292 CONTRIBUTIONS FROM nil NATIONAL HERBARIUM. 



aiuXD ndspersum M:i!in. -p. nov. I'i \n II. in. 



\\il\ ee triangular; slightly and evenly convex; Bides inflated; angles blunt, not pro- 

 duced, each one having an evident I 'ut \ cry minute spine; markings large, compound, 

 oval or rounded beads, loosely 1 >u » radially pHced at the middle of the valve, leaving 

 an indefinite central area, thence outward cl leelj se1 and radially arranged, n 

 ingth< ■ until within a ahoxi distance of the border, then rapidly diminishing 



to one-fourth - 



\\ idth bel ween api< :es, 0.066 mm. 



The general markings correspond to T. rutUeum Mann, which also has very minute 

 Bpines al the angles. The two are rather wide for one species. Compare this descrip- 

 tion with thai of the latter below. 



T> pe in the i . 8. National Museum, No 590157, from station 3699H, < tkhotsl 

 September 3, I s '.n;; L,584 fathoms, bottom of green mud an<l fine sand. 



Trigonium cinniunomeum Grev.) Mann. 



/' ict atium cinnamomeum Grev. Quart. Journ. Micr. Sci. n. >. 3: '2.V2. pi. 9. f. It. 

 L863. Moeb. Diat.-taf. pi. ;;./. It. L890. Schmidt, Alias pi. 151. I L890. 



D< Toni, Syll. Alg. 2: 974. 1894. 



discus cinnarnorneurnGTim.] Van. Beur. Synop. pi. t£6.f. i .. 1881. 

 P*< ium cinnamemeum Grun. Denkschr. Akad. Wien 48-: 83. i ss i. 



However thia diatom may contrasl with others by reason of the minute Bpines thai 

 terminate its apices, the hyaline lines leading up to these and thus bisecting the angles, 

 and the general I ul uol const am presence of still smaller spines along the margin, [do 

 nol recognize any reason for placing this in Cestodiscus. Thai genus as defined by 

 Greville is made to di ffer from Aulacodiscus by the absence of " furrows" or radial mu r 

 lines running from the center to the Bubmarginal processes, uol in any marked differ- 

 ence of the processes. To confuse such processes with the minute spines of this diatom, 

 to which "furrows" always do run. even to those occasionally present on the margin, 

 La hardly allowable; and when we add the further character of Cestodiscus added by 

 Castracane,& a submarginal hand with minuter and closer granulation, any resem- 

 blance to that doubtful genus vanishes. Schmidt and others rejeel this classification; 

 and Grunow subsequently proposes a new genus, Pseudotriceratium, i<> accommodate 

 this Bpecies. This would be warranted, in view of the uselessnese ol Triceratium, 

 did no1 Cleve's older genus Trigonium Berve the purpose as well. 



Found at stations 2807, 2920H, Galapagos and Hawaiian Islands. 



Trigonium coscinoides (Gr. & St.) Mann. 



7i coscinoides Gr. & St. Journ. Quek. Bficr. Club II. 2: 327. pi. t9.f. tS. 

 L886. De Toni, Syll. Alg. 2: 922. L894. 



i atium radioso-reticulatum Grun.; Schmidt. Atlas pi. 151. f. 35 .-'.(corrected 

 to T. coscinoides in Berichtigungen opposite pi. t5S). L890. Van Beur.Synop.pl. 

 11%. f. 5. L88] 



[ have been able by the discovery of numerous specimen- of this diatom to determine 

 the unity of the above. The specimens agree with the < tamaru figure excepl in having 

 a very minute spine al one or all of the apices, thus agreeing with the form described by 

 Grunow. As a rule, the tiny spine [g present on only one or two of the apices, thus 

 showing on the same valve the Grunow and the Grove and Sunt forms. I add a 

 question-mark after the citation to Van Heurck, because no specimens I have exam- 

 ined -how any trace of the secondary marking figured and mentioned by him, nor any 



"Tran- Micr. Soc. Lond. n. s. 13: 18 1865. 

 i. Rep. Voy. Chail. Bot. 2: L22. 1886. 



