THE GENUS A.LVEOPORA WITH THE EAYOSITID^. 503 



sents the line o£ fusion of two outer walls *. In the same figure 

 it will be noticed that at each point where the walls meet there 

 is a star. These stars appear to be of the same nature as the black 

 lines ; but as it is just possible that they may be secondary results 

 of fossilization and cracking, I would not lay too much stress 

 on them. If, however, they are what they appear to be, they 

 afford additional evidence as to the nature of the black line, 

 viz. that it was secreted under a rising ectodermal fold; its 

 starring at the angles may be intelligible from this point of view, 

 but is certainly unintelligible at the meeting-point of three 

 outer walls. 



We have thus, it seems to me, enough evidence to show that 

 there is apparently no fundamental morphological difference 

 between the recent Alveopora and the Palaeozoic Favosites. Both 

 start in an epithecal cup, new calicles are marked off from the 

 parent calicle in the same way *, i. e. primitively by septal spines, 

 which in fusing secondarily become lamellate walls, and this 

 j)rocess goes on until a colony is produced all within the same 

 epitheca. All the calicles produced in the central regions have 

 their walls purely septal, while those close against the epitheca 

 have their walls partly formed by it. As in Alveopora, again, each 

 new calicle is apparently marked off from one of the angles of 

 its parent. And, lastly, the skeleton of Alveopora in its earliest 

 stage has every claim to represent a very primitive form of the 

 Madreporarian skeleton (see below). 



So far, then, we see no valid reason why the modern Alveopora 

 should not be classed with Favosites as one of the Favositidae, 

 as proposed by Dana. But how far the genera usually placed in 

 this family t are related to one another can only be decided by 

 careful study of each separately from the new point of view. I 

 would now merely confine myself to calling attention to the 

 Devonian genus Arceopora of Nicholson and Etheridge jun. as 

 showing very striking resemblance to the modern Alveopora. 



The chief difficulties in the way of this association appear to 

 lie in the facts (1) that, with the exception of the little-known 

 genus KonincJda, M.-B. & H., from the Cretaceous, no con- 

 necting links have been found or suggested to tide over the 

 enormous interval of time which has elapsed since Favosites 

 flourished; (2) that it does appear possible* to explain the 



* But see Postscript, infra, p. 515. 



t Cf. the list given by Nicholson in his Pal. Tabulate Corals, pp. 35-36. 



36* 



