04 VEGETABLE ORGANS. terminated by a very minute dilatation or knob. The sting contains an acrid liquid,, which escapes when the little knob is broken off in wounding the skin, and produces the well-knoAvn irritation. By the side of the figure of the sting is represented the point of a fine needle (fig. 20), showing that the expression ' ' sharp as a needle " has no force when microscopic bodies are in question. Stomata ( an ^ ^ is represented magnified 150 times linear, or 150 dia- meters, as it is called. A knowledge of the number of times the object is magnified is of the greatest importance in making c 3 18 MAGNIFYING POWER. use of the drawings; for, without it, the observer will be unable to apply such a magnifying power of the microscope as will enable him to see the structural appearances figured in the drawings. The observer must also be acquainted with the magnifying powers of his microscope with the various object-glasses and eye-pieces. These are usually given when the instrument is purchased. Or they may be determined approximative^ thus : An ivory scale, with y^th of an inch engraved upon it, is placed on the stage, and viewed as an opake object, both eyes being kept open ; and the size of the image of one of the gradations is measured with compasses, upon the stage as seen with that eye which is not applied to the eye-piece. The number of T ^o tns f an inch contained in the measure obtained with the compasses represents the magnifying power. Thus, supposing the image of the T ^o tn f an mcn on the scale appears magnified to the length of 1 inch on the stage ; the magnifying power is 100 diameters, or 100 times linear. This proceeding is difficult to any one unaccustomed to the use of the microscope, yet by practice it becomes very easy. Other methods, which require the use of the camera lucida, are given in the Micrographic Dictionary. VEGETABLE ELEMENTS AND TISSUES. 19 CHAPTER III. VEGETABLE ELEMENTS AND TISSUES. WE may now enter upon the consideration of the microscopic structure of objects, beginning with those which are derived from the vegetable kingdom, as they are more easily procured and prepared for exa- mination than those belonging to the animal king- dom j moreover they are not so transparent, and hence are more readily distinguished under the micro- scope, which is of importance in the case of an un- practised observer. Cells. The elements of which all plants consist are cells. Cells, in their simplest condition, are microscopic, rounded, colourless, closed sacs or vesi- cles, resembling small bladders (Plate I. fig. 2), and consist of a thin, transparent, colourless, vegetable skin or membrane (a) called the cell-wall. The cells are well seen in a little of the pulp of an apple (fig. 2) , or in a section of almost any soft part of a plant. A high power is usually required to show them dis- tinctly, on account of their minute size. The outline of the cells is seen to be double, one line indicating the inner, the other the outer, surface of the cell-wall, the space between the two lines corresponding to the thickness of the cell-wall. In the pulp of the apple, the cells are loosely con- nected, and so retain their rounded form; but in most parts of plants, the cells become crowded and squeezed together, from their ordinary or normal expansion being limited in certain directions, so as mutually to alter each other's shapes. The sides then lose their originally rounded form and outline, becoming more or less straight (PL I. figs. 1 & 4), 20 VEGETABLE ELEMENTS AND TISSUES. the cells at the same time mostly adhering to each other, so as to be separated with difficulty. The forms thus produced are various and inter- esting, and have all received names by which they are distinguished. They are described in works on botany in two ways according to the outline (which is the most common, as this expresses the appear- ance usually presented in sections and on the surfaces of vegetable structures), or according to the entire or solid form, which it is often a difficult matter to determine. Cellular tissue. Cells aggregated thus form a tissue, which is called cellular tissue or parenchyma (Trapa, among, and ey^v/jLa, poured substance), be- cause it fills up the interstices of the other tissues of plants. In technical descriptions, the cell- structure is often left out of consideration ; and bodies composed of parenchymatous tissue are described as being reticu- lated or netted, because the united sides of the cell- walls appear as a network covering the surface. It must be understood that parenchymatous cells are such only as have the three dimensions of solidity (viz. the length, breadth, and depth) nearly equal. Intercellular passages. The observer will not have examined many sections of cellular tissue,, without noticing certain irregular black lines running be- tween the cells, as in a piece of a Geranium- (Pelar- gonium-) leaf (PL I. fig. 1). These lines arise from the existence of passages between the cells, contain- ing air; and they are called intercellular passages. By gently warming a section containing them in water over a spirit-lamp, or by moistening the section with a drop of spirit, the passages will be filled up with the liquid, so as to become transparent. When the intervals between the cells are larger and broader, they are called intercellular spaces. So far, cells have been considered simply in regard CHLOROPHYLL. 21 to their form, as vesicles, either rounded or altered in shape by mutual pressure. We have now to notice the matters contained within the cells, or the cell- contents. Cell-contents. In most cells, especially when young, a minute, rounded, colourless body may be seen, either in the middle or on one side, called the nucleus ; this is very distinct in a cell of the pulp of an apple (PI. I. fig. 2 b). And within this nucleus is often to be seen another smaller body, frequently appearing as a mere dot, called the nucle'olus. The nucleus is imbedded in a soft substance, which fills up the entire cell (PI. I. fig. 2 c) ; this is the pro'toplasm (TT/JCOTO?, first, TrXaer^a, formative sub- stance) . As it is very transparent, it is readily over- looked ; but it may usually be shown distinctly by adding a little glycerine to the edge of the cover with a glass rod, when it contracts and separates from the cell-walls, as in the lower cell of fig. 2. The proto- plasm in some cells is semisolid and of uniform con- sistence, while in others it is liquid in the centre, the outer portion being somewhat firmer and immedi- ately in contact with the cell-wall. In the latter case, it forms an inner cell to the cell-wall, and is called the primordial utricle. The terms "proto- plasm " and ' ' primordial utricle " are, however, used by some authors synonymously. The protoplasm is the essential part of the cell, and it forms or secretes the cell-wall upon its outer surface in the process of formation of the cell con- sidered as a whole. It is also of different chemical composition from the cell- wall, being allied in this respect to animal matter. Chlorophyll (^Xwpo?, green ; (biiXXov, leaf) . On examining a section of any green part of a plant, as the green substance of a Geranium- (Pelargonium-) leaf, it will be seen that the green colour does not arise from the whole substance being coloured, as 22 VEGETABLE ELEMENTS AND TISSUES. appears to be the case to the naked eye, but from the presence of little grains or granules of a green colour- ing-matter in the protoplasm of the cells. This green matter is called chlorophyll. If the cells be crushed, the granules will escape, and can be examined in the separate state. Chlorophyll is most abundant in those parts of plants which are exposed to the light. Starch. In many cells of plants, particularly those which have attained their full growth, other granules, larger than those of chlorophyll, and colourless, are met with ; these are the starch- granules (PL I. fig. 3). They are usually rounded or oblong, and exhibit on the surface a number of rings, one within the other, or concentric, as it is called. In the centre of the in- nermost ring is a black dot or streak, arising from the presence of a little pit or furrow, and called the hilum. The starch- grains may be readily seen within cells in a thin section of a potato (PL I. fig. 4) ; here they are very numerous, and larger than in most other plants. A separate grain is represented in fig. 3. The appearance of rings in the separate grains arises from the starch -granules being composed of numerous concentric coats or layers, like those of an onion. A very simple and striking method of determining whether any granule is composed of starch or not, consists in adding to it, when placed in water on a slide, a drop of solution of iodine. As soon as this touches the granule, it assumes a beautiful purple . colour, the depth of tint depending upon the quantity of the iodine-solution ; if this be very considerable, the granule appears almost black. The section of potato forms a very interesting object when mois- tened with the iodine- solution, the starch-granules becoming beautifully coloured, whilst the cell-wall re- mains colourless, and the protoplasm becomes yellow. The form of the starch-granules differs in differ- STARCH. 23 ent plants, so that the kind of plant from which starch has been derived may be distinguished by attention to the size, form, and structure of its starch- granules. Thus, the granules represented in PL I. fig. 3, which it will be noticed are all drawn under the same power, are derived from different plants, a being those of wheat-flour, in which the hilum is obscure, and the rings faint ; b is a granule of West Indian arrowroot, in which the hilum forms a trans- verse crack ; c is a granule of potato-starch, in which the hilum is a dot, and the rings are very distinct; d re- presents the compound granules of the oat, the separate granules being figured below ; e is a granule of lentil- starch, with its long dark hilum and elegant oval concentric rings ; and / represents a compound and separate granule of rice-starch. It will be noticed that the granules of oat- and rice- starch are angular, as it is called. The knowledge of the peculiar forms of the starch- granules is important in a practical point of view, for it enables us to recognize them when mixed as an adulteration with other substances, and also to distin- guish the different kinds of starch from each other. Thus table-mustard, as it is called, is principally com- posed of the cheaper wheat- or pea-flour, which is easily recognized by the structure of the starch-grains. Arrowroot is considerably dearer than potato-starch ; hence in trade the latter is fraudulently sold for the former, the adulteration being detected with diffi- culty by the eye, but easily under the microscope. Again, rice is largely mixed with wheat-flour, as it makes inferior flour into very white bread ; and this may also be readily detected under the microscope. The reader can now understand how valuable the microscope is in detecting adulterations, with a know- ledge of the various forms and structures of sub- stances, especially with the aid of a few chemical tests. 24 VEGETABLE ELEMENTS AND TISSUES. Starch- grains are altered by boiling in water, be- coming swollen and often changed into curious forms, the rings becoming faint or disappearing. If a piece of boiled potato be examined, the starch- granules will seem to have vanished from the cells, which are swollen and covered with an irregular kind of net- work. The network consists of parts of the proto- plasm situated in the interstices of the starch-granules, and solidified or coagulated by the heat. On crush- ing the cells by pressing upon the cover, the starch- granules will escape, swollen and partly fused toge- ther ; but they may easily be recognized as consist- ing of starch by the iodine test. The granules of " tons les mois " starch are par- ticularly well adapted for showing the concentric rings, the granules being about twice as large as those of the potato. Starch-granules are best examined in water; and a small quantity only of the starch must be placed on the slide, if the structure of the granules is to be seen clearly. They may be mounted in glycerine, although this makes them very transparent. To those who possess a polariscope, starch-gra- nules are particularly interesting, as they exhibit a black cross, and, with a plate of selenite laid beneath the slide, a beautiful play of colours. In addition to the starch and chlorophyll, the cells of plants contain other matters, as gum, sugar, &c. ; but as they are dissolved in the cell-liquid, they are not visible. In the cells of certain plants, how- ever, spherical globules, with light centres and black outlines, will be met with : these consist of oil. Raph'ides. Lastly, occurring in the cells of plants, especially such as are soft and juicy (succulent), will be found minute, hard, colourless crystals, called r aphides (pa$>l<$ } a needle) . These are most frequently needle-like or acicular (acus, a needle), but sometimes prismatic or rod-like with flat sides; they are also WOODY TISSUE. 25 not unfrequently grouped into little tufts. They may be readily found in a piece of the stem of garden-rhubarb (PL I. fig. 5 a), or of the common balsam. Porous and spiral cells. The walls of the cells of cellular tissue are sometimes covered with little dots (PL I. fig. 11 a), or slit-like markings ; the cells are then called porous cells. A specimen of them may be obtained from a section of the pith of the elder (Sambucus nigra). Sometimes cells exhibit the appearance of a spiral line marking their walls, as if a little bell-spring were coiled up in them (PL III. fig. 2 a). These are called spiral cells, or spiral fibrous cells, and the tissue formed by them is called fibro-cellular tissue. We now leave the cells of ordinary cellular tissue, to examine those in which the dimension of length predominates, so that they form tubular cells ; and first of those required to possess strength and firm- ness, combined with flexibility. These qualities are met with in the cells constituting Woody tissue. Of this there are two forms, called respectively wood-cells and woody fibres. The wood-cells are moderately long, more or less tapering and overlapping at the ends ; and the cell- walls are thickened, so as to possess considerable firmness. These cells are found in the wood of stems, as in the white woody portion of an ash stick, that of a lime-tree, the stem of a Chrysanthemum, &c. (PL I. fig. 6) . They are closely packed, and the tissue formed by their union is called prosenchyma (TT/OO?, close, ey^y^a, tissue). In the other kind of woody tissue the cells are very long and slender, strong, yet flexible, gradually tapering at the ends, where they overlap each other ; and they have thick walls, so that, when divided trans- versely, the cavity appears almost filled up (PL I. figs. 5 d, 9, & 7 b) . This tissue is called woody fibre 26 VEGETABLE ELEMENTS AND TISSUES. or pleurenchyma (jr\evpa, rib, e^vfjua), from its strength. The walls of the cells of woody tissue are often covered with dots, either simple or with an inner dot (PL I. fig. 6 b, fig. 11 b), or with streaks (PL I. fig. 6 a) or with a spirai fibre (fig. 11 b, c), either alone or with dots also. This tissue is of great importance in plants, from its strength and flexibility; it forms a considerable part of the veins of leaves, the inner bark (liber), and of the wood of the stems of trees. It is also very useful to man : for it constitutes hemp, of which rope and string are made ; flax, of which linen is made ; cocoa- nut fibre ; bast, used by gardeners for tying up plants, which is the inner bark of the lime ; and jute, which is the inner bark of an Indian lime-tree. In the white woody part of the stems of trees be- longing to the fir-order (Conif erse), as a piece of deal or pine, which is mainly composed of wood- (prosen- chymatous) cells, the cells exhibit rows of minute circular markings (PL I. fig. 10). These were for- merly supposed to be solid bodies or glands ; hence the tissue is still sometimes called glandular. Within the outer ring of each marking is an inner central dot, or sometimes an oblique streak. The side view of the cells (PL I. fig. 8 a), which is seen in a tangential section, shows that the markings are minute pits, each being opposite to one of an adjacent cell, and sunk inwards towards the centre of the cell, the inner dot or streak being a thinner portion of the cell-wall. This glandular tissue of the Coniferse is interesting as forming a test-object for the defining power of the microscope, which should show the two rings sharply and free from colour ; the section of the wood should be examined as a dry transparent object. The difference between the woody fibre and the wood- cells of coniferous wood may also be seen well in a piece of deal, as cut up for fire-wood. If the end of a stick of DUCTS. 27 this be examined with the naked eye, parts of brown rings will be seen traversing the whiter portion of the wood. These brown rings consist of woody fibre ; the white portion of wood- cells. On making a very thin transverse section, the interior of the woody fibres is seen to be almost entirely filled up (PI. I. fig. 7 b), while the cavity of the wood- cells is much more open (PL I. fig. 7 a) the former also contain globules of turpentine. It must be remarked here that some botanical authors include both forms of woody tissue under the term prosenchyma. But, as we shall see hereafter, the form of the prosenchymatous cells being some- times used as a character for distinguishing the cells of leaves, to which the term pleurenchymatous cells would be inapplicable, the above distinction will be found important. Vessels } vascular tissue. In the next form of tubu- lar cells, these are broader and softer than the cells of woody tissue, thin-walled, and the ends pointed ; and their walls exhibit spiral or ring-like markings, or rows of dots (PI. I. fig. 5 c, e, b) , indicating the existence of one or more spiral fibres or rings. When the vessels contain spiral fibres, they are called spiral vessels (PL I. fig. 5 c) ; when they contain ring-shaped por- tions of fibre, they are called annular (an'nulus, a ring) vessels (PL I. fig. 5 e) ; and when the spaces between the fibres are partly filled up, leaving only dots, the deposit forming a kind of network, we have a reticulated (rete, a net) vessel (PL I. fig. 5 b). This tissue can easily be obtained from a piece of cooked rhubarb, the stem of a balsam, or from any soft- stemmed plant. Vessels very frequently contain air. Ducts. The tubular cells forming ducts (PL I. figs. 5 b, lie) are large, more or less flattened or blunt at the ends (truncated) ; and the cell-membrane at first closing the ends is often removed or absorbed, so that the ducts communicate with each other, to allow 28 VEGETABLE ELEMENTS AND TISSUES. of the free passage of the sap through them. Their walls are invariably covered with markings, consisting of either simple or bordered dots, resembling those met with in the preceding forms of tissue. The ducts are often easily recognizable with the naked eye, in transverse sections of stems, by the large pores which they form in the wood. These may be well seen in a section of a piece of cane. The tissue com- posed of dotted ducts is called bothren'chyma (ftoOpos, pit) ; but the term is principally applied to those ducts in which the dots are simple, i. e. have no inner dot. The structure of the above forms of tissue may be best understood in relation to their development. It has been stated that the essential part of the cell is the protoplasm. As cells grow older, new matter is deposited by the protoplasm upon the inner surface of the cell-wall, either to a small extent, evenly and uniformly, as in ordinary parenchyma, or unevenly, in the form of spiral layers, forming fibres or bands, leaving bare spaces, where the original cell-wall exists alone. The matter thus deposited is called secondary deposit, the original cell- wall being the primary de- posit. When the secondary deposit covers the in- terior of the cells except at certain slit-like spaces, we have the appearance figured in PL I. fig. 6 a). When the deposit forms a spiral fibre, or a series of rings, we have the spiral or annular vessel or duct. And when the interspaces between the coils of a close spiral fibre are filled up except at certain spots, we have the dotted or reticulated vessel or duct. In many instances, these deposits are present to- gether : thus, sometimes the outermost deposit leaves rounded pits or dots, while an inner portion forms a spiral fibre (PL I. fig. 11 b) or one layer leaves simple rounded pits, while the other leaves smaller slits or dots placed opposite the former (PL I. fig. 6 b). In some cells the cavity is almost entirely filled up CELL-FORMATION. 