W. M. BALE ON SOME OF THE DISCOID DIATOMS. 



that observer quotes Grunovv as his authority. He describes 

 C. subtilis as apiculate, and differentiates other species from it 

 by the absence of apiculi. Yet Grunow says expressly that 

 C. subtilis is non-apiculate. " Der Ausgangspunkt fiir alle diese 

 Formen ist der stachellose C. subtilis (Ehr. partim), Gregory, 

 Grunow " (Diat., F.-Josef Land, p. 81). This form, which is similar 

 to C. symbolophorus, but without the stellate markings at the 

 centre, also agrees well with Rattray's own account of Ehren- 

 berg's original species. It is not common, and Yan Heurck 

 figures it from guano, not finding it in European gatherings. 

 But Peragallo, like Rattray, though claiming to follow Grunow's 

 authority for the type, has figured and described a totally different 

 form an apiculate variety. 



Actinocyclus. The excessive multiplication of specific names 

 which encumbers the Coscinodisci has not been carried out to a 

 corresponding extent in the much smaller group of the Actino- 

 cycli (ignoring, of course, Ehrenberg's multitudinous pseudo- 

 species) ; still there is no doubt that an undue regard for certain 

 points of structure has led to the establishment of several species 

 on insufficient grounds. Rattray's monograph admits about 

 seventy species : Fome of these have no claim to recognition, but, 

 on the other hand, I find that about fifteen out of thirty-four 

 species or varieties which I possess cannot be identified with any 

 of Rattray's descriptions. He has adopted in this monograph 

 the plan of furnishing extremely long and minutely detailed 

 descriptions, a method which renders identification more certain 

 when one is dealing with the precise form described, but does 

 not allow for the variations which constantly present themselves 

 even in a single gathering. In fact, as I have remarked in 

 reference to Coscinocliscus, many of these are not descriptions 

 of species, but of individual diatoms. Mr. Rattray uses five 

 places of decimals to express the fraction of a millimetre which 

 corresponds to the diameter of a pseudo-nodule ! Of what 

 possible use can such measurements be when applied to structures 

 so notoriously variable ? 



Before discussing the range of variation in the genus, and 

 as I shall refer repeatedly to the commonest species A. Ehren- 

 bergii I must premise that I use that name in the sense in 

 which it is used by Ralfs himself, and by Yan Heurck, Grunow, 

 Peragallo, and, so far as I know, by all other observers except 



