OF SOIL FERTILiTY: ThE LONG-TERM EH±CTS OF

OF SOIL FERTILiTY: ThE LONG-TERM EH±CTS OF
SOIL MICROBIAL ACTIVITY AS AN INDICATOR
OF SOIL FERTILiTY: ThE LONG-TERM EH±CTS OF
MUNICIPAL SEWAGE SLUDGE ON AN ARID SOIL
by
Jeffrey Walter Brendecke
Copyright© Jeffrey Walter Brendecke 1992
A Thesis Submitted to the Faculty of the
DEPARTMENT OF SOIL AND WATER SCIENCE
In Partial Fulfillment of the Requirements
For the Degree of
MASTER OF SCIENCE
In the Graduate College
THE UNIVERSITY OF ARIZONA
1992
2
STATEMENT BY AUTHOR
This thesis has been submitted in partial fulfillment of requirements for an advanced degree at The University of Arizona and is deposited in the University Library to
be made available to borrowers under rules of the Library.
Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgement of source is made. Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be
granted by the copyright holder.
J4Jk'
SIGNED:
APPROVAL BY THESIS DIRECTOR
This thesis has been approved on the date shown below:
2
Date
vLk
3
ACKNOWLEDGEMENT
No work is the result of any one individual's effort, and the present writing is no
exception. The following fellow human beings deserve, written or otherwise suitably recoMed, a mention of their names and contributions, if it be nothing more than the paltry
sounding of gratitude which I can offer here. I give my thanks to, first of all, Jean Brendecke, my best friend and wife, for her help in successfully navigating each and every
day, trying though the Herculean effort may have been, not to mention her help in proofreading. Thanks be given to Dr. Ian L. Pepper, my advisor, for guidance and assistance in
setting up the microbiological analyses and for evaluating the thesis, and to Drs. Hinrich
L. Bohn and P.J. Wierenga for for their editorial assistance in simplifying my often verbose, overembellished prose, thereby bringing the manuscript to publishable standards.
Also of special mention are Karen Josephson and Rachel Bailey, for never-failing logistical support in times of experimental panic. I also thank Rick Axelson, for the many, long
hours of statistical revelation, examination, and not to be forgotten, instruction. Due to
his efforts, otherwise opaque data saw the light of day. Other portentous contributions
were made by Tom Hilbert and David Confer for assistance, both material and instructional, in the determination of direct bacterial counts in the soil samples; Dr. Michael Ottman of the Department of Plant Science, for supplying the cotton plant growth data; Dr.
J. Artiola, for assistance in interpreting sewage sludge chemical analyses; workers in the
Soil, Water, and Plant Testing Laboratory in the Department of Soil and Water Science
for the analyses they performed on the soil samples; and to Dr. Steven Smith of the Department of Plant Science for supplying the alfalfa for the carbon dioxide evolution study.
4
TABLE OF CONTENTS
LIST OF ILLUSTRATIONS
6
LIST OF TABI FS
7
ABSTRACT
8
INTRODUCTION
9
LITERATURE REvIEw
12
History and Social Acceptance of Land Application
13
Effects of Sewage Sludge Application on Soil Physical and Chemical
Properties
15
Heavy Metal Toxicity as a Function of Soil Properties and the Sludge
Matrix
17
Sewage Sludge as a Crop Fertilizer
20
Soil Microbial Activity as an Indicator of Soil Fertility
24
The Long-Term Effects of Sewage Sludge and Heavy Metals on Soil Microbial Activity
30
MATERIALS AND METHODS
35
Experimental Design
35
Statistical Analyses
37
Analyses of Variance
37
Measures of Association Between Two Variables
38
Sewage Sludge Analyses
40
Soil Physical and Chemical Analyses
40
Soil Microbial Analyses
42
5
Viable Heterotrophic Plate Counts for Bacteria
42
Acridine Orange Direct Counts for Bacteria
44
Viable Heterotrophic Plate Counts for Actinomycetes
47
Viable Heterotrophic Plate Counts for Fungi
49
Fungal Counts by Serial Dilution
49
Fungal Counts by the Drop Plate Method
50
Dehydrogenase Assay (DHG)
51
Carbon Dioxide Evolution Analysis
54
RESULTS
57
Effects of Sewage Sludge on Soil Physical and Chemical Properties and
Heavy Metal Contents
Effects of Sewage Sludge Application on Cotton Plant Growth Parameters
Effects of Sewage Sludge on Soil Microbial Activity
57
.
. . .62
64
Enumeration Assays
64
Dehydrogenase Assay
65
Carbon Dioxide Evolution Analysis
68
DISCUSSION
70
CoNcLusioNs
83
APPENDIX: TABLES OF ASSOCIATIONS BETWEEN DEPENDENT VARIABLES
84
LIST OF REFERENCES
95
6
LIST OF ILLUSTRATIONS
Figure 1: Significant trends in physical and chemical properties of the soils as a
function of total sludge application over a four-year period.
58
Figure 2; Significant trends in soil heavy metal contents (total and DTPA-extractable) as a function of total sludge application over a four-year period.
60
Figure 3: Significant trends in stand of cotton as a function of total sludge application over a four-year period.
63
Figure 4: A significant trend in DHG with GC amendment as a function of total
sludge application over a four-year period.
66
Figure 5: CO2 evolution as a function of time for the three sludge treatments in response to moisture amendment.
68
Figure 6: CO2 evolution as a function of time for the three sludge treatments in response to moisture + alfalfa amendment.
69
7
LIST OF TABLES
Table 1: Mean chemical analyses of the sewage sludge applied at Marana by year
35
Table 2: Means of the values for soil physical and chemical properties by sludge
treatment.
57
Table 3: Means of the soil acid digestible and DTPA-extractable heavy metal contents.
59
Table 4: Theoretical added metal concentration in the field soil due to sludgeborne heavy metals as calculated from sludge metal contents and amounts of
sludge applied to the soil on a liquid volume basis.
61
Table 4: Mean cotton plant growth parameter values.
62
Table 5: Means of the microbial enumeration assays and DHG results.
65
8
ABSTRACT
This study measured the effects of four years of municipal, anaerobically digested
sewage sludge application on long-term soil microbial activity in a Pima clay loam
(Typic Torrifluvent) growing Upland cotton (Gossypium hirsutum L.). Treatments were
one unfertilized control, sludge applied at optimum rates for plant growth (based on N requirements), and sludge applied at three times the optimum rates (a total of 8.01 and 24.2
Mg ha1 (dry weight over four years)). Soil microbial activity was measured by viable heterotrophic plate counts for bacteria, actinomycetes and fungi; acridine orange direct
counts for bacteria; the dehydrogenase assay; and carbon dioxide evolution analysis. As
the high sludge treatment significantly reduced cotton plant stand and significantly stimulated some parameters of microbial activity (dehydrogenase activity and CO2 evolution),
soil microbial activity may not serve as a reliable predictive index of plant response to
sludge-applied pollutants.
9
INTRODUCTION
In 1986 a study was begun at the University of Arizona Marana Agricultural Center north of Tucson to determine the effects of long-term sewage sludge application on
crops under the environmental conditions of the Sonoran Desert. The investigation was
initiated by Pima County Wastewater Management Department to evaluate the benefits
of sludge on agricultural land, as all of the sewage sludge produced by the City of Tucson
is applied to agricultural land. So far, yields obtained by area farmers using the sludge
have been good with a number of crops, including cotton. By disposing of the sludge on
land, the County has rid itself of a waste problem, and the farmers have received a valuable organic soil amendment.
However, good yields of cotton may only be a transient, superficial phenomenon.
Soil analyses at the Marana site have shown that the heavy metal copper has been increasing in plant availability to make it the first heavy metal to limit the lifetime loading rate
of the sludge. Taylor etal. (1989) recommended using microbial activity as an "indirect
index of soil fertility" in the Marana cotton plots, as preliminary data had suggested that
"microbial numbers and activity of sludge amended soil at Marana are different than non
sludged soils, following 4 annual applications of sludge." To be sure, high levels of microbial activity are synonymous with high soil fertility. Microbes convert nutrients from
a form unavailable to higher plants into a soluble form readily available to plants. Without the activities of the soil microorganisms, plant residues and mineral nutrient forms
would remain immobilized and unavailable as organic matter, minerals, or gas (Teutscher
10
etal., 1960). It was believed that if the soil were rendered infertile by contaminants in the
sludge, the population or activity of soil microorganisms that play an important role in
maintaining soil fertility would be harmed in some way, possibly before the plants growing in that soil were visibly affected. In such a fashion, one could evaluate whether the
given method of waste disposal damaged the soil, hopefully before the field had been irreparably polluted. Although it is well established that a totally infertile soil would be essentially void of microbial activity, could slight variations in measurements of soil
microbial activity be positively correlated with agronomic plant growth parameters?
Could indices of microbial activity be used as an indicator of increasing or decreasing fertility of a soil?
An auxiliary study was begun to evaluate the effect of four years of sewage
sludge application on the soil microorganisms at the Marana study site. The null hypotheses tested in the study were:
The application of sewage sludge to the plots at Marana had no significant effect on any of the measured microbial parameters.
There was no significant association between microbial activity and soil properties that could be affected by sludge application.
There was no significant association between soil microbial activity in the
sludge treated plots and cotton plant growth parameters.
11
Given the arid conditions under which the study was performed, valuable conclusions could be derived for the practice of land disposal of sewage sludge in the unique climate and soils of arid regions, as little information on this subject is currently available.
12
LITERATURE REVIEW
Sommers and Barbarick (1986) have described the main constraints to land application of sewage sludge. These include pathogens, organic pollutants, heavy metals, and
nitrate pollution. Modern sludge does share many traits with the sludge commonly applied to agricultural land in the past. The potential for spreading disease still exists, as
does the potential for groundwater contamination by nitrates. However, due to higher
population density, the problems have become exacerbated. Furthermore, toxicant pollu-
tion stemming from the present level of industrialization poses new problems that must
be dealt with. The effects of municipal sewage sludge on the health and fertility of the
soil and its inhabitant populations, both plant and microbial, are as complicated as the issues that surround its acceptance by the broad public. In this review, the history of land
application and some of the concerns raised about its use in a historical context will be
briefly summarized, followed by a discussion of the effects of sludge on soil properties
associated with fertility and factors influencing heavy metal toxicity from sewage sludge.
The review will conclude with a synopsis of research on the viability of sludge as a crop
fertilizer and an overview of research performed on using microbial activity as an indicator of soil fertility and the effects of sewage sludge pollutants, i.e., heavy metals, on soil
microorganisms.
13
History and Social Acceptance of Land Application
Applying human excrement to land for agricultural purposes or disposal dates
back over 2000 years, although success with the practice for raising crops has historically
been mixed, ranging from good to abysmal, with failure often associated with an ignorance of soil properties (Fuller, 1983). Back in the nineteenth century, it was commonplace for the inhabitants of large cities, above all in New England and the Middle
Atlantic States, to collect their excrement in privies or cesspools, which, as required by local ordinances, were emptied at night by workers hired by the city (hence, the term
"night-soil"). The waste was then sold to farms for fertilization of their fields. However,
this practice came to an end due to post-Civil War urban growth and with the advent of
running water and the flushed toilet. Concern about outbreaks of diseases, such as typhoid fever, led to the promotion of the flushed toilet (or water closet) which in turn overloaded the cesspools with the diluted waste. Hence, sewerage, i.e., a system of sewers for
carrying sewage, was developed for disposing of the waste; the age of the "night-soil
men" came to an end (Goldstein, 1976). Sewage treatment plants which could process
sewage year-round instead of during periods of plant growth were developed. This development rendered costly land application by irrigation (Tarr, 1975) and cleaning of the
cesspools (Folwell, 1922) unprofitable. Consequently, discharging the sewage into rivers
and oceans became a common practice (Goldstein, 1976). Yet, objections were raised
that the use of sewerage was polluting rivers and streams and agriculturally valuable flutrients were being lost: "The consensus of opinion is that the two latter objections are
14
more than balanced by the advantages which it [sewerage] offers in comparison with any
other method yet tried for the removal of excreta and liquid wastes" (Foiwell, 1922). Yet,
with increasing environmental awareness, much money has been put into treating the
sludge to make it less of a hazard, but a proportionally small amount is being devoted to
responsible management of the treated material (Goldstein, 1976).
The current resurgence of interest in land application began in the U.S. in the
1930s with more impetus in the 1960s as environmental concerns grew (Tarr, 1975). Yet,
as there has been a time lapse since land application has fallen into disfavor, effort must
be made by the organizations involved to educate the public, farmers and city dwellers
alike, that land application of sewage sludge can be perfonned in an environmentally
sound and socially acceptable manner.
Whether or not land application of sludge becomes wide-spread may hinge on persuading the public and administrative agencies that the practice is safe. Forster et al.
(1986) summarized the basic "institutional constraints" that have inhibited wide-spread
acceptance of land application. Such constraints include attitudes as public fears of disease, odors, and decreased land values; as well as the bias of municipal officials not familiar with the subject, although regulatory constraints appear to be relaxing, especially at
the local level of government. Proper marketing and public relations (Goldstein, 1976);
the implementation of well managed and economically viable programs, such as the
Madison, Wisconsin Sewerage District's Metrogro program (Barnes, 1989; Taylor,
1989), combined with good research, such as that being done at the University of An-
15
zona Marana Agricultural Center, towards developing responsible sludge management
practices; and regulations controlling the input of harmful materials into the sludge
(Boyle, 1990) can help rekindle public acceptance.
Effects of Sewage Sludge Application on
Soil Physical and Chemical Properties
Sewage sludge has proven to be valuable in the amelioration of infertile land
(Goldstein, 1976). Many examples exist where, for example, sludge has been successfully used to restore mine spoils. One case involves the reclamation of coal mine surface
sites in Pennsylvania (Seaker and Sopper, 1988ab). The one-time application of woodchip-amended, composted sewage sludge from the Philadelphia area to five limed test
sites planted to tall fescue (Festuca arundinacea Shreb.), orchard grass (Daclylis glomerata L.), and birdsfoot trefoil (Lotus corniculatus L.) over a period of five years, was
shown to result in higher plant dry matter yields, plant litter accumulation, soil organic
matter content, and Kjeldahl nitrogen content than an inorganically fertilized comparison
site. Additionally, microbial populations and carbon dioxide evolution measured in the
soil were higher in the sludge amended sites than in the inorganically fertilized site, possibly, as the authors concluded, due to the complexation of heavy metals found in the
acidic mine spoils by organic compounds from the sludge.
Similar results on coal mine spoils, this time in illinois, were obtained (Pietz et
al., 1989ab). A combination of sewage sludge and lime, or sewage sludge alone was
shown to be effective in raising spoil pH (5.0 as compared to a control plot value of 2.8)
16
and reducing KC1 extractable aluminum. Revegetation of the spoils as measured by plant
cover and dry matter yields of a mixture of alfalfa, bromegrass, and orchardgrass were
highest in the sludge + lime treatment, however, plant tissue heavy metal content results
were mixed. At another acidic strip mine site in illinois, Stucky et al. (1980), found that
after a one-time application of sewage sludge and three growing seasons, reed canary
grass (Phalaris araundinacea L.), switchgrass (Panicum virgatum L.), and orchardgrass
accumulated heavy metals at levels not considered to be phytotoxic despite being grown
in a substrate containing potentially large amounts of Cd, Cu, Mn, Pb, and Zn. The
authors contributed the establishment of the grasses to the increase in pH brought about
by the sludge (3.0 initially to 4.4-5.5).
Studies in an agricultural setting have also yielded evidence for the improvement
of soil properties through sludge application. Fresquez etal. (1990) found an increase in
the soil organic matter (SOM) content after two years of sludge application to a degraded,
semiarid grassland. Water stable aggregates (WSA) may be enhanced through sludge ap-
plication, as Metzger etal. (1987) found using a Typic Haplargid, due to an increase in
fungal activity. Epstein etal. (1976) noticed an increase in soil moisture retention in
Aquic Hapludult plots amended with a dry sludge compost and anaerobically digested
sludge over the unamended control. Soil cation exchange capacity (CEC) increased by a
factor of three in a high sludge treatment. They did, however, notice a decrease in the
CEC with time, possibly due to the oxidation of CEC-causing organic matter associated
with the sludge. Soil organic carbon increased with sufficiently large application rates as
17
did total nitrogen and Bray-extractable phosphorous. Additionally, O'Connor et al.,
(1986) working with calcareous soils in New Mexico, observed that NaHCO3-exiractable
P increased in sludge amended soils, attributing the phenomenon to slight pH reductions
and concomitant mobilization of the P tied up in various forms of calcium phosphate in
the sludge amended soils.
Heavy Metal Toxicity as a Function of Soil
Properties and the Sludge Matrix
As mentioned above, the current procedure for disposing of sewage involves processing at a sewage treatment plant. Although the actual methods and degree of treatment
vary, the end result is often the separation of the sewage into the heavier sludge and effluent, which may be discharged into bodies of water, used for irrigation of turf or other
vegetation, or used to recharge ground water. The separation of sewage sludge from effluent has been found to be an effective method of removing heavy metals from the effluent,
although certain types of digestion can increase the amount of metals released from the
sludge into the effluent (Ghuman, 1985). Yet, the problem of sludge disposal remains. In
any case, the best method for keeping the levels of heavy metals low in the sludge is to
prevent them from entering the sewers in the first place; proper municipal regulation has
been shown effective in maintaining a quality suitable for land application (Goldstein,
1976).
Heavy metals contaminate land by a variety of pathways including the air (Pb
from gasoline), fossil fuel ash, agricultural fertilizers (especially phosphate fertilizers)
18
and pesticides, animal manures (from Cu and As in feed additives), urban and industrial
wastes, metal industry waste, and mining waste (Alloway, 1990b). Copper and zinc are
common pollutants in sewage sludge through the use of copper plumbing (Baker, 1990).
The potential for heavy metals to pose a hazard is as much a function of the form
in which the metals arrive in the soil and properties of the soil, as it is of the metal concentration. Heavy metals undergo a variety of complex reactions in the soil which drastically affect their availability. These include adsorption to and diffusion into clay
minerals, coprecipitation with other ions in the soil solution, and complexation with organic matter, with the stability of these complexes in the order Cu > Fe = Al > Mn = Co
> Zn (Alloway, 1990a). Furthermore, additional complexation could occur with organic
matter during the digestion process at the sewage treatment plant and by incorporation
into microbial biomass. Toxic metal availability can vary over time as the metal ions diffuse into stronger absorption sites, are incorporated into solids, and as reactive and toxic
forms of the metals are transformed into stable and less toxic variations (Bohn et al.,
1985).
Given the many factors which affect metal availability, it can be difficult to determine the concentration of available metal in the soil solution. As the soil cation exchange
capacity (CEC) is a function of the types of clays in the soil and their abundance, as well
as the soil organic matter content (CAST, 1976; Fuller, 1983), a number of states have
adopted guidelines which limit the amount of a given hazardous metal that can be applied
to the land on the basis of the CEC of the soil at the application site (Arizona Department
19
of Environmental Quality, 1990; Department of Land, Air & Water Resources, University of California, Davis (1984); compare with USEPA, 1983). Soil pH also strongly affects heavy metal availability, with metals generally available at acidic conditions and
less so under basic conditions, although there are exceptions. Pepper et al. (1983) found
that the plant availability of Cd, in contrast to Zn, increased at higher p11.
