Appi IED AND ENVIRONMENTAL MICROBIOLOGY, May 1983. p. 1474-1480
Vol. 45. No. 5
Entrainment of Viruses from Septic Tank Leach Fields
Through a Shallow, Sandy Soil Aquifer
JAMES M. VAUGHN,* EDWARD F. LANDRY, AND McHARRELL Z. THOMAS
Department of Energy and Environment. Brookhaven National Laboratory. Upton, New York 11973
Received 17 November 1982/Accepted 8 March 1983
A study was conducted which focused on movement of naturally occurring
human enteroviruses from a subsurface wastewater disposal system through a
shallow aquifer. The potential for significant entrainment of virus particles was /.
evidenced by their recovery at down-gradient distances of 67.05 m and from 7
aquifer depths of 18 m. A significant negative correlation was observed between
virus occurrence an- ie distance from the "septage" (leaching pool) source.
Virus occurrence could not be statistically correlated with either total or fecal
coliforms. indicating the limitations of current microbial water quality indicators
for predicting the virological quality of groundwater.
The past several years have witnessed the
development of considerable information con-
cerning the ability of human viruses to penetrate
soil and contaminate underlying shallow ground-
water systems (7, 16, 21-24). Given the potential
for virus contamination of an aquifer by sewage
disposal (e.g., subsurface disposal systems,
wastewater recharge basins), it is vital that infor-
mation concerning the tentt of viral entrain-
ment in groundwater be developed. The need for
such information is particularly acute in areas
such as Long Island where sewage-contaminat-
ed aquifers may serve as the sole source of
potable water for local residents.
The earliest evidence for the presence of
viruses in groundwater, based upon virus recov-
ery rather than epidemiological evidence, was
provided in 1972 by Mack et al. (10). They
isolated poliovirus type 2 from a 30.5-m-deep
drinking-water well located 91.5 m from the edge
of a wastewater drain field. The data suggested
both vertical penetration of soil and significant
lateral virus movement through the aquifer. Sim-
ilar findings were later reported by Wellings
(22), who described soil penetration and aquifer
entrainment of viruses resulting from the dis-
charge of secondarily treated wastewater
through a cypress dome. In a later study, Well-
ings et al. (25) recovered echovirus 22/23 from a
well located 30 m from a wastewater disposal
site. Vaughn et al. (21) noted sporadic isolations
of enteroviruses from two rapid wastewater in-
filtration sites located on Long Island. At the
first site, echovirus type 12 as well as several
unidentifiable isolates were recovered from a
groundwater observation well adjacent (3.3 m)
to a sand recharge basin receiving daily applica-
tions of secondarily treated wastewater. Virus
recoveries at the second site provided additional
evidence for lateral movement in the aquifer.
Here, viruses were recovered from an observa-
tion well located 45.7 m down-gradient from
recharge basins which received tertiarily treated
effluents (20). Schaub and Sorber (15), studying
groundwater beneath rapid wastewater infiltra-
tion basins, recovered viruses at soil depths of
30 m and at lateral distances of 182.8 m. Koerner
and Haws (8), observing the results of rapid
infiltration of primary effluent through coarse
gravel and sand, recovered virus from ground-
water at soil depths of 16.8 m and down-gradient
distances of 250 m.
The above data clearly demonstrate virus en-
trainment as a result of surface disposal. How.
ever, few studies have specifically focused on
vertical and lateral movement of viruses emanat.
ing from subsurface disposal systems. Vaughn
and Landry (19) isolated unidentifiable virus
types from a subsurface leaching pool which
received secondarily treated effluents. In the
same account, they reported a single isolation of
coxsackievirus B3 from an observation wel
located some 402-m down-gradient from a sani.
tary landfill. Marzouk et al. (11, 12) isolated
enteroviruses in 20.2% of groundwater samples
tested in Israel. Although the precise source of
sewage contamination was not identified, the
authors suggested the likely sources as being
subsurface seepage from septic tanks and under.
ground sewer lines. Wellings et al. (25) demon.
