THE FILAMENTOUS BACTERIA OF ACTIVATED SLUDGE 
 
 HARVEY LAWRENCE LONG 
 
 THESIS 
 
 FOR THE 
 
 DEGREE OF BACHELOR OF ARTS 
 
 IN 
 
 BOTANY 
 
 COLLEGE OF LIBERAL ARTS AND SCIENCES 
 UNIVERSITY OF ILLINOIS 
 
 1922 
 
/ 922 
 L85I 
 
 
 UNIVERSITY OF ILLINOIS 
 
 .May. 24 1 92. 2. _ 
 
 THIS IS TO CERTIFY THAT THE THESIS PREPARED UNDER MY SUPERVISION BY 
 
 Har v ey_ _Lawr_e n_c e_ Lo ng 
 
 ENTITLED T _he_ ^ e nt o_u s_ _B_a c_t e r_i a _ of _ Ac t_i y a. t ed_ _S lu dge 
 
 IS APPROVED BY ME AS FULFILLING THIS PART OF THE REQUIREMENTS FOR THE 
 
 DEGREE OF 
 
 Approved 
 
Digitized by the Internet Archive 
 in 2016 
 
 https://archive.org/details/filamentousbacteOOIong 
 
TABLE OF CONTENTS 
 
 Page 
 
 I. Introduction , 1 
 
 II. Historical 
 
 1. The Biology of Various Methods of 
 
 Sewage Disposal 2 
 
 2. A Review of the Taxonomy of the Chlamedo- 
 
 bacteriaceae ...... 4 
 
 3. A Review of the Taxonomy of the Beggiato- 
 
 aceae 18 
 
 III. Material and Methods .22 
 
 IV. Observations , . 23 
 
 V. Summary , 31 
 
 VI. Acknowledgement 32 
 
 VII. Bibliography ..... 33 
 
 VIII. Plates and Descriptions .34 
 
THE FILAMENTOUS BACTERIA OF THE ACTIVATED SLUDGE 
 PROCESS OF SEWAGE DISPOSAL. 
 
 By 
 
 Harvey L. Long 
 I 
 
 INTRODUCTION 
 
 The activated sludge process of sewage disposal is 
 said to have had its inception at Lawrence, Mass. , in 1912. 
 
 In "The 44th Annual Report, State Board of Health of Mass., 
 1912" (p.275-367) a sub-chapter (p. 2S0-2S2) records the 
 results of an experiment on the aeration of sewage as an aid 
 to filtration (Porter, p. 7, sec. 1). 
 
 Previous observations on aeration led to a series of 
 experiments begun in April 1912, to test the efficiency of 
 aeration. The sides of a slate tank became covered with a 
 brown growth which removed the suspended matter and a large 
 portion of the colloids from the sewage after a few hours’ 
 treatment. Experiments were continued at the Lawrence (Mass. 1 ) 
 experimental station, and, at the present time, the principal 
 sewage studies there are concerned with the activated sludge 
 process. 
 
 Numerous municipal and private interests have been 
 conducting experiments with the activated sludge process since 
 
2 
 
 that time. The earlier municipal experimental plants are at 
 Milwaukee (1912), Houston (1915), Cleveland (1915), Worcester, 
 England (1915), Sheffield, England (1915). Several industrial 
 plants have conducted experiments on the disposal of industrial 
 wastes by the activated sludge process. More recently (1920) 
 the Chicago Sanitary District has conducted experimental and 
 semi- exper imental plants (Porter, Indexes). Since 1914, the 
 Illinois State Water Survey at the University of Illinois has 
 maintained what Porter (p.106) believes to be the first exclusive 
 activated sludge plant in the United States. Studies on that 
 plant have been conducted along many lines. 
 
 II 
 
 HISTORICAL 
 
 1. The Biology of Various Methods of Sewage Disposal. 
 
 Certain investigations have been carried out upon the 
 flora and fauna of the filters. In general, a very wide range 
 of life is represented. Dr. Parkinson (1919) at the University 
 of Birmingham found many algae and definitely isolated the 
 following from the slimy growth on filters: Stichococcus, 
 
 Chlor ella. Scenedesmus . Nitzschia, Navicula . Oscillator ia . and 
 Phormidlum . The most common insect found was the "Spring Tail" 
 Aohorutes. J.W. Haigh Johnson (1914) found the dominating 
 or ganisrr^'on sewage filters to be (a) Flies, (b) Achorutes, (c) 
 Bacillus colon and Thiothrlx nivea . In the upper layer of 
 the filter bed he lists the following organisms: Zoo gloea 
 
 rami ?;era, Sphaerotilus natans . Begyiatoa alba. Thiothrlx nivea , 
 
3 
 
 Chromatium okenii. Polytoma uvela , Saprolegnla , Hyphae of 
 Mucors . Lower down in the filter bed, the Zoogloeal masses 
 
 were fewer and Rotifers, Ciliates, Nematodes, and aquatic 
 earthworms abounded. 
 
 In the study of the biology of the sprinkling filter 
 ( N. J. Ag. Ex ten. Bull No. 352), a gradual change in flora 
 and fauna from top downward was observed, the greatest activity 
 being in the center of the bed. 
 
 Cox (.1921), in discussing the no n-bacterial population 
 of the sewage trickling filter, reports that the types of 
 organisms are very limited. He made ho attempt to determine 
 the species or family in some cases "because little practical 
 importance is attached to detailed biological data". He lists 
 three common algae, the same number of protozoa, and a few 
 worms and insects. Schi zomycetes found were Zoogloea . Spjrell a . 
 Cr eno thrix , and Beg^iatoa. 
 
 There have been a few studies made of the biology of 
 the activated sludge, but it seems that all have neglected the 
 most prominent organisms in the sludge, the filamentous forms. 
 
 Hommon (1918, p.45), in his studies on the treatment 
 and disposal of industrial wastes, ran a test with activated 
 sludge. His experimental chambers were an inch in diameter and 
 four feet long. He made counts of the animal organisms present 
 and he lists several Ciliates, Flagellates, Rotifers and molds. 
 
 He records zoogloeal masses in standard units, the highest being 
 255,000 per cc. 
 
 In a series of articles, Paul Courmont and co-workers 
 

 
 
 
 
 
 
 
 
 
 
 
4 
 
 present the results of an investigation of the bacterial flora 
 of sewage purified by activated sludge. Their study, however, 
 is confined to the determination of the reduction in count in 
 the effluent of forms that grow on the different media. 
 
 Russell (1916, p.354) made a study of the bacterial 
 flora of sewage purification by aeration in the laboratories of 
 the State Water Survey of Illinois. Again no attention was paid 
 to the filamentous forms present. Observations were made only 
 on the organisms that would grow on the ordinary media. In as 
 much as filamentous forms do not grow on the media used, they 
 were not observed. 
 