29 by secondary deposit, which, leaves minute canals radiating from a small cavity in the centre to the cir- cumference, as seen in the transverse section of a plum-stone (PL I. fig. 56) ; here the canals appear as dark lines. In others, again, the secondary deposit forms several distinct layers, leaving channels very similar to those of the last ; an example is met with in the gritty tissue of the pulp of a pear. The obvious use of the pits and channels in the above tissues is to preserve the permeability of the walls of the elements, which would be destroyed if the walls were equally thickened all over. Cell-formation. New cells are formed by the division of old or parent cells. The actual process of division is difficult to observe, as it requires pro- longed observation ; but cells are often met with in all stages of division, of which some instances will be pointed out hereafter. The cell-division takes place in two ways, either according to the endo- genous (evBov, within, ryevvaw, to produce), or the exogenous (efo>, outside, yevvda)) method. The manner in which the division takes place in the former is this -. At first a slight indentation or con- striction of the protoplasm occurs at the line of division ; this deepens until the protoplasm is com- pletely divided. The freshly divided surfaces then become coated with a new portion of cell-wall, so as to make two or more new cells, which either remain in contact or separate from each other. In some cases, the divided portions of protoplasm become coated all over with new cell- walls. In the exogenous process, a portion of the proto- plasm protrudes from the surface of the cell, carry- ing the cell-wall before it, so as to form a little bud- like body ; this is next cut off at its point of junction with the parent-cell, and coated, as in the first case, with a new cell-wall, so as to form a new cell. Preparation. In examining the vegetable ele- D3 30 VEGETABLE ELEMENTS AND TISSUES. ments and tissues, very thin sections must be made with a razor or thin sharp knife ; these are then to be placed in a little water on a slide. As the struc- tures are all minute, the distinctness with which they are seen will mainly depend upon the proper thin- ness of the sections. When sections of dry stems are to be examined, the black margins of the air- bubbles contained in the cells often render the struc- ture indistinct ; these must therefore be displaced by first wetting the tissue with methylated alcohol, and then adding water to it in a watch-glass or on a slide ; or the tissue may be soaked in warm water for some hours : and this is mostly requisite in preparing thin sections of dry tissues. Attention must also be paid to the manner in which the section is made, or the direction in which the portion of the plant is cut. There are three im- portant directions which must be distinguished, pro- ducing transverse, longitudinal, and tangential sec- tions. If the cuts be made across the length of a stem, for instance, the section is called transverse. If the cuts be made in the direction of the length, through the centre, the section is longitudinal ; and if the cuts are made in a direction parallel to a line running down the centre of the stem, but nearer its margin, it is a tangential section. It is scarcely necessary to mention that an oblique section is intermediate between a transverse and a longitudinal section. EPIDERMIS. 31 CHAPTER IV. VEGETABLE ORGANS. THE vegetable elements and tissues which have been described form, either separately or by their com- bination in various ways, the organs of plants. To these we shall now pass, and consider the structure of the principal organs of the members of the vege- table kingdom. Leaves. Leaves in their simplest form consist of a single sheet or layer of parenchymatous cells or cellular tissue, an example of which may be found in almost any moss (PI. III. fig. 30). The granules of chlorophyll will often be very distinctly seen in these cells. The first addition to this form of leaf is a row or two of prosenchymatous cells running longi- tudinally down the middle of the leaf, so as to form a rudimentary vein or nerve. In other and more highly developed leaves, the layers of cells are nume- rous, and traversed by bundles of wood- cells, vessels, and ducts (fibro-vascular tissue), forming the veins, the entire surface being covered with a skin or mem- brane, called the epidermis. Epider'mis (eVl, upon, Se/^a, skin). This mem- brane is composed of one or more layers of colourless, closely packed cells (PI. I. figs. 13&28), the colour it occasionally exhibits usually arising from some of the underlying cells of the leaf being seen through it, or remaining adherent to it when stripped from the leaf. It is easily separated, by making a cut in a soft leaf, and peeling it off with a fine pair of for- ceps, or by soaking a leaf for some time in water and then stripping it off. It must be remarked that 32 VEGETABLE ORGANS. the epidermis covers not only the leaves, but every part of the plant. Hairs. Arising from the epidermis are the hairs of plants. These are thread-like or filamentous pro- longations of the epidermis beyond the surface of the leaf (PI. I. fig. 12), consisting of cells arranged end to end. They are often branched, sometimes star- shaped (stellate) (fig. 28), and present great varieties in form, as shown in the figures, the plants from which these were drawn being mentioned in the De- scription of the Plates. Sometimes hairs terminate in a little head (PL I. figs. 12 c, d, e\ the cell or cells composing which secrete a colouring or a viscid sub- stance ; they are then termed glandular. The hairs of plants are particularly interesting to the micro- scopic observer, not only on account of their curious forms, but in connexion with the remarkable pheno- menon of the circulation of the cell-contents, or rotation, as it is called, observable in them. This is difficult to be perceived by any one unaccustomed to microscopic observation, because the particles by which the motion of the cell-contents becomes evi- dent are exceedingly minute ; but practice in the use of a high power will overcome this difficulty. The hairs which exhibit the phenomenon best are those of the American Spiderwort (Tradescan'tia Virgin' ica\ which is to be found in every garden. It may, per- haps be recognized thus : The plant is about a foot and a half high; the leaves are sword-shaped and channelled, and the flowers are purple, in heads, and \\ inch in diameter. The hairs are attached to the sides of the stamens, towards the lower part or base. The stamens should be carefully picked off with for- ceps, and placed on a slide in a drop of water ; the hairs should then be separated with the mounted needles, and a cover applied. Under a low power, the hairs are seen to be beaded or momTiform (momle, a necklace), and of a fine purple colour (PL I. fig. 22). STINGS. 33 On applying a high power, as the J-inch, the indivi- dual cells will come distinctly into view, and the nucleus will be seen very clearly as a roundish granu- lar mass (PI. I. fig. 23 a). On carefully examining the cell-contents, delicate lines will be observed radi- ating irregularly from the nucleus, some passing to the top of the cells, while others run towards its base, as in the figure ; and on very close inspection, the portions of protoplasm of which these lines con- sist, will be found to move slowly and steadily, the motion becoming perceptible by means of the minute granules of which the protoplasm consists. The currents return at the ends of the cell, there being no passage of the contents of one cell into the cavity of either of those adjacent. During this examina- tion, it will be noticed that the surface of the cell- wall is striated with fine wrinkles. It may be remarked that the hairs should be taken from flowers which have only just opened ; for this curious and inexplicable rotation is connected with the growth of the cell ; and when this has at- tained maturity, it no longer occurs. The phenome- non may be observed in many other hairs of plants, as those of common groundsel (Senecio vulgdris) (PL I. fig. 12 a, b), and in the cells of the leaves of some water-plants; but I must refer to the article "Rotation" in the Dictionary for further information. The most important variety of hair is that derived from the Cotton-plant (a kind of Mallow), and form- ing the cotton of commerce. These hairs spring from the epidermis of the seeds. The cells com- posing it are very long and soft, becoming flaccid and easily bent when dry (PI. IX. fig. 13). Stings. Stinging hairs or stings may be well illustrated by reference to the common large nettle (Urtica dioica). In this plant they consist of a thick- walled cell, bulbous at the base, which is imbedded in the epidermis (PL I. fig. 21), the pointed end being 34 VEGETABLE ORGANS. terminated by a very minute dilatation or knob. The sting contains an acrid liquid, which escapes when the little knob is broken off in wounding the skin, and produces the well-known irritation. By the side of the figure of the sting is represented the point of a fine needle (fig. 20), showing that the expression " sharp as a needle " has no force when microscopic bodies are in question. Stomata (oTo^a, mouth) . On viewing a strip of epidermis, the observer will be sure to notice certain oval or roundish bodies (PL I. fig. 13 a), composed of mostly two kidney- shaped cells in apposition but leaving a chink between them ; these are the stomata. They communicate beneath with the intercellular pas- sages, of which they may be considered the mouths ; and by their agency a direct communication is esta- blished between these passages and the air. The two cells which guard the orifice are termed the " guard cells." Stomata are most numerous on the under surface of leaves; they are entirely absent in plants growing under water, and in most of the lower plants. In many of the stomata, viewed in the ordinary way, the air situated between the guard cells is indicated by the black spot or dot present ; but after a time, or by the application of a gentle heat to the slide, the air becomes displaced by the water, and their structure becomes very distinct. In certain plants, the epidermis is imbued with flint or sil'ica ; so that even when burnt to an ash the stomata are still quite distinct. Examples of this may be found in the stalk or culm of grasses, as in straw, the shining epidermis of which is siliceous ; or the epidermis of canes. Among the lower plants, this peculiarity is especially curious in the species of Equisetum, or mares'- tails. The manner in which the veins of leaves are ar- ranged is worthy of special attention, as it forms one STEMS. 35 of the characters by which the two leading divisions of the Vegetable Kingdom are characterized. Thus in one of these divisions the veins are branched, so as to form a network throughout the leaf; the plants with these netted veins, to which belong our trees, shrubs, and most herbs, are the Dicotyledons, or Ex'ogens ; while in the second division, the veins run parallel to each other, being little or not at all branched, and not forming a network. The plants with parallel veins, among which are our grasses, lilies, &c., are the Monocotyledons or En'dogens. Stems. In the stems of plants, the tissues are arranged round a centre; otherwise, in the sim- pler and lower plants, they agree in structure with leaves, the centre being occupied by some element of fib ro- vascular tissue, as simple wood-cells, a few vessels or ducts. In the higher or flowering plants, the stem exists in two distinct forms, corresponding to the differences above noticed in the arrangement of the veins of the leaves ; these must be considered separately. In the Dicotyledons or Eocogens (PL I. fig. 36), the centre of the stem, in a transverse section, is seen to be occupied by the pith or medulla, which is repre- sented in the figure by the innermost circle. Imme- diately outside and around this is a narrow ring, in- dicating the section of a sheath to the pith, and called the medullary sheath. Next comes a broad ring of wood of the first year's growth (fig. 36 a], traversed, from the pith to the bark, by wedge-shaped paler rays, termed the medullary rays. Outside the first year's wood is the newer and paler wood of the second year (b) ; and so on, a new ring of wood being added outside the preceding layer for each year of growth of the stem. On the outer side of the wood is the inner bark or liber (fig. 36 c) ; and outside this is the spongy outer bark (d) , covered by its epidermis. 36 VEGETABLE ORGANS. These structures are of different composition, as may be best seen in longitudinal sections. The pith and the medullary rays consist of cellular tissue, the cells being mostly rounded in the former, and more closely pressed together and squarish in the latter. The medullary sheath consists of vascular tissue ; and the wood, of wood- cells traversed longitudinally by bundles of vascular tissue and ducts, the latter being larger and more distinct towards its outer boundary. The liber is composed of woody fibre, and the outer bark of cellular tissue. The new woody matter being deposited outside the old, between the bark and the previously formed layer, gives origin to the term exogen (efo), outside, yevvaw, to produce). These structures may be ex- amined in the section of a branch of the lime-tree or lilac. In the Monocotyledons or Endogens (PL I. fig. 37), there is no distinct bark, nor pith, nor medullary rays the entire stem consisting of cellular tissue with isolated bundles of fibro-vascular tissue scat- tered through it. Moreover the new substance is added to the centre of the stem, or within the old ; hence the term endogen (eVcW, within, ryevvdw) . A section of a piece of cane will exhibit this structure. To examine the structure of stems, sections must be made in various directions. The relative position of the component parts of a stem are best seen in a transverse section ; but the structure of the tissues is most evident in longitudinal sections, and under the higher powers. The annual rings of the Exogens are best observed in transversely sawn-off pieces of per- fectly dry stems, which have been polished with sand- paper, and varnished with spirit varnish. Roots. The structure of roots is very similar to that of stems ; there is, however, no distinct pith, nor are there stomata on the epidermis ; and the vessels are replaced by ducts. The very fine rootlets or FLOWERS. 37 radicles of water-plants often show the rotation of the protoplasm very distinctly. Flowers. The various parts of flowers, being each a modified leaf, present the same general structure as the latter. As the reader may not be acquainted with the names of these parts or organs in the higher plants, and as we shall have to compare them with their representatives in the lower forms of vegetable life, it will be well briefly to indicate them. A com- mon and beautiful yet despised flower (PI. I. fig. 32) may serve for illustration ; this is chickweed (Stella- ria media) } which can be found everywhere. The outermost circle of flower-leaves, which forms a kind of cup to the rest of the flower (a), is the calyx ; the separate leaves being called the sepals. The row within this, in most flowers consisting of brilliantly coloured pieces, forms the corolla (b) ; the individual pieces being the petals. When the two kinds are equally coloured, or not distinguishable, the whole is called the perianth, as in a tulip. When the seg- ments of the perianth are dry and chaffy, as in the flowers of grasses, the outermost are said to con- stitute the glumes, and the innermost the palea. Within the ring of petals are certain thread-like organs called stamens (c) ; and these consist of a filament (fig. 39 ), surmounted at the top or apex by the anther (fig. 39 b), which is usually coloured, and consists of two lobes. The anthers when ripe burst, and discharge a coloured dust; this is the pollen. Lastly, within the stamens is the central organ of the flower, the pistil, and sometimes there are several of them. The pistil consists of three parts, viz. a swollen base, the ovary (fig. 41 b), surmounted by a column or style (fig. 41 a), and which is crowned by a viscid and often hairy summit, the stigma (fig. 40*). In chickweed there are 3 styles. It must be remarked that, in the flowers of some plants, stamens alone are present, while others COn- 38 VEGETABLE ORGANS. tain pistils only, although most flowers contain both organs. When the stamens and pistils occur in sepa- rate flowers on the same plant, the plant is said to be monoecious (fwvos, single, ottfo?, family) ; when all the flowers of distinct plants contain either stamens only or pistils only, the plant is dioecious (St?, twice, ol/cos) ; and when the stamens and pistils occur together in all the flowers of the same plant, the plant is said to be hermaphrodite. These terms had their origin in the idea that the differences of plants in respect to these organs were analogous to those of the sexes in animals. All the parts of a flower have their special uses : thus the calyx and corolla protect the delicate organs en- closed by them, until they attain maturity. The petals also, by their brilliant colours, attract insects which feed upon or collect the honey of the flowers ; these at the same time conveying the pollen which adheres to their bodies from one flower to the stigma of another. The stamens and pistils are organs of fructification, it being essential for the fertilization of the flowers that the pollen should come into contact with the stigma. We will now consider some inter- esting points of structure in these organs. Petals. The petals often form most beautiful mi- croscopic objects, on account of the curious shape and structure of the cells of their epidermis, and the splendid tints of the colouring matters contained in them. As petals are mostly too thick to allow of the cells being distinctly seen in the entire state, a little cut should be made in them while gently stretched on the finger, and the epidermis carefully stripped off with forceps; the strip should then be laid on the slide in water as usual : in this way the curious patterns of the epidermic cells will become very dis- tinct. The petals of a red geranium (Pelargonium) may be used to illustrate them (PL I. fig. 24) . The structure may be best understood by reference to the epidermis of the leaf of a geranium (PL I. fig. 13), POLLEN. 39 in which the cells present wavy or undulate walls. In the petal (fig. 24), the walls are inflexed at tole- rably regular distances, so as to give rise to the ap- pearance of a row of teeth lining the cell. If the strip of petal be folded, so as to exhibit the side view, it will also be seen that the cells project outwards from the surface to form a bluntish point or papilla, or the petals are papillose as it is called ; and the surface of the membrane around the papillae is finely wrinkled, so as to present the appearance of very deli- cate radiating lines or striae. Intermediate degrees of this inflexion may be found in various flowers, between the slight condition seen in fig. 13 and the extreme state of fig. 24, as in the snapdragon (An- tirrhinum mdjus). Anthers. The cavities of the anthers are lined with fibro- cellular tissue, the fibres of which aid in discharging the pollen ; this may be seen by dissect- ing an anther of London pride (Saxifraga umbrosd], or of a wallflower (Cheiran 'thus cheiri] in water. It also exists in chickweed. Pollen. The pollen consists of minute grains called the pollen-granules. They may be viewed either in the dry state as opake objects, or when immersed in water as transparent objects. As it is often difficult to moisten them, they may be touched on the slide with a little spirit, and then a drop of water added. Their forms are very varied and curious, but they are difficult of observation from their minute size. They consist of one or more coloured cells, and these cells are remarkable for their surfaces exhibiting spines, networks, folds, and markings of various kinds. Thus in the primrose the pollen- granules are cylindrical, the surface being furrowed (PI. I. fig. 16) ; in the sun- flower the granules are spherical, and covered with tubercles surmounted by spines (fig. 17) ; in the garden convolvulus the surface of the spherical gra- nules is covered with an elegant network, in the E2 40 VEGETABLE ORGANS. meshes of which are also situated spines (fig. 18) ; and in the granules of chickweed the surface pre- sents pits, with minute tubercles in the centre (figs. 30 & 31). The pollen-granules are often con- siderably altered by immersion in water; so that, in judging of their structure when examined in water, the resulting alteration must be taken into account. When ripe pollen -granules have been immersed in water for a short time, one or more minute tubes will be seen protruding from their surface \ these are the pollen-tubes, and the granular protoplasm con- tained in them is called the fovilla. In the process of fertilization of the flower, the pollen-granules fall upon the viscid stigma ; the pollen-tubes are then protruded, and, passing down the intercellular spaces of the style (PL I. fig. 14), enter an aperture in the ovule or young seed, which is thus endowed with the power of growing into a new plant. The pollen- tubes are often very long, and they do not exist fully developed in the pollen-granules, but grow down the style, just as the little rootlet of a seed grows into the soil. The style of a crocus will serve for dissecting out with mounted needles the long and very slender pollen-tube (PL I. fig. 15). O'vary. The ovary by its growth and enlargement becomes the fruit. There are many interesting mi- croscopic structures to be found in fruits and the seeds they contain, a few of which may be noticed here. On examining the surface of the rind or pericarp epl, around, Kapiros, fruit) of an orange, little dots will be seen, paler than the rest of the surface. These are receptacles of secretion, or glands, containing the evaporable or volatile oil upon which the fragrance of the orange depends. They consist of loose cells, sur- rounding a central cavity, and are imbedded in the rind. OVARY. 