strated the presence of echovirus in a series of
10.6- to 12.1-m-deep groundwater wells which
were clustered in an area surrounded by septic
tanks. A similar virus source was proposed by
Vaughn and Landry (19) after the isolation of
echovirus types 11 and 23, and coxsackievirus
VIRUS ENTRAINMENT IN GROUNDWATER 1475
A-16 from a groundwater observation well locat-
ed in the center of a cluster of single family
dwellings which discharged wastes to individual
The present study was designed to assess the
lateral transport of indigenous human viruses
through a shallow sole-source aquifer which
received discharges from a subsurface disposal
system. The study focused on the likely contri-
bution of such systems to shallow aquifer con-
tamination, and their potential impact on nearby
MATERIALS AND METHODS
Site selection and well installation. The site chosen
for the study was the Hampton Gateway apartment
complex, Speonk, N.Y., located on the south shore of
eastern Long Island. The complex consisted of three
buildings, each containing 40 apartment units. Since
the original plans had called for the eventual construc-
tion of three additional buildings (80 units), the dispos-
al system (Fig. 1) had been designed to accommodate
six apartment buildings, serving approximately 300
people. The system consisted of a 5.7 x 10*-liter septic
tank, three distribution pools (4.25 m3), and 40 leach-
ing pools (4.95 m'). Each pool was installed approxi-
mately 3 m below grade, which resulted in a 3.6-m
unsaturated zone above the static groundwater level.
As the system operated at approximately one-half
capacity, peripheral pools received little flow from the
central septic tank.-The major leachate plume was
found to be concentrated around the central pools.
Plume and groundwater-gradient measurements were
conducted by the Groundwater Resources Section of
the Suffolk County Department of Health Services.
Their investigations showed that the groundwater flow
was south 400 west. with a hydraulic gradient of 2.22
m/1.61 km. The configuration of the leachate plume
was determined on the basis of physical and chemical
measurements made on groundwater samples taken
from eight test wells driven down-gradient from the
leaching pools. Measurements included conductivity.
ammonia, nitrate, nitrite, and chloride determinations.
Subsequent data indicated that the septage percolate
was concentrated down-gradient from the center of the
sewage disposal system, moving slightly away (south)
from the hydraulic gradient. On the basis of these
examinations, a series of monitoring wells was driven
along the line of predicted plume movement. Using the
center of the system as the point of origin, wells (5.08-
cm diameter) were driven at distances of 1.52, 3.05,
4.57, 7.62, 10.67, 15.24, 22.86, 30.48, 45.72, and 60.35
m. An additional well was driven 5 m north of the
central pool array to serve as an up-gradient control.
Each well was fitted with steel piping, having a 0.91-m
stainless steel screen section set 1.52 to 1.83 m below
the top of the water table. Also tested during the study
was an existing drinking-water well located 67.05 m
(18.29-m depth) down-gradient from the septage
source. Sample water from this well was collected
from a tap located on the side of one of the apartment
As seen in Fig. 1, a sewer line serving building
number 3 crossed the study area between the 15.24-
and 22.86-m monitoring wells. Since any virus isola-
tions from these wells could be questioned on the basis
of possible leaking from cracks or defective joints, the
Health Department provided a detailed television in-
spection of the sewer line to determine its status. In
addition to optical inspection, each joint was individ-
ually pressure tested (testing conducted by PENGAT
Contracting Corporation, Setauket, N.Y.). All joints
in the vicinity of the testing wells were found to be
intact. Inspection of the entire pipe indicated no other
defects at any point.
Field sampling. Sample collection. Groundwater
samples (378 liters) were collected from each sample
well at biweekly intervals during the initial phase of
the study. Samples were collected in 209-liter acid-
sterilized tanks (Plasti-Cube; Grief Brothers Corp.,
Staten Island, N.Y.) which had been arranged in
permanent sets for exclusive use at each sample well.
