ANATOMY OP THE MADREPOEARIAN CORAL FUNGI A. 311 



shown, is perforate in its more peripheral portions, imperforate 

 in its central portion, and as age increases the imperforate 

 area increases largely. The canals passing through the per- 

 forate portion, and putting the intra-thecal in communication 

 with the extra-thecal coelenteron are, no doubt, homologous 

 with the system of canals described by Fowler in Hhodo- 

 psammia parallela. I can offer no explanation of the 

 origin and significance of the synapticula. Physiologically 

 they seem to serve as stays or buttresses, giving solidity and 

 coherence to the corallum. 



The most important result of my researches seems to rae to 

 be the strong evidence furnished in favour of von Koch^s 

 theory of the formation of the skeleton in the Madreporaria, 

 the evidence in favour of the existence of extra-thecal coelen- 

 teron being, as I think, particularly conclusive. 



The Mesoglcea, Mesoderm, or Stiitzlamelle in 

 Coelenterata. 



Throughout my paper I have used the name Mesoglcea 

 for the (structureless) supporting membrane which separates 

 the ectoderm from the endoderm in Fungia, as in all the 

 Coelenterata. 



The names given to this layer by German authors are 

 Stiitzlamelle, Zwiscliensubstanz, Gallertschichte, or Mesoderm. 

 Among English authors the use of the name mesoderm has 

 become general in describing it. Whilst the exact significance 

 of this layer in the Coelenterata and its homology with the 

 mesoblast of the higher Metazoa are, to say the least of it, far 

 from being settled, it seems to me that the use of the name 

 mesoderm is highly productive of confusion and error. 



The names ectoderm and endoderm, meaning simply outer 

 and inner skin, were first given by AUmann to the outer and 

 inner cell layers of the Coelenterata (G. J. Allmann, "On the 

 Anatomy and Physiology of Cordylophora,'' * Phil. Trans./ 

 cxliii, 1853), and had they always retained this their original 

 signification there could have been no objection to the use of 



