JOURNAL OF CLINICAL MICROBIOLOGY, Oct. 1979, p. 419-424
Vol. 10, No. 4
Periurethral Anaerobic Microflora of Healthy Girls
INGELA BOLLGREN,'t* GUNILLA KALLENIUS,2t CARL-ERIK NORD,' AND JAN WINBERG3
Department ofBacteriology, National Bacteriological Laboratory, S-105 21 Stockholm,' Department of
Pediatrics, Huddinge Hospital, S-141 86 Stockholm,2 and Department ofPediatrics, Karolinska Hospital,
S-104 01 Stockholm,3 Sweden
Received for publication 2 July 1979
The periurethral anaerobic and aerobic microfloras were investigated in 18
healthy premenarcheal girls, 5 to 14 years of age, by using a quantitative sampling
method. Colonization of the female periurethral area with enterobacteria seems
to be an important step in the development of urinary tract infections, and the
present study was undertaken as a stage in elucidating factors that might control
the establishment of urinary tract pathogens periurethrally. The study showed
that obligate anaerobic bacteria constituted 95.0% (standard error, ±5.8%) of the
total colony-forming units per square centimeter of periurethral area. An average
of 7.0 different anaerobic and 2.7 different aerobic strains per specimen was
obtained. The flora was dominated by anaerobic gram-positive cocci and gram-
positive rods, whereas anaerobic gram-negative rods comprised a minor part. The
most commonly encountered anaerobic isolates were peptococci and peptostrep-
tococci, propionibacteria, bifidobacteria, eubacteria, and bacteroides in decreasing
order of frequency. The aerobic flora consisted most commonly of nonhemolytic
streptococci and diphtheroids. The findings suggest that the periurethral microen-
vironment is a distinctive ecological niche, separate from the fecal and skin biotas,
although it has some characteristics in common with the vaginal flora.
The periurethral region of healthy girls prob-
ably forms a barrier against urinary tract infec-
tions (UTI). A previous study of ours on the
periurethral aerobic bacterial flora revealed no
or very small numbers ofgram-negative bacteria
and enterococci in healthy girls, whereas UTI-
prone individuals were colonized with urinary
tract pathogens even during infection-free pe-
riods (2, 3). This was in accordance with studies
on the introital aerobic flora of adult women (8,
mechanisms creating a microenvironment de-
void of enterobacteria in the normal individual
but failing to do so in UTI-prone persons. Prop-
erties of the periurethral mucosal cells seem to
be one important factor, cells in UTI-prone girls
having a significantly higher capacity for adher-
ence of bacteria than those in healthy girls (12).
With other human microbiotas, it has been
shown that the indigenous flora might interfere
with the establishment of potential pathogens;
e.g., alpha-streptococci in the pharynx interfere
with the beta-hemolytic group A streptococci
(4), cervical lactobacilli interact correspondingly
with Neisseria gonorrhoeae (20), and interfer-
is fragmentary about the
t Present address: Department of Pediatrics, Karolinska
Hospital, S-104 01 Stockholm, Sweden.
t Present address: Department of Bacteriology, National
Bacteriological Laboratory, S-105 21 Stockholm,Sweden.
ence between strains of Staphylococcus aureus
has been useful in curtailing staphylococcal dis-
ease among newborn infants (22). Thus, it is
conceivable that the indigenous periurethral mi-
croflora may also play a role in controlling the
establishment of urinary tract pathogens in
healthy individuals and that this balance for
some reason may be altered in UTI-prone ones.
To study possible interactions between orga-
nisms in the periurethral area, it is necessary to
have a clear picture ofthe normal flora. For that
reason, the periurethral indigenous bacterial
flora was studied in healthy girls. A sampling
method was used that permitted a quantitative
evaluation of the microflora. The main purpose
was to characterize the anaerobic bacterial flora
since the aerobic flora has been analyzed earlier
(2). Aerobic cultivation was, however, also per-
formed to allow comparison of the relative fre-
quencies of aerobic vis-a-vis anaerobic bacteria.