Pumping equipment and hosing (also arranged in sets
for certain well groupings) were acid sterilized before
each collection and thoroughly rinsed with 50 to 100
liters of sample water before the filling of the tanks. Bi
weekly, 38-liter septage samples were also collected in
containers used exclusively for this sample type. The
above measures (plus others described below) were
undertaken to avoid the likelihood of cross-contamina-
tion between different sample types.
Virus concentration and enumeration. Concentration
of viruses from large-volume samples involved a filter
adsorption-beef extract elution-organic flocculation
method previously described by Landry et al. (9).
Briefly, samples were acidified to pH 3.5, supplement-
ed with 0.5 mM AICI3, and passed through a virus-
concentrating filter series consisting of a fiberglass
depth cartridge filter (K-27) and a pleated cartridge
filter (Duo Fine, Timonium, Md.). Viruses were eluted
from concentrating filters with 1-liter volumes of 3%
beef extract-0.5 mM Tris (pH 9.5). Virus reconcentra-
tions from I-liter eluates followed the technique of
Katzenelson et al. (6). Eluate pH was adjusted to 3.5,
causing the formation of a virus-adsorbing protein
precipitate. After 30 min, the percipitate was collected
by centrifugation (5,000 x g for 10 min), and the
resulting pellet was dissolved in 10- to 25-ml volumes
of 0.15 M dibasic sodium phosphate (pH 9.0). Concen-
trates were then neutralized to pH 7.2 and stored at
-600C to await assay.
Virus concentration units were acid sterilized be-
tween each sample processing. Sterilization was ac-
complished by filling all portions of each system
(hoses, filter holders with appropriate filters, pumps)
with 1.2 N HCI and allowing a 30-min contact period.
Systems were then thoroughly iinsed with tap water
and appropriate sample water. Two separate units
were constructed for use in the study, one for the near-
well samples (1.52- to 10.67-m wells), the other for far-
well samples (45.72- to 67.05-m and control wells). The
relative efficiencies of both units were tested before
initiation of the sampling program, using poliovirus-
seeded groundwater collected from various study
wells. Recovery efficiencies for both units ranged
between 65 and 87% (data not shown).
Viruses were enumerated on monolayers of Buffalo
green monkey kidney cells which had been propagated
in Eagle minimum essential medium with Earle bal-
anced salt solution supplemented with 5% calf serum,
glutamine, and antibiotics (penicillin, streptomycin,
and gentamicin). Sample concentrates were diluted
VOL. 45, 1983
1476 VAUGHN, LANDRY, AND THOMAS
APPL. ENVIRON. MICROBIOL.
FIG. 1. Virus-monitoring well locations. Each well is numbered, and its distance from the septage source was
as follows: well 1, 1.52 m; well 2. 3.05 m; well 3, 4.57 m; well 4, 7.62 m; well 5, 10.67 m; well 6, 15.24 m; well 7,
22.86 m; well 8. 30.48 m; well 9, 45.72 m; well 10, 60.35 m; well 11, 67.05 m; well 12 (north control well), 5 m up-
gradient; well 13 (northeast control well), 34.6 m up-gradient. Unlabeled wells were used for determining the
direction of groundwater flow and were not sampled during the present study. Well 12 was abandoned early in
the study and well 13 was substituted.
(1:3), inoculated (4 ml) on'Buffalo green monkey
kidney cell monolayers in 75-cm2 flasks, and incubated
with rocking for 2 h to facilitate virus adsorption. After
decanting excess inoculum, monolayers were overlaid
with 15 ml of minimal essential plaquing medium
supplemented with 2% calf serum, agar, and antibiot-
ics (20, 21). The flasks were incubated for 1 to 3 days
at 36C under 5% CO2 and then stained with a second
overlay (10 ml) containing agar and neutral red. The
flasks were then returned to the incubator and ob-
served over an 8-day period for plaque development.
All PFU were confirmed by two passages on cell
monolayers. Only samples showing consistent cyto-
pathic effect were considered positive. Because of the
relatively low virus numbers expected, entire sample
concentrates were assayed.
