PHOTOTYPE PLATE NO. XXXII. 



53 



identity is indeterminable. Here and there, however, may be recognised the discoidal tests of 

 the Foraminifer, Orbitolitcs ; and, on the opposite side to that illustrated, a fragment of the corallum 

 of a Goniastrasa, nearly an inch in diameter, is conspicuously embedded. 



The fine molecular or granular calcareous deposit that encrusts the coarser constituent 

 particles of the platform-rock specimens above referred to represents a very important element in 

 the function of reef-rock construction. It is, in point of fact, the paste-like cement that binds all 

 these independent elements into one consolidated mass, previousl}' held in solution in the sea- 

 water, and precipitated by its evaporation on the retreat of the tide. This cementing action of 

 the sea-water in coral seas, through its saturation with carbonate of lime, is particularly well 

 illustrated on the foreshore of Thursday Island, immediately below ordinary high-water mark ; this 

 area, after submersion and complete infiltration with salt water, is then left for a long interval to the 

 active evaporating agency of the tropical sun. Not only are shell and coral fragments bound 

 together by the lime cement, but even granite pebbles of considerable size are found, on 

 attemptmg to pick them up separately, to be firmly coherent. In a similar manner, washed-up 

 shells, apparently freshly-deposited and lying loosely on the surface of the platform rock, prove to 

 be firmly attached to it by an almost invisibly thin film of lime cement. 



The upper figures, Nos. i to 4 of Plate XXXII., illustrate this binding property of the 

 repeatedly evaporated, lime-impregnated sea-water in a highly instructive manner. These 

 specimens were collected by the author on the foreshore of Sweer's Island in the Gulf 

 of Carpentaria. Nos. 3 and 4 represent a loosely coherent breccia formed of small, almost 

 perfect shells and larger shell-pieces, mingled with fragments of various species of coral, in 

 which the coralla of the several genera Madrepora, Turbinaria, and Porites may be distinctly' 

 recognised. Scattered among these, there is finally a considerable intersprinkling of ironstone 

 gravel. In neither of these two specimens is there any admixture or infiltration of finer grit or 

 sand, the entire constituent fragments being loosely adherent, and falling readily asunder if roughly 

 handled. In Nos. i and 2 from the same beach there is a substantial basis of red feruginous 

 siliceous sand, so consolidated as to form a compact sandstone, among which shells and ironstone 

 gravel, without any admixture of coral, are irregularly scattered. 



This inter-tidal area throughout the lime-saturated, tropical, coral seas undoubtedly re- 

 presents one of the most active and visibly effective of Nature's petrological laboi'atories. Loosely- 

 aggregated breccia, conglomerate, or limestone of such fine and solid texture that it rings with the 

 hammer, with every intermediate variety, are here in visible course of manufacture with a celerity 

 unparalleled under any other conditions. How far the study of this active rock-forming phe- 

 nomenon may or may not aid in a final settlement between the conflicting beliefs in subsidence 

 and elevation under which the Great Barrier Reef of Australia is thought to have been originally 

 constructed is a matter which will receive some attention in the following chapter. 



Had space permitted, another highly interesting and instructive photograph would have 