41 Other receptacles of secretion, called mttcB (vitta, a band), occur in the wall (pericarp) of the fruit of the Umbelliferae, or Parsley Order of plants, and their arrangement forms characters for distinguishing the genera. They may be well seen in caraway- seeds ; for the caraway-plant is one of the Umbelliferse. It must be observed that a caraway " seed " is not really a seed, but consists of half the fruit ; for, on careful examination, one side of it will be found to be flat- tened, the flattening resulting from the mutual pres- sure of the two half-fruits at that part ; moreover the dried style exists at its summit. In the figure (PL I. fig. 19), the flattened part of the seed is next the ob- server. The seed has five evident longitudinal ridges, one at each corner or angle. The vittse are dark- coloured (fig. 19 a), and placed one between each pair of these ridges ; and they consist of long flattened spaces in the substance of the pericarp, with trans- verse markings, indicating internal cross partitions, In botanical works, the presence of five ridges, with single vittse in the intervals, is given as a character by which the half-fruits (carpels) of the caraway are to be distinguished. But on closely inspecting the flattened surface, another ridge is seen running down its middle ; so that the seed really has six ridges, one of which is smaller than the rest from the pressure of the other half. Hence the character of five ridges with single vittse is incorrect. The vittse contain the volatile oil to which the fragrance and pungency of the fruit is owing, although some of the oil exists also in the cells of the kernel or albumen, which forms the white and greater part of the seed. The skin of a reddish apple, peeled off in the manner described for petals, exhibits beautifully the red colouring matter of different tints in adjacent cells, while the pulp displays the cell-contents, as already mentioned. The latter may also be easily 42 VEGETABLE ORGANS. examined, from their large size, in most of the softer fruits, as that of the snowberry or the cucumber. As the ovary or fruit approaches maturity, the petals and stamens wither and fall off, the calyx often remaining, and being sometimes adherent to the ovary, at others free or unattached to it. Seeds. During the ripening of the fruit, the seeds contained within it are gradually becoming further developed. The seeds themselves are covered outside by a skin or coat called the testa (testa, a shell) . This is re- markable for frequently displaying various kinds of figured patterns, consisting of raised networks, ridges, little knobs or tubercles, &c. Examples of these may be found in the seeds of the poppy (PL I. fig. 27), mignonette (fig. 29), and chickweed (fig. 51). Some seeds are winged, as it is called, i. e. fur- nished with an extension of the testa beyond the margin of the seed. This not unfrequently consists of aggregated fibre-cells, the spiral fibre being very distinct, as in the seeds of Eccremocar pus scdber (PL I. fig. 52). In the seeds of another curious plant in this respect, viz. Collomia grandiflora, the fibre- cells are separate, so as to resemble hairs, and very mucilaginous, and in the dry seed are closely pressed to its surface. If a portion of the testa of these seeds, which can be procured at the seed-shops, be cut off, laid on a slide, a cover applied, and when the object is in focus, a drop of water be added, in a short time water softens the mucilaginous walls of the cells, the power of the spiral fibres comes into play, and the cells expand so as to form a very interesting object; the cells, in their expansion, apparently writhing like so many minute worms (PL I. fig. 35). The seed itself, which is contained within the testa or seed-coat, consists essentially of the young plant or embryo. This is composed of three parts, viz. the plumule (plumuktj a little feather), or the young stem ; SEEDS. 43 the racficle (radicula, a little root), or the young root ; and one or two, rarely more, imperfectly developed or rudimentary leaves, the cotyledons (fcoTv\rj&7rTpa, a cover), which is carried up as the cap- sule and its stalk grow, so as to be either entirely thrown off, or to remain covering a greater or less portion of the capsule, when this attains maturity. PLATE III. [PAGE 54.] MOSSES. Fig. 1. Sphagnum acutifolium, expanded leaf. 2. Sphagnum acutifolium, cells of leaf. 3. Spermatozoa of Polytrichum pili- ferum. 4. Sphagnum acutifolium. 5. Sphagnum acutifolium, capsule. 6. Gymnostomum truncatulum. 7. Gymnostomum truncatulum, leaf. 8. Gymnostomum truncatulum, cap- sule and operculum. 9. Gymnostomum truncatulum, spore. 10. Dicranum heteromallum. 11. Dicranum heteromallum. 12. Dicranum heteromallum, leaf. 13. Dicranum heteromallum, opercu- lum. 14. Dicranum heteromallum, calyptra. 15. Dicranum heteromallum, capsule; 15 a, peristome. 16. Torttda muralis. 17. Tortula muralis, leaf. 18. Tortula muralis, capsule : a, tooth of peristome. 19. Tortula muralis, operculum. 20. Tortula muralis, calyptra. 21. Tortula muralis, archegonia. 22. Polytrichum piliferum. 23. Polytrichum piliferum, leaf. 24. Polytrichum piliferum, calyptra. 26. Polytrichum piliferum, antheridial stems. Fig. 26. Polytrichum piliferum, single head. 27. Polytrichum piliferum, antheridia and paraphyses. 28. Funana hygrometrica. 29. Funaria hygrometrica. 30. Funaria hygrometrica, leaf. 31. Funaria hygrometrica, capsule ; a, operculum. 32. Funaria hygrometrica, atalk-like body. 33. Funaria hygrometrica, young ar- chegone. 34. Funaria hygrometrica, more ad- vanced archegone. 35. Funaria hygrometrica, section of young capsule. 36. Funaria hygrometrica, calyptra. 37. Funaria hygrometrica, antheridia. 38. Funaria hygrometrica, spores. 39. Funaria hygrometrica, annulus. 40. Funaria hygrometrica, archego- nia. 41. Funaria hygrometrica, antheridial head. 42. Funaria hygrometrica, peristome. 43. Hypnum rutabulum. 44. Hypnum rutabulum, leaf. 45. Hypnum rutabulum, capsule. 46. Hypnum rutabulum, spores. 47. Hypnum rutabulum, peristome. 48. Hypnum rutabulum, cells of leaf. 49. Bryum capillare. MOSSES. 55 The calyptra is either simply mitre-shaped, or mitri- form (PI. III. fig. 24), or it is half-cleft, or dimid'iate (figs. 14, 36) . When the capsule is ripe, the upper part usually separates at a circular horizontal line (fig. 8) as a kind of lid, which is called the oper- culum (operculum, a lid), and thus the spores are enabled to escape. The rim of the capsule, from which the operculum has separated, forms its mouth, and this often exhibits a fringe of teeth (figs. 15, 18, 31), arranged in one or more rows; sometimes the teeth are replaced by a membrane, or, again, both teeth and a membrane may be present. This mouth- fringe is iheperistome (-Trepl, around, oro/i-a, mouth). In many mosses, an elastic row or ring of cells is situated between the mouth of the capsule and its operculum, called the annulus (figs. 18 & 39) ; this, when the capsule is ripe, aids in throwing off the operculum. It is important to become acquainted with the structure and arrangement of these parts, as they form characters by which the families and genera of mosses are distinguished. The capsules of the mosses form very beautiful microscopic objects, especially those furnished with a toothed peristome. Most of the mosses produce their fructification in the winter and spring. The class of mosses is divided into two Orders, according to whether the fruit-stalk is terminal, i. e. arises from the end of the stem or its branches, or whether it is lateral, arising from the side of the stem. Those with the fruit-stalk terminal, or the end fruited (PL III. fig. 22), form the Ac'rocarpi (a/cpa, summit, /capTros, fruit) ; while those with the fruit-stalks lateral, or the side-fruited mosses (fig. 43), constitute the Pleurocarpi (irXevpa, side). The new shoots or young branches of the stems of mosses are termed innovations. 56 MOSSES. We will now examine a few common mosses more in detail, beginning with the ACROCARPI. Sphagnum acutifolium (PI. III. fig. 4) is found in pools or bogs, growing at the margins so as "to be partially immersed. In this moss, the upper branches are grouped into a head. The leaves are crowded, and overlapping or im'bricate (imbrex, a tile) on the elongate stems; they are egg-shaped (ovate) on the main stems (fig. 1), and narrower or ovate-lanceolate on the branches; they are nerveless, and finely toothed at the apex. The capsule (fig. 5) is roundish- ovate, without a peristome, and the operculum is flattened. The grouped arrangement of the upper branches renders the species of Sphagnum easily re- cognized. The structure of the leaves is also very peculiar and characteristic (fig. 2) . The cells of which they consist are of two kinds, one (fig. 2 a) being colourless, elongate, pointed, and containing a spiral fibre ; the other consisting of shorter and narrower obtuse cells, containing chlorophyll, and situated be- tween the former. In many of the former kind of cells, little round apertures exist on the under surface, and minute animals may sometimes be found im- prisoned in them. Another species of Sphagnum, S. obtusifolium, is common, and greatly resembles the above, but has shorter and thicker stems, and rounded-ovate, very concave, and obtuse leaves. Gymnos'tomum truncat'ulum (PL III. fig. 6) is a common little moss, found on banks and in fields and gardens. In this there is no peristome, although, in the young condition, a membrane extends more or less over the interior of the mouth of the capsule. The stem is slender, rigid, and simple, or but little branched. The calyptra is dimidiate ; the operculum is present (fig. 8), and terminates above in an oblique beak, or it is obliquely rostrate (ros'trum, a beak) as DICRANUM. 57 it is called. The leaves are obovate (fig. 7) or ovate with the broader part remote from the stem, and narrowed at the apex, where the nerve protrudes or is ex'current (excurro, to run out). The spores (fig. 9) are reddish brown and smooth. Dicrdnum heteromal' lum (PI. III. figs. 10 & 11) is probably the first moss the reader will meet with on banks and heaths in the early spring ; and it will be sure to be noticed on account of the bright green colour of the patches and the beautiful orange- brown capsules. In this moss the capsule is nodding (cer'nuous) (PI. III. fig. 15), and has a single peristome, con- sisting of sixteen equidistant teeth, each being deeply cut or cleft longitudinally (fig. 15 ), so that there are thirty- two teeth altogether; and these are marked with internal cross-bars, or transverse ridges. The calyptra is dimidiate (fig. 14) ; and the lid is furnished with a long oblique beak (fig. 15 b). The leaves are crowded, strongly nerved (fig. 12), lanceolate at the base, and very narrow towards the apex, which is toothed ; they are, moreover, curved^ and bent towards one side, or secund. Tor' tula murdlis (PI. III. fig. 16) may be found on the top of almost every wall and on waste ground. In this moss the peristome is single (fig. 18), con- sisting of thirty-two spirally twisted teeth, arranged in pairs. They are narrow and slender, and each is composed of two longitudinal portions (fig. 18 #), one of which is pale yellow, the other reddish brown, like the capsule, and both are fringed and covered with very minute papillae. The capsule (fig. 18) is oblong, the ring or annulus remaining for some time. The lid is conical (fig. 19), with a longish somewhat ob- lique beak, and the calyptra is dimidiate (fig. 20). The stems are very short; the leaves (fig. 17) are oblong, obtuse; the nerve strong, and projecting as a colourless spirally striated bristle. The bristles often 58 MOSSES. give the patches of the moss a hoary appearance on wall-tops. The margins of the leaves are folded back or recurved, giving them a peculiar thickened ap- pearance. The largest of our mosses are contained in the next genus, viz. Polyt'richum, some of them having the stems from 2 to 4 inches, or even more, in height ; they are common on heaths and in woods. Polyfrichum piliferum (PL III. fig. 22) is very common on open dry heaths. This moss has simple stems, with the leaves crowded on the lower part of those which are fertile or fruit-bearing. The fruit- stalk is terminal (acrocarpous) ; the capsule ovate, 4-sided or quadrangular, with a knob or struma (struma, a swelling) at the base, the lid having a short beak. The calyptra (fig. 24) is half-cleft (di- midiate) and very hairy. The peristome is single, and consists of sixty-four teeth. The leaves (fig. 23) are lanceolate, nearly upright, the margins folded inwards or inflexed ; and they end abruptly in a saw- edged or serrated hair-like point. Poly'trichum commune, which is also very common, is larger than the last species, and may easily be dis- tinguished by the curved and serrate leaves, which have no bristle-point. In the early spring, patches of both these mosses may be found, in which the stems are terminated by little rosettes (figs. 25 & 26) ; these will be referred to presently. Keeping still to the end-fruited or Acrocarpous mosses, we have next to mention Funaria hygro- met'rica (PL III. figs. 28, 29), which is readily dis- tinguished from most other mosses by the pale apple- green colour which it possesses before the capsule ripens. It is extremely common on walls and waste ground. The capsule of this moss (fig. 31) differs from those of the preceding mosses in the peristome being BRYUM. 59 double (fig. 42), or composed of an outer and an inner row of teeth. The outer row consists of sixteen ob- lique reddish teeth, which are marked with trans- verse bars or trabe'culae (trabecula, a little beam), and their points are connected by a net-like thin plate. The inner row contains also sixteen teeth, arising from the division of the membrane lining the cap- sule ; these are yellowish, thin, and placed opposite the outer teeth. The capsule itself is pear-shaped or pyriform, orange-red when ripe, curved, and with the mouth oblique. The calyptra (fig. 36) is half-cleft, and expanded as if blown out below. The lid (fig. 31 a) is convex and obtuse ; and the annulus (fig. 39) is large and easily separable. The fruit-stalks are curved near the top. The leaves (fig. 30) are ovate, concave, entire, with a nerve reaching the apex, which is acute and prolonged into a little point, or apic'ulate. The spores (fig. 38) are small and reddish brown. The specific name (hygromet'ricd) of this moss expresses its hygrometric property ; for if either the recent and moist moss be dried or the dry moss wetted, the fruit-stalk gradually twists in opposite directions in the two cases. The last of the Acrocarpous mosses which we shall notice, Bry'um capiHare (PI. III. fig. 49), is tolerably common on trunks of trees, on the ground, and sometimes on walls. The capsule of this moss has a double peristome or mouth-fringe ; the outer consisting of sixteen red- dish-brown, equidistant, transversely striped teeth; the inner composed of sixteen thin keeled teeth, more or less split down the middle, and with two or three intermediate cilia. The capsule is nodding, smooth, oblong, pear-shaped, slightly narrowed below the mouth ; the lid being somewhat convex, and furnished with a short slender beak. The calyptra is dimidiate. The leaves (fig. 50) are obovate, the nerve extending 60 MOSSES. beyond the point, rendering them bristle-pointed. The seeds (fig. 49 a) are small and green. This moss serves to illustrate a great difficulty, which will often occur to the student, in determining whether a moss is end-fruited or side-fruited. For in this, as in many other end-fruited mosses, a little side-shoot or young branch (innovation) grows from the main stem immediately below the leaves sur- rounding the base of the fruit-stalk, so that the fruit- stalk appears to arise from the side of the stem. The only method of overcoming the difficulty is to ex- amine carefully the comparative size and thickness of the stem and the shoot, and to determine which is the weaker and so the newer. The leaves surround- ing the base of the fruit-stalk, which are mostly somewhat different in structure from the stem- leaves, are called the perichce tial (Tre/ot, around, %a/r?7, bristle) leaves. From among the side-fruited or Pleurocarpous mosses we shall select one only, Hypnum rutab'ulum (PL III. fig. 43), which is common on the trunks of trees and on banks. In this moss, the nodding unequal curved capsule (fig. 45) has a double peristome, resembling that of Bryum (fig. 47). The calyptra is half-cleft, and the lid conical and shortly beaked. The stem is reclining or procumbent, and the pale green imbricated leaves (fig. 41) are ovate and pointed, faintly saw-edged, the nerve becoming indistinct at about the middle. It will be noticed that the cells of the leaf (fig. 48) have the prosenchymatous form, or are elongate with pointed ends ; and that the fruit-stalk (fig. 45) is rough with little grains. Fructification. The fruit-producing organs of the mosses are of two kinds, comparable to those of the flowering plants, but with their names changed, as in the case of the ferns; the representatives of the anther being called antheridia, and those of the pistil FRUCTIFICATION, 61 archegonia. The antheridia may be best examined in Polytrichum piliferum or commune, the patches of stems with red rosette-like heads (figs. 25, 26) be- ing readily found in the spring on open heaths. The coloured leaves forming these heads differ in form from those of the stem, being broader and very sharp- pointed, and have received the distinctive name of perigonial (irepl, around, 701/09, offspring) leaves. In the centre of these leaves, which must be separated with mounted needles in a drop of water, the anthe- ridia (fig. 27), forming oblong cellular green sacs, will be seen ; and intermingled with them will be found some slender pale filaments, composed of mostly two rows of cells, which are the paraptiyses (TrapdQvais, a side growth). If the antheridia are quite ripe, they swell somewhat in the water, and from the free or unattached end a very delicate, colourless, cellular mass gradually escapes. If the antheridia are not quite ripe, the mass must be liberated by dissection. On carefully examining this mass under a high power, it will be seen to consist of very delicate rounded cells (fig. 3 a), each containing a coiled fila- ment, revolving more or less rapidly. After a time, these filaments (fig. 3 b) escape, so that they may be examined more minutely. They are excessively de- licate, and are best seen when dried on the slide. Each consists of a very slender curved filament, with a still finer filament, or cilium, arising from it on each side. These are the spermatozoa or spermatd- zoids (a-Trippa, seed, fcoov, animal, eZSo?, resemblance). In Fundria the antheridia (fig. 37) may also be found, by careful examination, in the little green heads terminating some of the stems (fig. 41, of the natural size). In this moss, the paraphyses are inflated at the summit into little knobs, or they are capitate (fig. 37). The pistil-like organs of mosses, or the archegonia, from which the capsule is formed, must be looked for in the winter or early spring. 62 MOSSES. They occur in the parts of the stems from which the fruit-stalk subsequently arises, and are surrounded by perichsetial leaves, so as to resemble in general aspect the antheridial heads. They are readily found in Tor tula and Funaria, which are always at hand. The archegonia (PL III. fig. 21) differ in form from the antheridia, being flask-shaped, with a neck and a dilated base. The neck contains a slender canal, and within the base is a special embryonal cell, from which the capsule is subsequently formed. The spermatozoa of the antheridia pass down the canals of the archegonia, and fertilize the embryonal cells ; but one archegonium only comes to maturity in each head, the others ceasing to grow, and withering, in which condition they are found at the base of the fruit-stalk when the capsule is fully formed. The embryonal cell grows by subdivision, so as to form a stalk-like body, which as it rises extends the arche- gonium upwards until it splits across near the base. Thus the archegonium becomes split horizontally into two parts, the upper and longer of which forms the calyptra, whilst the lower remains as a very short tube or sheath (vagi'nula) surrounding the base of the fruit-stalk. The cellular stalk-like body then swells at the summit, the swollen portion gradually be- coming developed into the capsule, by resolving itself into an outer wall lined inside with a coat forming the outer row of teeth at the top, and within this a thinner membrane or spore-sac, the cleft upper margin of which forms the inner teeth ; and within this are contained the spores. The mass of cells within the spore-sac remains, forming a central column, called the columella. These stages of growth may be readily traced in Funaria. In Plate III., fig. 40 represents two ferti- lized archegonia of the natural size, surrounded by the perichsetial leaves ; fig. 33 is a still more advanced archegone. In fig. 34 the calyptra has separated from MOSSES. 