Additional measurements. Total and fecal coliform
enumerations involved the use of a three-tube, most-
probable-number method performed in accordance
with standard methods (2). Ambient sample pH was
measured with a portable pH meter (Hach model
1975). Sample conductivities were measured with a
portable salinity-conductivity-temperature meter
(model 33; YSI Corp.). Data were statistically ana-
lyzed by methods described previously by Sokal and
Rohlf (17) and Steel and Torrie (18). Statistical analy-
ses were performed on a Hewlett-Packard HP9845B
desk-top computer, using preprogrammed statistical
Virus occurrence in septage distribution pools
and shallow (1.82-m) groundwater observation
wells. The system described above provided a
unique opportunity for studying the characteris-
tics of in situ virus entrainment through an
aquifer receiving a constant input of domestic
septic leachate. A summary of the virus recov-
ery data from the distribution pool, as well as the
control and down-gradient observation wells
VIRUS ENTRAINMENT IN GROUNDWATER 1477
screened at water table depths of 1.82 m, is
presented in Table 1. Although isolates were not
specifically identified, the concentration-recon-
centration methods and host cell system used
considerably favored the recovery of enterovi-
ruses (21). The proximity of the test wells to a
domestic wastewater disposal system left little
doubt as to the human origin of the isolates.
Initial samplings conducted at the control well
located some 5 m to the north of the leaching
pool area (well number 12) yielded virus isolated
in two of five samples tested. As there was no
likely up-gradient source of viruses, it was con-
cluded that the isolations were the result of
mounding of septage-contaminated groundwater
occurring beneath the leaching pools. With the
control value of this well compromised, sam-
pling was transferred to a new up-gradient well,
located 34.6 m to the northeast of the main
leaching pool area (well number 13). All subse-
quent testing of this well failed to yield any virus
isolates, a fact which tended to support the
contention that any viruses isolated from test
wells located below the leaching system (down-
gradient) had originated from the system, rather
than from some unknown up-gradient source.
Analysis of samples collected from the central
distribution pools, which were often highly toxic
to cell cultures, indicated the routine occurrence
of viruses. Virus concentrations recovered from
these samples (ranging from 0.07 to 148 PFU/
liter) were of little predictive value, as virus
residence times in distribution and leaching
pools could not be determined. It was, there-
fore, impossible to quantitate either the total
number of viruses being discharged to the leach-
ing pools or the level being released by the
leaching pools to the unsaturated soil. Qualita-
tively, however, the data served to indicate the
regular discharge of viruses into the leaching
The data from analysis of samples collected
over a 12-month period from shallow, down-
gradient wells indicated extensive entrainment
of indigenous viruses through the aquifer (Table
1). Near-well samples (i.e., those at lateral dis-
tances of 1.52, 3.05, and 4.57 m) were frequently
too toxic for tissue culture assay. Toxicity was
rarely noted in samples collected from beyond
the 10.67-m well, a fact which strongly suggest-
ed that the leaching pools were the principal
source of near-well toxicity. Virus isolates were
recovered from each of the down-gradient wells
on at least one occasion. In general, virus con-
centrations varied with well distance from the
septage source, with levels ranging from 0.002 to
10.8 PFU/liter detected in near-well samples and
concentrations of 0.002 to 0.05 PFU/liter occur-
ring in the most distant (i.e., 45.72- and 60.35-m)
wells. When the data were subjected to a weight-
ed regression analysis technique (17, 18), a sig-
nificant negative correlation (P < 0.05; r =
-0.6492) was revealed between the frequency of
virus occurrence and increasing distance from
the septage source. The weighted analytical
method was used to compensate for the relative-
ly low sampling frequency at some of the wells
(i.e., numbers 3, 5, and 9), giving more credence
to data from wells which had been sampled more
Coliform organisms, although occurring in
high concentrations in distribution pool samples
TABLE 1. Virus recoveries from shallow-screened (1.82-m) test wells
Distance No. of
Observation well' from samples No. No. Percent
source () collected tested positive positive'
source (m) collected
Septage distribution 0 15 8 7 87.5
1 1.52 28 19 8 42.1
2 3.05 16 14 5 35.7
3 4.57 3 2 1 50.0
4 6.09 16 14 5 35.7
5 10.67 10 7 3 42.8
6 15.24 16 16 1 6.2
7 22.86 26 25 4 16.0
8 30.48 22 22 4 18.2
9 45.72 3 3 1 33.3
10 60.35 22 22 2 9.1
12J 5.0 5 5 2 40.0
13' 34.6 12 12 0 0.0
SData from sample well number 11, screened at an 18-m depth, appear in Table 2.
