BY G. I. PLAYFAIR. ^39 



should be just a little more curved. It was quite evident, 

 from the appearance of this frustule, that each pole had 

 twisted round through an angle of 90°, so as to bring the 

 spines into the position shown in figs. 30, 31. The emarginate 

 portions now grow outwards on either side, and two new 

 spines complete the double frustule. Compare figs. 34, 35. 

 The two frustules then break apart down the central septum. 

 It is unfortunate that I was unable to find a perfect redupli- 

 cated form, var. Zacharmst , in the Sydney Plankton ; figs. 

 36-38 reproduce three forms from Brisbane. Cf. Schroter, 

 Schwebeflora, figs. 31, and 316. 



This series affords indisputable proof of the growth of the 

 siliceous membrane of diatoms. 



Var. GiiACiLis li. L. Smith. (PI. Ivi., fig. 39). 



Frustula angusta, apicibus conicis incrassatis, anuulis im 

 bricatis aut parallelis aut nullis. 



Long. corp. 75-83; lat. 4-7; long. sp. •J4/x. 



Rare among quantities of the other. It is probably a form 

 produced by strong longitudinal growth, which opens out the 

 annuli and causes them to assume an imbricate appearance. 

 Specimens were noted without annuli, and, in the form figured, 

 they were imbricate at one end and parallel at the other. Cf. 

 Apstein, Siisswasserplankton, p. 143, f.37; and Peragallo, Monog. 

 llhizosolenia, PI. iv., f.l7. 



PEKIDINIE.E. 



Genus C E K A T 1 u M Schrank. 



Ceratium HiRUNDiNELLA O. F. Mliller. 



This species was abundant in all the samples throughout 

 the year. In every case, the specimens were large and well- 

 grown, with strongly reticulate membrane. No examj)les 

 wei-e found of the common European variation figured by 

 Lemmermann in Plankt. Schwed. Gewass., T.2; and by W. 

 &G. S.West in Frw. Plankt. Scottish Lochs, p. 494, and in 

 Plankt. Irish Lakes, p. 94. 