63 the vaginule, and contains the stalk-like body, which is represented alone in fig. 32, the dark summit indi- cating the commencing formation of the capsule. Fig. 35 represents the young capsule, in which all the parts are more advanced in growth. When the seeds of mosses germinate, they produce at first a green Conferva-like filament, which branches at one end, the cells containing green endochrome, while brownish little roots are given off from the other end. The young leafy buds or young stems arise from these confervoid filaments. Examination. In the examination of the mosses, the capsules should be viewed as opake objects while fixed in the forceps; and to discover the minute structure of the teeth of the peristome, a capsule should be wetted with spirit, then immersed in water, slit up with fine scissors, and spread out with the mounted needles, so as to form a transparent object. In this way, the curious structure of the teeth be- comes very distinct. It must be noticed that, in the mosses, the anthe- ridia and the archegonia usually occur in separate flower-like heads ; or the mosses are either monoecious or dioscious (p. 38). Preservation. The mosses may be dried under pressure, and preserved entire in the same manner as the ferns or the flowering plants. If simply dried without pressure, their structure can be readily made out at any future time, by immersing them in water, or by keeping them for a few hours in a moist atmo- sphere. The minute structures of mosses may be mounted in solution of chloride of calcium, or in glycerine ; they keep extremely well without closing the cells. 64 CHAPTER VII. (ALGA, SEA- WEED). THE plants belonging to the Class Algae grow in water, either in that of the sea or in fresh water ; a few of them, however, being found on damp earth, damp walls, &c. The marine Algae are commonly known as sea-weeds ; but the fresh- water Algae gene- rally receive but little popular notice, forming, as they do, slimy masses or strata, of a green or brownish, sometimes red, colour. Algae are of simple structure, consisting entirely of cells ; in some these are single, in others, united end to end, to form threads or filaments, or grouped into a leaf-like expansion, or collected few together into a little spherical group or a flat plate. They possess none of the fibres, vessels, or ducts of the higher plants, although some long and slender cells, existing in the stalks of the fronds of the larger kinds, bear considerable resemblance to woody fibre. They ex- hibit no distinction of stem and leaf, but consist of fronds representing the stem and leaf combined and undistinguishable. And the term frond must be un- derstood to signify the separate parts arising from the point of attachment when they are fixed ; and in the case of those which are unattached or free, the entire plant is called a frond. The Algae are divided into three Orders, viz. the Fucoid'eae or olive-coloured Algae, the Florid' eae or red, and the Confervoid'eae or green Algae. FUCOID'EAE, Fucoid Algae, or Melanospor'eae (yu-eXa?, black or dark). The plants composing this order form our largest sea- weeds, and are found everywhere PLATE IV. [PAGE 64.] MABINE Fig 2. 3. 4. 5. 6. 7. 8. 9. 10. 11. 12. 13. 14. 15. 16. 17. 18. 19. Dasya coccinea, piece of. Dasya coccinea, portion with cap- sule (ceramidium). Dasya coccinea, portion of main filament. Dasya coccinea, section of filament. Meldbesia polymorpha. Melobesia polymorpha, portion with capsules (ceramidia). Jania rubens. Jania rubens. Lithocystis AUmanni. Ceramium nodosum. Ceramium nodosum, filament. Ceramium rubrum, filament. Ceramium rubrum, tetraspore. Ceramium rubrum, end of fila- ment. Ceramium rubrum, capsule (fa- veUa). < Fucus vesiculosus, receptacles of. Fucus vesiculosus, capsules (con- ceptacles). Fucus serratus, antheridial con- ceptacles. Spore of Fucus vesiculosus. Fig. 20. Antheridia of Fucus serratus. 21. Plocamium eomWMW,sporophyll. 22. Plocamium coccineum, with cap- sule (coccidium). 23. Plocamium coccineum, portion of frond. 24. Plocamium coccineum, tetraspore from sporophyll. 25. Polysiphonia fastigiata, portion of. 26. Polysiphonia fastigiata, filament with capsules (ceramidia). 27. Polysiphonia fastigiata, portion of filament. 28. Corallina officinalis. 29. Corallina officinalis, portion of fil- ament. 30. Corallina ojfficinalis, capsule (cera- midium). 31. Enter omorpha compressa. 32. Enteromorpha compressa, cells of frond. 33. Hypnea purpurascens, capsule (coccidium). 34. Hypnea purpurascens, spores. 35. Hypnea purpurascens, filament. FTTCOIDE^E. 65 in the sea and on the sea-shore. They are of an olive- green or olive-brown colour, and usually become darker on drying. Fucus vesiculosus, with its parallel-sided or linear olive-brown fronds, is known to every one as the sea- weed which is hung up to act as a weather-glass. The fronds have a central stout vein, or midrib, and scat- tered air-bladders, mostly in pairs. The fructification consists of yellowish oval enlarge- ments of the ends of the fronds, called the receptacles (fig. 16) ; but these are somewhat variable in form, being often angular or truncate. On holding one of the receptacles to the light, it will appear to contain a number of little grains imbedded in its substance, slightly projecting above the surface, and in the centre of each is a minute dot or pore. These grains are the capsules, or conceptacles, and contain the spores. The substance of the receptacles is composed of a beautiful network of colourless, jointed, cellular fibres (figs. 17 a and 18 a), the meshes of which are filled with a transparent gelatinous substance ; but immediately around the conceptacles the cells are shorter and more closely packed. The spores (fig. 19) are arranged in the conceptacles in a radiate manner; they are brown, and surrounded by a colourless sac, called the perispore (Kepi, around, (nropa, seed) ; and between them are numerous slender, colourless, jointed fila- ments, the paraph'yses. The spores are not, how- ever, truly single spores, for they ultimately divide into eight segments or sporules, each of which is capable of producing a new plant. In the conceptacles of some fronds of Fucus no spores will be found, the conceptacles (fig. 18) being filled with elegantly branched colourless filaments (fig. 20), the ends of many of them being distended into little yellowish sacs ; these are the antherid'ia. The antheridia contain large numbers of exceedingly minute spermatozoa, furnished with two cilia, and 66 ALQM. very similar to those existing in the antheridia of the mosses ; these, escaping through the pore of the con- ceptacle, fertilize the spores. The figure (20) in the plate was drawn from a con- ceptacle of Fucus serrdtus, another common species, differing from F. vesiculosus in having the margins of the frond serrate ; the antheridia of the two species do not, however, differ in any important respect. To examine the conceptacles of Fucus and their con- tents, the receptacles should be soaked in water, if not fresh, and thin sections made with a sharp knife. They form very beautiful objects, and may be pre- served in chloride of calcium or glycerine. FLORID'E^E, or Rhodosper'mese (poSov, rose, crTrep- fj,a, seed). The second Order of Algse, forming the Floridese (flos, a flower), comprises the red sea-weeds; a few of them are purple, or greenish-red ; so that by the colour alone they may be readily distinguished from the Fucoids, and from nearly all those of the next Order, the Confer' voids. A few of them are leaf- like, or possess flat fronds ; but most of them consist of finely divided or feathery fronds. They are often found upon the sea-shore of a dirty white colour, the colouring matter having been decomposed or washed out by rain. We shall consider a few of the genera and species under the heads of the families to which they belong. CORALLINA'CE^E, the Corallines, or calcareous Algae. In this family we have the beautiful Corallina offi- cindlis (PI. IV. fig. 28), the common Coralline, which is very abundant on the sea-shore, attached to larger sea- weeds, shells, and rocks. It is hard and chalky, from the presence of a large proportion of carbonate of lime in its minute cells. The fronds are composed of jointed and branched filaments. The fructification (figs. 29 and 30) consists of ovate cellular capsules, or ceramid'ia (/eepa/uov, earthen vessel), placed mostly at the ends of pinnate stalks, and containing a tuft RHODOMELACE.E. 67 of somewhat club-shaped jointed spores, springing from the base of the capsules (fig. 30) . When ripe, the spores escape from a pore or hole in the end of the capsules. The spores are 4-jointed, and hence are called tet'raspores (rerpa, four) . To observe these spores, the capsules must be soaked in strong vinegar for some hours, and then washed with water, to dissolve the calcareous matter. Jdnia rubens (PL IV. figs. 7 and 8) is another common and very elegant little coralline, and is of a pale red colour. It differs from the last in the branches being dichot'omous (S/%a, in two, ro/zo?, cutting) or forked, instead of pinnate. The capsules, or ceramidia, have also two short horn-like branchlets, placed one on each side, near the end. The genus Melobesia has the frond crustaceous, i. e. forming a hard crust or layer. M. polymor'pha (PL IV. figs. 5 and 6) is common on shells, stones, &c. The capsules (ceramidia) here form little blunt cones, scattered over the crusts, and containing the tufted tetraspores, as in Corallina. Lit ho cydtis Allman'ni (fig. 9) is very minute, and not uncommon upon sea- weeds, stones, &c. It con- sists of a single fan-shaped crustaceous layer of cells, closely investing the body to which it is attached ; its fructification is unknown. Leaving the family of crustaceous Floridese, we shall now pass to those of softer consistence, although all the marine Algae contain a considerable quantity of calcareous matter. RHODOMELA'CE^:. In this family we have the large genus Polysiphonia, in which the frond (PL IV. figs. 25 and 26) is filamentous, the filaments being apparently jointed and longitudinally striated. The filaments are composed of rings of cells (fig. 27), arranged end to end, and containing dark endochrome. The ends of the colourless cell- walls separating the endochromes of the cells of adjacent rings produce the jointed ap- 68 ALG^E. pearance ; while the striated appearance is caused by the dark cells being elongate and the cell-walls thick, so as to form white interspaces. The fructification consists of capsules (ceramidia), attached to the sides of the branches, containing pear- shaped spores, with tetraspores imbedded in swollen branches of separate plants. Polysiphonia fastigidta (fig. 25, a small piece) is common, attached to the fronds of Fucus. Its fila- ments are rigid, bristle-like, of the same breadth throughout, forked, and forming globular brown or yel- lowish tufts, from 2 to 4 inches long. The joints are broader than long, each with 16-18 of the dark cells. In the centre of the branches of this sea- weed is a row of curious objects (fig. 26 ), consisting of a dark- coloured body surrounded with irregular spiny mar- ginal processes, and with a colourless short process above and below. These require further investiga- tion. P. nigredcens is also common among masses of sea- weeds. Its filaments are brown, pinnate, the branches awl-shaped, and the joints about as long as broad. Ddsya coccin'ea (PL IV. figs. 1 and 2, representing small portions of a filament) is a very common fila- mentous red sea-weed of the same family. The fila- ments are 6-8 inches long, and bipinnate, the larger ones somewhat resembling those of Polysiphonia, in being composed of parallel longitudinal cells, arranged round the centre, but containing also smaller inter- mediate cells ; while the smallest branches (fig. 2), which arise in tufts, consist of a single row of cells, little longer than broad. The fruit consists of ovate capsules (ceramidia), placed at the base of the branches, and containing a round mass of spores. There is also another kind of fructification, occurring on distinct plants ; this is formed of one or two rows of tetra- spores, immersed in pod-like capsules, called stichid'ia (cr/%09, row). CERAMIACE.E. 69 DELESSERIA'CE^E. In this family, the typical or most highly developed genus of which, Delesseria, has beautiful leaf-like rose-red fronds, we shall examine the common Plocdmium coccin'eum (PL IV. figs. 23 and 22) . This is of a fine red colour ; the fronds are from 2 to 12 inches long, and consist of numerous branched and bushy filaments. These are compressed, with the branchlets arranged in alternate rows on the two margins of the stem. The end branchlets are acute and pectinate (pecten, a comb), or arranged like the teeth of a comb. The cells of which the fila- ments consist are small and angular, giving the surface the appearance of being elegantly netted under a high power. The fruit (fig. 22) consists of globular cap- sules, called coccid*ia (KOKKOS, a berry), placed in the axils or forks at which two branches separate, and containing a mass of angular spores. There are also tetraspore-pods (stichidia), as in Dasya ; and tetra- spores (fig. 24) in little leaf-like altered branches (fig. 21), called spor'ophylles (o-Tropa, seed, $>v\\ov, leaf), and antheridia are present. RHODYMENIA'CE^E. In this family we have Hyp'nea purpuras 1 cens (PL IV. fig. 35). The filamentous pale purple frond of this sea-weed is from 6 inches to a foot or more in length, the branches being alternate and spreading. The fructification consists of capsules or coccidia (fig. 32), immersed in the branches, and containing the spores (fig. 34). Tetraspores also occur in the cells of the surface of the filaments. CERAMIA'CE^;. This is the last family to be noticed. Cerdmium nodosum (PL IV. figs. 10 and 11), which belongs to it, is a most delicate and elegant filament- ous sea-weed, commonly found attached to other sea- weeds. The filaments are hair-like or capillary, irre- gularly dichot'omous ; they consist of colourless cells, 3 or 4 times as long as broad, and with thick walls. The junctions of the cells are swollen (fig. 11), and covered with very minute dark red cells, giving them 70 ALGJS. a knotty and jointed appearance to the naked eye or under a low power. The globular capsules, orfavel'ke (favus, a honeycomb) , containing the numerous spores, are situated at the ends of the branchlets, and the tetraspores (fig. 11) in twos or threes on the outer margins of them. In Cerdmium rubrum, which is also very common, being found attached to stones, rocks, and the larger Algae, the filaments (PL IV. fig. 12) are stouter than in C. nodosumiy branched so as to form tufts from 2 to 10 inches long, and their ends forked, with the tips hooked inwards (fig. 14). The central cells of the filaments are large and rounded, and their walls are entirely covered with a layer of very small angular red cells. The globular capsules (fig. 15), orfavella, are situated on the suter surface of the branches, stalked, and supported by 3 or 4 short branchlets. The tetra- spores (fig. 13) are imbedded in the branches, towards the ends. The capsules called favellse differ from the coccidia in the walls being simply membranous, while the walls of the coccidia, like those of the cera- midia, are composed of cells. The tetraspores are usually imbedded, among the cells of the superficial layer of the filaments, and are not very easily recognized by an unpractised eye ; it will be observed in the figures that they are some- times cleft horizontally, at others obliquely. CONFERVOID'E^E. This Order consists principally of the green freshwater Algae, although some of them are yellowish brown, purple, or red, and some are marine. Their general structure may be best illus- trated by selecting certain common examples from the families composing the order. The families are 13 in number. The species which are figured in the plates are found in fresh water, except when other- wise stated. CONFERVA'CE^E. On removing some of the soft green matter found adhering to the stems of water- PLATE V. [PAGE 70.] FRESHWATER ALGJB. Fig. 1. Conferva floccosa, single filament. 2. Lyngbya muralis, single filament. 3. Ulothrix mucosa (?), filament. 4. Synedra radians, prepared frus- tules. 5. Synedra radians, tuft of natural frustules. 6. Cladophora crispata, with zoo- spores (a). 7. Batrachospermum moniliforme, portion of filament. 8. Batrachospermum moniliforme, filament. 9. Closterium acerosum. 10. Draparnaldia glomerata. 11. Spirogyra quinina. 12. Spirogyra nitida, filaments con- jugating. 13. Zygnema cruciata. 14. Coleochfete scutata. 15. Xanthidia in flint. 16. Micrasterias rotata. 17. Gomphonema acuminatum. I Fig. 18. Gomphonema acuminatum, pre- pared frustules. 19. Ankistrodesmus falcatus. 20. Pediastrum Boryanum. 21. Hyalotheca dissiliens. 22. Pinnularia viridis. 23. Fragilaria capudna. 24. Fragilaria capudna, prepared frustules ; s* side view of F. virescens. 25. Scenedesmus quadricauda. 26. Schizoffonium, probably a form of Lyngbya. 27. Campylodiscus costatus. 28. Nitzschia minutissima, front view. 29. Nitzschia minutissima, valves. 30. Epithemia turgida. 31. Diatoma elongatum, natural frus- tules. 32. Diatoma elongatum, prepared frus- tules. 33. Cocconeis placentula. Plate V. i I & CH^ETOPHORACE^E. 71 plants in any pool or pond, one of the species of Con- ferva, C. flocculosa, is almost sure to be met with. On close inspection with the naked eye, the green filaments of which it consists are just visible, as ex- tremely fine, soft, silky threads ; and, under a high power of the microscope, the filaments are seen to be uiibranched, and composed of a single row of cells (PL V. fig. 1), or joints, as they are called in technical works ; these are 2 or 3 times as long as broad. In some specimens the joints are swollen, so as to present a rounded outline. In another common species, C. bombydina, the filaments are somewhat more slender, and the joints are from 3 to 5 times as long as broad. Cladoph'ora crispdta. This Confervoid forms large, entangled, dull-green masses, composed of branched, tufted, somewhat rigid and coarse filaments. It is often a troublesome overrunner of the fresh-water vivarium. The filaments are composed of thick- walled cells (PL V. fig. 6), from 4 to 6 times as long as broad, and often containing minute starch-granules. The Confervacese have two modes of reproduction. The first of these consists in the division of the endo- chrome of the joints into a number of distinct seg- ments, each of which becomes furnished at one end with two very slender cilia (PL Y. fig. 6 a). After a time, these ciliated bodies, which are called zo'ospores (wov, animal, airopd, seed) or gonid'ia (701/77, seed, eZSo?, resemblance), escape from the cells either by their rupture or through a papillary orifice, and swim about in the water, ultimately losing their cilia and growing into cells resembling those of the parent plant. In the second method, which occurs, for in- stance, in Conferva bombycina, certain of the joints enlarge so as to become rounded or inflated ; their endochrome then becomes coated with a new cell- wall, and so forms a spore, which subsequently escapes from the cell and germinates. :. Draparnal 'dia glomerdta forms 72 ALG^E. small green jelly-like masses, adhering to sticks and stones in water. These consist of branched filaments (PL V. fig. 10), prolonged at the ends into colourless hair-like points, and composed of single rows of cells, the green endochrome forming a band across the middle of each cell, the ends being colourless. In Coleochdte scutdta (PL V. fig. 14) the cells are closely united, so as to form a minute flat green disk. In the natural state, this beautiful little object adheres to the submerged leaves and stems of water-plants, and is therefore difficult to be found. But if a few healthy water-plants be kept for some time in a glass jar, the little Coleochcete, which is about as large as a pin's head, will often be found adhering to the side of the glass. BAT'RACHOSPEB/MEJE. The members of this family resemble to the naked eye the little masses of Drapar- naldia, and they are found in the same localities. They are of various colours, being green, brown, purple, or red. They consist, as in Bat'rachosper'mum monilifor'me (PL V. fig. 8), of branched filaments, which have a knotty appearance under a low power. The larger filaments are composed of cells arranged end to end, the knots consisting of numerous smaller whorled filaments, i. e. filaments arising from around them at the same level (fig. 7). The cells composing the whorled filaments are beaded or moniliform, and are prolonged into colourless hair-like points. The globules seen among the branches (fig. 7) consist of groups of spores. ZYGNEMA'CEVE. The members of this family re- semble the Confervacese in consisting of simple cel- lular filaments (Pl.V.figs.ll, 13), but differ from them in the elegant arrangement of the endochrome : this forms beautiful spiral bands, as in Spirog'yra quini'na (fig. 11), or star-shaped masses, as in Zy grid ma cru- cia'ta (fig. 13). A remarkably curious phenomenon met with in them is the manner in which the spores are DESMIDIACE.E. 73 formed, and which is known as conjugation. In this process the opposite cells of two distinctfilaments, lying near together, push out protrusions of the cell-walls, which meet and open into each other, forming cross tubes, as in Spirog'yra nit'ida (PL V. fig. 12). The contents of the opposite cells of the filaments then unite, forming large spores, which remain either in the cells of one of the filaments or in the cross tubes. The three species figured are common in clear pools. DESMIDLA/CE^E. The Desmidiacese are truly micro- scopic, few of them being even perceptible to the naked eye without the very closest examination. They