cringoli and others (2004) succinctly described how to prepare a saline solution of sodium nitrate and sodium thiosulfate (SNT; NaNO 3 + Na 2 O 3 S 2 ? 5H 2 O). Another saline solution, zinc chloride (ZC; ZnCl 2 ), was tested by Gibson and Walker (1967) to separate foraminiferal tests, but they considered the results unsatisfactory. However, the test case was carried out with wet samples, which influenced their results by de- creasing the recovery, as mentioned by the authors. In the last decade, sodium polytungstate solution (SPT) has been used as a non-toxic alternative to organic compounds despite the fact that the cost is relatively high (Corner and others, 1996; Buzas and others, 2002; Hayward and others, 2004; Husum and Hald, 2004; Abbene and others, 2006). Bearing in mind the potential health and environmental risks related to the use of organic compounds and the cost of reagents, this paper aimed to compare the effectiveness of four heavy liquids (TCE, SNT, ZCl and SPT) on the recovery of tests. MATERIAL AND METHODS A sediment sample rich in foraminiferal tests, collected in a marginal area of a Brazilian lagoon located at the Sa ˜o Francisco River delta (northeastern Brazil), was first homogenized, washed through two sieves (500- and 62-mm openings), dried, and then four sub-samples of 1 g each were taken from the 62-500-mm size fraction. These sub- samples were floated, respectively, in beakers with: (1) TCE, with a density of 1.46 g mL 21 ; (2) SNT solution with a density of 1.46 g mL 21 ; (3) ZC solution with a density of 1.70 g mL 21 ; and (4) SPT solution, with a density of 2.50 g mL 21 . No replicates were done. The floated and sink fractions were separated by decantation. The flotation liquid was drawn off both the float and sink fractions by vacuum filtration using a filter paper in a Bu ¨ chner funnel. The material was stored in acrylic flasks until picked under a stereomicroscope (Fig. 1). The tests found in both fractions were transferred to foraminiferal slides and identified to the genus level. The efficiency of the flotation procedure was verified by comparing the following quantitative and qualitative variables: percentage of tests recovered in the floated fraction relative to the total number of foraminifers in both fractions, absolute and relative abundance errors (defined in the following section); generic richness, diversity and equitability (Shannon-Weaver and McIntosh’s indices, Magurran, 2004); and microfaunal composition and dominant genera. ABSOLUTE AND RELATIVE ABUNDANCE ERRORS The absolute and relative abundance errors (n error and p error , respectively) express the deviation (%) of the values obtained from the floated fraction in relation to abun- dances obtained from the total sample. They were *Correspondence author: UNESP—Sa ˜o Paulo State University, Instituto de Geocie ˆncias e Cie ˆncias Exatas, Departamento de Geologia Aplicada, Avenida 24A, 1515, Bela Vista, C. P. 178, CEP 13506-900, Rio Claro, S.P., Brazil. E-mail: semensattojr@terra.com.br Journal of Foraminiferal Research, v. 37, no. 3, p. 265–269, July 2007 265