430 STUDIES IN EVOLUTION 



men of Cleistopora geometrica, illustrated by Edwards and 

 Haime, * represents the maximum size of the cells and their 

 equal development in this species. Although the tubes are 

 not long, the calices are nearly of the same size, and regu- 

 larly hexagonal. 



After the completion of a circle of calices about the parent 

 cell of the corallum enlargement takes place, (1) by buds 

 from the periphery, and (2) by intermural gemmation. The 

 first is not attended by any phenomena differing from the 

 production of the primary circlet of calices about the initial 

 cell. The second takes place under other conditions, and is 

 the chief method of increase in the growth of large corallums 

 having numerous corallites. 



The radial arrangement of the tubes in a large hem- 

 ispherical or cylindrical mass tends to make the axes of the 

 corallites diverge. This divergence can be taken up only by 

 an increase in the diameters of the tubes, or by the addition 

 of new calices between the others. The latter mode is called 

 intermural gemmation. In Favosites and allied genera the 

 maximum size of the corallites is soon reached, and the ex- 

 pansion of the coral is mainly derived from intermural growth. 

 The study of this method of increase properly begins after 

 one or more rows of calices have been developed about the 

 parent cell, and the calices have reached their full dimensions. 



The following description of a symmetrical system of in- 

 termural cell multiplication was observed in a hemispherical 

 specimen of Michelinia convexa D'Orbigny, from the Cornifer- 

 ous Limestone of the Falls of the Ohio. It shows very clearly 

 the stages of development of the interstitial buds, and their 

 modifications. Other corals were examined to the same end, 

 and were found to agree in all essential particulars, whenever 

 their growth was not irregular from their condition of fix- 

 ation, or from the excessive development or death of a number 

 of the corallites. An exact number of peripheral buds is not 



* Monographic des Polypiers Fossiles des Terraines Palaeozoiques, 252, pi. 17, 

 fig. 3, 1851. 



