52 THE AMERICAN MONTHLY [Feb., 



two-thirds full of water. Then for a small cube of the 

 material put in two or three equal sized lumps of soda 

 and let the water boil briskly for half an hour or more. 

 Then wash thoroughly in four or five changes of water, 

 each time filling up nearly to the top and rotating the 

 saucepan by the handle and allowing to settle for from 

 one to two minutes. Now pour off the water until the 

 saucepan is only about one-third full, add a teaspoonful 

 of pure nitric acid and boil for two or three minutes. 

 Wash again thoroughly to remove the acid, letting the 

 material settle a little longer each time than at the first 

 washing. Pour off well down and fill the saucepan iwo- 

 thirds full again; add a lump of soda, and boil for five 

 minutes. Wash again thoroughly to remove the soda. 

 Pour oft' water well down and transfer material to a tall 

 thin glass or ordinary tumbler and let it settle thor- 

 oughly. Then pour off and transfer material to a porce- 

 lain evaporating dish. Remove as much water as possi- 

 ble and add a good quantity of pure nitric acid. Put in 

 a few large grains of sand to prevent bumping of acid, 

 cover with a saucer, right side up. and boil over an alco- 

 hol lamp in a retort stand for three or four minutes. 

 Wash again several times in tall glass or tumbler, allow- 

 ing it to settle well, and then boil again in the evapoia- 

 ting dish with a small piece of soda. Then wash to re- 

 move the soda and the material will be found (juite clean 

 so clean at least that when mounted in Canada balsam, as 

 it should be, the forms will be transparent enough for 

 study. But if the result is not as satisfactory as one 

 could wish all that is necessary is to boil again in pure 

 nitric acid and then in soda. And if there is an exces- 

 sive amount of foreign material mixed M'ith the Radio- 

 hiria most of it can be removed l)y pouring the cleaned 

 material into a test-tube and allowing it to settle not 

 longer than thirty seconds. After pouring oft', the sedi- 

 ment will be found to be almost entirely Radiolaria. 



