With 2:2′-dipyridyl as reagent for reducible and total iron, determinations were made on samples from the western English Channel at midwinter 1946–47, at the time of the maximum of phosphate, and also in July 1936.

The midwinter results have been treated statistically and the errors of analysis assessed.

Unlike phosphate and nitrate, much of the inorganic iron occurs as particles of fair size, distributed at random in the water. In winter there is no statistically significant difference between the total iron content of samples drawn from 0·5,5,25 and 50 m. depth, the grand mean and standard deviation for the whole area being 0·25 ± o·11 mg.-atom/m.3. The reducible iron for the same depth range and area was 0·19 ± 0·14 mg.-atom/m.3. There appears to be about 0·06mg.-atom/m.3 of iron which is reasonably highly dispersed and not determinable until the water has been oxidized with bromine.

In the bottom layers of water, although the reducible fraction differs little from the overlying layers, the total iron shows a very much greater mean value and variance.

The very low content of total and reducible iron and of particulate ferric phosphate in bottom water at a position 49° 30′ N., 5° 00′ W. has been provisionally attributed to removal by a dense bed of suspension feeders, possibly the mollusc Pinna.

By a statistical examination of the analytical results it has been deduced that the particulate fraction in the upper 50 m. was distributed on an average amongst about 45 particles per litre, each containing about 0·24μg. Fe and with a diameter of about 160μ. This result has been compared with data on the distribution of iron in plankton organisms.

As a statistical average waters richer in phosphate were also richer in total iron, the atomic ratio Fe/P being about 0·57.

In July 1936 the ‘elegans’ water to the westward was much richer in both phosphate and total iron than the more eastern mixed water containing Sagitta setosa. The results agree well with the conclusions drawn from the cruises at midwinter 1946–47.

In 1933–34 concentration of iron was frequently observed in the surface layers; in 1946–47 it was not. In the earlier work the surface film was always cut and included in the sample, in the later it was not. The probable amount of a surface unimicellar film of electropositive unprotected ferric hydroxide has been calculated and agrees well with observation.

Particulate ferric phosphate was found to the extent of about 0·06 mg.- mols/m.3 This appears to increase towards the bottom.

The success of the work at Newlyn in January 1947 owes much to the willing co-operation of Capt. Creese and the crew of R.V. Sabella under very adverse weather conditions. To Mr P. G. Corbin grateful thanks are due for undertaking single-handed the supervision of all the scientific work at sea on stations never more than two hours apart for two or three days on end. The statistical development would not have been possible without the unstinted assistance of Mr G. M. Spooner.