BILLING,
E. (1955). J . gen. Microbiol. 13, 252-260
Studies on a Soap Tolerant Organism: a New Variety
of Bacterium anitratum
BY EVE BILLING*
Bacteriology Department, University of Manchester
SUMMARY :A soap-tolerant organism, commonly found in face-flannelsand sponges,
was compared with Bacterium anitratum, Moraxella lwofi var. brevis and Alcaligenes
viscosus. These organisms have many characters in common. It is proposed that
Bacterium anitratum and related organisms should be included in the family
Achromobacteriaceae and that the soap-tolerant organism be considered a new
variety of B. anitratum with the provisional name Achromobacter anitratum var.
saponiphilum. Some problems of classification of these and similar organisms are
discussed.
The soap-tolerant organism had two major antigens ; one was associated with the
capsule and largely strain specific, the other was common to all strains and was
present also in non-capsulated variants. Both antigens were detached from the cell
on heating, the specific antigen partially and the common antigen almost completely ;
they could be demonstrated in the saline medium after suspensions had been heated
a t 100" for 60 min. and though detached from the cell, could still stimulate antibody
formation. There was an antigenic relationship between the soap tolerant organisms
and Bacterium anitraturn.
Certain Gram-negative cocco bacilli isolated from a variety of pathological
sources were named Bacterium anitraturn by Schaub & Hauber (1948). Similar
organisms have been variously named Herellea vaginicola (tribe Mimeae) (de
Bord, 1939, 1942), B 5 W (Stuart, Formal & McGann, 1949) and Mormella
lzuofi var. glucidolytica (Pigchaud, Pikchaud & Second, 1951).
Gram-negative diplococci sometimes found in cerebrospinal fluid, distinguishable from the meningococcus by their ready growth on unenriched
media at room temperature, have been named Diplococcus mucosus (von
Lingelsheim, 1906, 1908; Cowan, 1938; McFarlan, 1941 ; Bray & Cruickshank,
1943). The majority of organisms from cerebrospinal fluid and other sources
identified as D . mucosus are however identical with Bacterium anitraturn
(Seeliger, 1953) and differ from Cowan's strains (the first t o be fully described)
in their inability to reduce nitrates to nitrites or to liquefy gelatin, in the
development of bacillary forms, and in feeble virulence for mice. Two other
organisms, Moraxella Zzuofi (Audureau, 1940) and Mimn polyinorpha (tribe
Mimeae) (de Bord, 1939, 1942) appear t o differ from Bacterium anitratum only
in their inability to produce acid in peptone water sugar media.
The present work arose from the isolation of a soap-tolerant (ST) organism
from face-flannels and sponges. It closely resembled Bacterium anitratum and
preliminary studies showed that it also had many characters in common with
* Present address: Ministry of Agriculture, Fisheries and Food, Olantigh Road, Wye,
Ashford, Kent.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:18:01
New variety of B. anitratum
saprophytic Gram-negative bacilli found in food and water, including Alcaligezzes viscosus. As there have been widely differing opinions regarding the
systematic position of Bacterium anitratum and similar organisms, the ST
organism was studied in detail, including its antigenic structure, and its
characters compared with strains of B . anitratum, Alcaligenes viscosus and
some saprophytic organisms with similar properties ; a single strain of Moraxella
Zwofi var. brevis was also included. It was hoped that this study would clarify
the taxonomy of what is a t present an ill-defined group.
GENERAL CHARACTERS AND TAXONOMY
Sources and isolation of strains
Soap-tolerant organisms were isolated on Lemco peptone agar containing 10 yo
(w/v) of 'Ivory' toilet soap. One ml. inoculum was spread over the surface
of the medium in a Petri dish and incubated a t 25" for 7 days. ST colonies
appeared earlier (2-3 days) than other soap tolerant organisms. The materials
examined included residual moisture squeezed from 42 face-flannels and
sponges, 30 specimens of infected urine, 14 faecal and 9 soil extracts (supernatant fluid from thick suspensions) ; the numbers of strains isolated were,
respectively, 2 6 , 1 , 3 and 0. An additional strain was from an abscess where the
predominant organism was a staphylococcus.
Bacterium anitratum : all twenty-seven strains had originally been isolated
from pathological sources including cerebrospinal fluid, urine and bed-sores
and included the 10 type strains of Ferguson & Roberts (1950), and NCTC
8101.