b Sample concentrates which were not toxic to cell cultures.
(Number positive/number tested) x 100.
SControl well located north of the septage source.
SControl well located northeast of the septage source.
VOL. 45. 1983
1478 VAUGHN, LANDRY, AND THOMAS
(105 to 108/100 mi), were rarely detected beyond
the 1.52-m sample well (data not shown).
Virus occurrence in deep groundwater observa-
tion wells. By the end of the first 6 months of
sampling, it had become apparent that viruses
could be extensively entrained through the up-
per (1.82-m) portion of the aquifer. It could not
be determined on the basis of the data, however,
whether viruses could be carried to greater
aquifer depths. The delineation of such a capaci-
ty was important, as most of the private drink-
ing-water wells in the region draw from consid-
erably deeper portions of the glacial aquifer. As
the only deep-screened well being tested at the
time was the apartment drinking-water supply
(well number 11), located 67.05 m from the
septage source and screened at approximately
18 m, we decided that additional wells should be
installed and tested. The new wells were located
in close proximity to existing test wells. One
well (number 14) having a 6.09-m screen depth
was sunk adjacent to the 1.52-m test well (well
number 1). Three additional wells installed near
the 30.48-m test well (well number 8) had screen
depths of 6.09, 12.19, and 18.29 m, respectively
(wells 15, 16, and 17). Sample collections, proc-
essing, and assay were performed as described
above. To accommodate the influx of new sam-
ples, we discontinued testing at wells 2, 4, 5, and
Attempts to recover viruses from well number
14 were unsuccessful (Table 2). This was pre-
sumably due in pan to the high incidence of
sample toxicity (45.5%). Isolates were recov-
ered, however, from wells 15 and 16 (screened at
6.09- and 12.19-m depths, respectively), located
some 30.48 m below the septage source (i.e.,
down-gradient). The presence of viruses in these
wells indicated the likely mixing of septage-
contaminated waters at these depths. This mix-
ing was attributed to draw-down effects caused
by sampling at the respective well points. Simi-
lar effects might be expected during normal
usage of any supply well. Although viruses were
not recovered from the deepest well at 30.48-m
distance (well number 17), they were isolated on
one occasion from the drinking-water well (well
APPL. ENVIRoN. MICROBIOL.
number 11), which was screened at a similar
depth (18 m). Virus recovery at this latter well
represented the most extensive lateral virus
movement observed during the study (67.05 m).
Virus occurrence related to other measure-
ments. Virus occurrence was compared with pH
and conductivity data via product-moment cor-
relation matrices. The analysis indicated no gen-
eral relationship between any of the parameters
(Table 3), although several isolated instances
were noted in which a significant positive corre-
lation occurred between pH (26.0) and virus
occurrence (e.g., wells 3, 10, and 15). These
results differed dramatically from the majority of
observation well samples, and no definitive rela-
tionship could be established. Likewise, a single
negative correlation was observed between con-
ductivity values and virus occurrence in well
number 3 (Table 3). Because this relationship
was based on only two observations, the corre-
lation was considered to be an isolated and
Product-moment correlation analysis of the
occurrence of total or fecal coliforms with the
presence of enteroviruses in groundwater sam-
ples indicated that no significant relationship
existed between these parameters (Table 4).
