Intkoductory. 



23 



close together, or are diactine rods (with triaxial foundation), or a combina- 

 tion of both, is not yet certain. Very small but distinct stauractins are 

 present. [Sometimes with tenninal sieve-plate (?)]." Although the characters 

 of the family can now be stated with greater precision, still the above 

 definition serves to indicate the structural features which distinguish this 

 from other groups of Lyssacine spcmges. 



The family Puotospongid/e, founded by IIinde and somewhat restncted 

 by K.AUFF, includes a number of Cambrian and lower Silurian forms 

 (Protospongia, Salter, Diagonella, DawfWn, Phormohella, IIinde), in 

 which the very thin net-work consists almost exclusively of a single 

 layer of cruciform spicules which lie at the angles of the quadrate meshes, 

 their rays forming the sides of the meshes, but according to Rattff, not 

 overlapping to form fascicles. In these species all the subordinate series of 

 meshes are similarly formed. In addition to the regular hexactins or pentactins 

 of the reticulmu there are also prostalia in the fonn of a few strong root-threads ; 

 sometimes there is a row of short rods about the margin of the osculum. 

 Small lateral surface rods (pleuralia) are also present in some instances. 

 These structures are shown in the species from the Little Metis slates 

 described by Sir William Dawson,* Protospongia mononema, P. tetranema, 

 P. coronata, etc., and it appears from an examination of some of this very inter- 

 esting material that the cruciform spicules may extend so far as to overlap 

 one another, and also that there are pi'esent a few fllifonn spicules lying along 

 the axes of the stauractins. In some of these sponges, (Diagonella coronata 

 and D. cyathiformis, Dawson) the meshes are not bounded by vertical and 

 horizontal sides but are arranged in oblique rows. Recognizing these as 

 characters of the Protospongid^, it seems somewhat difficult to find substan- 

 tial belief in the usefulness of the proposed family Plectospongid^, Rauff,f 

 which ajipeare from the description to have for its principal characteristic the 

 formation of spicular bundles by the overlapping of the filiform rays of the 

 stauractins, together with other rod-like spicules. Dr. Raitj"f distinguishes 

 in the skeleton a dermal layer which can not be discovered in the Proto- 

 SPONGLD^, but is always clearly defined in the Dictyospongidve. Cyatiiopiiycus, 

 its best known genus, also shows evidence of large circular mural openings, as 



"New Species of Fossil Sponges from the Siluro-Cambrian at Little Metis on the Lower St. Lawrence ; 

 Trans. Roy. Soe. Canada, vol. vii, Sect. Iv, pp. 31-55, figs. 1-27, pi. iii, 1889. 



tit may be observed that this name has not been formed strictly iu accordance with the rulesof nomen- 

 clature. There is no genus Plectosponoia and there should not, therefore, be a family termed Plec- 

 rospoNGiD/E. The type of the family is not specified but it is evident that the Cyathophycus reticulatun, 

 Walcott, has furnislied the conception of family characters, although the group also includes the genua 

 Pl,ECTODERMA, Hinde. 



