Confervaceæ are a genus of algals. The species consist of unbranched filaments composed of cylindrical or moniliform cells, with starch granules. Many are vesicular, and all multiply by zoospores generated in the interior of the plant at the expense of the granular matter. They are, for the most part, found in fresh water attached or floating, some in salt water, and a few in both, in colour usually green, but occasionally olive, violet, and red. The Confervaceæ proper are often divided into four families: 1. Hydrodictidæ; 2. Zygnemidæ; 3. Confervidæ; 4. Chætophoridæ. To the microscopist all the plants of this genera are extremely interesting as subjects for the study of cell multiplication. The process usually takes place in the terminal cell, the first step in which is the division of the endochrome, and then follows a sort of hour-glass contraction across the cavity of the parent cell, whereby it is divided into two equal parts. This is better seen in some of the desmids than in Fig. 284, Nos. 4, 5, and 6. Some species are characterised by a different mode of reproduction; these possess a number of nuclei, and multiply by zoospores of two kinds, the largest of which have either two or four cilia, which germinate directly the smaller are biciliated; conjugation has been seen to take place in a few instances.

Allied to the Confervaceæ is an interesting plant, Sphæroplea annulina, which has received careful attention from Cohn. The oospores of this plant are the product of a process partaking of a sexual nature, and when mature are filled with reddish fat vesicles which divide by segmentation.

The Ædogoniaceæ also closely resemble Confervaceæ in habits of life, but differ in some particulars, especially so in the mode of reproduction (only a single large zoospore being set free from each cell) and by the almost complete fission of the cell-wall or one of the rings which serve as a hinge. The zoospores are the largest known among algals, and each is described as having a red eye-spot. The Chætophoraceæ form an interesting group of confervoid plants, and are usually found in running streams, as they prefer pure water. One of the characteristics of the group is that the extremities of the branches are prolonged into an acute-shaped termination, as represented in Fig. 284, No. 6. A very pretty object under the microscope is Draparnaldia glomerata, belonging to this species. It consists of an axis composed of a row of cells, and at regular intervals whorls of slender prolongations, containing chlorophyll or endochrome of a deeper green; these attain to an extraordinary length.

The Batrachospermæ bear a strong resemblance to frog-spawn, from which they derive their name, and are chiefly a marine group of algals allied to the Rhodespermeæ or red seaweeds. The late Dr. A. Hassall first described them; they have since received more careful attention from M. Sirodot. They are reddish-green, extremely flexible, and nothing can surpass the grace of their movements in water; but when removed from their element they lose all form, and resemble a jelly-like substance without a trace of organisation; but if allowed to remain quiet they regain their original shape.

The presence of the cell-membrane will be best demonstrated by breaking up the filaments, either by moving the thin glass cover, or by cutting through a mass of them in all directions with a fine dissecting knife. On now examining the slide, in most instances many detached empty pieces of the cell-membrane, with its striæ, will be seen, as well as filaments partly deprived of protoplasm. On the application of iodine all these appearances become more distinguishable in consequence of the filament turning red or brown, while the empty cells remain either unaffected, or present a slight yellowish tint, as is frequently the case with cellulose when old.

With regard to the contents of the cell, the endochrome is coloured in the Oscillatoriæ, and is distinguishable by circular bands or rings around the axis of the cylindrical filament. Iodine stains them brown or red, and syrup and dilute sulphuric acid produce a beautiful rose colour. As to their mode of propagation, nothing positive is known. If kept for some time they gradually lose their green colour; a portion of the mass, becoming brown, sinks to the bottom of the vessel, and presents a granular layer.

Mesoglia vermicularis (Fig. 285) consists of strings of cells cohering and held together by their membranous covering. In the lowly organised plant Vaucheria (Plate I., No. 23, V. sessilis)—so named after its discoverer Vaucher, a German botanist—a genus of Siphonaceæ, we have an example of true processes of sexual generation. The branching filaments are often seen to bear at their sides peculiar globular bodies or oval protuberances, nipple-shaped buddings-out of the cell-wall, filled with a dark-coloured endochrome and distributed in pairs, one of which curves round to meet the other, when conjugation is seen to take place. Near these bodies others are found with pointed projections, which have been described as “horns,” but these, Pringshelm says, are “antherids which produce antherozoids in their interior,” while the capsule-like bodies constituting the oospores become, when fertilised, a new generation, which swarm out through a cavity or aperture in the parent cell-wall.