Alcaligenes viscosus group : three typical strains were obtained, all from milk.
Six similar but non-viscid organisms were isolated from water, meat and milk.
A further 3 strains included in this group liquefied gelatin.
Moraxella h o f i var. brevis: a single strain was obtained from the National
Collection of Type Cultures, NCTC 7976.
Cultural and biochemical methods
Morphological and cultural characters were determined from cultures on
Lemco peptone medium. Incubation temperature was 25" or 37" according
to the preference of the strain or 25" if there was no preference.
The substrate for fermentation tests was sterilized by Seitz-filtration and
1 % (w/v) added to peptone water with Andrade's indicator; tests were
incubated for 14 days. A 10% lactose agar medium, recommended by
Chilton & Fulton (1946) for detecting late lactose fermenters, was used in
addition to the usual 1 % lactose peptone water.
Lipase was detected on an olive oil medium with neutral red as indicator.
Horse-blood agar was used for haemolysis. Methods in general use were used
in all other tests and uninoculated media, and known positive organisms were
always included.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:18:01
E . Billing
254
RESULTS
The morphological, cultural and biochemical data for all strains are summarized
in Table 1.
Table 1. Summary of morphological, cultural and biochemical characters of the
soap-tolerant organism Bacterium anitratum and similar bacteria
Characters common to all strains
Gram-negative cocci or bacilli ; often pleomorphic ; strictly aerobic ; grow well on unenriched media and on MacConkey's agar a t 25" ;do not produce acid from 1 yo lactose, sucrose,
maltose, mannitol, glycerol ; catalase-positive ; indole-, H,S- and VP-negative ; anaerogenic ;
do not reduce nitrates to nitrites.
Characters not common to all strains
No. of strains
Capsules
Motility
Growth a t 3'7'
Growth a t 4 4 O
Citrate utilization
Acid from glucose
Acid from galactose
Acid from xylose
Acid from arabinose
Acid from 10 yo lactose agar
Oxidase
Lipase
Lecithinase
Gelatinase
Haemoly sis
Soap tolerance
Litmus milk
A . ziscosus
group
M . lwofi
var. brewis
Tr= trace, C= coagulation, A= acid, D= digestion, Alk= alkaline. Figures in parentheses
indicate number of exceptions to the general rule.
Special features of ST strains
The capsules are shown in P1. 1, fig. 1. At 37" some strains were diplococcal
or ovoid (Pl. 1, figs. 2, 3) but others grew as rods, filaments and bizarre forms
which were also capsulated (Pl. 1, figs. 4-6). The rods frequently showed
bipolar staining. Strains which were markedly pleomorphic a t 37' were predominantly ovoid a t 25".
In Lemco broth there was good growth at 5 , 25 and 37" but never more than
a trace a t 44". Ten strains preferred 25 to 37", the remainder were indifferent, or
preferred 37'. I n broth there was uniform turbidity, an iridescent surface ring
and a ropy deposit.
Colonies on agar, after 1-2 days of incubation, were 1-2 mm. in diameter,
circular with an entire edge, convex, smooth, glistening, greyish white,
moderately opaque and markedly iridescent. Ten strains were viscid, the
remainder butyrous in consistency. (All viscid strains grew better a t 37" than
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:18:01
New variety of B. anitratum
255
at 25" and showed little pleomorphism.) On further incubation colonies increased in size and usually became radially striated or segmented, the segments
being more translucent and less iridescent. After 5-7 days some strains showed
surface papillae.
Non-capsulated variants were obtained by plating old cultures or subculturing from segments. The variant colonies were smaller, more translucent
and less iridescent, butyrous or cheesy in consistency but never viscid. Some
were spontaneously agglutinated in saline and serological tests suggested that
surface antigens had been lost.
From two strains another type of variant was isolated. The colony was
normal in form but in consistency was much more viscid; it never became
striated or segmented. Morphologically the organisms resembled the normal
form and were capsulated. Serological tests showed that there was no loss of
surface antigen, and it appeared that this was a stable M (mucoid) form in
comparison with the normal form which invariably gave rise to segments
consisting of non-capsulated variants.
