Regarding the forms designated as "Leptothrix buccalis," which occur frequently in the mouth, especially in deposits on the teeth, little can be said that is satisfactory, since cultures have almost always failed.

Miller (Die Bakterien der Mundhöhle, II. Aufl., Berlin, 1894) appears to have cultivated no leptothrix. He cites the following uncultivated leprothrices, briefly and very insufficiently characterized:

Leptothrix gigantea Miller. Threads fixed at one end, small to very thick, with or without distinct septa. Iodin reaction?

Leptothrix maxima buccalis Miller. Jointed threads 1 to 1.3 μ thick, without iodin reaction.

Bacillus maximus buccalis Miller. Like the preceding, but with iodin reaction. It is not stated why this species is designated bacillus while the former are designated leptothrix.

Leptothrix innominata Miller is said to present slender, 0.5 to 0.8 μ thick, tangled, unsegmented, often wavy or bent threads, which sometimes stain violet with iodin.

Arustamow (C. B. vi, 349) described two unnamed leptothrices of the mouth, cultivated by him, both of which grow at incubator temperature, No. 1 being an exquisite anaerobe, No. 2 an outspoken aerobe. The agar growth corresponds in some measure to that of the L. epidermidis; potato and gelatin growths are not described.

Dobrzyniecki (C. B. XXI, 225) has described in detail a Leptothrix placoides alba, which was successfully cultivated aerobically on gelatin. The growth at first resembled anthrax, and then liquefied. The growth on agar was slow-hard, dense colonies being formed. The organism presents long, jointed threads, with a tendency to form tangles, and it stains blue with iodin and iodid of potassium solution and a little lactic acid. Stains by Gram's method. Is not motile.

Flexner has isolated an interesting organism from a rabbit which died of puerperal infection. It occurs in long threads, is always free of spores, has no motion, and does not branch. It is pathogenic, but very difficult to cultivate outside of the body. He names it Bacillus (Leptothrix?) pyogenes filiformis Flexner (Jour. of Exp. Med., Vol. I, 211, 1896).

Long and rather thick (1 to 5 ) unbranched threads, without membranes, and quiet or with a gliding motion. When fresh, there are in part no partitions, but often numerous very highly refracting bodies may be recognized. The granules consist of sulphur, which may especially be recognized if the threads are previously allowed to dry, and also by their solubility in H,S. In this way also the previously indistinguishable transverse partitions become apparent. According to Winogradsky, the breaking up of

the threads into smaller pieces, which later grow out, is the only method of multiplication. The statements of Zopf regarding other forms in the cycle of development of Beggiatoa are opposed by Winogradsky.

According to the old idea, Beggiatoa formed HS and sulphur from sulphates, and was the cause of HS being present in sulphur springs. According to Winogradsky, on the contrary, it is dependent for its nourishment upon the preexisting H,S, which it transforms into sulphur. (See Untersuchungen über Schwefelbakterien, C. B. II, 590.)

Fig. 25.-Beggiatoa alba Vauch (after Zopf).

If

B. alba is found especially in foul slime and dirty water; also sometimes as isolated individuals in pure water. abundant, they form whitish films.

B. nivea Rabenhorst is recognized in sulphur springs as the principal constituent of the slime in the spring.

B. roseo-persicina. Zopf. (Die Spaltpilze, 3. Aufl.) Bacterium photometricum Engelmann (Pflüg. Arch., Bd. 30, 95). This variety is very striking because of its rose-color. In the cooler parts of the year it spreads widely along the banks of small streams, pools, etc. It is always a sign of

contamination of water, but not specific (perhaps from sulphite wood-pulp factories or the like). According to Winogradsky, Zopf has improperly included in this variety a large number of other rose-colored inhabitants of water. Jegunow (C. B. II, 279) has reported many interesting observations regarding another sulphur bacterium, whose classification is still undetermined, but it most likely belongs to the spirilla.

Crenothrix polyspora. Ferd. Cohn.

(Cohn's Beiträge, Bd. 1, H. II, 130.)

Long, rigid, unbranched threads, consisting of a single row of low cells, unpigmented, included by a membrane which is very thin at the younger parts of the thread and thick at the older parts. The membrane is a product of the cuticle of the cell. In the membrane is deposited

Fig. 26.-Crenothrix polyspora. Cohn.

some iron hydroxid or carbonate, which stains it brown. Sometimes also the membranes for considerable lengths are surrounded by a yellow, ferruginous mass with a luster like oil, so that macroscopic, brownish flakes appear.

The thickness of the threads varies from 1.5 to 5.2 μ, it often being easily recognized that the older part of the

thread (where it is fixed) is wider and stronger. Also the height of the individual cells varies from one-half to four times the thickness, square forms being most common (Fig. 27, a and e).

Sometimes the terminal cell of a thread is large and oval (like a spore), in which case a deeper cell grows out laterally.

Propagation occurs through a peculiar breaking up of the cells at the end of a thread into fragments. Cohn distinguishes two types: (a) Formation of microgonidia: several individual cells in the thread break up by longitudinal and transverse division into not less than 16 very

small plasma spheres, which later are freed from the somewhat swollen ends of the threads and grow out into new threads (Fig. 27, d). (b) Formation of macrogonidia: several of the cells in a thread near its end are transformed by infrequent division into larger, roundish, oval or diplococci-like forms, which grow out into new threads (Fig. 27, b). One type passes over into the other.

The plant is widely distributed (especially) in waters containing iron (also in tap-water). A pure culture in the bacteriologic sense has not been obtained. According to

Rössler, cultivation readily succeeds in well-water in which pieces of brick are boiled and to which is added "some" ferrous sulphate. (Compare Ferd. Cohn, Beiträge zur

Biologie, Band 1, Heft 1, p. 108, Breslau; Rössler, Arch. f. Pharm., Bd. 233, 1895.)

Here also should be included the interesting Leptothrix ochracea Kützing, which does not belong to the leptothrices in a strict sense, because of its fully developed membrane. Winogradsky, who describes it in detail, gives the following characteristics: slender, jointed, fixed threads of bacilli, with membranes. The membrane is thick below, thin at the free ends, and the terminal rods are entirely without any membrane. The bacilli in the membrane are motile, but this motility is lost as soon as the membrane has reached a certain thickness. This variety thrives only in water containing ferrous oxid. In the metabolic process hydrated oxid of iron is deposited in the membrane. In hay decoction prepared from well-water and freshly precipitated hydroxid of iron the organism always grows easily and rapidly. It forms yellowish flakes and films, and in nature gives rise to extensive ocher deposits. (Compare Winogradsky, Bot. Zeitung, 1888, p. 261.)

Cladothrix dichotoma. Ferd. Cohn.

(Cohn's Beiträge, Bd. 1, Heft. III, p. 185.)

Long, apparently non-segmented threads, with thick or thin membranes, in part free, in part attached to putrefying algæ. Thickness 1 to 5 μ. The pseudodichotomy is especially interesting. It is dependent upon the growing of a lower segment of a thread along by the side of a higher one (Fig. 30). Pure cultures of this organism have been but little studied; we have not possessed any. According to Büsgen, the most recent investigator of this organism (Ber. der deutsch. bot. Gesellschaft, 1894, p. 147), it grows slowly and without perceptible liquefaction in gelatin containing a little meat extract.

The surface growth consists of a "round white patch," which is not elevated, and from which, as from the stab, after a few days delicate threads grow out.

The threads have thin membranes when grown on gelatin and thick ones in dilute solutions of meat extract. The membrane is patent at the end of the threads, and through this opening, as also through irregularly occurring tears of