 2. A Review of the Taxonomy of the Chlamedobacteri aceae 
 
 The Chlamedobacteri aceae . a group of filamentous 
 bacteria, have long been called iron bacteria because of the 
 occasional occurence of iron oxide in connection with the 
 organisms. The deposition of iron is not common to all members 
 of the group. In certain forms the deposition of iron oxide 
 in connection with the filaments is the usual thing, in others 
 occasional, and in others never. Furthermore, there is not yet 
 a technique developed for the obtaining of the filamentous 
 forms in pure culture. The classification of these organisms 
 is based on the original and only method remaining, the morpho- 
 logical basis. 
 
 At present, there is considerable variation in the 
 description of the respective forms and considerable uncertainty 
 as to the tenability of certain genera and species. Although it 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
5 
 
 is not the task of this paper to determine taxonomic questions, 
 it will be necessary, because of the mentioned uncertainty, 
 to define the basis assumed in the determination of the particu- 
 lar forms present in the activated sludge. Because of 
 variations in descriptions, because of the scattered sources 
 of material, because much of the original literature is not 
 in English, and because, in order to identify the filamentous 
 forms, the writer has had to spend months reviewing literature 
 on the subject, he has chosen to include the essential points 
 which are concerned in identification, 
 
 Ellis has contributed perhaps the greater part of 
 the literature and added a considerable fund of new information 
 during the last quarter century, Many of his writings will 
 be referred to in their original source, but recently much of 
 his published work and some of the unpublished were gathered 
 together in one volume, "The Iron Bacteria”, published by the 
 Frederick A. Stokes Co., New York, bearing no date of publication. 
 (Elli3, a). 
 
 One of the most recent reports of studies of the 
 literature of the Chlarnedobacteriales is that of Buchanan (1916 
 a, p.301), in which he records one family and five genera. 
 Following is Buchanan’s list including an additional genus - 3. 
 
 Order II. Chlarnedobacteriales. (Or do nov.) Buchanan 
 
 (a) 1913 p.301) 
 
 Family I. Chlamedobacter iaceae, Migula, 1394, p, 237 
 
 Genus 
 
 1 . 
 
 Leotothrix, Kuet zing, 1843, p. 
 
 198 
 
 Genus 
 
 2. 
 
 Didymohelix, 
 
 Griffith, 1353, 
 
 p. 438 
 
 Genus 
 
 3. 
 
 Spirophyllum, 
 
 Ellis, 1907, (a) 
 
 p. 21 
 
6 
 
 Genus 
 
 4. 
 
 Creno thrix. 
 
 Cohn, 1370, p. 130 
 
 Genus 
 
 5. 
 
 Sphaero tilus 
 
 , Kuetzing, 1833, p.385 
 
 Genus 
 
 6. 
 
 Clono thrix. 
 
 Schorler, 1904, p.689 
 
 Buchanan describes (1918, (a)p.30l), the Chlamedo- 
 bacteriaceae as filamentous bacteria, typically water forms, 
 frequently sheathed, without true brahching, although false 
 branching may be present. The sheath is frequently impregnated 
 with iron. Conidia may be developed, but never endospores. 
 Sulphur granules or bacter iopurpurin never present. Mature 
 cells or filaments neither motile, nor protozoan-like. 
 
 Migula’s description (1900, p. 1030) follows. Cells 
 are cylindrical, and arranged in a thread with a sheath. 
 Reproduction takes place through conidia which do not pass 
 through a resting period but develop immediately. 
 
 Genus LEPTOTHRIX (Kuetzing, 1843, p.198) emend. Buchanan, 1913. 
 
 Synonyms: 
 
 Chlamsdo thrix Migula, 1900, p.1030 
 
 Lepto trichia Trevisan, 1879, p. 929 
 
 Detoniella Trevisan, 1885, p. 929 
 
 The following description given by Buchanan embraces 
 the important characters of the genus. "Filaments of color- 
 less cylindrical cells, later thicker, yellortr'or brown, 
 becoming incrusted with iron oxide. The iron may be dissolved 
 
 by dilute acid, where upon the cells show up well. Multipli- 
 cation i 3 through the division and abstriction of cells, and 
 

7 
 
 motile cylindrical swarm cells. Swarm cells sometimes germinate 
 in the sheath and give the appearance of branching. Pseudo- 
 dichotomous branching may occur." 
 
 This genus has been re-described by two men previous 
 to Buchanan: by Cohn (1S75, p. 203) and by Ellis (1907, (b) 
 p.503). 
 
 Number of species, five. 
 
 In as much as Leptothrix ochracea e is the only one 
 that is sufficiently described for practical purposes, the 
 discussion will be confined to that species. 
 
 Leptothrix Qchracea e Kuetzing, 1843, p. 198 
 
 According to Ellis, L. Ochracea is by far the most 
 widely distributed of all of the iron bacteria, occuring in all 
 parts of the world. It occurs in most iron bearing waters. 
 
 Very little is known of the internal structure of the 
 filament. In nature it becomes encrusted and is easily 
 recognized because of the permanent tubular nature of the encrusted 
 sheath with its distinctly ochraceou3 or yellow-red coloring. 
 
 The average thread measures 1 1/2 to 2 micron, but it 
 may increase to 3 and reach a length of 200 micron. Plates 
 showing Leptothrix ochracea are to be found opposite page 2, 
 
 Ellis, "The Iron Bacteria". 
 
 The writer has found young filaments in the fuscous 
 colored growths on water faucets which are kept moist constantly. 
 
 In these young filaments the cellular nature of the filament 
 inside the sheath was quite discernable without staining. The 
 cells were from 1 to 1.5 microns broad and from two to thr ee 
 
8 
 
 times as long. 
 
 Genus DIDYMOHELIX Griffith, 1853, p. 438 
 
 Synonyms according to Buchanan in part. 
 
 Galionella Bory, 1823, in part 
 Gallionella Ehreriberg, 1838, p. 166, in part. 
 Gloeotila Kutzing, 1843, p.245, in part 
 Species, one broad species unless Spirophyllum be 
 included in the genus. 
 
 Didymohelix ferru^inea, (Ehr) Griffith, 1353 
 The filaments are twisted, simple, or two filaments 
 twisted together. Young cells colorless, later yellow-brown 
 to rust red, through the deposition of iron. Simple filaments 
 show no division into cells. Sheath not demonstrable, (Buchanan 
 1913 a, p. 304). 
 
 Ellis compares the structure (a. p.19) to the 
 twisting of a wire hairpin spirally on itself. Average thickness 
 of that thread is 1/2 micron to 3/4, although it may reach 
 1 1/2 micron. The loops formed by the twisting thread may b e 
 from 2 to 6 microns in amplitude. 
 
 The writer has found Didymohelix growing on water 
 faucets of the University water supply. 
 
 What are the phylogenetic relations of Spirophyllum 
 f errugineum Ellis and Uodophvllum ferrugineum Ellis with 
 each other, and with Didymohelix ferru^inea Griffith? Buchanan 
 (1318, (a), p. 304) left that a question. The writer can see 
 
 no particular likeness, morphologically, between the latter and 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 « 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
9 
 
 the two former. However, from the description that will 
 follow, it will be seen that the two forms first named have 
 some morphological characteristics in common and might be 
 included in one genus. 
 
