Vol. 55.] DEVONIAN EOCKS OP NEW SOUTH WALES. 61 



they are limited to a few spicules of siliceous sponges and some 

 minute spined or dentate plates, probably of some invertebrate 

 organism. The sponge-spicules are small, detached, fusiform acerates 

 and cylinders, from 0-16 to 0-36 mm. in length, which pro- 

 bably belong to monactinellid sponges. No trace of any diatoms 

 has been observed among the radiolaria of the siliceous lime- 

 stones ; if any had been present originally, it is probable that they 

 would have been preserved, seeing that the equally delicate lattice- 

 structures of the radiolaria have, in numerous instances, remained 

 practically unaffected by the fossilization. 



As mentioned by Messrs. David & Pittman, impressions of Lepido- 

 dendron aiist7rde are present in the same rock with numerous 

 chalcedonic casts of radiolaria, and specimens of the same plant also 

 occur in beds of tuS on two or three horizons. It may be rash to 

 express an opinion without having seen the facts in the field, but it 

 does not seem improbable that both the plant-fragments and the 

 tuff may have been carried by the wind and wave-currents to some 

 distance from land and deposited in deep water, in the same manner 

 as the masses of leaves, pieces of bamboo, sugar-cane, and other 

 debris which Prof. A. Agassiz dredged up from a depth of over 

 1000 fathoms, and at a distance of 10 to 15 miles from land, off 

 Caribbean Islands.^ 



It does not yet appear that radiolaria have been found in the 

 bluish coral -limestone associated with the radiolarian series at 

 Tarn worth, which in places reaches a thickness of 1000 feet. In 

 microscopic sections of this rock (jS^o. 258 b) which I have examined 

 there were only portions of corals which had been replaced by 

 crystalline calcite. This rock has probably been formed under 

 conditions, as to depth, different from those in which the radiolarian 

 limestones and claystones were deposited. 



VI. SUMMAET. 



In the important series of radiolarian rocks, some thousands of 

 feet in thickness, which have been traced out by Messrs. David & 

 Pittman in New South Wales at Bingara, Barraba, and Tamworth, 

 these organisms occur in beds of chert, siliceous limestones, clay- 

 stones and shales, and in sedimentary volcanic tuffs. In the cherts 

 the radiolaria are usually thickly grouped together, filling tJne rock ; 

 they are nearly entirely in the condition of casts infilled with clear 

 silica and without structure, thus precisely similar to those in chert 

 with radiolaria known in other parts of the world. In the siliceous 

 limestones, on the other hand, the radiolaria not infrequently 

 retain their delicate structures practically unaltered by fossilization ; 

 they are now embedded in a groundmass of crystalline calcite which 

 has infiltrated their hollow spaces, and when this is eliminated by 

 weak acid the rock is seen to be composed of an entangled tissue 

 of entire and fragmentary forms and fine debris. The claystones 



1 'Three Cruises of the Blake,' 1877-1880, Bull. Mus. Comp. Zool. vol. xiv 

 (1888) p. 291. 



