146 THE AMEKICAN MONTHLY [July, 



Figs. 5 and 36. — Staurastrutn cumbricum^ nov. sp. (S. magnum, 

 tertiam partem longius quam latum, semicellulis late ellipticis, a vertice 

 triangulare ; lateribus leviter convexis cum spinis velatis sed paucis ad 

 sinum, multis longioribus ad angulos.) 



Frond njther large, one-third longer than broad, semicells broadly 

 elliptical, end view triangular, with slightly convex sides, beset with 

 spines except at the constricted part, many of which are much longer 

 about the apices of the angles. 



Fig. 6. — Stau7'astruin cufnbricum. var. canibriczini^ nov. var. 

 (Var. semicellulis subangularibus et isthmo angustiori.) 



This differs from the type by having the semicells somewhat angular, 

 and the isthmus narrower. 



Fig. 7. — Staurast?'u??i osteonum nov. sp. (S. minutum, sinu latis- 

 simo et obtuso, semicellulis rotundato-ellipticis, a vertice triangulare, 

 angulis rotundatis et lateribus concavis.) 



Minute, front view shaped like a dumb-bell, end view trigonal with 

 slightly concave sides, cytoderm smooth. 



Fig. 8.^ — Staurastrufti coarctatuni Breb., var. subcurtum Nord. 



Fig. 9. — Staurastrum iotanutn Wolle. 



This tiny species may easily escape observation. Its arms are simi- 

 lar to those of St. tetracerufn Ralfs, but much finer and more delicate, 

 the end view being also triradiate. Its form in front view is not unlike 

 those of St. O' Mearii Arch., and St. pterosportim Lund., but it 

 is smaller, and has arms and not spines. 



Fig. 10. — Arthrodesnius tenuissi?mis Arch. 



This being a rare species a figure is given here. 



Fig. II. — Staurastru?n furcatum Ehrb. A faintly punctate variety 

 of this species is shown in the figure. The lateral bifid processes are 

 almost reduced to two spines, the other " bifid processes " being sharply 

 bispinate. 



Fig. 13. — As the figures of Staurastriim anatin?irnQooke Sl Wills, 

 as published by Cooke differ so greatly, a figure of the form which was 

 abundantly seen at Capel Curig, is given. Cooke's figure in Grevillea 

 is nearer the form herein figured than the one in his Brkish Desmids. 

 The extremities of the processes often appear as in Fig. 12^, caused 

 by the spines being placed somewhat vertically over each other. 



Fig. 13. — Alicrasterias ajnericana Ehrb., var. Lewlslana nov. var. 

 (Var. lobis polaribus subintegris et latioribus, incisuris angustioribus 

 infra lobos polares.) ^ 



This has the end lobes subentire and wider, the infra-apical incisions 

 are narrower also than in the var. recta Wolle, which it approaches. 

 This is a very distinct variety, and very distant from '' forma major 

 Reinsch," with which it is worth comparing, as this latter seems to be 

 at the other extreme. 



Fig. 14. — Staurastrutn tnuricatum Breb., var. acutum nov. var. 

 (Var. spinis brevibus (nee granulis), semicellulis truncato-pyramid- 

 atis.) 



This differs from the type in the acute though short spines in place 

 of " conic granules," as well as in the trapezoid or truncately pyra- 

 midal semicells. 



Fig. 15. — Staurastrutn ophluraY^und..^ i'ornvd nonaradiata (Forma 

 a vertice nonaradiata.) End view with nine rays. 



