310 J. DEBY ON THE STRUCTURE OP THE DIATOM VALVE. 



exist, should become enlarged. Now, the very parts which have 

 been called orifices by some and depressions by others, are the last 

 of all to disappear as the shell is dissolved. This mode of obser- 

 vation, besides establishing the fact that these are really the 

 thickest parts of the shells, reveals many interesting particulars of 

 structure in the various genera of Diatomaceas. Thus in the large 

 species of Pinnularia it may be seen with even a low power, that 

 the two parallel bands (separated by a canal) which reach from the 

 central knob to the terminal ones, and which appear smooth before 

 the application of the acid, become distinctly striated after their 

 surface is dissolved off, as does also the central spot itself, showing 

 that stride which existed in the young shell are covered up and 

 nearly obliterated by subsequent deposits. 



" In Stauroneis the Cross-band and the two longitudinal bands 

 are the last to dissolve, and these last bands, as in most of the 

 family, appear separated by what is either a canal or a very thin 

 portion of the shell. 



" In Grammatophora the undulating lines are internal plates 

 which are the last to dissolve. In Heliopelta, Actinoptychus, &c, 

 the polygonal central spot is the last to disappear. In Isthmia, 

 the sj)ots on the surface, ivliich at first appear like granular pro- 

 jections, are in reality thin portions of the shell, and under the action 

 of the acid they soon become real holes* The acid also proves that 

 the larger spots at the transverse bands are really a series of large 

 arcuate holes in the siliceous shell, and the piers of this series of 

 arches remain some time after all the rest of the shell has 

 vanished. Many other interesting facts are revealed by the action 

 of this acid on these shells, and no one can use it without learning 

 much with regard to their true structure. 



" A few directions with regard to the mode of manipulation in 

 these experiments will probably be useful. As the fumes of the 

 hydrofluoric acid, if they reached the lenses, would greatly injure 

 them, I would advise experimenters (even if they have a micro- 

 chemical stage) to protect the front face of their objectives by 

 temporarily cementing to them a thin plate of mica by means of 

 Canada balsam. This can be attached or removed in a few 

 moments, and completely protects the lens without materially 

 affecting its optical power. As mica resists the action of hydro- 

 fuoric acid much better than glass does, I prepare the cell in which 



* The italics are my own. — J. D. 