MATERLALS AND METHODS
Subjects. Eighteen girls attending either a child
health care center or a pediatric outpatientclinic for
minor surgical complaintswere studied. None of them
had any history of UTI and urine cultures werenega-
tive. None had any infectious disease at the time of
examination, and none had takenanyantibiotics dur-
ing the 2 months before sampling.Tengirlswere 5 to
10 years old and eight girlswere 11 to 14yearsold.
BOLLGREN ET AL.
Sampling technique. A quantitative method for
periurethral sampling, previously described in detail
(2), was used to determine the anaerobic and aerobic
periurethral floras. A 5-ml plastic syringe was filled
with sterile 12% (wt/vol) gelatin (Difco Laboratories,
Detroit, Mich.), kept anaerobically at 4°C to obtain a
firm consistency, and then kept at room temperature
for 10 min before use. The needle end of the syringe
was cut off, and the gelatin, with a cross-sectional area
A 2-mm slice (0.2 ml of gelatin) with the bacteria-
bearing surface was cut off and put into a sterile tube,
which had been gassed out with carbon dioxide. The
tube was transported to the laboratory in an anaerobic
jar (GasPak system, BBL Microbiology Systems,
Cultivation techniques. The further handling of
the specimens was performed within 2 h of collection.
A 2.3-mi portion of Virginia Polytechnic Institute di-
lution salts solution (11) was added to the gelatin
sample under a flow of oxygen-free CO2. The gelatin
was dissolved at 37°C for 10 min and homogenized by
agitation in a Vortex mixer. Portions (0.1 ml; i.e., 1/25
of the dissolved gelatin sample) were spread onto
freshly prepared blood agar plates (10% [vol/vol]horse
blood in agar base; Oxoid, London, England) for an-
aerobic and aerobic incubation and onto different se-
lective culture media (Table 1). Serial dilutions (1/10,
1/100, 1/1,000) of the dissolved gelatin sample were
also made in the Virginia Polytechnic Institute dilu-
tion salts solution, and 0.1-ml portions were inoculated
onto blood agar plates for anaerobic and aerobic in-
cubation. Selective media (Table 1) were also inocu-
lated with some of these dilutions.
Anaerobic plates were incubated at 37°C by using
the GasPak system and examined after 10 days. Aero-
bic media were incubated in air or 5% CO2 at 37°C and
examined after 48 h and once again after 10 days.
Total counts and counts of different organisms based
1 cm2, was pressed against the ure-
on colony morphology were determined and recorded
as colony-forming units (CFU) per square centimeter
of periurethral area. In general, the anaerobically in-
cubated blood agar plates inoculated with low dilu-
tions of the specimens showed heavy growth, so that
colony counting and isolation of different organisms
was possible only on blood agar plates bearing higher
dilutions of the specimens (1/100, 1/1,000).
Colony types were enumerated on plates containing
30 to 300 colonies. The selective media revealed some
additional strains in low numbers, but in some cases
organisms present in small numbers were probably
missed. Each different type of colony was isolated.