Adams and Sanders (1984) added Cu and Zn salts to the raw sludge prior to
chemical treatment in a pilot treatment plant with aluminum chiorohydrate or added Ni
salts during sedimentation and adjusted the pH of the frozen samples dispersed in water.
Using a cation exchange resin in the mixture, they found an exponential relationship between metals extracted from the sludge in the supernatant (removed by the resin) and PH,
with threshold values for Zn at 5.8, Ni at 6.3, Cu at 4.5, above which the level of metals
in the supernatant increased little with increasing pH. Additionally, they found lower levels of metals were released by a metal-contaminated, unamended control sludge than by
themetal loaded sludge at a given pH with 38.7, 47.2, 25.9 percent of the metals in the
supematant of the metal amended sludges and 18.2, 50.5, 1.4 percent for Zn, Ni, and Cu
in the control sludge at pH values of 4, 3, and 3, respectively. They concluded that the Cu
threshold was so low that Cu toxicity should not present a problem in arable soils, even
with pH lowering in the rhizosphere. However, a problem could exist for Zn and Ni.
Similar results with respect to the availability of indigenous sludge metals were obtained
by Bell etal. (1991) with cadmium.
20
In view of the above research, the relationship between soil heavy metal content
and plant toxicity has been difficult to ascertain. Recently, Chang et al. (1991) made a review of 260 research articles concerning the effect of sludge-applied heavy metals and
phytotoxicity and have found that up to the present no unequivocal evidence for phytotoxicity due to sludge-applied metals exists. They also determined that EPA guidelines for
sludge metal loading rates tended to overestimate the phytotoxicity of the metals. Consequently, they suggested introducing a new measure of loading rate based on the probability of a loaded metal to cause phytotoxicity in a plant when the metal was present in a
concentration equal to the phytotoxicity threshold. To this end, legal definitions of phytotoxicity would be needed that would 1) define some criterion that must be met for determining whether phytotoxicity has taken place, such as yield reduction, 2) eliminate other
causes of the observed symptoms, 3) define biochemical pathways that took part in producing the observed toxicity symptoms, and 4) ascribe a specific metal to the production
of the symptoms.
Sewage Sludge as a Crop Fertilizer
A number of investigators have researched the effects that land applied sewage
sludge has on crop growth. Higgins (1984) performed a study in New Jersey where
sludge was applied to a sandy loam (Typic Hapludult) at 0, 22.4, and 44.8 Mg ha' for
three years with the fourth year serving as a "residual" year. Dent corn (Zea mays L.) was
planted for grain and silage, with rye (Secale cereale L.) as the winter cover crop, which
21
was harvested in spring prior to seed formation. Concentrations of Zn and Cu significantly increased in the A horizon (pH 6.2) whereas Fe, Ni, Cr, and Pb did not (although
the author did not state so, these increases were probably noticeable because the background metal contents were so low). Heavy metal contents of corn stover or grain did not
rise with the exception of Cu, Zn, Ni, or Cd (stover) and Fe (grain), whereas increases of
Zn and Cu in the rye were noticed for some years with no increases in the residual year.
All of the measured yield parameters tended to increase with sludge application,
often with no significant difference between the two sludge application levels. In the
fourth year, yields of silage and grain rose with increased application of sewage sludge,
as nitrogen, which had accumulated in the soils of the higher sludge treatment, was mobilized.
Higgins (1984) mentioned that there is a need to balance crop fertilization requirements with sludge disposal. This may have resulted in overfertilization on the experimental plots and the problems with nitrate contamination and increases in electrical
conductivity (EC) in the groundwater measured in wells at the application site and down
gradient from the plots. Given the first year sludge content of 66.5 g kg' total N and
15.0 g kg1 N}13-N (100% available, no nitrate was measured), 51.5 g kg' organic N was
applied to the soil, with 40% available the year of application (Watson et al., 1985). This
resulted in an application of Ca. 800 kg ha' available N the first year, and, similarly,
160 kg ha1 N the last year, in addition to what was released from soil N accumulated
over the previous years. Yet, corn requires only 160 to 180kg ha' inorganic N for the
22
grain yields obtained at the study site (Zscheischler et al., 1979). It is possible that adding
more sludge than what was necessary for plant growth artificially accelerated heavy metal accumulation and led to excess nitrate contamination of the groundwater.
Crop plants can be successfully grown under conditions of the arid southwestern
U.S. deserts when sewage sludge is used as a soil amendment. In a six year study on the
growth of wheat (Triticum aestivum L.) on a calcareous soil (Typic Calciorthid) using
dried sewage sludge from the City of Phoenix, Day etal., (1987) examined three different fertilization schemes involving sewage sludge at 10 Mg ha' as compared to inorganic
fertilizer applications (suggested rates for N, P, and K and inorganic equivalents of N, P,
and K in the sludge). After six years, it was found that the sludge application did not decrease grain yield or straw yield, and all three treatments had roughly the same average
number of days to harvest, plant height, grain yield, grain volume weight, and grain:straw
ratio. Additionally, the grain was without excessive amounts of zinc, lead, copper, and
nickel.
Day et al. (1988), this time in the Avra Valley near Tucson, applied sludge at
rates of 78, 112, and 224 kg ha' N for one year to a Grabe sandy loam (Typic Torrifluvent) in addition to an inorganic fertilizer treatment (78 kg ha1 N). Similar to the previous study, wheat was planted. They found that sewage sludge increased the time
needed to head, but plant heights, number of heads per unit area, yields of hay, grain, and
straw were not affected by treatment. Higher levels of sludge did not significantly affect
grain yields. The behavior of heavy metals varied with the metal. Cadmium and Ni were
23
below detectable limits in all plant parts, and the accumulation of Cu, Pb, and Zn was not
a significantly affected by treatment. Higher concentrations of metals were observed in
the grain than in the straw of mature wheat.
Another experiment in southeastern Arizona was conducted to determine the
sludge loading rate which would maximize Upland cotton (Gossypium hirsutum L.) yield
without significant metal accumulation in the plant (Watson et al. 1985). Dried sludge
from the City of Tucson was applied to an alkaline (pH 8.2) Pima clay loam (Typic Torrifluvent). Lint yields in the inorganic fertilizer (at recommended rates of N and P) and
annual sludge treatments for three years (17, 21, and 21 Mg ha'; 34, 41, and 42 Mg ha';
and 85, 81 , and 83 Mg ha1) were significantly higher than in the unfertilized controls.
The yield from a one time application treatment of 85 Mg ha1 sludge was significantly
higher than the control only in the first year. Heavy metal concentrations in the seed were
not significantly different between treatments. Leaf heavy metal content showed Cd to be
the only metal significantly higher than the control for the two highest annual sludge application rates and the inorganically fertilized treatment; the investigators pointed to possible Cd contamination in the inorganic fertilizer as the cause.
In view of the above studies, sewage sludge can be used under certain conditions
as a safe crop fertilizer, at least from the perspective of continued agricultural productivity and public health concerns with respect to heavy metal accumulation in plant parts.
Nevertheless, it could be possible that the sludge may affect soil microorganisms through
heavy metal or organic toxin contamination, causing an eventual decrease in soil fertility
24
through disruption of fertility-maintaining processes. To this end, literature on studies using soil microbial parameters as an index of soil fertility will be reviewed, followed by
studies on the effects of heavy metals on soil microorganisms.
Soil Microbial Activity as an Indicator of Soil Fertility
By nature, studies involving the use of one parameter to predict another need to
make use of some sort of numerical measurement of the degree to which the parameters
are related. Measures of association of two variables, such as Pearson product-moment
correlation coefficient, are often used for this purpose. However, no cause and effect relationship can be drawn between two variables using a measure of association without having randomly assigned treatment levels to one or more independent variables, as hidden
variables, known or unknown to the investigator, may explain some or all of the variability in the dependent variable. In fact, inappropriate use of measures of association may in
fact be a contributing factor to the difficulty researchers have experienced in correlating
microbial parameters with other phenomena, as well as the inconsistency of results between soils or even treatments. Consequently, when the term "correlate," or any derivative thereof, is used in this review, this is merely for purposes of convenience and does
not imply a cause and effect relationship between variables.
Many of the reactions involved in nutrient recycling occur as a result of the activity of soil microorganisms. Without these organisms, dead tissue from both plant and animal sources would accumulate, preventing the essential nutrients within from being
25
utilized by other organisms. Most of the carbon dioxide in the atmosphere originates
from the oxidation of organic matter by soil microorganisms (Bohn et al., 1985), which
can produce 10,000 kg CO2 ha1 y1 (Schachtschabel et al., 1984). This CO2, is, of course,
the main source of carbon brought into biogeochemical cycles by higher plants and
chemolithotrophic microorganisms. Microorganisms mineralize organic matter, liberating
many important nutrients, e.g., nitrogen and phosphorous, and temporarily immobilize
nutrients in their bodies, preventing them from being leached from the soil (Schachtscha-
bel etal., 1984). Soil microorganisms also play an essential role in soil formation from
parent material and the formation of soil organic matter (Paul and Clark, 1989), which
heavily contributes to the tilth, water holding capacity, aeration, and cation exchange Capacity of the soil. Thus, the activity of soil microorganisms is essential in the maintenance of fertile soil, not only in natural settings, but also in agricultural fields. Certainly,
a soil with no soil microbial activity could not maintain its fertility for an extended period
of time.
A number of studies to determine the correlation of soil microbial activity and
soil fertility have been performed, but these have often served to show that a direct relationship between the two is difficult to quantify. In the second decade of this century, a
rather comprehensive series of articles by Selman Waksman were published in Soil Science under the title "Microbiological Analysis of Soil as an Index of Soil Fertility," re-
viewing a number of the more commonly used, contemporary techniques of quantifying
microbial activity, namely plate counts, ammonification, nitrification, cellulose decompo-
26
sition, carbon dioxide evolution, and nitrogen fixation, as applied to experimental plots
set up by the Soil Department of the New Jersey Agricultural Experiment Stations for
studying the availability of nitrogen in various fertilizers (Waksman, 1922abc; Waksman,
1923abc; Waksman and Starkey, 1924; Waksman and Heukelekian, 1924; Waksman and
Karunakar, 1924). In summary, Waksman found that levels of microbial activity were
higher in the more productive soils while lower in the more marginal soils. He, however,
cautioned that a direct correlation of soil fertility and soil microbial activity may be difficult to quantify:
These two factorsmicrobiological activity and soil productivityneed
not of necessity be related. The one, crop production, is at any moment dependent to a large extent upon the inorganic nutrients present of the soil,
while the other, microbiological activity, is regulated to a much larger extent by the abundance of soil organic matter. A soil composed of little else
than quartz sand with available elements essential to plant growth may
support plants temporarily and still lack any abundant microbial flora. In
such cases the microbiological activities and soil productivity are not correlated but the first may be better considered as forecasting the future possibilities of the soils (Waksman and Starkey, 1924).
In addition, soil microbial activity and soil fertility are by no means independent
of each other, nor are they independent of the soil:
In judging a soil [for its fertility], microbial activity should also be drawn
upon. If one is namely to assume that plant stand and soil properties influence microbial activity, it can just as well be assumed that microbial activity can have consequences for plant stand and soil properties (Glathe and
Thalmann, 1970c, translated from the Gennan by the author).
Other studies done in various parts of the world have indicated that there is some
correlation between management practices and microbial activity, although these results
27
do not apply universally. Experiments involving measurements of microbial activity tend
to revolve around three fonns of measurement: 1) enumerations of the organisms using
various techniques, such as counting the colonies formed on petri dishes containing a Suitable growth medium; 2) measurements of physiological activity, such as the rates of evolution of carbon dioxide due to respiration by the organisms or estimations of the activity
of particular enzymes involved in metabolism of all or selected groups of microorganisms; and 3) measurements of microbial biomass, which may be derived from 1) or 2) or
be determined by independent means. However, due to the diversity of the types of organisms involved and the diverse properties of the soils under study, the results of microbial
analyses in the soil environment may be difficult to quantify on an absolute basis, much
less interpret.
Much of the research on the effects of various treatments on soil microbial activity has involved comparing conventional and "organic" agricultural practices, using soil
microbial activity as a measure which management technique serves to maintain a
healthy soil. Unfortunately, not all of these studies compare crop yields. One study in
Washington (Bolton et al., 1985) found the activity levels of several enzymes (urease,
phosphatase, and dehydrogenase (DHG)) and microbial biomass were significantly
higher in the soil of a farm where organic farming had been practiced (higher organic
carbon and nitrogen levels), i.e., where leguminous crops planted to supply nitrogen and
maintain soil fertility, than in the soil of an adjoining farm where contemporary conventional farming practices were used, although the significance of the differences varied
28
with the season. Periods of higher DHG were said to coincide with times of heavy organic matter oxidation. In contrast, plate counts (bacteria, actinomycetes, fungi) showed
no real differences. Yields were higher than average for the area at the organic farm, but
no correlation between microbial activity and yield was made.
Fraser etal. (1988), comparing conventional and organic management practices
in Nebraska, found that soils fertilized with manure had significantly higher levels of organic C and significantly higher DHG (compare research on the relation of activity to soil
carbon content (Dutzler-Franz, 1977ab; Schuller, 1989)). The type of crop planted also
had some effect on microorganisms in that the management of the crop affected the soil
organic matter balance. No mention was made of yields. Similar findings were made by
Doran et al., (1987) on a farm converted from conventional to organic management, and
Martyniuk and Wagner (1978), in a study conducted at Sanborn Field in Missouri where
fertilization studies had been underway since 1888.
Verstraete and Voets (1977) in Belgium came to comparable conclusions. Nevertheless, as indicated by their research, not all aspects of microbial activity may be correlated with the changes in management practices. Their study involved various cropping
schemes using green manures, incorporation of crop residues, and farmyard manure application. They used microbial activity analyses to detect the effects of the various management practices on the soil. They were able to divide the activity analyses in two groups:
1) those that are not sensitive enough to detect minor differences in microbial activity,
e.g. protease and DHG, and 2) those that were sensitive to variable soil factors. However,
29
they found few correlations between crop yield and biological activity, which depended
to a great extent on the season in which the assays were performed.
Another series of experiments was performed in the Federal Republic of Germany
to fmd relationships between plant production and DHG, plate counts, and soil carbon dioxide evolution as well as to correlate the measurements of microbial activity with each
other (Glathe and Thalmann, 1970c). Using a number of different soil types and various
crop rotations plant yields and DHG were poorly correlated. The effect of organic fertilizers on DI-IG was generally positive, but dependent on the type of amendment used, the
season, soil, climate, and time after fertilization that the test was run. Little relation was
found between DHG and plate counts, but the correlation was better with CO2 evolution,
although differences between soils appeared. With the application of different organic
amendments to a vineyard soil, DHG was highest with peat and manure, followed by
brewer's grain and green manure. Bacterial plate counts did not correspond to DHG, although fungi were favored with the application of organic amendments without differences between treatments. Another result was that organic fertilizers had more of an
effect on microbial activity, and inorganic fertilizers had a greater effect on yield (compare this to (Waksman and Starkey, 1924), as quoted above); inorganic fertilizers could
affect microbial activity indirectly by increasing plant growth and root secretions. Therefore, as the researchers concluded, one cannot derive nutrient availability and possible
yield from microbial activity.
30
As is apparent from an overview of research conducted with the intent of correlating soil microbial activity with soil fertility, the factors involved may at times be so complex as to make the use of a few tests insufficient for finding subtle differences in
fertility. To remedy such a problem, the use of indices derived from the results of a battery of tests (Soil Microbiological Index (SM!), Beck, 1984ab) may be useful, but one
finds few reports of these in the literature, possibly due to the level of expertise needed
for each of the tests and the length of time involved in carrying out the analyses.
The Long-Term Effects of Sewage Sludge and
Heavy Metals on Soil Microbial Activity
We have seen that parallels between soil microbial parameters and soil fertility
can be made under certain circumstances. Now, let us turn our attention to using this tenuous indicator as a means of measuring the effect of pollutants on soils.
In his review, Duxbury (1985) discussed a number of ways in which heavy metal
pollution can affect soil microbial ecology. He cited research that indicated that environmental heavy metal pollution can affect the diversity of microbial communities as well as
the numbers of microorganisms, with some organisms more abundant in polluted than unpolluted soils. Furthermore pollutants may affect antagonistic relationships beteen soil microorganisms. He also summarized a number of the mechanisms by which microrganisms
are resistant to or can adapt to heavy metals evolutionarily. For example, numerous studies have shown that Gram-negative bacteria are more resistant to heavy metals than
Gram-positve bacteria, possibly because of inherent differences in cell wall structure. Ad-
31
ditionally, evidence for the existence of metal-tolerance plasmids and Iransposons has
been found.
Grossbard
(1973)
cautioned that one cannot use any one measurement of soil mi-
crobial activity for assessing pollution damage in a soil, as different activities may be affected to a different extent by the pollutant. Likewise, Duxbury
(1985)
pointed out that
much of the research on the effects of heavy metals on microbially mediated processes is
conflicting, and the mechanisms involved not well understood. Additionally, as in the
case with soil fertility, so many factors can affect microbial activity that any test performed at the current level of precision would probably be more affected by different environmental factors than by the pollutant. With sewage sludge, the pollutant often
translates to heavy metals, which currently are of great concern as they do not degrade
with time as many organic compounds do, although their availability may decrease with
time.
Studies on the effects of heavy metals on microorganisms can generally be placed
into two categories: 1) metal salt amendments to nutrient agar or broth, and 2) metal
amendments to soils, in the form of metal salts added directly to the soil or loaded in sewage sludge, or as sludge that is contaminated with heavy metals by origin. Heavy metals
applied to nutrient broth or agar would tend to be more available than if amended to soil
or digested in sewage sludge. Beck
(1981),
in a study to determine the effects of heavy
metals applied as salts on bacteria and fungi, observed rather high metal concentrations
were needed to bring about a 50% reduction in plate counts of bacteria and fungi (for ex-
32
ample, 15 to 22 ppm Cu2, for bacteria) if the metals were added as salts to agar. These
values are in the same range as those reported by Premi and Cornfield (1971) in sludge
amended soils exhibiting a decrease in nitrification. As one would expect, Beck found
drastically higher concentrations of metals were needed to reduce the plate counts of soil
bacteria when the metals were added to the soil. This phenomenon was especially appar-
ent in the case of Cu. Tests on DHG gave mixed results; Hg2 reduced DHG and Cr3 increased it when the salts were added to the reagent solution. Fungi were less sensitive
than bacteria to most metals. Similar findings were made by Angle et al. (1991) who argued for a new measure of soil metal concentration based on metal activities in aqueous
solution.
