loosely aggregated together as that a slight force should
occasion a lateral displacement of them. Still it must he
allowed to be very singular, that the frustules should not
rather separate altogether from each other, than allow so
great a displacement as is frequently seen to occur, and yet
should retain their adherence.
A separate frustule bears much resemblance to the undilated
form of the frustule of Diatoma elongatum. Colour
yellowish brown.
I am in some doubt as to the precise locality of this production.
I t was either gathered by the Rev. D. Landshorough
at Stevenston, in Ayrshire, and then probably found
near the sea, or (what is less likely) by myself, somewhere in
the neighbourhood of Cheshunt.
72. F R A G IL A R IA Lyngh.
Char. Filaments compressed, attenuated, fragile, rarely angular,
until, in most cases, tioo canals or grooves passing
round the centre o f each frustule, indicated hy the presence
o f puncta on the lateral margins.
Derivation. From frango, to break; in allusion to the
brittleness of the filaments.
The genus Fragilaria approaches very closely to tha t of
Diatoma as restricted in this work ; indeed the two genera
merge into each other, through Fragilaria virescens or rather
Diatoma virescens. The compressed character of the frond
is sufiicient to distinguish it and all other Diatomaceoe from
the genus Aleloseira.
1. F ra g ila r ia p e c t in a l is Lyngh.
P late XCV. Figs. 1. 4.
Char. Filaments large, dark hrown. Frustules in front view
quadrangular, two or three times broader than long, longitudinally
striated loith two evident puncta on the lateral
margins ; in end view curved with slight constrictions at
each extremity, striated.
/3 undulata Ralfs. —• Convex margin o f end view o f the
frustule M;^Z7¿ two indentations; centre o f concave margin
prominent.
Ag. Syst. p. 7. ; Consp. Diat. p. 62. ; Grev. in Br. Fl.
vol. ii. p. 403. ; Harv. Br. Alg. p. 197. Conf. pectinalis
Dillw., t. 4. ; Eng. Bot. t. 1611. ; Ju rg , Dec. 18. No. 12.
Fragilaria grandis Ehr., Die Infus. p. 2 0 3 .1.15. fig. 11.;
Pritch. Infus. p. 220. — jS undulata, fig. 171. ; Ralfs, in
Annals, vol. xii. pl. 2. fig. 3. F. pectinalis Ehr., pl. xvi.
fig. 1 a ; Jenner, loc. cit. p. 202.
Llah. Everywhere common in pools, ditches, &c. — /3 undulata.
Drws Ardudwy, near Barmouth : Rev. T. Salweg.
Scrubbs, Colebrook Park : Mr. Jenner.
The filaments of this species and others of the genus vary
exceedingly in size, or rather in diameter, on account of the
cells of the species of the genus developing themselves in
breadth as well as length, and this developement occasions
the greatest variations in the width of the cells, compared to
their length : thus sometimes the cells are longer * than broad
in the very slender filaments ; at others, as in the longest
threads, they are six or seven times broader than long, but
usually their breadth exceeds their length only by three or
four times.
On the anterior surfaces of the frustules at either extremity,
but not extending entirely across them, a delicate row of striæ
may be detected : these add greatly to the beauty and interest
of the plant, and do not appear hitherto to have been described :
they are doubtless formed by the continuation of the striæ
observed in the end view of the frustule. Mr. Jenner has
* The terms length and breadth, when employed in this work in the
description of the Diatomaceoe, are used in the same sense in which they
are employed in the definitions of the Confervæ, the breadth of the frustule
corresponding to the diameter of the filament, and the length to its longitude.
In the Diatomaceoe the breadth of the frustule usually exceeds the
length.