OF THE KAYICTJLE^. 92* 



rounded apices. SD. ii. p. 70, pi. 56. 

 f. 353. Dorset. Strict 28 in -001". In 

 the character of its frond this species 

 somewhat agrees with Rhaphidogioea j 

 but the frustules are arranged in series, 

 not in fascicles, as in that genus. 



Doubtful Species, 



C. ? ampJiioxys (E., K.). — Known only 

 from fragments. Naviculae parallelo- 



gramic, smooth; valves acutely lanceo- 

 late. KSA. p. 105. = Naunema am- 

 phioxys, EA. pi. 3. 2. f. 5. Mexico. 

 (xii. 55-57.) 



C. ? Americaman (E., K.). — Naviculae 

 striated, large, linear, with subacute 

 apices, densely arranged wdthin branched 

 tubes. IvA. p. 105. = Naunema Ameri- 

 ca mon, EB. 1845, p. 79. River Hudson. 

 Strife 18 in 1-1200". 



Genus SCHIZONEMA (Ag,, Kiitz.) (Monema, Grev. ; Monnema, Meneg. ; 

 Naunema, Eh7\), — Erustiiles naviculoid, arranged confusedly or in a single 

 file, within a capiUary, submembranaceous, single-tubed, more or less branched 

 frond, of nearly equal diameter throughout. This genus, constituted by 

 Agardh, has been repeatedly divided and reunited, and the generic names 

 altered and transposed in an arbitrary manner without regard to priority. 

 Dr. GreviUe founded Monema for the species with single tubes, retaining 

 those with compound fronds in Schizonema. This division seems judicious, 

 and indeed has been adopted by nearly every succeeding writer, although 

 Greville's names have been disused or differently applied. Agardh recognized 

 the distinctions, but retained Schizonema for the species with a frond of 

 simple structiuT, and founded Micromega for the species having a compound 

 structui-e. As this arrangement has been foUow^ed by Kiitzing, and acquiesced 

 in by Greville, we use it here. There is the greatest difference, how^ever, in 

 the distribution of the species, even amongst those who admit both genera. 

 " This discordance of opinion," observes Meneghiui, " as to the aiTangement 

 of some species in one or other of the two genera, which, independently of 

 their names, appear so distinct and so clearly defined, arises from the great 

 difficulty of discerning the parallel tubes including the particular series of 

 naviculae. In some species the wall of the external tube is clearly distinct, 

 and the naviculae are confused within ; but in some others it seems as if, 

 instead of a tube, there were a mucous mass in which the naviculae are im- 

 mersed." Professor Smith considered that " this great diversity of opinion 

 owes its origin to the variableness and inconstancy of the characters adopted 

 by the writers who arranged the species under two genera. The presence of 

 only one or of many files of frustules is certainly, to some extent, dependent 

 upon the stage of growth of the specimen examined ; and the appearance of 

 secondary tubes within the general mucus-envelope is more or less apparent 

 in different portions of the same frond, or according as it is examined in the 

 fresh or dry state. A very extensive comparison of specimens leads me to 

 beheve that in every case where the development of the frond is much 

 advanced, as in the older or basal portions, numerous files of frustules may 

 be observed." For these reasons Professor Smith united the genera and 

 divided Schizonema into two sections, " the first having frustules fir^mly 

 sHiceous, and fronds, in consequence, somewhat setaceous and robust ; and the 

 second including those species whose frustules are flaccid and dehcate in 

 character." As we consider the diagnostic differences sufficient, we have 

 retained, with slight alteration, the arrangement of the species in these 

 genera given by Meneghini in his memoir upon the Diatomacea?. The frond 

 in Schizonema is generally densely tufted and more sparingly branched than 

 in Micromega. It is always single-tubed, and usually very slender, with 

 even, parallel margins. The ends of the filaments, which in the early state 

 are often empty, iinaUy become ruptured and permit the escape of the 

 naviculae. In a recent state these characters will generally suffice to deter- 



