Prototheca of the Madreporaria. 29 



of those around it, as, for instance, in Daricinia and 

 Ph illipsastrcea . 



On turning to modern Astrseida?, we find that the tabulate 

 character of the Palaeozoic corals has become obscured, on 

 the other hand the septa have become prominent. These 

 conspicuous radial folds of the prototheca make it difficult to 

 discern the exact character of the concentric foldings of the 

 protothecal wall. 



I would suggest that, as a rule, the rising of the radial 

 septal folds has also raised the concentric rim-folds. We 

 might diagrammatically express it by imagining a cnlicle 

 like that in figure 13 j becoming changed into the form 

 shown in fig. 16, which represents a calicle with high double 

 walls, and on each side of it a smaller bud. We may assume 

 that the tall ring-fold has been formed at the expense of the 

 earlier horizontal tabulate area round the fossa. It is im- 

 possible here to attempt any review of the many Astrasid 

 forms, but, speaking roughly, they are built of groups of low 

 calicles with the protothecre modified in this way. The 

 difference between this and that shown in diagram fig. 14 is 

 that the calicle is shallower and more open. 



Without professing any intimate knowledge, I am inclined 

 to believe that most of the different forms now included 

 among the modern Astrseidae may be referred to variations : — 



(1) In the distances of the corallites from one another : 

 (a) they may be wide apart, as in Orbicella, Solenas/rcca, 

 Echinopora, &c. j (b) they may be close together, Favia, 

 Diploria, &c. ; (c) they may be so close thai the outer wall 

 of the parent supplies the inner wall of the bud, Prionastrcea, 

 Ooniastrcea, Leptastrcea*, &c. ; (d) even these single division- 

 walls may be incomplete, Hydnophora. 



(2) In the ways the intercalicinal valleys are filled up. 



(3) In the characters of the septa and in the way in which 

 they come over the edges of the fossa? and are distributed on 

 the surface of the intervening tissue. 



Concluding Notes on the Terminology of the Walls. 



A wall built by a direct continuation of the edge of the 

 prototheca should, I think, be called prototheca f- These 



* The Palaeozoic Michelinia might be regarded as the morphological 

 equivalent of these forms before the development of septa disguised the 

 protothecal cups. 



t The term epitheca may be retained in its usual sense, and be under- 

 stood to refer to all traces of the primitive undifferentiated protothrcal 

 wall aud rim, even though they have lost all signs of Laving been once 

 parts or expansions of a cup. 



