ON CHANGES IN THE SEA AND THEIR RELATION TO ORGANISMS.
The sea and air form the circulating media for the living organisms of the world, and although the air circulates so rapidly that no correlation between local faunas and floras and local composition of the air has been found, the local composition of the sea is distinctly affected by living organisms. The local changes in the composition of the sea are the subject of the present paper. These changes are due chiefly to organisms, but partly to meteorological causes. The circulation of the air certainly affects the sea, but the circulation is so rapid that my attempts at correlating meteorological observations (made for this purpose) with changes in the sea have not been encouraging.
The water evaporated is returned with addition of fixed nitrogen from electric discharges or falls on the land and is returned with various salts, chiefly and with fixed nitrogen and other products of organisms. Various seaweeds absorb thus leaving an excess of which has a very low solubility and is constantly being precipi tated in certain warm seas and is precipitated within the bodies of organisms in the surface waters of all seas. In working out the relation of H-ion concentration (pH) to the solubility of in sea-water, I found that all sea-water is supersaturated with and will lose some of it if shaken with calcite or aragonite crystals.
The study of the local changes in the sea is complicated by the pres ence of currents. The surface currents are due to winds, but, owing to the inertia of the water, they do not change as rapidly as the wind and hence are indicators of the prevailing winds (fig. 1), except where they are deflected by bodies of land or rotation of the earth. The largest currents may be constant (fig. 2), but the smaller currents show seasonal variations. Very few constant or seasonal vertical currents have been mapped, but vertical currents must be universal. McEwen has col lected an enormous mass of evidence to show the presence of vertical 216 Papers from the Department of Marine Biology.
currents off the California coast. Vertical currents occur also off the coast of Norway and in the eastern Mediterranean in winter. I have found local and diurnal changes in the and content of the sur face correlated with the presence of plants attached to the bottom (off the Florida Keys), indicating a complete inversion of the water or mix ing of top and bottom water several times a day; the mixing was not rapid enough, however, to equalize the temperature to the depth of 50 meters on a calm sunny day. Palitzsch (1912 b) found the average
02 in surface water of the North Atlantic to be 5.9 c.c. per liter during the day and 5.34 at night.
The opportunities offered permitted of continuous observations at two stations, stations A and B, figure 4 (east and west of Logger head Key, Dry Tortugas), more or less systematized observations around Tortugas (fig. 4), and a series of observations on board between Tortugas and New York City (fig. 3).
The number of grams of chlorine per kilogram of sea-water (abbrevi ated Cl) was determined by titration with silver-nitrate solution, using potassium chromate as indicator and standardizing the whole method with sea-water standardized by the International Commission. The number of cubic centimeters of per liter was determined by the Winkler method. The self-closing water-bottle was not adequate for this purpose, but some observations were made on water drawn up from about 60 meters through glass and rubber tubing attached to the wire of a Lord Kelvin sounding-ma chine, and run continuously through the analysis bottle before reaching the pump. Twice the volume of the tube was run through the sample bottle with the least possible suc tion before the analysis was made. The alkaline reserve was recorded as the number of cubic centimeters of 0.01 N HC1 used in titrating 100 c.c. of sea-water while boiling in a 500 c.c. Erlenmeyer "nonsol" flask, using di-brom-o-cresol-sulfon phthalein as indicator, until all trace of purple color had disappeared. The end-point is not sharp, but is sharper with this indicator than with any other yet found. The water must not dry on sides of flask. The pH was determined colorimetrically (McClendon, 1917), correction being made for Cl. The total per liter was determined from the alkaline reserve and pH by means of the conversion table.