130 SILICEOUS SPONGES. 



spicules in their proper relative positions, is strong evidence .that they were held 

 together by a common membrane ; and the presence of such a membrane in sponges 

 with a similar lattice-like framework of large spicules, as in Cincliderma quadratum 

 for example, also points to the same conclusion. 



Whether this exterior spicular framework constituted the entire sponge is an open 

 question ; at present it is the only structure which has been discovered. 



Prof. Zittel* is of the opinion that this and the following genus will probably 

 constitute a new family ; but as nothing is at present known of the interior cha- 

 racters of these sponges, I prefer to place them provisionally in the family of the 

 Staurodermidaj on account of their resemblance to the dermal layer of some of the 

 sponges of this family. 



Distribution. Menevian : St. David's, South Wales. Lower Lingula Beds : Tyddyn- 

 gwladis. Upper Mawddach : North Wales. Also in black shales of Cambrian age 

 in Norway and Sweden. 



Genus DICTYOPHYTON, Hall, 1863. 



DiCTTOPHTTON TUBEROSUM, Conrcid, sp. (Plate XXVIII. fig. 3.) 



1842. Hydnoceras tuberosum, Conrad, Journ. Ac. Nat. Sci. Philad. vol. viii. 1st ser. p. 267, 



t. 16. f. 1. 

 1863. Dictyophyton tuberosum, Hall, 16th Report of State Cab. New York, p. 90, t. 3. f. 1. 



1879. Dictyophyton, sp., Scliimper, Haudbuch der Pal. 2 Bd. 1 Lief. p. 69. 



1880. Dictyophyton tuberosum, Ferd. Eoemer, Letb. Geogu. 1 Th. p. 127, f. 3. 

 1882. Dictyophyton, Zitt. Neues Jahrbuch, II Bd. p. 204. 



Sponges inverted conical or elongated funnel-shaped, apparently subangular in 

 horizontal section when uncompressed, with projecting nodes disposed in horizontal 

 and vertical rows. Neither the basal extremity nor the summit has been preserved. 

 A fairly large, though incomplete, specimen is 190 mm. in length by 70 in width. 



The surface of the sponge is divided into quadrate and oblong spaces by horizontal 

 and vertical grooves or raised lines, of varying depth and strength, crossing each 

 other at right angles. The stouter lines and grooves form the larger squares, the 

 sides of which are from 4 to 6 mm. in length, and these are divided and subdivided 

 by finer lines precisely in the same manner as in Protos^ongia. No structure can be 

 detected in the interior of the funnel. 



Nothing beyond the outer form and quadrate impressions of the outer surface, in 

 a matrix of micaceous sandstone, remains of this sponge. The regular framework of 

 the surface, however, so closely resembles that of Protos])ongia that there is every 

 reason to suppose that it has been formed by the casts of spicules of different dimen- 

 sions, disposed so as to form a regular lattice-work, as in that genus. It is also 



* Jv^eues Jahrbuch fiir Mineralogie &c., 2 Ed., 18S:i, p. 204. 



