Coccospheres and the Origin of CoccolitJis. 307 



equally small, taken in Valencia Harl^our in 1899. It is difficult to 

 estimate with certainity the number of coc eoliths on the coccosphere 

 while it is intact, and one is apt to under-estimate it. But it is pos- 

 sible to make sure of accuracy by crushing coccospheres mounted in 

 glycerine jelly. The coccoliths are then sufficiently separated from 

 one another to enable one to coimt them easily, while the viscidity of 

 the momiting medium does not allow any of them to be lost. It was 

 from coccospheres treated in this manner that the maximum and 

 minimum numbers here stated were derived. 



The variation in the amount of overlap brings it to pass that some- 

 times coccospheres with a small number of coccoliths upon them have 

 an ecjual or even greater diameter than others which carry a larger 

 number of coccoliths. To quote an example : — Two coccospheres were 

 found, each measuring 0-021 mm. in diameter; one of these carried 

 ten coccoliths and the other only seven. 



The coccoliths, although they do not vary so much as the cocco- 

 spheres in their dimensions, yet differ from one another considerably 

 in size. From a large number of measurements of the long axis of 

 the oval, it was found that the extremes deviated from the mean by as 

 much as 20 per cent., the maximum being 0*018 mm. and the minimum 

 0*013 mm., while the mean was 0*015 mm. 



This variation in size of the component coccoliths is in part but 

 not, I think, to a large extent, responsible for the variation in the size 

 of the coccospheres, as it is found that coccoliths of different sizes 

 occur on one and the same coccosphere. It will be seen later that 

 there is some reason to believe that the smaller coccoliths are formed 

 in the early stages of the coccosphere's life history when it is itself 

 small. It may be noted that the smaller coccoliths on a coccosphere 

 are often, if not usually, without the transverse bar across the central 

 perforation. 



Some of the finer details of the structure of the coccoliths are more 

 easily made out when these latter are examined in situ on the cocco- 

 spheres, as then in the one field coccoliths may be examined in almost 

 every position, and views from different directions may be carefully 

 compared. Broken coccospheres, too, often afford valuable evidence as 

 to the relations of the coccoliths to one another and their interlocking. 

 This last point, as well as the shape of the body surrounded by the 

 valves, is greatly elucidated by observations made on microtome sec- 

 tions of material embedded in paraffin. The most instructive prepara- 

 tions were obtained from sections 5/x — 10/x thick; often those of lO/x 

 are more satisfactory than the thinner ones, as, despite the support 

 afforded by the paraffin, the coccoliths in the latter are so often badly 

 shaken and cracked that the fine details of their structure are not so 

 well seen as they are to the thicker sections. 



When a coccolith is examined in actual or optical vertical section, it is 



