252 PAPERS FROM THE DEPARTMENT OF MARINE BIOLOGY. 



defined outlines of the individual fragments of this structure, as also by the 

 fact that each plate is apparently always a single crystal, extinguishing, as a 

 consequence, simultaneously between crossed Nicol prisms. 



Crustacea are characterized particularly by a vaguely fibrous structure, 

 which tends, however, to be confused with a more pronounced fibrous struc- 

 ture in certain algal fragments. Not much familiarity with the group was 

 acquired. Where cirripeds or ostracods have been differentiated from other 

 Crustacea it was by external characters. 



Worm-tubes were recognized only by external form. Their microscopic 

 characters seemed difficult to recognize, especially because the samples of 

 known material generally contained, inside the tube, fragments of a number of 

 other organisms, which after crushing appeared on the slide mixed with the 

 fragments of the tube itself. 



Characters for differentiating millepores from madrepores under the 

 microscope were not worked out. On inspection they appeared extremely 

 similar. It seemed, however, as though millepores if present should be 

 recognizable externally by their cellular structures, even in very small grains. 



The manipulation of the count varied somewhat in different sizes. 

 Some effort was made to keep a proportion between the number of grains 

 determined in the particular size-portion and the ratio that the portion bore 

 to the entire sand, but to have kept the ratio at all exact would have necessi- 

 tated counting more grains in the coarser portions or fewer in the fine than 

 would be worth while, so that the ratio maintained was only rough. In the 

 coarser portions the weight as well as the number of grains taken for identi- 

 fication was determined — except in the "fine gravel" of No. 27353a, which was 

 the first studied, before the importance of determining the weight was recog- 

 nized. The object, then, was to preserve from the weighed portions a large 

 number of the original grains of each group and from their weight to calculate 

 the weight of the total of grains of each group originally present, in such a way 

 as to check with the original weight. 



The method of calculating the weights of these portions will be presented 

 later. At present the method of determining the number of different kinds 

 of grains is to be described. As finally developed, it consisted in identifying 

 as many grains as possible directly by their external appearance. These were 

 put aside in small vials. It was found that this identification could be best 

 conducted by grouping the grains that appeared to be most alike in color, 

 texture, translucency, form, etc. By identifying a certain number of these 

 from their internal structure under the microscope the identification, by 

 external appearance, of those most closely resembling them was assured. 

 It was found that generally the algae were most certainly recognized by a 

 certain rough, porous, opaque appearance. For determining the microscopic 

 character the effort was made, as often as possible, to take only part of the 

 grain. This on the one hand gives identified grains for future comparison 

 and on the other gives more material for weighing. In order to get the true 



