These are exclusively parasitic fungi, and grow within the tissues of various flowering plants, sometimes entirely destroying them.

As a type of this group we will select a very common one (Cystopus bliti), that is always to be found in late summer and autumn growing on pig weed (Amarantus). It forms whitish, blister-like blotches about the size of a pin head on the leaves and stems, being commonest on the under side of the leaves ([Fig. 33], A). In the earlier stages the leaf does not appear much affected, but later becomes brown and withered about the blotches caused by the fungus.

If a thin vertical section of the leaf is made through one of these blotches, and mounted as described for Mucor, the latter is found to be composed of a mass of spores that have been produced below the epidermis of the leaf, and have pushed it up by their growth. If the section is a very thin one, we may be able to make out the structure of the fungus, and then find it to be composed of irregular, tubular, much-branched filaments, which, however, are not divided by cross-walls. These filaments run through the intercellular spaces of the leaf, and send into the cells little globular suckers, by means of which the fungus feeds.

The spores already mentioned are formed at the ends of crowded filaments, that push up, and finally rupture the epidermis ([Fig. 33], B). They are formed by the ends of the filaments swelling up and becoming constricted, so as to form an oval spore, which is then cut off by a wall. The portion of the filament immediately below acts in the same way, and the process is repeated until a chain of half a dozen or more may be produced, the lowest one being always the last formed. When ripe, the spores are separated by a thin neck, and become very easily broken off.

In order to follow their germination it is only necessary to place a few leaves with fresh patches of the fungus under a bell jar or tumbler, inverted over a dish full of water, so as to keep the air within saturated with moisture, but taking care to keep the leaves out of the water. After about twenty-four hours, if some of the spores are scraped off and mounted in water, they will germinate in the course of an hour or so. The contents divide into about eight parts, which escape from the top of the spore, which at this time projects as a little papilla. On escaping, each mass of protoplasm swims away as a zoöspore, with two extremely delicate cilia. After a short time it comes to rest, and, after developing a thin cell wall, germinates by sending out one or two filaments ([Fig. 33], C, E).

Fig. 33.—A, leaf of pig-weed (Amarantus), with spots of white rust (c), one-half natural size. B, non-sexual spores (conidia). C, the same germinating. D, zoöspores. E, germinating zoöspores. sp. the spore. F, young. G, mature sexual organs. In G, the tube may be seen connecting the antheridium (an.), with the egg cell (o). H, a ripe resting spore still surrounded by the wall of the oögonium. I, a part of a filament of the fungus, showing its irregular form. All × 300.

Under normal conditions the spores probably germinate when the leaves are wet, and the filaments enter the plant through the breathing pores on the lower surface of the leaves, and spread rapidly through the intercellular spaces.

Later on, spores of a very different kind are produced. Unlike those already studied, they are formed some distance below the epidermis, and in order to study them satisfactorily, the fungus must be freed from the host plant. In order to do this, small pieces of the leaf should be boiled for about a minute in strong caustic potash, and then treated with acetic or hydrochloric acid. By this means the tissues of the leaf become so soft as to be readily removed, while the fungus is but little affected. The preparation should now be washed and mounted in dilute glycerine.

The spores (oöspores) are much larger than those first formed, and possess an outer coat of a dark brown color ([Fig. 33], H). Each spore is contained in a large cell, which arises as a swelling of one of the filaments, and becomes shut off by a wall. At first ([Fig. 33], F) its contents are granular, and fill it completely, but later contract to form a globular mass of protoplasm (G. o), the germ cell or egg cell. The whole is an oögonium, and differs in no essential respect from that of Vaucheria.

Frequently a smaller cell (antheridium), arising from a neighboring filament, and in close contact with the oögonium, may be detected ([Fig. 33], F, G, an.), and in exceptionally favorable cases a tube is to be seen connecting it with the germ cell, and by means of which fertilization is effected.

After being fertilized, the germ cell secretes a wall, at first thin and colorless, but later becoming thick and dark-colored on the outside, and showing a division into several layers, the outermost of which is dark brown, and covered with irregular reticulate markings. These spores do not germinate at once, but remain over winter unchanged.

Fig. 34.—Fragment of a filament of the white rust of the shepherd’s-purse, showing the suckers (h), × 300.