Thus photosynthesis has an effect on the of the air that is not entirely counteracted by the respiration of animals, because animals live not only at the surface but at all depths. The amount of photo synthesis must affect the content of the air, but so many factors affect photosynthesis that probably no one factor is the limiting factor for the whole surface of the sea. Photosynthesis varies directly with both CO2 tension and illumination, and either of these might become a limiting factor under certain conditions. There is an optimum pH range for plants, and since the temperature of the sea-surface as a whole is dynamically constant, there might be constructed an optimum COrtension map of the sea. But the range of pH compatible with the life of seaweed is rather broad and may be different for different species and has not been worked out for the whole life-history of a single species, and besides, there may be some other limiting factor. It seems probable that the limiting factors are different in cold and warm seas. Photosynthesis is more than doubled by a rise of 10° in tempera ture, but the number of grams of plant tissue per square meter of sea surface is less in warm seas than in cold seas. The limiting factor in very cold seas may be temperature. The limiting factor at Tortugas may be fixed nitrogen, since I found less than 0.02 mg. of fixed nitrogen per liter of sea-water. There is a more abundant growth of eel-grass on the west side of Loggerhead Key than on the east, and sewage and garbage contaminates the water of the west side. On the other hand, the water of the west side is less agitated by the wind. At the Marque sas, fixed nitrogen is washed into the sea from decaying organic matter on shore. There is an abundant growth of eel-grass in shallow water around Marquesas and in the lagoon where the water is not extremely shallow or perhaps disappears at low tide.
In the shallow pool between Bush Key and Long Key, Tortugas, blocked off from the sea except at extreme high tide by sand-bars thrown up in a recent hurricane, there is a dense growth of seaweed, which probably gets its fixed nitrogen from the adjacent shores. It seems probable that the fixed nitrogen formed by electric discharges, decomposition of organisms, and nitrifying bacteria that finds its way to Tortugas is so reduced by the action of Drew's bacilli as to be the limiting factor for the growth of seaweed. At any rate, the amount of
phytoplankton per unit volume of sea-water is much smaller than in cold seas. Only on the bottom, and especially along shores where the plants may have access to fixed nitrogen (from the decay of organisms) before Drew's bacilli have had time to decompose it, do we find an abundant plant growth.
The limiting factor for animals seems to be food, but under some unusual conditions oxygen might become the limiting factor. Fish in an aquarium open to the air, and with a constant stream of oxygenated sea-water flowing in, suffocate if the stream of water is not sufficiently rapid. The same is true of sponges and other animals. If we take the lowest oxygen concentration found right at the sea surface, 1 kg. of fish would use all of the oxygen in 4,300 liters of sea-water in 24 hours and would show symptoms of altered metabolism before it was used up. Oxygen diffuses into the water from the air very slowly, and the fact that fish can come to the surface does not help them much. It seems improbable that fish would congregate in such numbers as to suffocate. On the other hand, dinoflagellates sometimes multiply or congregate in such numbers as to die and make the water foul. In the summer of 1907 I observed such swarms of Gonyaulax polyedra in the Pacific Ocean off La Jolla, California, that the water was red by day and turned to fire at night. The infected water welled up in spots which grew larger until nearly all of the water was infected. In about 3 days the water stunk and some dead fish were cast on shore by the waves. The death of large numbers of fish from unknown causes are reported by Taylor. It is not intended to imply that these were due to lack of oxygen, but the oxygen content of the water was not investigated and no satisfac tory explanation was found.
No direct determinations showing absence of oxygen in sea-water have come to my notice, but in the water is destroyed by oxygen and the presence of indicates oxygen-lack. In some Scandinavian fjords sometimes rises to within 4 feet of the surface and kills the oysters in "oyster polls." In the Black Sea appears at a depth of 180 meters, according to Palitzsch.