The fruit of fresh-water and most olive-green algals is enclosed in spherical cavities under the epidermis of the frond, termed conceptacles, and may be either male or female. The zoids are bottle-shaped and have flagella; the transparent vesicle in which they are contained is itself enclosed in a second of similar form. In monœcious and diœcious algals the female conceptacles are distinguished from the male by their olive colour. The spores, when developed, are borne on a pedicle emanating from the inner wall of the conceptacle. They rupture the outer wall at its apex; at first the spore appears simple, but soon after a series of changes takes place, consisting in a splitting up of the endochrome into six or eight masses of spheroidal bodies. A budding-out occurs in a few hours’ time, and ultimately elongates into a cylindrical thread. The Vaucheria present a double mode of reproduction, and their fronds consist of branching tubes resembling in their general character that of the Bryophyta, from which indeed they differ only in respect of the arrangement of their green contents. In that most remarkable plant Saprolegnia ferox, which is structurally so closely allied to Vaucheria, though separated from them by the absence of green colouring matter, a corresponding analogy in the processes of development takes place. In the formation of its zoospores, an intermediate step is presented between that of the algæ and a class of plants formally placed among fungi.

The Ulvaceæ.—The typical form of seaweeds is the Ulva lactuca, well known from its fronds of dark-green “laver” on every coast throughout the world. Ulvæ are seen to differ but little from the preceding group of fresh-water algals. The specific difference is that the cells, when multiplied by binary subdivision, not only remain in firm connection with each other but possess a more regular arrangement. The frond plane of the algal is either more simple or lobed, and is formed of a double layer of cells closely packed together and producing zoospores. The whole group is chiefly distinguished from Porphyra by their green colour, the latter being roseate or purple. Ulvæ are mostly marine, with one or two exceptions. One species (U. thermalis) grows in the hot springs of Gastein, in a temperature of about 117° Fahr. The development of Ulvæ is seen in Fig. 286. The isolated cells, A, resemble in some points those of the Protococcus; these give rise to successive subdivisions determining the clusters seen at B and C, and by their aggregation to the confervoid filament shown at D. These filaments increase in length and breadth by successive additions, and finally take the form of fronds, or rows of cells.

The marine greenish-olive algæ present a general appearance which might at first sight be mistaken for plants of a higher order of cryptogams. Their fronds have no longer the form of a filament, but assume that of a membranous expansion of cells. The cells in which zoospores are found have an increased quantity of coloured protoplasm accumulated towards one point of the cell-wall; while the zoospores are observed to converge with their apices towards the same point. In some algæ, which seem to be closely related in form and structure to the Bryophyta, we notice this important difference, that the zoospores are developed in an organ specially destined for the purpose, presenting peculiarities of form and distinguishing it from other parts of the branching tubular frond. In the genus Derbesia distinct spore cases develop, a young branch of which, when destined to become a spore case, instead of elongating indefinitely, begins, after having arrived at a certain length, to swell out into an ovoid vesicle, in the cavity of which a considerable accumulation of protoplasm takes place. This is separated from the rest of the plant, and becomes an opaque mass, surrounded by a distinct membrane. After a time a division of the mass takes place, and a number of pyriform zoospores, each of which is furnished with flagella, are set free.

In Cutleria (Fig. 288) we have a special feature of interest with two kinds of organs, seemingly opposed to each other with regard to their reproductive functions. The sporangia not only differ from those of other species, but the frond consists of olive-coloured irregularly-divided flagella, on each side of which tufts (sori) consisting of the reproductive organs, intermixed with hair-like bodies, are scattered. The zoospores are divided by transverse partitions into four cavities, each of which is again bisected by a longitudinal median septum. When first thrown off they are in appearance so much like the spores of Puccinia that they may be mistaken for them, although so very much larger than those of other olive-coloured algæ.

