310 \ i i:ir.r i i<»\> PROM rm: NATIONAL HERBARIUM. 



H • ouated pi • ticulation, 



and a strikingly distinct b Raits, De Toni, and Boyer unite it with the above, 



hini.lt ;u>«l others consider them distinct I agree with this latter view. I also 

 bich Schmidt • looks upon :i- synonymous with / 

 Bail The resemblance Is Dot worth considering. Grunow's claim that 

 Bail, is onlj a triangular form of Biddulphia granuioi R per is equally 

 untenable. Th< ,.iy different, and the whole build of /»' grtxnulata l 



iculation, is much finer and more delicate than in /'. tpinotwn. 

 he name Biddulphia spinosa has been applied by Grevilled i" a quite different 

 diatom and as I consider Grun gnment of this spei • Denti- 



cells ' i" be m;i« 1« • on inadequate grounds, the original specific name of Bailey is pre- 

 empted, and the choice lies between Bailey't and Roper's T. armcQum. 



Both were published in 186 1 ; but as Bailey's article appeared in February and Roper's 

 somewhere near the close of the year, I have selected the Dame assigned by Bailey. 

 My form is a large and elegant variety of this variable Bpeciee h shows a distinct 

 inner triangular area -> mmetrica] with the outer triangle, its reticulation showing do 

 radiation; but outside of this triangle the reticulation is radial, running vertically to 

 . bile a fe^ rowB of the Del work form broad lines running from each apex of 

 the inner triangle to the ; ch horn-like process in the angles of the outer tri- 



angle. No BUture exists between these various portions of the valve, but the pattern 

 is made very evident by the arrangement of the reticulation. 



Pound at station 2807, Galapagos Islands. 

 Biddulphia shadboltiana Grev. Mann. 



n? gibbosum Barv. & Bail. Proc. Acad. Phila. 6: 181. pi 9. J 

 7: 130. L854. Schmidt, Atlas pi. 80. f. U 15, 17, U. 1882. 



n orbiculatum Shadb. err. det. Bright. Quart. Journ. Micr. Sci. 4: 276. pi. 

 17. f. :>>. L856. 



m 8hadboltianum Grev. Trans. Micr. Soc. Lond. n. b. 10: 28. 1862. Van 

 Beur. Synop. pi 108. f. 5 7. L881. Schmidt, Atlas pi 80. f. 18 10 L882. De 

 Toni, Syll. Alg. 2: 964. L894. 



/// tlongatum Grun. in Schmidt, Atlas pi 80. f. It. L8f 

 Lampriscus bittoni Schmidt, Atlas pi 80. f. 11. II 



I think there is sufficient ground for rejecting Trioeratium orbiculatum Shadb. in 

 connection with this species. Greville explains with car.- f that Brightwell confused 

 another form with the original '/'. orbiculatum of Shadbolt. This is borne out by 

 Shadbolt's description and figure, the figure being reproduced byMoebius.* Shad- 

 bolt mentions do spines, and they are omitted from the figure by Tuffin West, which 

 clearly argues they were uotthere. Ii is true, as Boyer i has pointed out, that Bright- 

 well's species may be with or without -pine-, a statement that I can confirm. But 

 though Brightwell's Bpecies may resemble Shadbolt's in Dot having spines, that i\<>c< 

 uot mean that Shadbolt's species resembles Brightwell's further than in this aegative 

 quality. Greville, in conferring the name shadboltianum on the misnamed species 

 of Brightwell, clearly emphasizes their differences and RalfsJ repeats this distinction. 



" Quart. Journ. Micr. Sci. 1: 2 is. pi .;./'. .;. ; 

 6 Ehrenb. Mikrog. pi 19. f. 18. L854. 

 - Schmidt, A Ma- pi 87. f. 18. L885. 



Micr. Soc. Lond. n. -. 13: 6. pi l.J. 5. 1865. 

 i Denkschr. Akad. Wien 48 J : 58. 1884. 

 (Trans. Micr. Soc. Lond. d. .-. 10: 28. L£ 



as. Mi.r. Soc. Lond. n. s. 2: 1 1. pi t.f. 6. L854. 

 b. Diet.-taf. pi ../. 6. 1890. 

 Acad. Phila. 1900: 710. L901. 

 [nfus. ed. 4. 853. 1861. 