These findings were not unexpected in light of
earlier studies which had empirically demon-
strated that no correlation existed between virus
and coliform occurrence in sewage-impacted
groundwaters on Long Island (21). The present
findings provided an important statistical verifi-
cation of earlier data.
Virus occurrence in groundwater as related to
season. Of 280 samples collected during the
course of the study, 218 were sufficiently non-
toxic to permit testing on the cell culture system.
Eighty-eight of these were from two spring
(March through May) collections, whereas 29,
50, and 51 were from a single summer (June
through August), fall (September through No-
vember), and winter (December through Febru-
ary) collection, respectively. Data were ana-
lyzed for seasonal fluctuations by using
Friedman's test for two-way classification (18),
which ranked the frequency of virus occurrence
TABLE 2. Virus recoveries from deep-screened test wells
Observation Distance n No. of No. No. Percent
well sorv nfro n) depth ( ) c ple tested' positive positive"
well source (m) depth (m) collected
14 1.52 6.09 11 6 0 0.0
15 30.48 6.09 11 11 1 9.1
16 30.48 12.19 11 11 2 18.2
17 30.48 18.29 10 10 0 0.0
11 67.05 -18 13 13 1 7.7
Sample concentrates which were not toxic to cell cultures.
S(Number positive/number tested) x 100.
- -------- -- u--------
VIRUS ENTRAINMENT IN GROUNDWATER 1479
TABLE 3. Product-moment correlation matrix
defining the relationship between virus occurrence in
observation wells and sample pH or conductivity
Values in parentheses represent number of obser-
Significant linear correlation at P < 0.05.
Coefficients not calculated owing to inability to
recover virus at indicated sample wells.
(i.e.. total number of positive samples divided
by the total number tested) at each well for each
season. The rankings were then summed, and a
chi-square value was calculated. The test was
applied only to data from down-gradient, shal-
low (1.82-m) wells which had been sampled
during all four seasons of the year. The analysis
revealed no significant difference in overall virus
occurrence frequency as a function of season.
This result indicated that the likelihood of recov-
ering virus from any point along the test gradient
could not be confined to a particular time of the
year and that viruses appeared to be homoge-
neously dispersed throughout that portion of the
aquifer under study.
Microbial contamination of groundwater sys-
tems represents a major source mechanism for
waterborne disease outbreaks (4, 5, 13, 14). In
most cases, contamination of shallow aquifers
has been the direct result of wastewater disposal
at or near the soil surface, with subsequent
lateral movement through the aquifer (1). The
most often identified sources of contamination
have been subsurface septic systems and cess-
The most commonly identified agent in drink-
ing water-associated outbreaks occurring be-
tween 1971 and 1977 was Giardia lamblia (5).
However, 57% of the recorded outbreaks could
not be associated with a specific etiological
TABLE 4. Product-moment correlation matrix
defining the relationship between v,:us occurrence
and the presence of total or fecal coliform organisms
well Total Fecal
1 -0.0812 (17) -0.0911 (15)
2 -0.0472 (13) NF"
3 NT' NF
4 -0.0862 (14) -0.0263 (13)
5 0.1208 (7) -0.1196 (7)
6 -0.0706 (16) NF
7 -0.0448 (25) NF
8 -0.0181 (24) NF
9 0.5000 (3) NF
10 -0.0808 (21) NF
11 -0.1280 (12) NF
12 0.6053 (5) NF
13 NT NF
14 NVd NF, NV
15 NT NF
16 NT NF
17 NT, NV NF, NV
Values in parentheses represent number of obser-
b NF, Coefficient not calculated owing to inability to
recover fecal coliforms at indicated sample wells.
c NT, Coefficient not calculated owing to inability to
recover total coliforms at indicated wells.
d NV, Coefficient not calculated owing to inability
to recover viruses at indicated wells.
agent, a fact which led Craun (5) to speculate on
the possibility of a virus etiology. The potential
for waterborne virus disease transmission un-
derscores the need for identifying the fate of
viruses in a sole-source aquifer.