Gelatin was liquefied in 2-3 days, except for the non-fermenting strain which
took 7 days. This strain produced an alkaline reaction in litmus milk, but the
remainder coagulated milk by a rennin-like enzyme, followed by slow formation of acid and slight digestion. Changes occurred more rapidly if there was
a residual cream layer in the medium. Using Hugh & Leifson's method, it was
shown that acid production from glucose was due to oxidation, not to true
fermentation. Haemolysis was slight. The MR test was variable. On Christensen's (1946) urea medium a few strains were weakly positive.
All strains grew well on Lemco agar containing up to 10% of toilet soap;
they were equally tolerant of disinfectant soaps such as 'Lifebuoy' and
' Hexaphene '. The maximum tolerances of other organisms were : Escherichia
coli, 2.5 % ; Aerobacter aerogenes, 2-5yo;Staphylococcus albus, 0-6yo,S . aureus,
; Bacillus subtilis, < 0.6 yo.
The ST organism utilized soaps as the sole source of carbon when these were
substituted for citrate in Simmons's (1926) medium; sodium oleate gave a
more profuse growth than sodium palmitate or sodium stearate.
Nine strains that were tested failed to kill mice when injected intraperitoneally except in massive doses containing about lo9 viable organisms.
Similar numbers did not cause necrosis when injected intradernially into
guinea-pigs.
< 0.6
Comparison of the ST organism with similar bacteria
Bacterium anitratum. This was an ovoid bipolar stained organism with less
pleomorphism. Its colonies on agar were larger and more opaque and there was
denser turbidity in broth. It grew better a t 37" than at 25" and, except for one
strain, grew well at 44". It differed, too, in having a low soap tolerance and in
not liquefying gelatin (Table 1). It formed a biochemically homogeneous
group except that one strain did not produce acid from glucose, two did not
utilize citrate, and one was lipase and lecithinase negative. Five strains had
a viscid growth.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:18:01
256
E . Billing
Alcaligenes viscosus group. The twelve strains examined were diplococcal
or ovoid diplobacilli, some showed small capsules. Like some ST strains those
able to grow were often highly pleomorphic a t 37". Abd-el-Malek & Gibson
(1952) have remarked on the coccal morphology of some of their strains of
A. viscosus. Three strains liquefied gelatin and differed from the non-fermenting ST strain only in their low soap tolerance (0.6yo). One was unable
to grow at 37". The typical A . viscosus strains differed from the non-fermenting
Bacterium anitratum strain only in their viscous growth and in failing to grow
at 44". Also capsules were not demonstrated.
Moraxella Zzuofi var. brevis. This organism, a diplobacillus, showed no
pleomorphism a t 37". It differed from the non-fermenting strain of Bacterium
anitratum in being non-capsulated and in failing to utilize citrate and to grow
at 44".
SEROLOGY
Materials and Methods
Suspensions were made in saline from 20 hr. agar cultures at 37" washed
once and resuspended in saline, containing 1/10,000 t hiomersalate. For
agglutination they were equal in density to Brown's tube no. 3; 5 times as
dense for immunizing rabbits and 25 times as dense for agglutinin absorption
and for preparing extracts.
Preparation of antisera. Rabbits were immunized intravenously by five
injections from 0.1 to 2.0 ml. of living organism a t 5-day intervals and the
serum collected 7-10 days later. The sera from two rabbits were pooled and
preserved with 1/ 10,000 t hiomersalate.
Sera were prepared against four ST strains (ST 1, 2, 3, 5) and two Bacterium
anitratum strains, B 1 (NCTC 8101) and BT 1 (capsular type I, B. anitratum,
Ferguson & Roberts, 1950).
Agglutination tests were incubated a t 37" for 2 hr., and read after standing
overnight a t 5".
Heated suspensions were kept at 100" for 1 hr., washed twice in saline and
resuspended in saline. This washing was important to remove antigen detached
during heating.
Extracts for precipitin tests were the supernatants from saline suspensions
which had been heated a t 100" for 1 hr. and centrifuged.
Precipitation tests. Twofold dilutions of extracts were mixed with an equal
volume of 115 antiserum, incubated a t 37" for 2 hr. and read after standing
overnight a t 5".
Antigenic components of ST strains
All 31 ST strains were agglutinated to a high titre by the 4 ST antisera.