 Genus SPIROPHYLLUM Ellis, 1907, ( (a), p. 21 (b) p. 507) 
 
 Type species (monotyp.) - S. ferrugineum . Body of 
 cell elongated, flattened, and spirally twisted. Number of 
 spiral turns may vary from a quarter turn to fifteen or more. 
 Width varies from 1 to 6 microns, length may reach 200 micron. 
 
 No definite membrane but edge is thickened so as to form a 
 sort of rampart all around the cell. Multiplication by the 
 formation of conidia, by external constriction, which are oval. 
 Conidia are formed before twisting begins. Ferric hydroxide 
 deposited on the surface. Found only in iron water, and 
 always in connection with Lentothrix ochracea, when found at all. 
 (Enlows, 1920, p. 86). 
 
 Nodophyllum ferrugineum is also flattened, but 
 restricted at intervals, being arched rather than spirally 
 twisted. Otherwise it is quite similar to Soirophvllum and 
 deserves the same generic name. But, in as much as there 
 have been no recorded observations of Nodophyllum since the 
 original, perhaps there should be no definite decision as to 
 their position until further study has been made. 
 
 It is apparent that in many cases, as shall be 
 noted later, new genera have been created unnecessarily. 
 
Genus CRENOTHRIX. 
 
 10 
 
 Cohn, 1870, p. 130 
 
 Synonyms according to Ellis (a, p. 41) 
 
 Leptothrix kuhniana R ab enho r s t 
 
 Cr eno t hr i x kuh n i ana Zopf 
 
 Crenothrix manganifera Jackson 
 Leptothrix meyeri Ellis 
 
 Megalothrix discophora Schwers 
 
 Number of species, one. 
 
 Crenothrix polyspora . Cohn 
 
 Following is Migula' s (190C, p. 1033) description 
 of Crenothrix. 
 
 Non-branching thread bacteria, with contrast of 
 base and tip. Attached, with a thickening of the free end. 
 Sheath moderately thick, and encrusted with iron oxide if in 
 iron bearing water. Cells cylindrical to disc-like. Reproduce 
 through spherical non-mo tile conidia formed from the vegetative 
 cells which divide in three planes and cause the expansion of 
 the tip of the filament. The conidia become free and germinate 
 often on the sheath of the mother filament. 
 
 Filaments vary from 1.5 to 5 microns at the base; 
 from 4 to 9 at the tip. Filaments may reach a length of 2 to 
 3 mm. Sheath on the thin threads is very delicate. Young 
 threads very often show no sheath at the tip. Threads usually 
 vary in thickness. 
 
 Cr eno thrix is very widely distributed (Ellis, a, p. 
 43). It occurs in brooks, drain pipes, water pipes, and in 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 u 
 
 
 « 
 
 
u 
 
 small quantities, in open waters. It is present perhaps in a 
 large percentage of ferruginous water. 
 
 in addition to the description given by Migula, Ellis 
 (a, p. 49) gives some additional facts from his observations. 
 
 Growth takes place by fission, and, as the mucilaginous sheath 
 hardens, continued division and growth break the septa of the 
 now fixed sheath, and a tube results. In this manner, large 
 cells may be forced from the end of the filament as macrogonidia. 
 
 It often happens that the or ganism retains throughout 
 life the cylindrical form of its immature stage, and that it does 
 not assume the trumpet shape customarily assumed by the adult 
 organisms. 
 
 An important difference between Creno thrix and Sphaero tilus 
 cladothrix is that the cells of the former are visible without 
 staining. A plate showing a young filament of Cr eno thrix . 5 
 microns in diameter, is shown opposite page 42, Ellis, The Iron 
 Bacteria. 
 
 Cohn (1870, p. 130), in the original description, gives 
 the size of filaments 1.5 to 5.2 microns broad; vegetative cells 
 in length are 1/2 to 4 times the breadth of the filaments. They 
 occur in stagnant and running water containing organic matter and 
 iron salts. 
 
 Benecke (1912, p. 491) gives the size of the micrcgonidia 
 (the spherical spores formed as the result of the division of the 
 spiral cells into three planes'! as 2 to 4 microns in diameter, and 
 the macrogonidia may reach a length of 7 microns. 
 
 Jackson (1902) gives the size of the vegetative cells 
 
-L lj 
 
 varying from 1.5 microns in width to 2 to 5 microns in length. 
 Schorler (1904, p. 691) says that the cells are twice as 
 long as wide. 
 
 Summary of Crenothrix 
 
 Non-br anching thread bacteria, which may or may 
 not have contrast in base and tip. Sheath delicate, moderately 
 thick until encrusted with iron oxide. Cells of the filament 
 may vary in thickness; in length from cylindrical to discoidal; 
 in size 1.5 to 2 micron wide by 2 to 7 microns in length, in 
 general, twice as long as wide. Cells visible without stain- 
 ing. Growth by fission. Reproduction by spherical microgonidia, 
 2 to 4 microns in diameter, as a result of division of vegeta- 
 tive cell in three planes. Also by macro go nidi a, escaped or 
 isolated vegetative cells. Occurs in stagnant or running 
 water containing organic matter. 
 
 Genus SPHAEROTILUS: (Kutzing, 1333) emend. Buchanan, 1913 
 
 Synonyms; 
 
 Cl ado thr ix Cohn, 1875, p. 185 
 
 Chlamedo thrix Molish, 1910 
 Following is a summary of Buchanan’s description of 
 the genus (1918, 1, p. 305): Filaments of rods or oval cells, 
 attached, colorless, showing pseudodi cho tomous or false 
 branching; multiplication by motile swarm cells and non-mo tile 
 conidia, the former with a clump of flagella near one end. 
 
 Number of species described, two: 
 
 S. natans Kutzing, 1833 (p. 385) 
 
 (Cohn) era. Migula ( 1900, p . 1035) 
 
 S. dichotoraus 
 
13 
 
 For some time this genus has been designated by the 
 name Cl ado thr ix . but the generic name Clado thrix belongs to 
 the Hyphomyoeten and above all not to the bacteria (Migula, 1900 , 
 p. 1036). Undoubtedly the species that Cohn called Clado thrix 
 is more nearly related to S. natans than Creno thrix polyspor a , 
 but, as shall be observed later, Creno thrix and S , natans are 
 sufficiently similar, on the present basis of classification 
 to be considered species of the same genera, if not varieties 
 of the same species or variants of a common type. 
 
 Some have thought (Benecke, 1912, p. 492), because 
 members of the genus Sphaero tilus do not cause deposition of 
 iron oxide, that the fact was sufficient to separate this genus 
 from others. But Creno thr ix is not always iron incrusting and 
 so separation on that premise is untenable. 
 
 Variation of environment influences not only the 
 morphology of these forms but also the physiology, as shall be 
 noted later, a fact that has not been appreciated by former 
 investigators. Ellis advises retaining the name Clado thrix 
 for this genus (a, p.92), evidently because it is difficult to 
 abandon the familiar name for the unfamiliar. 
 
 Sphaero tilus natans Kutzing, 1833 
 
 Following is Migula' s description of this species. 
 (1900, p. 1035). 
 
