Species of Poly zoa from the Devonian Rocks. 83 



from tlie base, showing the branches cut across. Above, the 

 branches are separated by the fenestrules; and immediately 

 beneath this are seen the dark oval spaces which represent 

 the cavities of the biserial cells, and two of which are contained 

 within the cavity of each branch. Below this, again, each 

 branch is seen to give off an immense keel or ridge, which is 

 directed inwards towards the interior of the frond. These 

 ridges are separated by deep intervening grooves ; and there 

 can be no doubt that the cells open at the bottom of these 

 gi'ooves, those of one side of the branch opening on one side 

 of the base of the great keel, and those of the other opening 

 upon the opposite side of the same. 



g. This exhibits a greatly magnified transverse section of 

 the frond at a point a little above the base. As in the pre- 

 ceding, we can recognize without difficulty the shallow fenes- 

 trules, the divided branches carrying in their interior each a 

 pair of cells, and the great internal keels. Here, however, 

 we have two new features : fii-st, the deep grooves between 

 the keels are subdivided by delicate calcareous laminfe, which 

 connect the opposite sides of contiguous keels, and divide 

 the intervening grooves into shallow transverse chambers ; 

 secondly, the grooves between the keels are closed internally 

 by a continuous calcareous membrane which has a minutely 

 porous or vesicular structure. 



h. This represents a fragment taken from near the base of 

 the funnel and exhibiting the internal surface of the polyzoary. 

 We see here the inner faces of the great longitudinal keels ; 

 but instead of these being separated by deep sulci as in the 

 upper portion of the frond, they are here separated by shallow 

 rounded grooves formed by a continuous calcareous membrane, 

 which is not penetrated by either the fenestrules or the cells. 

 No apertures, therefore, of any kind appear on the interior of 

 the coenoecium near the base, the open interstitial grooves of 

 the upper portion of the frond being here closed by a continuous 

 connecting layer. At the right-hand comer of the figure the 

 keels and their connecting membrane are broken away, so 

 that we see the cavities of the rows of cells ; whilst the ex- 

 treme corner is still further broken away, so that the fenestrules 

 come into view. 



i. This simply represents a single branch in transverse 

 section, greatly enlarged, and shows the cells in the interior 

 of the branch and the great triangular keel pi-oceeding from 

 its internal surface. 



From the above description it will be evident that the 

 structure of Carinopora is quite anomalous, and wholly unlike 

 any thing that has hitherto been observed in any member of 



