210 ANNUAL REPORT SMITHSONIAN INSTITUTION, 1917. 



normally pure, that is, not contaminated by refuse or other abnormal 

 impuritites. 



It is relatively easy to get large numbers of planulae to attach 

 themselves to disks by using the culture methods above described. 



DISTRIBUTION OF CORALS BY MARINE CURRENTS. 



It has already been stated that because of its bearing on the pos- 

 sibility of the distribution of coral species by oceanic currents, it is 

 highly important to know the duration of the free-swimming larval 

 stage. Observations were made on four species. The range was 

 from 2 to 23 days. Should an ocean current have *a velocity of 3 

 knots per hour, in 23 days planulae might be carried 1,65G knots; at 

 2 knots per hour, 1,104 knots; at 1 knot per hour, 552 knots. It is 

 known that every species of shoal water coral in the Bermudas is 

 found in Florida and the West Indies; while not only is the Hawai- 

 ian fauna Indo-Pacific in its affinities, but several of the species (at 

 least four) also occur on the east coast of Africa or in the Red Sea, 

 and I seriously doubt any part of the Hawaiian fauna being pe- 

 culiar to those islands. The clue to the cause of the wide distribu- 

 tion of living coral species is given by the possibly long duration of 

 the free-swimming larval stage. It should be mentioned here that 

 numerous instances of the transport of coral colonies attached to 

 floating pumice or to driftwood are on record, but it seems to me that 

 the transportation of larval corals is more important in the dis- 

 tribution of corals by ocean currents than the transportation of 

 attached colonies. 



RATE OF GROWTH OF CORALS. 



The growth-rate of corals was studied on colonies developed from 

 planulae that were reared in the laboratory according to methods 

 already described and then planted in the sea (pi. 21), and on 

 colonies from planulae naturally attached but known to have settled 

 in a certain season; on colonies fastened with Portland cement to 

 terra cotta or reinforced concrete disks and then fixed on the heads 

 of iron stakes driven into the sea bottom (pis. 22-25) ; and on 

 colonies naturally living in the ocean. 



Two methods were used for rearing to subsequent stages the larvae 

 that settled in the laboratory culture. One was to fasten the disks 

 bearing the young polyps to the bottom of a floating live car; the 

 other was to j)lant the disks directly on stakes. Both methods suc- 

 ceeded. Plate 20, figures A and B, illustrate the method of plant- 

 ing in a floating live car; Plate 19, figure B, the apparatus for 

 planting on iron stakes. A long iron bar, with a cap on the lower end 