Beck and Suss (1979), investigating the effects of gamma-irradiated and pasteurized sludge on soil microbial activity, observed in field experiments with three soils (all
pH 7.0) amended with sludge of low heavy metal contents (from a residential area) at
three application levels, 130 (yearly for 3 years), 400 (one time), 800 m3 ha' (yearly)
(5 to 10% solids), a 250% increase of bacterial populations after the first year; by the second year the counts had leveled off to the values found in the untreated control. Likewise,
the activity of reductases (catalase and DHG) and hydrolases (amylase, cellulase, protease, alkaline phosphatase) increased 150 to 250%. By the fourth year of the experiment,
microbial activity and biomass were at least as high if not higher than the control with the
highest activity almost exclusively corresponding to the highest rate of sludge applica-
33
tion. There were, nevertheless, some differences to which degree the measured microbial
parameters rose in response to sludge application.
Field studies on the effects of sludge on microbial populations have at times taken
advantage of long-term studies set up to evaluate the sludge as just another fertilizer.
Long-term studies should prove to be invaluable for assessing the effects of heavy metals
in a way that cannot be simulated in the laboratory or by modelling alone. One such longterm study was conducted in Great Britain (Brookes and McGrath, 1984; McGrath and
Brookes, 1986). A sandy clay loam was fertilized in three treatments: 1) inorganic fertilizer (1942-1967), 2) farmyard manure at 5.2 or 10.4 Mg ha' y' (1942-1967), and 3) anaerobically digested, air-dried sewage sludge at 8.2 or 16.4 Mg ha' y1 from 1942-1961.
Soil pH values were found to be Ca. 6.5 to 6.8. Copper, as extracted by 0.05 M EDTA,
was found to be 34-9 1 mg kg1 in the sludge treated soil. After 1967, all plots were given
equal treatments of inorganic fertilizer. The researchers found that viable heterotrophic
plate counts for bacteria, actinomycetes, fungi, and protozoa were unaffected by sludge
treatment. DHG was lowest in the sludge amended soils, as was oxidation of added N}1
and N0, although NO production was not much different than the inorganic fertilizer
treatment (but lower than the manure treatment). Total microbial biomass was 50% of the
value for the manure and inorganic fertilizer treatments. Based on CO2 evolution from unfumigated soil and soil ATP concentration values which were similar for all treatments,
the authors concluded that under the stress of the heavy metals more substrate was
34
needed to form an equivalent amount microbial biomass as compared with unpolluted
soils, or that the organisms lived less long. Twenty years after the last sludge application,
metal contamination was still at U.K. maximum allowable values and reduced microbial
activity as measured by biomass production was still evident. The authors correlated
biomass depression with Cu and Ni content in the soil.
35
MATERIALS AND METHODS
Experimental Design
The study was carried out at the University of Arizona Marana Agricultural Cen-
ter (Field C-4) on a Pima clay loam soil (382 g kg' sand, 367 g kg' silt, 261 g kg1 clay
site described in Post a al., 1978). The field was laid out as a randomized complete block
design with four blocks and three treatments: 1) an unfertilized control (no sludge); 2) liquid, anaerobically digested sewage sludge from the Pima County Treatment Plant at ma
Road applied at a rate to supply approximately the recommended levels of nitrogen
Table 1: Mean chemical analyses of the sewage sludge applied at Marana by year.
Solids, Total C, Kjeldahl N, and Total P are in g kg' dry sludge, and all other values
in mg kg' dry sludge. nr=not recorded.
Year of Application
Variable
Solids
1986
1987
1988
1989
9.5
14
13
24
Total C
nr
252
250
281
Kjeldahl N
48
44
52
41
NH4-N
3550
391
NO3-N
516
210
nr
nr
nr
nr
Total Inorganic N
nr
nr
100
300
TotalP
17
8.2
30
27
Zn
1140
800
1130
1590
Cu
898
568
761
957
Pb
221
89.3
152
175
Ni
50.8
26.0
46
50.0
Cr
77.0
32.0
53
94.5
Cd
10.2
7.33
13
14.5
Ag
45.8
2.67
63
59.7
36
(202 kg ha1, as measured as the sum of total Kjeldahl N and NH4-N in the sludge filtrate)
to the soil (low sludge); and 3) liquid sewage sludge applied as in the low sludge treatment but in three successive applications (high sludge). Mean analyses for the sludge are
given in Table 1. The sewage sludge was applied to the soil prior to planting by injection
20-30 cm deep in the soil using a terragator with floatation tires, beginning in February,
and followed by ripping to minimize soil compaction. The fields, measuring 6.10 x 172
m, were planted in April to Upland cotton (Gossypium hirsutum L.) cultivar DPL-41,
which was harvested twice successively in the fall.
Soil samples for the microbiological analyses were taken on December 22, 1989
using an Oakfield Soil Sampling Kit (Soiltest, Inc., Lakebluff, illinois, USA) with a
1.905 cm 0 corer, taking Ca. 100 cores by walking down the length of the field beginning
and ending ca. 10 m from the field edge to avoid areas affected by uneven application
rates at the start of the sludge application run. As the center of the field was judged visually, it is believed that samples were taken without respect to the position that the sludge
injector might have taken. The soil was collected after the fields had been harvested for
the final time (November 30, 1989) and cleared by disking. This was important as the
goal of the study was to measure any long term effects of the sludge on soil micmorganisms. Collecting soil after harvest eliminated interferences in the microbial assays that
could have been caused by the presence of growing plants (compare Coppola, 1986). Additionally, by the time the soil was collected, most of the easily available substrate that
was originally in the sludge had been utilized in the field. After sampling, the soils were
37
placed in a 4°C cooler and later sieved through a 2 mm sieve (U.S.A. Standard Testing
Sieve No. 10, A.S.T.M.E.-. 11 Specification, W.S. Tyler, Inc. Mentor, Ohio, USA). There-
after, the soils were mixed to ensure homogeneity in each sample and stored at 4°C until
analyzed. Dates of the microbial analyses are given below under the description of each
analysis. A subsample of the soil was air-cli-ied for determination of soil physical and
chemical properties, in which the gravimetric water content was determined by drying at
105°C.
Statistical Analyses
Analyses of Variance
Unless otherwise specified, all statistical analyses for treatment effect of the discrete treatment levels were performed using an Analysis of Variance (ANOVA) for a ran-
domized complete block design with a = 0.05 using the SAS GLM procedure (SAS,
1990). The means were compared using Fisher's Protected Least Significant Difference
(FPLSD) at a = 0.05. As the treatment was also a continuous variable (when expressed
as the total amount of sludge applied over the past four years), the two treatment degrees
of freedom were split to find the functional form of the treatment effect, testing for significant linear and quadratic treatment effects on the dependent variable in question.
Since the treatment levels were not equally spaced, orthogonal polynomial coefficients to
make the linear and quadratic effects independent of each other were calculated by Rick
Axelson from total amounts of sludge applied over four years with a Fortran program us-
38
ing the OPOLY subroutine in IMSL (IMSL, 1987). The assumption of equal variance between Ireatments was tested by plotting the residuals against the predicted values, and the
assumption of normally distributed residuals was evaluated using the Shapiro-Wilk statis-
tic at a = 0.05 using the NORMAL option in the SAS UNIVARIATE procedure (SAS,
1989). Where a significant curvilinear trend was found, a regression analysis was per-
formed using the SAS REG procedure (SAS, 1990) to determine the equation describing
the effect of sludge treatment on the dependent variable employing the functional form
calculated above. For this purpose, the blocks were averaged by treatment as too few
treatment levels existed to supply enough degrees of freedom for a regression analysis for
each treatment x block combination. In the case of non-normally distributed residuals or
of variables where the assumption of equal variance could not be upheld by weighting
with the reciprocal of the variance or by a transformation, the Friedman two-way analysis
of variance by ranks was used (Daniel, 1978).
Measures of Association Between Two Variables
The partial Pearson correlation coefficient (1.y2.) was calculated (using the SAS
CORR procedure (SAS, 1989)), correlating dependent variable yl with dependent variable y2, with treatment (total sludge applied over four years held constant) to control for
variation (treatment held constant) in the dependent variables due to sludge application.
Significance tests for partial rrn, testing the null hypothesis that the partial population
correlation coefficient,
was equal to zero, were also calculated using the CORR
39
procedure in SAS. Bivariate normality, necessary for using r12
as a measure of linear
association, was evaluated for the two variables as described by Johnson and Wichern
(1988) using QUIKCHEK, a C program written by the author. The Pearson partial correlation coefficient was reported only in cases where there was some evidence for a bivariate normal distribution between the dependent variables. In all cases, including those
situations where the assumption of bivariate normality was violated, a partial Kendall t b
(t11) was calculated similarly as described for
above (SAS CORR procedure,
KENDALL option with treatment held constant (SAS, 1989)) to estimate the concor-
dance between the two dependent variables yl and y2. Significance tests for
not calculated, and
were
> 0.8 was used as a limit for significant concordance. The Ken-
dall t statistic is a nonparametric measure of association (concordance), and was chosen
as it is more reliable for small sample sizes than the Spearman Rank Correlation Coeffi-
cient (Gibbons, 1985). Unlike r, t is not a measure of linear association, and unlike the
case for r, t =0 does not imply independence between the two dependent variables, although independence implies no association (t =0) (Gibbons, 1985). Discussions involving measurements of association between two dependent variables will use the term
correlated to refer to measurements involving
involving
and concordant for measurements
40
In accordance with the findings of Dutzler-Franz (1977a), measurements of association between dehydrogenase activity and soil organic matter content (total organic
carbon and total nitrogen) were made after transforming the enzyme activity data using
the natural logarithm, in d1iflon to the untransformed values.
Sewage Sludge Analyses
All sewage sludge analyses were performed by contracted laboratories. Total C,
P, and Kjeldahl N were determined on air-dried sludge after cenirifugation and filtration,
as were the metals Zn, Cu, Pb, Ni, Cr, Cd, and Ag. NH4-N, NO3-N, and total inorganic N
were analyzed in the supernatant and converted to an air-dried sludge basis.
Soil Physical and Chemical Analyses
All analyses of physical and chemical properties other than pH and gravimethc
moisture content were performed by the Soil, Water, and Plant Testing Laboratory (SWPTL) at the University of Arizona. All tests were performed using air-dried soil, except
where otherwise noted, with results converted to an oven-dry basis, based on the moisture content of the air-dried soil.
Soil texture was determined using the hydrometer method of Gee and Bauder
(1986), as described in Artiola (1989). Soil water holding capacity (WHC) was determined using the pressure plate method of Klute (1986) as described in Artiola (1989).
Gravimetric moisture content was periodically determined using moist soil to account for
any drying which may have taking place during storage of the soil at 4°C. For this analy-
41
S1S, Ca. 20 g of moist soil were dried overnight at 105°C, cooled in a desiccator, and
weighed. Electrical conductivity was determined from a saturated paste extract using dis-
tilled water and 50-100 g soil vacuum filtered through Whatman 42 filter paper, as described in Artiola (1989). The results are reported as dS m1.
Soil pH was determined in a 1:2 suspension of 10 g air-dried soil and 20 ml of
0.01 M CaCJ2 (Schofield and Taylor, 1955), stirring periodically for 30 mm and allowing
30 mm for the soil to settle before taking the readings with a Broadley-James Model
9405 combination pH electrode (Cole-Parmer, Chicago, Illinois, USA). For the determination of soluble cations, a 1:2 extract was made with 30 g of soil and 60 ml of distilled
water and vacuum filtered through Whatman 42 paper. Soluble Na, K, and Ca were then
determined by inductively coupled plasma emission spectrophotometry (ICP, Leeman
Labs, Lowell, Massachussetts, USA) as described in Artiola (1989), and results are reported as mg Y' of extract. Exchangeable cations were determined for Na, K, and Ca during the course of the CEC determination using the sodium acetate method followed by
analysis by ICP as described in Artiola (1989). The results are reported as mmol(+) kg'
dry soil.
Both total organic carbon and total nitrogen were determined by high temperature
analysis on ball-milled soil (<100 jim 0 particle size) after treatment with 100 g 11
H3PO4 and subsequent oven drying at 60°C, using a Carlo Erba NA 1500 CNS Ma-
lyser (Carlo Erba Strumentazione, Milan, Italy) as described in Artiola (1989). Available
PO4-P was estimated using the Olsen-bicarbonate extraction procedure (0.5 M NaHCO3
42
adjusted to pH 8.5), with the color developed using a molybdate reagent and mixed ascorbate reagent, and analyzed colorimetrically as described in Artiola (1989).
Total, acid-digestible heavy metals were determined on one gram samples of bail-
milled soil (<100 p.m 0 particle size) subjected to wet oxidation using concentrated
HNO3 and 3% 11202 according to Method 3050 (USEPA, 1986), and analyzed by ICP us-
ing Method 6010 (USEPA, 1986) for Zn, Cu, Pb, Ni, Cr, Cd, and Ag as described in Artiola (1989). The amount of heavy metals available to plant and microorganisms was
estimated using 10 gram samples of soil which were extracted with diethylenetriaminepentaacetic acid-triethanolamine (DTPA-TEA) (Lindsey and Norvell, 1978) as described
in Artiola (1989). The extracts were analyzed for Zn, Cu, Pb, Ni, Cr, Cd, and Ag using
ICP as described in Artiola (1989). "DTPA-TEA-extractabje metals" will be referred to
as "DTPA-extractable metals" in the text for simplicity.
Soil Microbial Analyses
Viable Heterotrophic Plate Counts for Bacteria
The analysis was performed in January 1990. A sample of field-moist (ca. 100 g
kg' moisture) soil (10 g of oven dry soil) was added to a 50 ml amber plastic vial and incubated covered with Paraflim® (American Can Company, Greenwich, Connecticut,
USA) at 22°C for three hours. Media were prepared as follows: Each liter of peptone-
yeast agar (PY) contained 5.0 g peptone (Difco, Detroit, Michigan, USA), 3.0 yeast extract (Difco), and 15 g agar (Difco). At this point the pH of the agar was found to be 6.8.
.43
After autoclaving 20 mm at 145 kPa at 121°C and cooled to Ca. 45-50°C, the agar was
amended with an aqueous solution of CaC12 (Mallinckrodt, Paris, Kentucky, USA) to
bring the final concentration of added CaCl2 to 10 mM. The soil extract agar (SE) was
made from a soil extract of Pima clay loam, which was collected from one of the no
sludge plots at the study site the previous year. To make the extract, a 1000 g sample of
the Pima clay loam was mixed with 1000 ml tap water and stirred briefly prior to autoclaving 20 mm at 145 kPa and 121°C. To the suspension, 1 g CaCO3 (Mallinckrodt, St.
Louis, Missouri, USA) was added to the mixture, and the suspension was poured into a
2000 ml graduated cylinder, where it was allowed to settle. The supernatant was then decanted, filtered through Whatman 2 paper, and centrifuged at 7000 rpm for 10 mm. The
total yield of soil extract was 400 ml. The SE agar was made by combining 1.0 g glucose
(Sigma, St. Louis, Missouri, USA), 0.5 g K2HPO4 (Mallinckrodt, Paris, Kentucky, USA),
0.1 g KNO3 (Matheson, Coleman and Bell, Norwood, Ohio, USA), 100 ml soil extract,
15 g agar (Difco), and 900 ml distilled water. The pH was measured prior to adding the
agar and found to be 8.4. The agar was autoclaved 20 mm at 145 kPa and 121 °C.
The incubated soil was added to a sterile 95 ml dilution blank containing distilled
water. The suspension was shaken on a reciprocating shaker (Eberbach Corporation, Ann
Arbor, Michigan, USA) set on high for 10 mm. For each 95 ml dilution bottle, a dilution
series was made from Ca. 10.2 to 10 g ml' dry soil by pipetting 1.0 ml of the shaken sus-
pension to each successive tube and vortexing. Of each of the 10 to 10 dilutions,
44
0.10 ml of suspension was plated Out on PY and SE agar in 10 x 100 mm 0 petri dishes,
and incubated at room temperature (23°C) for seven days prior to counting. Counts were
performed without magnification, and actinomycetes were not counted with the bacteria.
The results of the study were reported as colony forming units (CFU) per gram of ovendried soil.
Acridine Orange Direct Counts for Bacteria
The procedure, performed March 1991 to April 1991, was derived from Schmidt
and Paul (1982), with modifications, as noted. All glassware was acid rinsed (0.1 M HCI)
followed by two rinses with water Milli-Q (Millipore, Bedford, Massachussetts, USA) reagent-grade water that had been filtered through a 0.22 xm Mihipak filter (Millipore,
Bedford, Massachussetts, USA). From a review of the literature, it is apparent that a
given extracting solution and/or procedure will not work on all soils, and, indeed, empirical trials using the Pima clay loam showed that an extractant and accompanying extracting procedure needed to be devised that would effectively separate the bacteria from the
soil particles and break up the soil aggregates to optimize the counts.
The extracting solution was made in batches large enough for 23 150 ml dilution
bottles and for 10 five ml fillings of staining tubes. The solution was 1 g 1.1 peptone
(Difco, Detroit, Michigan, USA) (Schmidt and Paul, 1982), 2 g li sodium hexametaphosphate (calgon) (Pfalz and Bauer, Waterbury, Connecticut, USA) (Troildenier, 1973), and
6 J.LM in Zwittergent 3-12 (Calbiochem, San Diego, California, USA) (Camper etal.,
45
1985), with the volume made up using distilled water. The pH was generally around 5.9
before being adjusted to 9.0 using 150 g 1' NaOH (MCB Manufacturing Chemists, Inc.,
Cincinnati, Ohio, USA). The solution was filtered through a 0.2 jim polycarbonate filter
(Nucleopore, Pleasanton, California, USA). French bottles (ca. 150 ml, Corning, No.
1367, Corning, New York, USA) were ifiled with Ca. 104 ml of the extractant, with an ex-
tra bottle reserved for final volume measurement after autoclaving. The remaining solution was saved for use in the staining tubes and autoclaved in its own bottle. The bottles
were capped and autoclaved at 145 kPa, 121°C for 15 mm. The final volume was about
100 ml at a pH of 8.7 after one week of sitting tightly capped at room temperature
To prepare the acridine orange staining solution, 1 g of acridine orange (Fluka,
Ronkonkoma, New York, USA) was added to 100 ml of the above described, filtered reagent-grade water, and stirred until saturated (ca. 15 mm). The solution was filtered
through Whatman 2 filter paper to remove the heavy sediment and refiltered through a
0.2 jim polycarbonate filter (Nucleopore), changing the filter every 15 ml. The solution
was stored under refrigeration (Schmidt and Paul, 1982) and replaced periodically.
To one of the French bottles containing the extractant, 10.00 g moist soil was
added, and the bottle was capped, shaken 30 s by hand, and sonicated for 3 minutes in an
ultrasound cleaner (L&R Transistor/Ultrasonic T-7, L&R Manufacturing Co., Kearny,
New Jersey, USA). After sonication, the bottle was shaken 30 s by hand, and 1.00 ml of
the suspension was immediately removed and pipetted into a fresh bottle of the extractant, taking care to immerse the pipette to the same depth for each soil (ca. 5 cm below
46
the surface of the suspension). The bottle was capped and shaken 30 s by hand and allowed to stand 10 mm. After settling, 1.00 ml of the suspension was removed by immersing a 1 ml serological pipette about 1 cm deep in the suspension and added to a sterile 16
x 125 mm test tube (staining tube) containing 3.75 ml of the extracting solution. Then,
0.25 ml of the acridine orange staining solution was added to the staining tube, bringing
the total volume to 5.00 ml with a final acridine orange concentration of Ca. 0.5 g 1*
The suspension was vortexed vigorously followed by staining for 2 mm, vortexed
again, and added to a glass filter assembly with a sintered glass filter support fitted with a
0.2 p.m factory-stained black polycarbonate filter (Nucleopore or Porectics Corporation,
Livermore, California, USA) supported by a 0.6 p.m polycarbonate filter (Nucleopore).
