from adaptive or "homoplastic" modifications fitting the plants for the peculiar conditions of their life. The last-mentioned order has but very slight affinity with the two first, and indeed the group hung but loosely together, and is now generally abandoned; all three orders are doubtless dicotyledonous with well-marked affinities to other orders of that class.

Before we examine the peculiarities in question, it will be as well to recall to the mind of the reader the usual constitution of an ordinary phanerogam. Functionally, all its organs are either nutritive or reproductive; that is, they are employed in processes which either contribute directly to the maintenance of the life of the individual to which they belong, or are directed towards the formation of new individuals in an embryo state which shall reproduce the species. The former system of plant-organs includes the green leaves spread out in the sunshine, the absorbent root-hairs buried in the soil or immersed in water-both food-obtaining organs, and both periodically renewed-and the whole system of axes, stem, branches and root, which carry these temporary structures, and are themselves of more solid and durable make, fitted to be not only supports, but also channels through which pass in various directions, according to the necessities of the plant, the sap and other juices. The structure of the widely separated leaves and roothairs is very different, and adapted in either case to the surrounding conditions. The delicate walls of the latter readily allow the passage of fluid, and through them is constantly streaming into the plant a current of water charged with carbon-dioxide, and containing, copiously diluted, the varied mineral and nitrogenous constituents of the soil needed by the plant. This fluid is then transmitted upwards from cell to cell, through the structure of the axial system, and is especially drawn to its extremities and to the leaves where growth is going on rapidly, and evaporation is great. The skin which covers these parts is, on the leaves especially, provided with minute orifices (stomata), so that the atmospheric air freely passes into the loosely-built cell-structure below. The walls of these cells are thin; the substance chiefly needed for the formation of plant-structures is carbon, and this is one constituent of carbon-dioxide, a gas which always exists in the atmosphere in small quantity as the result of combustion and animal respiration. This gas the leaf-cells extract from the air, it passes through the walls, and is reduced to its elements, almost the whole of the oxygen being returned to the air whilst the carbon is retained and assimilated. At the same time are produced, by combination of the constituents of the sap, the organic compounds, often very complex, which are proper to the plant. These very remarkable processes occur chiefly in those parts which possess the characteristic

green colour of healthy vegetation, which is due to the presence in the cells of a substance called chlorophyll. There can be little or no doubt that this is the main agent in the assimilation of carbon. It is well known that plants deprived of sunlight develope none of this green colouring matter, and that their tissues are weak and flabby;

Pale, fleshy, as if the decaying dead

With a spirit of life had been animated.

Such plants cannot take up carbon-dioxide; on the contrary, and this is also true of the parts of all plants which are not green -as the bark and the flower-they expire that gas, and so vitiate the atmosphere like animals.

By these processes the plant increases in size and complexity, and at the same time is able to store up in its tissues starch and other carbonaceous substances to be employed as fuel during the flowering and fruiting seasons. These reproductive functions do not commence till a period pretty definite in the life of the species, when, instead of unfolding green leaves, the buds produce those variously modified organs which form the flower and the end of all of which is directed to the production of ripe seed. The process of their evolution, and the changes they pass through, are all wasteful to the plant and involve the consumption of its hoarded stores: so great an effort is flowering that it frequently causes the death of the plant.

The modifications in external anatomy which are met with in parasites are chiefly in the organs of nutrition. Disregarding altogether for the present the green parasites, such as the Mistletoe and Loranthus, the first thing that strikes the observer is the absence of leaves. True there are not wanting a few fleshy or dry scales, but of green foliage, and indeed of green colour at all, the plants are absolutely devoid. Nor can we find root-hairs; indeed, in many no root at all can be said to exist, but where there is a branched root, it possesses instead of root-hairs suckers applied to the tissues of the supporting plant. Even the supporting axis may be greatly reduced, and so the whole nutritive system be wanting, as in the Raflesiacea, where the base of the flower is in close apposition to the tissues of the alien stem upon which it grows. From this extreme case we may pass to parasites with a short axis closely covered with large scales (Cytinus), with a prostrate, half-buried rhizome, angular and branched (Hydnora) or very large and swollen (Balanophoraceae), and so reach plants with an erect, welldeveloped scale-bearing axis, as the Broomrapes (Orobanche). In another type, as Dodder (Cuscuta) and Cassytha, the axis is present in the form of long slender twining threads, upon which are placed the absorbent suckers, but scarcely a vestige of any

leaf-organs. On the other hand there is usually a great development of the inflorescence. Parasites seem, and sometimes are, all flower, and their flowers are generally handsome. The largest flower known is Rafflesia Arnoldi, a parasite on vines in Sumatra, and a profusion of blossom is produced by our more familiar Dodders and Broomrapes.