are very beautiful, on account of their bright green co- lour and often elegant forms. Many of them are very common, existing in every pond or ditch; but they abound most in clear open boggy pools on heaths. On placing some water containing them in a glass jar and exposing it to the light, they will often be found adhering to the glass, or forming a layer on the surface of the muddy sediment. The Desmidiacese consist mostly of single cells (PL V. figs. 9, 16) ; and these consist of two equal halves or segments, as indicated either by a paleness of the endochrome or a deep constriction at the line of junction, which is called the suture. The cells are often elegantly lobed and cut, or spiny; and in many the surface exhibits minute markings, consist- ing of little protrusions of the cell-wall outwards, or inflations, as they are called. Their reproduction is effected by division and con- jugation. In the process of division the cells gradually separate at the suture, and a new half-cell is formed upon each old half, which grows until it attains the size and form of the original half of the parent-cell. The conjugation is effected by two cells approximating so that their sutures are near together, the cells then 74 ALG.E. open at the sutures, and the effused contents become united to form a spore or sporange, from which one or more individuals are formed. These spores are often elegantly spinous on the surface. Among the species selected for illustration is Clos- terium acerosum (PI. V. fig. 9), in which the cells are single, elongate, very slightly curved or lunate ; the endochrome forms long bands, often containing nu- merous globules or transparent vesicles. At each end of the cells is a round transparent vesicle, con- taining exceedingly minute granules, which exhibit a trembling kind of motion. Between the cell- wall and the cell-contents very fine currents may also be detected, forming a circulation resembling that in the hairs of Tradescantia. In Micrasterias rotdta (PL V. fig. 16) the cells, which are single, are deeply cleft into two segments at the suture, the segments being again regularly cut into five lobes, which are toothed or dentate. In Hyalotheca dissil'iens (PL V. fig. 21) the cells are united into a cylindrical filament, and are sur- rounded by a very delicate gelatinous sheath. In Ankistrodes'mus falcdtus (PL V. fig. 19) the cells re- semble those of Closterium in shape, but are aggre- gated into faggot-like bundles, and are very much smaller. In the beautiful little Pediadtrum borydnum (PL V. fig. 20) the cells are aggregated into a disk, the marginal cells being bidentate or having each two points, so that the whole resembles a star. The species of Pediastrum are reproduced by the contents of each cell subdividing into numerous ciliated segments or zoospores, which subsequently escape in a mass from the cell, ultimately losing their cilia, and reuniting to form a new individual. In Scenedes'mus quadricauda (PL V. fig. 25) the oblong cells are united, side by side, the outermost cells being furnished with a bristle at each end. The division of these cells takes place obliquely, so that DIATOMACE.E. 75 in the divided groups the cells are situated in two alternate rows. The spores of many of the Desmidiacese are spi- nous, and they are often found fossil in flint (PL V. fig. 15). To detect them in this substance, thin slips of flint may be examined under a half-inch power ; or the chips of flint may be cemented to a slide with balsam, and ground down on a hone. The Desmidiacese must be mounted in the moist state : the smaller ones will keep well in chloride of calcium \ but the larger ones are injured both by that liquid and by glycerine. The remarks made upon mounting, at page 15, are especially applicable to these delicate organisms. DIATOMA'CE^E, or Siliceous Algae. The members of this family are singly very minute ; but when ex- isting in large numbers, as they are often found at the bottom of ditches and ponds, on the submerged stems of water-plants, or upon damp ground, they form yellowish-brown evident masses or strata. They occur both in sea- and in fresh water. They usually consist, like the Desmidiacese, of single cells, which are called frustules. But they are especially charac- terized by the cell-walls being imbued with silica or flint, so that if the frustules be heated to redness upon the point of a knife or a slip of platinum-foil, which destroys the organic part of the cells, the coat of silica remains, exhibiting the perfect form of the original cells or frustules. The form of the frustules is very different in the various genera and species, as represented in PL V. tigs. 22, 23, 27, 30, 31, and PL VI. figs. 16, 17, 23; and it will be noticed that, in the figures, two views are given of each frustule, / indicating the front view, and s the side view. In all the front views, as in PL V. fig. 22, one or more lines will be observed running longitudinally down the middle of the frustules, and corresponding to the indications of division existing in the cells of the H2 76 . ALG.E. Desmidiaceee. Each half of a frustule is called a valve, and the line at which these valves meet is called the suture. That side or aspect of the frustule in which the suture lies (fig. 22/) is the front view, and the other aspect of the frustule (fig. 22 s) is the side view. The frustules are mostly four-sided the main breadths of the two opposite valves forming two sides, and the bent margins of the valves, with the back and front of the hoop, forming the two other sides - y so that the view presented by the side of a frustule is the same as that of a single valve. The suture is the line at which the division of the frustules takes place in the formation of new individuals. In this process the cell-contents divide into two parts, as in ordinary endo- genous cell-formation, the two new surfaces thus produced becoming coated with a new portion of cell- wall or valve, so that two frustules now occupy the place of the original one. At the same time a silice- ous band, encircling the frustules at the line of suture, is formed to fill up the interval between the edges of the parent valves; this is the hoop (PI. V. fig. 22 /;. PL VI. fig. 10/), and beneath it lie the two newly formed valves. In many cases I believe that each half-frustule becomes coated with a new entire cell- wall, with its siliceous valves. The frustules of the Diatomaceae are constantly undergoing division when in vigorous growth. After the frustules have divided, the new ones either sepa- rate entirely, as is perhaps most commonly the case ; or they remain united, sometimes completely, so as to constitute a filament (PI. V. fig. 23), while at others the frustules cohere only at the angles (PL VI. fig. 23), so as to form a zigzag chain. In some species, the frustules are attached to foreign bodies by means of a gelatinous cushion (PL V. fig. 5 ; PL VI. fig. 7) j while in others they are situated upon a simple or branched gelatinous stalk (PL V. fig. 17) or stipes (stipes, a stem) . DIATOM ACE ;E. 77 When the frustules are examined in the living state, the cell- contents resemble those of ordinary vegetable cells, excepting in regard to the colour, and exhibit granules and globules, and sometimes a nucleus is visible. It will also be noticed that many of the free frustules move slowly across the field of the microscope ; but the cause of the motion is unknown. When the frustules have been properly prepared, the surface of the valves exhibits a number of coarser or finer markings, consisting of dots, lines (striae), flutings, or networks, &c., arranged with great re- gularity and symmetry, often of extreme minuteness, and rendering them exquisite objects under the mi- croscope. The exhibition of these markings requires not only that the valves shall be properly prepared and mounted, but that the object-glasses be of good quality, and that the management of the light be thoroughly understood ; so that to a beginner, their examination is often a matter of great difficulty ; for only the very coarsest or largest of these markings can be perceived in the natural frustules. The appearance of these markings, and even their apparent absence or presence, frequently depends upon the kind of illumination used : thus, under one kind of illumination the valves may appear simply white or coloured, while under another they appear covered with lines, and under a third with dots. It will often be observed, also, that the colour of the valves varies according to the illumination and the power used the same valve appearing white, yellow, brown, blue, &c; and the wet or dry state of the frustules often cause a decided difference in their ap- pearance as regards colour. To illustrate the forms and markings of the frus- tules and valves, we may select the following species taking first those which occur in fresh water. In Epithemia tur'gida (PI. V. fig. 30), the side- view or valve (s) exhibits transverse or slightly radia- H 3 78 ALG.E. ting lines, with intermediate rows of dots these markings being continued over the margins of the valves so as to appear also in the front view (fig. 30/), ceasing at the hoop. The frustules are curved or arcuate (ar'cus, a bow) in the side view, oblong and narrowed at the ends in the front view. In Fragildria capucina (PL V. fig. 23), which is ex- tremely common in fresh- water pools, &c., the frus- tules are united side by side into long filaments, which are often twisted. In the separate and pre- pared frustule, the front view (PL V. fig. 24/) is rect- angular, the valves (s) being narrowly lance-shaped or lanceolate. The valves under ordinary illumina- tion appear colourless and without markings, but, by proper management of the light, very fine transverse striae are seen upon them, consisting of rows of very minute dots. Fig. 24 s* represents the valve of Fra- gilaria vires' cens, a nearly allied species. Diat'oma elongdtum (PL V. fig. 31) is often found with the above. Its frustules are coherent at the angles. The front view (fig. 32 /) is rectangular, often slightly narrowed in the middle ; and the valves are narrowly linear, and capitate at the ends ; they are also transversely striated. In Synedra splen'dens {PL V. fig. 5) the frustules radiate from a soft gelatinous cushion. They are linear in the front view (fig. 4 f) , the valves (fig. 4 s) being gradually narrowed or attenuated from the middle to the ends, and exhibit transverse striae interrupted opposite a middle longitudinal line. In Campy lodis 'cus costdtus (PL V. fig. 27) the frus- tules are disk-shaped and curved, so as somewhat to resemble a saddle. The markings consist of central dots, with radiating coarse flutings. Nitzsch'ia minutis' sima (PL V. fig. 28) has oblique valves, i. e. the front half of the suture is not opposite the back portion ; the valves (fig. 29) are constricted in the middle, and the ends narrowed and prolonged. DIATOMACE.E. 79 The markings consist of a row of oblong dots or puncta (punctum, a point) . This species often forms yellowish layers upon damp paths, &c. In the next group, the valves have a longitudinal line running down the middle of the valves, with a little knob or nodule in its centre (PL V. fig. 22 s), both consisting of internal thickened portions of the valves. In Cocconeis placen'tula (Pl.V. fig. 33) the valves are oval, and the markings consist of longitudinal rows of minute dots, with a marginal row of puncta ; these markings are invisible under ordinary illumination. InGomphonema acumindtum (Pl.V. fig. 17) thefrus- tules are attached to a branched stalk (stipes) ; they are wedge-shaped or ciineate (cuneus, a wedge) and trans- versely striate (fig. 18), the striae consisting of dots. In Pinnuldria vir'idis (PL V. fig. 22), which is very often seen slowly traversing the field of the micro- scope when a drop of pond-water is examined, the frustules in the front view are linear, the valves being elliptic oblong, and transversely striated, the striae consisting of furrows. In Gyrosig'ma (Pleurosig'ma) attenudtum (PL VI. fig. 16) the valves are sigmoid, or somewhat resemble a Greek 9 (sigma) in outline, and the markings con- sist of rectangularly crossed rows of very fine dots ; in the front view, the frustules are linear-oblong with truncate ends. Tabelldria flocculosa (PL VI. fig. 23) has the frus- tules adherent only at the angles, as in Didtoma. They are rectangular, and in the front view exhibit a row of longitudinal dark lines interrupted in the middle ; these have been compared to the vittse of the fruit of the Umbelliferae, and have received the same name. Among the marine species may be mentioned Melosira nummuloides (PL VI. fig. 9), in which the frustules are united into a chain or cylindrical fila- ment. This is very common among sea- weeds, &c, ; 80 ALG.E. and it illustrates well the process of division of the frustules (fig. 10 /) . The valves are covered with fine dots, and near each end of the frustules is a projecting rim encircling it, and appearing as a curved line extending beyond the margin of the frustule in the front view. In Actinocy 1 clus un- duldtus (PI. VI. fig. 5) the frustules are separate, disk-shaped, and the valves are divided into six equal parts by six rays, each alternate portion of the surface of the valves being situated on a lower level than those adjacent, so that an alteration in the focus is required to bring into view the dots on any two ad- jacent divisions of the valve. The surface of the valves is covered with easily recognized dots. The form of the surface is best seen in the front view (fig. 5 /) when the frustule is placed on its edge. Rhabdonema arcudtum (PI. VI. fig. 7), which is very commonly found attached to sea- weeds, resem- bles Tabelldria in the frustules having the vittae (PI. VI. fig. 8/). The frustules form short filaments, attached by a little gelatinous cushion. The valves have transverse striae, interrupted in the middle (fig. 8 *). Gyrosig'ma (Pleurosig ma) anguldtum (PL VI. fig. 17) resembles G. attenudtum in the sigmoid form ; but the markings consist of lines crossing each other obliquely; and these are resolvable into rows of dots (fig. 17 a) under suitable illumination. In Coscinodis'cus radidtus (PI. VI. fig. 3) the frus- tules are disk-shaped, the valves being elegantly sculptured with easily recognized cell-like markings or dots, so as to resemble a piece of vegetable cellular tissue. But in some other species the dots are very minute, and difficult to be shown satisfactorily. These markings consist of depressions or pits in the surface of the valves. That this is the case may easily be seen by examining a fragment of the valve, when the shadings of the broken ends of the netted DIATOMACE.E. 81 thicker portions, which project like teeth, strongly contrast with the difficultly distinguishable portions of the thin interspaces. The fossil forms from the Bermuda deposit are best for the investigation of this structure ; many of these are extremely beautiful microscopic objects, their markings resembling those on the engine-turned back of a watch. The detection of the finer markings of the Diato- macese, which, according to my view, consist of de- pressions like those upon the valves of Coscinodiscus t is a matter of great difficulty to those who are un- accustomed to the use of the microscope, and who have not a complete set of apparatus. The main point to be attended to in bringing them into view, is to use one-sided oblique light, i.e. to turn the mirror by its stem as much as possible to one side, and then to incline it so as to throw the light upon the object. In this way the valves of the species of Gyrosigma, for instance, appear covered with lines (PL VI. figs. 16 and 17) ; but the lines are spurious, i. e. they are the optical expression of rows of minute dots (figs. 16 0, 17 a) ; and when oblique light is thrown upon the valves from all sides, by means of a special achromatic condenser, in which the central rays are excluded, the dots become distinct, and the mark- ings resemble those on the valves of Coscinodiscus. To show the finer dots clearly, a valve should be crushed, so as to obtain a fragment as flat as possible ; for the surface of the valves is curved more or less in all the species. The valves of G. angulatum are generally used to test the quality of the object-glasses of the microscope, and also for practice in " making out " the lines and dots ; there are, however, many Diato- macese more difficult. As the nature of these markings is a disputed point, the discussion of which is not adapted for an elementary work, I must refer for further details to the ' Micrographic Dictionary ; 3 it may be remarked, 82 ALG^E. however, that some observers have regarded them as cells, and others as elevations or tubercles on the surface of the valves. The preparation of the valves for showing the markings should be effected by burning the frustules, or the mass containing them, on a strip of platinum- foil over a spirit-lamp. The incinerated mass should then be transferred to a slide, and the valves separated with the greatest care by a bristle mounted in a hair- pencil stick under a low power of the microscope. This is, however, a substitute for the proper method, which is dangerous in the hands of one unpractised in chemical manipulation. It is this : The mass of Diatomacese (the water containing it having been carefully poured off as far as possible) is put into a Florence oil-flask, and strong nitric acid (aquafortis) gently added, more than sufficient to cover it. The mixture is then carefully boiled over a spirit-lamp for some time. When it is cold, distilled water is added, the whole shaken, and allowed to settle. The watery part is then gently poured off, more water added, and this poured off after settling, and the pro- cess repeated until a drop of the water evaporated to dryness on a slide leaves no residue. The Diato- maeese then form a white sediment at the bottom of the water, and can be transferred to a slide with a dipping-tube. The drop is then dried with a gentle heat, and the valves mounted as dry transparent objects (p. 12). If the valves have coarse markings, they may be mounted in balsam; but if the markings are fine, balsam makes them much more difficult of detection. Many of the most beautiful Diatomacese are found in the fossil state ; and specimens of these are sold already mounted. I would advise those unacquainted with them to purchase a slide of the ' ' Bermuda " or " Richmond " earth, which abounds in the species of Coscinodiscus ; and of the " San Fiore deposit," VOLVOCINE.E. 83 which contains many species of Epithemia, Navicula, Pinnularia, &c. These may be procured from Mr. Norman, 178 City Road, or from the microscope- makers. VOLVOCIN'EJE. The Volvocinese are inhabitants of clear fresh-water pools, on heaths and bogs. They are very minute, of a rounded or plate-like (tabular) form, of a green colour, and are pretty readily distin- guished from most of the other Algse by their free motion ; for they swim about in the water like animals, as which they were formerly considered. They consist usually of groups of thick- walled soft cells, each being furnished with one or two cilia, by means of which the movement of the compound bodies is produced. In the beautiful Volvox globdtqr (which is not uncommon) the cells form a hollow sphere (PL VI. fig. 18), studded with exceedingly minute green spots or zoospore-like bodies, representing the green endo- chrome of the component cells, and from each of which very fine radiating lines extend, so as to give the surface a netted appearance ; the lines consisting of delicate processes of the endochrome, which may be compared with those existing in the cells of the hairs of Tradescantia. In the interior of the parent globes are often seen several young organisms, usually eight, of a deep green colour ; these escape by the rupture of the parent, so as to form independent beings. Sometimes they are found of a yellow colour, and furnished with a thick transparent coat ; these are called ' ' resting spores," as they remain for some time before undergoing their full development. The cilia of Volvox, of which there are two to each of the component cells, are difficult to detect ; they are best seen when the organism is dried without a cover, or after moistening them with a little solution of iodine, which dyes them brown. Synura volvox (PL VI. fig. 13) is a still more minute member of this family, and is often found ALG.E. rolling along among Conferva. The greenish zoospore- like bodies of this Alga have one cilium only, and arise from a common centre by a narrowing of the base (%. 