amber colour, but leaves the capsule and globular cell unaffected.

Many globular vesicles of different sizes, which appear to be entirely empty, are also to be observed, and here and there a solitary starch-globule.

But the composition of the irregular bodies, which appear to vary in size from a minute granule to the largest starch-globule, is not so evident. These bodies, which have a faint yellow opake colour, and apparently granular structure, are like the circular disks loosely scattered through the mucus, or enclosed in plurality within transparent globular vesicles otherwise empty, or attached to or imbedded singly, in the wall of one of these cells, which also seems to be their normal appendage. In Chara verticillata (Roxb.) these bodies in the early part of their growth are club-shaped, after which the large end appears to expand into an irregular, globular or agariciform head, to which the small end then forms a kind of pedicel, and thus they are also found within vesicles. At first these irregular bodies look very much like starch-granules, and particularly the agariciform ones, from the eccentric lines on their surface; but iodine, even when assisted by sulphuric acid, only turns them of a deep brown amber colour like that which it produces in the nucleus of the disk; sometimes it seems to have little or no effect upon them. They differ from the circular disk in the extreme irregularity of their form, their apparent want of capsule, greater thickness, deeper yellow colour, greater opacity, and in their apparent origin from granules infinitely smaller than the circular disks.

Axial fluid. This, as before stated, fills the centre of the internode; it is of an aqueous consistence, colourless, and frequently contains bunches of acicular raphides (oxalate of lime ?), starch-globules, and many of the faint yellow irregular bodies just mentioned, all of which, except the raphides, appear to have accidentally dropt into it from the mucus-layer.

Thus we have the internode of Nitella complete, and we have only to conceive the mucus-layer moving round the axial fluid and propelling it and its particles in the same direction (by the projections on its surface), to obtain a true idea of the motion which takes place within the cell of the Characeæ.

Let us now follow the passage of the cell-contents into gonidia.

All are aware, that in the freshwater Algae commonly called Confervæ, the formation of the spore is preceded by a breaking up or displacement of the cell-contents, after which a condensation and re-arrangement of them takes place, and they are then invested with a capsule which remains entire, until the time arrives for the spore thus formed to germinate. Now, under

certain circumstances, which appear to be the approaching dissolution or death of the cell-wall, a similar process takes place in the cells of the Characea, and following this from the beginning, we find, that it first commences with a cessation of the circulation, after which the lines of green disks forming the green layer become displaced, and, as if obeying a still continued but inappreciable movement of the mucus-layer, they roll themselves up into lines which assume a more or less irregular arrangement across the internode, or into groups of different sizes, more or less connected by narrow lines of mucus and single disks, so as to present an areolar structure in contact with the inner surface of the cell-wall. The next stage is the separation of the disks into still more distinct groups, which, having become more circumscribed and circular, leave the cell-wall and evince a certain amount of polymorphism and locomotion. The cavity of the internode, hitherto rendered turbid by the breaking up of the green layer, now clears off and becomes transparent, save where the circular masses, which have changed from their original green into a brownish-green yellow colour, intercept the light. After a day or two, but the time seems to vary,—the green disks become entirely brown, and the group, assuming a more circumscribed and circular form, shows that it is surrounded by a transparent globular cell; this we shall henceforth call the gonidial cell. A new substance, consisting of a bluish semitransparent mucus, more or less charged with minute granules (from which its colour appears to be derived), and refractive globules of a faint yellowish-green and sapphire-blue colour, makes its appearance in different parts of the brown mass, or to one side of it, and afterwards, becoming botryoidal or mulberryshaped, separates into gonidia. The brown chlorophyll with the other effete contents then shrinks up into a structureless, homogeneous, more or less defined, circular nucleus, of a dark brown colour, and the cell, frequently projecting on one side in a conical form, bursts at the apex and gives exit to the gonidia.

The gonidia are globular, ovate or spindle-shaped, and of a light blue colour. They average 1-4300th of an inch in diameter, and contain, together with, the blue substance mentioned, more or less also of the refractive globules, and a transparent vesicle. Each gonidium is provided with one or two cilia according to its form, that is to say, the globular ones present one and the spindle-shaped two, which may be perceived while they are yet grouped or separate in the transparent gonidial cell, where they already exhibit a certain amount of polymorphism. Shortly after they have become free in the internode, the wall of the latter gives way and they pass into the water, where, for a certain time, they remain so active, that it is almost impossible to

describe their form; but here and there, that which I have stated may be seen in those which are less active in their movements than the rest.

There are, however, certain peculiarities about the elongated and spindle-shaped gonidia which it is desirable to notice, viz. that one cilium appears longer than the other, and that while the short one floats almost motionless backwards, so as to appear as if it proceeded from the posterior extremity, the latter, for the most part, keeps up a constant whipping-movement in front which frequently renders it imperceptible. Both cilia however appear to be of the same length, if we add the length of the gonidium to the one which floats behind, and which is concealed in the first part of its course by lying in contact with or underneath the body of the gonidium; both also appear to be occasionally brought together anteriorly. There is also frequently a kind of proboscis extended from the rostrum or beak of the gonidium which moves incessantly, both in the ovate and spindle-shaped forms, and seems to have a suctorial extremity by means of which it fixes the gonidium to the glass, while the floating cilium also appears to be provided with a similar power and to exert it for detaching the gonidium, when so fixed, by pulling it backwards, which it does with a peculiar jerk; when this little proboscis has been present, I do not think I have ever observed the anterior or active cilium.

