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



to exhibit a dark, semi-opaque aspect. But all these species 

 usually include forms of the same size, contour and arrangement 

 of markings, but of a soft brown colour, uniform throughout 

 or nearly so, and generally with fine punct'a. Are these complete 

 valves, or secondary plates, or primary plates from which the 

 secondary ones have been detached ? Some of these brown discs 

 have the silex of the subulate areas so thickened as to appear 

 black under a low power. Valves of A. Ehrenbergii with 

 sharply defined granules and clear, distinct subulate areas mostly 

 appear blue under low powers, with the subulate spaces white. 

 Others, such as that described above, from Holler's Typen-Platte, 

 are more commonly green or purple, and show no white streaks, 

 though having large subulate areas, the substance of the valve 

 itself appearing to have a dusky tint. The bright colours of 

 these species can only be seen when dry or mounted in balsam 

 or a similar medium, while in water they are colourless. 



No other diatom known to me presents such endless variety 

 of marking as A. Barklyi, and occurring, as it does, in such 

 profusion, it is especially suitable for a study in variation. This 

 diatom is of interest as being probably the first to be named in 

 Australia, it having been described by Dr. Coates in the Trans- 

 actions of the Royal Society of A^ictoria for 1 860, under its 

 present name. Rattray incomprehensibly calls it " Actinocyclus 

 Barklyi (Ehr.) Grun.," though he knew that it was named by 

 Coates, and not by either of the authors cited. He quotes a 

 reference to it in the Q. J. M. S. for 1861 (wrongly quoted as 

 "Plate CXXXVIII." instead of "Page 138 "), but does not refer 

 to Coates' original description. It is distributed by Moller under 

 the name A. dubius Grunow. It is one of the largest of the 

 genus (perhaps the largest), specimens in my slides attaining a 

 diameter of 0*24 mm., or more than double the maximum size 

 assigned to it by Rattray. 



In normal valves the fasciculi are arranged much as in A. 

 Ralfsii, but great variety exists in the denseness or otherwise of 

 the granules, which, as in A. Ehrenbergii, also vary greatly in 

 sharpness. But it is in individual departures from the normal 

 arrangement that the tendency to variation exhibits itself in 

 such an extraordinary degree. In many cases the markings are 

 interrupted at a uniform distance from the centre, so as to form 

 a ring, and several such concentric rings may exist on one valve, 



