APPENDIX. 123 



tallogenic forces; on the contrary, the structure was completely amorphous, and such as 

 could have resulted only from a mere aggregation of extremely minute granules, them- 

 selves crystalline. Although there seemed to be nothing of the tessellated structure, which 

 has been detected in the shells of many molluscs, yet the observations may point to a 

 similarity of formation ; and it is possible that with more skilful manipulation and higher 

 lenses, something more satisfactory might be ascertained. It should be observed, however, 

 that the beautiful results obtained by Mr. Carpenter, in his investigations into the struc- 

 ture of shells, were generally distinct when a power of only fifty diameters was used. 



The structure of the spicula in the Alcyonia appears to be somewhat different. These 

 spicula are commonly five or six times as long as their greatest breadth, and are more or 

 less pointed at each extremity. The surface is usually rough with minute prominences, 

 so that a profile, seen in any direction, gives nearly the appearance in the figures on 

 page 54, though often still more irregular. In the Spoggodia florida of Lesson, their 

 length is equal to full twenty diameters, and the spicula are mostly a little curved. They 

 lie in every position in the thin integuments which constitute these zoophytes (figure 4 c, 

 plate 59), and are exsert about the small clusters of polyps. But in the Alcyonia they 

 are generally much shorter, and often lie nearly parallel, through much of their texture; 

 and near the bases of the tentacles, there are two oblique divergent series, corresponding 

 apparently with the lines of tissue. 



One of the spicula of the Spoggodia is figured, enlarged, on plate 59, fig. 4 d. Al- 

 though unusual in length, the knobby character of its surface is the prevailing one in 

 the Alcyonaria. When polished down very thin, the appearance in figure 4 e, is pre- 

 sented. Faint lines varying in distinctness are seen to run parallel with the edge, 

 through all its uneven outline; that is, there is evidence of a concentric structure, evinc- 

 ing that the spicula are formed by successive superpositions over the irregular surface. 

 About the central portions of the figure there are a few oblong dark spots, each of which 

 pertained to one of the surface knobs that had been polished off". The concentric layers 

 in these transsected knobs, have their edges towards the observer, and consequently they 

 are not as transparent as the flat parts between. It is also seen that these prominences, 

 traced inward, become a little oblique at the centre, from which they appear to radiate, 

 showing that the spicula, as they were formed, increased most rapidly towards one ex- 

 tremity. In polarised light the spicula exhibit brilliant colours ; and the same magnify- 

 ing power which gave scarcely visible points of colour with the lamelke of the Astrtea, 

 here afforded sheets half an inch or more in breadth, of rich green and flame tints. The 

 spicula appeared therefore to be the result of a simple crystalline superposition of the 

 calcareous material from the depositing secretions. With a power magnifying three 

 hundred diameters, no trace of animal cellules was distinguished, and no regular texture 

 apart from the evidences of a concentric structure above stated. 



The hardness of these coral secretions, which is much above that of common carbonate 

 of lime, as stated by Mr. Silliman on a following page, is not fully explained by the 

 peculiar chemical composition detected by this chemist. We suggest, as one cause, that 

 the calcareous portion may have, in its intimate texture, the structure of Arragonite, or 

 prismatic carbonate of lime, instead of that of common rhombohedral calc spar. The 

 Arragonite structure has been shown to be due to crystallization at a higher temperature 

 than that which is required for calc spar, the two minerals being identical in composition: 

 in consequence of this higher temperature, a different crystalline form is assumed ; and, 