Organisms isolated from anaerobic media were incu-
bated anaerobically and aerobically to determine air
Identification of microorganisms. All isolates
were Gram stained, and the slides were evaluated by
two persons independently of each other. Gram-neg-
ative rods belonging to Enterobacteriaceae were iden-
tified to species by the API 20E test kit system (An-
alytab, Plainview, N.Y.) (17). Biochemical identifica-
tion of streptococci was performed according to Fack-
lam et al. (7). The following tests were performed:
growth on agar containing 6.5% (wt/vol) NaCl (Difco),
growth in Trypticase soy broth (BBL) containing 40%
(wt/vol) bile (Difco), growth on hematin agar (Oxoid)
with 0.04% (wt/vol) potassium tellurite, hippurate hy-
drolysis in peptone broth (Difco) containing 1% (wt/
vol) sodium hippurate, hydrolysis of esculin in Enter-
ococcosel agar (BBL), and acid production from arab-
inose and raffinose. Identification of Streptococcus
agalactiae was verified by a precipitation test with
group-specific antiserum (14). The following biochem-
ical tests were used for the identification of staphylo-
cocci (S. aureus and S. epidermidis): oxidation-fer-
mentation test with glucose and tests for production
of coagulase and deoxyribonuclease. Free coagulase
was detected in a tube assay using citrated rabbit
plasma. Production of deoxyribonuclease was detected
1. Media used for identification of anaerobic and aerobic microorganisms in the periurethral area
according to the Wadsworth Anaerobic Bacteriology Manual (25)
Blood agar (10% horse blood;
ficient agar (Oxoid)
Enterococcosel agar (BBL)
Mitis salivarius agar (Difco)
Hematin agar (Oxoid)
Blood agar (10% horse blood)
', 10-2, 10-
Total aerobic counts, predomi-
Total anaerobic counts, pre-
', 10 2, 10-;
Phenylethyl alcohol agar
Egg yolk-neomycin agar (Ox-
Rogosa SL agar (Difco)
Veillonella agar (Difco)
10", 101, 10
J. CLIN. MICROBIOL.
PERIURETHRAL MICROFLORA IN HEALTHY GIRLS
on deoxyribonuclease agar (Difco) (5). When staphy-
lococcal colony counts exceeded eight colonies per
plate (>200 CFU/cm2; three strains), isolates were
identified to species according to Kloos and Schleifer
(13) for identification of, e.g., Staphylococcus sapro-
phyticus. The following tests were performed: hemol-
ysis of blood agar (5% [vol/vol] bovine blood in agar
base), nitrate reduction in peptone broth (Difco) con-
taining 0.2% potassium nitrate, and aerobic acid pro-
duction from various carbohydrates in a broth medium
(Oxoid). The following carbohydrates were tested:
fructose, arabinose, ribose, maltose, lactose, sucrose,
trehalose, xylitol, xylose, and mannitol. Diphtheroids
were identified by colony appearance, typical Gram
staining, and a positive catalase test. Catalase-nega-
tive, gram-positive aerobic rods producing lactic acid
from glucose fermentation (identified by gas-liquid
chromatography) were characterized as facultative
lactobacilli. Yeasts were identified according to stan-
dard laboratory procedures (23).
Anaerobic bacteria were identified to genus level by
their morphological appearance in Gram staining and
by analysis of end products of glucose metabolism by
gas-liquid chromatography according to Holdeman
and Moore (11).
Total viable counts and relative frequen-
cies of anaerobic versus aerobic microor-
ganisms. Total viable counts (including anaer-
obic and aerobic bacteria and yeast cells) per
square centimeter of periurethral area showed
great variations between individual specimens.
The mean total count was 1.4 x 106 CFU/cm2,
the median count was 4.8 x 105 CFU/cm2, and
the range was 6.6 x 103 to 6.3 x 106 CFU/cm2.
Aerobic and anaerobic microorganisms were
recovered from 17 samples. The ratio ofaerobic/
anaerobic bacteria was fairly constant in all spec-
imens. Thus, anaerobic microorganisms consti-
tuted 95.0% (standard error, ±5.8%) of the total
CFU per square centimeter. The remaining
specimen yielded a pure culture of facultative
Anaerobic microflora. An average of 5.5
(range, 0 to 11) different anaerobic strains (i.e.,
anaerobic isolates identified to genus level) per
specimen was obtained from the anaerobically
incubated blood agar plates. Some additional
types of microorganisms present in small num-
bers could be found bythe use ofselectivemedia,
and with these strains included, an average of
7.0 (range, 0 to 14) different strains per specimen
was found. In the following presentation, the
mean numbers of strains per specimen take ac-
count of findings from both blood agar plates
and selective media.
Table 2 lists the counts of different anaerobic
bacteria grouped according to Gram staining
properties. The proportions of different anaero-
bic bacteria expressed as a percentage of total
CFU per square centimeter are shown in Fig. 1.