14). In Gonium pectordle (fig. 11 a) the green bodies, which are sixteen in number, and furnished each with two cilia, are grouped into a flat square plate ; and in the very minute Gonium tranquil' lum (fig. 11 c) these bodies are also sixteen in number, and arranged in a tabular form, but are without cilia. SIPHONA'CE^E. The structure of this family may be illustrated by the genus Vaucheria, of which two or three species are common on damp ground or in freshwater pools, forming a green layer. At first sight, the filaments of which the little plants consist appear like those of a stout Conferva ; but on close examination they are found to be branched, and not jointed, consisting of a single cell from end to end (PL VI. fig. 26). The reproduction is effected by the agency of two kinds of organs, antheridia and capsules (sporangia), situated near each other (fig. 26 a) on the walls of the filaments, of which they are protrusions or outgrowths their cavities being separated from that of the filament by a partition or septum. The antheridia produce spermatozoa, and the sporangia each a spore, the one fertilizing the other in the ordinary manner. In addition to this method of fructification, zoospores are also produced the ends of the filaments becoming swollen, the con- tents cut off by a septum, and forming single large zoospores covered with cilia, the farther development of which resembles that occurring in the Confervacese. OSCILLATORIA'CE/E. The members of this family are commonly found in stagnant water or on shaded damp ground, . especially in the cold seasons of the year, forming green strata or masses. Oscillatoria autumndlis (PL VI. fig. 1) occurs every- where upon damp shaded banks of ditches, espe- PLATE VI. [PAGE 84.] FRESHWATER Fig. 1. Oscillatoria aittumnalis. 2. Oscillatoria nigra. 3. Coscinodiscus radiatus. 4. Nostoc minutissimum. 5. Actinocyclus undulatus. 6. Bacterium. 7. Rhabdonema arcuatum. 8. Rhabdonema arcuatum, prepared frustules. 9. Melosira nummuloides. 10. Melosira nummuloides, prepared frustules. 11. a, b, Oonium pectorale : 11 c. Go- nimn tranquittum. 12. Spirulina oscillarioides. 13. Synura volvox. 14. Synura volvox. 16. Gyrosigma attenuatum, front view. Fig. 16. Gyrosigma attenuatum, side view ; 16 a, portion of a valve. 17. Gyrosigma angulatum ; 17 a, por- tion of a valve. 18. Volvox globator. 19. Glozocapsa. 20. Chara vulgaris, globule. 21. Chara vulgaris, portion of fila- ment. 22. Chara vulgaris, branch with nu- cule and globule. 23. Tabellariajlocculosa. 24. Tdbellaria flocculosa, prepared frustules. 25. Palmella cruenta. 26. Vaucheria Ungeri (sessilis). 27. Vaucheria Ungeri, capsule. Plate VI OSCILLATORIACEjE. 85 cially when newly made, forming a greenish-black closely adherent stratum. Under the microscope it is seen to consist of innumerable palish-green filaments ; these are jointed or transversely striated, some being straight, others curved, the ends often exhibiting a writhing or worm-like movement. The appearance of these fibres is peculiar, seeming as if they were solid throughout, and so differing from that of the Confervacese, in which the cell-walls are readily dis- tinguishable from the cell- contents. The fibres easily break across at the joints ; and the last few segments are often narrowed and rounded, so as to form a blunt point. When they have been left in water, they ex- hibit colourless tubular sheaths surrounding and extending beyond them. These sheaths consist of the consolidated outer portions of the cell- walls ; for when the cells undergo transverse division, and ex- pand by growth in the direction of the length of the filament, the original septa or inner walls are broken through, and their remains may often be seen on the inner surface of the sheath, appearing as little teeth. Oscillatoria nigra (PL VI. fig. 2) is another very similar species, forming blackish-green masses, and is common in ditches. It has longer filaments than the last, with narrowed and slightly curved ends ; and the endochrome is distinctly granular. In two other genera of this family, Vib'rio and Spirulma, the filaments are spiral. Vib'rio spiril'lum is excessively minute, colourless, and found in decom- posing vegetable mixtures. The short filaments move rapidly through the water, with a corkscrew-like motion. In Spirulina oscillarioides (PL VI. fig. 12), which is more rarely found in clear pond- waters among Conferva, the filaments are greenish, and form a beautiful simple spiral, resembling that of a very slender spiral vessel. Lyngbya murdlis (PL V. fig. 2) is very common on damp walls, gravel walks, &c. It forms a bright 86 ALGM. grass-green layer, consisting of somewhat rigid curled filaments. The endochrome is usually broader than long ; and the cells of the filaments are often found empty, the endochrome having escaped in the form of gonidia. PL VI. fig. 6 represents a species of Bacterium which is not uncommon in decomposing vegetable liquids ; the filaments are short, curved, pointed at the ends, and have four joints. Fig. 26 represents a Schizogonium, found upon damp paths. The filaments resemble those of Lyngbya, but are united in pairs. Fig. 3 represents a filament of a Vlothrix, which is common in freshwater pools, showing the curious manner in which the endochrome is arranged in the cells, forming bands partially lining the cell-walls. NOSTOCHA CEM. Two species of the typical genus Nos'toc will serve to represent this family. Nos'toc commune is found on damp ground or in ponds, and forms to the naked eye firmish, olive-green, skin-like 7 plaited masses, an inch or more in diameter. Under the microscope it is seen to consist of numerous beaded fibres, imbedded in worm-like gelatinous sheaths ; these are curved and interwoven to form the compound mass. In the middle of many of the fila- ments is an enlarged colourless cell, called the vesi- cular cell, which is related to the reproduction, but in a manner not yet determined. Nostoc minutissimum (PL VI. fig. 4) forms solid gelatinous bluish-green masses, varying in size from a pin's head to a pea; it is found upon unhealthy water-plants kept in glass vessels. The component filaments are very slender, wavy, and the sheaths often have a brownish tinge. ULVACE.E. These Algae are mostly marine some, however, being found in brackish or fresh water, or on damp ground, thatch, moss, &c. They are gene- rally of considerable size, forming flat or tubular PALMELLACE.E, 87 fronds, often several inches long, a few being fila- mentous. They consist of one or more sheets or layers of cells, containing mostly green endochrome. This at first fills the cells, but subsequently becomes converted into single spores, or subdivided into nume- rous ciliated zoospores. Ul'va latissima is very common on the sea-coast, being found attached to stones, shells, &c. It forms a broad, flat, green, rounded or^oblong, thin frond, wavy and crumpled at the margins, and from 6 to 18 inches in length. The minute cells form two layers, adherent to each other. The zoospores formed are numerous in each cell. Enteromorpha compres sa (PL IV. fig. 31) is also common in the sea and in brackish ditches ; it is often found floating. The frond is green, tubular, flattened or compressed, and branched, the branches being usually simple and narrowed at the base. The frond consists of two layers of minute cells, separated .by a space rendering it hollow. The zoospores are numerous in the cells (fig. 32) . PALMELLA'CE^. These Algae are found in fresh or salt water, or on damp earth, wood, &c. They are green or red, forming round or irregular masses or strata. They consist of loosely connected cells, im- bedded in a gelatinous mass or matrix, thus forming a frond. Chlorococcum vulgdre (PL II. fig.l) is very common upon the bark of elm-trees, palings, &c., forming a green granular crust. It consists of minute rounded or oval cells, mostly undergoing division into twos, fours, or eights. These cells are attached to the sides or ends of very fine colourless filaments. It is most probable that this organism, which is usually placed among the Algae, consists of the gonidia of a Lichen. Chlorococcum muror'um forms a somewhat similar .but soft and thin green layer, upon damp walls or 88 ALGJE, other porous bodies. It consists of very minute oval green cells,, with thick walls, and imbedded in the ends of prolongations of a gelatinous matrix. Palmel'la cruerita (PI. VI. fig. 25) forms a portwine- red layer at the bottom of damp walls or on the ground. It is composed of pale red cells, imbedded in no definite order in a colourless gelatinous matrix. The cells are filled with red granules, and are often found undergoing division. PI. VI. fig. 19 represents a species of Glceocap'sa, in which the cell- envelopes do not soften and unite to form a gelatinous matrix, as in Palmetto, and other members of the family, but are persistent. This species occurs in fresh water containing Confervae. CHARA'CE^E. This family consists of the single genus Chdra, the systematic position of which is not agreed upon by authors ; as however its structure will be better understood after what has been gone over, it may be conveniently considered here. There are several species of Chara, the one illus- trated, Char a vulgdris (PI. VI. fig. 21), being com- monly found in ditches and pools. It consists of long main stems, often a foot in length, which are branched, and surrounded at tolerably regular intervals by whorls of branchlets. In some species, the stems and branches consist simply of elongated cells, arranged end to end ; while in others, of which Chara vulgaris is one, the central cells are surrounded by a number of narrower spirally arranged cells, forming an outer coating. The Charse have long formed interesting micro- scopic objects, on account of the circulation of the protoplasm being visible in the cells, as in the hairs of Tradescantia. This is best seen in those species in which the outer layer of cells is absent from the stems, and which were formerly arranged in a sepa- rate genus (Nitella) . But it may also be seen in the stems and especially the young branchlets of any of the other species ; and as the granules of the proto- CHAEACE^E. 89 plasm are large, the phenomenon is more easily wit- nessed than in Tradescantia. The fructification consists of two kinds of organs, viz. red globules (PL VI. fig. 22) representing the anther-organ, and green capsules (fig. 22), or nucules, corresponding to the ovaries. The structure of the globules is very curious. Their transparent walls (fig. 20) consist of eight somewhat triangular plates, each of which is composed of cells radiating from a centre ; and from the inside of each of these centres arises a tubular cell extending to the middle of the globule, the unattached ends giving origin to nume- rous colourless coiled filaments, consisting of minute cells arranged end to end, each containing a very minute coiled spiral fibre, to which are attached two exceedingly slender cilia. These ciliated fibres are the spermatozoa. The capsules or nucules (fig. 22), which are situated near the globules, are urn-shaped, coated with spiral cells, and crowned with five shorter cells. When the globules are ripe, they become rup- tured by the separation of the valves ; and the sperma- tozoa, escaping from the cells of the coiled filaments, swim about and enter a canal in the capsules to fertilize the ovule contained within. The Charge grow readily in a glass jar of fresh water, with a few pebbles at the bottom ; and if the plants be not overgrown with Confervoids, the fructification will continue to be produced almost throughout the year. The circulation is best seen in the whorled branch- lets, a portion of the growing ends being placed in a live-box, or simply laid upon a slide in water and covered with thin glass. Preservation. The Algae are best preserved in two ways, the entire fronds being dried upon paper under pressure, as directed for the Ferns ; and small portions, showing the minuter structures and fructification, being mounted in chloride of calcium or glycerine. If i3 ALGM. it is required to preserve the marine Algae according to the first method, they should first be immersed for a time in fresh water, to dissolve out the saline matters derived from the sea-water, which would keep them damp and ultimately spoil them. After these matters have been removed, the fresh water should be changed, and pieces of paper placed beneath them while sus- pended in the water ; on withdrawing the paper care- fully, keeping the Algse at the same time spread out, they may be made to retain the required position; and when the water has drained away, and the re- maining moisture has mostly evaporated, they may be submitted to pressure in a press. The Confervoid Algse may be conveniently spread out upon paper and preserved in the same manner, as some of the distinguishing characters are founded upon their appearance in the dry state, their adhesion to the paper, &c. Moreover they can then at any time be minutely examined, by the immersion of a small portion in water. LICHENS. 91 CHAPTER VIII. LICHENS. THE Lichens are found growing upon the bark of trees, old palings, &c. Those most easily seen with the naked eye form grey or coloured dryish patches or pendulous tufts; while the smaller ones are singly easily overlooked, from their minute size and close adhesion to the matrix or hody upon which they grow, forming, by their aggregation, the grey or otherwise-coloured dry and brittle coatings of almost every tree or decaying branch. The Lichens derive their nourishment from the air, and not from the matrix in this respect differing from the Fungi, with some of which, as we shall pre- sently see, they agree in the structure of the fruit. The structure of the Lichens is simple, no distinc- tion of root, stem, and leaves existing in them, although certain dry root-like fibres exist in many of them, by which the plant is fixed to the matrix. The whole consists mainly of a frond or thal'lus (0a\\o$, a leaf) . This is either raised above the surface of the matrix in a shrubby form, or spread upon the surface as a flexible lobed layer (PL II. fig. 2), or it is dry and brittle (crustaceous) and closely adherent (PI. II. fig. 26), The fructification consists of little saucers, disks, or streak-like furrows, often of a different colour from the thallus, the structure of which will be best illus- trated by reference to a few common species. PARMELIA'CE^E. Parmelia parietina (PL II. fig. 2) is a very common Lichen, found on the bark of trees, on old palings, &c. It is of an orange-yellow colour, 92 LICHENS. the thallus being flat, lobed, and scalloped (crenate) at the margins. The structure of the thallus may serve to represent that of most of the Lichens. It consists of three layers, an upper cortical or epider- mic layer, which is continued over the margins to the under surface of the thallus, to form the under layer ; and between these is the middle or medullary layer. The medullary layer consists of interwoven fibres, which are more closely packed towards the upper and under surfaces, so as to give them a cellular appear- ance. Near the upper part of the medullary layer, a number of minute rounded green cells exist, lying loosely in its meshes (PI. II. fig. 4 a). These green cells (gonidia) appear to correspond to the buds of the higher plants, and, when detached from the plants, they are capable of growing into new individuals. On the upper surface of the thallus of Parmelia (fig. 2) the fructification may be observed. This consists of saucers or shields (apothecia, airoBriK^ a repository), formed of a raised and expanded portion of the thallus (fig. 3), and containing the spores. The spores are enclosed in closely set upright cells, or spore-sacs (figs. 4 b and 5 b), called as'd (acro9, a bottle) or thecse ; and intermingled with the asci are filaments, enlarged and coloured at the ends (para- physes), which are probably abortive asci. In systematic works upon the Lichens, the saucers and their contents are included in the term apothecia, the saucer alone being called theexcip'ulum (excipulum, a receiver) ; the mass of asci and paraphyses forming the nucleus or thaldmium (OdXafios, a bed) . The yellow spores are very minute, each ascus con- taining eight of them, and they are divided by a trans- verse partition or septum. Near the margins of the lobes of the thallus are small dark points. These are the pouting mouths of little capsules (spermogonia) sunk in the substance of the thallus, and containing numerous filaments, GRAPHIDE^E. 93 terminated by very minute stick-shaped bodies (sper- matia), which break off and escape through the orifices of the capsules. These are probably the representa- tives of the anthers of flowering plants and of the antheridia of the ferns. They are, however, so dif- ficult to find and examine, that I must refer to the Dictionary for a further description and figures of them. LECIDIN'E^;. This family contains the genus Cladonia, three or four species of which are common on boggy heaths, banks, &c., viz. Cladonia coccifera, the Scarlet Cup-moss (PL II. fig. 21); C. pyxiddta, the Common Cup-moss (PL II. fig. 23) ; C. vermicu- Idris (PL II. fig. 22) ; and C. rangiferma, the Rein- deer Moss (PL II. fig. 24) . The thallus of these Lichens forms little rounded irregularly overlapping scales, with scalloped edges, overgrowing the surface upon -which the Lichens are found. The fruit-stalks, or podetia (irovs, a foot), are hollow (fistulose), and either simple and dilated into cups (PL II. fig. 23), or branched with the corners or angles between the adjacent branches perforated. The apothecia in the young state resemble those of Parmelia on a small scale; but as they approach maturity, the centre becomes pushed up, so that the spore-layer is extended over the ends of the stalks. In C. coccifera and pyxidata the cups are proliferous at the margin ; i. e., branches upon which the apothecia are placed arise from it. The asci and paraphyses are very minute, but do not differ essentially in structure from those of Parmelia. In C. vermicularis the podetia are pointed and more solid than in the other species, the apothecia forming very minute spots at their apices. GRAPHID'E^E. To this family belong Gr aphis scripta (PL II. fig. 26) and Opeg'rapha betulina (PL II. fig. 30). These little Lichens are easily overlooked, from the thin and but slightly raised thallus being only visible to the naked eye as a discoloration of the bark 94 LICHENS. of the trees upon which they grow ; while the fructi- fication is very minute, forming little black streaks or lireVla (lira, a furrow), irregularly arranged, and resembling somewhat the letters of some of the Ori- ental alphabets. In Gr aphis script a (fig. 26) the thallus is thin, somewhat membranous, smoothish, shining, greyish white, and faintly bordered with black. The lirellse (fig. 27) are partly sunk in the bark, winding and narrow, some being simple, others branched; and they are surrounded by a raised border, formed by the thallus. The lirellse are lined at the sides with a black (carbonaceous) layer or excip'ulum, within which are situated the asci and paraphyses. The spores (PL II. fig. 29) are 8-cleft, the segments being again divided longitudinally into little spores or sporidia. In Opegrapha betulina (PL II. fig. 30), which is found on the bark of the birch-tree, the thallus is thin, dirty yellowish white, bordered with black. The lirellse (figs. 31, 32) are mostly simple, without a raised border of the thallus, and the excipulum forms a complete lining to them. The spores (fig. 33 a] are 3-cleft, and taper at the ends. CALICI'EJE. Calidium claveVlum (PL II. fig. 6) is a pretty little Lichen, growing upon old boards and farm-buildings. The thallus is granular and greyish white. The apothecia (fig. 7) are stalked and black, but of a lighter colour than the mass of spores form- ing the nucleus. The spores are very minute, black, oblong, and divided by a transverse septum. The Lichens are divided into two Orders, according to whether the apothecia are open before the spores are ripe, as in the species noticed above, or whether the apothecia only open to discharge the ripe spores. The first Order forms the Gymnocar'pi (yt>/-tvo?, naked, /capTrbs, fruit) ; the second forms the Angiocar'pi (dy- yeiov, vessel, capsule). LICHENS. 95 Preservation. The Lichens are readily preserved, on account of their dry nature ; they need simply be kept in a dry place, and glued to pieces of card. If room is an object, they may be dried under pressure, as in the case of the flowering plants. When re- moistened, the minute structures may be easily made out by sections. The smaller ones may be mounted dry, in cells made of the wax cement (p. 16). The minute structures keep well in chloride of calcium or glycerine. 96 FUNGI. CHAPTER IX. FUNGI. THE Fungi form the lowest class of plants : as ex- amples of them, may be mentioned mushrooms, toad- stools, puff-balls, the mould of paste, the blue mould of cheese, &c. The more minute Fungi are very common, forming beautiful microscopic objects, al- though they are rarely studied by the microscopic observer. Fungi live usually upon rotting or decaying vege- table substances, as rotten wood, the dead leaves and stems of plants, &c. ; but sometimes they are found upon living plants, and some of them exist upon de- caying animal matters, and even in living animals. Fungi exhibit no separation of root, stem, or leaves, as exists in the higher plants ; nor do they contain chlorophyll, the presence of which is so generally associated with the idea of a plant. But they consist of aggregations of mostly elongate cells, forming branched and interlacing colourless fibres, buried like roots in the substance (matrix) upon which they grow, and from which they derive their nourishment ; this portion of the Fungus is called the mycelium (JJLVKTIS, a fungus). The portion of the Fungus projecting beyond the surface of the matrix is the fructification ; and this is the part usually called the fungus, the mycelium being overlooked by a casual observer. So that here we have a character distinguishing the Fungi from the Lichens, which derive their nourish- ment from the air, and not from the matrix. The absence of the green cells, or gonidia, forms another PLATE VII. [PAGE 96.] FUNGI. Fig. 1. Agaricus micaceus : a, gills. 2. Agaricus campestris : a, spores ; b, basidia. 