 The organism will be found in yellowish-brown "floes" 
 at times free floating in dirty water, which under the microscope 
 are composed of bundles of colorless threads. The threads are 
 
 composed of cylindrical, somewhat rounded cells, which in the 
 
14 
 
 younger threads only are close together, while in the older 
 there are wider intervals between the cells. 
 
 Altogether, the cells of S. natans are very much like 
 S. dichotomus as thick (2 microns , but seldom so pressed 
 together. The sheath, opposed to that of dichotomus . is 
 extremely delicate, slimy and very difficult to see. (This does 
 not mean thin, for the sheath of S. natans is thicker but less 
 visible) . 
 
 Eidam found, throughout the homogenous protoplasm, 
 exceedingly numerous, small spherical parts, which, in sharp 
 definition showed strong refraction in every separate cell 
 (1876, p. 133). Every cell is a sporangium and a multitude of 
 small round spores. The spores germinate very soon and form thin 
 delicate filaments to which, when the mother filament is present, 
 they become attached. More and more of the spores develop 
 sporelings, until the slowly decomposing mother filament becomes 
 thickly surrounded with the young sporelings. 
 
 Migula in all of his observations never saw this 
 state of affairs. This phenomenon was of common occurrence in 
 the studies made by the writer of the flora of the activated 
 sludge, and it remains to be determined whether Eidam really 
 had Sphaero tilus or whether it was a new or allied form (see 
 Plate I, Figs. 1 and 2). 
 
 Eidam is the only one to describe a member of the 
 Chi amedobacteriaceae as having endospores. 
 
 J.W.Haigh Johnson (1914, p.121) gives the following 
 dimensions for Sphaero tilus natans . Filaments, 6 to 8 microns 
 
 thick, with cells 2 microns wide by 4 to 6 microns long. 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 . 
 
 
 
 
 
 
 
 
 . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
15 
 
 Gelatinous sheath thick but not readily visible. Multiplication 
 by transverse cell division or fission also by swarm cells. Can 
 be found in almost every stream receiving unpurified sewage, 
 also in the effluent of industrial wastes. 
 
 Zoogloea form is often assumed by S. natans . These 
 gelatinous masses, called Zoogloea ramiaera . 1000 to 2000 by 
 15 microns in size, contain rod-like cells 1 micron thick im- 
 bedded within the matrix. See Plate IV, Fig. 7 for a type of 
 Zoogloea ramigera. 
 
 A habit growth of S. natans is in many parallel 
 strands with a common mucilaginous sheath (Benecke, p.203). 
 
 Ellis (a) also makes the same observation, S. natans closely 
 resembles S. dichotoma but differs from it in that the threads 
 run parallel, all being enveloped by a common mucilaginous 
 cover ing. 
 
 Summary of Sphaerotilus nutans 
 Parallel filaments, enveloped in a more or less 
 common mucilaginous covering. Filaments attached, colorless, 
 composed of cylindrical, somewhat rounded cells, varying from 
 1 to 2 microns wide to 4 to 6 microns long, close together in 
 young stages, farther apart in older. Multiplication by fission, 
 by swarm cells, and possibly by endospores, with sporelings 
 developing on mother filament. Occurs in streams receiving 
 unpurified sewage and in the effluent of industrial wastes. 
 
 Sphaerotilus dichotomus (Cohn) em. Migula (1900 p. 1035) 
 Synonyms: 
 
 Cladothrix dichotoma Cohn, 1875 
 
15 
 
 Chi arnydo thrix sideroporus Molish 1910* 
 
 The characters of S. dicho teems are well established, 
 since Migula (1900, p. 1026) has described the species and 
 since Ellis (a, pp. 72-77) has figured the organisms so well in 
 plates and drawings. 
 
 Migula (1900, p. 1036) describes S. dicho tomus as 
 follows: Thin, fine colorless floes, which under the microscope 
 appear to be composed of dichotomous threads. The threads are 
 surrounded by a thin, firm and scarcely visible, sheath. Migula 
 and Busgen both found filaments 2 microns thick. Dichotomy is 
 the result of a cell breaking out of the sheath but still 
 remaining in contact and covered by its own slime sheath and by 
 division developing a new filament which appears as a branch to 
 the mother filament. 
 
 Multiplication takes place through swarm spores, 
 which appear as vegetative cells with a cluster of flagella near 
 one end. They leave the sheath before germination. Found very 
 commonly in swamp water, occurs in iron water amongst T.,epto thrix 
 ochraceae . 
 
 As noted under the description of S. natans the cells 
 of S. dichotomus are very close together, the interval between 
 cells being hardly visible ( see Plate III, Fig. 6). 
 
 Ellis (a, p.76) gives the following characterization: 
 In a majority of cases, 5. dichotomus appears as isolated long 
 thin filaments, usually in great numbers attached to a common 
 object. The cells are invisible until stained. Growth takes 
 
 * 
 
 Ellis, p. 97,. Considers Chlamydo thrix siderooorous 
 s y no ny mou 3 with S . d i ch o tbW5 ' . ~~ 
 
 Molish to be 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
17 
 
 place in length of thread only, and, as the filament grows 
 older and the sheath becomes hardened and detached from the 
 cells, growth results in the disappearance of the transverse 
 septa as in Crenothrix . Ellis represents in diagrams and shows 
 by plates that the cells of S. dichotomies are four to ten 
 times as long as thick; that there is very little space between 
 cells; that the sheath is very thin, and that dichotomy may 
 not always be expected. 
 
 Ellis gives three methods of multiplication, (1) by the 
 liberation of thread fragments, (2) by the liberation of re- 
 juvenated single cells which become motile, called swarm spores 
 by Migula, and (3) by the formation of spiral threads. 
 
 Summary of Sphaerotilus dicho tornus 
 
 Filaments, that may show dichotomy, composed of 
 rod-shaped cells approximately 2 microns in width and varying 
 from 3 to 10 microns long. Sheath thin, firm, visible. Cells 
 invisible until stained. Occurs in thin, fine, colorless floes, 
 composed of a great number of filaments attached to a common 
 object. Growth by fission. Multiplication through motile swarm 
 spores which are ciliated vegetative cells by fragmentation of 
 the filament, or the formation of spiral threads. Occurs in 
 swamp water and iron water where Lento thrix ochracea is found- 
 
 Genus CLONOTHRIX Schorl er, 1904, p. 68S 
 
 Buchanan (1918, a, p. 305^ gives the following summary. 
 
 Filaments with false dichotomous or irregular 
 branching attached, with contrast of base and tip, tapering to tip. 
 

 
 
18 
 
 Sheath always present, thin on young filaments, later becoming 
 thicker and encrusted with iron or manganese. Multiplication by 
 small non- mo tile gonidiaof spherical form, formed from 
 disc-shaped cells near tip by longitudinal division or rounding 
 up. 
 
 Number of species, one. 
 
 Clone thrix fuse a Schorler 
 
 Discovered by Schorler in the water works of Dresden 
 and described in 1914, but has not been found elsewhere since. 
 Young threads are 2 to 3 microns thick, older ones 5 to 7. Cells 
 measure 2 microns in thickness, cylindrical, some longer and 
 some shorter. Cells are visible without staining. 
 
