1. Schlieper's theory of the function of increased oxygen intake by "homoiosmotic" marine invertebrates in dilute sea water in maintaining their body fluids hypertonic to the surrounding water is discussed, and objections are brought forward to the methods used in the experiments on which his conclusions were based.
2. By periodic weighings, and measurements of respiratory rate (under narcotic) by Barcroft manometers, it was found that the weight of N. diversicolor, on transference to water of low salinity, at first increases and then falls, and that the respiratory rate is at first increased and later tends to decrease.
3. With N. cultrifera the weight increases to a higher value and does not sub sequently fall, and the respiratory rate is also increased but to a lesser extent than with N. diversicolor.
4. These differences in the amount of increase in respiratory rate are more marked in water containing only 16.6 per cent, sea water than in water containing 25 per cent, sea water.
5. N. diversicolor maintains its activity while N. cultifera becomes practically inert in dilute water. The latter does not actually die in 25 per cent, sea water after 100 hours, but dies in 16.6 per cent, sea water after about 50 hours.
6. Exposure to M/1000 KCN or to anaerobic conditions in dilute water tends to break down the mechanism by which the free osmotic inflow of water in N. diversicolor is prevented, and the weight curves under these conditions approach the N. cultrifera form.
7. The respiratory rate of G. ulvae increases progressively with dilution of the sea water, and is roughly proportional to the initial difference of osmotic pressure inside and outside the animal.
8. The swelling of Gunda in dilute water is due to swelling of the gut cells, which become much vacuolated. The other tissues appear unaltered.
9. M/1000 KCN or anaerobic conditions cause a greater amount of swelling in Gunda in a given salinity than normally occurs.
10. These experiments seem to give reasonably good support to Schlieper's hypothesis.
11. The mechanism responsible for this "osmotic resistance" in N. diversicolor must be of a somewhat different nature from that in G. ulvae.
12. A rigid distinction between "homoiosmotic" and "poikilosmotic" marine animals cannot be supported.
