E. M. NELSON ON NAVICULA RHOMBOIDES AND ALLIED FORMS. 95 



for it. Another diatom fairly common in the gathering is 

 Navicula serians. A more minute examination shows that the 

 rhomboid form is strongly marked, especially in those of a 

 medium size. 



The average length of the diatom is 0-0036, which is 4*9 times 

 its breadth ; the transverse striae count 72 to 73 in one-thousandth 

 of an inch. The shape of the ends of the raphe are very impor- 

 tant; when examined by ordinary illumination, such as would 

 have been used at that day, they appear like a tiefoil-headed 

 Gothic window. This diatom will be called in this paper, for the 

 -sake of identification, the English rliomhoides. 



To return to our historical narrative : a few years later quite a 

 commotion was aroused in the microscopical world by an 

 announcement from America that somebody's |-inch objective had 

 resolved the N. rhomhoides. Subsequently it was found that 

 the rhomhoides in question was a large and coarse form that 

 had been found at Bennis Lake. (This is now known as the 

 Cherryfield rhomhoides, which a few years ago Mr. Mainland 

 rediscovered in " Sozodont " tooth powder, from which some of 

 our finest specimens have been obtained.) 



This diatom is about twice as long as the English o^hoynhoides , 

 and it differs from it in the coarseness of its striation, which is 

 pretty constant at 60,000 per inch ; it could therefore be resolved 

 by the 5-inch objectives of that day. Moreover, the shape of 

 the end of the raphe resembles a Romanesque or Norman 

 arch, and the ratio of its length to its breadth is about 5*9 ; 

 further, the rhombic angle is not nearly so distinctly marked in 

 this diatom. If the diatom were of the same species as the 

 English 7'homhoides, one would expect to find the same rhombic 

 angle, the same trefoil-headed raphe, and a striation nearer 

 50,000 than 60,000, for the large forms of this Cherryfield 

 rhomhoides appear to be finer than the small. But to proceed : 

 another Kavicula brought forward as a test about this time 

 was called the Frustulia saxonica. It much resembled the 

 English rhomhoides, for it had both the rhombic angle and 

 the Gothic termination to its raphe distinctly visible, but it 

 <liffered, inasmuch as it was smaller and its striation was 

 finer, being about 82,000 to the inch ; there was also far less 

 variation in the size of the specimens, its ratio being about 4"3. 



The last diatom of the series we have to consider is the 