The suspension was filtered at 15 to 84 kPa vacuum, with the vacuum adjusted during
the filtration to keep a slow-paced drip. The filter was removed and let dry briefly (Zim-
mermann etal., 1978) followed by mounting in immersion oil (Carl Zeiss, Oberkochen,
FRG, 518C DIN 58 884, fle=1518
D=1515 at 23°C) with one drop under the filter, one
drop on the filter, a coverslip (VWR Scientific, Inc., San Francisco, California, USA, 24
x 30 mm No. 1), and one drop of oil on top of the coverslip. The bacteria were counted
under a Zeiss Standard microscope using the "standard FJTC filter set," also recom-
mended for acridine orange assays (exciter filter BP 450-490 nm, chromatic beam splitter FT 510, barrier filter LP 520 (catalog no. 48 77 09)). The microscope was fitted with a
200 W halogen lamp and a Zeiss IV FL epifluorescence condenser, with the illuminator
47
set at 10 V. The objective used for counting was a Planapochromat 63/1.4 Oel 160/- with
8x oculars and was calibrated using a stage micrometer to have a field of view of 0.278
mm 0 Troildenier (1972a) recommends that the fields of view be as large as possible.
The final total magnification was 504x.
As was pointed out by Troildenier (1972b), a reduction of the number of replicate
samples leads more quickly to an increase in the standard error than a reduction in the
number of fields of view examined per replicate. Three replicate extractions were made
for each soil, counting six fields of view per replicate (after Trolidenier, 1972b). The
fields of view were chosen positioning the stage by looking from the outside (T. Hubert,
personal communication). The results were reported as bacteria per gram of oven-dried
soil.
Viable Heterotrophic Plate Counts for Actinomycetes
In this analysis, performed in July 1990, ten grams of field-moist soil (ca. 90 g
kg1 moisture) were added to 50 ml amber plastic vials, covered with plastic wrap, and
ventilated with two holes in the top using a dissecting probe. The soil samples were incubated at 23°C for seven days to control the growth of the more moisture-sensitive bacteria (Tsao er al., 1960).
The ingredients for the glycerol-casein medium used in this assay were taken
from Küster and Williams (1964), and the procedure for making it was derived from Williams and Davies (1965). Each liter of medium was made from 10.0 g glycerol (cone-
48
sponds to 8.5 ml) (Sigma), 0.3 g vitamin-free casein (Sigma), 2.0 g KNO3 (Matheson,
Coleman and Bell, Norwood, Ohio, USA), 2.0 g NaC1 (U.S.P., Fisher Scientific, Fair
Lawn, New Jersey, USA), 2.0 g K2HPO3 (Mallinckrodt), 0.05 g MgSO47H2O (Mallinckrodt) 0.02 g CaCO3 (Mallinckrodt), 0.01 g FeSO47H2O (Sigma) 18 g agar (Sigma, Plant
Tissue grade), 50 mg cycloheximide (50 j.tg mi-i final concentration) (Sigma, this antifungal agent is heat stable (Williams and Wellington, 1982)), and 945 ml distilled water.
Very little of the casein actually dissolved. The fmal pH before heating and adding agar
was 8.5. The medium was autoclaved is mm at 450 kPa at 12 1°C. After cooling to Ca.
50°C, nystatin (antifungal agent, 50 ig ml1 final concentration) (Sigma), was added as a
suspension while stirring the agar, which had been cooled to Ca. 50°C. The antibacterial
agents polymyxin B sulfate (5 .tg m11 final concentration) and sodium penicillin G (1 j.tg
mr1 final concentration) were not added as earlier trials had demonstrated that the actinomycete populations in these soils were sensitive to these antibiotics (Williams and
Davies, 1965) and growth of bacteria on the medium was not high enough to warrant any
procedures for inhibiting their growth.
After incubation of the soils, the contents of the vials were added to 95 ml of distilled water, shaken 30 mm on a reciprocating shaker (Williams and Davies, 1965), and a
serial dilution was made as described above for the bacteria, but only up to 10 g soil m1
Five spread plates were made for each dilution using 0.10 ml of suspension from the
three most dilute tubes on 10 mm x 100 mm 0 petri dishes as trials had shown that in the
49
case of pour plates, as described in Williams and Davies (1965), bacteria were too difficult to distinguish from actinomycetes. The plates were incubated at 22°C for 17 days
prior to counting. The results were calculated as CFU per gram of dry soil.
Viable Heterotrophic Plate Counts for Fungi
Fungal Counts by Serial Dilution
This analysis was performed in July 1990. Ten grams of moist soil were added to
50 ml amber plastic vials, covered with plastic wrap, ventilated with two holes, and incubated for one day at 22°C. The medium, rose bengal-streptomycin agar (Martin, 1950),
was made by combining the following in 1000 ml of tap water: 10.0 g glucose (Sigma),
5.0 g peptone (Difco), 1.0 g K2HPO4 (Mallinckrodt), 0.5 g MgSO42H2O (Mallinckrodt),
0.033 g rose bengal (Baker, Philhipsburg, New Jersey, USA), 15.0 g agar (Sigma, Plant
Tissue Culture grade). At this point the pH of the medium was 7.7. The agar was autoclaved 15 mm at 145 kPa and 121°C. After cooling to Ca. 50°C, 10.0 ml of sterile strepto-
mycin sulfate (Sigma) stock solution (7.0mg ml) was mixed into the agar for a final
concentration of 70 p.g m1' (Rodriguez-Kabana, 1967).
To plate, a serial dilution for each of the soils was made as described above for
the actinomycetes except that the suspension was shaken for 30 mm and the dilution was
performed using 9 ml distilled water blanks up to 10 g dry soil mY1. For each soil,
0.1 ml of suspension was applied to each of five replicate 10 mm x 100 mm 0 petri
dishes, and pour plates were made with the cooled agar, The plates were incubated for six
50
days at 22°C before counting. The counts were converted to CFU per grain of oven-dried
soil.
Fun gal Counts by the Drop Plate Method
This method was modified from that described by Rodriguez-Kabana (1967), and
the analysis was performed in July 1990. The soils for the assay were prepared and incu-
bated as described above for the dilution plates. Before plating, a 10 g 1' carboxymethylcellulose (CMC) solution was made by adding 20.0 g CMC (Sigma) to 2000 ml of
distilled water at 100°C with constant stirring. After dissolution, 325 ml of the CMC solution was added to 500 ml erlenmeyer flasks fitted with 7.6 cm magnetic stirbars. The
flasks were covered with foil, and autoclaved at 145 kPa for 15 mm at 121°C. One flask
was prepared for each of the soils to be plated. To plate, the incubated soil was added to a
sterile flask and stirred 5 mm on high for 5 mm. While stirring, a sterile 15.2 cm pasteur
pipette fitted with a latex bulb was filled with suspension and one drop of suspension was
pipetted into each of five 15 mm x 150 mm 0 sterile petri dishes. Immediately thereafter,
four tared, aluminum weighing dishes were each filled with 10 drops of suspension, moving quickly from one dish to the next, for the purposes of estimating the amount of soil
added to each dish. Trial experimentation showed that the weight of CMC added in each
drop was negligible in comparison to the amount of soil (about 0.0003 g). The plates
were incubated six days at 22°C before counting.
51
Dehydrogenase Assay (DHG)
This procedure, perfonned May 1990 to July 1990, was modified from Klein er
al. (1971). The following solutions were prepared as nutrient amendments (electron donors) as preliminary trials showed no measurable dehydrogenase response in unamended
soil: glucose. 10 g Y1 glucose (Sigma), Peptone yeast broth: 5 g 11 peptone (Difco) and
3 g 1.1 yeast extract (Difco) broth, glycerol-casein broth: 0.85 ml glycerol (ca. 1.0 g,
Sigma) in 100 ml distilled water with 0.03 g vitamin-free casein (Sigma). Only the clear
supernatant was used, as very little of the casein dissolved. All of the above solutions
were autoclaved 10 mm at 145 kPa and 121°C. A 30 g 11 solution of 2,3,5triphenyltetrazolium chloride (TFC, Difco) solution was prepared and filtered through a Millipore type
HA 0.45 m filter. A triphenyl formazan (TPF) standard curve was prepared from a stock
solution of TPF (Sigma, fmal concentration 125 g m11 TPF) at 0.00, 2.50, 10.00. 12.50,
and 15.0 pg ml1 in 100% methanol (MeOH).
One gram of moist soil (Ca. 90 g kg1 moisture) was weighed into each of three 16
x 125 mm screw cap test tubes. Under aseptic conditions, the following solutions were
added to each tube: 200 .il of 30 g l' UC and 500 .tl of the appropriate electron donor
solution (or 500 j.tl sterile distilled water for the blank. After adding the solutions, the
tubes were tightly capped and vortexed on low to minimize the amount of soil on the
sides of the tube yet thoroughly mix the soil and solutions. The amount of liquid used
saturated the soil, leaving a film of liquid on the soil surface. The tubes were incubated at
52
30°C for four days for the glucose and peptone-yeast amended soils, and eight days for
the glycerol-casein amended soils.
To extract TPF from the soil, 10 ml methanol was added to each of the tubes at
the end of the required incubation time, and the tubes shaken on a reciprocating shaker
for 30 s. Then, each tube was individually shaken briefly by hand and its contents poured
into a polypropylene funnel plugged with absorbent cotton (Absorbent Cotton Co.,
U.S.P. grade), which drained into a 25 ml volumetric flask. The samples in the funnels
were repeatedly extracted with 3.5 to 10 ml quantities of methanol until the cotton plug
was colorless. The volumetric flasks were diluted to volume with methanol and covered
with Paraflm® (American Can Company). The absorbance of the solutions was measured
at 485 nm using a Bausch and Lomb Spectronic 100 with the vacuum for the Micro Flow-
Thru cell set at 20 to 27 kPa using 100% methanol as the blank. In the case of the peptone-yeast amendment, it was necessary to dilute the filtrate by a factor of 0.2 with 100%
methanol. The absorbance values were converted to p.g TPF per grain of oven-dried soil.
The study was set up to be analyzed using a Multivariate Analysis of Variance
(MANOVA) with the measured dehydrogenase response to the three electron donors (glucose, peptone-yeast, and glycerol-casein) being the three dependent variables, and sludge
treatment the independent variable. Multivariate normality was tested by performing a
univariate ANOVA on a linear combination of the independent variables followed by a
Q-Q test on the residuals (Johnson and Wichern, 1988) with a = 0.05. Equality of the
within-group covariance matrices was tested for using the likelihood ratio test in the SAS
53
DISCRIM procedure (SAS, 1990). The Bartlett test of sphericity was performed using
SPSS-X release 3.1 (SPSS, 1988) and evaluating the significance at a = 0.05 to evaluate
whether a multivariate analysis of variance was appropriate. A significant result from this
test would mean that some or all of the dependent variables were correlated; otherwise
the variables could be analyzed using three independent univariate ANOVAs for each of
the dependent variables.
To determine the presence of unextracted TPF in microbial cells (after Glathe and
Thalmann, 1970a), one of the soil samples from the dehydrogenase assay from a no
sludge treated plot was air-dried after extraction with MeOH during the TPF analysis.
From this sample, 100 mg was suspended in 10 ml of distilled water by vortexing. One inoculating loop-full of the suspension was heat-fixed to a glass microscope slide and exam-
ined under oil on a Nikon Optiphot (Nippon Kogaku K.K., Japan) microscope using the
Ph 4 Plan 100 DL 1.25 oil 160/0.17 objective with a 1 Ox ocular. Photographs were taken
on a Nikon FX 35A 35 mm camera on a Nikon Microflex HFXll photomicrographic attachment using Kodak Kodacolor "Gold 100" color print film (ISO 100/2l°D, EastmanKodak, Rochester, New York, USA). The objects were photographed with phase contrast
settings 4 and 0.
54
Carbon Dioxide Evolution Analysis
This procedure, performed from May 1991 to June 1991, was modified from
Bartha and Pramer (1965) and Stotzky (1965). The incubation jars for these studies were
946 ml regular-mouthed glass Mason jars fitted with two-piece screw lids (Kerr Glass
Manufacturing Corporation, Los Angeles, California, USA). The lids were fitted with a
hook from which to suspend the CO2 traps, which were plastic vials (60 x 27 mm with
15 mm internal diameter mouths) containing a standardized KOH solution.
In the water amendment study, enough distilled water to bring 100 g of moist soil,
originially at a moisture content of Ca. 90 g kg, to 150 g kg' moisture content (corre-
sponds to -0.1 MPa water potentialNeilson and Pepper, 1990) was added to the jars
with a pipette. On top of this, 100 g of moist soil was added and the jars sealed with the
KOH traps, containing 20.00 ml of standardized 0.1 M KOH, suspended from the lids;
moisture amendment thus occurred by capillary action, avoiding puddling the soil.
In the alfalfa amendment study, enough distilled water to bring 100 g of moist
soil, originally at Ca. 90 g kg' moisture content, to 150 g kg' moisture content was added
to the jars along with sufficient water to bring the ground alfalfa amendment (Medicago
sativa L. ground in a Wiley mill with a 60 mesh screen), amended 2.5 g kg' on a dry
weight basis to the soil, to its original fresh moisture content (773 g kg'). The alfalfa was
determined to have 48.3 g kg1 total nitrogen and 451 g kg' total organic carbon as deter-
mined using an NC+S analyzer (Carlo-Erba Strumentazione, Milan, Italy) and
55
3.24 g kg1 sulfur (Artiola and Mi, 1990). The moist soil (100 g) and alfalfa mixture was
added to the jars on top of the water, and the jars were sealed with a KOH trap containing
20.00 ml of standardized 1 M KOH. However, due to the lime needed for the amendments, the jars in this case were put on ice after adding the soil, and all of the jars were
stored in a cooler (4°C) overnight prior to incubation.
One jar with no soil, water, or alfalfa amendment added was set up in a like manner for each amendment study to be used as a blank. All of the soils in both amendment
studies were incubated for seven days at 28°C. After each day of incubation, the jars
were aerated 10 mm for the water amendment study and
15
mm for the alfalfa amend-
ment study, and fresh CO2 traps were added. Carbon dioxide evolution was measured iitrimetrically by measuring the amount of neutralized KOH by adding 1 ml of 1.0 M
BaC12 (Sigma) and a 5.00 ml aliquot of the alkali sample to a flask containing distilled
water and back-titrating with standardized HC1 (0.01 M for the water amendment study
and 0.1 M for the alfalfa amendment study) using phenolphthalein as the indicator. The results were reported as .tg CO2 evolved per gram of oven dried soil by subtracting each
day's blank value from the sample value.
The studies were set up as a repeated measure design at equal time intervals to
simplify the statistical analyses. The statistical analysis consisted of three phases: 1) The
functional form of the time effect on CO2 evolution was found by using the REG procedure in SAS (SAS, 1990) to calculate the intercept and linear, quadratic, cubic, and quar-
56
tic parameter estimates using CO2 evolved as a function of linear, quadratic, cubic, and
quartic orthogonal coefficients taken from Lentner and Bishop (1986) in the MODEL
statement. The orthogonal coefficients were necessary as time is autocorrelated. It was decided in advance to use no higher than the cubic term. 2) The highest order significant pa-
rameter estimates from 1) (a = 0.05) were used in analyses of variance with orthogonal
linear contrasts as described earlier to fmd the functional form of the time x treatment effect. 3) A regression was performed on the orthogonal coefficients based on sludge
amendment level described earlier to find the functional form of the treatment effect on
CO2 production. 4) The equation that described the effect of time and treatment on CO2
evolution was estimated by performing a regression using the functional forms derived in
1), 2), and 3) in the model statement of the GLM procedure using the SOLUTIONS option in the MODEL statement (SAS, 1990). Given the randomized complete block design, block and treatment were used in the CLASSES statement and block and treatment
were used as discrete variables in the MODEL statement. However, averaging over
blocks resulted in little reduction in fit of thç model (small block effect), and, thus, only
those equations are reported as they are easier to interpret. Due to the complex time x
treatment interaction of the CO2 evolution data, no measures of association were calculated with any other dependent variables.
57
RESULTS
Effects of Sewage Sludge on Soil Physical and
Chemical Properties and Heavy Metal Contents
After four years of application of sewage sludge to the fields at Marana, few of
the more common soil properties appear to have been influenced (Table 2). Both soluble
Na and EC showed a significant linear increase with sludge application (Figure 1, A and
B). Soluble Na was not concordant with EC (t12 = 0.427) (a more complete listing of
Table 2: Means of the values for soil physical and chemical properties by sludge
treatment. pH is in pH units; EC in dS m1; CEC and exchangeable cations in
mmol(+) kg' dry soil, soluble cations in mg 1.1 extract; WHC, total organic C, and total N in g kg' dry soil; and PO4.P in mg kg' dry soil. Means superscripted with the
same letter are not significantly different at a = 0.05. Italicized values are medians
as the data was not normally distributed. NS=Not Significant, NA=Not Applicable
(Friedman analysis of variance by ranks was used).
Variable
WHC
No Sludge
229
Low Sludge
242
High Sudge
252
FPLSD
pH
7.70
7.70
7.69
NS
EC
CEC
0.32
0.37
0.39
NS
249
262
270
NS
Exchangeable Na
2.87
2.97
2.96
NS
Exchangeable K
13.5
13.4
13.7
NS
Exchangeable Ca
240
239
227
NS
Soluble Na
23.9
27.1
26.9
NS
Soluble K
18.2
19.6
20.7
NS
Soluble Ca
36.7
15.oa
40.0
271b
NS
Available PO4-P
32.1
16.9a
6.36
Total Organic C
4.50
5.03
5.94
NS
Total N
0.75 1
0.747
0.820
NS
NA
58
0.50 -
30.0
0.40 -
25.0-i
20.0-
0.30
Cl)
o
15.0y=0328-I01ti276x
0.20
y=24.B+0.102x
.odd =0 .260
r
LU
S
I
w
S
.
.
S
2 10.0
0.10
5.0-i
0
Cl)
0.00
0
I
I
I
I
I
5
10
15
20
25
0
I
I
5
10
A
15
20
25
B
40.0 y=16.9-0i59x+0.0406x2
-2r
-
.
c)
30.0
7.
5
._________v.