3. Physarum album, on a piece of stick. 4. Physarum album, spores. 5. Uredo segetum, spores. 6. Uredo caries, spores. 7. Uredo Candida, on leaf of Shep- herd's ~Pu.T8,e(Capsella) : s, spores. 8. JEcidium grossularice, sorus. 9. jEcidium grossularice : p, spore- capsules (peridia): s, anther- capsules (spermogonia). 10. Nemaspora crocea : a, spores. 11. Tonda casei. 12. Torula herbarum, on a piece of stick. 13. Torula herbarum, spores. 14. Phragmidium bulbosum, on bram- ble-leaf. 15. Phragmidium bulbosum, stylo- spores and paraphyses. 16. Puccinia graminis, on a piece of straw. Fig. 17. Puccinia graminis, spores. 18. Sporocybe alternata, filament and 19. Botrytis parasitica, on Shepherd's Purse. 20. Botrytis parasitica, spores and fil- aments. 21. Rhinotrichum, species of. 22. Rhinotrichum, heads of spores. 23. Rhinotrichum, spores detached. 24. Rhinotrichum, spores. 25. Penicillium glaucum, 26. Penicillium glaucum, ,head of spores. 27. Coremium leucopus. 28. Tuber cularia vulgaris. 29. Tubercularia vulgaris, divided re- ceptacle. 30. Tubercularia vulgaris, filaments. 31. Tubercularia vulgaris, spores. 32. Spharia fragiformis. 33. Trichothecium roseum, on a piece of stick. 34. Trichothecium roseum. 35. Trichothecium roseum, filaments and spores. Plate VII I.onSon,-John Mm Fee HYMENOMYCETES. ' 97 character by which the nearly allied members of this class of plants can be distinguished. The fructification of the Fungi occurs in two dis- tinct forms, in one of which the seeds or spores are naked, and situated at the ends of slender cells or filaments, whilst in the other the spores are contained in usually flask-like cells, called asci, similar to those occurring in the Lichens. In a few Fungi, antheridial organs, called spermogonia, as in the case of the Lichens, have also been detected. The Fungi are divided into six Orders, from each of which a few species may be selected to illustrate their structure more in detail. HYMENOMYCE'TES (v^v, membrane, ILVKVIS, fungus). This is the highest Order of Fungi, containing a large number of genera and species ; as examples of which may be mentioned the common Mushroom, Toad- stools, &c. Their general structure may be illustrated by the examination of the common Mushroom (Agai^icus campes'tris) ; the species figured (PL VII. fig. 1), how- ever, being Agaricus micdceus, which is common at the root of trees, the bottom of decaying posts, &c. The vegetative part of the fungus consists of a cotton-like mycelium, which is composed of slender, colourless, interwoven filaments, popularly known as the spawn. The portion commonly called the mush- room corresponds to the fructification, and consists of certain parts visible to the naked eye. These are an expanded portion at the top, forming a hemi- spherical cap, receptacle, or pileus (pileus, a cap), and a stalk, or stipes, upon which the cap is supported. On the under surface of the cap are a number of nearly parallel, radiating, dark-coloured plates or gills, somewhat resembling the gill-plates of a fish. The dark colour of the gills arises from the presence of the spores, which are coloured, although in some species they are white. The surface of the gills, upon K 100 FUNGI. caying stems, sticks, &c. The mycelium consists of inconspicuous, fine filaments, which run beneath the epidermis and bark of leaves and stems, or exist in the intercellular passages, the fruit bursting through the surface. The spores are short-stalked, forming stylospores (o-ruXo?, stalk, O-TTO/OO?, seed) or conid'ia (tcov&iov, little dust). But there is great confusion in the descriptions of the spores of the same Fungus by different botanical authors, some describing the fruit (in PI. VII. fig. 15, for instance) as composed of rows of spores, while others regard it as forming a single septate (septum, a partition) or partitioned spore. Torula herbdrum (PI. VII. fig. 12) is very common on the decaying stems of plants, especially those be- longing to the Parsley order (Umbelliferse), forming greenish-black streaks or patches. The spores (fig. 13) are grouped into chains or beaded (moniliform) rows, with very short stalks, and these are crowded to form the black patches visible to the naked eye. Under the microscope the spores appear of a brown colour. Torula cdsei (PL VII. fig. 11) forms reddish or white patches upon decaying cheese. It consists of branched, interwoven, tufted filaments (flocci), with comparatively large spherical spores arranged in rows at their ends. Nemas'pora crocea (PL VII. fig. 10) is a very curious member of this Order, and is found upon decaying beech-sticks. It appears as an orange-coloured tendril- like gelatinous mass of spores, bursting through a little pore on the surface of the bark. The spores (fig. 10 a) are very minute, slender, and curved, and under a high power appear jointed. dEcid'ium grossuldriae (PL VII. fig. 8) is found very commonly on the leaves of the gooseberry-bush. It forms to the naked eye oval or rounded spots (sori), of a red colour ; and on close examination, the spots appear dotted with yellow points. Each point is the rifice of an open capsule ( peridium), which has burst CONIOMYCETES. 101 through the epidermis of the leaf (PI. VII. fig. The capsules are split or lacerated at the margins, and form little cups containing the spores. The spores are very minute, yellow, and are arranged in closely packed moniliform rows. The red colour depends upon the altered chlorophyll of the leaf. On the leaves containing the spore-capsules or peridia will be found smaller, brownish-yellow capsules (spermo- gonia) partly imbedded in their substance (PL VII. fig. 9 s). These contain minute filaments (sterigmatd], terminated by short rows of rounded cells (spermatia], which are supposed to exert an antheridial function. The species ofJEcidium are very numerous, and many of them are extremely common as those upon the nettle, the barberry, the dandelion, the wood-ane- mone, the violet, and buttercups. The groups of capsules form exquisite opake objects under a low power of the microscope. PhragmicFium bulbosum (PL VII. fig. 14) is another very beautiful coniomycetous Fungus. It forms little reddish, afterwards sooty dots upon the under surface of the leaves of various species of Bramble (Rubus) . The oblong spores (fig. 15) are from 2- to 4-septate, and stalked, the stalks being swollen or bulbous at the base. The spores, which appear brown when magnified, are covered with little knobs (tuberculate) on the surface; and the uppermost little spore or sporidium is terminated by a minute point (apiculate) . Among the spores are numerous barren filaments or paraphyses. Pucciriia gram'inis (PL VII. fig. 16) is to be found everywhere upon damp rotting straw, and upon grasses. It forms sooty irregular streaks, consisting of densely crowded, one-partitioned (uniseptate) spores (fig. 17), which appear brown under the microscope. This Fungus is sometimes called "mildew." There are numerous other species of Puccinia which occur upon common plants. K3 102 FUNGI. Uredo seg'etum is the " smut" of wheat, barley, and oats a fungus too well known to the farmer. It forms sooty masses, bursting through the epidermis of the stalk and ears of the corn, and soiling the fingers when handled. The spores (PL VII. fig. 5) ar^ exceedingly minute, and the stalks are so slender and loosely connected with them that they are not readily detected. Under the microscope the spores appear brown and faintly dotted, this appearance arising from a reticulated structure of the surface, similar to that of the poppy-seed on a very small scale. Uredo caries is the " bunt " of corn. It grows within the grain, filling it with a sooty, foetid mass. The spores (PL VII. fig. 6) are considerably larger than those of the last species, and their surface is distinctly reticulated. They are attached to the fila- ments of the mycelium, as in Uredo segetum. The spores of both these species of Uredo may be found in most kinds of flour and bread, especially in those of inferior quality. Uredo can'dida (PL VII. fig. 7) is another species, forming white dots upon the leaves of the common Shepherd's Purse (Capsel'la bursa pastor' is] which is easily recognized by the form and arrangement of the pods (fig. 19). The spores (s) are rather large and white. Other species of Uredo are very common upon numerous species of weeds or wild flowering plants ; and they so closely resemble each other that, when one is known, the others are easily recognized. Usually each species occurs upon a distinct species of plant, as is the case with parasites generally. In many of them the spots (sori) exhibit a thin membrane cover- ing the spores, which bursts down the middle, so as to bear some resemblance to a capsule. But there is no true capsule, the membrane consisting of the epider- mis of the leaf or stalk of the plant, which is raised and HYPHOMYCETES. 103 torn by the expansion of the growing fungus ; so that the peridium is spurious, as belonging to the matrix, and not to the fungus. It may be mentioned here that the so-called species of Uredo are not truly distinct species, but are the forms of species of Puccinia } Phragmidium, &c. ; so that the latter genera have two kinds of fruit, one of which is a Uredo, the other a Puccinia. But I must refer to the Dictionary for further details upon this point. HYPHOMYCE'TES (vcfxia), to weave, fjLV/crjs, fungus). In this, the 4th Order of Fungi, are contained many of the commonest moulds which are found growing upon decaying substances, and sometimes upon living plants. The mycelium creeps among the particles of the substance, or the elements of the tissues, upon which the Fungus lives, in the form of slender threads or filaments. The spores, which are either simple or partitioned (septate), and naked, occur either singly or in rows at the ends of fine interwoven cottony threads or floc'ci (floc'cus, a flock of wool), which are generally very evident to the naked eye. The threads support- ing the spores form the ped'icels (pedicel'lus, a little foot). In technical descriptions, these filaments, which are usually composed of cells arranged end to end, are said to be septate (PI. VII. fig. 26), and not jointed, as in the case of the filaments of the Conferva, which are constructed in a similar manner. When not septate, the filaments are said to be continuous. STILBA'CEI. To this family belongs Tuberculdria vulgaris (PI. VII. fig. 28), which is found upon decay- ing sticks and branches of trees, especially the lime- tree. It forms little firm red knobs or tubercles, each of which is a receptacle. On making a section of a receptacle (PI. VII. fig. 29), the interior is seen to be paler than the bright red surface, and a short broad stalk comes into view. The receptacle is com- posed of crowded cell-filaments, so short near the base as rather to resemble cellular tissue (fig. 30) ; but 104 FUNGI. towards the surface the filaments become extremely slender and branched ; and each branch is terminated by a minute oblong spore, or a short row of them (fig. 31). If a stick with this Fungus upon it be kept for some time in a damp place, short whitish fibres, branched at the ends, and visible to the naked eye, will be seen sprouting from around the base of the receptacle (PL VIII. fig. 1). These, when examined under the microscope, appear composed of fine filaments (PL VIII. fig. 2), resembling those of Tuber cularia, and having the minute spores at the ends. After a con- siderable time, the entire receptacle of the Tubercu- laria becomes resolved into these fibres. In this state the Fungus assumes the characters of an Isdria, a genus of a different family of Fungi (Isariacei), so that we have here an /s0H#-form of Tubercularia. Sometimes the tubercles of the Tubercularia be- come darker, almost black, harder, and granular on the surface. On making a section of them in this state, the whole of the under portion of the surface is found to contain little roundish capsules, containing asci and spores, and it constitutes Sphdriafragifor'mis (PL VII. fig. 32). As the Spharia is the more complex and highly organized condition of this Fungus, the other two conditions must be regarded as forms, and not as species of separate genera. DEMATIE'I. In this family the filaments upon which the spores are placed are not compacted as in Tubercularia, but separate ; and they are of a dark brown or black colour. Sporoc'ybe alterndta (PL VII. fig. 18) is occasion- ally found upon decaying vegetable substances, form- ing little black velvety spots or patches. The my- ceHal filaments are exceedingly minute, septate, ta- pering at the ends, and terminated by a little tuft of pear-shaped cells, from which the black simple spores arise singly. PLATE VIII. [PAGE 104.] FUNGI. Fig. 1. /sana-form of Tubercularia. 2. /sana-form of Tubercularia, fila- ments. 3. Aspergillus glaucus. 4. Aspergillus glaucus, filaments and heads of spores; a, separate spores. 5. Aspergillus glaucus, head of spores. 6. Peziza omphalodes. 7. Peziza stercorea. 8. Peziza stercorea, cup (receptacle). 9. Peziza stercorea, asci and para- 10. Peziza stercorea, divided recepta- cle. 11. Peziza stercorea, bristles. 12. Dothidea typhina, on leaf-stalk of grass. 13. Dothidea typhina, surface of patch (stroma). 14. Dothidea typhina, capsules (peri- thecia). 15. Dothidea typhina, ascus contain- ing spores. 16. Sphceria rubella, on nettle-stem. 17. Sphceria rubella, asci. 18. Sphceria rubella, capsules (peri- thecia). 19. Sphceria rubella, ascus and spores. 20. Sphceria buttata, on piece of stick ; Fig. 20 a, section of tubercle (recep- tacle). 21. Sphceria buttata, asci and spores. 22. Sphceria complanata, on piece of stick. 23. Sphceria complanata, tubercles (re- ceptacles). 24. Dothidea ulmi, on elm-leaf. 25. Dothidea ulmi, asci. 26. Dothidea ulmi, section of recep- tacle. 27. Dothidea ulmi, spores. 28. Chcetomium elatum ', 28 a, spores ; 28 b, filaments. 29. Chcetomium elatum, on piece of stick. 30. Hysterium fraxini, on piece of stick. 31. Hysterium fraxini, receptacle. 32. Hysterium fraxini, ascus with spores. 33. Erysiphe guttata, on hazel-leaf. 34. Erysiphe guttata, capsule. 35. Erysiphe guttata, capsule (concep- tacle) with fulcra. 36. Mucor mucedo : a, columella ; s, spores. 37. Acrostalagmus : a, spores. 38. Gall on oak-leaf. 39. Gall on oak-leaf. Plate VIII. r HYPHOMYCETES. 105 MUCED'INES. Many of the Fungi belonging to this family are extremely common on decaying vege- table substances, and some are found upon living plants, to which they are very injurious. To the naked eye they usually appear as mouldy or cottony masses, either white, black, or coloured blue, yellow, &c. The spores are attached singly or in rows to branchlets arising from the ends of the filaments, so as to form little heads. Bot'rytis parasit'ica (PI. VII. fig. 19) is common upon the flower-stalks of the Shepherd's Purse, form- ing white mealy patches. The fruit-stalks are com- paratively large and thin-walled, the branchlets being slender, mostly curved, and terminated each by a large, spherical, smooth, simple, white spore. Botrytis vulgdris is also common on various decay- ing plants. Its filaments are grey, and the branchlets lobe-like ; the spores being minute, spherical, either white or greenish, and placed simply at the tips. Botrytis infes'tans is the potato- Fungus. It forms white spots upon the under side of the leaves of the potato-plant, and by some authors is considered to be the cause of the potato-disease. The filaments are branched at the ends, and terminated by single oval spores, which are apiculate at the free end, and con- tain minute little spores or sporidia. Oid'ium Tuck'eri is the well-known destructive grape- Fungus. It forms white cottony masses upon the vine and its grapes, the fruit-stalks being short and terminated by one or two end- to- end oblong spores. It appears to be the Coniomycetous form of another Fungus (Erysiphe). Trichothecium roseum (PL VII. fig. 33) is found upon rotting sticks; very frequently upon willow- baskets kept in a damp place. It forms little rounded, slightly raised, pinkish spots, less than the size of a pin's head. The branched and septate foot-stalks (figs. 34, 35) are terminated each by a little group of 106 FUNGI, obovate spores, divided by a transverse partition (uni- septate). Sometimes this little Fungus is quite white, at others greenish; when perfectly ripe, the spores become oblong. PeniciVlium glaucum (PL VII. fig. 25) is the com- mon Blue Mould found upon decaying substances, as cheese, &c., the interwoven mycelial filaments often forming large cakes or crusts upon the surface. The septate fruit-stalks (fig. 26) are fork-branched at the ends, the branchlets being terminated each by a row of very minute spherical smooth spores. On some decaying substances, as apples, gum, &c., the fruit- stalks are found aggregated into a thick stalk, the branchlets and spores forming a rounded head, so that the whole resembles a little blue mushroom (fig. 27) . In this form the Fungus has been placed in a distinct genus, and called Coremium leucopus. In other species the spores are pink and white. This little Fungus is of special interest, on account of one form of it constituting the yeast-plant, or yeast as it is commonly called. This consists of rounded or oblong cells, which grow very rapidly in ferment- ing liquids by budding the large quantity of sugar and gluten present favouring the vegetative or simple growing process, at the expense of the fructifying process. But this is only an instance of what we constantly find in flowering plants, the use of very rich soil rendering flowers double, which is really re- ducing their organs to the state of leaves. When the sugar has become exhausted, the cells of the yeast become longer and thinner, as if starved ; they then form a more recognizable mycelium, which extends to the surface of the liquid, and produces finally the fruit-stalks and the Penicillium fruit. Aspergil'lus glaucus (PL VIII. fig. 3) is an ex- tremely common mould upon cheese, jams, &c. It resembles the last in appearance to the naked eye, except that it has rather a green tinge, the heads of ASCOMYCETES. 107 fruit being much more compact and rounded. The fruit-stalks (fig. 4) are large, bulbous or inflated at the ends (fig. 5), and from the inflations arise the crowded rows of spores. The spores are rounded, and rough (scabrous) on the surface. On removing most of the spores from the head of fruit, each row of spores is found to arise from a very short stalk. Plate VII. fig. 21 represents a beautiful species of Rhinot'richum, which is found upon decaying and sickly plants, and upon rotting sticks, forming a minute grey mould. The fruit-stalks (fig. 22) are large, sparingly branched, septate or jointed, appearing brownish under the microscope. Their ends are branched, mostly biternate (fig. 23), i. e. each branch dividing into three branchlets, and these again into three still finer ones. The ends of the branchlets are inflated, and coated with little points, upon each of which a smooth white spore (fig. 24) is placed. ASCOMYCETES (aoveo?, a bottle, fjLv/cijs, fungus). The Fungi belonging to this Order are found upon the stems and leaves of plants, and upon decaying sub- stances, as dung, &c. They are usually evident to the naked eye, some even equalling the Hymenomy- cetous Fungi in size ; and many of them are brilliantly coloured. They are in general distinguishable with facility from the Fungi of other Orders, by the arrange- ment of the spores in colourless sacs or asci (PL VIII. fig. 9), resembling those noticed in the case of the Lichens. These asci are usually enclosed in a cap- sule or perithecium. The mycelium is usually buried in the matrix, so as not to be conspicuous. Helvelldcei. To this family belongs the large genus Peziza, some of the species of which are beautifully coloured, yet scarcely microscopic. Among these may be mentioned Peziza omphalodes (PL VIII. fig. 6), which forms little red masses upon damp ceilings. It does not possess the ordinary form of a Peziza, which is that of a cup fixed at the end of a stalk, like a 108 FUNGI. mushroom with the cup turned inside out, the asci lining its interior. Peziza cocciriea is not uncommon in woods. It is whitish outside, the interior of the cup being of a brilliant scarlet colour. It is from half an inch to an inch in height. Peziza stercor'ea (PI. VIII. fig. 7) is often found upon dung. The surface of the cup of this Fungus is granular and covered with bristles (figs. 8 & 11). The cup is concave (fig. 10), and lined with the asci (fig. 9), among which are simple paraphyses. The Pezizce are excellent Ascomycetous Fungi for exhibiting the asci, as they are more or less soft, and thus sections of them may be easily prepared, or they may readily be picked to pieces with the mounted needles. TTTBERACEI. In this family is contained the Truffle (Tuber cibdrium). The asci are situated upon the inner surfaces of the winding canals traversing the substance of the fleshy fruit (peridium) of which the truffle consists. PHACIDIA'CET. To this family belongs Hysterium frax'ini (PL VIII. fig. 30), which is found upon ash- twigs. The drawn-out capsules or perithecia (fig. 31) are black and elliptical, with a longitudinal fissure or orifice, and contain the asci (fig. 32) with the spores. SpH^RilcEi. Dothid'ea typhina (PL VIII. fig. 12) is a common Fungus upon the stems of living grasses. It forms an orange-coloured patch or layer encircling the stem, and covered with little dots. On making a section (fig. 14), it appears composed of a row of oblong or obovate closely placed capsules (perithecia) immersed in and continuous with a finely fibrous re- ceptacular mass (stroma). The asci (fig. 15) are very slender, arising in a tuft from the bottom of the cap- sules, and containing eight still more slender spores. Except under a very high power, the spores appear as interrupted lines running down the interior of the SPH./ERIACEI. 