Florideæ, the red algæ (Plate II.), present many varieties of structure, although less appears to be known of their reproductive processes than of lower forms of cryptogamic plants. These are, however, of three kinds. The first, to which the term polyspore has been applied, is that of a gelatinous or membranous pericarp or conceptacle, in which an indefinite number of zoospores are contained. This organ may be either at the summit or base of a branch, or it may be concealed in or below the cortical layer of the stem. In some cases a number of spore-bearing filaments emanate from a kind of membrane at the base of a spheroidal cellular perisporangium, by the rupture of which the zoospores formed from the endochrome of the filaments make their escape. Other changes have been observed; however, they all agree in one particular, namely, that the zoospore is developed in the interior of a cell, the wall of which forms its perispore, and the internal protoplasmic membrane endochrome, the zoospore itself, for the escape of which the perispore opens out at its apex.

The second form is more simple, and consists of a globular or ovoid cell, containing a central granular mass; this ultimately divides into four quadrate-shaped spores; these, on attaining maturity, escape by rupture of the cell-wall. Another organ, called a tetraspore, takes its origin in the cortical layer. The tetraspores are arranged either in an isolated manner along the branches, or in numbers together; in some instances the branches that contain them are so modified in form they look like special organs, and have been called stichidia; as, for example, in Dasya (Fig. 289). Of the third kind of reproductive organ a difference of opinion exists as to the signification of their antheridia; although always produced in precisely the same situations as the tetraspores and polyspores, they are agglomerations of little colourless cells, either united in a bunch, as in Griffithsia, or enclosed in a transparent cylinder, as in Polysiphonia, or covering a kind of expanded disc of peculiar form, as in Laurencia. According to competent observers, the cells contain spermatozoids. Nägeli describes the spermatozoid as a spiral fibre, which, as it escapes, lengthens itself in the form of a screw. Thuret, on the contrary, says the contents are granular, and offer no trace of a spiral filament, but are expelled from the cells by a slow motion. The antheridia appear in their most simple form in Callithamnion (Plate II., Nos. 32 and 34), being reduced to a small mass of cells composed by numerous little bunches which are sessile on the bifurcations of the terminal branches. The spores are simpler structures than the tetraspores, and mostly occupy a more important position. They are not scattered through the frond, but grouped in definite masses, and generally enclosed in a special capsule or conceptacle, which may be mistaken for a tetraspore case. The simplest form of the spore fruit consists of spherical masses of spores attached to the wall of the frond, or imbedded in its substance, without a proper conceptacle; such a fruit is called a favellidium, and occurs in Halymenia; the same name is applied to the fruits of similar structures not perfectly immersed, as those of Gigartina, Gelidium, &c., where they form tubercular swellings on the lobes. In some, the tubercles present a pore at the summit, through which the spores emerge forth. In other cases, as in Ceramium (Plate II., Nos. 27 and 37), the spores occupy a more conspicuous place; a characteristic species is Delessaria (Plate II., No. 39), the coccidium either occurring on lateral branches, or is sessile on the face of the frond, when it consists of a case filled with angular-shaped spores attached to the wall of the case. The general external appearance of the red algæ is very varied, but it seems to attain to its deepest colouring in the Red Sea, which, it is said, is entirely due to the peculiarly vivid red seaweed. They are all exquisite objects for the microscope, as may be surmised from the few varieties presented in Plate II. The Florideæ of the warmer seas exhibit most elegantly formed fronds, as will be seen on reference to the “Phycologia Australica” of the late Dr. William Harvey, F.R.S.

The Characeæ may be placed among the highest of the algals, if only for the complexity of their reproductive organs, which certainly offer a contrast in their simplicity of structure. Chara vulgaris, stonewort, is a simple fresh-water plant, preferring still freshwater ponds or slow-moving rivers running over a chalky soil. It thus derives the calcareous matter found in the axis of the plant, together with a small portion of silica. Its filaments (or branches, as some botanists prefer to call them) are given off in whorls. The Characeæ are a small family of acrogens, consisting of only two or three at most. They are monœcious and diœcious, the two kinds of fruit being often placed close together. They may easily be grown in a tall glass jar for observation. All that is necessary is to put the jar occasionally under the house tap and let the water run slowly over the top for a short time, thus renewing the contents without disturbing the plant. The hard water supplied to London suits chara better than softer water. Both chara and nitella are objects of great interest to microscopists, since in the former the important fact of vegetable circulation was first observed. A portion of the plant of the natural size is shown in Fig. 290, No. 1.

