f 



108 ROYAL SOCIETY OF CANADA 



Genus JIYALOSTKLIA, Hindo. 



7.— Hyalostelia Metihsica, Dawson. 



(Vi^. 17. PI. III., Fig. 10.) 



F\C: n.—Hyaloatclia Metissica. Spicules x 5. , 



Geneml form ln-oad, witli a large oscuhim and a numher of stout 

 anchoring-n),ls. Body-wall lornied of single long slender rods, woven into 

 a very loose mesh and HU])porting cruciform and other spicides of varied 

 form, attached only hy the soft memhranes, so that they are almost always 

 found loose and disarranged. Up to 1889 I had seeii the species only "in 

 this condition ; out was so fortunate in 1895 as to find a small specimen 

 retaining its form, which I have figured in Plate III., Fig. 10. The spiral 

 ancht)ring-rod8 figured with this species in 1889 really belong to Pala^o- 

 saceus, which at one time I was disposed to connect Avith Hyalostelia, but 

 am now convinced that they are altogether distinct, though Hyalostelia 

 may form a connecting link IxMween the Protospongida^ and the Dictyo- 

 spongida', its body-wall being formed not of cruciform spicules, but'of 

 long slender and single rods woven together into squaro or rhombic 

 meshes. , 



As to the use of the generic name Hyalostelia for this species, I am, 

 by no means certain, since the sponges included pi-eviously in that genus 

 are very imperfectly known to me. 