In the present study, viruses were isolated
from groundwater at distances of up to m
down-gradient from the leaching pools o-a sub-
surface wastewater disposal system. Although a
majority of isolates were recovered from wells
drawing'from the upper portion of the aquifer,
the potential for extensive vertical entrainment
of viruses within the aquifer was evidenced by
virus recovery from wells which had been
screened at 6-, 12-, and 18-m de ths. The pif-
ic limits o virus entrainment through the system
could not be determined, as wells had not been
installed beyond 67 m dwngradient) r deer
Current building codes in many portions of the
county where groundwater serves as both a
potable water source and an eventual recipient
of domestic wastewater require that a minimum
distance of 3 8 m be observed in the place-
ment of domestic s s u 'ace-disposalsystems
and down-gradient private drinking water wells.
In those instances in which the minimum dis-
tance cannot be applied, a supplementary regu-
lation requires that wells be installed an addi-
VOL. 45, 1983
1480 VAUGHN, LANDRY, AND THOMAS
tional 0.91 m in death for each 0.3 m under the
recommended minimum distance. The results of
the present study indicate the prudence of ques-
tioning this regulation in areas having a hydro-
geological profile similar to that of Long Island.
In such regions, should virus-free water be de-
sired, it may be advisable to consider either
increasing the lateral placement distance or drill-
ing significantly deeper wells. The virological
ramifications of such measures are currently
being pursued by this laboratory and by the
Suffolk County Department of Health Services.
Statistical analysis indicated no overall pat-
terns of significant correlation between virus
occurrence and occurrence of fecal and total
coliform organisms. This result, in conjunction
with similar findings from previous studies (3,
15. 16), provided strong evidence repudiating
the use of current microbial water quality stan-
dards for predicting the likely virological quality
This project was conducted by Brookhaven National Labo-
ratory under a contract with the Suffolk County Department of
Health Services. N.Y. We wish to acknowledge the invaluable
efforts of Aldo Andreoli. Joseph Baier. and Craig Wherle,
Suffolk County Department of Health Services, and Lois
Baranosky. Marilyn Dahl. and Cheryl Beckwith, Brookhaven
National Laboratory. in the preparation and conduct of this
research. Statistical analyses were kindly provided by Keith
Thompson. biostatistician. Brookhaven National Laboratory.
I. Allen, M. J. 1978. Microbiology of groundwater. J. Water
Pollut. Control Fed. 50:1342-1344.
2. American Public Health Association. 1975. Standard meth-
ods for the examination of water and wastewater. 14th ed.
American Public Health Association, Inc., Washington,
3. Burge, W. D., and P. B. Marsh. 1978. Infectious disease
hazards of landspreading sewage wastes. J. Environ.
4. Cramn, G. F. 1979. Disease outbreaks caused by drinking
water. J. Water Pollut. Control Fed. 51:1751-1760.
5. Craun, G. F. 1980. Disease outbreaks caused by drinking
water. J. Water Pollut. Control Fed. 52:1833-1839.
6. Katzenelson, E., B. Fattal, and T. Hostovesky. 1976.
Organic flocculation: an efficient second-step concentra-
tion method for the detection of viruses in tap water.
Appl. Environ. Microbiol. 32:638-639.
7. Keswick, B. H., and C. P. Gerba. 1980. Viruses in ground-
water. Environ. Sci. Technol. 14:1290-1297.
8. Koerner, E. A., and D. A. Haws. 1979. Long-term effects
of land application of domestic wastewater. Environmen-
tal Protection Agency Report 600/2-79-047. Ada, Oklaho-
APPL. ENVIRON. MICROBIOL.
9. Landry, E. F., J. M. Vaughn, M. Z. Thomas, and T. J.
Vicale. 1978. Efficiency of beef extract for the recovery of
poliovirus from wastewater effluents. Appl. Environ. Mi-
10. Mack, W. N., Y. Lu, and D. B. Coohoon. 1972. Isolation
of poliomyelitis virus from a contaminated well. Health
Serv. Rep. 87:271-274.