Each plant is composed of an assemblage of long tubiform cells placed end to end, with fixed intervals, around which the branchlets are disposed with great regularity. In nitella the stem and branches are composed of simple cells, which sometimes attain to several inches in length. Each node, or zone, from which the branches spring, consists of a single plate, or layer, of small cells, which are a continuation of the cortical layer of the internode (Fig. 290, No. 3) as an outgrowth. Each cell is partially filled with chlorophyll granules, and it is these that are seen under the microscope taking the course shown by the arrows (Fig. 290, No. 2). The rate of movement of the granules is accelerated by moderate warmth and retarded by cold. It is in viewing the circulation in water plants that the warm stage of the microscope is brought into use. Borne along with the protoplasmic stream are a number of solid particles consisting of starch granules and other matters. The method of viewing the circulation is by cutting sections off a portion of the plant with a very sharp knife, and arranging them in a growing cell with a few drops of water, and covering over with a thin cover-glass.

The reproductive process of Chara is effected by two sets of bodies, both of which are placed at the base of the branches (Fig. 291, E and D) either on the same or different plants, one set known as globules or antherids, and the other as nucules, containing the oospores or archegones. These are often of a bright red colour, and have covering plates, or shields (B and E), curiously marked, and the central portion is composed of a number of filaments rolled up (as in E) or free (as seen at B), projecting out from the centre of the sphere. The antherid is supported on a short flask-shaped pedicle, which projects into the interior. At the apex of each of the eight manubria is a roundish hyaline cell, termed a capitulum, and at its apex again six smaller or secondary capitula. The long whip-shaped filaments are divided by transverse septa into a hundred or more compartments, every one of which is filled with an antherozoid (as at A), consisting of a spiral thread of protoplasm packed into two or three coils; these escape and become free (as seen at C), each having two long fine flagella. The young antherozoid swims off with a lashing action, and the whole field appears for a time filled with life. They swim about freely, but their motion gradually ceases, and soon they arrive at a state of inaction.

Nitella appears to have a somewhat different mode of fructification to that of its congener. It puts forth a long filamentous branch from one of its joints, which, on reaching the surface of the water, terminates in a whitish fruit-like cluster. It is even a more delicate and less robust algal than chara, and every care should be taken to imitate the still water in which it grows. It delights in shady woods and in calcareous open pools.

Similar care is requisite with regard to Vallisneria; and a more equal temperature is better suited to the growth of this aquatic plant. It should be planted in the middle of the jar or aquarium, about two inches deep in mould, closely pressed down, then gently fill the jar with water. When the water requires changing, a small portion only should be run off at a time. It appears to thrive in proportion to the frequency of changing the water, and taking care that the water added rather increases the temperature than lowers it.

The natural habitat of the Frog-bit, another water-plant of much interest, is found on the surface of ponds and ditches; in the autumn its seeds fall, and become buried in the mud at the bottom during the winter; in the spring these plants rise to the surface, produce flowers, and grow throughout the summer. Chara may be found in many places around London, and in the upper reaches of the Thames.

Anacharis alsinastrum.—This remarkable plant is so unlike any other water-plant that it may be at once recognised by its leaves growing in threes round a slender stem. It is also known as “Waterthyme,” from a resemblance it bears to that plant.

The colour of the plant is deep green; the leaves are nearly half an inch long, by an eighth wide, egg-shaped at the point, with serrated edges. Its powers of increase are prodigious, as every fragment is capable of becoming an independent plant, producing roots and stems, and extending itself indefinitely in every direction. The specific gravity of it is so nearly that of water, that it is more disposed to sink than float. A small branch of the plant is represented, with a hydra attached to it, in a subsequent chapter.

The special cells in which the circulation is most readily seen are the elongated cells around the margin of the leaf and those of the midrib. On examining the leaf with polarised light, the cells are observed to contain a large proportion of silica, and present a very interesting appearance. A bright band of light encircles the leaf, and traverses its centre. In fact, the leaf is set, as it were, in a framework of silica. By boiling the leaf for a short time in equal parts of nitric acid and water, a portion of the vegetable tissue is destroyed, and the silica rendered more distinct, without changing the form of the leaf.

It is necessary to make a thin section or strip from the leaf of Vallisneria for the purpose of exhibiting the circulation in the cells, as shown in Fig. 290, No. 4. Among the cell granules, a few of a more transparent character than the rest, are seen to have a nucleolus within.