The meaning of these modifications in the structure of parasites is in part obvious. Organs for the acquirement of raw food are clearly not needed; all that is required is a union of the tissues of the parasite and nurse-plant so close that the fluid nutriment of the latter may readily pass from one individual to the other. Nor, as this pabulum is already in a somewhat thickened condition, is an extensive surface for evaporation needed. The usual absence of chlorophyll is less readily explained. It is commonly stated that the food of the parasite is already in a prepared state, and that therefore no elaboration on its part is necessary; but a very little examination is sufficient to show that this is not the case. A very large number of parasites possess an abundance of green colouring matter, and there is no reason to suppose it useless to them. But a more direct argument is found in the fact, that the juices of stock and parasite have usually an entirely different chemical composition. It must, indeed, be evident that a parasite, such as the Dodder, which often holds in its strangling embraces half-a-dozen or more different plants of various composition, must have the power of selecting what it needs from each; and when we find compounds in the parasite which do not occur in any of its victims, we are forced to allow that processes of assimilation go on in its tissues. The chemistry of these plants has been but little studied; they, however, contain stores of starch, and are generally far richer in hydrocarbonaceous principles than the species upon which they grow. Thus Orobanches and Cytinus contain abundance of oleo-resinous matter, and a species of Balanophora is said to be employed in Java to make candles. This excess of hydrocarbons is especially worth remark, as it exists along with a loss of large quantities of carbon by the constant expiration of carbon-dioxide, for in relation to the atmosphere parasites act like the coloured parts of other plants. Nearly all of them possess stomata, but in smaller numbers than foliage-bearing vegetables. Many contain strongly astringent principles, and most become black when dried.

We must then admit that the power of elaboration of new substances is possessed by parasites, and it seems likely that along with the constant expiration of carbon-dioxide, there is concurrently an absorption of the same gas in smaller quantity continually going on in the ordinary way; the amount of carbon-dioxide apparently given out would then in reality only

represent the difference between a small absorption and a great expiration. The expiration of carbon-dioxide is, of course, very exhausting to the host-plant, weakening it quite in the same way as a disproportionate and excessive production of its own flowers would do.

It is also to be noted that parasitism is a condition which may vary in degree. The plants most absolutely dependent upon others are those which, like the Rafflesiacea, are quite without provision for a separate existence; the seeds, doubtless, though this has not been observed, germinate on the branches upon which the fully developed plants are found, as is the case with those of the various green parasites. The numerous leafless root parasites present several gradations from such absolutely parasitic forms as Cynomorium and the Balanophora, through Orobanche, Harveya, Lathræa, Striga, &c., to those Scrophulariacea and Santalacea which are but slightly dependent for supplies of food on the plants to whose roots they are attached. Some of these forms are completely parasitic only in early life: this is probably the case with our Neottia (Bird's-nest) and Monotropa. In that singular stem-parasite the Dodder, the embryo, entirely without cotyledons, is able to live for a short time after germination, and rises into the air until it attaches itself to some plant, when its connection with the ground is speedily broken.

In attempting to classify the parasites all these conditions must be taken into account; but unfortunately the mode of attachment has not been carefully noted in numerous cases, whilst in several presumed parasites the actual union with another plant has never been observed. We may, however, at once divide them into the essentially distinct groups of chlorophyllous and non-chlorophyllous plants, and in classifying the latter we cannot do better than follow A. P. De Candolle, and group them into those with a single sucker (monobasic), those where the attachments are numerous (polystomal), and those where along with but one large sucker there are numerous thick fleshy roots (polyrrhizal). Further division is shown in the following table, though it must be admitted that the distinctions between the first and third groups are not very well defined:

-

*Without green leaves or chlo

rophyll:

MONOBASIC.

On stems:

Rafflesiacea.

On roots:
Cytinacea.
Balanophora, &c.

POLYSTOMAL.

On stems:

Cuscuta.
Cassytha.
On roots:

Lathræa, &c.
Cynomorium, &c.

In the plates accompanying this article are figures, drawn partly from nature and partly from various published memoirs on the subject, of non-chlorophyllous parasites, selected as examples of the various modes of attachment above mentioned. To illustrate the monobasic condition a species of Pilostyles (P. aethiopica, Welw.) has been chosen. This genus is placed in the Rafflesiaceae, an order consisting of some twenty-five species, mostly tropical, and contains a few forms, chiefly from South America, but some occurring in Asia Minor and in Africa. Each plant (see Plate XCIX., figs. 1-4) consists simply of a little unisexual flower, surrounded with several rows of bracts, and containing in its centre either a column, round the head of which is a circle of one-celled anthers (fig. 3), or a hemispherical knob, the stigma, below which is situated the unilocular ovary, containing numerous minute ovules attached to its sides (fig. 4), and succeeded by a globose fleshy berry. These little flower-plants occur in groups sessile and half immersed on the younger twigs of trees and shrubs, and, so far as known, all the species grow on Leguminosa only. They appear to originate beneath the bark, little round knobs upon which are the first indication of their presence. These increase in size, split open, and allow the contained Pilostyles to emerge. The bark thus forced up forms a prominent ridge round the parasite, and after the latter has fallen off a little cup-shaped cavity long remains (see fig. 2). The union of stock and parasite is not difficult to rupture, and the latter is readily detached, its base not penetrating into the woody structure of the branch. In the great species of Rafflesia, Sapria, and Brugmansia of the East Indies, there is, at least in the first, actual vascular connexion, the vessels of the stock passing absolutely into the base of the parasite.

Cytinus Hypocistis (also figured on Plate XCIX., figs. 5–9), is another monobasic parasite. This pretty plant is not uncommon in the South of Europe and North Africa, growing, as its name implies, on the roots of species of Cistus in the spring. The whole plant is brilliant orange-yellow, and consists of an axis clothed with overlapping scales, and bearing at the top a tuft of flowers. These are unisexual, the antheri