484 Mr. C. Mereschkowsky on 



the ends of the valve ; they do not form any regular row, as 

 in other species, but are disposed at rather variable intervals 

 one from another, and their apices are usually not very acute, 

 often bent in a hook, sometimes bifid (fig. 6). The surface 

 of the valve shows very peculiar, coarse, irregularly contorted 

 lines or markings, resembling somewhat Arabic inscriptions 

 or shorthand writing signs ; these markings are sometimes 

 distributed all over the surface of the valve, and in this case 

 the diatom has a striking resemblance to Liradiscus ovalis, 

 Grev., the more so as the general outlines of both are nearly 

 the same * ; but usually the lines or markings are not so nume- 

 rous, sometimes even more scantily represented than in the 

 figure 4, and sometimes there are only a few of them, as 

 shown in the figure 5. As will be seen from fig. 7, these 

 lines or markings are irregular membranaceous ribs elevated 

 over the surface of the valve. 



Length : 0024 0-0245 00265 Q-Q27 0031 0042 Q-042 0049 

 Breadth :0 : 0T3 0014 X 00145 X 0014 X 0-015' 



13. Chaetoceros diadema (Ehr.), Gran. Common. 



A very characteristic species, of which I have seen not 

 only the endocysts, but the frustules also. Filaments of this 

 species with endocysts enclosed in their cellules have been 

 recently observed and figured by H. Gran f. 



14. Chsetoceros didymus, Ehr. Very common. 



15. Chsetoceros incurvus, Bail. Rather rare; known only 



by its endocysts. (PI. XVI. figs. 1, 2.) 



This very peculiar species is easily recognized by the 

 disposition of the four awns with which the endocysts are 

 provided. The figures given by Brightwell are correct as to 

 their general outlines, but this author has failed to represent 

 the granulation of the surface of the valves ; the granules 

 are small and irregularly scattered over the whole surface of 

 the body of the endocysts (figs. 1, 2). In the above-cited 

 work of H. Gran we find the description of a new species of 

 Chcetoceros — the Chcetoceros cinctus, Gran, which has endo- 

 cysts very similar to those of C. incurvus, Bail. J, differing 



* It would not be surprising if Liradiscus should prove to be the endo- 

 cyst of some unknown species of Chcetoceros. 



f H. Gran, in ' Norske Nordhavs-Expedition/ plate ii. fig. 18. 

 \ H. Gran, I. c. plate ii. figs. 23-27. 



