KOXGL. SV. VET. AKADEMIENS HANDLINGAR. BAND 27. N:0 3. 163 



Aclinaiitliese. 



Frustule with dissimilar valves, the upper without, the lower with, central nodule and me- 

 dian line, usually bent along the longitudinal (Cocconeis) or along the transverse axis (Achnanthes). 

 As early as 1783 O. F. MtJLLBR observed an Aehnanthes, named by him Conferva armillaria 

 (= A. longipes). Another species was in 1819 named by Lyngbye Echinella stipitata {A. brevipcs). 

 The generic name Achnanthes was given in 1822 by Bory St. Vincent. In his Conspectus criticus 

 diat, 1832, J. Agardh defined the genus as »frustilla (1. articuli) convexa, pauca, in frondem vexilli- 

 formem (vexillum) stipitatum coadunata* (coordinata?) KtJTZiNG, who correctly observed that the 

 lower valves only of Achnanthes and Cocconeis have a central nodule, formed in 1844 a section 

 ^Monostomaticce^ (Bac. p. 70) comprising the Cocconeidece, with the genera Cocconeis and JDory- 

 phora, and the Achnanthece, including the genera Achnanthes, Achnanthidium and Cymbosira, di- 

 stinguished by the genuflexed frustules, Achnanthes comprising stipitate, Achnanthidium free living 

 forms, and Cymbosira those, in which the frustules are connected as in Diatoma. Heiberg (Consp. 

 Grit. Diat. 1863) rejected these distinctions, as founded exclusively on the manner in which the 

 frustules occur in the living state, but retained the names Achnanthes and Achnanthidium, the 

 former for those forms which have a double structure, of costse, alternating with double rows of 

 puncta; the latter for those which have no costse. This view has not been adopted. Grunow 

 (Verb. 1860 p. 511) included in the family Achnanthece, besides Achnanthes and Achnanthidium, 

 both in the sense of Kutzing, also Cocconeis, and Rhoicosphenia, a genus formed for receiving 

 Gomphonema curvatum of older authors; but in the year 1862 (Verh. p. 116) he separated as a 

 distinct family Cocconeidece, comprising Cocconeis, Campyloneis and Rhaphoneis. He published in 

 1880 (Arct. Diat. p. 17) a very valuable synopsis of Achnanthes, retained A. longipes in the 

 same genus as A. subsessilis, but reserved the name Achnanthidium for A. flexellum, or Cocconeis 

 Thwaitesii W. Sm. 



The genus Cocconeis was formed in 1838 by Ehrenberg. The species of this genus usually 

 have a broad, elliptical outline, and live attached to algse by their lower valves, which is probably 

 the cause that the frustules are genuflexed along the longitudinal axis, thus differing from Ach- 

 nanthes, in which genus the frustules are genuflexed along the transverse axis. In course of time 

 a number of diatoms with elliptical outline were described by authors as Cocconeis, although their 

 valves were similar, and in 1867, Grunow (Novara p. 8) tried to bring order into this chaos. He 

 there defined the family Cocconeideee as follows: 



»Naviculoid diatoms, attached by their flatter side to algse, with or without gelatinous 

 envelopes; with dissimilar and bent, or similar and straight valves. Interior costate stratum of 

 the lower or of both valves absent, present in a rudimentary state, or strongly developed by 

 the vertical elevation of the ribs into marginal loculi». 



He included in this family Campyloneis, Cocconeis, Orthoneis and Mastogloia In the genus 

 Anorthoneis, formed for receiving Cocconeis excentrica, he sees a connecting link between Cocconeidece 

 and Cymbellece. In the year 1880 (Arct. Diat. p. 16) he separated Mastogloia and Orthoneis, in- 

 cluding them in a new family, Mastogloiacece. 



The true Cocconeideai and Achnanthese resemble each other in the dissimilarity of their 

 valves, the lower being naviculoid, the upper without central nodule and median line, but with 

 an axial pseudo-raphe or area. The only respect in which they differ is in the manner in which 

 the frustules are genuflexed, but the latter characteristic is not of generic importance. One may 

 easily feel inclined to unite in one group or family all diatoms with dissimilar valves, and form 

 new "genera of species which are related. Although this course appears to me to be the only 

 one, by which one may hope to bring order into the chaos of forms, I consider that such a 

 family would be far from a natural one, and would comprise widely different types. The dissimi- 



