ZOOLOGY AND BOTANY, MICROSCOPY, ETC. 211 



Sphacelariaceae.* — In a lengthy paper on this family of Phseospome, 

 M. C. Sauvageau makes the following remarks on their general structure. 

 He agrees with Magnus in regarding the most common mode of branch- 

 ing as sympodial rather than monopodial as stated by Pringsheim. A 

 character which distinguishes the Sphacelariaceae from the most nearly 

 allied order, the Ectocarpaceae, is the black coloration of the thallus by 

 eau de Javelle, due to the presence of a substance, the nature of which 

 is at present unknown, and apparently not identical in all the species. 

 The chromatophores always have the form of discs or grains, never of 

 plates or bands. Tannin is almost invariably present, either generally 

 diffused or limited to special cells. The hairs are of endogenous origin, 

 differing in this respect from those of the Ectocarpaceae, and agreeing 

 with the Myrionemaceae. 



M. Sauvageau has no doubt about the existence of sexuality in 

 the Sphacelariaceae ; there are (at least in Sphacelaria hystrix), two 

 kinds of plurilocular organ, one of which certainly consists of antheridia, 

 the other probably of oogonia. The act of impregnation has not, how- 

 ever, been actually observed. The classification of the species into 

 autonomous and parasitic is not a natural one ; the parasitism, where it 

 occurs, is not obligatory. 



The greater part of the paper is occupied by a special account of 

 the following species : — Battersia mirabilis, Sphacella subtilissima, Spha- 

 celaria pulvinata, S. bracteata (Sauv. MS.), S. pygmsea, S. fsecunda 

 (Sauv. MS.), S. Borneti, S. sympodicarpa (Sauv. MS.), S. chorizocarpa 

 (Sauv. MS.), S. Beinkei (Sauv. MS.), S. spuria (Sauv. MS.), S. radi- 

 cans, S. britannica (Sauv. MS.), S. olivacea, S. plumula, S. plumigera, 

 S. racemosa, Chsetopteris plumosa. 



Classification of Diatoms.] - — C. Mereschkowsky proposes to abolish 

 the accepted classification of Diatoms into Raphideae, Pseudoraphideae, 

 and Anaraphideae, and to substitute one primarily into two classes, 

 Mobiles and Immobiles, derived from the presence or absence of move- 

 ment, this power being dependent on the presence or absence of a slit 

 in the wall of the frustule. Whether this slit be placfd at the centre 

 of the valve or near one of its margins is of secondary importance. 

 The Mobiles are again divided into two families, — the Raphideae, char- 

 acterised by the presence of nodules and the usually central position 

 of the raphe (arranged in three groups, the Heteroideae, Naviculoideae, 

 and Tropidoideae), and the CarinataB (Surirelloideae and Nitzschioideae), 

 characterised by the absence of nodules (except Bhopalodia) and the 

 lateral position of the slit (except some species of Nitzschia). The 

 Immobiles again consist of two families, — the Bacilloideae (Pseudo- 

 raphideae, except Carinatae, marked by the elongated or bacillar form of 

 the valves, the usual presence of a pseudoraphe, and the usual absence 

 of appendages or spines (again divided into the FragilarioideaB and 

 Tabillarioideae, according to the presence or absence of internal parti- 

 tions ; and the Anaraphideae (Biddulphioideee and Discoideae). In the 

 Anaraphideae the two valves of the same frustule do not generally corre- 

 spond with respect to the position of their pattern or of their processes. 



* Journ. de Bot. (Morot), xiv. (1900) pp. 213-34, 247-59, 304-22; xv. (1901) 

 pp. 22-36, 50-62, 94-116, 137-49 (24 figs.). 



t Ann. Nat. Hist., ix. (1902) pp. 65-8 (4 fi^s.). 



p 2 



