292 CONTKl BUTTONS FROM THE NATIONAL IIERBAiilUM. 



Trigonium adspersum Mann, sp. now Plate LI, future 5. 



Valves triuniruTar; Hliglitly and evenly convex; sides inllated; angles blunt, not prn- 

 duced, each one Laving an evident but very minute spine; markings large, compound, 

 oval or rounded Ix-ads, loos(^Iy but nidially placed at the middle of the valvc^, leaving 

 an indefiniti' coutral an^u, thence outward closely set and radially arranged, renuun- 

 ing tlie same size until ^vithin a short distance of the border, then rapidly diminishing 

 to one-fourth size. 



• Width l)ct\\een apices, O.0G5 nun. 



The gen(^ral markijigs correspond to T. rusticwm Mann, whit-li also lias very minute 

 spines at the arigles. The two are ratluu* wi(h^ for one species. C(mipare this (U^sfTip- 

 tion with that of I Ik; latter Ix^low. 



Type in the U. K. Xational Museum, No. 590157, from station :5(i!)01I, Oldujtsk Sea, 

 September 8, 1890; 1,584 fathoms, l)ottom of greentJnud and fine sand, 



Trigonium cinnaniomeuni (Grev.) Ahuin. 



Triceratium dvvnmomeum Grev. Quart. Journ. Micr. Sci. n. s. 3: 232. pi. 9. f. U. 

 1863. Moeb. Diat.-taf. pi. 47. f. 1!. 1890.- Schmidt, AtUus pi. loL f. 23-24. hSOO. 

 DeToni, Sylh Alg, 2: 974. 1894. 



CeHtodiscus vititianionicuin Gvnn.; Van. Hour. Synop. pi. ljr,.f. 1-2. 18SL 

 rseudotrlceratiuttt cinuainonenm Grim. Dnd^schr. Akad. Wicu 48-: 83. 1884. 



However this (Hatom may contrast with others by r<>ason of tlie nunute spim-sthat 

 terminat(^ its apic^es, thehyaUne lim^s It^uling up to these and thus Itisecting the angles, 

 and th(^ gc^neral !)Ut not constant i)resence of still smallcT spines along the margin. I do 

 not recognize any reason for i)hu-irvj: thi^s in Cestodiscus, That genus as defuHHi l>y 

 Grevilh*« is made to<U ffer from Aulacodiscus l)y the absence of ^'furrows" or radiating 

 lines nuuiing from the center to the submarginal processes, not in any marked dil'fcT- 

 ence of tht^ processes. To confuse such i)n>cesses witli the minute spines of this diatom, 

 to which ''furrows" always do run, (*v(^n to those occasionally present on the margin, 

 is hardly allowable^; and when we add tlie further (character of Cestodiscus added Ijy 

 Castra<'ano,?> a submarginal l>and with minuttT and (doser granulation, any resem- 

 l)lance to tbat doubtful genus vanislies. Schmidt and others reject this classification; 

 and Grunow subsequently proposes a new genus, Pseudotriceratium, to accommodate 

 this speei(>s. This would l)e Avananted, in view of the ust^lcssness ot Triceratium, 

 did not eleven's older g(^nus Trigonium serve the purpose as well. 



Found at stations 2807, 2920II, Galaj.agos and Hawaiian Islands, 



Trigonium coscinoides (Gr. & St.) Mann. 



Triceratium vosdnoidcs Gv. & St. Journ. Quek. Micr. Club II. 2: :i27. pi Vj.f. 13^ 

 18SG. Do Toni, Syll. Alg. 2: 922. 1894. 



Tricerallum nuUoao-relirulaUnn Grnn.; Schmidt, Atlas ;>?. 751. /. -^5-^6* (corrected 

 to 7\ coscinoides in Kerichtigungen opposite p?. 153). 1890. Van Ileur. Synop. pL 

 112, f. 5. 1881.(?). 



I have b(Hm able by the discovery of lunuerous specimens of this diatom to determine 

 the unity of the above. The si)eciniens agree with the Oamaru figure except in having 

 a very minute spine at one or all of th(^ apices, thus agreeing wiili th<' form described by 

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

 showing on the same valve the Grnnow and the Grove and Sturt forms. I add a 

 question-nuirk after the citation to Van Ilcurck, because no specimens I have exam- 

 ined show any trace of the secondary marking figured and nn^ntioned by him, nor any 



« Trans. Micr. Soc. Lend. n. s. 13: IS. 1S(J5. 

 ^Castr. Hep. Voy. Chall. J5ot, 2: 122. 188G. 