20.0-
S
io.o-
-
0.0
0
5
10
I
I
15
20
25
C
Figure 1: Significant trends in physical and chemical properties of the soils as a function of total sludge application over a four-year period (the abscissa is in units of Mg
dry sludge ha-'). Numbers next to symbols denote the number of overlapping data
points.
measures of association are given in the appendix), indicating that the increase in EC may
be attributed to an increase in soluble Ca, possibly due to organic acids in the soil organic
matter in the sludge amended soil, although no cause and effect relationship was established. Although the sludge was high in Ca (ranged from 30 to 48 g kg1 dry sludge over
the four year period), the soils are known to be high in native Ca (26 g kg' soil as CaCO3
equivalentsPost et aL, 1978). In fact the same relation of EC to treatment was observable in the seedbed soon after sludge application (possibly due to salts added in the sludge
59
Table 3: Means of the soil acid digestible and DTPA-extractable heavy metal contents. All values are in mg kg' dry soil. Means superscripted with the same letter are
not significantly different at a = 0.05.
Variable
Zn (acid)
No Sludge
95.5
Low Sludge
93.5
High Sudge
90.7
FPLSD
NS
Cu (acid)
58.0
57.0
56.9
NS
Pb (acid)
21.8
21.7
21.2
NS
Ni (acid)
22.0
21.2
19.8
NS
& (acid)
18.7
19.3
17.4
NS
Cd (acid)
2.44
2.44
NS
Zn (DTPA)
1.06'
Cu (DTPA)
0.799'
2.98'
2.36
194b
3.11'
394b
Pb (DTPA)
2.39
2.54
2.64
NS
Ni (DTPA)
0.103
0.162
0.264
0.0619
Cr(DTPA)
0.406
0.518
0.525
NS
0.4 15
0.28 1
not being leached significantly by irrigation), though at a slightly higher level (0.67, 0.74,
and 0,92 dS m for the no, low, and high sludge amended soils (Ottman er al., 1989)).
Sludge application significantly affected NaHCO3-extractable PO4-P, due possibly to an in-
teraction between localized reductions in pH and the large amounts of PO4-P added to the
soil in the sludge (Figure 1) (O'Connor etal., 1986). The soil water holding capacity, total
organic C, and total N were unaffected by sludge application (Table 2).
Heavy metals were believed to be the factor that would limit the length of time that
land disposal could continue on the fields in southern Arizona. Yet, measurements of total
metal concentrations in the soils showed no sign of increasing. In Table 3 and Figure 2A
we see that the only total metal to have changed in concentration is Ni, which showed a
60
D
20)
25.0 -
3.00 -
20.0
2.50 - rWØdlO.8O1
---------_
y=O.799+O.0252x+O.000909x2
:
'2.00-
15.0 -
10.0
2150
r1=O512
1.00 -
.j 5o
050-'
z
0.0
S
y=22.O-O.900x
i
5
10
15
I
I
20
25
0.00
5
10
A
I
I
15
20
25
B
5.00 -
3.00
4.00
2.50
$
20)
2.00 -
3.00
E 2.00
E
1.00
.c
0.00
5
-
hbooT
VmOdd=O896
0
.=O.436
modt
1.50
y=2.98+000345x+O.00149x2
y=2.4l +O.00982x
I
I
10
15
0.50 0.00
I
25
20
5
I
I
I
10
15
20
25
D
C
0.300
2.
y=O.105+O.00660x
0.250
.2
0 0.200
2
0.150
4
¶
.2
z_b 0.050
0.000
0
10
15
20
25
E
Figure 2: Significant trends in soil heavy metal contents (total and DTPA-extractable) as a function of total sludge application over a four-year period (the abscissa is
in units of Mg dry sludge ha'). Numbers next to symbols denote the number of overlapping data points.
61
Table 4: Theoretical added metal concentration in the field soil due to sludge-borne
heavy metals as calculated from sludge metal contents and amounts of sludge applied to the soil on a liquid volume basis.
Metal
Zn
Theoretical Added Metal Concentration
(mg kg1 dry soil)
Low Sludge
High Sludge
4.9x1()3
L4x102
Cu
3.3 x i0
9.2 x i0
Pb
5.5x10
1.8x103
Ni
1.7 x i0
4.9 x i0
Cr
2.6x1O
7.8x iO4
Cd
4.6x105
Ag
l.8xl0
1.3x104
4.9x104
significant decrease. As the sewage contained measurable amounts of Ni (see Table 1,
page 35), the significant increase seen was probably due to field variability. In fact, if one
calculates the theoretical change in soil heavy metal concentration due to sludge application, native metal content is apparently being measured in the total metal assays of the
soil, as the sludge-applied metal values are very low (Table 4).
In contrast to the total metal contents of the soil, several DTPA-extractable metals
(Zn, Cu, Pb, and Ni), exhibited significant increases with sludge application. Some fraction of the organic matter in the high sludge treatment may have been chelating these
ions, enhancing their availability, although there is no apparent relationship between soil
pH, total organic C, and total N and DTPA-extractable metals (Appendix Table 8). Total
metal concentrations were not associated with DTPA-extractable metal contents (Appen-
62
dix Table 7), indicating that some other factor, such as the nature of the organic matter,
which may be a function of sludge application but was not measured, may have affected
the availability of the metals. DTPA-extractable metal measurements are less prone to
field variability than the total metal determinations, as a 10 g soil sample was used for exiraction as opposed to the 1 g sample for the acid digest (J. Artiola, personal communication; Artiola, 1989). Total Ag and DTPA-extractable Cd and Ag were below detection
limits.
Effects of Sewage Sludge Application on
Cotton Plant Growth Parameters
Although sewage sludge application significantly reduced plant stand as measured in June and November, there was no significant effect of sludge application on lint
yield nor plant height (Table 5). Reduction in plant stand (Figure 3) could not be ex-
Table 5: Mean cotton plant growth parameter values. Yields are in Mg ha', plant
height in cm, and plant stand in plants m' of row. Italicized values are medians as
the data was not normally distributed. Means superscripted with the same letter are
not significantly different at a = 0.05. NS=Not Significant, NA=Not Applicable
(Friedman analysis of variance by ranks was used).
Variable
Yield 1
No Sludge
175
180
166
FPLSD
NS
Yield 2
8.45
10.6
10.6
NA
Total Yield
182
187
173
NA
PlantHeight
113
125
NS
Plant Stand (June)
15.3k
125
15.62
1291)
1.45
Plant Stand (November)
13.12
105b
Low Sludge High Sludge
98O
0.7 16
63
20.0
15.0
-:
4
E
I
10.0
aD 10.0 : y=15.3+O.1O&x-O.00851x2
C
-2r,k,=O.7l2
(.1)
C
5.0
y=l2.4-O.123x
r,=O.689
5.0
0
0
November
C)
0.0
I
0
5
10
I
I
I
15
20
25
0.0
0
I
I
5
10
15
20
25
A
B
Figure 3: Significant trends in stand of cotton asa function of total sludge appplication over a four-year period (the abscissa is in units of Mg dry sludge ha'). Numbers
next to symbols denote the number of overlapping data points.
plained by any relation with heavy metal data (see Appendix Tables 5 and 6), and there
was no significant concordance between EC and plant stand as measured in June
= -0.472) and November
= -0.330) (see Appendix Table 9). Ottman et al.
(1989) could not explain this relationship, as the EC values were too low to reduce germination of the plants. Additionally, they found that tillage alleviated problems of clod formation in the high sludge soils, with no significant difference in seedbed particle size
distribution due to treatment, leaving the causes of the reduced stand unaccounted for. In
fact, despite a lower stand in the high sludge treated plots, yields for the plots were above
the state average for that year of 120 Mg ha' (Dr. M. Ottman, personal communication).
64
Effects of Sewage Sludge on Soil Microbial Activity
Enumeration Assays
None of the investigated parameters showed any decrease in the level of soil microbial activity with increased sludge application. On the contrary, the only detectable differences between treatments showed an elevation of activity (Table 6). Viable
heterotrophic plate counts for bacteria on the different media, peptone yeast extract and
soil extract agar, although different in magnitude, were significantly correlated
(ryly2tll = 0.897) with each other, despite the vastly different pH values of the respective
media (6.8 for peptone yeast and 8.4 for the soil extract agar). Acridine orange direct
counts for bacteria, approximately an order of magnitude larger than the plate Counts on
peptone yeast, were, in contrast, not concordant with plate counts on either media
(t121 = 0.079 and
= -0.008 with bacteria on peptone yeast and soil extract agar,
respectively). This was not surprising as a given medium will represent only a small fraction of the total species in a soil (Atlas and Bartha, 1987).
Heterotrophic plate counts for actinomycetes were about an order of magnitude
lower than the bacterial plate counts, and those for fungi still an order of magnitude lower
than the actinomycete values. The two methods for plating fungi were significantly corre-
lated with each other (r1
= 0.770). On the one hand, one would have expected the se-
rial dilution technique to have resulted in an overemphasis of heavy sporulating fungi,
such as Aspergillus sp. and Penicillium sp.; observing the plates inoculated by either
65
Table 6: Means of the microbial enumeration assays and DHG results. Means superscripted with the same letter are not significantly different at a = 0.05. Viable heterotrophic plate counts are in CFU g' dry soil, bacteria by direct count in cells g'
dry soil, and DHG values in ig TPF g' dry soiL NS=Not Significant.
Variable
No Sludge
Low Sludge
High Sludge
FPLSD
Bacteria (PY)
1.76x iø
1.62x i07
1.90x i07
NS
Bacteria (SEA)
9.23 x 10
7.88 x 10
1.05 x 107
NS
Bacteria (direct count)
1.15 x i08
1.52 x 108
1.30x 108
NS
Actinomycetes
1.86x 106
1.76 x 106
1.77 x 106
NS
Fungi (serial dilution)
3.37x i05
3.13 x iø
3.49x 105
NS
Fungi (drop plate)
1. 12 x i05
1.32x
1.47 x iø
NS
DHG (glucose)
8.50
6.74
7.44
NS
DHG (PY)
31.1
5.82a
29.0
6.00a
32.1
NS
743b
1.00
DHG (GC)
method, those same heavy-sporulating genera dominated the plates. However, as was observed by Rodriguez-Kabana (1967) the coefficient of variation was lower by treatment
for the drop plate method (24, 26, and 28% for the no, low, and high sludge amended
soils) when compared to the serial dilution technique (45,9, and 39%, respectively).
Dehydrogenase Assay
The dehydrogenase assay (DHG) measures the amount of the pale yellow 2,3,5
triphenyltetrazolium chloride (TTC), a competitive NAD inhibitor, which is reduced to
the red triphenylformazan (TPF) by various dehydrogenases. Theoretically, DHG should
be useful for measuring the level of activity of that ubiquitous group of enzymes regardless of species, if no nutrient amendment were used which would skew the results
66
through selection (Stevenson, 1959). However, preliminary studies on the organic matter
poor soils in this study showed that readings obtained without an amendment were too
low after four days of incubation to be analyzed (results not shown) due to the lack of degradable organic substrate in the soil. Similar findings were obtained by Klein et al.
(1971) in Oregon on an arid soil.
As adding an amendment would favor the activity of organisms that readily utilize that substrate, three separate experiments were performed in which three diverse nutrient amendments were used. Given the design of the dehydrogenase assay study, the most
informative analytical procedure for evaluating the results was a multivariate analysis of
variance (MANOVA).
In this study, TPF formation as a result of adding glucose (similar to the plating
medium used for fungi), peptone yeast broth (similar to what was used in the bacterial
plate counts), and glycerol-casein broth (similar to what was used for the actinomycete
10.00 8.00 -.
c 6.00
y=5.82+O.00123x+O.00270x2
2.00
Y2m=O2O4
0.00
0
5
I
I
10
15
I
20
25
Figure 4: A significant trend in DHG with GC amendment as a function of total
sludge application over a four-year period (the abscissa is in units of Mg dry sludge
ha').
67
counts, but without the antifungal agents and micronutrients) thus served as three dependent variables in the MANOVA. A MANOVA is intuitively important here as some organisms stimulated by one amendment broth may be stimulated by another. However, as
Bartlett's test of sphericity showed that the dependent variables were not correlated with
one another, the use of three univariate analyses of variance was justified. In other words,
it seems that the populations responding to each of the amendments were mutually exclusive. Whereas the MANOVA showed a significant difference between the mean vectors
for each sludge treatment, univariate ANOVAs attributed the difference to a significant
difference in TPF formation in response to the glycerol-casein amendment. Here, the
high sludge amendment was significantly higher in TPF formation than the no and low
sludge amended soils (Table 6), with a significant curvilinear increase with additional
sludge amendment. Note, however, that the fit of the model was poor (Figure 4).
Microscopic examination of the MeOH-extracted soil from a no sludge plot found
numerous bacterial cells still containing unextracted TPF. Fungal spores and amorphous
particles, most likely containing soil organic matter, also retained the red stain. These observations affirmed the findings of Glathe and Thalmann (1970a), that MeOH does not
completely extract all of the TPF from the soil.
68
Carbon Dioxide Evolution Analysis
Carbon dioxide evolution proceeded quickly and had probably peaked within the
first 24 h of incubation, before the first reading was taken in the soils of both the water
and alfalfa amendments. In both the water and alfalfa amended studies, a significant increase in the initial CO2 production (intercept) was observed. The organisms in the water
amendment study (Figure 5) possibly responded to a flush of native soil organic matter
made available through chemical changes in the organic material which may have occurred through drying of the soil in the field prior to collection (Waksman and Starkey,
1924). The easily degraded alfalfa promoted much larger initial levels of evolution (Fig-
100
y=148-73.4:+14.8r0.995t (No Sludge)
- - -y=148-73.41+14.8:2-O.995r3 (Law Sludge)
- y=lSI-73.4:14.8r -0.995: (High Sludge)
80(1)
-D
0)
0)
0
0)
>
0
>
6040-
w
C20
C)
0
0
1
I
I
I
I
I
2
3
4
5
6
7
time (days)
Figure 5: CO2 evolution as a function of time for the three sludge treatments in response to moisture amendment. The equations were averaged over blocks. Each data
point represents the mean of the four blocks of each sludge treatment.
69
ure 6). White, fungal growth on the soil surface peaked by the second day as did a heavy
musty smell on the third day in the jars in the alfalfa amended soil.
The time x interaction was not significant in the water study but was significant at
the linear level in the alfalfa amendment study. That is to say, sludge treatment had no
significant effect on the rate at which CO2 production occurred in the water amendment,
while in the alfalfa amendment study there was a significantly more rapid decrease in the
rate of CO2 production in the high sludge treated soil than in the no and low sludge treatments.
700
600 -
y=787-214r+16.1t (No Sludge)
--r-792-2151+16.112(Low Sludge)
0
C?)
V
C)
C)
- y--822-22Ot+16.1: (High Sludge)
2
r
=O.99
500400300-
>
0
>
w
200-
c'J
0o
1000
0
I
I
I
2
3
4
5
6
7
time (days)
Figure 6: CO2 evolution as a function of time for the three sludge treatments in response to moisture + alfalfa amendment. The equations were averaged over blocks.
Each data point represents the mean of the four blocks of each sludge treatment.
Note the larger magnitude of initial CO2 production as compared to the moisture
amendment depicted above in Figure 5.
70
DISCUSSION
Experiments to determine the microbiological activity of the soils at Marana
showed no signs that four years of sludge application in accordance with plant nutritional
needs have harmed soil microbes. Indeed, there is some indication that sewage sludge applied at levels providing nitrogen in excess of what plants need may actually, at least for
the present, be stimulating the overall activity of the microorganisms.
As mentioned earlier, the rationale for collecting soil at the end of the growing
season was to avoid any stimulation which the rhizosphere effect may have had on the microorganisms in a given core sample and to give an estimate of the long-term microbial
activity, which was residual in the soil after the crops had been removed and the most easily degraded residues decomposed. To be sure, the activities would have been vastly dif-
ferent had the soil been collected soon after sludge application due to the high influx of
sludge nutrients, but that would have confounded attempts at finding any long-term toxic
effect of, e.g., heavy metals on the microorganisms.
On the other hand, collecting the soil well after the end of the growing season
may have decreased measurable microbial activity due to the harsh conditions in the soil
(moisture and nutrient stress). Using soil collected after crop removal under arid conditions lead to the soil being a harsh environment for organisms to survive in, a factor
which may have made it difficult to measure the activity of the organisms. Roszak and
Colwell (1987) reviewed a number of mechanisms by which organisms survive periods
of harsh environmental conditions, such as desiccation and starvation. They found that
71
the numbers of organisms detennined by direct counts increased with time as compared
with the numbers of organisms measured using heterotrophic plate counts, the difference
being a segment of the population that may be surviving in some form of dormancy, still
alive and respiring, but not capable of dividing-the viable but non-culturable organisms.
As a result, the activities measured in this study, including plate Counts and the physiological assays DHG and CO2 evolution, could have been measuring only a segment of
the population that was able to quickly return to an active state despite high levels of nutrients in the growth medium that may have been toxic to some organisms, as in the plate
counts on peptone yeast, or enzyme assays in glucose, peptone-yeast, or glycerol-casein
broth. However, this explanation seems unlikely as direct counts were not significantly affected by sludge application, and CO2 evolution did show differences between sludge
treatments both where only water was added to the soil, and where water and alfalfa were
added to the soil. What appears to be the case is that a segment of the population, which
was being somehow stimulated by a factor in the sludged soils such as increased micronu-
trient availability (Zn, Cusee Table 3 on page 59, and Figure 2 on page 60), was ready
to come out of dormancy despite desiccation and possible starvation in the soil as soon as
moisture and substrate were available.
In this study, viable heteroirophic plate counts for bacteria, actinomycetes, and
fungi were unaffected by sludge treatment. Similar fmdings were made by McGrath and
Brookes (1986) in a 20 year sludge land treatment experiment for plate counts of bacteria, actinomycetes, fungi, and protozoa, in soils with higher levels of available heavy met-
72
als (EDTA-extractable). Viable heterotrophic plate counts serve as an estimation of the
number of viable, heterotrophic organisms in the soil. However, as Verstraete and Voets
(1977) pointed out, there may be too many inherent errors in these analyses for them to
be sensitive to subtle population variations in the soil. For example, viable heterotrophic
plate enumeration assays assume that one organism will result in one colony (and, hence,
the results are reported as Colony Forming Units or CFU) and will result in one colony
on the plate. Yet, in reality, a single colony may be the result of a group of organisms that
were bound together, and an underestimation might result, which will depend on the nature of the organisms and how they bind to each other and to soil particles. Also, not all
organisms will grow equally well on the same medium, and some organisms may not
grow at all, and, hence, not be counted. For just these reasons, direct counts were performed, which make none of these assumptions.
In his studies using a series of treatments ranging greatly in productivity due to
management, Waksman (1922c) found some influence of management on numbers of
soil microorganisms, bacteria, actinomycetes, and fungi. But in the soils of this study,
which were similar both physically and chemically (Table 2, page 57), the methods used
were not sensitive enough to determine small differences. On the other hand, the soils do
reflect the status of a fertile soil for which Alexander (1977) reported 108.1010 bacteria
(heterotrophic plate counts 10% of the direct count values), 105 108 actinomycetes, and
20000.106 fungal "propagules" (spores, hyphae, or hyphal fragments) per gram of dry
soil (see Table 6, page 65).