The phenomenon of cell cyclosis occurs in other plants beside those growing in water. The leaf of the common plantain or dock, Plantago, furnishes a good example, the movement being seen both in the cells of the plant and hairs of the cuticle torn from the midrib.

Cell-division.—In order to study the process of cell-division the hairs on the stamens of Tradescantia should be taken. Remove one from a bud on a warm day and let a drop of a one per cent. sugar solution fall upon it, and cover it with a thin glass cover. Place it for a short time in a moist-chamber (Fig. 256), and then examine it with a magnifying power of 500 diameters. The nucleus of the cell will be seen, near its terminal position, to gradually elongate in the direction of the longer axis of the cell and become more granular, while the protoplasm moves towards the extreme end; the nucleus at the same time will present a striated appearance, with the fibrilla arranged parallel to the longer axis of the nucleus, and at length approach each other at the poles. A nuclear spindle will now be produced, and the fibres ruptured in the equatorial plane, so that two nuclei will be found in place of the one. The best preparations of nuclei are obtained by making thin longitudinal sections of actively-growing plants (young rootlets of Pinus, for example), and staining them with hæmatoxylin in the manner described in a former chapter.

Desmidiaceæ and Diatomaceæ.

The two groups of Desmidiaceæ and Diatomaceæ differ so little in their general characters that they may be spoken of as members or representative families of microscopic and unicellular algæ alike in their remarkable beauty and bilateral symmetry, and of such peculiar interest as to call for special notice. Desmids differ from diatoms chiefly in colour, in lacking a non-silicious skeleton, and in their generative process, which for the most part consists in the conjugation of two similar cells. Diatoms, on the other hand, have dense silicious skeletons and a general absence of green colouring matter. Ralfs, in his systematic monograph, enumerates twenty genera of desmids. The limiting membrane is alike firm and flexible, since it exhibits some elasticity and resistance to pressure, and is not readily decomposable. Traces of silica are found in only a few of the desmids, while the frustule of the diatom is chiefly composed of this substance; both have an external membranous covering, so transparent and homogeneous in structure as to be in danger of being entirely overlooked, unless some staining material is used, together with a high-power objective possessing considerable penetration. In some species, however, the mucous covering is more clearly defined, as in Staurastrum and Didymoprium Grevelli. Openings occur in the outer membrane of other species, as the Closterium.

Many species of desmids have a power of motion, the cause of which must be due either to cilia or a flagellate organ. This, however, is denied by some observers, who regard their movements as due to an exudation of the mucilaginous contents of the cell, to exosmose, or diffusion, neither of which hypotheses will at all help us to understand the gliding movements of the Oscillariæ or the sharp jerky movement of the Schizonema. The movements of desmids are especially exerted when in the act of dividing, and by sunlight, towards which they are always observed to move. The force with which some diatoms move about is very great, and this can only be satisfactorily explained by admitting a specialised organ.

The appearance of the Desmidiaceæ (Plate X.) is much modified by their eminences, depressions, and processes, as well as that of the surface, the margin of the fronds, and the depth and width of the central constriction. The surfaces may be dotted over irregularly, the dots themselves being elevated or depressed points in their structural character. The margins of some have a dentate appearance, as in Cosmarium. In the elongated forms, such as Penium, the puncta are disposed in lines parallel to the length. In several these lines are either elevations or furrows, it is not always easy to say which; they are peculiar, however, to the elongated forms of Closterium. When the lines are fine they produce a striation of the surface, but in order to discover this the fronds should be viewed when empty and by a fairly good power. The modification of surface in several genera seems to be due, not to mere simple appendages, but to expansion of the limiting membrance into thickened processes, and which may terminate in spines, as in Xanthidium and Staurastrum (Plate X., Nos. 8-19 and 22). A general distribution over the surface is characteristic of the former, but in Euastrum the surfaces are very irregular, and therefore described as “swellings or inflations.” Micrasterias has its margin deeply incised into lobes, which in some have a radiating arrangement; when the lobes on the margin are small they constitute crenations or dentations. The fronds of Euastrum binatum are bicrenate on the sides, as are those of Desmidium and Hyalotheca and other species. Another variety of margin exists, known as undulating or wavy, while the general concavity or convexity of the margins furnish other specific characteristics.

