done in text-books, has failed absolutely. We can understand and know the pathogenic varieties only if we study simultaneously the non-pathogenic, from which the former have once originated and still always originate1 (see Pest).

The doctrine of the absolute constancy of bacteria, which for ten years was almost a dogma, is now scarcely at all seriously advocated.

II. The Nomenclature of Bacteria.

The nomenclature at present employed in bacteriologic works written by medical men is characterized by a limitless arbitrariness and inconsistency. Since these nomenclators often possess absolutely no sentiment for their arbitrariness, and the simple rules of scientific nomenclature are often entirely unknown to them, I allow myself to set down, as briefly as possible, the most essential rules, which are, by international agreement, accepted by all educated peoples, especially as they bear upon bacteriology.

1. Every plant and also every fission-fungus belongs to a species, every species to a genus, every genus to a family. 2. Following the precedent of Linné, every vegetable or animal organism, therefore every variety of bacterium, should have two Latin names: the first designating the genus to which the concerned organism belongs, which name is a substantive; the second indicating the variety (species), and being an adjective (not two) or the genitive of a substantive, only rarely a substantive in the nominative case. Thus, in the genus bacillus belong the species Bac. subtilis (hay bacillus), also the species Bac. anthracis (anthrax bacillus), and Bac. megatherium.

3. Genera must only be founded upon important morphologic characteristics; so-called "biologic genera," such as photobacterium for all light-emitting bacteria, pyobacterium for rods causing suppuration, etc., are only calculated to produce confusion.

1 If the pathologist may, perhaps, say that the pathogenic bacteria alone interest him, such a statement- -as I have often heard-from the mouth of a hygienist is almost beyond understanding.

4. As designations for species many authors have used, instead of one adjective or substantive, a plurality of adjectives, evidently with the object of furnishing a description through the name: Bac. rosettaceus metalloides, Staphylococcus pyogenes aureus, Bacillus pyogenes fœtidus, Bacillus mesentericus panis viscosi I and II. This effort can be understood, but it has been abandoned as entirely impractical by all descriptive naturalists since Linné. The name of the species should indicate only the variety unequivocally; the characterization belongs to the description. It does no harm if two or more organisms possess names that mean the same, if they do not sound alike. Besides, a Micrococcus albus, also a Micr. niveus, albissimus, candicans, and purus are entirely right; the description must give more exactly the kind of differences existing between these white cocci.

5. Improperly (i. e., contrary to the binomial rule) formed names may be replaced. We have done this with the greatest consideration for the existing name whenever possible. We have not changed names like Bac. acidi lactici, because acidum lacticum represents a single idea, and names like Sempervivum Regina Amalia, Pedicularis Friderici Augusti, Trigonella Fonum græcum, Pedicularis Sceptrum carolinum have remained, although certainly not convenient, still uncontested. Varieties which we have not studied more closely or which in our opinion should be suppressed, have not been renamed; on the contrary, Mez has conducted this renaming in the widest extent in a most acceptable manner.

6. If names are properly formed in the binomial manner and correctly published, then they must not be changed by the author himself, much less by others, even if subsequently another name appears better. Also, the reason that the name is philologically incorrect or not beautiful gives no occasion for change. Even if, for example, it was

1 We regret that we had to do this also in the case of a number of convenient and very familiar names; for example, those of Flügge. Unfortunately, also, Kruse has formed a large number of new names contrary to rule. Our names have the priority over his, because published about two months earlier, but they are to be preferred besides, in so far as Kruse's are formed contrary to rule.

literally more correct to call the genera which we call "mycobacterium" "tuberculomyces," such a proposition is absolutely unallowable. Renaming is only required if the name given was employed earlier with another signification. Thus, Cohn founded upon a certain organism the new genus streptothrix, without knowing that Corda, about thirty years previously, had conferred this name upon a fungus that is totally different from his. His new variety must, therefore, receive a new genus name, which he who first observed Cohn's oversight is justified in establishing.

7. It happens that an author differs from his predecessor regarding the bounds of the genera, that therefore he transfers a species from one genus into another, pre-existing or newly formed by himself. This is permissible; still, the designation of the species must not be changed. So we had the right, when we broke up the very large genus bacillus, following the suggestion of Hüppe, into the two genera, bacillus and bacterium, to rename a number of varieties (for example, Bacillus pyocyaneus being renamed Bacterium pyocyaneum), but we did not have a right (however much the name pyocyaneum was disliked) to rename the organism Bacterium coeruleo-viride or Bacterium Gessardi or anything else.