73
In contrast to enumerations, physiological assays can measure activity independent of the actual number of organisms present. Although much research has been
conducted on the role extracellular enzymes may play in the environment (Burns, 1982;
Skujins, 1967; Skujins, 1976), dehydrogenase is not an enzyme believed to be active on
the outside of the cell (Skujins, 1967). McGrath and Brookes (1986) stated that DHG is a
function only of living cells, and as such, is a better measure of activity than phosphatase,
which has been shown to be associated with extracellular enzymes (Skujins, 1976). In
this study, there was no sign of significant inhibition of dehydrogenase activity in metal
salt amended soils, as was reported by Rogers and Li (1985) and Beck (1981), or in highCu sludge amended soil (McGrath and Brookes (1986); Brookes and McGrath (1984))
but rather an indication of stimulation of some segment of the population (organisms that
can utilize the glycerol-casein amendment), possibly due to enhanced micronutrient availability (compare Figures 2, B and C, and 4 on pages 60 and 66).
At the outset, we believed that increases in activity due to the broth amendments
could be attributed to those organisms that grew on those compounds in the agar form.
However, as can be seen above, this was not the case (note the lack of any association between enumeration assays and DHG values in Appendix Table 10). One explanation
could be the anaerobic conditions present in the incubation tubes at the end of the incuba-
tion periodon the one hand, few fungi and actinomycetes are anaerobes and on the
other, the plate counts exclusively enumerated aerobic organisms. In the glycerol-casein
amendment, however, the samples had to be incubated twice as long as the other amend-
74
ments in order for measurable levels of TPF to be formed; the actinomycetes grown on
agar required 17 days to be counted as opposed to one week for the bacteria on peptone
yeast. Notwithstanding, are these results representative of an actual phenomenon in the
soil or a result of the many factors which can lead to erroneous results by the DHG assay?
Microscopic examination of some samples of soil after complete extraction by
methanol found water-insoluble TPF on soil organic matter particles and in the interior of
fungal spores and bacterial cells, as was described by Glathe and Thalmann (1970a), leading to the conclusion that the TPF was not completely removed from the soil. In fact,
Glathe and Thalmann (1970a) found that the common practice of repeated methanol extraction removed only about 70% of the total TPF in the soil, with extraction efficiency
dependent on the soil moisture content, organic matter content, and the texture of the soil.
To this end, they developed a method of extraction using methanol, acetone and CC14,
which they found to extract 100% of the TPF in the soil. Organisms may be able to continue reducing TTC only up to a point, as the material collects in the cells and may cause
them to rupture (Glathe and Thalmann,. I 970b). These findings bring up the question as
to whether differences in the type of soil organic matter, and the type of organisms and
their ability to accumulate TPF could lead to skewn dehydrogenase values in the analyzed samples.
Glathe and Thalmann (1970a) also demonstrated that the toxicity of TTC, a com-
petitive NAD inhibitor, was advantageous, as an "atypical propagation of the microorganisms, which could falsify the results, is largely avoided" (Glathe and Thalmann
75
(1970a), translated by the author). This may be important as the addition of nutrients,
such as glucose and peptone yeast broth, has been shown to lead to measurable multiplication of organisms in as little as 24 hours (Casida, 1977). Furthermore, Glathe and Thaimann (1970a) suggested that because reductase activity should begin immediately if the
enzyme is present, and dehydrogenase activity in air-dried soil has been shown to com-
mence only after a short incubation period, the term "dehydrogenase activity" may not be
appropriate; "TTC-reduction" should be used instead.
However, Glathe and Thalmann (1970a) cautioned against using incubation periods longer than the linear portion of the TPF production curve (as a function of time), as
they noticed in fertilization studies that incubation periods longer than 24 h led to results
indicating significant differences between fertilization treatments which were not significantly different with a 24 h incubation period. At what portion of the TPF production
curve the readings in this study were taken is not known, but as suggested by the work of
Casida (1977), they were most probably taken outside of the linear range in the glucose
and PY amendment studies.
Additionally, DHG results are known to be sensitive to extended soil storage
time, even at 4°C (Ross, 1970), as was the case here. Drying of the soil, as took place to
some degree in the field prior to collection and during storage, has also been known to adversely affect results (Glathe and Thalmann, 1970b). All of these factors may have made
the obtained results difficult to relate to the state of the soil. Indeed, Versiraete and Voets
(1977) classified the dehydrogenase assay as one of the methods "not sufficiently sensi-
76
tive or accurate to detect and reflect in a statistically significant way minor differences be-
tween soil microbial populations," and Ross (1973) cautioned against using DHG as a
measure of soil respiration until the factors that affect the results are understood.
Use has been made in the past of basing enzyme activity results on an organic C
basis instead of a soil weight basis (Ross, 1973; Dutzler-Franz, 1977a; Schuller, 1989).
Beck (1984ab), in relating his Soil Microbiological Index (SMIsee page 30) to soil
types and management practices, found good correlation between the SMI of a diverse ar-
ray of soils and the soil organic matter content, justifying this relationship as representing
the tendency of microorganisms to colonize humus particles. He pointed out that although there is good correlation between values represented on a soil weight basis with
those on an organic C basis in many agricultural soils, the relationship changes drasti-
cally in soils that are extremely poor or rich in humus. Note that in the soils of Beck's
study (1984b) where sewage sludge had been applied, the predicted values of the SMI
from a quadratic formula generally greatly overestimated the observed values. Dutzler-
Franz (1977a) found that basing enzyme activities on an organic C basis improved correlation with bacterial numbers obtained by heterotrophic plate counts, while Schuller
(1989) observed correlation of DHG on an organic C basis with microbial biomass. How-
ever, a trial using DHG values converted to an organic carbon basis yielded no new insight into the data in this study and was not pursued any further. Possibly, the organic
material derived from sewage sludge behaves differently than that derived from crop residues, Logarithmic transformations of the enzyme assay results, as discovered by Dutzler-
77
Franz (1977a) to improve correlation with some soil properties across soils of increasing
humus content, showed a significant positive correlation with total soil nitrogen, but not
with total organic carbon (Appendix Table 11).
As for the dehydrogenase activity, CO2 evolution is also believed to be the result
of activity of living cells and not of soil enzymes (Stotzky and Norman, 1961). Waksman
and Starkey (1924) theorized that CO2 evolution is an index of soil respiration as CO2 is
the end product of the oxidation of carbon substrate by many heterotrophic organisms.
Waksman and Starkey (1924) classified CO2 analyses into two categories, based on how
the experiment was carried out. "Respiratory power" is derived from CO2 evolution in
soil incubated under optimum temperature and moisture conditions, and is a function of
the number of organisms present in the soil and the suitability of the soil for their growth.
"Decomposing power" refers to CO2 evolution experiments where an organic amendment
is added to the soil and measures the rate at which it is decomposed. It, too, is a function
of the factors which determine the respiratory power of the soil. In this study, the water
amended experiment could be considered a measure of respiratory power and the alfalfa
amended experiment a measure of decomposing power. Starkey (1924) observed that
soils under different management practices did not vary significantly in their decomposing power as they did in respiratory power. Yet In both experiments performed here, similar results were obtained for both respiratory and decomposing power, with the high
sludge amended soils respiring at significantly higher levels (i.e., the intercepts were
higher) than the no and low sludge amended soils (Figures 5 and 6 on page 68). Possibly,
78
some property in the soils associated with sludge application, such as increased micronuirient availability (see Figure 2, page 60) may have stimulated microbial metabolism.
However, the two amendment studies differed in the way that sludge treatment affected the rate at which CO2 was produced. As can be seen in Figure 5, all three sludge
treatments exhibited the same rate of CO2 production in the moisture amendment study,
i.e., the CO2 evolution curves were parallel. On the other hand, CO2 production in the
high sludge treated soils in the moisture + alfalfa amendment study decreased significantly more rapidly than in the no and low sludge treated soils (Figure 6). Those organisms responsible for the elevated respiration in the high sludge treated soils apparently
exhausted the substrate more quickly than in the soils of the other two treatment regimes
and returned to low levels of activity. An enhanced micronutrient availability status in the
high sludge treated soils may have played a role here as well.
Phosphorous availability has been demonstrated to be a limiting factor to soil microbial respiration as measured by CO2 evolution in laboratory studies (Stotzky and Nor-
man, 1961; Nannipieri eral., 1978). Available PO4-P was significantly increased by
increasing sludge treatment (Table 2 and Figure 1, on pages 57 and 58) as was CO2 evolution (Figures 5 and 6). However, whether increased PO4-P availability may have caused
the elevation in CO2 cannot be determined from this study given the experimental design.
In view of the results of the above DHG (GC) and CO2 evolution studies, the null
hypothesis that the sludge had no effect on the soil microorganisms is rejected. Nevertheless, as the results indicate stimulation of microbial activity with sewage sludge treat-
79
ment, there is no tangible support for concerns about any detrimental effects that sludge
could have had on the organisms that play an important role in maintaining soil fertility.
In their long-term study, Brookes and McGrath (1984) were able to detect decreased microbial biomass in soils twenty years after the last sludge application, with the lowest levels found in soils with the highest heavy metal contents. However, the EDTA-extractable
metal concentrations in their study were much higher than what was measured at Marana,
even in the high sludge amended soils, with values for Cu up to 90 mg kg1 and Ni up to
10 mg kg1, as compared to the high values at Marana of 2.86 and 0.103 mg kgt (DTPAextractable). As the Marana soils had an alkaline pH (7.7 in 0.01 M Cad2) and lower ap-
plication rates (2.0 Mg ha' y1 (low) and 6.0 Mg ha1 y1 (high) air-dried sludge) over
four years, while the soils of the Brookes and McGrath study were neutral (1: 25 (wlv)
soil : 1120), with sludge applied at higher rates over a longer period of time (8.2 Mg ha1
y' (low) and 16.4 Mg ha1 y1 (high) for 20 years), it is possible that sludge application
could continue at Marana for many years to come without exhibiting any negative effect
on the soil microorganisms.
Measurements of plant gmwth have often been difficult to definitively associate
with measurements of soil fertility (Verstraete and Voets, 1977), and many researchers
studying the effects of management on soil microbial activity have avoided associating
the results with plant yield (Bolton etal., 1985; Fraser et al., 1988; Doran etal., 1987;
Martyniuk and Wagner, 1978). In contrast, Waksman (Waksman, 1922c; Waksman and
Starkey, 1924) demonstrated a general relation between crop yields and heterotrophic
80
plate counts of bacteria, actinomycetes, and fungi and carbon dioxide evolution in soils
under strikingly different management practices for 13 years. It appears that measurable
association can exist between crop yields and microbial activity, given the proper measure of activity and the use of soils vastly different in fertility. Yet, the usefulness of soil
microbial activity in a predictive fashion with regard to soil fertility is dependent on
whether the following question can be conclusively answered: Are soil microorganisms
more sensitive to environmental pollutants than crop plants, i.e., can changes in microbial
activity be used to reliably predict changes in yield as affected by pollutant stress?
In the present study, heavy metal data were not significantly associated with any
of the plant growth results, except for a significant negative correlation of DTPA-extractable Cr with plant stand as measured in November (see Appendix Table 6). Copper, the
metal approaching limiting levels in the soil, with DTPA extractable levels increasing
with additional sludge application, did not contribute to the plant growth results, probably
as it is one of the least mobile trace elements and is readily fixed (Baker, 1990).
In contrast to the cotton, some measurements of microbial activity were associated with higher levels of certain DTPA-extractable metals (see Appendix Table 2) where
several significant positive correlations were observed, thereby rejecting the null hypothesis that there was no significant association between soil properties and soil microbial activity. Bacteria counts on soil extract agar were significantly positively correlated with
DTPA-extractable Cu and bacteria by direct count with Cr. Fungi (by the serial dilution)
were significantly positively correlated with Zn and Cu, fungi by the drop plate method
81
with Cr, and DHG (glycerol casein) with Zn and Ni. As explained earlier, it is not surprising that no similar relation existed between microbial parameters and total metals (Appen-
dix Table 1), possibly, in part because the metals added to the soil in the form of sludge
were too dilute for detection (Table 4, page 61). McGrath and Brookes (1986) (see also
Brookes and McGrath, 1984), in soils treated with a low metal sludge contml, and sludge
amended with Zn, Cu, Ni, and Cr, found that only EDTA-extractable Cu and Ni were associated with reductions in soil microbial biomass, and they suggested that these metals
were possibly the most toxic to microbial biomass. The difference in results between
their study and the one at Marana may be due to the difference in soil pH (6.0 as opposed
to 7,9 at Marana) or the nature of the microbial test used. A significant positive association also existed between available PO4-P and total N and heterotrophic bacteria counts
on peptone yeast (Appendix Table 3). However, as these correlations are the result of an
observational study, no cause and effect relationship between higher microbial activity
levels and increased heavy metal availability can be drawn, and any of a number of unknown variables could have led to the foregoing associations.
In view of the above, it seems that organisms may not have been as sensitive to
adverse effects of the sludge as plants may have been, and the validity of using soil microbial measurements, at least as performed hem, as a measure of soil fertility as defmed by
the suitability of the soil for continued acceptable yields needs to be questioned. The interactions between soil, sludge, microorganisms, and plants are complicated. Microorganisms may react differently to high concentrations of heavy metals than plants, and how
82
microorganisms may react is not well understood. For example, Quraishi and Cornfield
(1971) found that EDTA-extractable Cu concentrations (Cu added as a salt) as high as
1000 mg kg1 stimulated N mineralization, and Grossbard (1973) speculated from this research that plants may not tolerate concentrations as high as the microorganisms. In con-
trast, Duxbury (1985) pointed out that research on the effect of heavy metals of
microbially mediated process in situ is often conflicting and the mechanisms poorly un-
derstood. Angle et al. (1991), working with soil bacteria isolated from polluted soil sampled near a zinc smelter in Pennsylvania, found that bacteria have a high intrinsic
resistance to heavy metal concentrations in soil. Additionally, the more rapid rate of proliferation of microorganisms may complicate comparisons between the state of the soil after harvest and the conditions at planting time: The soil at the time of seedling
establishment may not be conducive to plant growth, consequently leading to plant stand
or even yield reduction in the mature plants, while the microorganisms may have time to
recover from any initial adverse conditions no longer present. This study, possibly for
some of the reasons given above, determined that the only detectable treatment effect on
cotton plant growth was negative, while the only detectable treatment effect on soil microbial activity was stimulatory. Tests for association between plant growth and soil microbial parameters failed to reject the null hypothesis that no association between microbial
activity and plant growth parameters existed (Appendix Table 4). Consequently, the use-
fulness of a soil microbial index for measuring changes in soil fertility as a result of pollutant stress has not been substantiated.
83
CONCLUSIONS
A study to evaluate the use of sewage sludge as an organic soil amendment and
plant fertilizer has not demonstrated any detrimental effect of land-applied sewage sludge
on cotton production after four years of sludge application, as cotton yields have remained high. An auxiliary study has, in addition, indicated that potential recalcitrant pol-
lutants in the sludge, most notably heavy metals, do not appear to have accumulated to a
degree sufficient to harm soil microorganisms, which are important in the maintenance of
soil fertility. In fact, some evidence showed stimulation of microbial activity with increasing sludge treatment. However, this study suggests, also, that soil microbial activity may
not be useful as a predictive indicator of soil fertility in response to pollutant stress, as
soil microorganisms may not be as sensitive under field conditions to pollutants as plants.
Significantly reduced cotton plant stand had no corresponding decreasing parameter
among the common microbial assays performed here. Therefore, caution should be used
when diagnosing a field soil for potential phytotoxicity exclusively using soil microbio-
logical assays. In view of the results of this study, performing research to evaluate how
much of a recalcitrant pollutant can be applied to a soil without diminishing its plant production potential may not be appropriate as a long-term solution to current sludge disposal problems. The future of land application as an environmentally sound sewage
sludge disposal avenue will ultimately be a function of the success of efforts to reduce
the amount of recalcitrant pollutants, such as heavy metals, entering the waste stream.
84
APPENDIX: TABLES OF ASSOCIATIONS
BETWEEN DEPENDENT VARIABLES
Appendix Table 1: Correlations between total soil heavy metal concentrations and
microbial parameters. Normal face values are r1 yZm' italicized values are the probability of rejecting H0, that P)l)l2=0' and bold face values are the Kendall
b.
NA and NA denote violation of the assumption of bivariate normality.
Bacteria
(PY)
Bacteria
(SEA)
Bacteria
(direct count)
Actinomycetes
(GC)
Fungi
(serial dilution)
Zn
(acid)
0.520
0.101
0.249
Cu
(acid)
0.729
0.011
0.554
NA
0.094
0.595
0.054
0.454
NA
NA
NA
DHG
(glucose)
DHG
(PY)
01
(acid)
-0.171
0.615
-0.143
(acid)
0.005
0.988
0.041
-0.155
0.223
0.509
-0.039
-0.121
0.724
-0.100
0.066
0.847
0.013
-0.333
0.214
0.183
NA
NA
-0302
NA
NA
NA
NA
0317
0328
0390
NA
0.086
-0.143
0.151
0.227
0.139
0.117
0.731
0.149
0.256
0.449
0.271
NA
NA
0.218
0.338
NA
NA
0320
0309
-0.039
0.233
0.216
0.404
NA
0.036
0.917
-0.053
-0.182
0.591
-0.047
0.114
0.739
-0.172
NA
NA
NA
NA
-0.036
0.024
NA
-0.022
0.948
-0.092
-0.02 1
NA
0.952
0.056
0.162
-0.210
NA
-0.069
0.840
-0.168
.0.187
-0.272
0.418
0.0498
NA
NA
NA
NA
NA
NA
NA
NA
0.233
0.177
-0.140
0.098
-0.049
0.886
-0.030
NA
0.662
0.027
0.522
NA
0.411
NA
NA
NA
0210
NA
NA
-0333
0.151
-0.177
-0.188
NA
NA
NA
NA
NA
NA
NA
NA
0.540
.0.776
0.205
-0.080
-0.346
0.298
-0.0334
NA
0.282
DHG
(GC)
Cr
0.149
-0.168
Fungi
(drop plate)
Ni
(acid)
0.419
0.199
0.113
Pb
(acid)
NA
0.559
0.074
0.316
NA
NA
0336
.0.152
85
Appendix Table 2: Correlations between DTPA-extractable metals and microbiological parameters. Normal face values are
italicized values are the probability
of rejecting H0, that p3,1
and bold face values are the Kendall
b. NA
and NA denote violation of the assumption of bivariate normality.
