prominent feature, and hence termed Coniomycetes; the other, in which the threads are most noticeable, is Hyphomycetes. In the former of these, the reproductive system seems to preponderate so much over the vegetative, that the fungus appears to be all spores. The mycelium is often nearly obsolete, and the short pedicels so evanescent, that a rusty or sooty powder represents the mature fungus, infesting the green parts of living plants. This is more especially true of one or two orders. It will be most convenient to recognize two artificial sub-families for the purpose of illustration, in one of which the species are developed on living, and in the other on dead, plants. We will commence with the latter, recognizing first those which are developed beneath the cuticle, and then those which are superficial. Of the sub-cuticular, two orders may be named as the representatives of this group in Britain, these are the Sphæronemei, in which the spores are contained in a more or less perfect perithecium, and the Melanconiei, in which there is manifestly none. The first of these is analogous to the Sphæriacei of Ascomycetous fungi, and probably consists largely of spermogonia of known species of Sphæria, the relations of which have not hitherto been traced. The spores are produced on slender threads springing from the inner wall of the perithecium, and, when mature, are expelled from an orifice at the apex. This is

the normal condition, to which there are some exceptions. in the Melanconici, there is no true perithecium, but the spores are produced in like manner upon a kind of stroma or cushion

formed from the mycelium, and, when mature, are expelled through a rupture of the cuticle beneath which they are generated, often issuing in long gelatinous tendrils. Here, again, the majority of what were formerly regarded as distinct species have been found, or suspected, to be forms of higher fungi. The Torulacei represent the superficial fungi of this family, and these consist of a more or less developed mycelium, which gives rise to fertile threads, which, by constriction and division, mature into moniliform chains of spores. The species mostly appear as blackish velvety patches or stains on the stems of herbaceous plants and on old weathered wood.

Much interest attaches to the other sub-family of Coniomycetes, in which the species are produced for the most part on living plants. So much has been discovered during recent years of the polymorphism which subsists amongst the species in this section, that any detailed classification can only be regarded as provisional. Hence we shall proceed here upon the supposition that we are dealing with autonomous species. In the first place, we must recognize a small section in which a kind of cellular peridium is present. This is the Ecidiacei, or order of "cluster cups." The majority of species are very beautiful objects under the microscope; the peridia are distinctly cellular, and white or pallid, produced beneath the cuticle, through which they burst, and, rupturing at the apex, in one genus in a stellate manner, so that the teeth, becoming reflexed, resemble delicate fringed cups, with the orange, golden, brown, or whitish spores or pseudospores nestling in the interior.* These pseudospores are at first produced in chains, but ultimately separate. In many cases these cups are either accompanied or preceded by spermogonia. In two other orders there is no peridium. In the Caomacei, the pseudospores are more or less globose or ovate, sometimes laterally compressed and simple; and in Pucciniai, they are elongated, often subfusiform and septate. In both, the pseudospores are produced in tufts or clusters direct from the mycelium. The Caomacei might again be sub.

* Corda, "Icones Fungorum,” vol. iii. fig. 45.

divided into Ustilagines* and Uredines. In the former, the pseudospores are mostly dingy brown or blackish, and in the latter more brightly coloured, often yellowish. The Ustilagines include the smuts and bunt of corn-plants, the Uredines include the red rusts of wheat and grasses. In some of the species included in the latter, two forms of fruit are found. In Melampsora, the summer pseudospores are yellow, globose, and were formerly classed as a species of Lecythea, whilst the winter pseudospores are brownish, elongated, wedgeshaped by compression, and compact. The Pucciniai ‡ differ primarily in the septate psendospores, which in one genus (Puccinia) are uniseptate; in Triphragmium, they are biseptate; in Phragmidium, multiseptate; and in Xenodochus, moniliform, breaking up into distinct articulations. It is probable that, in all of these, as is known to be the case in most, the septate pseudospores are preceded or accompanied by simple pseudospores, to which they are mysteriously related. There is still another, somewhat singular, group usually associated with the Pucciniai, in which the septate pseudospores are immersed in gelatin, so that in many features the species seem to approach the Tremellini. This group includes two or three genera, the type of which will be found in Podisoma.§ These fungi are parasitic on living junipers in Britain and North America, appcaring year after year upon the same gouty swellings of the branches, in clavate or horn-shaped gelatinous processes of a yellowish or orange colour. Anomalous as it may at first sight appear to include these tremelloid forms with the dust-like fungi, their relations will on closer examination be more fully appreciated, when the form of pseudospores, mode of germination, and other features are taken into consideration, especially when compared with Podisoma Ellisii, already alluded to. This family is technically characterized as,

*

66 Tulasne, Mémoire sur les Ustilaginées," "Ann. des Sci. Nat.” (1847), vii.

12-73.

