POLYZOA FROM THE SILURIAN FORMATION. 263 



tinctly marked off by impressed lines ; but they are not arranged be- 

 tween elevated longitudinal lines. The margins of the cells are very 

 thick and conspicuous, not granulated, tuberculated, or spinigerous. 



The best preserved fragment examined had a length of eight and a 

 half lines, dividing at its summit into two branches, its diameter being 

 rather more than one-third of a line. 



From its cylindrical form, and the absence of a laminar axis or of non- 

 poriferous margins, it wauld seem certain that this singular form is not 

 a Ptilodictya; but I am at a loss to know where it should properly be 

 placed, its minuteness rendering it very difficult to make out the details 

 of its internal structure, and thus leaving its generic affinities uncer- 

 tain. It has, however, some affinity with Ptilodictya (T) raripora, Hall, 

 from the Clinton group ; and I have, therefore, retained it provisionally 

 in this genus. 



Position and locality : Cincinnati group, Cincinnati. From the cabinet of Mr. U. P. 

 James. 



Ptilodictya fenestelliformis, Mcholson. 



Plate 25, figs. 8, 86. 



Polyzoary palmate or sub-palmate towards the base, dividing distally 

 into small branches. Basal expansion and branches flattened and sharp- 

 edged, the branches being acutely elliptical in cross section, and about 

 three-fourths of a line in thickness centrally. Cells covering the whole 

 surface on both sides, with the exception of the sharp lateral margins, 

 and of certain non-poriferous areas to be afterwards noticed. The cells 

 on the two aspects of the flattened frond have their bases separated by a 

 thin laminar axis. The cells in the middle of the frond are about three- 

 eighths of a line in height, gradually diminishing towards the margins. 

 Cell-mouths ovate, slightly longer than broad, arranged in longitudinal 

 rows, alternate or sub-alternate in contiguous rows. About five cells in 

 one line measured longitudinally, and six in the same space measured 

 diagonally. The longitudinal spaces between the rows of cells are broad 

 and slightly elevated, and are faintly striated longitudinally, or obscurely 

 punctate. On the other hand, the spaces between the ends of the cells 

 are very much narrower; and the surface thus closely resembles that 

 of a small JFenestella, the cell-mouths looking like " fenestrules," the 

 longitudinal interspaces between the cells representing the " interstices," 

 and the narrow spaces between the ends of the cells corresponding with 

 the " dissepiments." The only specimens examined exhibit numerous, 



