( THE society's note-books. 199 



before, and then decant again at five minutes ; and if the diatoms 

 are very small, at ten minutes. The dealers prepare large quanti- 

 ties at a time. They get rid of the coarse sand by " rotating," 

 and of the light sand by filtering through fine gauze. The great 

 point is to have the stuff properly cleaned before decanting and 

 rotating, so that you have clean sand and cleaji diatoms. But you 

 cannot get such stuff as Oamaru clean and bright with acid only. 

 It must be boiled up afterwards with alkali, either carbonate (in 

 which case you must boil it for a considerable time), or caustic (in 

 which case you must use a very small piece and only boil for a 

 minute or two) ; but the material must be bright and clean before 

 you can separate -densities properly. Should the flocculent matter 

 not disappear, try the Liquor Ammonia process. Pour off the 

 water from the deposit closely, put about half-an-inch o( Afmnonia 

 fortissimum into it, shake it well, leave it standing for a day or two 

 (at least twenty-four hours), dilute it well with water, and then pour 

 in enough muriatic acid to neutralise the ammonia. Shake vigor- 

 ously before decanting several times in distilled water. — —Another 

 plan. — After washing away the sulphuric acid, boil with common 

 soda, and then while still hot neutralise with muriatic acid ; wash 

 and repeat the boiling with soda, etc., until all the flocculent 

 matter has disappeared. It should then be well shaken in a 

 bottle and decanted as before. But, oh, the patience required for 

 this work ! 



By the courtesy of our Hon. Secretary, I am able to add some 

 additional Notes, taken from my second Box of Slides. 



Therein I endeavoured to show that we must no longer arrange 

 diatoms by their shapes ; that increase of knowledge brings increase 

 of accuracy. I illustrated this by the genus Stictodiscus. Formerly 

 Triceratium parallelum was arranged in the genus Triceratium. 

 It is now removed from that class, and is put under Stictodiscus, 

 which now assumes either a discoidal, or triangular^ or quadran- 

 gular form. Hence, Stictodiscus Californiciis may be S. Californi- 

 ciis, discoidal form ; or S. C. forma trigona ; or 6". C. forma trigona, 

 var. gibbosa ; or S.C. forma quadrata. The other discoidal forms 

 were S. C, var. areolata, S. Hardmanianus, and S. fiitidus. I may 

 add the description of these as then sent. 



Stictodiscus Californicus has large puncta somewhat remotely 



