OF THE COSCINODISCE^. 



827 



but, from oui' insufficient acquaintance with them, we are unable to decide on 

 their proper position, and have not attempted their removal. Are they akin 

 to Goniothecium ? 



Xanthiopyxis glohosa (E.). — Frus- 

 tiiles subglobose, hispid, with short setae. 

 = Pyxidicula glohosa, KS A. p. 23. Fossil. 

 Bermuda. Diam. 1-552". 



X. ohlonga (E.). — Frustiiles oblong, 

 equally and broadly rounded at each end, 

 densely hispid, with short setae, which 

 are sometimes joined by a membrane. 

 EM. pi. 33. 17. f. 17. = Pyxidicula(K.). 

 Fossil. Virginia. L. 1-552". (v. 76.) 



X. constricta (E.). — Frustules oblong, 

 constricted at the middle, and broadly 

 rounded at each end, hispid, with short 

 setae, which are often joined by mem- 

 brane. =PJ/a^iV7^c?<^rt constricta, }L^ A., p. 23. 

 Fossil. Bei-muda. L. 1-384". _ Ditfers 

 from X. ohlonga by its constriction. 



X. hirsuta (E.). — Frustules bivalved, 

 subglobose, not cellulose, rough with 

 simple and obsoletely forked hairs. = 

 Pyxidicula ? hirsuta, ERBA. 1845. Fos- 

 sil. Maryland. Diam. 1-115". Habit 



of Xanthidium, but silicious. 



X. urceolaris (E.). — Valve urceolate, 

 with the summit bristly ; margin revo- 

 lute. EM. pi. 33. 16. f. 14. Fossil. Vir- 

 ginia. Diam. 1-1560". " I [EhrenbergJ 

 have only met with single valves. In 

 form they resemble Stephanogonia, but 

 are not angular." 



X. alata (E.). — Frustules smooth, ob- 

 long, each end equally and broadly 

 rounded ; margin of the valves bordered 

 by a lacerated or deeply dentate, not 

 setose membrane. = Pyxidicida alata, 

 KSA.p.23. Fossil. Bemmda. D. 1-552"- 



Douhtfid Diatom. 

 X. aculeata (Ft.) = Pyxidicula acideata. 

 The figure in Microg. pi. 18. f, 124 shows 

 a globular spinous body, resembling 

 some sporangia of the Desmidieae. Eh- 

 renberg himself regards this as a very 

 doubtful Diatom. 



Genus INSILELLA (Ehr.). — Frustules simple, equally bivalved, cylindrical 

 (fusiform), with a turgid ring interposed in the middle between the valves. 

 Marine. Eesembles a cylindrical Biddulphia. 



Insilella Africana (E.). — Frustules 

 fusiform, smooth, foiu" times constricted ; 

 the middle joint largest, subglobose ; the 



others decreasing at each end, oblong; 

 each apex acuminated. KSA. p. 32. 

 Mouth of the river Zambese, Africa. 



FAMILY VIII.— COSCINODISCE^. 



Frustules disciform, mostly simple ; lateral valves or discs flat or convex, 

 cellulose, areolate or granulate, without processes, but sometimes fui-nished 

 wdth spines or teeth ; connecting zone ring-like and generally smooth. The 

 Coscinodisceas are closely allied to the Melosirese, — a fact noticed by Kiitzing 

 himself, although in his arrangement the families are widely separated. The 

 distinction between them is by no means satisfactory ; according to Kiitzing, 

 it consists in the cellulose or areolated structure of the Coscinodisceae. But 

 whilst on the one hand we find in the Melosirese some species of Pyxidicula 

 and Stephanopyxis with cellulose valves, on the other hand, in this family 

 some species are merely granulate or punctate. Practically, however, the 

 proper situation of the species can generally be determined without much 

 difficulty. In the Coscinodisceae the fi-ustules never form filaments, the con- 

 necting zone is always present, narrow and ring-like, and the lateral valves 

 are never so convex as to be hemispherical or campanulate ; so that the disk 

 is almost always in the field, it being difficult to obtain a good front view. 

 Most of the forms included in this family are marine, and many are remark- 

 able for their exceeding beauty. 



Genus COSCINODISCUS (Ehr.).— Fmstules simple, discoid; disc ceUular 

 or dotted, without processes, defined border, internal septa, or division into 

 radiating compartments. '' The only essential character that distinguishes 



