280 PROCEEDINGS OF SECTION D. 



tion of Pen. Mooreanum, Arch., and also of Pen. poly morp hum, 

 Perty ; Cos. minimum var. stibrotundatum, W. & G. S. West, 7 x 6u, 

 the lowest form of Cos. ellipsoideuni, Elf v. (and of others) ; and Cos. 

 concinnnm v. pyginaiini (Hantzsch), Rein., 8 x 8z/, a minute poly- 

 morphic form of Cos. anisochondrum, Nord. Very minute forms 

 of Cyclotella Meneghiniana (greatest observed diameter 32m) are 

 found in the plankton of the Sydney Water Supply, measuring only 

 4m across. It is evident, of course, that Desmids, on account of 

 their bilateral symmetry, cannot be expected to descend to as small 

 a size as others of the lower algae. Most of my recent observations 

 have been upon forms of Cosmarium, but in one species of Closte- 

 riiim, viz., C. decorum, Breb, whose variations are common here, 

 I have been able to note the strength of this degenerating influence. 

 In this genus and others, where the length of the cell is the promi- 

 nent feature, it might be supposed that the length would suffer 

 rather than the breadth, but this is not the case when all the varia- 

 tions have been passed in review. The length of this species falls 

 from 756 z^ to 198//, and the breadth from 48m to 13// — one-quarter 

 of the maximum in each case. These figures also give a very 

 inadequate idea of the immense difference in appearance — the 

 optical area being reduced to the sixteenth part. C . gracile also, 

 whose greatest length is about 450//, I have found as low down as 

 56//. 



Form. — Quickly repeated division of the cell always tends to 

 reduce the outline eventually to the very simplest shape — circle, 

 oblong, ellipse — and this is true even of the very long cells of 

 Docidium and Closterium. Even the most salient angle gets 

 rounded off. Compare Delponte, Desm. suhalp., T. xviii, f. 46, 

 where a number of degraded cells of Doc. trahecula are shown in- 

 volved in mucus, which has been thrown out to protect them in 

 time of drought. Under favourable conditions these forms will 

 develope into perfect specimens of the type. 



Ornament. — With regard to ornamentation, those adornments 

 which are most prominent are generally the first to go, as they are 

 the last to come. Arthrodesmiis and Xanthidium lose their spines at 

 once and become to all appearance forms of Cosmarium, with the 

 result that at present there are several young forms connected with 

 these genera masquerading as Cosmaria. Indeed, there are very 

 good reasons for believing that the lower forms of all species of 

 Xanthidium and Arthrodesmus are well known " species " of 

 Cosmarium, just as the genus Arthrodesmus itself is merely a collec- 

 tion of certain lower forms of Xanthidium. 



Strige, puncta, scrobiculae, crenations, granules, verrucae, all 

 gradually disappear in degenerate forms, generally also in the 

 reverse order to that in which they make their appearance. 

 Neither outline, size nor markings are, per se, indications of specific 

 identity or even of generic rank. Cf. Ar. glaucescens, Wittr., Om 

 Gotland's Sotv. Alg., T. iv, f. 11, which, however much it may look 

 like an Arthrodesmus, has been shown to be a Theapedia. Of course 



