8 F. Cliajrnian: 



These weathered terminals frequently present a rough, resetted 

 appearance, due to the replaoement of the original substance 

 of the pillars by crystalline calcite. The centres of the pillars 

 often contain a residual crystal of calcite towards which the 

 outer and adjacent ones converge, which seems to point to the 

 former existence of a central canal. The oalcite crystals sur- 

 rounding the central area in our specimens sometimes exhibit 

 in transverse section under the microscope a radial grouping 

 of brush-like clusters of fine dark lines, possibly indicative of a 

 previous organic structure as was pointed out by von Giimbel.^ 



Beekite, which was recorded by Etheridge and Dun as oc- 

 curring in the specimens from New South Wales, has not yet 

 been noticed in our Victorian specimens. 



A specimen from Buchan, in grey limestone, shows the rarely 

 occuiring actual form of the summit plates of the lower or 

 outer surface of the body-wall of the sponge. On the opposite, 

 weathered face of the same specimen the condition of the 

 pillars shows that they were axially perforated by a slender 

 canal. 



The microscopical exaanination of a number of thin slices of 

 Receptaculites in Middle Devonian limestone, both from Vic- 

 toria and New South Wales, has revealed isolated and frag- 

 mentary horizontal spicules apparently detached from the ex- 

 ternal layers of the sponge, and distributed through the in- 

 filling material or mud forming the black limestone (Plate IV., 

 Figs. 2, 3). These spicules resemble the rest of the sponge- 

 remains in the same beds in being now in the form of calcite. 

 Hinde has remarked- with regard to the examples of Recep- 

 taculites found in the Ttenton Limestone, that, whilst the axial 

 canals are distinctly shoAvn in the siliceous portions of speci- 

 mens, those parts replaced by calcite have the canals entirely 

 obliterated. One of our specimens, however, which has been 

 replaced by crystalline calcite, shows unmistakable traces of 

 the central canal (Plate IV., Fig. 2). As regards the identity 

 of the spioular body shown on Plate IV., Fig. 3, it may be 

 noted that the space between the two circular prominences 



1 Abhandl. k. bayer. Akad. VViss, 1875, Bd. xii., Abtli. i., p. 192. 



2 Quart. Journ. Geo). Soc, vol. xl., 1884, p. Sin. 