Agglutination was rapid with formation of large floccules, different in size for
each strain.
Absorption of each of the 4 ST sera with another strain or a non-capsulated
variant (VST1) removed the common antibody but left a specific antibody
which agglutinated its own strain to a low titre. The specific antigen appeared
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:18:01
New variety of B. anitratum
257
to be capsular as it was not demonstrated in non-capsulated variants. Only
four strains, including ST 2 and ST 5, possessed related specific antigens.
Suspensions heated to 100" were only agglutinated by their own antiserum
and failed to absorb common antibody. I n precipitation tests with extracts
of heated suspensions, however, both antigens were readily demonstrated. The
common antigen produced large grey floccular aggregates which formed in a
few minutes whereas those of the specific antigen were smaller, white and
€ormed more slowly. Sera prepared against extracts gave titres of up to 640
and 5120, respectively, of both specific and common agglutinin, showing that
the extracted antigens were still active agglutinogens.
Table 2. Cross-agglutination with four strains of the soap-tolerant
( S T )organism and their respective antisera
Titre with
I
Antiserum
ST 1
ST 2
ST 3
ST 5
ST 1
ST 2
ST 3
ST 5
Absorbed
with
Nil
Nil
Nil
Nil
VST 1*
VST 1*
VST 1*
VST 1*
ST 1
5,120
20,480
20,480
10,240
20
0
0
0
A
Unheated suspensions
ST 2
2,560
40,960
40,960
20,480
0
160
0
0
ST 3
5,120
20,480
20,480
10,240
0
0
ot
0
Heated suspensions
ST 5
5,120
40,960
40,960
20,480
0
0
0
20
ST1 S T 2 S T 3 S T 5
40
0
0
0
0
640
0
0
0
640
40
0
0
0
10
160
0
0
0
0
0
160
10
0
0
40
0
640
0
40
0
640
*
t
Non-capsulated variant of ST 1.
When tested against a suspension grown at 25' titre=40.
Bold figures show homologous systems.
Sera prepared against heated suspensions also contained specific agglutinins
with titres of up to 640. I n contrast, common agglutinin was almost absent,
a small amount (titres of 10 and 160) being found in only two of the four sera.
This was in keeping with the other evidence that heating removed most of the
common antigen from the cell. Two of the non-capsulated variants, VST 1
and VST 3, became autoagglutinable when the common antigen was removed
by heating. VST 2 and VST 5, on the other hand, were autoagglutinable before
heating and common antigen was hardly detectable in extracts. This suggested
that variation involved the progressive loss of specific and common antigen
and that loss of both antigens led to saline instability.
Eflect of growth temperature on arLtigen production
Reference has been made to ten strains which grew better a t 25" than at 37".
These strains were not agglutinated by the four antisera when grown a t 37"
and common antigen was barely detectable in extracts. It appeared that the
formation of common antigen had been suppressed a t 37". I n one such strain
against which an antiserum had been prepared, specific antigen was readily
demonstrated by agglutination and precipitation tests on cultures grown a t
both temperatures.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:18:01
258
E . Billing
One strain, ST 3, when grown at 37", was not agglutinated by a specific
(absorbed) serum, whereas at 25" it was (Table 2). This was not due to the
non-production of specific antigen at 37" as heated suspensions were agglutinated and specific antigen was present in extracts. I n this case the common
antigen appeared in some way to prevent specific agglutination.
Antigenic diflerence of a non-fermenting ST strain
The non-fermenting ST strain (Table 1) differed antigenically from the other
strains. A suspension grown a t 25" was agglutinated by the 4 ST antisera but
only to 10-20% of titre. The strain did not absorb an appreciable amount of
common antibody, but it appeared to have a common antigen related to but
not identical with that of the other strains. Heated suspensions were inagglutinable; the specific antigen was not investigated.
A suspension grown at 37" was not agglutinated by any of the four sera,
suggesting that the common antigen was not produced a t this temperature.
