34 Prof, H. A. Nicholson — New Palaeozoic Polyzoa. 



the frond are straight, but those of the lateral rows are oblique. 

 Cell-mouths unknown. 



In most essential characters, and in general appearance, the genus 

 Heterodictya entirely resembles Ptilodictya. We have, however, the 

 very anomalous and very important feature, that the cells in the 

 present genus are as thoroughly and as regularly tabulate as the coral- 

 lites of Chaetetes or Favosites. This clearly necessitates the removal 

 of Heterodictya from Ptilodictya, and establishes a very interesting 

 transitional link between the Polyzoa and the Tabulate Corals. I 

 am only acquainted with a single, exceedingly large, species which 

 can certainly be referred here. I should however, suspect that Ptilo- 

 dictya (Flustra) lanceolata, Golclfuss, will very probably turn out to 

 be an example of this genus. 



Heterodictya gigantea, Nicholson. Plate II. Figs, la-le. 



Polyzoary forming a single, flattened, unbranched, two-edged 

 frond, the dimensions of which are unknown, though certainly very 

 great. The largest specimen observed expands gradually in width 

 in proceeding from the base upwards. Its length is three and a 

 quarter inches ; the breadth of the broken base is nine lines, and the 

 breadth of the broken distal extremity is fifteen lines. Both ends of 

 this fragment are broken away, and its total length, when perfect, 

 may be estimated with every probability as having been at least 

 half a foot, with a width of not less than two inches. The edges 

 of the frond are quite sharp, and its width in the centre is two lines 

 or a little more. Its cross-section is thus acutely elliptical, the two 

 poriferous surfaces being gently and regularly convex, without any 

 central angulation. The frond is completely divided into two halves 

 by a central laminar axis, which is marked with" longitudinal striae, 

 conforming with the rows of cells, but does not exhibit arched 

 transverse strias. The cells are arranged in longitudinal rows, in 

 three series. The first series is central, and consists of a few rows 

 in which the successive cells are themselves longitudinal, and are not 

 obliquely disjwsed. The remaining two series of rows are lateral, 

 and each consists of a number of rows in which the cells are directed 

 obliquely outwards and upwards as regards the margin of the frond 

 in the direction of the row itself (PI. II. Pigs, lb and le). The 

 general arrangement of the cells is thus penniform. There are about 

 six rows of cells in one line measured transversely, and thus there 

 are about ninety rows altogether at the wider end of the frond. 

 There are four or five cells in the space of one line measured longi- 

 tudinally, and the cells are alternate or sub-alternate in contiguous 

 rows. The cells have the form of cylindrical tubes directed upwards 

 towards the surface at an angle of about 70° with the laminar axis. 

 Each tube is partitioned off transversely by well-developed tabular, 

 of which there are five or six in the space of one line, some of them 

 not quite extending across the tube, but the most of them complete. 

 The bases of the cells, as seen by decortication of the laminar axis, 

 have mostly the form of narrow ovate slits. The free surfaces of the 

 frond, and consequently the characters of the cell-mouths, are un- 

 known. 



