58 THE 'PALEONTOLOGY OF MINNESOTA. 



[Reoeptaoulltes owenl. 



the outer side of the vertical rays of the spicules they are not septa]. The distance 

 of the cells [circular perforations of the matrix once occupied by the vertical rays of 

 the spicules] from each other is variable, those near the center being closer together, 

 though, in receding from the center, there are at intervals intercalated rows of cells, 

 which take the same direction, and give the cells a closer arrangement towards the 

 margin than in the intermediate space before the intercalation of the additional 

 rows. The apertures [impression of head-plates] both above and below are essen- 

 tially rhomboidal [from 3 to 5 mm. in width] ; but in well preserved surfaces there 

 are remains of rays, which, however, are rarely observed ; and I have not seen them 

 on opposite sides of the same specimen." 



A small specimen of this species from Goodhue county, Minnesota, has the lower 

 surface preserved as crystalline calcite, while all other portions of the sponge are 

 missing. The outline of the plates cannot be determined, but their arrangement is 

 well indicated by a series of knobs arranged in quincunx. These were regarded at 

 first as having been produced by the wearing away of the softer matrix surrounding 

 the crystalline calcite, usually filling the interior of the vertical rays of the spicules 

 in Minnesota specimens. Upon grinding the specimen transversely to the surface, 

 it was discovered that the vertical rays of the spicules are not present, and that only 

 the lower or outer surface of the sponge is preserved. We therefore conclude 

 that each head-plate in this specimen had originally a central knob similar to those 

 figured by Dr. Hinde in Sphcerospongia tessellata Phillips, sp.* In the latter these 

 knobs are comparatively smaller than in Eeceptaculites oweni Hall. 



The upper or inner layer is never preserved entirely in Minnesota specimens of 

 R. oweni, and we shall therefore give Dr. Hinde's description of this integument as it 

 occurs in R. occidentalis Salter, a closely related species : " The vertical rays in this 

 species of Receptaculites continue cylindrical to near their basal extremities, and then 

 abruptly expand into horizontal plates. These plates have four straight sides, but 

 at each of the corners there is a semicircular or semi-elliptical vertical hollo w.f 

 Each plate appears also to be traversed by four horizontal canals, which radiate 

 from the center, where they are in connection with the canal of the vertical ray. 

 These plates are intimately united together so as to form a continuous 

 inner or upper layer. The delimitations of the separate plates in this layer are not 

 always preserved ; in many specimens they appear to have been completely oblit- 

 erated, and the layer resembles a continuous plate with numerous cylindrical or 

 elliptical canals which penetrate through it at right angles'' (loc. cit. p. 825). In 

 many specimens from Minnesota, the horizontal canals of the upper or inner layer 



*Loc.cit. pl.37, fig. lb. 



tHerr liautf states, that these vertical hollows or pores did not originally exist in the " gastral " wall, but are the result 

 of fussilization (Nicholson " Manual of Palaeontology, vol. ii, p. 1564). 



