8 Dr. Schumanns Formula} for Diatom-lines. By W. J. Dickie. 
from that of the transverse lines. Consequently the entire structure 
consists of numerous rectangles, standing upon and near one 
another. 
On the contrary, in other species of Navicula the puncta of the 
next succeeding series are, as it were, put out of place, and not till 
the third series do we find a repetition of the first. 
In this case the structure is divided into oblique-angled paral- 
lelograms. 
The former case I call the corresponding, the latter the alter- 
nating series. 
In Figure 1 let G A be the longitudinal axis of the frustule (or 
a line parallel to it), and the horizontal lines G H, C J, A K, the 
so-called transverse striae, and the lines standing perpendicular to 
them, the longitudinal striae. 
The structure then consists of rectangles, one of which is 
A C B F. This I divide by a diagonal into two right-angled tri- 
angles, and take the triangle A B C as the basis of the structure. 
Let a. denote the width apart of the horizontal striae, and /3 the 
width* apart of the vertical striae, and 7 the shortest distance apart 
of the inclined striae ED, AH, &c. 
Ehrenberg more than thirty years ago drew attention to the 
value of the magnitude a, as he found it, in the same species, 
pretty nearly a constant. He ranked it therefore as a characteristic 