Zn
TPA
0.484
0.132
0.423
Cu
TPA
NA
NA
NA
0.397
0.156
NA
0.397
0.632
0.037
0.386
0.067
0.845
0.165
NA
NA
0.588
NA
NA
0.174
0.176
0.057
0.339
0.236
0.485
0.239
0.467
0.147
0.381
Fungi
(serial dilution)
0.613
0.045
0.057
Fungi
(drop plate)
Bacteria
(PY)
Bacteria
(SEA)
Bacteria
(direct count)
Actinomycetes
(GC)
DHG
(glucose)
DHG
(PY)
DHG
(GC)
Pb
ITPA
NA
Ni
PTPA
-0.074
Cr
TPA
NA
0.829
0.222
NA
-0083
NA
-0.205
NA
0545
0.218
-0.197
0.254
0.451
0.020
0.607
0.048
0.415
0.397
0.309
-0.204
0.548
0.108
0.703
0.016
0.144
0.446
0.169
0.135
-0.041
0.905
-0.178
0.469
0.146
0.330
0.364
0.271
0.043
0.419
NA
-0.345
0.299
-0.148
0.67 1
NA
NA
NA
NA
0226
0.338
0200
NA
0.169
0.144
NA
NA
NA
NA
0.138
0.288
0.031
0.928
0.296
0.336
-0.115
0.737
-0.070
0.470
0.145
0.373
0.390
0.235
0.353
NA
0.355
NA
NA
0283
NA
0.192
0.343
0.016
0.67 1
0.478
0.137
0.374
0.554
0.077
0.448
0.656
0.028
0.452
NA
0.024
0.388
NA
0.024
0.341
NA
0.259
86
Appendix Table 3: Correlation between various soil physico-chemical properties
and soil microbiological parameters. Normal face values are
italicized values
are the probability of rejecting H0, that p,1 2.ut=0 and bold face values are the Kendall
b. NA and NA denote violation of the assumption of bivariate normality.
pH
WHC
CEC
PO4-P
0.455
0.160
0.359
NA
NA
0.669
0.024
0.514
0.389
0.376
0.742
0.009
0.648
0.598
0.052
0.419
0.254
0.451
0.316
0.533
0.091
0.508
NA
NA
NA
NA
-0.050
-0.027
0.394
0.230
0.168
NA
NA
Total
Total N
Or:anicC
Bacteria
(PY)
0237
NA
NA
0.082
0.016
NA
0.251
-0.838
0.001
0.054
0.180
0.034
0.922
0.295
-0.272
0.418
.0.245
-0.348
0.294
-0.282
-0.213
0.530
0.105
NA
NA
NA
0.085
0.000
0.999
0.046
0.078
0.022
0.537
0.088
0.385
-0.369
0.264
-0.164
-0.670
0.024
-0.171
NA
NA
NA
NA
NA
NA
NA
NA
-0.027
0.005
0.350
0.201
-0.338
NA
NA
NA
-0.242
-0.183
0.418
0.200
0.144
NA
0310
-0.103
0.764
0.029
-0,059
0.862
-0.055
NA
NA
NA
NA
NA
NA
NA
-0.285
0.092
0.450
-0.063
-0.007
0.984
0.300
DHG
0.352
NA
-0.390
NA
NA
0.5 22
(PY)
0288
NA
0236
NA
NA
0.263
0.128
-0.083
0.614
0.192
0.100
0.488
DHG
0.324
-0.001
NA
NA
NA
NA
(GC)
0332
0.998
0.047
NA
NA
NA
NA
-0.248
0.597
0.000
0.674
Bacteria
(SEA)
Bacteria
(direct count)
Actinomycetes
(GC)
Fungi
(serial dilution)
Fungi
(drop plate)
DHG
(glucose)
0.343
0301
0.244
NA
NA
NA
-0.075
NA
NA
0.103
87
Appendix Table 4: Correlations between cotton growth paramters and soil microbiological parameters. Normal face values are
italicized values are the probability of rejecting H0, that P)l)2=0 and bold face values are the Kendall
b.
NA and NA denote violation of the assumption of bivariate normality.
Yield 1
Bacteria
(PY)
NA
NA
-0.037
Bacteria
(SEA)
Bacteria
(direct count)
Actinomycetes
(GC)
Fungi
(serial dilution)
Fungi
(drop plate)
DHG
(glucose)
DHG
(PY)
DHG
(GC)
Yield 2
-0.434
0.182
-0.329
NA
Total
Yield
NA
NA
-0.143
-0.159
0.641
.0.075
-0.279
0.140
0.682
0.062
0.094
0.783
0.027
-0.111
0.371
0.262
0.283
-0.185
0.586
-0.048
0.275
0.414
0.178
0.033
0.923
0.151
NA
-0.200
-0.137
0.688
-0.016
0.164
0.630
0.109
0.097
0.776
0.083
NA
-0.357
0.282
-0.188
Plant Plant Stand
June
0.245
-0.026
0.468
0.939
0.101
-0.210
Hei ht
NA
-0.186
Plant Stand
ovember
0.232
0.4 92
0.036
0.412
0.208
0.067
NA
0584
0.128
-0.216
0.425
0.192
0.206
0.348
0.295
0.133
-0.296
NA
0.483
0.132
0.292
0.095
0.781
0.029
NA
-0.013
-0.022
0.948
0.031
-0.081
0.016
-0.045
0.896
-0.105
0.229
0.499
0.206
-0.260
0.440
0.012
NA
NA
NA
NA
NA
NA
NA
0.174
0.609
NA
NA
NA
-0.070
0.111
0.745
NA
NA
NA
NA
NA
NA
NA
.0.009
NA
-0.731
NA
-0.705
0.013
-0.292
-0.151
NA
-0.195
NA
NA
NA
NA
NA
-0.204
0565
NA
NA
NA
NA
-0.151
-0.175
-0.004
-0.248
-0.249
NA
NA
NA
NA
NA
NA
NA
NA
-0.163
NA
NA
0.1784
NA
NA
-0.201
-0.362
-0.220
.0.291
88
Appendix Table 5: Correlations between cotton plant growth parameters and total
heavy metals. Normal face values are r1 y2m' italicized values are the probability of
rejecting H0, that p3,1>,2=0, and bold face values are the Kendall
b. NA and
NA denote violation of the assumption of bivariate normality. NA and NA denote violation of the assumption of bivariate normality.
Yield 1
Yield 2
Total Yield
Plant Height
Zn
acid
NA
Plant Stand
(November)
Pb
acid
NA
Ni
acid
NA
Cr
acid
NA
Cd
acid
0.380
NA
NA
NA
NA
NA
0249
-0.109
-0.129
0.170
0.000
0.126
0.247
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
-0.381
-0.284
0.009
-0.274
-0.231
0.078
0.818
0.023
NA
NA
NA
NA
NA
0.396
NA
NA
NA
NA
NA
0227
-0.216
-0.210
0.206
-0.075
0.042
0.254
NA
NA
NA
NA
NA
NA
NA
NA
NA
-0.041
0.525
0.097
0.406
-0.031
0.104
0.272
0.304
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
0.073
-0.206
0.20972
0.264
0.330
0.430
0.187
0.216
NA
NA
NA
NA
NA
NA
NA
NA
-0.012
-0.146
0.355
0.284
0.296
0.028
-0.006
NA
Plant Stand
(June)
Cu
acid
NA
0.238
0.480
0.275
89
Appendix Table 6: Correlations between cotton plant growth parameters and DTPAextractable heavy metals. Normal face values are
italicized values are the
probability of rejecting H0, that
and bold face values are the Kendall
b. NA and NA denote violation of the assumption of bivariate normality. NA
and NA denote violation of the assumption of bivariate normality.
Yield 1
Yield 2
Total Yield
PlantHeight
Plant Stand
(June)
Plant Stand
(November)
Zn
TPA
-0.187
Cu
Pb
Ni
Cr
'TPA
'TPA
-0.040
0.906
-0.166
-0.415
0205
NA
-0.270
0.248
-0.109
0.750
.0.097
-0.068
0.844
-0.079
NA
-0.425
0.193
-0.343
'TPA
0.582
-0.006
aTPA
-0.282
0.400
-0.220
NA
NA
NA
NA
-0.070
-0.049
-0.283
-0.347
0399
0296
-0.075
-0.264
-0.085
0.803
-0.266
-0.123
0.718
0.158
-0.229
0.498
0.134
0.110
0.748
0.159
0.361
-0.117
0.731
-0.110
-0.169
0.619
.0.062
NA
-0.324
NA
0330
0.450
0.165
0.066
-0.218
0.325
NA
0.143
0.675
-0.263
-0.559
0.074
-0.458
-0.426
0.191
-0.166
-0.678
0.022
NA
-0.316
0275
0.072
NA
NA
0.121
0.283
0399
0.116
-0.147
0.666
-0.178
-0384
90
Appendix Table 7: Correlations between DTPA-extractable and total heavy metals.
Normal face values are r1 yZm' italicized values are the probability of rejecting H0,
that p3,1 T=° and bold face values are the Kendall
b. NA and NA denote violation Jthe assumption of bivariate normality.
Zn
(DTPA)
Cu
(DTPA)
Zn
(acid)
NA
Cu
(acid)
NA
Pb
(acid)
NA
Ni
(acid)
NA
(acid)
NA
Cr
NA
NA
NA
NA
NA
0.165
0.330
-0.224
0.157
-0.207
NA
NA
NA
NA
NA
NA
NA
NA
0.200
0.282
-0.132
0.174
NA
NA
-0.039
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
0.163
0.212
-0.229
0.174
0.112
Ni
NA
NA
NA
NA
NA
(DTPA)
NA
NA
NA
NA
NA
0.401
0.248
-0.221
0.383
-0.075
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
0.006
-0.090
-0.116
0.105
0.161
Pb
(DTPA)
Cr
(DTPA)
91
Appendix Table 8: Correlations between soil physico-chemical properties and
DTPA-extractable heavy metals. Normal face values are
italicized values are
the probability of rejecting H0, that p,1=0, and bold face values are the Kendall
b. NA and NA denote violation of the assumption of bivariate normality.
Zn
NA
Cu
TPA
NA
Pb
TPA
NA
NA
NA
NA
NA
NA
0.263
0.044
-0.023
0.071
-0.042
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
-0.237
-0.356
-0.060
0.015
-0.231
NA
NA
NA
NA
-0.144
NA
'TPA
pH
WHC
CEC
PO4-P
Total
Organic C
TotaiN
Ni
'TPA
NA
Cr
'TPA
NA
NA
NA
NA
-0.165
0.077
-0.010
NA
NA
0.021
NA
NA
NA
NA
NA
NA
NA
NA
NA
0.511
0.492
0.335
0.485
-0.040
NA
NA
NA
NA
NA
NA
NA
NA
0.187
0.127
NA
NA
-0.185
-0.105
0.029
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
0.531
0.586
0.467
0.315
0.270
NA
92
Appendix Table 9: Correlations between cotton plant growth parameters and soil
physico-chemical properties. Normal face values are
italicized values are the
probability of rejecting H0, that P)lrn=°' and bold face values are the Kendall
tyly2.trt b. NA and NA denote violation of the assumption of bivariate normality.
pH
EC
WHC
CEC
PO4-P
-0.210
NA
NA
NA
NA
0.210
0335
NA
NA
NA
NA
0536
0.119
-0.385
0.320
-0.135
-0.448
0.026
-0.1768
0.603
-0.111
NA
NA
NA
NA
-0.283
-0.266
0.430
-0.296
Total
Total N
Or:anic C
Yield 1
Yield 2
Total Yield
Plant Height
Plant Stand
(June)
Plant Stand
(November)
-0.276
0.410
0.024
NA
NA
NA
NA
-0.027
-0.163
-0.176
-0.214
-0.2688
0.424
-0.097
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
-0.397
0.338
-0315
-0.481
-0.046
NA
NA
NA
NA
NA
-0.152
0.657
-0.162
0399
-0.228
0.382
0.246
0.346
NA
NA
NA
NA
NA
0.496
-0.241
.0.105
0.114
-0.204
0.110
0.748
0.086
0.000
1.000
0.153
NA
NA
-0.471
0.203
NA
NA
NA
NA
0349
NA
NA
NA
NA
0.196
0.086
-0.392
-0.110
0.006
-0.029
0.932
-0.078
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
NA
-0.330
0.200
0.310
-0.334
-0.130
-0353
93
Appendix Table 10: Correlations between DHG analyses and enumerations results.
Note that although similar media were used between DHG with added PY broth bacteria plated on PY, and DHG with GC broth and actinomycetes on GC, there is no
italicized values
noticeable measure of association. Normal face values are
and bold face values are the Kenare the probability of rejecting H0, that
b. NA and NA denote violation of the assumption of bivariate normality.
dali
Bacteria
(PY)
Bacteria
(SEA)
Bacteria
(direct Count)
Actinomycetes
(GC)
Fungi
(serial dilution)
DHG
lucose
-0.219
0.518
0.202
NA
NA
0.397
DHG
GC
0.473
0.141
0.388
-0.058
0.865
0.216
0.288
0.319
0.204
0.547
0.167
NA
NA
NA
0390
NA
NA
NA
0.047
-0.206
0.162
NA
NA
NA
NA
NA
NA
-0.063
0.070
0.200
NA
NA
NA
NA
NA
-0.233
NA
0.036
Fungi
(drop plate)
DHG
NA
NA
-0.038
NA
NA
-0.348
0.073
NA
NA
-0.045
94
Appendix Table 11: Correlations between DHG analyses (natural logarithm transformation) and enumeration results. Normal face values are
italicized values.
are the probability of rejecting H0, that Py12m=°' and bold face values are the Kendall
b. NA and NA denote violation of the assumption of bivariate normality.
Total
Organic C
Total N
DHG
(glucose)
-0.371
0.262
-0.063
0.078
0.820
0.300
DHG
0.777
0.192
DHG
(GC)
0.011
0.974
0.000
0.539
0.087
0.488
0.776
0.005
0.674
(PY)
0.097
95
LIST OF REFERENCES
Adams T.McM. and Sanders, J.R. (1984) The effects of pH on the release to solution of
zinc, copper and nickel from metal-loaded sewage sludges. Environmental Pollution (Series B) 8, 85-99.
Alexander, M. (1977) Introduction to Soil Microbiology, 2nd edn. John Wiley & Sons,
New York.
Alloway, B.J. (1990a) Soil processes and the behaviour of metals. In Heavy Metals in
Soils (B.J. Alloway, Ed.), pp. 7-28. Glasgow, Blackie.
Alloway, B.J. (1990b) The origins of heavy metals in soils. In Heavy Metals in Soils
(B.J. Alloway, Ed.), pp. 29-39. Glasgow, Blackie.
Angle, J.S., Chaney, R.S. and Rhee, D. (1991) Bacterial resistance to heavy metalsMetal activity vs. total metal concentration in soil and media. Agronomy Abstracts.
83rd Annual Meeting, p. 256. American Society of Agronomy, Madison.
Arizona Department of Environmental Quality (1990) In Guidelines for the Agricultural
Land Application of Sewage Sludge, Phoenix, AZ.
Artiola. J.F. and Ali, A.-M.S. (1990) Determination of total sulfur in soil and plant samples using sodium bicarbonate / silver oxide, dry ashing and ion chromatography.
Communications in Soil Science and Plant Analysis 21, 94 1-949.
Artiola, J.F. (1989) Manual of Operating Procedures for the Analysis of Selected Soil,
Water, Plant Tissue and Waste Chemical and Physical Parameters, Soil and
Water Science Report #89-101. University of Arizona, Tucson.
Atlas, R.M. and Bartha, R. (1987) Microbial Ecology: Fundamentals and Applications,
2nd edn. The Benjamin/Cummings Publishing Company, Inc. Menlo Park.
Baker, D.E. (1990) Copper. In Heavy Metals in Soils (B.J. Alloway, Ed.), pp. 15 1-176.
Glasgow, Blackie.
Barnes, D. (1989) Metrogro sludge program boosts farm profits. BioCycle 30, 30.
Bartha, R. and Pramer, D. (1965) Features of a flask and method for measuring the persistence and biological effects of pesticides in soil, Soil Science 100, 68-70.
96
Beck, Th. (1984a) Mikrobiologische mid biochemische Charakterisierung landwirtschaftlich genutzter BodenI. Mitteilung: Die Ermittlung einer Bodenmikrobiologischen Kennzahl. Zeitschrftflir Pflanzenernahrung und Boden/wnde 147,
456-466.
Beck, Th. (1984b) Mikrobiologische und biochemische Charakterisierung landwirtschaftlich genutzter Bfidenll. Mitteilung: Beziehungen zum Humusgehalt.
Zeitschrftflir Pflanzenernahrung und Boden/cunde 147,467-475.
Beck, Th. (1981) Untersuchungen fiber die toxische Wirkung der in Siedlungsabf1len
haufigen Schwermetalle auf die Bodenmikroflora. Zeitschrrft für Pflanzenernahrung undBodenkunde 144, 613-627.
Beck, Th. and SÜSS, A. (1979) Der Einfluss von Klärschlamm auf die mikrobielle Tätigkeit im Boden. Zeitschrftflir Pflanzenernahrung und Boden/unde 142, 299-309.
Bell, P.F., James, B.R. and Chaney, R.L. (1991) Heavy metal extractability in long-term
sewage sludge and metal salt-amended soils. Journal of Environmental Quality
20,481-486.
Bohn. H.L., McNeal, B.L. and O'Connor, G.A. (1985) Soil Chemistry. John Wiley &
Sons, New York.
Bolton, H, Jr., Elliott, L.F., Papendick, R.I. and Bezdicek, D.F. (1985) Soil microbial
biomass and selected soil enzyme activities: Effect of fertilization and cropping
practices. Soil Biology and Biochemistry 17, 297-302.
Boyle, M. (1990) Biodegradation of land-applied sludge. Journal ofEnvironmental Qua!ity 19, 640-644.
Brookes, P.C. and McGrath, S.P. (1984) Effects of metal toxicity on the size of the soil
microbial biomass. Journal of Soil Science 35, 341-346.
Burns, R.G. (1982) Enzyme activity in soil: Location and a possible role in microbial
ecology. Soil Biology and Biochemistry 14,423-427.
Camper, A.K., LeChevallier, M.W., Broadaway, S.C. and McFeters, G.A. (1985) Evaluation of procedures to desorb bacteria from granular activated carbon. Journal of
Microbiological Methods 3, 187-198.
Casida, L.E., Jr. (1977) Microbial metabolic activity in soil as measured by dehydrogenase determinations. Applied and Environmental Microbiology 34, 630-63 6.
97
Chang, A.C., Granato, T. and Page, A.L. (1991) Land application of municipal sludge
Assessing the phytotoxicity of metals. Agronomy Abstracts. 83rd Annual Meeting, p. 38. American Society of Agronomy, Madison.
Coppola S. (1986) Summary of investigations in Italy into effects of sewage sludge on
soil microorganisms. In Factors Influencing Sludge Utilisation Practices in
Europe, (R.D. Davis, H. Haini, and P. L'Herrnité, Eds.), pp 72-79. Elsevier Applied Science Publishers, London.
Council for Agricultural Science and Technology (1976) Application of Sewage Sludge
to Cropland: Appraisal of Potential Hazards of the Heavy Metals to Plants and
Animals, EPA-430/9-76-013. Office of Water Program Operations, United States
Environmental Protection Agency, Washington, DC.
Daniel, W.W. (1978) Applied Nonparametric Statistics. Houghton Mifflin Company,
Boston.
Day, A.D., Ottman. M.J., Taylor, B.B., Pepper, I.L. and Swingle, R.S. (1988) Liquid
sludge as fertilizer for wheat. BioCycle 29, 60-61.
Day, A.D., Thompson, R.K. and Swingle, R.S. (1987) Effects of sewage sludge on yield
and quality of wheat grain and straw in an arid environment. Desert Plants 8, 104105, 142-143.
