464 ASTROSCLERA WILLEYANA, THE TYPE OF A NEW FAMILY OF SPONGES. 



but no certain indication of the presence of that element could be detected in the small 

 portion at my disposal. 



" All the properties described above lead to the conclusion that the substance is 

 composed of aragonite." 



The Funafuti specimen furnished a larger fragment for analysis and determination 

 of the specific gravity, and of this Mr Hutchinson has made a further examination, with 

 the following confirmatory result : — 



"A small fragment was placed in bromoform, and the liquid diluted with benzol till 

 the substance sank. The tube was now connected to an air-pump, and gently warmed 

 until the liquid boiled under diminished pressure ; the substance gave off a stream 

 of bubbles, and its density had apparently increased. More bromoform was now added 

 until the substance just remained suspended, and on taking the specific gravity of the 

 liquid it was found to be 284 at 15° C. 



"Another portion reduced to powder gave similar results. 



" The substance is therefore considerably denser than calcite, and as organic matter 

 is present, the specific gravity of the mineral constituent is no doubt greater than 

 that observed. 



"The fragment was next dissolved, with the exception of an organic residue, in 

 hydrochloric acid, and tested ; it appeared to be pure calcium carbonate, and no trace 

 of magnesium could be detected." 



The organic basis of the skeleton, which remains after treatment with acids, varies 

 remarkably in amount. In one specimen (to which the following description applies) 

 it is abundant, while in the two others it is very scanty. At the central regions 

 of the skeletal elements a highly refracting material remains, associated with a 

 deej^ly staining substance. These are disposed in radiating lines, and often give rise 

 to the appearance of a bright Maltese cross or a star with more than four rays, 

 whose rays of light and dark shift as the focus is altered (Fig. 10, sic.). The central 

 region of the skeletal element is often sharply limited by a circular boundary, the 

 limit of the deeply staining substance, and measures from 7 — 9;ii in diameter. At 

 the peripheral parts of the skeletal elements a less deeply staining and not highly 

 refracting substance remains, which may have a radial arrangement as in Fig. 6, 

 coming in contact at its edge with adjacent elements, and joining with them by a 

 well-marked dividing line. In other cases the peripheral parts of the skeletal 

 elements are represented by a granular matrix, without radial arrangement, and 

 merging, without a boundary line, into the periphery of neighbouring elements 

 (Fig. 10, sk.). I am unable to account for the diiference between individuals in the 

 amount of the skeletal matrix, or for that between skeletal elements in their structure. 

 It is to be observed that the round central regions of the skeletal elements are of about 

 the same size as the smallest spherules, which are to be found in the jelly. The 

 peripheral region of the element (distinguished by its low refraction and feeble staining 

 power of its matrix) is that which is laid down by concentric additions in the course 

 of growth. In sections of the undecalcified skeleton the two regions are undistinguish- 

 able. The organic basis of the skeleton takes a blue stain with picronignisin, in 



