KONGL. SV. VKT. AKADKMIKNS II AXI)I,IN<;A1{. I'.AM) 26. N:() 2. 7 



as for instance Navicula liaeana and Nav. Yarrensis, Nov. tumida and Nav. plifnta, Navicula 

 SUfina and Nav. superimposita, Cocconns he{cropleura and C. pellncida. 



Tlio direction of the terminal fissures in most NavionlfB is the same in both extremities, 

 hut there are forms in which they are disposed in contrary directions. The latter characteristic 

 on which Grunow was iiiclined to found a separate genus, Fscudopleurosigma, occurs in widely 

 different groups, and it would not be in accordance with a natural system to class in one genus 

 all forms with the terminal fissures in contrary directions. Such forms are frequently allied to 

 others with the fissures in the same direction. On the other hand as a specific characteristic it 

 seems to be of very great value. 



Areas, or structureless parts of the valve, occur usually along the median line and around 

 the central nodule. I call these areas axial and central areas. There are also, sometimes, lateral 

 areas, or blank spaces on botli sides of the median line, separated from the latter by a striated 

 portion. The form, the presence, or abseiice, of such areas is in most cases a valuable specific 

 characteristic, and, fre(|uently, a useful distinction in groups of allied forms. 



Longitudinal lines. In a great number of forms there occur lines, parallel to the median 

 line, in some cases dilated into longitudinal bands or lateral areas, limiting an interior elevated 

 or depressed portion. Although such longitudinal lines or lateral areas occur in certain groups of 

 forms most of which are devoid of them, as in the Navicula lineolatoi, and in the whole group of the 

 Nariculcc lyratfB, the}- are nevertheless in my opinion of great systematic importance. In the Navicuhe 

 lyratce the longitudinal lines or lateral areas are expansions of the central nodule, corresponding with 

 tlie horns of Diploneis. In other cases they are merely non-striate portions of the valve, but in 

 AmpJriprora, Diploneis, (Jaloneis they are of a different nature. In Aniphiprora they mark the limit 

 between the elevated interior and the Hatter exterior parts, which fre(|uently have a different struc- 

 ture. They appear to be of the same nature in (Jaloneis. In Diploneis they limit a depressed 

 portion, and the same is the case in several groups of Amphora, in which they are frecj^ently so 

 placed as to be distinctly seen only in favoural)le positions of the valve. In several cases the 

 longitudinal lines are double. In Diploneis I call the space between the median line and the in- 

 terior longitudinal line the farroir, and the space between two longitudinal lines the honila. The 

 presence or absence of longitudinal lines appears to me to be of very great importance in the con- 

 struction of genera. 



Structure of the valve. The valves of the naviculoid diatoms vary greatly in structure. In 

 most cases the valves are decorated with small or large puncta, disposed in different manners. 

 At present it seems to be generally admitted that these piincta are small cavities in the silicious 

 mass of the valve. They are rarely scattered irregularly, but are usually arranged in transverse 

 rows, more or less radiate or parallel, called strife. The puncta are also so arranged as to form 

 •straight, or undulating, longitudinal rows, parallel to the median line, or decussating straight, or 

 curved, rows as in Fleurosigma and others. When the puncta are large they are called alveoli 

 and have the appearance of more or less quadrate spaces, as in several forms of Diploneis and Am- 

 phora. The silicious mass enclosing the alveoli in such cases has a reticulated appearance. I call 

 these siliceous walls of the alveoli, if forming continuous silicious ribs, costrc. The strong, trans- 

 verse silicious ribs of Diploneis Crahro are such costee. The alveoli or puncta are frequently united, 

 as in Finnularia, (Jaloneis etc. In such cases the stria? appear to be channels, communicating with 

 the interior, in the larger Pinnnlarice, by a larger foramen or ocellus. The outside part of the 

 valve consists in several cases of a thin, minutely punctate stratum, usually seen only with difficulty. 

 In a variety of Diploneis S2)lendida, called Dip. diplosticta, there is such a finely punctate stratum 

 outside the layer of alveoli. In the genus Dietyoneis it is also visible, the minute puncta forming 

 transverse, and at the same time obliquely decussating stria\ This is also the case with the genus 

 TraeJnjneis, where the fine puncta form longitudinal or somewhat oblique lines. In the group Navi- 

 cidce lineolatrf the transverse strict are crossed by fine lines, which appear to be formed by closely 

 crowded alveoli, as transitions exi.st from strife of this nature to strife composed of distinct 



