26S MESSES. W. AVEST AND G. S. WEST ON 



159. Staurastrum Cerastes, Lund. Desm. Siiec. p. 69, t. 4. tig. 6. (PI. XVIII. fi^^ 4.) 



luong. 50-53 n ; lat. c. proc. 57-5-61-5 /n ; lat. istlmi. 10-12-5 u. 



All the American specimens of this species that we have seen agree exactly with the 

 original Swedish and also with British specimens. TJie tigure given hy Wolle (Desm. 

 U.S. pi. 43. tigs. 6, 7) — which he states to be " a smoother specimen " — certainly does 

 not represent this species. In all the hundreds of examples of >S'. Cerastes that we have 

 examined, we have never found any but the slightest variations, the general characters, 

 such as the arrangement of the emarginate verrucse, the smoother under side of the 

 processes, and the elegant curvature of the semicells, remaining quite constant, the 

 " margins and areas " not being " variously roughened." We give a figure of a typical 

 American specimen. 



160. S. Zyg.exa, nov. sp. (PI. XVIII. fig. 5.) *S'. parvum, tam longum quam latum (cum 

 processibus) ; semicelluke longitudinaliter rectangulares, lateribus leviter concavis 

 glabris, angulis inferioribus leviter rotundatis, apicibus convexis, angulis superioribus 

 in processus validos subincurvatos productis, processibus et apicibus dentieulatis, 

 apicibus processuum subcapitatis cum annulo spinarum parvarum ad extremum, 

 annulis duobus granulorum minutorum ad basin semicellularum ; a vertice visse 

 triangulares, lateribus dentieulatis valde concavis, angulis in processus breves 

 subcapitatos denticulatos j)roductis. 



Long. 84 ju; lat. c. proc. 35 ju ; lat. isthm. 7 /u. 



161. S. RoTULA, Nordst. in Vidensk. MeddeL 1869 (1870), p. 227, t. 4. fig. 38. 

 (PL XVIII. fig. 11.) 



Long. c. papill. 46*5 n ; lat. c. proc. 58-80 yu. 



The Xorth American forms vary considerably in size ; the processes are somewhat 

 convergent, and the back of the semicells is not so high as in the Brazilian form {cfr. 

 Xordst., Preshw. Alg. of X. Zeal. & Austr. p. 37). I)r. Xordstedt also states that, *'to 

 judge by a drawing of living specimens by Mr. C. Lofgren tlie crenation in the margin 

 should have been sharper (the minute spines would probably have been broken off)." 

 In the specimens we examined the processes were generally trinodulose, but some 

 examples were seen in which the processes had two rings of minute teeth pointing forward. 

 Most of the specimens were 8-radiate in vertical view, but one example seen was 10-radiate, 

 and had 10 apical papillae, one above the base of each process. Wolle's fig. 14, pi. 44 

 (Desm. U.S.), does not represent the front view; it is somewhat tilted, and the apical 

 papillse are not rectangular as he figures them, but obtusely conical : the apices of the 

 processes are not bifid as figured by him; this remark also applies to several other 

 figures of his, such as S. Ophiura. 



162. S. Ophiura, Lund. Desm. Suec. p. 69, t. 4. fig. 7. 



U.S. specimens of this are very variable both in dimensions and number of rays, the 

 latter varying from 4 to 8. The variety tetracerum, WoUe (long. 60 a*, lat. c. proc. 

 102-132 A*, lat. isthm. 11-5 p), as we have observed it, has longer and more slender 

 processes than figured by Wolle. (PI, XVIII. fig. 16.) 



