Fig. 296.—Gomphonema constrictum. (From a microphotograph.)
This beautiful sight was not confined to one frustule, but was exhibited in all that were in a healthy condition. I examined several, and watched them for a long time. The phenomenon described depends much upon the healthy condition of the frustule at the time; as the movements of the diatoms became sluggish, the circulation gradually slackens and then ceases altogether. I also saw a somewhat similar action in the more active specimens of P. hippocampus and Navicula cuspidata, but the coarser markings and thickness of the wall of these diatoms seemed to place greater difficulties in the way of observation than the finer valves of the P. angulatum. One thing I believe is certain, that the circulation described is precisely similar to that seen in the Closteria, or, on a much larger scale, in Chara and the leaf of the Anacharis, bearing in mind also that in the Closterium the cell is divided by a transverse suture, and in P. angulatum by a longitudinal one ([Plate II]., Nos. 38-40). About the same time some very lively specimens of the Pinnulariæ were sent to me, and the movements of these frustules were more closely observed. One or two of the more active would attack a body relatively larger than itself, it would also force its way into a mass of granular matter, and then recede from it with a jerky motion. In more than one instance a cell of Palmoglæa was seized and carried away by the Pinnularia, the former at the time being actively engaged in the process of cell division. Other diatoms present among my specimens were also in an active condition, and the circulation of granular matter in all was distinctly visible. In the Pinnulariæ two large colourless vesicles were seen on either side of the median nodule, each having a central nucleus, as represented in the accompanying sketch, made while under observation in two positions. The central portion of each frustule was closely packed with a rich yellowish-brown coloured endochrome, interspersed with a few fat globules. The phenomenon of cyclosis was not seen in any of these diatoms, but I have satisfied myself, by staining, of the presence of a delicately fine external protoplasmic covering in many diatoms. That their movements resemble the gliding movements exhibited by the Amœba can scarcely be doubted. Numerous forms of Diatomaceæ are found growing on or attached to water-plants or pieces of detached stalks, which, although generally simple, are sometimes compound, dividing and subdividing in a beautiful ramous manner. Pinnulariæ, Nitzschia, &c., are seen adherent by one extremity, about which they turn or bend themselves as on a hinge. By the process of cell-division, groups of Synedræ become attached by a point, in a fan-like form. The fan-like collection of frustules is said to be flabellate, or radiate. In Licmophora, Achnanthes and other species ([Plate II]., Nos. 29-33) the double condition of union of frustules and of attachment by a pedicle are illustrated. When a stipe branches it does so normally in a dichotomous manner, each new individual being produced by a secondary pedicle. This regular dichotomy is seen in several genera: Cocconema and Gomphonema, the latter more perfectly in [Fig. 296], from a microphotograph, in which a branching, or rather longitudinal, rupture of the pedicle takes place at intervals, and the entire organism presents a more or less complete flabella, or fan-like cluster, on the summit of the branches, and imperfect or single frustules irregularly scattered throughout the whole length of the pedicle.
Isthmia enervis (Fig. 297).—The unicellular frustule of this species is extremely difficult to define, owing to the large areolations of the valves; it has a remarkable internal structure. The olive-brown cell contents are found collected, for the most part, into a central mass, from which radiating, branched, granular threads extend to and unite with the periphery. When viewed by a magnifying power of 600 or 700 diameters, these prolongations are seen to be composed of aggregations of ovate or spindle-shaped corpuscles, held together by protoplasmic matter. These bodies are sometimes quiescent, but more often travel slowly to and fro from the central mass. The general aspect under these conditions so nearly corresponds to the characteristic circulation in the frustules of unicellular plants and of certain rhizopoda, that it is difficult to realise that the object when under examination is an elegant marine diatom.
Fig. 297.—Isthmia enervis. Microphotograph.
There is a large section of diatoms in which the frustules are diffused throughout a muco-gelatinous envelope in a definite manner. Histologically this is homologous with the pedicles and connecting nodules thrown out during the act of self-division, and in some species (Cocconeis, Fragillaria, &c.) it often persists after that act is complete.
Diatomaceæ, Recent and Fossil.
Fig. 298.—Fossil Diatoms from Springfield (Barbadoes).