594 APPENDIX. 



are seen thickened spots of siliceous matter (like drops of glass on the surface of 

 a bottle) ; but by no means a round hole, as it is represented by Ehrenberg and 

 many others. That such a hole is decidedly sometimes not present in the centre, 

 is seen unanswerably in such fragments as fig. 3., and especially fig. 4., which may 

 be easily obtained by crushing the shield. On each side the central clefts are 

 seen a great number of oblique clefts, which, according to the direction of the 

 light through the focus, appear smaller or larger. In these spots the shield con- 

 sists of two leaves lying one over the other. These leaves are penetrated with 

 the small clefts which, where both the lamellae touch each other, are somewhat 

 broader, which explains the varying breadth of the clefts according to the altera- 

 tion of the foci. Fragments in which this structure is clearly represented may 

 be frequently obtained by crushing the shield (fig. 6.). 



Fig. 2. Lateral view of the shield. The three enlargements are seen here on 

 both sides, from which it is very evident, and what we might have supposed at 

 first sight, that the central enlargement is a little depression upon the external 

 surface. The two sides of the shield exhibit only a few of the oblique stripes : in 

 the centre is seen a broad smooth surface, which is traversed in its whole length by 

 two parallel clefts. In this figure is seen more strikingly than'^in fig. 1. the double 

 contour which denotes the thickness of the wall of the shield, and which sud- 

 denly ceases both above and below. This clearly shows that a passage exists 

 from the top to the bottom of the shield, which may be easily confirmed, if the 

 shield, or, what is better, the same obliquely fractured, is looked at from above. 

 This may be done by taking some of the siliceous earth of Erbsdorff and mixing it 

 with mucilage, and before it is perfectly hardened cutting off delicate plates with 

 a razor. Fig. 5. exhibits a section of the upper part of a shield prepared in this 

 way. 



Such an artificial and complicated structure amongst plants has no explanation, 

 and is entirely without signification. In all actual plants we find the silica pre- 

 sent in quite a different form, as little separate scales or drops, and distributed 

 through the substance of the cell-wall. 



Fig. 7. Spirogym quinina ( 82.) The end of a filament of the plant. A three- 

 fold wall of the cell may be distinguished. Externally, a gelatinous covering (a), 

 which extends over all the cells of the fibre ; under it lies the special cell-mem- 

 brane (b). Both are transparent, and separated by a delicate black line, and are 

 easily distinguished from one another, especially at the commissures where 

 two cells unite. The cell-membrane is clothed on its inner surface with a deli- 

 cate but clearly distinguishable pale yellow layer, of a semi-fluid proteinaceous sub- 

 stance (d). Upon this layer lie the bands, dentated at their edges, of chlorophyll (*), 

 the basis of which is probably wax. These bands are externally furrowed, and 

 take up in the furrows a clear firm substance (c), which may be easily distinguished 

 from a mere interspace if the whole fibre be moistened with tincture of iodine. 

 In the continuity of this transparent substance (vegetable jelly) there exist in- 

 dividual larger or smaller granules (/), which, at least at certain times, consist 

 of starch. Accidentally in the midst of the cell there exists a somewhat elongated 

 cytoblast (g), which contains a nucleolus surrounded by an areola of mucous 

 matter, from which proceed on all sides little streams towards the wall of the 

 cell. There is always in the nitrogenous layer a circulation consisting of innu- 

 merable very varying streams which unite together in a reticular manner. The 

 direction of some of these is shown in the plate by the direction of an arrow. 

 These little streams are so changeable, that if a drawing from nature be made 

 of the system of streams, in the course of a quarter of an hour, on comparing it 

 with the original it will be found that every one of the streams has taken a dif- 

 ferent direction. If a representation of the new streams is made, and thus on 

 from one quarter of an hour to another, it will be found that they keep on chang- 

 ing, and that the whole layer of nitrogenous matter takes part in the movement 

 of the little streams. The remainder of the contents of the cell contains a trans- 

 parent fluid. 



Fig. 8. Mould found growing on the stems of Passiflora alata ; the upper part 

 of the little plant with a lateral branch. In this case also the nitrogenous layer 

 circulates in little streams. This little plant exhibits a tolerably complete history 



