Ini'koductoby. 17 



Recent species and the later fossil forms show that fixation may be effected 

 either by the base {l>asiempliytic) or by one side {pleuremphytiv) of the sponge. 

 The ancdent sponges were largely, indeed almost exclusively basiem2)hytic. 

 This term, however, is quite general, as basal attachment is effected by several 

 special appliances ijopliopliytes). 8ome basiemphytic 82)onge8 spread their 

 substance over the opposing surface so that the Ijasal part may be of greater 

 diameter than the superior portions of the sponge and show no difference 

 therefrom in spicular composition. In others, such as Polylopuus, tufts of 

 spicular threads from the lateral lobes curve downward, and are interwoven 

 into a short basal rope ; while in Phkronema attachment is attained by 

 means of several distinct basal tufts. In Euplectklla the strong vertical 

 threads of the skeleton terminate below in a tuft of fibres which constitutes 

 the tip of the elongate obconical body. IIyalonema possesses a long twisted 

 rope of thick strands, Holascus and Dictyospongia have shorter, untwisted 

 basal bundles. In Tiiysanodictya and Phuagmodictya attachment was effected 

 by means of a frill or collar (^perilAipli) encircling the truncated base, and 

 in the latter genixs appearing to have been aided by a small tuft extended 

 from the basal diaphragm. The Devonian Dictyosponges, Hallodictta, 

 AcTixoDicTYA and Cp^yptodictya, ^vere either spreading forms encrusting the 

 surface upon which they grew, or pleuremphytic sponges attached by special- 

 ized processes yet iinknown. There are many forms among the Dictyosponges 

 whose mode of life must have been like that of Euplkctella, buried in 

 the mud or sand for a considerable portion of their length. Such are 

 Dictyospongia and the long, slender, annulated horns of Ceratodictya. 



The composition of the sponge-skeleton is generally regarded as of funda- 

 mental significance in the classification of the entire division, and the combi- 

 nation of the skeletal elements, the shape of their component parts and their 

 mode of uiuon, are highly essential taxonomic characters. Tlie proper under- 

 standing of the fossil s^ionges is, therefore, gi-eatly embarrassed by the usual 

 nuitilation and destruction of the skeletal parts, or their modification by the 

 process of fossilization. This knowledge can be acquired only by most patient 

 investigation of the best preserved material. 



Since this ^vork does not deal -with the ceratose or calcareous sponges they 

 may be left out of present consideration. 



The Siliceous Sponges — Sheletal Stvuctxire. The structure of the 

 siliceous sponges is simple, though often difficult to elucidate on accoimt of 

 the isolation of a part or all of the spicular elements. The siliceous sponges 

 are either thick or thin-walled. In tlie former the considerable substance of 

 the cup permits a f idl development and differentiation of the external, medial 