Gram-positive cocci were frequently isolated,
with an average of 2.9 different strains per sam-
ple. Specimens with a preponderance of gram-
positive cocci (>50% of total CFU per square
centimeter) originated in six instances out of
seven from girls of less than 11 years of age.
Anaerobic nonsporulating gram-positive rods
also often dominated the periurethral flora. Five
of six samples with a preponderance of gram-
positive rods were obtained from girls in the
older age group. The mean number of different
types of bacteria identified per sample was 2.2.
Figure 2 shows the distribution of the different
genera of gram-positive rods in terms of per-
centage of the total CFU per square centimeter.
In contrast, clostridia were only isolated in one
instance, from a 10-year-old girl (7.5 x 102 CFU/
Anaerobic gram-negative rods were isolated
from several specimens, but were never predom-
inant. An average of 1.7 different strains per
specimen was obtained, but the majority ofthese
TABLE 2. Types and concentration oforganisms
present in periurethral samples from 18 healthy
No. of specimens at
(CFU/cm2 of periure-
S1 Si0 S50 >-50
:S1 s1o SW >S50
FIG. 1. Presence ofanaerobic gram-positive cocci,
gram-positive rods, and gram-negative rods,
pressed in percentage of total CFUper square centi-
meter in the periurethral samples from 18 healthy
VOL. 10, 1979
BOLLGREN ET AL.
S10 S50 >50
'10 S50 >50
Not identif ied
most frequent aerobic bacteria (Table 2). One 5-
year-old girl was heavily colonized with S. aga-
lactiae (1.3 x 106 CFU/cm2), and one 11-year-
old girl carried a homogeneous lactobacillus flora
(6.3 x 106 CFU/cm2).
Of potential pathogens of the urinary tract,
gram-negative rods belonging to the Enterobac-
teriaceae and enterococci were isolated from
only one and two samples, respectively, and in
low counts. S. epidermidis was isolated from
several specimens, but in small numbers (Fig. 4).
Three strains of coagulase- and deoxyribonucle-
ase-negative straphylococci were identified to
<10 550 >50
o10 S50 >50
Per cent of total
FIG. 2. Distributions of different anaerobic gram-
positive rods, expressed in percentage of total CFU
per square centimeter in the periurethral samples
from 18 healthy girls.
strains were recovered only from selective media
and in very small numbers. They represented
less than 1% of total CFU per square centimeter
(Fig. 1). The quantitative distributions of differ-
ent genera of gram-negative rods are given in
Fig. 3 as a percentage of the total CFU per
square centimeter. All strains of Bacteroides
species, which constituted 1% or more of the
total CFU per square centimeter (Fig. 3), were
isolated from the blood agar plates, whereas the
blood agar [Oxoid]) yielded no bacteroides or
very few colonies.
Anaerobic gram-negative cocci (Veillonella)
were isolated in one instance, from a 12-year-old
girl (2.3 x 103 CFU/cm2).
Aerobic microorganisms. The mean num-
ber of various aerobic strains isolated per speci-
men was 2.7 (range, 1 to 4). Gram-positive cocci
(mainly .onhemolytic streptococci) and gram-
positive rods (mainly diphtheroids) were the
S10 '50 >50
of total CFU/cm2
FIG. 3. Distributions of different anaerobic gram-
negative rods, expressed in percentage of total CFU
per square centimeter in the periurethral samples
from 18 healthy girls.
_ , ---.
CFU / cm2
0 So3 >103
FIG. 4. Viable counts of Escherichia coli, entero-
cocci, and S. epidermidis in the periurethral samples
from 18 healthy girls.
J. CLIN. MICROBIOL.
PERIURETHRAL MICROFLORA IN HEALTHY GIRLS
species. None turned out to be S. saprophyticus.