 This genus perhaps represents a minor local variation 
 of the common type. 
 
 3. A Review of the Taxonomy of the Beggiatoaceae. 
 
 The order, Thiobacteriales is represented by three 
 families (Buchanan 1918, p. 461). 
 
 Family I. Achromatiaceae - Buchanan, 1916, p.462 
 Family II. Besa atoaceae - Migula, 1895, p. 41 
 Family III. R h o do b ac t er i a c e a e - Migula, 1900, p. 1042 
 Of this order only one genus, family Beggiatoaceae 
 was observed by the writer in the flora of activated sludge. 
 
 Beggiatoaceae Migula. 
 
 Buchanan (1918-p.463) gives a summary of the characters 
 of the family and a key to the three genera. These he follows 
 
 with a description of each, naming a type species. 
 

 . 
 
 . 
 
 
 
 
 
 
 
 ' 
 
19 
 
 The Beggiatoaceae are filamentous bacteria, usually 
 showing an oscillating motion similar to Qscillatoria not 
 attached and no differentiation into tip and base. Cells contain 
 sulphur granules. Spore formation and conidia unknown. 
 
 The three genera are : 
 
 Cenus 1. Thio thrix - Winogradsky, 1888, p. 39 
 Genus 2. Be^eiatoa - Trevisan, 1842, p.76 
 Genus 3. Thionloca .- Lauterborn, 1307, p. 238 
 
 Beggiatoa Trevisan 
 
 Free, sheathless threads composed of discoidal cells. 
 Motility by means of undulating membrane similar to Qscillatoria . 
 Reproduction by partial fracture and subsequent fragmentation. 
 Widely distributed, occurs in situation varying from potable 
 waters to foul, sulphur springs commonly, fresh water and marine. 
 Number of species, five. 
 
 Type of species, according to Buchanan, Ber^iatoa alba . 
 The species of Be.y iatoa may be distinguished by the 
 following key based on Migula's description: 
 
 Key to the Species of Beggiatca 
 
 1, (4) Cross wall indistinct - 2 
 
 2, (3) Filament 3.3 - 3.8 microns thick B . alb a 
 
 3, (2) Filament 1.8 - 2.5 microns thick B. Lentoformis 
 
 4, (.1) Cells (articulations) of the filament very distinct - 5 
 
 5, (8) Found in salt water (marine) situation - 6 
 
 6, (7) .f ilament more than three microns thick B. mirabilis 
 
 7, (6) Filament less than three microns thick B.minema 
 
 8, (5) Occurs in fresh water (swamps or sulphur springs 
 
 not marine B. abachnoidae 
 

 
 
 
 
 
 
20 
 
 Description of the Species of Beggiatoa 
 Note: The descriptions following are based on Mi gula* 
 
 (1900) description of the Be gel atoa . particularly in the 
 treatment of this group. References are not so complete in 
 detail as to enable one to arrive at all of the sources. 
 Nevertheless, as much as could be found in respect to dates and 
 particular references, has been included. 
 
 Beggiatoa alba (Vaucher) em. Trevisan 1842 
 Synonyms: 
 
 Oscillator la alba Vaucher (?) 
 
 Hy-rocrocis vandellii Meneghini (?) 
 
 Beggiatoa punctata Trevisan, 1842, p.56 
 Filament without recognizable cross walls, 3.3 to 
 
 3.8 microns thick. Occurs in pale slimy masses in foul water, 
 sulphur springs and swamps. Most common species. 
 
 Beggiatoa 1 eu to formis (Meneghini) Trevisan 1842 
 Syno nyms: 
 
 Qscillatoria leptof itiformis Meneghini 
 
 Filament without recognizable cross walls, very thin, 
 
 1.8 to 2.5 microns thick. Occurs in great quantities in pale 
 and slimy masses in sulphur springs, also in swamps and waste 
 waters. Not as common as B . alba but found with it. 
 
 The only difference between the descriptions of B . alba 
 and B. leptoformis is one of size. For all practical purposes 
 
 the two species are identical especially since they occur 
 together . 
 
21 
 
 Beggiatoa mirabllls Cohn 1865, p. 81 
 
 Filaments 13 microns thick, articulated and variously 
 curved and tangled, with rounded ends. Length of cells 1/2 
 times width. Filaments associated with a snow-white slime. 
 Marine. 
 
 Be^iatoa minema Warming (?) 
 
 Similar to B. leptoformis . except that the filaments 
 are articulated (Compound of cells) 1.8 to 2 microns thick. 
 Length of cells 1/2 times width. Longest threads 40 microns. 
 Marine. 
 
 The same can be said of these two marine forms as was 
 said of the two previous fresh water forms: the descriptions are 
 quite identical except as to the size of the filaments. 
 
 Be-?,£iatoa arachnoidae Rabenhorst, p. 94 
 Synonyms: 
 
 Oscillator i a arachnoidae ( ? ) 
 
 Beg-viatoa pellucida . Cohn (?) 
 
 Articulated filaments, 5 to 6.5 microns thick, cells 
 length cells 1 to 1.2 times width. Not marine. Occurs in 
 swamps and sulphur springs. 
 
 Summary of Bergiatoa . 
 
 There are two distinct types of B egsl ato a . One type 
 articulated, the other without visible cross walls. The former 
 occurs both in fresh and salt waters, while the latter has been 
 found only in fresh water. 
 
22 
 
 III 
 
 MATERIAL AND METHODS 
 
 The microscopical observations with which this paper 
 is concerned were made from November 15th to December 29th, 1921. 
 It was desired (1) to trace the development of the sludge from 
 raw sewage to a climax sludge, (2) to observe the evolution of 
 forms that takes place, and (3) to identify the types present. 
 
 As before stated, the present study is concerned with the 
 filamentous forms only. 
 
 Samples of sludge taken from the up cast well of the 
 aeration chamber were brought in daily, excepting Sunday. 
 Observations were made on these samples and permanent slides 
 prepared from each. 
 
 Counts of various forms including zoogloeal masses, 
 were made. The results are reported in an unpublished article. 
 
 In as much as we are concerned with Beggiatoa here the method 
 used in counting will be described. The one cubic centimeter 
 rectangular cell was filled with sludge, and counts were made of 
 ten fields one millimeter square. Estimations of forms or units 
 present per cc. were based on these ten counts. (Generally, two 
 mounts 'were made with five fields counted from each.) 
 
 In the preparation of permanent slides, a few drops 
 of sludge were evaporated on the slide and fixed by heating. 
 Carbol fuchsin and gentain violet were used to stain different 
 areas of the same slide. 
 
 In as much as the quantitative observations on the 
 
 filamentous forms can be determined only relatively, direct 
 count being impossible, three terms — f ew , common, abu no. ant , are 
 

 . 
 
 
 
 
 
 
 - 
 
 . 
 
 
 
 
 
 
 . 
 
 
 
 t 
 
 ■ 
 
 
 
 ■ 
 
 
 
 
 ■ 
 
 
 
 
 
 
 
 
 
 
 
 
 
23 
 
 used by the writer to describe the quantitative relationships. 
 
 Low power magnification as used in the observations indicates 
 a magnification of 100 diameters. 
 
