112 p. T. CLEVE, SYNOPSIS OF THE NAVICITLOID DIATOMS. 



median line of the lattei*. I have proposed (18«3 Vega p. 474) to include these forms in a Section 

 Vegrr, biit I now prefer to connect them with the other species of N. ortJiostichce. 



17. N. porto-moiitaiia Cl. N. Sp. — V. broad, linear, slightly gibbous in the middle, with 

 broad rounded ends. L. 0,07; B. 0,017 mm. Median line with approximate median pores and 

 bordered by a narrow silicious rib. Terminal nodules thick, terminal fissures indistinct. Trans- 

 verse strise 19 to 20, longitudinal 19 in 0,oi mm. The puncta close to the median line are 

 larger than the others. — PI. Ill f. 36. 



Fresh water: Puerto Monte, Chile, fossil (Kinker Coll.)! 



Gyrosigma Hassall (1845). 



Valve more or less elongated and sigmoid. Central nodule small. Ends of the median line 

 in contrary directions. Central area small or indistinct. Axial area indistinct. Structure: puncta 

 disposed in transverse and longitudinal rows. — Cdl-confents (of the freshwater forms) with two 

 chromatophores along the connecting zone, which long before the division of the cell are trans- 

 versely cut oif and migrate in pairs to the inside of the valve. The opening between the halves 

 of the chromatophores becomes oblique, and each half increases to a chromatophore. The margins 

 of the chromatophores entire (Pfitzer, Ban und Entw. p. hi). Marine species (Gr. balticum) liave 

 irregularly serrated chromatophores the indentations l)eing directed towards the central nodule. 

 The median part of the chromatophores is ol)liquely striate, their substance being alternately 

 thicker and thinner. The stria^ of the two chromatophores cross each other in an oblique angle 

 (0. MlJLLER Ber. d. Deutch. Bot. Ges. 1883 p. 481). 



The sigmoid NaviculcB were named Narieula Sigma by Ehrenberg. Hassall proposed for them 

 the name Gyrosigma, which was adopted b}' Rabenhorst (Die Siissw. Diat. 1803), but not by other 

 diatomists, who prefered the newer name Fleurosigma, formed by W. Smith, 1852, who published the 

 first monograph of the species (Ann. Xat. Hist. 2 ser. IX p. 1). The genus Pleurosigma, as accepted 

 by all diatomists, includes forms with a structure of small puncta or alveoli, disposed in transverse 

 rows, which are crossed by other rows, either longitudinal, or obliquely decussating. There are no 

 intermediate forms between these two types, and I tliink they may justly be considered as different 

 genera. For the forms witli tlie puncta in transverse and longitudinal rows, I adopt the name 

 Gyrosigma, although, as (tRunow remarks, this name involves taiitology. For the forms with the 

 puncta disposed in transverse and oblique rows I reserve the name Pleurosigma. — Among the forms 

 of Gyrosigma are several with carinated valves, for which Ralfs 1861 (Pritch. Inf. p. 920) proposed 

 the generic name Donl-inia. In my opinion this genus is not acceptable, as founded on a charac- 

 teristic which is subject to too much variation. The same may be the case with lilioicosigma, pro- 

 posed 1867 by Grunow (Hedwigia VI p. 10) for forms with genuflexed or arcuate frustules. The 

 genus Rhoicosigma seems at first sight to be better founded, as the valves of the same frustnle 

 of 11. compactum are (as Peragallo has shewn) dissimilar. But on the other hand some forms 

 (as R. rohustum) have evidently similar valves. Besides, the flexure of the frustnle differs in 

 different species to all degrees. The manuscript-name Endosigma BriiIB. for the forms living like 

 Schizonema, in gelatinous tubes is not admissible, on the same grounds as Schizonema, Colleto- 

 nema, Endostauron etc. — From Gyrosigma may be removed Pleurosigma staiirophorum Grfn., 

 which has no close affinity to any of tlie other forms, but has the characteristics of C'aloneis, being 

 a sigmoid form of that genus. — The division of the sigmoid forms of Navicula into two groups, 

 founded on the disposition of the puncta in transverse and longitudinal, and in transverse and 

 oblique, rows, was first proposed by W. Smith, and has been accepted by all later diatomists. 

 In the year 1880 Grunow published (in Arctische Diat.) an elaborate monograph, in MJiich he 



