Iron solubilities of Fe(III) hydroxide in coastal and oceanic waters and in the ultraviolet (UV)-irradiated seawaters over a pH range of 5.7-8.2 at 20°C were determined by a sim:Dle filtration (0.025 pm) involving y-activity measurement of 59Fe. At pH ranges of 5.7-7.2 (coastal water) and of 5.7-7.6 (oceanic water), only the Fe(OH),+ species is significant. The calculated solubility products, log*&,, for coastal and oceanic waters were 4.8-5.0 and 4.4-4.6, respectively. The solubilities within the pH range of 7.8-8.2 are relatively independent of pH and aging time. Solubility in the oceanic water was about one order of magnitude lower than that in the coastal water, and UV irradiation reduced solubility to 40.1 r-&I. The vertical profiles of ambient Fe(III) solubility (pH 8.0-8.2) in oceanic waters have the following features in common: solubility in the surface mixed layer is high and variable (0.3-0.6 nM), generally corresponding with the depth of high chlorophyll a concentrations; solubility minima (0.15-0.2 nM) occur at a depth of 50-200 m. These results suggest that natural organic Fe(III) chelators exist in significant concentrations and control the dissolved iron concentration in seawaters.Although iron is an element of great biological and geochemical importance, its oceanic chemistry, such as inorganic speciation and organic complexes, is very complex and not yet fully understood. The dissolved inorganic species of Fe(III) in seawater are predominantly the hydrolysis products Fe(OH),+, Fe(OH),O, and Fe(OH),- (Byrne and Kester 1976;Zafiriou and True 1980;Motekaitis and Martell 1987). However, there are discrepancies between the estimated concentrations of dominant hydroxo-complex species of Fe(III) and values (0. l-l 0 nM) for the thermodynamic solubility of Fe(III) hydroxide in seawater (e.g. Byrne and Kester 1976;Zafiriou and True 1980;Zhuang et al. 1990). These discrepancies are probably due to the existence of natural organic Fe(III) chelators in seawater. Recent reports (Gledhill and van den Berg 1994; Wells et al. 1995;Rue and Bruland 1995) pointed out that iron complexation with organic ligands is possible in oceanic waters, although most researchers have argued that organic iron complexes do not seem to be significant in the open ocean.The oceanic distributions and biogeochemical behavior of dissolved and particulate iron are controlled by com-L Present address: Biology Department, Central Research Institute of Electric Power Industry, 1646 Abiko, Chiba, 270-l 1 Japan. AcknowledgmentsWe thank K. Toya, K. Suzuki, T. Takabayashi, H. Kawakami, and A. Katsumoto for help with the field and technical assistance. We also thank the crews of the Oshoro Maru and the Hokusei Maru of Hokkaido University for help in sampling. We are grateful to Eden Rue and Kenneth Bruland for offering forthcoming data and to anonymous reviewers for comments that helped improve the manuscript.This research was partially supported by a Grant-in-Aid (06640626) for Scientific Research from the Ministry of Education, Scie...

We have performed an in situ test of the iron limitation hypothesis in the subarctic North Pacific Ocean. A single enrichment of dissolved iron caused a large increase in phytoplankton standing stock and decreases in macronutrients and dissolved carbon dioxide. The dominant phytoplankton species shifted after the iron addition from pennate diatoms to a centric diatom, Chaetoceros debilis, that showed a very high growth rate, 2.6 doublings per day. We conclude that the bioavailability of iron regulates the magnitude of the phytoplankton biomass and the key phytoplankton species that determine the biogeochemical sensitivity to iron supply of high-nitrate, low-chlorophyll waters.