 IV. 
 
 OBSERVATIONS 
 
 The characteristic appearance presented by an unstained 
 field under lower power magnification is as follows: One to three 
 
 Bep.;giatoa filaments may be in the field, as many or more 
 peritrichs, a few hypotrichs, and perhaps a nematode, - these, 
 moving around amongst the masses of sponge-like material 
 composed of a network of filamentous forms, the so-called iron 
 bacteria. 
 
 Sludge has been described as amorphous colloidal 
 material. The writer would describe climax sludge material as 
 consisting primarily of a network of filamentous bacteria, a 
 delicate sponge-like mass, intersperced with enmeshed amorphous 
 material. In the amorphous material can be found representatives 
 of every morphological type of bacteria, with the zoogloeal 
 growth habit common. One is impressed by the mass of filaments 
 that compose the foundation of the sludge, and the writer does 
 not doubt that they are organisms of greatest sanitary significance. 
 
 Just what these filamentous forms are and their 
 prominence and time of appearance in the formation of the sludge 
 will be the subject matter of the following paragraphs. 
 
 In the consideration of the filamentous bacteria it 
 is well to keep in mind that all of the descriptions given in the 
 foresoins: discussion are of European xorms, with one minor excep— 
 
 O O 
 
 tion. We should expect to find such widely distributed forms 
 

 
 
 
 
 
 
 
 
24 
 
 fairly constant in morphological characters the world over. 
 
 On the other hand, observations have been more or less fragmen- 
 tary in many cases, and we may find upon Continued observation 
 that forms believed to be different are variations of a common 
 type. 
 
 Occurrence of Beggiatoa. 
 
 The occurrence of Beggiatoa was fairly constant 
 throughout the period of observation, especially after the 
 mature sludge was developed. A few filaments were observed 
 during the earlier days. During the period of climax sludge 
 the counts per cubic centimeter varied from 100 to 5500. 
 
 Beg- iatoa was easily recognized because of the motility and 
 easily counted because of the small number as compared with 
 other filamentous forms. 
 
 Filaments present varied from 1.5 microns to 3.5 
 microns in thickness, all without visible cross walls. The 
 larger form described as B. alba was the most abundant, occurring 
 in a proportion of about three to one with the smaller form 
 described as B. arachnoidae . 
 
 Occurrence of Sphaerotilus dichotomus 
 
 The first appearance of S, dichotomus was noted on 
 November 28th, thirteen days after the beginning from raw sewage. 
 While making counts under low power branching colorless filaments 
 were observed. Stained mounts revealed the character of the 
 filaments a^ represented in Plate III, Fig. 6. Such filaments 
 were very few until December 12th when they became common, but 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 ■ i */■ f 
 
 ■ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
25 
 
 never abundant. S. dichotomus and Beggiatoa alba occurred 
 in approximately equal proportions. 
 
 Occurrence of Crenothrlx polyspora 
 
 Crenothrix in the immature filaments was common in the 
 observations made on the first samples taken. Rapid increase 
 occurred so that at all times the typical Crenothrix filament 
 was in predominance. After fifteen days operation the appearance 
 of the sludge in the stained mount was characteristic of that 
 which continued throughout the rest of the period, with a 
 possible increase in the percentage of filamentous forms. 
 Crenothrix as indicated, predominates over all other forms 
 present. 
 
 The writer wishes to define what he has included 
 under the name Crenothrix . Some of the facts necessary in the 
 consideration of the filamentous bacteria have been enumerated 
 in a previous paragraph. Already many forms have been recog- 
 nized as variants and have been included as synonyms, perhaps 
 in sludge different varieties are represented, and, indeed, 
 had the several growth habits and morphological types been 
 observed by themselves, they might have been mistaken for new 
 species, or possibly new genera. But rather than make the error 
 which earlier observers have made and cause still more confusion, 
 the several variants of a common type are included under the 
 genus Crenothrix because the description of that genus is broad 
 
 and quite definitely accepted. 
 
 In as much as in some sanitary discussions ^.Johnson, 1914 
 

 
 • 
 
 
 - 
 
 
 ■ 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
26 
 
 the name Sphaerotilus natans is used in the description of the 
 flora, the writer has included the description of the genus 
 in the foregoing consideration. Perhaps the genus has not been 
 sufficiently described to bear consideration or comparison, 
 if not, then, with Ellis, we need not consider it a definitely 
 established genus. Ellis believes (a, p.93) that the generic 
 and even specific distinctness of Sphaerotilus natans is 
 doubtful. The writer believes the observations which follow 
 
 will corroborate his statement. (This does not invalidate the 
 entire genus, however ). 
 
 In the first place, we note that the description of 
 Sphaerotilus natans and Crenothrix polyspora are quite similar. 
 There are three differences noted: 
 
 1. S . net ans apparently has less power to precipitate 
 iron oxide than Crenothrix . 
 
 2. The habit growth of S. natans is in parallel 
 
 strands . 
 
 3. There is a difference in method of multiplication. 
 
 In case of the first difference it is not an absolute 
 difference. Not all Crenothrix filaments are iron incrusting, 
 and as indicated some Sphaerotilus natans may be iron incrusting. 
 
 Secondly, it is highly possible for Crenothrix to 
 simulate the growth habit which characterized Sphaerotilus natans . 
 If a group of filaments with gelatinous sheaths were pressed 
 together as they developed, it would be very natural for them to 
 maintain such an or ganization. Such growth habits have occurred 
 in activated sludge composed of filaments which in the writer's 
 
27 
 
 judgment could be classified only as Creno thrix . The writer 
 has observed similar growth habits in great abundance in 
 sludge fed by starch wastes. 
 
 The question of multiplication will be considered 
 under variation. 
 
 The most conclusive evidence that the writer can find 
 for the identity of Creno thrix and Sphaerotilus natans is the 
 fact that no author has ever described both of them as occurring 
 together, but when described they came from identical situations. 
 If they occur in identical situations why have they not been 
 described together? In fact, an author who recognizes one in 
 his observations never mentions the other. In one case, 
 Sphaerotilus natans was found in effluent from industrial wastes. 
 That would indicate that what was called S . natans due, to 
 environment, might be merely a variant of Creno thrix . In fact, 
 the original form of Cr eno thr ix described is, perhaps a variant 
 of a common type, as shall be noted in the following paragraphs. 
 If such be the case, we might expect a great variation in an 
 environment such as is encountered in the activated sludge. 
 
 Variations from the Common Type . 
 
 (1) Ellis observes that we must not always expect to 
 
 find Creno thrix assuming the "trumpet” shape at the free end, 
 a character which is so often used to describe it. The writer has 
 never found it assuming such a form, neither have other 
 investigators of American forms recorded such forms. 
 
 Evidently the method of multiplication which causes 
 the formation of the "trumpet" is uncommon to American forms. 
 
28 
 
 On the other hand, the type of multiplication as observed 
 by Jackson (1302) is quite common under certain conditions. 
 
 The cells of the filament break up into a number of small 
 spores causing a disc-like appearance through the filament. 
 
