_' s '' iNTBIBUTIONS PROM I ill. NATIONAL HERBARIUM. 



in interna] secondai atour, \\ iih 



both valves alike or different, in Bonn instances hyaline and in others variously 

 sculptured, without : or various]} ornamented with spines and othei 



internal cases ha> ae thicker siliceous walls than the external valves 



;m<l girdles which inclose them, and consequ< ntlj are frequently met with in -« a dredg- 



i t he external walls thai ga> e 

 form to the complete filaments have been 1 b1 Vj these various parts have beensuc- 

 • ly found they have received ic and specific names, and we there 



fore lia \ <• to deal with a tangle i f nomenclature unusual even among the Diatomi 

 The generic mum- firsl to be constituted and Bubsequentl) applied t<> these forms is 

 Actinia us. Be< auseof its priority Grunow gives it to the form known elsewh 



Laud. This name, however, was nol originally applied to a 

 *1 i ;i 1 1 .in. bul to a stellate Bilicious Bpicule, which Ehrenberg called Dictyocha Actxnu- 

 and was not used for the diatoms until L855< 1. quina iut Ehrenb.d being 

 indeterminate, but probably also a Bpicule . It is therefore not available for the 

 Diatomaceae, but musl give place to Chaetoceros, I s l">. In this connection I wish i<> 

 call attention to a quotation from Van Heurck's Synopsis given below under ( Tun I 



In the above list of synonyms "in part" we find Dicladis and Goniothe- 

 cium, bestowed like Cha< ad preceding it in thai volume. I am, however, 



opposed to accepting any of these names for the following reasons: I They represent 

 what are nol necessarily structural parte of the plants, Castracane,' holding them to be 

 Bporangial encasements; 2 we are not even able to affirm thai these spicules are con- 

 fined to whal we known- the Chaetocerae; (3) both the above generic names were 

 based on fossil fragments from which we can build do conception of the forms to which 

 they belong, and in using them we should, therefore, be constituting a genus without 

 any fixed idea of its Btructure. Dicladia was found in African guano and Goniothe- 

 ciuin in diatomaceous earth from Richmond, Virginia. ' 



I can nol follow Castracane in maintaining the separate i baetoceroe and 



Bacteriastrum, as his grounds of distinction o appear to be trivial. They are I that 

 the valves of Bacteriastrum are round and of < Ihaetoceros generally oval; 2 that the 

 former have more awn- or setae than the latter; (3) thai these are in Bacteriastrum 

 arranged around the valve, bu1 in Chaetoceros on opposite sides; I that these awns 

 in Chi interlace. Anyone who will examine a gathering of Bacteriastrum 



, arians Laud, will Bee thai these characters are in di fferenl parts of the same filament 

 too inconstant to form a generic concept. Thus, in II. I.. Smith"- type no. 57, both 

 round and oval valv< s are Been in the same filament and bearing precisely the same 

 kind of setae. Again, the terminal valves often have two Betae, while those below 

 them may have more; and in the case of Lauder's figures of this BpeciesA we fmd 

 terminal valves with eighl Betae and the intermediate ones with many more. As to 

 the interlacing of the setae, it is by do mean- constant, many specimens examined 

 showing no Buch tendency. In fact, that condition is hardly to be discovered in 

 < lastracane's on d fig i 



Van Hem-. Synop. pi. 82bi8.f. 10. L881. 

 b Phys. Abh. Akad. Wiss. Berl. 1839: L49 L50. 1841. Cf. Mikrog. pi. 18. f. 59 SO. 



Ber. Akad. Wiss. Berl. 1854: 237. 1855. 

 d Ber. Akad. Wiss. Berl. 1844: 76. L845. cf. also Griff. & Henf. Micr. Diet. ed. 3. pi. 



43. f. i '■■ - 



' < astr. Hep. Voy. ('hall. Bot. 2: 81. I 



rCf. article by Brightwell in Quart. Journ. Micr. Sci. 4: 106. 1856. 



i r. op. cil . v -'. 

 /'Tran-. Micr. Soc. Lond. d. -. 12: pi. 8. L864. 

 • istr. op. cit. pi. 19. J 



