r.Y L HEAD BDW LRDS. 2 I 



, the vessel is filled up with water, stirred, allowed to 

 settle for the same length of time, and poured into the 



1. This is repeated until it has been done at least six 

 times, when we shall find all of the sand, free from diatoms, in 

 the small beaker. This can be thrown away, and as soon as the 

 material in the large beaker has settled, it is returned to the 

 small one, and the same process gone through with, on!; 

 tending the time of settling now to about ten seconds. The 

 next density is that which settles in twenty seconds; and on on, 

 five or six densities may he obtained, and if carefully prepared 

 they will be found to contain forms varying very much one from 

 the other. The large species of Jriceratium, Aulacodiscus, and 

 the like, will be found in the coarsest density, and the broken 

 diatoms in the lightest. 



Preserving and mounting specimens so (is to have them in </ 

 condition for study at any fvtu Of course, when possi- 



ble, diatomacese should be studied in the living condition. But 

 there are many forms which have not been as yet found Living, 

 and these can only be studied as dead skeletons; and, in fact, it 

 is in the dead skeletons of the diatomacese that many of the 

 most marked characteristics are to be found; and on such char- 

 acteristics species have been founded. Besides, the most beau- 

 tiful sculpturing of the valves is only to be seen after everything 

 has been removed but the siliceous cell wall I have termed the 

 .skeleton. Therefore I advocate the cleaning of a portion, at 

 least, of every gathering in the manner described, so that noth- 

 ing will be left but the clean siliceous cell-wall. 



If we desire to keep specimens in a slate as near that they 

 present when living as possible, we have to put them up in some 

 preservative lluid in which they will not decay, ami in which the 



softer parts will be preserved. Unfortunately these soft parts 



do not keep well; hi it t he tl ! i id which 1 have found to be the 



best for the purpose Is distilled water, which has to every fluid 



ounce two or three drops of wood creosote added, and there- 

 after a snthcient number of drops of alcohol, which will be 



about double the number of the drops ol en > make the 



OreOSOte Soluble in the water, which it is only to a very slight 

 degree under ordinary conditions. I do not advocate any fluid 



containing glycerine, or. in ; the pr. fluids 