Pediastreæ (Plate X., Nos. 24-29).—The members of this family formerly included the Micrasterias and Arthrodesmius of Ehrenberg. From their arrangement of cells in determinate numbers and definite forms, it has been thought by some observers that they should be removed from the desmids to a special or sub-family. The points of difference consist in the firmness of the outer covering, in the frequent interruptions on the margin of the cells, and in the protrusion of “horns,” or rather a notch more or less deep. It is true that the cells are not made up of two symmetrical halves, and that they are in aggregation, which is not (except in the Scenedesmus, a genus that distinctly connects this group with desmids) in linear series, but in the form of discoidal fronds. They, however, divide into 8, 16, or 32 gonidia, and these move about for some time before the formation of a new frond. It was Nägeli who first instituted a sub-genus of Pediastrum, under the designation of Anomopedium, the chief characteristic of which is the absence of bilobed peripheral cells. In Cœlastrum the cells are hexangular, the central ones very regularly so; in Sorastrum they are wedge-shaped, or triangular, with rounded-off angles. Viewed laterally the cells appear oblong. The cells of Pediastrum are considerably compressed, so that when aggregated they form a flattened tubular structure; in figure they are polygonal, frequently hexagonal, a shape owing, in all probability, to mutual lateral pressure during growth. There is a pervading uniformity in the contents of the cells of the different genera, which consist of protoplasmic endochrome. At first the colour is pale green, but it becomes deeper with full maturity, while the decaying cells are seen to change to a deep reddish-yellow or brown. The protoplasm is also clear and homogeneous, but in time granules appear, enlarge, and multiply in number; moreover, each cell presents a single bright green vesicle, around which are collected clear circular spaces or globules, recalling those of Closterium, and varying in number from two to six or more, their position not being regulated by the partition wall as in Palmellæ, but by the centre of the entire frond. Oil globules are also contained in the cells; their presence is indicated by the addition of a drop of tincture of iodine. On one occasion Nägeli saw in Pediastrum boryanum the endochrome disposed in a radiating manner, an arrangement which often obtains in algals and in other vegetable cells with a central nucleus. The cells of Pediastreæ are always united together in compound fronds, as represented in Plate X., Nos. 24 and 29.56

The differences pointed out in no way constitute a claim to remove Pediastreæ from among Desmidiaceæ, certainly not to rank as a distinct species.

Reproduction of Desmidiaceæ.—A true reproductive act is presented by the conjugation or coupling of two fronds, and by the resulting development of a sporangium and subsequent interchange of the contents of the two cells. At another time self-division is frequently seen to take place in all respects as in the cells of other algæ. The proceeding is varied in some essential particulars by the form of the fronds and by other circumstances; as in fission of Euastrum, for instance (seen in Plate X., Nos. 1, 2, and 12), when the narrow connecting bands between the two segments of the fronds are rapidly pushed aside by growth and finally divide. Two modes of conjugation of fronds are represented in Plate X., Nos. 25 and 33, in Closterium and Penium. The act of conjugation admits of variations in character, as shown in Staurastrum and Microsterias; the contents of both fronds are discharged into a delicate intermediate sac; this gradually thickens and produces spines (Plate X., Nos. 8 and 19). In Didymoprium the separate joints unite by a narrow process pushed out from each other, often of considerable length, through which the endochrome of one cell is transferred to the other, and thus a sporangium is produced within one of two cells, just as in the conjugatæ (No. 5). In Penium Jennereri the conjugation takes a varied form; the fronds do not open and gape at the suture, but couple by small but distinct cylindrical tubes (No. 27).