No strains of S. aureus or neisseriae were
isolated. Haemophilus vaginale was not
countered, which may either reflect the real
absence of this organism or be due to lack of the
use of a selective agar. Candida albicans was
recovered from three girls, 11, 10, and 7 years
old (1.0 x 103, 1.3 x 103, and 5.0 x 102 CFU/cm2,
respectively), and Saccharomyces
was recovered from one 14-year-old girl (2.5 x
This investigation of healthy premenarcheal
girls shows an indigenous periurethral flora that
is multibacterial and consists mainly of anaer-
obes, the most common being gram-positive
cocci and rods. There were great variations
among individuals with respect to the total via-
ble counts per square centimeter of periurethral
area and to the numbers of bacterial genera
represented by the isolates. However, the ratio
of aerobes to anaerobes was remarkably con-
Herein, a quantitative method has been used
to describe the periurethral microflora. Most
other studies of the urogenital region concern
only qualitative aspects, and certain bacteria
may have been overemphasized with a disregard
for others. In some previous studies, the speci-
mens were precultured in broth before being
inoculated onto solid media for isolation of dif-
ferent bacteria, a method that might favor rap-
idly growing species.
By using qualitative methods, certain bacte-
ria, e.g., Bacteroides species, may receive undue
emphasis, especially when present in low num-
bers, as a result of the choice of particularselec-
tive media. In this respect, quantitativemethods
are potentially advantageous since it may be
more obvious when certain bacteria constitute a
minor part of the flora. The risk of overestimat-
ing transient contaminants in small numbers
from adjacent microbiotas is accordingly re-
The findings of this study suggest that the
periurethral region is a bacteriological niche,
with a flora different from those of feces and
skin. The aerobic fecal flora comprises mainly
enterobacteria and enterococci, whereas these
bacteria are absent or very scanty in the periu-
rethral region of healthy girls (2). The present
study shows a similar dissociation between these
areas concerning the anaerobic flora. Thus, al-
though the anaerobic fecal flora is dominated by
Bacteroides species (6), these bacteria were ab-
sent or found in comparatively small numbersin
the periurethralarea. In addition, allbacteroides
strains constituting 1% or more of total CFU
were isolated from the blood agar plates but did
not grow on the selective medium (kanamycin-
vancomycin blood agar) used for isolation of
bacteroides of fecal origin. This phenomenon
may be indicative of distinct properties of the
bacteroides strains peculiar to the periurethral
With regard to the skin surrounding the uro-
genital area, the findings are more complex. The
majority of the bacteria isolated from the peri-
cocci, bifidobacteria, and eubacteria, have not
been reported in the skin flora, indicating that
these two microenvironments may be distinct
from each other (16). Nevertheless, such orga-
nisms as S. epidermidis, diphtheroids, and pro-
pionibacteria, which are typical constituents of
the skin flora (16), were often isolated also from
the periurethral area. However, the regular per-
iurethral findings ofS. epidermidis and diphthe-
roids make it probable that these bacteria are
true constituents of the flora of the periurethral
area, just as they are considered to be of the
flora of the vagina (1). Although propionibac-
teria were found in fairly high counts in several
periurethral samples, contamination seems un-
likely since the samples did not show corre-
spondingly high counts of the other species typ-
ical of the skin flora.
A more complicated matter is whether the
periurethral niche is distinct from the adjacent
urethral and vaginal microenvironments. Simi-
larities could well be expected as embryologi-
cally this whole area derives from the urogenital
sinus and is covered with squamous epithelium,
which in adults shows cyclic variations due to
hormonal influence (26). Other factors such as
different secretions may, however, create differ-
ences in the microenvironment.
Existing data suggest that the aerobic micro-
flora is similar throughout the distal urogenital
area, i.e., constituted of S. epidermidis, diphthe-
roids, nonhemolytic and a-streptococci, and, in
adult women, lactobacilli (8, 10, 19), indicating
that this whole area might reasonably be looked
upon as an entity. The whole area is also char-
acterized by scanty findings of potential uro-
pathogens such as enterobacteria and entero-
cocci (2, 10, 19). S. saprophyticus,another com-
mon cause of UTI in young adult women (18),
was not registered with the method used in this
study. More extensive studies concerning the
urethral flora (21) and the periurethralflora of
adult women (Bollgren, unpublished data) sug-
gest that S. saprophyticus rarelyis a constituent
of the normal flora in the distal urogenitalarea.