109 asci. The little dots visible to the naked eye are the slightly projecting mouths of the capsules, which are more distinctly seen in the magnified portion of the Fungus (fig. 13). In the young state, this Fungus is whitish. This Fungus cannot be mistaken for a UredOj two species of which occur upon grasses Uredo linedris forming yellowish-brown spots, and Uredo rubigo yellow spots. Dothidea utmi (PL VIII. fig. 24) forms black, slightly raised, and somewhat star- shaped spots upon the upper surface of the leaves of the elm. In a sec- tion (fig. 26) the cavities are seen, containing the very delicate asci (fig. 25). The spores (fig. 27) are oval, with a minute septum at one end. Sphdria rubel'la (PL VIII. fig. 16) is extremely common on the dead stems of the nettle, &c. In this Fungus the black bottle-like perithecia (fig. 18), con- taining the asci and paraphyses (fig. 17), are at first situated beneath the epidermis, through which they at length burst. The spores (fig. 19 a) are spindle- shaped, and from four- to seven- sept ate. When ripe, they escape by a hole or pore in the neck. Sphce'ria complandta (PL VIII. fig. 22) is another common species, found in hedges, on dead sticks of the softer (herbaceous) plants, as the parsley-order (Umbelliferae) . Here the minute capsules, which are scattered over the stems, are at first rounded, then flattened on the top (depressed), the neck being very minute (fig. 23). The spores in this species are exceedingly minute, oblong, and not contained in asci. Sphdria bulldta also belongs to this family. It occurs upon decaying birch-sticks, presenting to the naked eye the appearance represented in PL VIII, fig. 20. The black, raised tubercles (receptacles) in their growth burst through the bark, splitting the epidermis. They consist of a white stroma (fig. 20 a), 110 FUNGI. in which, the bottle-shaped capsules (perithecia) are immersed, the necks projecting slightly above the surface as little points (papillae). The tufted spore- sacs or asci (fig. 21), with the thread-like paraphyses, are contained within the capsules; and within the asci are the densely packed, very numerous and minute curved spores. Another species, Sphceria discifor'mis, is also com- mon on birch-sticks. It differs from the last in the tubercles being perfectly flat; the spores are also longer, straight, and spindle-shaped (fusiform). PERISPORACEI. Erys'iphe guttdta (Pl.YIII. fig. 33) is a member of this family. It appears on the under side of the leaves of the common hazel as a pale spot ; and on closely examining it with the naked eye, little black dots are seen scattered on the surface. These are the capsules (conceptacles), which are seated upon straight white filaments. The filaments (fulcra) are six or seven in number, and are placed under the capsule, like the legs of a stool (fig. 34) ; they are rigid, and swollen or inflated at the base (fig. 35). The asci are broad and short, and contain only two spores. Erysiphe maculdris is the very destructive hop-mil- dew ; and other species are common on various plants. Ch&tomium eldtum (PI. VIII. fig. 29) resembles little tufts of brown hairs, occurring upon decaying herbaceous stems. The capsule (fig. 28) is crusta- ceous, and covered with interlaced, rough, branched hairs (fig. 28 b). The spores (fig. 28 a) are oval, with a little point at one end (apiculate). PHYSOMYCETES ($i)ov, animal), and that the name of polypidom has been altered to polyzoary. ENTOMOSTRACA. 127 CHAPTER XL ARTICULATA (ARTIc'lJLUS, A JOINT). THE animals belonging to this subkingdom are spe- cially distinguished by the body and limbs being jointed : as familiar instances, may be mentioned the lobster, the wood-louse, spiders, insects, and worms. Taking the class Crustacea, to which the two first animals belong, we find interesting microscopic forms in the subclass Entomos'traca (evrofiov, insect ; ocrT/ao- KOV, shell). ENTOMOSTRACA. The animals contained in this Order are met with in every pool or pond, some of them inhabiting the sea. They are mostly minute, yet visible to the naked eye, forming specks swimming actively or leaping through the water ; hence some of them have been called water-fleas. The body of the animal is protected by a shell or car'apace, which in some consists of a single piece (PL IX. fig. 30), while in others it consists of two similar parts or valves (fig. 31), in the latter case the joints of the body being indistinctly visible. The head is furnished with usually two projecting feelers or antennae (an- ten'na, a sail-yard), one of which is uppermost or su- perior (PL IX. figs. 30, 31, 34 a], the other lowermost or inferior (figs. 30 and 34 b) ; and these are often used for swimming. The antennae are jointed, and sometimes beautifully plumose (pluma, a feather) or feathery, i. e. furnished with rows of long and very slender filaments. There are several pairs of jointed legs, some of which serve as jaws (foot-jaws), while others are finely filamentous to serve for swimming 128 ENTOMOSTKACA. and as respiratory organs (branchial feet). The four species figured are very common. Cypris tristridta (PL IX. fig. 34) is found in ponds and ditches. The carapace is bivalve, or has two valves, which are convex and oval ; and it is of a greenish colour, with three irregular dark stripes behind. The superior antennae (a) are jointed and finely fea- thery, the inferior antennae (b) having a tuft or pencil of fine filaments arising from their anterior margin. The eye is single. The animal swims steadily and freely through the water. The eggs of Cypris (PL IX. fig. 35) are often found in glasses of water containing the animals. They are rounded or oblong, of a red colour, glued together by an amorphous jelly, and adherent to pieces of stick or the sides of the glass. They are enclosed in a thick shell, which exhibits a cellular appearance in the surface view, and is striated in the side view ; so that the structure of the shell is prismatic, as in that of the oyster. When the eggs escape from the shell, they present the appearance represented in fig. 35 #, the body of the young animal being enclosed in a transparent envelope, one end of which forms a blunt protrusion; there is also a separate slender process enclosing the superior antennae. After a time, the envelope is cast off (fig. 35 b) } when the animal begins its active stage of life. Theeast-ofF envelopes (fig. 35 c), with the protruded portions wrinkled, are often found in the sediment of water containing the animals. The structure of these ova is that of what are called winter ova, which agree with the resting-spores of the lower plants or the Algae. Cy clops quadricor'nis (PL IX. fig. 30) is another common species. In this the body is closely sur- rounded by the jointed shell, as in a lobster. The superior antennae (a) are very long and many-jointed, each joint having short bristles arising from it, while the inferior antennae are short and four-jointed. There DAPHN1A. 129 is no separate head, this being united to or conso- lidated with the first joint of the thor'ax or chest, the head and thorax together comprising fonr joints. The remaining joints enclose the belly, or abdomen, which has the appearance of a tail ; but the tail is constituted by the two last parallel pieces, which are furnished with fine feathery filaments. The female is most commonly met with, and is easily known by having the egg-pouch, or ovary (o) external on each side, and filled with eggs or ova. The little Cyclops is readily recognized by its form and jerking motion through the water. Daph'nia pulex (PI. IX. fig. 31) is a very common Entomostracan, and is very well adapted to illustrate the structure, on account of its size and transparence. In this animal the body is loosely connected with a bivalve shell, which, on careful examination, is seen to be reticulated or marked with net-like lines. The superior antennae (a) are very small, placed under a small beak, and have at the end a minute tuft of hairs. The inferior antennae (b) resemble arms, being large and branched ; and by means of them the animal rows itself through the water. The structure of the eye is curious, consisting of a number of round lenses aggre- gated together, the fine muscular threads by which it is moved being easily distinguished with a high power. The legs are flattened, and furnished with elegant feathery setae (seta, a bristle), serving as gills or branchiae. They are constantly in motion, fanning the water so as to change incessantly the portion with which they are in contact. About the middle of the back is placed the little transparent heart, with its colourless blood, which may be distinctly seen beating, or contracting and dilating, in the living animal ; and between the back of the animal and the shell are seen the ova, which remain there until they are hatched. The genera and species of the Entomostraca are very numerous. Those mentioned above will serve 130 ARACHNIDA. to illustrate the general structure of the order. To distinguish the man'dibles (mandib'ula, a jaw) or proper jaws, the foot-jaws, and the branchial legs, the animals must be dissected in water with the mounted needles. The very delicate feathery fila- ments of the branchiae may be best observed when these organs are dried on a slide. The Entromostaca may be kept alive in a jar of water with water-plants for a long period. They may be removed from the water for examination by the dipping-tube, and are best observed in a live- box. ARACH'NIDA (apa^vrj, spider) is the Class of spiders, scorpions, and mites. Araneida. This Order contains the more highly developed forms of the Class, among which are the common spiders of houses and gardens ; and some of their structures are very curious and interesting. The head of spiders is united or fused with the thorax, forming one piece, which is called the ceph'alo- ihorax (tcea\rj, head, copal;, chest). The claw-jaws, or mandibles, are terminated by a curved and pointed claw, with which the spiders hold their prey. It is traversed by a slender canal, con- taining a slender tube or duct leading from a poison- gland, and opening near its point; and when the insect prey is transfixed by the mandible, the poison is pressed out and enters the wound. Near the root or base of the mandibles on each side is a jointed feeler, or pal'pus ; but spiders have no anten'nse. The eyes are simple, forming separate round shining dots, and are called ocel'li (ocel'lus, a little eye) ; they are usually placed on the top of the head, and are often arranged in a geometrical form, as a triangle, &c. The legs are four pairs ; they are hairy, and ter- minated by two or three claws, which are fringed with minute teeth, or pec'tinate. These claws serve ACARINA. 131 to comb the fibres of the web, just as we comb our hair with a common comb. The spinnerets, with which spiders form their web, are very curious organs. They are situated at the under and hind part of the body, and consist of two or three cones, or papillae, on each side. On the summits of these papillae are very numerous bristle- like tubes, through which the secretion of certain glands passes ; this secretion, when hardened by ex- posure to the air, forms the fibres of the web. On carefully examining a spider's web, the radial fibres, or those which pass from the centre to the cir- cumference, will be found to be smooth, these fibres serve to fix the web ; while the cross fibres are covered with numerous viscid globules, which serve to attach flies or other prey to them. This difference of the fibres is best observed with a hand-lens. ACARINA, or the Order of Mites. Here belongs cheese-mite, Ac? arm domes' ticus (PL IX. fig. 36) . Its body is somewhat milky white, oval, and furnished with feathery hairs. When viewed from beneath, there is seen a transverse line, indicating the separa- tion of the thorax from the abdomen ; and another line in front of this, with four minute tubercles, from each of which arises a hair. The head is pointed and beak-like, forming a ros'trum (rostrum, a beak), con- sisting of two mandibles pressed together ; these can only be seen to be separate when dissected apart with the mounted needles. Each mandible is chelate (jCflW) forceps), or has the form of a lobster's claw ; and beneath the two mandibles is a flat membranous under lip or labium, consolidated on each side with a palp. The legs are four pairs, as in all the Arachnida ; they are pinkish, 6-jointed, and terminated by a leaf- like sucker and a minute claw. The males (fig. 38) are smaller than the females, the fore legs being much stouter, and furnished with a blunt tooth (fig. 38 a). The eggs can often be dis- 132 ACARINA. tinguished within the body of the female (fig. 36) ; they are oval and granular. Another species of Acarus, A. sac'chari, is found abundantly in ordinary moist sugar. If a little of the sugar be placed in a wine-glass, some water added, and the mixture be stirred until the sugar is dissolved, the Acari will be found both in the sediment and floating on the surface. A somewhat larger member of the order occurs as a parasite upon a species of Dung-beetle (Geotrupes ster cor arms] which is vulgarly known as the Lousy Watchman. The beetle is black, shaded with purple, about three-quarters of an inch long, and is found under cow-dung. The mites cling pertinaci- ously to the under parts of the beetle, and can easily be seen with the naked eye. They are whitish, with the mandibles, the sucker, and two claws very dis- tinct ; and the palpi are unattached to the labium, or free. These mites form the species Gam'asus coleoptrator'um. Another species, Gamasus teldrim, is the red spider of the greenhouse. Trombirfwm fuliginosum (PL IX. fig. 37 d) is a common red spider of gardens. It is of a scarlet colour, appearing velvety from the presence of a dense coat of feathery hairs (fig. 37/) . The palpi of this mite are large, free, the last joint but one (PL IX. fig. 37 a) being furnished with a claw, while the last joint is obtuse, and resembles a lateral appendage. The mandibles (fig. 37 b) are furnished with a sharp curved claw. The legs are long, especially the an- terior pair, and terminated by two claws, with a de- licate sucker-like appendage (fig. 37 c). Another species of Trombidium, T. holoseric'eum, greatly resembling the last, is also found in gardens. It may be easily distinguished from the last by the club-shaped hairs (fig. 37 e) existing upon the body. The harvest-bug, which causes such irritation of the legs of persons who frequent corn-fields in the PLATE X. [PAGE 133.] INSECTS. Fig. 1. Atropospukatorius : * natural size. 2. Aphis of Geranium: a, foot; &, anal tube j c, antenna. 3. Scales on wing of Apollo-butter- fly. . 4. Lithobius forcipatus. 5. Litho1riwforcipatw,]iea&of: #, an- tennae ; b, mandibles ; c, labial palpi ; d, labium. 6. Dytiscus marginolis, head of larva. 7. Young larva of Dytiscus margi- nalis. 8. Pupa of Gnat ( Cidex pipiens). 9. Larva of Gnat. 10. Head of Gnat, male. 11. Head of Gnat, female. 12. Coccinella 7-punctata (large Lady- bird), labium of. 13. Coccinella 7-punctata, mandible of. 14. Coccinella 7-punctata, labrum of. 15. Coccinella 7-punctata, antenna of. 16. Coccinella 7-punctata, maxilla of: a, palp ; b, c, lobes of maxilla. 17. Head of Musca domestica (House- fly) : a, antenna ; b, labial palpi ; c, proboscis. 18. Head ofStomoxys calcitrans : 18 a. antenna. 19. Scales of insects : a, scale of Po- dura] b, of House-moth {Tinea Fig. vestianelld) j c, of Podura ; d, of Lepisma', e, hair of Podura', f, scale of Cabbage-Butterfly (Pontia brassicce). 20. Head of Cabbage-Butterfly: ^an- tennae broken off"; b, palp ; c f tongue (antlia)j d, club of antenna. 21. Head of human Flea (Pulex irri- tans), female ( ) : a, palpi ; b, niaxillsB. 22. Flea of the Rat (Pulex muris), male (cJ). 23. Pterostichus (Steropus) madidus; 23 a, antenna. 24. Part of leg of Pterostichus madi- dus : a, tibia j b, tarsus. 25. Labrum of Pterostichus madidus. 26. Mandible of Pterostichus madidus. 27. Labium of Pterostichus madidus : a, mentum ; b, labial palp. 28. Maxilla of Pterostichus madidus : a, claw j b, c, maxillary palps. 29. Proboscis of House-fly. 30. Larva of Flea. 31. Larva of Chironomus plumosus. 32. Foot of House-fly. 33. Eye of House-fly. 34. Leg of Ant (Formica fused) ; 34*, pectinate process. Plate X. INSECTS. 133 autumn, is also a species of Trombidium T. autum* ndle. Preparation. The organs of the mouth, &c., of the Spiders are easily prepared for examination, by care- fully pulling them off with forceps or the mounted needles, then drying them under pressure between two glass slips, macerating in turpentine, and mount- ing in balsam. Those of the Acarina should be dis- sected out with the needles, after the body has been crushed in a drop of water on a slide, and the internal substance has been gently washed away with a hair pencil. They may then be dried on a slide, with a cover laid on, and turpentine applied to the edge of the cover, balsam being added when most of this has evaporated. The various parts may also be mounted in chloride of calcium or glycerine. INSECTS. The members of the class of Insects are extremely interesting to the microscopic observer, not only on account of the beautiful structures which they present, but also from these being comparatively large, usually coloured, and easily distinguished under the lower powers. Hence they form admirable ob- jects for study to those who are but little accustomed to the use of the microscope. MYBJAP'ODA (yitu/oto?, myriad, TTOI)?, foot. This Order contains those insects which are popularly known as the hundred-legs and millepedes ; by many zoologists they have been arranged in a distinct class. The most common member of this Order is Litho- bius forcipdtus (PL X. fig. 4), which is found under stones, in cellars, and among garden-rubbish. It is of a yellowish-brown colour, with long, many-jointed, gradually tapering or setaceous (seta, a bristle) anten- nae (figs. 4 0, 5 fl), and two large and powerful mandi- bles (fig. 5 b) resembling those of the spiders. It has also a broad, notched, and toothed lower lip, or labium (fig. 5 d), above which are two toothed jaws, 134 INSECTS. or maxillae, and two lip-feelers, or labial palpi (c). The eyes consist of a group of ocelli on each side (e) . The body is protected by alternately larger and smaller dorsal plates, which are fifteen in number; and there are fifteen pairs of legs, which are terminated by a single claw. On the sides of the body will be found some oval dark-looking bodies, fringed with hairs; these are the spir'acles (spirac'ulum, a breathing-hole) or breathing-pores. They form the orifices of certain branched and transversely striated tubes, which are distributed throughout the body ; the tubes are called tracheae (trachea, the windpipe), and their walls con- tain an elastic spiral fibre which keeps them open. These parts of the insect can only be distinctly seen when the body has been slit up on the under side ; and, after washing away the animal matter with water by the aid of a hair pencil, pressed between two slides with a clip, dried, soaked in turpentine, and mounted in balsam. THYSAN^JRA (Ovaavoi,, fringe, ovpa, tail). The in- sects belonging to the genus Podura, of this Order, are very minute and difficult to examine ; but they are specially interesting, on account of the structure of their scales. They are common in gardens and cellars, under flower-pots, &c., and are about one-tenth of an inch long. They are of a brownish or silvery- leaden colour, wingless, with six legs, and when touched they leap like a flea. The leaping motion is produced by the action of the tail, which is forked and bent under the body. The body is usually covered with minute scales (PI. X. fig. 19, #, c), and these are used as test-objects. The structure of the scales varies in the different genera and species; those usually used (fig. 19 #) are stated to belong to Podura plum'bea- it appears, however, that this is not correct. The scales sold as test-objects under this name are covered with minute and short raised lines (fig. 190), arranged in irregular ANOPLURA. 135 but somewhat parallel wavy rows. It requires a good microscope and a high power to show them dis- tinctly, and they should appear perfectly black and separate. The little lines are much coarser in some scales than in others; so that there are easy and difficult scales, as they are called. The Podura may be caught by holding a sheet of paper near their haunts and disturbing them; and when they have jumped upon the paper, a slide laid upon them and gently pressed will remove some of the scales for examination. The scales should be mounted as dry transparent objects ; for if wetted, they become very transparent, and the markings appear removed, which however is not really the case. The scales of Lepisma saccharma (PI. X. fig. 19 d), a member of this Order, were formerly used as test- objects ; but they are too easily made out to serve for this purpose with modern microscopes. The in- sect is not common. The scales (fig. 19 d] exhibit continuous nearly parallel longitudinal lines or ribs. ANOPL