DISCUSSION
A study of the morphology, cultural characters and biochemical reactions of the
ST organism shows that it is closely related to Bacterium anitratum. Both
organisms are similar to Alcaligenes viscosus and other saprophytic organisms
found in food and water, which supports the suggestion that Bacterium
anitratum and related organisms should be included in the genus Achrornobacter (Brisou & Morichau-Beauchant, 1952;Brisou, 1953). It is proposed that
the ST organism should be considered as a new variety of Bacterium anitraturn,
and, as the generic name Bacterium is now invalid (Judicial Commission, 1953),
that both varieties should be included in the genus Achromobacter (family
Achromobacteriaceae), the ST organism being named A. anitratum var.
saponiphilum. By Rule 7 of the International Bacteriological Code of Nomenclature (1948),typical strains of Bacterium anitratum are automatically renamed Achromobacter anitraturn var. anitratum.
The status of acid production in peptone water sugar media as a differential
character for strains of these organisms is doubtful. Single strains of both
Bacterium anitratum and the ST organism failed to ferment any sugars. In the
latter case, the non-fermenting strain was agglutinated by antisera to the
fermenting strains. Pigchaud et al. (1951) also observed serological crossreactions between their fermenting and non-fermenting strains of JlormelZa
lzuofl, and Seeliger (1953)doubted whether fermenting and non-fermenting
Bacterium anitratum strains should be considered as distinct species. Brisou
(1953)has suggested that the genus Alcaligenes should be merged with the
genus Achromobacter.
The significance of growth temperature ranges and optima in the classification of this group is also doubtful. It is not unusual for strains of well-defined
species, isolated from different sources, to vary in their temperature requirements and ST strains showed such variation to some degree. while some members of the Alcaligenes viscosus group differed from one another only in their
ability to grow at 37".
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:18:01
New variety of B. anitratum
259
There is no evidence that ST organisms are pathogenic, though they were
found twice in pathological material. Viscous strains may be one of the causes
of sliminess of face-flannels and sponges.
Two main antigens of ST organisms have been clearly defined. The 'specific'
antigen is capsular and is not found in non-capsulated variants. It resists
heating a t 100"for 1 hr. and is partly detached by this treatment. It is largely
strain specific and pure specific agglutinin can be prepared by absorption.
The common antigen is widely distributed, resists heating at 100" for 1 hr.
and is completely (or almost completely) detached in the process. It is present
in non-capsulated variants, but appears to be a surface antigen. In its general
properties it resembles the Vi rather than the 0 antigens as the former are
surface antigens, which are detached by heat and are heat-resistant (Stuart &
Kennedy, 1948; Felix, 1952). With one strain there was evidence to suggest
that it prevented specific agglutination of cultures grown at 37".
On the basis of capsular antigens Ferguson & Roberts (1950) found ten
serological types of Bacterium anitraturn and observed cross-reactions with
unabsorbed sera due to another antigen. Preliminary tests made in this study
suggest that it may be similar to the common antigen of the ST strains.
Cross-reactions were observed between ST strains and four Bacterium
anitratum strains. Antiserum to Ferguson & Roberts type I strain agglutinated
all ST strains, and this appeared to involve the common antigen, but none was
agglutinated by antiserum against NCTC 8101. None of the other organisms
included in this study was agglutinated by ST or B. anitratum antisera.
I am indebted to Dr C. L. Greenbury, Dr T. Gibson, Dr W. W. Ferguson and the
late Dr A. L. P. Peeney for the provision of strains. Most of this work was done
while the author was in receipt of a training grant from the Agricultural Resea.rch
Council.
REFERENCES
ABD-EL-MALEK,
Y. & GIBSON,
T. (1952). Studies in the bacteriology of milk. IV. The
Gram-negative rods of milk. J. Dairy Res. 19, 294.
AUDUREAU,
A. (1940). Gtude du genre Moraxella. Ann. Inst. Pmteur, 64, 126.
BORD,G. G. DE (1939). Organisms invalidating the diagnosis of gonorrhoea by the
smear method. J. Bact. 38,119.
BORD,G. G. DE (1942). Descriptions of Mimeae trib. nov. with three genera and
three species and two new species of Neisseria from conjunctivitis and vaginitis.
Iowa St. Coll. J . Sci. 16, 471.
BRAY,P. T.& CRUICKSHANK,J. C. (1943). Meningitis due to Diplococcus mucosus.
Brit. med. J . i, 601.
BRISOU,J. (1953). Essai sur la systematique du genre Achromobacter. Ann. Inst.
Pasteur, 86,812.