 Masses of Cr eno thrix filaments have been found in such a state 
 in a minature experimental sewage disposal plant in the labora- 
 tories of the State Water Survey. 
 
 These same filaments show ed a readiness of fragmentation 
 into segments containing 10 to 15 of the disc-like cells. Such 
 fragmentation is ascribed to S, dichotomies (Ellis, a,p.88). 
 
 Whether such fragments possess motility has not been determined. 
 That these fragments are not S. dichotomus is certain because 
 they are visible unstained. 
 
 (2) The description of Creno thrix is quite broad, 
 includingf i laments of both discoidal and cylindrical cells. 
 Filaments made up of discoidal cells characteristically Cren o thrix 
 occur in abundance throughout the activated sludge. These 
 filaments range from the very young and immature filaments, .5 
 to .75 microns in thickness, to the older, 2 micron in thickness. 
 They occur quite characteristically in parallel heavy thick 
 masses (Plate II, Fig. 4), although they occur very commonly as 
 single filaments. 
 
 The most striking variation is shown in Plate II, Fig. 
 
 3, in which a filament composed of cylindrical cells has divided 
 up into discoidal cells leaving the interval between the 
 original rods plainly visible. In filaments from 2 to 3 mm. 
 
 long, it is possible to find cells at one end normally rod-shaped. 
 

29 
 
 Toward the center of the filaments many refractile areas 
 appear through the length of the cells (which is of common 
 occurrence just before or during the process of assuming the 
 discoidal form). At the opposite extreme, the rods have become 
 a segment of discs. Plate I, Fig.l in the upper left hand 
 corner 3hows a short filament in which the refractile areas 
 occur . 
 
 Eidam, (1376, p.133) already referred to in the con- 
 sideration of Sphaerotllus natans found such refractile areas 
 and called them what would be interpreted as endospores. He 
 describes a condition which the writer found and reproduced in 
 Plate I, fig. 2. He describes the spores escaping through the 
 gelatinous matrix and germinating on the mother filament. Of 
 course, spores may be attracted to, or become attached to the 
 surface of the filament from the exterior, but it is more 
 reasonable to assume that the multitude of short sporelings 
 as shown in Plate I, Fig. 2 originated from the cells within the 
 filament rather than that they are merely gathered from the 
 surrounding medium. The fact that refractile areas appear in the 
 cells and that not all filaments are equally covered with 
 sporelings, some having none, others many, lends evidence toward 
 that assumption. 
 
 Eidam believed the form to be Sphaerotilus natans 
 since it answered that description. But, as far as morphological 
 characters are concerned, it answers the description of Ore no thri x . 
 We have already described a condition very similar under (1) in 
 the reference made to Jackson (1902). Even in Plate I, Fig.l, 
 

 
 
 
 
 ■ 
 
 
 
 - 
 
 
 
 
 
 
 . 
 
 
 - 
 
 
 
 
 
 
 
 
 
 
 
 
 
30 
 
 a few short sporelings appear, but, in addition, many long 
 sporelings which give outline to the invisible sheath, of 
 course, the origin of the sporeling could, be determined certainly 
 if one could isolate the forms in pure culture. Until such a 
 technique is developed one can only speculate on the matter. 
 
 Again, the rods of the filaments shewn in Plate III, 
 
 Fig. 5, seem identical with those figured by Migula (1900, 
 
 Plate XVIII, Fig. 8). The writer found such filaments growing 
 in the parallel habit as shown by Migula (1900, Plate XVIII, Fig. 7). 
 But on comparison, Plate ill is identical with the forms 
 represented in Plates 1 and II. There may be a difference in 
 development in the particular filaments shown, but the morpho- 
 logical characters are identical, as well as the refractile 
 areas common to both. 
 
 Sphaero tllus has been described as forming swarm 
 cells. This character has been used to separate this genus 
 from Crenothrix. Since these forms have not been isolated in 
 pure culture it is difficult to determine the origin of swarm 
 spores. In the second place, granting the formation of swarm 
 spores it cannot be said that such a character is less likely 
 to be a variation from the common method than the other mode of 
 multiplication cited. 
 
 By way of summary of this comparison, then it has 
 been shown that Sphaero tilus natans is synonymous with Cr enc thri x 
 polyspora , or that they are both variants of a common type; 
 that the type represented is quite subject to its environment, 
 assuming various modifications of a common type. (Fig. 1, 2, 3, 4, 5 ) , 
 
31 
 
 V. 
 
 SUMMARY 
 
 1. Filamentous forms increase from "few" to "abundant" 
 during the first ten to fifteen days of sludge formation. 
 Following the first period of ten to fifteen days, under the 
 particular conditions, the sludge continued primarily as a 
 network of filaments. 
 
 2. Creno thr ix is present at all times and is the pre- 
 dominating type of organism. 
 
 Filaments of the type Creno thrix are subject to 
 great variation. Perhaps some of the variants deserve -che 
 designation of species, but, in as much as they are without a 
 doubt due to immediate environmental influences, they should be 
 considered merely as growth habits, at least until isolation in 
 pure culture is accomplished. 
 
 Sporelings, short and long, occurred commonly in 
 connection with Creno thrix . never in connection with Sphaero tilus 
 aichotomus and, therefore, they possibly originate from spores 
 produced by filaments of the Creno thrix type. 
 
 3. Sphaero tilus dichotomus (Syn: Cladothrix dichotoma 
 Cohn) appeared thirteen days after the beginning from raw 
 
 sewage and remained a constantly occurring form being "common" 
 after December 12th. 
 
 4. Two species, one type, of Be^giatoa were observed 
 sparingly during the formation period of the sludge and became 
 common in the climax sludge, occurring approximately in same 
 proportion as Sphaero tilus dichotorms . 
 
32 
 
 VI. 
 
 ACKNOWLEDGEMENT 
 
 The writer wishes to thank Dr . A. M. Buswell 
 (Chief of the State Water Survey) for the opportunity 
 to conduct this investigation and for the U3e of 
 laboratories and materials. To Dr. Stella M. Hague, 
 under whose supervision this investigation was con- 
 ducted, the writer is indebted for invaluable criticisms 
 and suggestions. 
 

 
 
33 
 
 VII. 
 
 BIBLIOGRAPHY. 
 
 Benecke, Wilhelm, 1912, Ban and Leben der Bakterien 
 
 Buchanan, R. E., ISIS, (a) Studies in the Classification and 
 nomenclature of the Bacteria. Journal of Bacteriology, 
 
 Vol.III, pp. 301-306. 
 
 (b) Same title, Journal of Bacteriology, Vol.III, pp. 481-474 
 
 Cohn, Ferdinand, 18 65, Hedwigia 
 
 1870, Unter suchungen uber Bakterien, Beit.z. 
 
 Biol. d. Pflanzen, Vol. 1, h.l, Breslau. 
 
 1875, U nter sue hungen uber Bakterien, II, Beit.z. 
 Biol. d. Pflanzen, Vol. 1, h.3, Breslau. 
 
 Cox, Charles R., 1921, Non Bacterial Population of Sewage Trickling 
 Filters. Eng. News Record, Vol. 87, 18, p. 720. 
 