Comparisons with studies concerning the an-
aerobic microflora of these different partsof the
distal urogenital area are more complex. Re-
e.g., peptococci, peptostrepto-
VOL. 10, 1979
424 Download full-text
BOLLGREN ET AL.
cently, a study on the vaginal anaerobic micro-
flora of girls was published (9). This study, in
accordance with ours, revealed a multibacterial
anaerobic flora, with a mean of 5.3 anaerobic
species per vaginal sample, compared to 7.0 dif-
ferent strains per specimen in the present peri-
urethral investigation. The results of the vaginal
flora study, however, diverged from ours in other
respects, e.g., high isolation rates of clostridia
and Bacteroides species. Apart from the meth-
odological differences, more important may be
the fact that the two investigations studied dif-
ferent populations. Hammerschlag et al. (9) thus
reported that Bacteroides species were more
frequently isolated from girls under 3 years of
age than from girls 3 to 15 years of age. These
findings of anaerobic bacteria of probable fecal
origin in the vagina correspond well with our
previous studies showing high numbers of aero-
bic fecal bacteria in the periurethral area of
young girls (2).
Most other studies on the vaginal and urethral
anaerobic microfloras concern adult women,
where the high glycogen content of the mucosal
cells due to hormonal influences creates a mi-
cromilieu quite different from that in premen-
archeal girls. In spite of this, our findings in
many respects agree well with some quantitative
studies on the anaerobic vaginal flora in adult
women (1, 15), which showed that anaerobic
bacteria often constituted about 90% ofthe adult
vaginal flora, with a preponderance of gram-
positive cocci and rods.
From a functional point of view, the introital
area in adult women seems, at least, to have
characteristics in common with the periurethral
region in girls, since both areas may act as bar-
riers against UTI.
We thank Lena Pettersson for skilled technical assistance
in the laboratory. We are indebted to Cyril J. Smyth for
helpful comments and critical review of the manuscript.
Financial aid for this study was provided by the Swedish
Medical Research Council (grant 19X-765).
1. Bartlett, J. G., A. B. Onderdonk, E. Drude, C. Gold-
stein, M. Anderka, S. Alpert, and W. M. Mc-
Cormack. 1977. Quantitative bacteriology of the vagi-
nal flora. J. Infect. Dis. 136:271-277.
2. Bollgren, I., and J. Winberg. 1976. The periurethral
aerobic bacterial flora in healthy boys and girls. Acta
Paediatr. Scand. 65:74-80.
3. Bollgren, I., and J. Winberg. 1976. The periurethral
aerobic flora in girls highly susceptible to urinary infec-
tions. Acta Paediatr. Scand. 65:81-87.
4. Crowe, C. C., W. E. Sanders, and S. Longley. 1973.
Bacterial interference. II. Role of the normal throat
flora in prevention of colonization by group A Strepto-
coccus. J. Infect. Dis. 128:527-532.
5. Dornbusch, K., C.-E. Nord, B. Olsson, and T. Wad-
strom. 1976. Some properties of coagulase-negative
deoxyribonuclease-producing strains of staphylococci
from human infections. Med. Microbiol. Immunol. 162:
6. Drasar, B. S., and M. J. Hill. 1974. Human intestinal
flora. Academic Press, London.
7. Facklam, R. R., J. F. Padula, L. G. Thacker, E. C.
Wortham, and B. J. Sconyers. 1974. Presumptive
identification of group A, B, and D streptococci. Appl.
8. Fair, W. R., M. M. Timothy, M. A. Millar, and T. A.
tions of the urethra and vaginal vestibule in normal,
premenopausal women. J. Urol. 104:426-431.