BRISOU,
J. & MORICHAU-BEAUCHANT,
R. (1952). Identit6 biochimique entre certaines
souches de B. anitratum et Moraxella lzvofi. Ann. Inst. Pasteur, 82,640.
CHILTON,
M. L. & FULTON,
M. (1946). Presumptive media for differentiating paracolon and Salmonella cultures. J. Lab. clin. Me&. 31, 824.
CHRISTENSEN, W. B. (1946). Urea decomposition as a means of differentiating
Proteus and paracolon cultures from each other and from Salmonella and
Shigella types. J. Bact. 52,461.
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:18:01
E . Billing
COWAN,S. T. (1938). Unusual infections following cerebral operations. With a
description of Diplococcus mucosus (von Lingelsheim). Lancet, ii, 1052.
FELIX,A. (1952). The properties of different Salmonella Vi antigens. J. Hyg., Camb.,
50, 515.
FERGUSON,
W. W. & ROBERTS,
L. F. (1950). A bacteriological and serological study
of organism B 5 W (Bacterium anitratum). J. Bact. 59, 171.
HUGH,R. & LEIFSON,E. (1953). The taxonomic significance of fermentative versus
oxidative metabolism of carbohydrates by various Gram-negative bacteria.
J. Bact. 60, 24.
International Bacteriological Code of Nomenclature (1948). Ed. by BUCHANAN,
R. E.,
ST JOHN-BROOKS,
R. & BREED,R. S. J. Bact. 55, 287.
Judicial Commission (1953). Minutes of the Judicial Commission meetings held a t
Rome in connection with the V I International Congress for Microbiology,
September 1953. Minute 9. Int. Bull. Bact. Nom. Tax. 3, 153.
LINGELSHEIM,
W. VON (1906). Die bakteriologischen Arbeiten der Kgl. Hygienischen
Station zu Beuthen 0. Schl. waihrend der Genickstarreepidemie in Oberschlesien im Winter 1904/5. Klin. Jb. 15, 373.
LINGELSHEIM,
W. VON (1908). Beitrage zur hiologie der epidemischen Genickstarre
nach den Ergebnissen der letzten Jahre. 2. Hyg. InfektKr. 59, 457.
MCFARLAN,
A. M. (1941). Diplococcus mucosus von Lingelsheim. A description of
a strain with comments on the systematic position of the organism. J . Path.
Bact. 53, 446.
PIECHAUD,
D., PIECHAUD,
M. & SECOND,
L. (1951). gtude de 26 souches de Mormella
lwofi. Ann. Inst. Pasteur, 80, 97.
SCHAUB,
I. G. & HAUBER,
F. E. (1948). A biochemical and serological study of a
group of identical Gram-negative bacilli from human sources. J. Bact. 56, 379.
SEELIGER,
H. (1953). Zur Systematik des Bacterium anitratum (Schaub & Hauber).
Zbl. Bakt. (Abt. 1. Orig.), 159, 173.
SIMMONS,
J. S. (1926). A culture medium for differentiating organisms of colontyphoid-aerogenes groups and for isolation of certain fungi. J. infect. Dis. 39,
209.
STUART,C. A., FORMAL,
S. & MCGANN,
V. (1949). Further studies on B 5 W, an
anaerogenic group in the Enterobacteriaceae. J. infect. Dis. 84, 235.
STUART,
C. A. & KENNEDY,
E. R. (1948). Thermostability and alcohol solubility of
Vi antigen. Proc. SOC.exp. Biol., N . Y . , 68, 455.
EXPLANATION OF PLATE
Films of ST strains from 18 hr. cultures on Lemco agar, incubated a t 37". Magnification,
x 1250. Figs. 2-6 were stained with dilute carbol fuchsin.
Fig.
Fig.
Fig.
Fig.
Fig.
Fig.
1. ST 1 stained for capsules by Howie and Kirkpatrick's method showing capsules.
2. ST 1. Coccal forms.
3. ST 3. Ovoid forms.
4. ST 3. Rod forms showing bipolar staining.
5. ST 21. Branched bizarre forms.
6. ST 22. Long filamentous forms.
(Received 21 January 1955)
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:18:01
Journul of General Microbiology, Vol. 13, No. 2
Downloaded from www.microbiologyresearch.org by
IP: 88.99.165.207
On: Sat, 17 Jun 2017 10:18:01