 Ehrenberg, C.G., 1838, The Infusionstheierchen alvollkommende 
 Organ! smen. Le ip zing. 
 
 Eidam, 1876, Ueber die Entwickelung des Sphaerotilus 
 
 natans. Jahres ber d. Schlesisch. Geselisch. f, vaterl. 
 
 Kultur, p. 133. 
 
 Ellis, David, 1907, (a) Proc. Roy. Soc. Edinburg, vol.27,p.21 
 
 1907, (b) A Contribution to our Knowledge of the 
 Thread Bacteria. Centrabl. f. Bakteriol. Abt. 2, Vol. 19, 
 p. 503. 
 
 (a) The iron Bacteria, (Pub. by F. Stoakes 
 & Co. bearing no date.) 
 
 Enlows, Ella, M. A, 1920, The Generic Names of Bacteria. U.S.P.H. 
 Service, Hygienic Lab. Bull. 121. 
 
 Griffith, J.W. , 1853, On Gallionella ferruginea. Ann and Magaz. 
 of Nat. Hist. London Ser.2, Vol. 12, p.438. 
 
 Hommon, Harry B., 1918, The Treatment and Disposal of Strawboard 
 Waste. Public Health Bulletin, No. 97, pp. 58-69. 
 
 Jackson, D.D., 1902, A New Species of Crenothrix. Trans. Am. Hie. 
 Society, Vol 23, p.33. 
 
 Johnson, J.W.Haigh, 1914, A . Contribution to the Biology of Sewage. 
 Jour, Ec, Biology, Vol. 9, pp. 105-124, 127-134, 
 
 Keutzing, F. T. , 1833, Linnaea, Berlin, Vol, 8, pp.19,385 
 
 1843, Phytolgia Generalis, Leipzing. p.198 
 

 
 
 
 
 
 
 
 
 
 
 
Bib. -2 
 
 Lauterborn, Root., 1307, Ein neue Gattung der Scwef elbakter ien. 
 Ber. d.d.bot. Gesellschaf t . , vol. 25, pp.23S-242 
 
 Migula, W. , 1895, Schizomycetes, in Engler and prantl. Natur- 
 lichen Pflanzenf amilien, p. 20. 
 
 Migula, W. , 1900, System der Bakterien, Berlin. 
 
 Molish, H. , 1892, Die Pflanzen in ihrer Beziehungen, zum Eisen. 
 Jena. 
 
 1310, Die Eisenbakter ien. Jena 
 New jersey Ag. Exten. Bull. 352. 
 
 Parkinson, 1319, Parkinson and Bell. Insect Life on Sewage Filters. 
 San. Pub. Co . Ltd. London. 
 
 Porter, J. Edward, 1921, A Bibliography on "The Activated Sludge 
 Process of Sewage Treatment". Gen. Filtration Co. Inc. 
 Rochester, N. Y. 
 
 Rabenhorst, Flora Europ. Alg. , Bd.II, p.94. 
 
 Russel and Bartow, 1916, Bacteriological Study of Sewage Purifi- 
 cation by Aeration, Water Survey Series, Illinois, No. 13, 
 pp. 348- 358 . 
 
 Schorler, 1904, Beitrage zur Kenntniss der Eisenbakteri en. 
 Centrabl. f. Bakteriol. Abt.2, Vol. 12, p. 689. 
 
 State Board of Health of Mass, 1912, The 44th Annual Report. 
 
 Trevisan, V., 1842, Prospetto della Flora Eugana, p. 56. 
 
 1879, Prime linee d’ introduzione, alio studio dei 
 Batter j. italiani. Rendiconti. Reale Instituto Lombardo di 
 Scienze e letters IV. Ser. 2, Vol. 12, p. 133-151. 
 
 1885, Caratteri di alcumi nuovi generi di batteroi- 
 acee. Atti della A ccademia Fiso-Medico-Stati3tica, in Milano. 
 Ser. 4, Vol. 3, pp. 92-107. 
 
 Warming, Om nogle ved Danmarks Kyster levende Bakterier.p. 356. 
 
 Winogradsky, 1888, Fur Morphologie und Physiologie der 
 
 Sohwef elbacterien Beit..z. Morph, und Physio. d. Bakterien h.l. 
 

 
 
 
 
 
 . 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
34 
 
 VIII. 
 
 PLATES A ITT DESCRIPTIONS 
 
 The microphotographs composing the following plates 
 were taken by Arthur Granville Eldridge, Instructor in 
 Photography in the Laboratories at the University of Illinois 
 from slides prepared by the writer. 
 
 Fig. 1. 
 
 Crenothrix polyspora. 
 
 Main filament, a semi-mature filament of 
 The sheath is well outlined by short and 
 lings. 
 
 Arrow indicates short Crenothrix filament 
 refractile areas. 
 
 Creno thr ix. 
 long spore- 
 showing 
 
 Fig. 2. 
 
 Fig. 3. 
 
 Crenothrix polyspora. 
 
 Shows filaments heavily decorated with sporelings. 
 Arrow indicates a filament of Sphaerotilus dichotomus. 
 Near edges of dark masses spores and coccoid bacteria 
 are shown. The dark areas are chiefly such forms in 
 connection with the filaments and amorphous material. 
 
 Crenothrix polyspora. 
 
 Figure shows normal filament 
 
 in which the individual 
 
 rods have divided up into disc-like 
 
 segments. 
 
 F i g. 4 . 
 
 Fig. 5. 
 
 Same as above, but showing a common growth habit of 
 younger filaments. The filaments of this mass 
 apparently developed in the disc-like form at the 
 origin of the filament. The mass is r o thick that a 
 focus can be obtained only on one level. Arrows 
 indicate place at which structure is best represented, 
 Thi3 is the commonest form appearing throughout the 
 sludge. Such forms are distributed through the heavy 
 
 masses m 
 
 greater 
 
 numbers than any other type. 
 
 Crenothrix polyspora. 
 
 Two main filaments are of same type shown in Plate I., 
 but in a different state of development. Sheath not 
 recognizable except as suggested by few particles 
 which mark the sheath limits. Upper filament shows 
 fission. 
 
 Fig. 6. Sphaerotilus dichotomus. 
 
 Arrow indicates the part of filament best in focus 
 Very thin sheath, long rods, with short interval 
 
 between. Contrast with 
 
 Fig. 
 
 1 . 
 
 Fig. 7. 
 
 A 
 
 zoogloeal 
 
 habit of growth common in the 
 
 sludge. As indicated by the forming filament. 
 
 activated 
 this 
 
 is perhaps a zoogloeal state a^ 
 form that is evidently related 
 Plate II, Fig. 4. 
 
 sumed by a filamentous 
 to the type shown in 
 
PLATE I 
 
 *ig. 1 
 
 x 75C 
 
 Fi g. 2 
 
 x 500 
 

 
 
 
 
 
 
 
 
 
 
 
 
 
 
PLATE II 
 
 Fig. 3 
 x 500 
 
 i'ig. 4 
 x 500 
 
PLATE III 
 
 Fig. 5 
 x 500 
 
 Fig. 6 
 x 500 
 
PLATE IV