Department of Land, Air & Water Resources, University of California, Davis (1984) Irrigation with Reclaimed Municipal Wastewater: A Guidance Manual (G.S. Pettygrove and T. Asano, Eds.), Report Number 84-1 wr. California State Water
Resources Control Board, Sacramento.
Doran, J.W., Fraser, D.G., Culik, M.N. and Liebhardt, W.C. (1987) Influence of alternative and conventional agricultural management on soil microbial processes and nitrogen availability. American Journal of Alternative Agriculture 2,99-106.
Dutzler-Franz, G. (1977a) Der Einfluss einiger chemischer und physikalischer Bodenmerkm ale auf die Enzymaktivität verschiedener Bodentypen. Zeitschrft für Pflanzenernahrung und Bodenkunde 140, 329-350.
Dutzler-Franz, G. (1977b) Beziehungen zwischen der Enzymaktivitaten verschiedener
Bodentypen, der mikrobiellen Aktivität, der Wurzelmasse und einigen Klimafaktoren. Zeitschr ift für Pflanzenernahrung und Bodenlwnde 140, 351-374.
98
Duxbury, T. (1985) Ecological aspects of heavy metal responses in microorganisms. In
Advances in Microbial Ecology (K.C. Marshall, Ed), pp. 185-235. Plenum Press,
New York.
Epstein, E., Taylor, J.M. and Chaney, R.L. (1976) Effects of sewage sludge and sludge
compost applied to soil on some soil physical and chemical properties. Journal of
Environemtnal Quality 5,422-426.
Foiwell, A.P. (1922) Sewerage: The Designing, Constructing and Maintaining of Sewerage Systems and Sewage Treatment Plants. John Wiley & Sons, New York.
Forster, D.L., Logan, T.J., Pierzynski, G.M. and Southgate, D.D. (1986) Institutional constraints on residuals use. In Utilization, Treatment, and Disposal of Waste on
Land (E.C.A. Runge, Ed.), pp. 235-243. Soil Science Society of America, Inc.,
Madison.
Fraser, D.G., Doran, J.W., Sahs, W.W. and Lesoing, G.W. (1988) Soil microbial populations and activities under conventional and organic management. Journal of Environmental Quality 17, 585-590.
Fresquez, P.R., Francis, R.E. and Dennis, G.L. (1990) Sewage sludge effects on soil and
plant quality in a degraded, semiarid grassland. Journal of Environmental Quality
19, 324-329.
Fuller, W.H. (1983) Soil injection of sewage sludge for crop production, Technical Bulletin 250. University of Arizona Agricultural Experimental Station, University of
Arizona, Tucson.
Gee, G.W. and Bauder, J.W. (1986) Particle-size analysis. In Methods of Soil Analysis,
Part 1, Physical and Minerological Methods, 2nd edn (A. Kiute, Ed.), pp. 383411. American Society of Agronomy, Inc., Soil Science Society of America, Inc.,
Madison.
Ghurnan, G.S. (.1.985) Heavy Metal Composition of Treated Municipal Wastewater and
Sludge Residues, ERC 05-85, PB86-164191. Environmental Resources Center,
Georgia Institute of Technology, Atlanta.
Gibbons, J.D. (1985) Nonparametric Methods for Quantitative Analysis. American Sciences Press, Inc., Columbus.
Glathe, H. and Thalmann, A. (1970a) Uber die mikrobielle Aktivitat und ibre Beziehungen zu Bodenfruchtbarkeitsmerkmalen einiger Ackerböden unter besonderer
99
Berucksichtigung der Dehydrogenaseaktivitat (TTC-Reduktion)--1. Mitteilung:
Methodische Versuche zur Bestimmung der TTC-Reduktion. Zentralblatt für Bakteriologie, Parasitenlwnde, Jytektionskrankheiten und Hygiene: Abteilung 11Allgemiene, landwirtschafthiche und rechnische Mikrobiologie 124, 1-23.
Glathe, H. and Thalmann, A. (1970b) Uber die milcrobielle AktivitAt und ihre Beziehungen zu Bodenfruchtbarkeitsmerkmalen einiger Ackerböden unter besonderer
BerUcksichtigung der Dehydrogenaseaktivitat (T1'C-Reduktion)-2. Mitteilung:
Bestimmung der T1'C-Reduktion im Boden in Laboratoriumsversuchen. Zentralblati für Bakteriologie, Parasitenlwnde, lnfelaionskrankheiten und Hygiene: Abteilung IlAligemiene, landwirtschafthiche und technische Mikrobiologie 124,
24-35.
Glathe, H. and Thalmann, A. (1970c) Uber die mikrobielle Aktivität und ihre Beziehungen zu Bodenfruchtbarkeitsmerkmalen einiger Ackerböden unter besonderer
Berticksichtigung der Dehydrogenaseaktivitat (TTC-Reduktion)--3. Mitteilung:
Mikrobiologische Untersuchungen an Proben von Freilandversuchen auf BOden
mit unterschiedlichen Fruchtbarkeitsmerkmalen. Zentralbiart für Bakteriologie,
Parasitenkunde, I'fektionskrankheiten und Hygiene: Abteilung JIAllgemiene,
landwirtschafthiche und technische Mikrobiologie 124, 36-55.
Goldstein, J. (1976) Sensible Sludge: A New Look at a Wasted Natural Resource. Rodale
Press, Inc., Emmaus.
Grossbard, E. (1973) Problems of assessing the effect of pollutants on microbial activity.
Bulletins from the Ecological Research Committee (Stockholm) 17,457-463.
Higgins, A.J. (1984) Land application of sewage sludge with regard to cropping systems
and pollution potential. Journal of Environmental Quality 13,441-448.
IMSL (1987) User's Manual: Stat/Library, Fortran Subroutines for Statistical Analysis.
Version 1.0 IMSL, Inc., Houston.
Johnson, R.A. and Wichern, D.W. (1988) Applied Muluvariate Statistical Analysis. Prentice Hall, Englewood Cliffs.
Klein, D.A., Loh, T.C. and Goulding, R.L. (1971) A rapid procedure to evaluate the dehydrogenase activity of soils low in organic matter. Soil Biology and Biochemistry
3, 385-387.
Klute, A. (1986) Water retention: Laboratory methods. In Methods of Soil Analysis, Part
1, Physical and Minerological Methods, 2nd edn (A. Kiute, Ed.), pp. 635-662.
100
American Society of Agronomy, Inc., Soil Science Society of America, Inc.,
Madison.
KUster, E. and Williams, S.T. (1964) Selection of media for isolation of Streptomycetes,
Nature 202,928-929.
Lentner, M. and Bishop, T. (1986) Experimental Design and Analysis. Valley Book Company, Blacksburg.
Lindsay, W.L. and Norvell, W.A. (1978) Development of a DTPA soil test for zinc, iron,
manganese and copper. Soil Science Society of America Journal 42,421-428.
Martin, J.P. (1950) Use of acid, rose bengal, and streptomycin in the plate method for estimating soil fungi. Soil Science 69,215-232.
Martyniuk, S. and Wagner, G.H. (1978) Quantitative and qualitative examination of soil
microflora associated with different management systems. Soil Science 125, 343350.
McGrath, S.P, and Brookes, P.C. (1986) Effects of long-term sludge additions on microbial biomass and microbial processes in soil. In Factors Influencing Sludge Utilisation Practices in Europe, (R.D. Davis, H. Haini, and P. L'Hermité, Eds.), pp
80-88. Elsevier Applied Science Publishers, London.
Metzger, L., Levanon, D. and Mingelgrin, U. (1987) The effect of sewage sludge on soil
structural stability: Microbiological aspects. Soil Science Society of America Journal 51, 346-35 1.
Nannipieri, P., Johnson, R.L. and Paul, E.A. (1978) Criteria for measurement of microbial growth and activity in soil. Soil Biology and Biochemistry 10,223-229.
Neilson, J.W. and Pepper, I.L. (1990) Soil respiration as an index of soil aeration. Soil
Science Society of America Journal 54,428-432.
O'Connor, G.A., Knudtsen, K.L. and Connell, G.A. (1986) Phosphorus solubility in
sludge-amended calcareous soils. Journal of Environmental Quality 15, 308-3 12.
Ottman, M.J., Pepper, I.L., Artiola, J.F. and Taylor, B.B. (1989) Sewage sludge effects
on soil properties, nitrogen availability, and yield at Marana, 1989. In Cotton: A
College of Agriculture Report. (J. Silvertooth and M. Bantlin, Eds.), pp. 175-180.
Office of Agricultural Sciences Communications, University of Arizona, Tucson.
101
Paul, E.A. and Clark, F.E. (1989) Soil Microbiology and Biochemistry. Academic Press,
San Diego.
Pepper, IL , Bezdicek, D.F., Baker, A.S. and Sims, J.M. (1983) Silage corn uptake of
sludge-applied zinc and cadmium as affected by soil pH. Journal of Environmental Quality 12,270-275.
Pietz, R.I., Carlson, C.R., Jr., Peterson, J.R., Zenz, D.R. and Lue-Hing, C. (1989a) Application of sewage sludge and other amendments to coal refuse material: I. Effects
on chemical composition. Journal of Environmental Quality 18, 164-169.
Pietz, R.I., Carlson, C.R., Jr., Peterson, J.R., Zenz, D.R. and Lue-Hing, C. (1989b) Application of sewage sludge and other amendments to coal refuse material: II. Effects
on revegetation. Journal of Environmental Quality 18, 169-173.
Post, D.F., Hendricks, D.M. and Pereira, O.J. (1978) Soils of the University of Arizona
Experiment Station: Marana. Agriculture Engineering and Soil Science Report
78-1. College of Agriculture, University of Arizona, Tucson.
Premi, P.R. and Cornfield, A.H. (1971) Incubation study of nitrogen mineralisation in
soil treated with dried sewage sludge. Environmental Pollution 2, 1-5.
Quraishi, M.S .1. and Cornfield, A.H. (1971) Effects of addition of varying levels of copper, as oxide or phosphate, on nitrogen mineralisation and nitrification during incubation of a slightly calcareous soil receiving dried blood. Plant and Soil 35,
5 1-55.
Rodriguez-Kabana, R. (1967) An improved method for assessing soil-fungus population
density. Plant and Soil 26, 393-396.
Rogers, J.E. and Li, S.W. (1985) Effect of metals and other inorganic ions on soil microbial activity: Soil dehydrogenase assay as a simple toxicity test. Bulletin of Environmental and Contamination Toxicology 34, 858-865.
Ross, D.J. (1973) Some enzyme and respiratory activities of tropical soils from New Hebrides. Soil Biology and Biochemistry 5, 559-567.
Ross, D.J. (1970) Effects of storage on dehydrogenase activities of soils. Soil Biology
and Biochemistry 2, 55-61.
Roszak, D.B. and Colwell, R.R. (1987) Survival strategies of bacteria in the natural environment. Microbiological Reviews 51, 365-379.
102
SAS (1990) SAS Procedures Guide: SASIS TAT User's Guide. Version 6, 4th edn. SAS
Institute, Inc., Cary.
SAS (1989) SAS Language and Procedures: Usage. Version 6, 1st edn. SAS Institute,
Inc., Cary.
Schachtschabel, P., Blume, H.-P., Hartge, K.H., Schwertman, U., Brummer, G. and Renger, M. (1984) Lehrbuch der Bodenkunde, 11th edn. Ferdinand Enke Verlag,
Stuttgart.
Schmidt, EL. and Paul, E.A. (1982) Microscopic methods for soil microorganisms. In
Methods of Soil Analysis, Part 2, Chemical and Microbiological Properties, 2nd
edn (A.L. Page, R.H. Miller and D.R. Keeney, Eds.), pp. 803-814. American Society of Agronomy, Inc., Soil Science of America, Inc., Madison.
Schofield, R.K. and Taylor, A.W. (1955) The measurement of soil pH. Soil Science Society Proceedings 19, 164-167.
Schuller, E. (1989) Fluoreszenzmikroskopisch bestimmte mikrobielle Biomasse und
deren Beziehung zum Gesamtkohlenstoffgehalt und zur Dehydrogenaseaktivitat
in ausgewahlten Bodenproben. Ze itschrzft für Pflanzenernahrung und Bodenkunde 152, 115-120.
Seaker, E.M. and Sopper, W.E. (1988a) Municipal sludge for minespoil reclamation: I.
Effects on microbial populations and activity. Journal of Environmental Quality
17, 591-597.
Seaker, E.M. and Sopper, W.E. (1988b) Municipal sludge for minespoil reclamation: II.
Effects of organic matter. Journal of Environmental Quality 17, 598-602.
Skujins, J. (1976) Extracellular enzymes in soil. CRC Critical Reviews in Microbiology
4, 383-421.
Skujins, J.J. (1967) Enzymes in soil. Soil Biochemistry (A.D. McLaren and G.H. Peterson, Eds), pp. 371-414. Marcel Dekker, New York.
Sommers, L.E. and Barbarick, K.A. (1986) Constraints to land application of sewage
sludge. In Utilization, Treatment, and Disposal of Waste on Land (E.C.A. Runge,
Ed.), pp. 193-216. Soil Science Society of America, Inc., Madison.
SPSS (1988) SPSS-X User's Guide, 3rd edn. SPSS, Inc., Chicago.
103
Starkey, R.L. (1924) Some observations on the decomposition of organic matter in soils.
Soil Science 17,293-314.
Stevenson, I.L. (1959) Dehydrogenase activity in soils. Canadian Journal of Microbiology 5,229-235.
Stotzky, G. (1965) Microbial respiration. In Methods of Soil Analysis, Part 2, Chemical
and Microbiological Properties (C.A. Black, Ed.), pp. 1550-1572. American Society of Agronomy, Inc., Madison.
Stotzky, G. and Norman, A.G. (1961) Factors limiting microbial activities in soilI. The
level of substrate, nitrogen, and phosphorous. Archiv für Mikrobiologie 40, 341369.
Stucky, D.J., Bauer, J.H. and Lindsey, T.C. (1980) Restoration of acidic mine spoils with
sewage sludge: I. Revegetation. Reclamation Review 3, 129-139.
Tarr, J.A. (1975) From city to farm: Urban wastes and the American Farmer. Agricultural History 49, 598-612.
Taylor, B.B., Ottman, M.J., Pepper, I.L., Artiola, J.F. and Gerba, C.P. (1989) Agricultural Sludge Reclamation Study: Final Report for 1989. Submitted to: Pima
County Wastewater Management Department. University of Arizona, Tucson.
Taylor, D.S. (1989) Sludge reuse aided by low-tech methods. BioCycle 30, 28-29.
Teutscher, H., Adler, R. and Seaton, J.P. (1960) The Soil and its Fertility. Reinhold Publishing Company, New York.
Troildenier, G. (1973) The use of fluorescence microscopy for counting soil microorganisms. B ulletins from the Ecological Research Committee (Stockholm) 17,53-59.
Troildenier, G. (1 972a) Fluoreszenzmikroskopische Zahlung von BodenbakterienI.
Historischer Uberblick und Beschreibung cines Verfahiens zur Zahlung von
Bodenbakterien in Trockenpraparaten nach Farbung mit Acridinorange. Zentralblatt für Bakteriologie, Parasiten/wnde, Infektionskrankheiten und Hygiene:
Zweitenaturwissenschaftliche--Abteilung 127,25-40.
Troildenier, G. (1972b) Fluoreszenzmikroskopische Zahiung von BodenbakterienII.
Biometrische Beirachtungen über em Verfahren zur Zahiung von Bodenbakterien
in Trockenpraparaten, Zentralblatt für Baicteriologie, Parasitenkunde, Infektionskrankheiten und Hygiene: Zweitenatur.vissenschafthiche--Abteilung 127,41-51.
104
Tsao, P.11., Leben, C. and Keitt, G.W. (1960) An enrichment method for isolating actinomycetes that produce diffusible antifungal antibiotics. Phytopatholoy 50, 88-89.
United States Environmental Protection Agency (1986) Methods of Analysis of Hazardous Solid Wastes, SW-846. 3rd edn. Office of Solid Waste and Emergency Response, Washington, DC.
United States Environmental Protection Agency (1983) Hazardous Waste Land Treatment, SW-874. Office of Solid Waste and Emergency Response, Washington, DC.
Verstraete, W. and Voets, J.P. (1977) Soil microbial and biochemical characteristics in relation to soil management and fertility. SoilBiology and Biochemistry 9, 253-258.
Waksman, S.A. (1922a) Microbiological analysis of soil as an index of soil fertility: I.
The mathematical interpretation of numbers of microorganisms in the soil. Soil
Science 14, 8 1-101.
Waksman, S.A. (1922b) Microbiological analysis of soil as an index of soil fertility: II.
Methods of the study of numbers of microorganisms in the soil. Soil Science 14,
283-298.
Waksman, S.A. (1922c) Microbiological analysis of soil as an index of soil fertility: III.
Influence of fertilization upon numbers of microorganisms in the soil. Soil Science 14, 32 1-346.
Waksman, S.A. (1923a) Microbiological analysis of soil as an index of soil fertility:
W,[sic] Ammonia accumulation (ammonification). Soil Science 15,49-65.
Waksman, S.A. (1923b) Microbiological analysis of soils as an index of soil fertility: V.
Methods for the study of nitrification. Soil Science 15,241-260.
Waksman, S.A. (1923c) Microbiological analysis of soil as an index of soil fertility: VI,
Nitrification. Soil Science 16, 55-67.
Waksman, S.A. and Heukelekian, 0. (1924) Microbiological analysis of soil as an index
of soil fertility: VIII. Decomposition of cellulose. Soil Science 17, 275-291.
Waksman, S.A. and Starkey, R.L. (1924) Microbiological analysis of soil as an index of
soil fertility: VII. Carbon dioxide evolution. Soil Science 17, 141-161.
105
Waksman, S.A. and Karunakar, P.D. (1924) Microbiological analysis of soil as an index
of soil fertility.[sic] IX. Nitrogen fixation and mannite decomposition. Soil Science 17, 379-393.
Watson, J.E., Pepper, I.L., Unger, M. and Fuller, W.H. (1985) Yields and leaf elemental
composition of cotton grown on sludge-amended soil. Journal of Environmental
Quality 14, 174-177.
Williams, S.T. and Davies, F.L. (1965) Use of antibiotics for selective isolation and enumeration of actinomycetes in soil. Journal of general Microbiology 38,251-261.
Williams, S.T. and Wellington, E.M.H. (1982) Actinomycetes. In Methods of SoilAnalysis, Part 2, Chemical and Microbiological Properties, 2nd edn (A.L. Page, R.H.
Miller and D.R. Keeney, Eds.), pp. 969-987. American Society of Agronomy,
Inc., Soil Science of America, Inc., Madison.
Zimmermann, R., Iturriaga, R. and Becker-Birck, J. (1978) Simultaneous determination
of the total number of aquatic bacteria and the number thereof involved in respiralion. Applied and Environmental Microbiology 36,926-935.
Zscheischler, J., Estler, M.C., Gross, F., Neumann, H. and Geissler, B. (1979) Mais: Anbau und seine Verwertung. DLG-Verlag-GmbH, Frankfurt am Main.
Was this manual useful for you? yes no
Thank you for your participation!

* Your assessment is very important for improving the work of artificial intelligence, which forms the content of this project

Download PDF

advertisement