BOTTOM SAMPLES, MURRAY ISLAND, THE BAHAMAS, AND FLORIDA. 25 1 



(in addition to predominant CaCOs) have about 15 per cent MgCOs and 

 8 per cent Ca3P208. While the loose spicules, which are probably by far the 

 most abundant skeletal representative of the alcyonarians, are easily recog- 

 nized by their external form, the laminated solid skeleton of such forms as 

 Lepidisis and his have characters very much like the madreporarian skele- 

 ton; the loose spicules, too, when their external form has been lost by wear, 

 are not readily differentiated. Fortunately, the alcyonarians seem scarcely 

 to be included in the materials studied from Murray Island. In the two 

 samples only one spicule was noticed. Where spicules are so scarce it is 

 improbable that there is an appreciable amount of the solid skeleton present 

 and the failure to recognize any alcyonarian material may therefore have been 

 justified. In the first examination of some of the portions a number of grains 

 were tentatively classified as alcyonarian, but on re-examination the char- 

 acters suggesting this group did not seem pronounced enough, in the absence 

 of any evidence that the group was appreciably represented, to justify 

 leaving them there. They were therefore placed with the madreporarians, 

 which they resembled in their general characters. 



The two distinct chemical groups in the algae are the Corallinaceae and 

 the genus Halimeda. The Corallinaceae that have been analyzed contain 

 about 19 per cent of MgCOs, Halimeda only about 0.5 per cent. This chem- 

 ical difference was learned too late for the differentiation of the two groups 

 in the study of the Murray Island sands; but it is very doubtful in any case 

 whether such differentiation was possible except in the coarsest sizes of 

 material. However, here is a question requiring the most careful attention 

 and persistent effort, for both groups are well represented in the sands and 

 the difference in their content of magnesium carbonate, the substance with 

 which this study is most concerned, far exceeds that between the members of 

 any other group considered. 



Mollusks are one of the groups having generally the best-defined char- 

 acters under the microscope. They tend a little at times to confusion with 

 crushed madreporarian material; but, even if the two are occasionally not 

 properly discriminated, it is not important from the chemical point of view, 

 since both are nearly pure CaCOs. 



Foraminifera, at least those present in the Murray Island sands, are 

 more often recognizable by their external characters than any other group. 

 Under the microscope, too, the small perforations of the Perforata are unmis- 

 takable, but the Imperforata are a little more difficult, since they tend to 

 resemble certain fragments of algae or bryozoa. 



Bryozoa were probably very scarce, if at all present. Their microscopic 

 characters seem to be rather mixed, partly resembling corals, partly perhaps 

 algae, but recognition by external character would probably be possible with a 

 large proportion of bryozoan grains. 



Echinoid fragments are by far the most easily recognized under the 

 microscope, by their reticulation in three planes and the curved, sharply 



