206 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1917. 



The ratio of the weight of the salts in the sea water to any given 

 weight of water is expressed as parts per thousand and is called the 

 salinity of the water, for instance, a salinity of 36 means that there 

 are 36 pounds of salt in 1,000 pounds of water. 



The factors affecting the life of corals, so far considered, are all 

 more or less correlated ; for instance, at gi^eat depths in the ocean the 

 temperature is low, there is no light, and surface agitation of the sea 

 is not felt. The relative salinity of the ocean differs from these fac- 

 tors in that it is not definitely related to any one of them, except that 

 in the Tropics the surface concentration of sea salts is somewhat 

 greater than that at depths considerably below the surface, and that 

 there the concentration is somewhat greater than that on the surface 

 in higher latitudes. In other words, there is in the surface waters of 

 the Tropics some concentration of salts due to evaporation, but the 

 difference in the salinity of the different parts of the ocean, away 

 from the mouths of great rivers, is not sufficient to affect the life of 

 corals. Notwithstanding these facts, it is important to know the 

 maximum and minimum salinities that corals can endure. 



The average salinity of the Tortugas water according to Dole is 

 36.01. Of the 17 species of Tortugas corals kept in a tank of water 

 with a salinity of 18.28 for 24 hours, all were damaged or killed except 

 Maeandra areolata, Siderastrea radians, and Pontes astreoides ; but 

 no specimen of 16 species showed any evidence of harm after remain- 

 ing 48 hours in water of a salinity of 27.87. Apparently corals would 

 not be hurt if the salinity of the ocean were reduced to about 80 per 

 cent of its present salinity. Mayer obtained similar results in his 

 work on the corals of Murray Island, Australia. 



Although I did not experiment with concentrated sea water, the 

 studies made by Goldfarb and others on the effect of concentrated and 

 diluted sea water on regeneration in hydroids and in the jelly-fish 

 Casslopea are here pertinent. The combined results of the experi- 

 ments are in accord with the deductions made by oceanographers and 

 geologists from other data, viz, the ocean is becoming more salt, and it 

 appears that marine organisms are now living in an environment 

 which is considerably below the optimum condition for their exis- 

 tence. 



HOW CORALS CATCH THEIR FOOD AND WHAT THEY EAT. 



I made no more interesting experiments on corals than those to 

 discover how they catch their food and what they eat. Although 

 nearly all the species abundant in the Tortugas were used in making 

 the experiments, one species, Maeandra areolata, was studied more 

 than any other. It was fascinating to bring a colony with the ani- 

 mals composing it entirely retracted, as in plate 22, and induce it to 