11. Marzouk, Y., S. M. Goyal, and C. P. Gerba. 1979. Preva-
lence of enteroviruses in groundwater of Israel. Ground
12. Marsouk, Y., S. M. Goyal, and C. P. Gerba. 1980. Rela-
tionship of viruses and indicator bacteria in water and
wastewater of Israel. Water Res. 14:1585-1590.
13. McDermott, J. F. 1974. Virus problems and their relation
to water supplies. J. Am. Water Works Assoc. 66:693-
14. National Academy of Science. 1977. Drinking water and
health, chapter 3. In Waterborne microorganisms. Na-
tional Academy of Science. Washington. D.C.
15. Schaub, S. A., and C. A. Serber. 1977. Virus and bacteria
removal from wastewater by rapid infiltration through
soil. Appl. Environ. Microbiol. 33:609-619.
16. Sobsey, M. D., and J. Scandura. 1980. Enteric microor-
ganism removal by on-site wastewater disposal systems in
coastal plain soils. Interim Project Report. Sanitary Engi-
neering Section. University of North Carolina, Chapel
17. Sokal, R. R., and F. J. Rohlf. 1969. Biometry. W. H.
Freeman and Co.. San Francisco.
18. Steel, R. G. D.. and J. H. Torrle. 1960. Principles and
procedures of statistics. McGraw-Hill Book Co., New
19. Vaughn, J. M., and E. F. Landry. 1977. Data report: an
assessment of the occurrence of human viruses in Long
Island aquatic systems. Brookhaven National Laboratory
Report 50787. Upton. New York.
20. Vaughn, J. M., and E. F. Landry. 1978. The occurrence
of human enteroviruses in a Long Island groundwater
aquifer recharged with tertiary wastewater effluents, p.
233-243. In H. L. McKim led.). State of knowledge in
land treatment of wastewater, vol. 2. U.S. Army Cold
Regions Research and Engineering Laboratory. Hanover,
21. Vaughn, J. M.. E. F. Landry, L. J. Baranosky, C. A.
Beckwit,. M. C. Dahl, and N. C. Dellhas. 1978. Survey of
human virus occurrence in wastewater-recharged ground.
water on Long Island. Appl. Environ. Microbiol. 36:47-
22. Welllngs, F. M. 1976. Viral aspects of wetland disposal of
effluent, p. 297-305. In Proceedings of the Symposium on
Freshwater Wetlands and Sewage Effluent Disposal. Am
23. Wellings, F. M., A. L. Lewis, and C. W. Mountain. 1977.
Survival of viruses in soil under natural conditions. In
F. M. D'ltri (ed.), Wastewater renovation and re-use.
Marcel Dekker. Inc., New York.
24. Wellings, F. M., A. L. Lewis, C. W. Mountain, and L V.
Plerce. 1975. Demonstration of virus in groundwater after
effluent discharge onto soil. Appl. Microbiol. 29:751-757.
25. Wellings, F. M., C. W. Mountain, and A. L. Lewis. 1977.
Viruses in groundwater in individual on-site wastewater
systems, p. 61-65. In N. I. McClelland (ed.). Proceeding
of the Second National Conference. Ann Arbor Science
Publishers, Inc., Ann Arbor, Mich.
Selection of bactei
has been desired as
pool of phage-unrelate
cocci available for u
cultured dairy product
Most isolated mutan
production of their pa
sensitivity to the ori
were attacked by new
stability of lactic store
both rate of acid prod
specificity is well doct
the possibility of a dirt
two properties has ne
rence (14) mentioned
producing activity of t
that mutations alterin
might interfere with
transport systems also
fce. Lactose metabolic
ty in lactic streptococci
19, 20), and whether th
ally linked has not beer
23). Some proteinases
associated with the cell
cocci (5-7, 22, 27), ane
associated structures an
sorption, their loss mi
phae sensitivity. We stt
dlo of phage-resistant
producing variants to
whether these properties
APPLIED AND ENVIRO,
Copyright C 1983, Amt