[.June, 1857.] 121 



THE DIATOMACE^: 



Their Characteristics and Structure ; together with Remarks on 

 Collecting and Examining this Order of Plants. Communicated 

 by Henry J. Church. 



[The following pages are condensed from Rabenhorst's work, 

 'Die Siiszwasser Diatomacese ' (Berlin, 1853) — 'The Fresh- water 

 Diatomacese/ — a cheap and useful book. They form by far the 

 larger part of the Introduction j a few passages only being omit- 

 ted, in order to bring the present Paper within due limits.] 



The Natural Order Diatomacect* consists of unicellular organ- 

 isms, possessing a prismatic flinty skeleton or frame, and con- 

 taining a peculiar golden-yellow or brown colouring matter. 

 Propagation — by the division of the parent-cell into two new 

 cells. 



No order in organic natm'e is more strictly defined than 

 that of the Diatomacece. Their inflexible siliceous frame- 

 work — destructible neither by a red-heat nor by decay — distin- 

 guishes them at once from all known organized bodies. If their 

 near neighbours, the Desmidiece, approach them as regards form, 

 a simple experiment at once removes all doubt. This experi- 

 ment consists in placing a small portion of the specimen in ques- 

 tion upon mica or platinum foil, and exposing it to a red heat 

 by the aid of a spirit-lamp. AU organic matter is thus destroyed, 

 and only the frame of the Diatom remains, its form unchanged ; 

 while, on account of the removal of its contents, its structure is 

 more sharply and clearly defined. Care must be taken however 

 that no salts of potash or soda be present ; otherwise, as is well 

 known, they would enter into chemical combination with the si- 

 lica, and produce glass. 



If the Diatom be infallibly distinguished by the material of 

 its exterior, we meet with yet other characters hardly less pecu- 

 liar. The form of its framework is highly symmetrical ; round 

 {rund), discoid, cylindrical, or prismatically quadrangular, with 

 acute angles, and generally plane extremities. The two surfaces 

 which correspond in position, almost invariably correspond also 

 in form; hence, in diagnoses, we need only describe one side 

 and one extremity. Sometimes the former is the more deve- 



* From the Greek 5ict, through, and Tejxvta, I out. 

 N. S. VOL. II. 11 