After a while, perhaps some hours, the gonidia become stationary, and while they appear to be fixed by the proboscis mentioned, the long cilium floats motionless, or presents a languid kind of whip-like undulation; the latter then disappears, and a day or two after, the gonidia both small and great, for there are many sizes, as will presently be mentioned, are seen creeping about the watch-glass (into which they were transferred for observation) under as active polymorphism as any amabous cell could present; diffluent, digitated, and in the form of that beautifully radiated figure called Actinophrys Sol (Ehr.). They also now present the "contracting vesicle," as well as other vesicles, which do not appear to alter their dimensions, but vary, like the former, in distinctness with their change of position and the varying form of the gonidium. After a few days' existence in this state their polymorphism becomes very sluggish, they remain for some time under a slowly changing rhizopodous figure, which is more or less common to all, and then disappear.

Whether they germinate or not, I have not been able to determine.

Lastly, a development of transparent mucus which becomes filled with vibrios and Bacterium termo (Dujar.), immediately follows the elimination of the gonidia both in the gonidial cell

and in the internode, presenting itself in a mass or branched organized form, more particularly at those parts of the latter which have become ruptured. This also disappears after a few days, and a thorough dissolution of the internode and its original contents seems thus to be completed.

We have now then seen that a breaking up or displacement of the green layer, a grouping of its green disks, the investment of the groups with a mucus-covering, their complete separation, their endowment with a certain amount of polymorphism and locomotion, the turning brown of the chlorophyll, and the passage of the investing mucus into a globular transparent cell, precede the evolution of the gonidial substance and its subsequent selfdivision into gonidia; let us now see if there be anything else within the gonidial cell during the time this process is going on.

It will be remembered, that a great number of discoidal bodies exist in the mucus-layer of the internode, and that many of these are loose, while others are fixed singly in the wall of a transparent vesicle; now (apparently under an arrest of development), a gonidial cell frequently presents itself, in which a single disk, with or without its vesicles, precisely similar to one of these, is seen within or to one side of the brown, chlorophyll, surrounded by a shrivelled, crenulated membrane, appearing, from its irregular mulberry-like surface, want of motion, dingy colour, and absence of refractive granules, as if it were the remains of the gonidial substance which had perished from some cause or other just before it began to separate into gonidia. Moreover, in many of the disks, the nucleus is not only seen to be separated from its capsule by an annular interspace, but its granules have become larger and more distinct, and an irregular cavity like a contracted vesicle appears to exist in its centre. When the circular disk is enclosed within the brown matter, it may be rendered more evident by the addition of alcohol, which extracts the colouring matter completely, while the application of iodine deepens the colour of the brown matter, and gives a dark brown tint only to the nucleus of the disk.

That the gonidial cells should contain a nucleus within the brown matter is easily conceived, for on truncating a young internode, sometimes, in the way which I have mentioned, particularly where the green layer is soft and previously disturbed, the whole of the contents rush out together, and all the vesicles, both large and small, become surrounded with green disks, which, under these circumstances, present a similar appearance to that which is witnessed in the internode when the green layer has been broken up and its green disks separated into groups, preparatory to the development of the gonidia.

But many of the gonidial cells are too small to contain the circular disk, and we have still to account for the disappearance of the irregular bodies of a faint yellow colour and granular structure, which appear to form a much larger proportion of the cell-contents than the circular disks.

As the smallest of the gonidial cells, which may appear in the internode, is but a very little larger than the single gonidium. which it contains, and others are sometimes present which may contain fifty or more, it is obvious, that although the latter may also contain a circular disk, there is no room for it in the former. Nevertheless, the small gonidial cells containing one, two, and three gonidia, as they vary in size from the 4300th to the 2150th of an inch in diameter, are provided with a body precisely similar to the irregular ones mentioned, which is not in the interior, but attached to or imbedded in their cell-wall, and with the latter seems to comprise all the elements of which the small gonidial cells are composed. In this way then we can account for the disappearance of the irregular bodies.

It is also not unusual to see in the older internodes, when their contents are passing into gonidial cells, a few of the green disks in situ, as well as loose in the cavity, and their disappearance also calls for explanation, which would be difficult, if we did not frequently see some of them actually in situ, under the form of gonidia, among the other green disks which have not passed into this state or departed from their original linear arrangement. Whether the gonidium has here taken the place of the green disk, or whether it has been developed in its transparent vesicle, I have not been able to determine; but in an old internode, where the basal structure of the green layer has become more or less hardened, the remains of the transparent vesicles may occasionally be seen in their original position, while the green disks have disappeared. In the fan-shaped groups of cells too, which form part of the capsule of the globule, the red granules, which are equivalent to the green disks, may frequently be seen attached, like the irregular bodies, singly or in groups, to the periphery of small gonidial cells which contain one or more gonidia. This appears to be invariably the case at the dehiscence of the globule, while the absence of circulation in these cells from the commencement would indicate a corresponding scarcity of the mucus-layer and its contents. With the exception of the central cavities of the globule and nucule, I have not seen any kind of cell in the species of Chara and Nitella, which I have had under observation, that has not produced gonidia.

Lastly, we have to account for the genuine starch-globules and raphides, both of which may be seen lying among the gonidial