Observations on the Analysis of Iron in Seawater with Particular Reference to Estuarine Waters

Certain aspects of the analysis of particulate and soluble iron in estuarine waters have been investigated. Determinations of total particulate iron by the method of Lewis and Goldberg showed an average coefficient of variation of about 8%. It is suggested that this degree of variability is attributable to sampling errors and inhomogeneity of iron in the particulate matter. The use of thioglycollic acid to render iron reactive has been examined for the measurement of total soluble iron. The method gave reproducible blank values and an average coefficient of variation of about 9% under normal laboratory conditions. The accuracy of the procedure may be affected by incomplete recovery of iron in some organic compounds and by positive interference from copper, which, however, exceeds 10% only when the weight ratio of copper to iron is greater than 7. On the basis of experiments with59Feas a tracer, it is recommended that filtered samples be acidified to about pH 2 if they must be stored before analysis.