KONGL. SV. VET. AKADEMIENS HANDLINGAR. BAND. 27. N:0 3. 99 



terminal fissures. Axial area indistinct; central area a broad fascia, widened towards the margins. 

 Strise strongly radiate in the middle, convergent at tlie ends, 9 to 10 in 0,oi mm. Zonal view of 

 the frustule rectangnlar with strongly divergent striae. — Alloioneis Stauntonii GtRUN. Cl. M. D. 

 304. Eoss. D. Öst. Ung. p. 142 Pl. XXX f. 36. Amphora naviculacea Donk. M. J. 1861 p. 11 

 Pl. I f . 12? 



Marine, aestuaries: Scotland! 



Ampliora Ehb. (1840). 



The first known species of Amphora is A. ovalis, described as Nnvicula Amphora by Eheen- 

 BBRG 1831. The genus Amphora was established by the same author 1840 (Ber. p. 11). In the 

 »Bacillarien» Kutzing 1844 described 18 species only, but this uumber was greatly increased by 

 Gregokt (Diat. of Clyde 1857), wbo named 32 new species and first made the distinction between 

 forms with complex and not comples connecting zone. Several other authors have since added 

 new species and in the year 1873 Professor H. L. Smith published (Lens p. 65) a synopsis of all 

 the known forms. By the issue of the plates XXV to XXVni (1875) and XXXIX, XL (1876) 

 of A. Schmidt's Atlas the number of species was greatly increased. Since then new species 

 have been added, and in the Sylloge of De Toni (1891) the number amounts to 221. 



An inspectiou of these species shews that it is impossible to give any diagnosis of the 

 genus Amphora, which is sufficient to distinguish it from Cymbella. The following seems to be 

 the only possible diagnosis of Cymbella and Amphora together: 



Naviculoid diatoms, with both valves similar and asymmetrical along the longitudinal axis. 



The distinction between Amphora and Cymhella is, so far I can see, no other than the 

 degree of asymmetry; the ventral and dorsal side of Cymbella being in the same plane, but in 

 Amphora in planes crossing each other in an angle, which is variable. 



Amphora and Cymbella are only asymmetrical forms of Naviculfe, belonging to difFerent 

 types. There are in the same species gradiial passages from perfectly symmetrical to asymme- 

 trical amphora-like forms (as in Trachyneis aspera). In several groups of Navicula more or less 

 asymmetrical forms occur (for instance Pinnularia Stauntonii and others, formerly named Alloioneis) 

 so closely connected with the symmetrical that it would be unnatural to separate them. The 

 asymmetrical form is not a sufficient characteristic for a natural family, but is merely a facies, 

 which may occur in groups of very different types and seem to depend on the method of growth, 

 Amphorse occuring attached to algse and other objects. This genus is in short to be considered, as 

 well as Achnanthes and Cocconeis, as degenerated forms. To trace the origin of these forms is in 

 most cases diflicult, as the intermediate passages are löst or unknown, but we may get some ap- 

 proximate knowledge of the original types by the study of the structure of the valve and by 

 comparing it with that of difFerent types of Navicula. 



The Cymbellfe appear, to a great extent, to be asymmetrical forms of the section Naviculce 

 lineolatce, and the same may be the case with the still imperfectly known Amphora labuensis. 



Amphora Clevei is no doubt nearly akin to the genus Trachyneis. 



Amphora elegans Pbrag. is with great probability allied to the section Naviculce orthostichm. 



As to the other large number of Amphorse, they may be classed in forms with and with- 

 out longitudinal lines. Those with lines are probably asymmetrical forms of Diploneis or 

 allied genera. There are in some species of this section forms with a structure so closely re- 

 sembling that of Diploneis that the idea of their connection presents itself at once to the mind, 

 notwithstanding the diiferent shape of the valves. In tlle large Amphora nodosa we have a form 

 with coarse, transverse costée, alternating with rows of ocelli, as in Diploneis JBeyrichiana, D. lesi- 



