DuTYOSPONGIDiK 



2i» 



the terminal plate in these genera, and their continuity with the rest of the 

 skeleton, of itself demonstrates the fact that this is the base of the sponge, 

 and is not in the remotest sense homologous with the sieve-plate in Eupli-xj- 

 tella; this evidence is corroborated by the structure of the peripheral frill 

 and its evident adaptation to the purposes of attachment as seen in the 

 character of its distal margin. 



Prom this consideration of the characters of the Dictyosponges and the 

 groups allied to them, may be deduced the following definition of the family : 



Family DICTYOSPONGIDiE, Hall. 



Obcanical, prisirmtic, cylindric, explanate, vase, funnel and saucer-sfiaped, 

 ihin-ioalhd sponges, with the inner layer of the shileton composed of series of reticu- 

 lating vertical and horizontal bundles rnosthj of smooth diactine rods ; the outer 

 layer composed of regidarly au^gianged stauractins forming quadrate meshes 

 with more or less irregularly disposed spicules scattered over the quadrules thus 

 produced (^pinulu^y clavules, etc.)- Regular liexactins in the parenchyma. 

 Single lophophyte generally present. Periloph, marginalia and pleuralia 

 present or absent. No mural areoles / no term,inal sieve-plate. 



Structure of the Skeleton. 

 In contrast to the very considerable geneiic variation and abundant 

 specific representation now recognized in this family, our knowledge of the 

 spiciilar elements of the skeleton is, and must remain, very incomplete. This is 

 the more to be regretted, since the structure of this frame-work is the essential 

 basis of taxonomic arrangement. Among the abundant species of the 

 Devonian, this skeleton, almost without exception, is lost, however clear and 

 undisturbed the original form of the sponge and the impression of its surface 

 characters. In the sands and sandy shales of the Chemung group from which 

 the great maj(jrity of these sponges has been obtained, the original hydrated 

 silica of the skeleton has undergone change, probably through the agency of 

 organic acids resulting from decomposition, and has been replaced first by 

 pyrite, and by oxidation this has passed into limonite and been quickly 

 i-emoved from the porous rock. In rare instances portions of the spicular 

 rods in the condition of limonite have been seen in specimens from the upper 

 Devonian and more often in those from the sandy layers of the lower 

 Carboniferous (Waverly group) of Ohio. The calcareous shales in which the 

 sponges of the Keokuk group at Crawfordsville and Indian Creek, Indiana, 

 are embedded, have, however, retained a greater or less portion of the 