Among those enumerated, the compressed and deeply constricted cells of Euastrum offer the more favourable opportunities for studying the manner of their division; for although the frond is really a single cell, in all its stages it appears like two, the segments being always distinct, from the earliest stage. The segments, however, are separated by a connecting link, which is subsequently converted into two somewhat round hyaline bodies. These bodies gradually increase and acquire colour, and as they grow the original segments are further divided, and at length become disconnected, each taking a new segment to supply the place of that from which it is separated. It is curious to trace the progressive development of the newer portions, which at first are devoid of all colour; but as they become larger a faint green tint is observed, which gradually darkens, and then assumes a granular appearance. Soon the new segments attain their normal size, while the covering in some species shows the presence of puncta. In Xanthidium, Plate X., Nos. 9, 10, and Staurastrum, Nos. 15-18, the spines and processes make their appearance last, beginning as mere tubercles, and then lengthening until they attain their perfect form and size, armed with setæ; but complete separation frequently occurs before growth is fully completed. This singular process is repeated again and again, so that the older segments are united successively, as it were, with many generations. When the cells approach maturity, molecular movements may be at times noticed in their contents, precisely similar to what Agardh and others aptly term “swarming.” Meyen describes this granular matter as starch.57 Closterium, early in the spring, when freshly secured and exposed to light, presents a wonderful appearance, these bodies being kept continually in motion at both ends of the frustule by the ciliary action within the cell, and the whole frond is seen brilliantly glittering with active cilia. When a gleam of stronger light is allowed for a moment to fall on the frond, the rapid undulations of the cilia produce a series of most delicate prismatic Newton’s rings. The action and distribution of the cilia, together with the cyclosis of the granular bodies in the frond, are better seen by the aid of Wenham’s parabola or a good condenser with a central stop. One of the wide angular objectives shows the circulation around the marginal portions of the whole frond. The stream is seen to be running up the more external portion, internal to which is another stream following a contrary direction; this action, confined to the space between the mass of endochrome and the outer portion of the cell-wall, is seen to pass above or around the space in which cyclosis of the spores is taking place.

During the summer of 1854, the late Rev. Lord Sidney Godolphin Osborne and myself became much interested in the remarkable family of Closteria. Fig. 292 is a highly magnified view of Closterium lunula which I drew by the aid of the camera-lucida at the time. There could be no doubt about the ciliary action within the frond: it was in every way similar to that of the branchiæ of the muscle, the same wavy motion, which gradually became slower as the death of the desmid drew near. This was brought about earlier when the cell was not kept supplied with fresh water.

In diagram A, line b points to a cluster of ovoid bodies; these are seen at intervals throughout the endochrome within the investing membrane. These bodies are attached to the membrane by small pedicles, and are occasionally seen in motion about the spot, from which they eventually break away, and are carried off, by the circulating fluid, to the chambers at the extremities of the frond; there they join a crowd of similar bodies, in constant motion within the chambers, when the specimen is quite fresh. That the action of these free granules or spores is “Brownian,” as surmised by some writers, is in my opinion entirely fallacious. It is doubtless in a measure due to the current brought about by the ciliary motion of the more fluid contents of the cell.

The circulation, when made out over the centre of the frond, for instance at a, is in appearance of a wholly different nature from that seen at the edges. In the latter the matter circulated is that of granules, passing each other in distinct lines, but in opposite directions; in the circulation as seen at a, the streams are broad, tortuous, of far greater body, and passing with much less rapidity. To see the centre circulation, use a Gillett’s illuminator and a ¹⁄₈th or a ¹⁄₁₀th immersion; work the fine adjustment so as to bring the centre of the frond into focus, then almost lose it by raising the objective; after this, with great care, work the milled head until the darker body of the endochrome is clearly brought out.

At B is an enlarged sketch of one extremity of the frond. The arrows within the chamber pointing to b denote the direction of a strong current of fluid, which can be occasionally followed throughout. It is acted upon by cilia at the edges of the chamber, the greater impetus appearing to come from the centre of the endochrome. The fluid is here acting in positive jets, that is, with an almost arterial action; and according to the strength with which it is propelled at the time, the loose floating bodies are sent to a greater or less distance from the end of the frustule; the fluid is thus impelled from a centre, and kept in activity by the lateral cilia, that create a rapid current and give a turning motion to the free bodies. The line—a, in this diagram, denotes the outline of the membrane which encloses the endochrome; on both sides cilia can be seen. The circulation exterior to it passes and repasses in opposite directions, in three or four distinct courses; these, when they arrive at—c, seem to encounter a stream making its way towards an aperture at the apex of the chamber; then they appear to be driven back again by a stronger force. Some, however, do occasionally enter the chamber, but very rarely will one of the bodies escape into the outer current, and should it do so, is carried about until it becomes adherent to the side wall of the frond.

