314 



Dr. H. H. Dixon. On tlie Structure of 



lead to a readjustment of the coccoliths already in position. In order 

 to effect this, they must slide past each other, and take up such rela- 

 tive positions with one another and the new coccolith as will change 

 the curvature of the surface and increase the size of the sphere. The 

 dished shape of the coccoliths and their oval outline are apparently 

 adaptations to accommodate this peculiar method of growth. The 

 oval form and the method of intercalating new plates in the skeleton, 

 necessitates the overlap ; but this expenditure of material is com- 

 pensated by the great rigidity so obtained, and the complete protection 

 of the organism. 



It is evident that these observations on the internal origin and 

 development of coccoliths dispose of those theories which regard the 

 coccospheres either as reproductive bodies of the coccoliths or as inde- 

 pendent organisms which aggregate coccoliths on their surface, as some 

 Ehizopoda gather diatom skeletons. The observations also render im- 

 probable the view that the coccoliths are formed as precipitates in 

 dead organic slimes, as has been suggested. The history of the develop- 

 ment shows that the coccosphere is the organism which secretes the 

 coccoliths as its skeleton, and it is probable that these latter only occur 

 free in the water after the death and disintegration of the parent 

 coccosphere. 



Summary. 



The following conclusions may be drawn from the foregoing obser- 

 vations : — 



1. The body of the coccolith extends for a short distance inside the 

 internal valve. 



2. The coccolith is composed of calcium carbonate and a trace of 

 some substance soluble in 1 per cent, of sodium carbonate. 



3. Coccospheres are covered over with an extremely delicate pel^ 

 licle, which is less readily soluble in dilute acids than the coccoliths 

 within it. 



4. The coccoliths on a coccosphere are partially embedded in a slimy 

 proteid material. 



5. Within the slimy layer there is a somewhat stratified internal 

 spherical membrane. 



6. The specimens of coccospheres examined contained no chromato- 

 phore. 



7. In many instances the presence of a minute internal body, pre- 

 sumably a nucleus, was demonstrated. 



8. Coccoliths are secreted internally in close proximity to the 

 nucleus ; the collar uniting the valves is first formed, then the valves 

 are developed, and finally the central body of the coccolith is secreted. 



9. The coccolith, when complete, is probably extruded to the sur- 

 face, and takes up its position among its predecessors. Its valves 



