BY G. 1. PLAVFAIE. 537 



largest, the successive terms of which differ from one another by 

 no more than 1 /a in diameter. It is appaUiug to have to believe 

 that each size is a separate evolutionary variation of the type, 

 absolutely unconnected in growth and development with even the 

 size next above it. 



Var. MAJOR, n.var. (PI. Ivi., f.25). 

 Diam. 32 fx. Undulate in side-view. The specimen is figured 

 with the chromoplasts in situ. 



Var. STELLiGERA Cleve et Grun., forma. (PI. Ivi., f.26). 



Frustula in medio areolis cuneatis radiantibus 6-10 ornatse. 



Diam. as in the type. Cf. P. T. Cleve, New and little 

 known Diatoms, PL v., f.63c. 



There is an error in the description of plates ; it should 

 read:— "c. from France; b. var. s^eZ/«/«/e/« Grun.," instead 

 of vice versa, see p. 22. Note also Schroter, Schwebeflora 

 unserer Seen, f.54, but the form is undoubtedly a variation 

 of C i/cl. Meneghiniana. 



Var. KuETZiNGiANA(Thw.). (PI. Ivi., f. 27). 



Diam. 12; crass. 8 /*. 



Cf. Thwaites, Ann. Nat. Hist., Ser.2, 1848, Pl.xi., f.Dl, 

 2. This form was found in quantity in a duckpond in the 

 Botanic Gardens, fed from the Sydney Water-Supply. It is var. 

 iiiniatissimu with the marginal striag beginning to form. They 

 begin to form as mere points round the inner edge, and 

 gradually extend inwards. In f.27, the striae have been 

 wrongly drawn ; they should be short wedge-shaped dashes. 

 Thwaites, fig. 5, loc. cit., is my var. major, but diam. 40 /x. 



Schroeter's form. (PI Ivi., f.28). 



Occasionally two or four Cyclotella frustules might be seen 

 embedded in a cylinder of mucus, with pores radiating out 

 on all sides. Cf. Cycl. comta var. quadrijuncta Schroter, I.e., 

 f.58. Such, however, are merely caused by repeated mitosis 

 under conditions of partial dessication. Schroter 's form has 



