32 On the PrototJieca of the Madreporaria. 



The curved line a represents the secretion of the basal skin 

 after it has been dragged (?) out of the prototheca by the 

 growth of the walls in height; b represents the secretion 

 formed by the skin after it has become detached from a. 



Fig, 2. The same regarded hypothetically as three separate cups, the 

 lowest thick-walled cup being the prototheca (sens, strict.) ; in 

 it cup a a a is inserted, and cup hhb m aaa. 



Fig. 3. A diagram to explain the morphology of Montlivaltia, The 

 early cups rapidly expand and eventually become a series of 

 saucers, ee . ., supported above one another by the septal folds 

 which run continuously upwards. On the right half of the 

 figure the upper part is in section, showing the tabulate floor 

 and the irregularly bent-up rim. On the left half these rims 

 are shown from the outside as irregular bauds of epitheca 

 running round the coral. 



Fig. 4. An early stage like that of fig. 1, but, having fallen over and 

 resecreted itself at ff, it bends upwards again. The bagging 

 of the detached basal skins will take the shapes shown, and 

 the fossula in the bases of the cups will be on the convex or 

 dorsal side of the curved skeleton. 



Fig. 5. A diagram to show how, if the prototheca proper was wide- 

 mouthed when it fell over, the fossula will come over to the 

 ventral or concave side, a is again the spot where the coral 

 secretes a new attachment. 



Fig. 6. The diagrammatic representation of the arrangement of the 

 septa in the so-called Tetracorallia. It receives a simple 

 explanation as due to tlie necessary rearrangement of the septa 

 in a coral which fell over and was bending up again. See text, 

 p. 11. 



Fig. 7. Diagram to illustrate the method of budding of a prone proto- 

 theca and the subsequent bending upwards of parent and buds 

 which might give rise to such a form as Ilulijsites. 



Fig. 8. Two figures of radicle-formation, aftor Lacaze-Duthiers. 



Fig. 9. Diagram to illustrate the one-sided bend-over of tlie prototheca 

 such as it is suggested would give rise to the Znphrentis 

 giga7itea of Milne-Edwards and Haime. See text, p. 14. 



Fig. 10. Diagram to explain the early flattening out of the prototheca in 

 the Perforata. The rim of the cup creeps outwards all 

 round, generally with successive slight bendings up and theo. 

 down again. | 



Fig. 11. Diagram of the early stages in Palceocyclus. The protothecd 

 proper seems to have fallen over and then suddenly to have 

 widened out, the repetition of this is still more widened out^., 

 and so on. What appears to have been a wa-inkled basail 

 epitheca is not a continuous groNsth like that in fig. 10, but a 

 repetition of so many separate protothecal rims. ) 



Fig. 12. Diagram of the early stage of Cgclolites. The prototheca i^. 

 nearlj' flattened out, but it is still repeated continually, only- 

 instead of the secretions of the successively detached skin/s 

 forming continuous tabulse, they are broken up into vesicular 

 dissepiments. Here also what appears to be the wrinkled, 

 epithecal floor is in reality a concentric series of separate rimsi. 



Figs. 13 a-g. Various forms assumed by the protothecss in Palseozoid: 

 corals, all in the direction of becoming flattened out. The! 

 developmental transitional stages between the deep proto-j 

 theea and these adult forms have still in many cases to bd 

 worked out. ' 



/ 