✦ Tulasne, “Mémoire sur les Urédinées,” “Ann. des Sci. Nat.” (1854), ii. 78. Tulasne, "Sur les Urédinées,” ""Ann. des Sci. Nat." 1854, ii. pl. 9.

§ Cooke, M. C.," Notes on Podisoma," in "Journ. Quek. Micr. Club," No. 17 (1871), p. 255.

Distinct hymenium none. Pseudospores either solitary or concatenate, produced on the tips of generally short threads, which are either naked or contained in a perithecium, rarely compacted into a gelatinous mass, at length producing minute spores =CONIO

MYCETES.

The last family of the sporifera is Hyphomycetes, in which the threads are conspicuously developed. These are what are more commonly called "moulds," including some of the most elegant and delicate of microscopic forms. It is true of many of these, as well as of the Coniomycetes, that they are only conidial forms of higher fungi; but there will remain a very large number of species which, as far as present knowledge extends, must be accepted as autonomous. In this family, we may again recognize three subdivisions, in one of which the threads are more or less compacted into a common stem, in another the threads are free, and in the third the threads can scarcely be distinguished from the mycelium. It is this latter group which unites the Hyphomycetes with the Coniomycetes, the affinities being increased by the great profusion with which the spores are developed. The first group, in which the fertile threads are united so as to form a compound stem, consists of two small orders, the Isariacei and the Stilbacei, in the former of which the spores are dry, and in the latter somewhat gelatinous. Many of the species closely imitate forms met with in the Hymenomycetes, such as Clavaria; and, in the genus Isaria, it is almost beyond doubt that the species found on dead insects, moths, spiders, flies, ants, &c., are merely the conidiophores of species of Torrubia.*

The second group is by far the largest, most typical, and attractive in this family. It contains the black moulds and white moulds, technically known as the Dematiei and the Mucedines. In the first, the threads are more or less corticated, that is, the stem has a distinct investing membrane, which peels off like a bark; and the threads, often also the spores, are darkcoloured, as if charred or scorched. In many cases, the spores are highly developed, large, multiseptate, and nucleate, and sel

Tulasne, L. R. and C., "Selecta Fungorum Carpologia," vol. iii. pp. 4-19.

In

dom are spores and threads colourless or of bright tints. the Mucedines, on the contrary, the threads are never coated, seldom dingy, mostly white or of pure colours, and the spores have less a tendency to extra development or multiplex septation. In some genera, as in Peronospora for instance,*

PIG. 40.--Rhopalomyces

candidus.

*

a

secondary fruit is produced in the form of resting spores from the mycelium; and these generate zoospores as well as the primary spores, similar to those common in Alge. This latter genus is very destructive to growing plants, one species being the chief agent in the potato disease, and another no less destructive to crops of onions. The vine disease is produced by a species of Oidium, which is also classed with Mucedines, but which is really the conidiiferous form of Erysiphe. In other genera, the majority of species are developed on decaying plants, so that, with the exception of the two genera mentioned, the Hyphomycetes exert a much less baneful influence on vegetation than the Coniomycetes. The last section, including the Sepedoniei, has been already cited as remarkable for the suppression of the threads, which are scarcely to be distinguished from the mycelium; the spores are profuse, nestling on the floccose mycelium; whilst in the Trichodermacei, the spores are invested by the threads, as if enclosed in a sort of false peridium. A summary of the characters of the family may therefore be thus briefly expressed :

Filamentous; fertile threads naked, for the most part free or loosely compacted, simple or branched, bearing the spores at their apices, rarely more closely packed, so as to form a distinct common stem HYPHOMYCETES.

Having thus disposed of the Sporifera, we must advert to the two families of Sporidiifera. As more closely related to the Hyphomycetes, the first of these to be noticed is the * De Bary, A., "Recherches sur les Champignons Parasites," in Sci. Nat." 4me sér. xx. p. 5; Grevillea," vol. i. p. 150.

66

"Ann. des