(manganese), and 2,138.6 A (zinc), with the 

 appropriate single-element hollow cathode dis- 

 charge tube. Solutions were fed to the burner 

 through a 15-cm. length of 0.381-mm. inside 

 diameter polyethylene tubing. Aspiration rates 

 were measured by timing the flow of 1.0 or 

 5.0 ml. from a 10-ml. graduated cylinder. We 

 also caught and measured the flow from the 

 drain trap, and estimated the atomizer effici- 

 ency as the ratio of sample consumed by the 

 burner to the total sample input. 



Copper, manganese, and zinc standards 

 were prepared by dissolving 100 mg. of reagent 

 metal in 20 ml. cone. HC 1 (manganese and 

 zinc) or 16 ml. cone. HNO3 (copper) and 

 diluting to 1 1 . with deionized distilled water. 

 Working standards containing 1, 2, or 4 p. p.m. 

 of each of the metals were made by diluting 

 the stock standards with 0.25 N HCl. Similar 

 standards containing up to 32 g./lOO ml. of 

 Ca(N03)2.4H20 were also prepared. Metal 

 standards and dissolved shell samples were 

 stored in polyethylene bottles for analysis. 



Locally obtained American oysters, Cras- 

 sostrea virginica , were scrubbed and scraped 

 to remove fouling organisms. The shells were 

 removed, rinsed in distilled water, dried, and 

 weighed. The shells were digested in cone. 

 HNO3, which was subsequently evaporated; the 

 residues were dissolved in 250 ml, 0.25 N HCl. 

 A small amount of sand, which had been 

 entrapped m the shell matrix, remained in- 

 soluble and was removed from some samples 

 by filtration. 



Effect of Calcium Concentration on Aspiration 

 Rate and Atomizer Efficiency 



Concentrated solutions of shell digest were 

 aspirated at conspicuously lower rates than 

 were standard solutions containing no calcium; 

 with each stepwise dilution of the original 

 solution, instrumental response per gram of 

 shell increased. Sensitivity did not increase 

 proportionally with each dilution, however, 

 even though aspiration rate increased linearly 

 with decreasing concentration of calcium. 

 Sensitivity appeared instead to be correlated 

 with atomizer efficiency, which decreased 

 concomitantly with the aspiration rate. Flow 

 rates observed for shell solutions were dupli- 

 cated by corresponding standard solutions of 

 calcium nitrate. 



Effect of Calcium on Instrumental Response to 

 Metal Standards 



Solutions of reagent calcium nitrate con- 

 taining no added metals produced significant 

 instrumental responses for copper, manganese, 

 and zinc even at concentrations lower than 1 

 percent (fig. 7). Response to metal standards 

 was linear with concentration at any specific 

 level of calcium, but the sensitivity of the 

 analysis decreased as the concentration of 



""'^X^ 

 ^*>?^ 



C.CNO,., lO""" ' 



'' c.iNOjH ion*"* 



111 



pEmOOMIUll-l'E"*! 



Figure 7. — Atomic absorption by copper, manganese, and 

 zinc in the presence of high concentrations of dissolved 

 calcium nitrate. 



calcium was increased. This decrease, as 

 well as the nonlinear relation of absorption 

 with increasing calcium concentrations, is 

 consistent with the decreased atomizer effi- 

 ciency at high solute concentrations and the 

 corresponding low aspiration rates. The large 

 response from calcium at the analytical lines 

 for copper, manganese, and zinc necessitates 

 the addition of calcium concentration as a 

 third coordinate on the generalized "standard 



If 



