520 On the Madreporarian Getitis Turbinavia. 



certain cliaracters of the caliclcs were recognized as of more 

 value than mere superficial resemblance, that I found myself 

 compelled to separate the groups into different species. The 

 remarkable resemblances are due entirely to similarity in the 

 general aspect of the coenenchyma. The Tongatabu speci- 

 mens, for instance, have a velvety appearance. The Formosan 

 specimens have a rough look, like a gritty sponge. The 

 Shark's Bay specimens have a solid stony look, the gyrating 

 furrows being separated by broad granulated ridges. Other 

 examples of the same phenomenon might be mentioned. Its 

 cause is no doubt to be souglit in the varying physical con- 

 ditions of their several environments. AH corals must in 

 some way be influenced by the varying climatic and other 

 conditions under which they develop. The effects of these 

 are perhaps specially visible in Turbinaria, owing to the 

 great abundance of the coenenchyma in this genus. 



EXPLANATION OF THE PLATES. 

 Plate XIX. 



Fit/. 1. Section through tlie stalk of a very young cup-sliaped corallum, 

 showing the cavity of the parent polyp, the walls greatly 

 thickened by coenenchyma. 



Fiff. '2. A young calicle at the edge of a corallum, growing in the direction 

 of the arrow. The coenenchyma is built up by true costae con- 

 nected by concentric trabeculae ; distally the primitive costae are 

 distinct, proximally they are already secondarily obscured. 



Fiy. 3. Diagram to explain the method of budding pecuUar to the genus 

 Turbinaria. It is seen to be lateral throughout. The downward 

 " flow " of the coenenchyma has both thickened the stalk and 

 submerged the axial polyp. 



Fig. 3 a. Diagram to illustrate the relation of the buds and coenenchyma 

 to the axial polyp in Madrepora. 



Fig. 3 b. Diagram to show the budding of a radial polyp from below, 

 i, the axial parent polyp ; ~, a radial polyp ; iS, three polyps 

 budding from ~, one on each side and the middle one from 

 below, after 2 has bent upwards as shown in hg. 3. 



Fig. 4. Fragment of an erect fi'ond (of the hifrons type), showing the 

 appearance of periodicity in growth, due to the downward 

 streamings of the coenenchyma. at, the most recent downward 

 flow, submerging the polyps below it (see fig. 5) ; A, a previous 

 very abundant flow, which threw out a ridge along which young 

 calicles appear ; c, the line of progressive decay, which is ad- 

 vancing more rapidly on the left than on the right, where it 

 stands at d. 



Fig. 5. A calicle on the line a in the preceding figure, being submerged 

 by the downward flow of the crenenchyma. 



Fig. G. The initial modification of the young cup which gives rise to the 

 mesenteriform type of <^rowth (diagrammatic). 



Fig. 7. A portion of a section through a glomerate Turbinarian, showing 

 that the method of budding peculiar to the genus need not be 

 affected by the enormous thickness of the coenenchyma. 