9. Hammerschlag, M. R., S. Alpert, A. B. Onderdonk,
P. Thurston, E. Drude, W. M. McCormack, and J.
G. Bartlett. 1978. Anaerobic microflora of the vagina
in children. Am. J. Obstet. Gynecol. 131:853-856.
10. Hammerschlag, M. R., S. Alpert, I. Rosner, P. Thur-
ston, D. Semine, D. McComb, and W. M. Mc-
Cormack. 1978. Microbiology of the vagina in children:
normal and potentially pathogenic organisms. Pediat-
11. Holdeman, L. V., and W. E. C. Moore (ed.). 1975.
Anaerobe laboratory manual, 3rd ed. Virginia Polytech-
nic Institute and State University, Blacksburg.
12. Kallenius, G., and J. Winberg. 1978. Bacterial adher-
ence to periurethral epithelial cells in girls prone to
urinary tract infections. Lancet ii:540-543.
13. Kloos, W. E., and K. H. Schleifer. 1975. Simplified
scheme for routine identification of human Staphylo-
coccus species. J. Clin. Microbiol. 1:82-88.
14. Lancefield, R. C. 1933. Serological differentiation of hu-
man and other groups of hemolytic streptococci. J. Exp.
15. Levison, M. E., L. C. Corman, E. R. Carrington, and
D. Kaye. 1977. Quantitative microflora of the vagina.
Am. J. Obstet. Gynecol. 127:80-85.
16. Noble, W. C., and D. A. Somerville. 1974. Microbiology
of human skin. W. B. Saunders Co. Ltd., London.
17. Nord, C.-E., A. A. Lindberg, and A. Dahlback. 1974.
Evaluation of five test kits-API, Auxo Tab, Entero-
tube, PathoTec and R/B-for identification of Entero-
bacteriaceae. Med. Microbiol. Immunol. 159:211-220.
18. Pereira, A. T. 1962. Coagulase-negative strains of staph-
ylococcus possessing antigen 51 as agents of urinary
infection. J. Clin. Pathol. 15:252-253.
19. Pfau, A., and T. Sacks. 1977. The bacterial flora of the
vaginal vestibule, urethra and vagina in the normal
premenopausal woman. J. Urol. 118:292-295.
20. Saigh, J. H., C. S. Sanders, and W. E. Sanders. 1978.
Inhibition of Neisseria gonorrhoeae by aerobic and
facultatively anaerobic components of the endocervical
flora: evidence for a protective effect against infection.
Infect. Immun. 19:704-710.
21. Sellin, M., D.
Sylvester, and J. D. Andersson. 1975. Micrococcal
urinary-tract infections in young women. Lancet ii:570-
22. Shinefield, H. R., J. C. Ribble, and B. Marvin. 1971.
Bacterial interference between strains of Staphylococ-
cus aureus, 1960 to 1970. Am. J. Dis. Child. 121:148-
23. Silva-Hutner, M. 1974. Yeasts, p. 352-363. In E. H.
Lennette, E. H. Spaulding, and J. P. Truant (ed.),
Manual of clinical microbiology, 2nd ed. American So-
ciety for Microbiology, Washington, D.C.
24. Stamey, T. A., M. Timothy, M. Millar, and G. Mihara.
1971. Recurrent urinary infections in adult women. The
role of introital enterobacteria. Calif. Med. 115:1-19.
25. Sutter, V. L., V. L. Vargo, and S. M. Finegold. 1975.
Wadsworth anaerobic bacteriology manual, 2nd ed.
University of California, Los Angeles.
26. von Rutte, B., and I. Delnon. 1968. Die Urethralzyto-
logie und ihre Bedeutung zur Diagnose hormonal be-
dingter Miktionsstorungen. Praxis 16:555-561.
1970. Bacteriologic and hormonal observa-
I. Cooke, W. A. Gillespie, D. G. H.
J. CLIN. MICROBIOL.