With regard to the propagation of the C. lunula, I have never seen anything like conjugation; but I have repeatedly seen self-division (shown at D a a). This act is chiefly the work of one half of the frond. Having watched for some time, one half is seen to remain passive, while the other has a lateral motion from side to side, as if moving on an axis at the point of juncture; the motion increases, is more active, until at last with a jerk one segment separates itself from the other, as seen at E. It will be noticed that each end of the segment is perfectly closed before separation finally takes place; there is, however, only one perfect chamber, that belonging to the extremity of the original entire frond. The circulation continues for some time previous to and after subdivision, in both fronds, and by almost imperceptible degrees increases in volume. From the end of the endochrome symptoms of elongation of the frond take place, the semi-lunar form gradually changes, elongates, and is more defined, until it takes the form and outline of the fully-formed frustule at the extremity. The obtuse end—b of the other portion of frond is at the same time elongating and contracting, and in a few hours from the division of the one segment from the other the appearance of each half is that of a nearly perfect frustule, the chamber at the new end is complete, the globular circulation exterior to it becomes affected by the circulation from within the said chamber, and, shortly afterwards, some of the free bodies descend, and become exposed to the current already going on in the chamber. E is a diagram of one end of a C. didymotocum, in which the same process was well marked, and completed while it was under observation.

It will appear to most observers that if the continuation of the widely-spread family of Desmidiaceæ was wholly dependent upon conjugation and subdivision of their frustules, a process requiring several hours for its completion, the whole species must have long ago disappeared. It may be presumed then that some other mode of reproduction must prevail. In the fresh-water algæ the two more general methods of multiplication are clearly governed by the conditions of the seasons; the resting-spores securing continuity of life during the winter, the swarm-spores spreading the plant profusely during the warmer portion of the year, when rapid growth is possible. I therefore regard the actively swarming bodies seen in continuous motion at the two extreme portions of the frustule of Closterium lunula as being either oospores or zoospores, by means of which reproduction takes place.

Diatomaceæ, commonly called brittleworts, Plate XI., are chiefly composed of two symmetrical valves, narrow and wand-like, navicular, miniature boat-shaped, hence their name Navicula (little ship). Hitherto they have excited the deepest interest among microscopists because of their wonderfully minute structure, and the difficulty involved in determining their exact nature and formation. Each individual diatom has a silicious skeleton, spoken of as a frustule, frond, or cell, having a rectangular or prismatic form, which mostly obtains in the whole family, the angles of the junction of the two united valves being, as a rule, acute, and enclosing a yellowish-brown endochrome. Deeply-notched frustules, like those of the Desmidiaceæ, do not occur, and the production of spines and tubercles so common in that family is rare in the Diatomaceæ. Great variety of outline prevails, so much so that no rule in this respect can be formulated.

The frustules, however, are usually composed of two equal and similar halves, but exceptions to this are found in the Actinomtheæ, Cocconcidæ, and one or two other families. The extremities of some species, e.g., Nitzshia and Pleurosigma, are extremely elongated, forming long, filiform, tubular processes; in Biddulphia and Rizoselenia, short tubular processes from their margins. Great variety of outline may prevail in a genus, so considerable indeed that no accurate definition can be given, the characteristics shading off through several species until the similarity to an assumed typical form is much diminished, which may again be modified by accidental circumstances that surround the development of the silicious frustule. It must not be forgotten that the figure is greatly modified or entirely changed by the position of the valves, whether seen in one position or another, as already explained in connection with “Errors of Interpretation.” Again, in the genera Navicula, Pinnularia (Plate II., Nos. 33, 38, and 40), and others, the frustules are in one aspect boat-shaped, but in the other either oblong with truncated ends, or prismatic. In the genus Triceratium (Plate XI., No. 10), the difference of figure is very remarkable as the front or side view is examined.

The sudden change in appearance presented to the eye as the frustule is seen to roll over is rather peculiar. As a rule, therefore, we must examine all specimens in every aspect, to accomplish which very shallow cells should be selected, say of ¹⁄₁₀₀th of an inch deep, and covered with glass ¹⁄₂₅₀th of an inch thick. A good penetrating objective should be used, and careful illumination obtained. The Diatomaceæ are perhaps more widely distributed than any other class of infusorial life; they are found in fresh, salt, and brackish water; many grow attached to other bodies by a stalk (Plate II., No. 33, Licmophora and Achnanthidium); while others, as Pleurosigma, No. 40, swim about freely.