8. The author who names a genus places his name after it. We speak of the Bacillus Cohn, and mean the genus bacillus as Cohn established it; of the Vibrio Ehrenberg emend. Löffler, and mean the genus vibrio as established by Ehrenberg and afterward more accurately described by Löffler.

9. Whoever discovers a new species or names one not previously named lege artis, gives it a genus and a species name, and places his name after the latter. Flügge, who first named a large number of bacteria, gave, for example, the name Bacillus pyocyaneus Flügge to the long-known cause of bluish-green suppuration.

10. When one places a species in a new genus he puts his own name after the new name, thus, Bacterium pyocyaneum Lehmann and Neumann, but it is always to be recommended to add, in parentheses, the name of the author who first named the species. Therefore we always write,

where it does not become too cumbersome (in titles, etc.), Bacterium pyocyaneum (Flügge) Lehmann and Neumann. While we desire that all names which express the systematic position of the variety of bacterium shall conform to the general rules of nomenclature, still we believe that names currently used in bacteriologic literature, as gonococcus, pneumococcus, staphylococcus, tubercle bacillus, diphtheria bacillus, can be still used, but as so-called ordinary names. Thus also the strictest botanist, if not speaking in a strictly systematic sense, often speaks of the oak instead of quercus, and strawberry instead of fragaria. We must, however, strictly avoid smuggling into the literature as names of genera such names as gonococcus, etc.

III. The Formation of the Families and Genera of Fission-fungi.

The families of the fission-fungi are given fairly uniformly by the more recent investigators. Here a better division does not seem possible at present; on the contrary, regarding the genera, the comprehension is most variable. The simplest and most natural division is that of Flügge (retained by Kruse in Flügge, third edition), which so properly includes the genera micrococcus (streptococcus), sarcina, bacillus, and spirillum, but without rejecting energetically such genera as staphylococcus, or separating the causes of diphtheria and tuberculosis. more copious selection of genera is made by Hüppe, still more by Migula, and the most extensive by A. Fischer. After mature deliberation we have followed Flügge most closely as to the coccaceae and bacteriaceæ, on the other hand, the works of Löffler and Migula as to the spirillaceæ.

A

I. Family Coccaceæ Zopf, emend. Migula. Spherical Bacteria.

Cells, when free, are perfectly globular;1 division in one, two, or three directions of space, in which each spherical cell divides into halves, quarters, or eighths of a sphere,

1 Unfortunately this applies, only imperfectly to the Strept. lanceolatus and Micrococcus gonorrho

which again grow out into perfect spheres. Endospores and flagella very rare. Before division the cells may be one and a half times as long as broad, faint staining then revealing an unstained line of division.

1. The cells divide (almost) only in one direction of space at right angles to the direction of growth, so that if the products of division remain attached, they form (especially in bouillon) shorter or longer rosary-like chains, the chains often consisting of distinct pairs of cocci. Under certain circumstances there are only (or largely) pairs of cocci instead of chains. Streptococcus Billroth. 1

2. The cells regularly divide, at least on the most suitable nutrient medium (hay decoction), in three directions of space, 2 and remain united in larger or smaller cubical family groups. Sarcina Goodsir.

3. The cells divide irregularly in various directions, so that there occur single cocci, single groups of from two to four cells, and, finally and preponderantly, irregular grouped bunches. Here belong all forms that do not appear as undoubted streptococci or sarcina. Micrococcus Cohn.

The recognition of these three genera of cocci is largely artificial, and there occur perfect transitions.

The genus Staphylococcus Ogston has no botanical rights, for the property of forming "grape-like" clusters is possessed at times by all varieties described to-day as micrococci. The name staphylococcus does not primarily designate any "new" genus. Ogston found (microscopically) two forms of micrococci in pus (without cultivating them), grape cocci and chain cocci, and designated them by the well-chosen names of Staphylococcus and Streptococcus (Billroth). Rosenbach later cultivated the varieties

1 Here belongs Leuconostoc Cienc., which is only a streptococcus with at times enormously thick capsules (see below). Also part of the "diplococci" are naturally included here.

2 The varieties which, by division in two planes at right angles to each other, form flat groups, and which are described by authors as pediococcus, merista, merismopedia, we leave among the micrococci. Since even the genus Sarcina is separated with difficulty, we do not recognize the need for the genera planococcus and planosarcina of Migula, which are founded upon one or two flagellated varieties, especially as the formation of flagella varies (see below).