Bạn đang xem bản rút gọn của tài liệu. Xem và tải ngay bản đầy đủ của tài liệu tại đây (19.23 MB, 371 trang )
era1 grasses, including Lolium prenne L., Dactylis glomerata L., and Kentucky
bluegrass (Poa pmtensis L.) (Slangen and Kerkhoff, 1984; Waddington et al.,
1989). This chapter is restricted to major food and fiber crops of the world: rice,
corn, wheat, grain sorghum, potato, sugarcane, and cotton.
The wet conditions that exist during rice production and the preference of rice
for NH,+-N over NO,-N suggest that the application of NIs with NH4+-or
NH,+-producing fertilizers, such as urea, would be a sound N management practice. Prasad et al. (1986) suggested the use of NP for increasing N efficiency in
rice. Field experiments were conducted with NP, AM, and ST (Lakhdive and Prasad, 1970; Reddy and Prasad, 1977) and these clearly showed that on rice soils
with high percolation rates, nitrification inhibitors can be usefully employed for
increased rice yields and N efficiency. Nitrification inhibitors were specifically
effective in reducing N losses under alternate wetting and drying conditions frequently encountered in rice fields (Rajale and Prasad, 1972). Thomas and Prasad
(1987) reported that for direct-seeded rice, NP-blended urea produced 4.7 mg
grain ha ' compared to 3.7 mg ha I with urea. However, under similar conditions, DCD showed no advantage (Sudhakara and Prasad, 1986b).
Results from experiments conducted at different centers in Japan showed that
ammonium sulfate treated with NP increased rice yield by 15-20% over untreated
ammonium sulfate (Nishihara and Tsunyoshi, 1968). Similarly, in field tests carried out with AM in Japan, they showed that yields of transplanted as well as
direct-seeded rice were increased by the use of 5-6 kg ha-' AM along with ammoniacal fertilizers.
In the United States, Wells (1976) reported rice grain yield increases from the
addition of 1.12 and 2.24 kg h a - ' of NP applied with 67 to 178 kg h a - ' of
preplant-applied urea-N. In another study in Arkansas and Louisiana, no increase
in yield due to NP was recorded in 1977, but in 1978 there was a positive grain
yield response to NP (Touchton and Boswell, 1980). In Louisiana, Patrick et al.
(1968) reported no advantage with NP for rice. Wells ef al. (1989) summarized
results with DCD from Arkansas, California, Louisiana, Mississippi, and Texas.
DCD delayed nitrification and tended to result in rice grain yield increases compared to urea-applied preplant without DCD in drill seeding. In water-seeded continuously flooded rice, using DCD was advantageous only if the flood was delayed
for more than 14 days after urea application. At the International Rice Research
Institute, application of prilled urea with 10 or 15%DCD during the final harrowing produced lowland rice yields comparable to those with split applied prilled
urea without DCD (De Datta, 1986).
Bains et al. (1971) reported the effectiveness of neem (A. indica Juss) seed
RAJENDRA PRASAD AND J. F. POWER
extract-treated urea for increasing rice yields and N efficiency. Reddy and Prasad
(1975) showed that the coating of urea with neem cake controlled nitrification for
a period of about 2 weeks and resulted in a significant increase in rice grain yield
over prilled urea. Prasad and Prasad (1980) reported increased rice yields and N
efficiency with neem cake-coated urea. These results have been confirmed by a
large number of workers in India (Budhar et al., 1987, 1991; Govindaswamy and
Kaliyappa, 1986; John et al., 1989; Joseph et al., 1990; Latha and Subramanian,
1986; Mishra et al., 1991; Prasad et ul., 1989; Singh et al., 1984, 1990a,b; Singh
and Singh, 1991; Velu et al., 1987). NIs have therefore a definite place in rice
culture, especially in conditions where N losses due to leaching and denitrification
The results of field experiments with corn in the eastern corn belt of the United
States (Nelson and Huber, 1980) illustrated that 70% of the trials in Indiana
showed increased yields with NIs (NP and Terrazole); the average corn yield increase from NI was 24 and 5.2% for fall-applied anhydrous ammonia and urea
liquid solution fertilizers, respectively. Yield increases were also obtained in Kentucky, Michigan, and southern Illinois but not in Wisconsin and northern Illinois.
Hergert and Wiese (1980), summarizing the results of experiments with NP in the
western corn belt of the United States, observed that the data obtained from Minnesota, Kansas, and Nebraska indicated the largest impact of NIs on irrigated
sandy soils, particularly where rainfallhrrigation provides excess water; the response of NIs on fine-textured soils was rather limited. From the results of later
experiments with DCD and NP in the north central states, Malzer et al. (1989)
also concluded that the greatest benefit for NI use was obtained on coarse-textured
soils; their results are shown in Table IV. The data in Table IV also suggest that
DCD was superior to NP when used with urea ammonium nitrate or urea. This
was further confirmed in a later study conducted on installed lysimeters at the
Herman Rosholt Bonanza Valley irrigation farm located in west central Minnesota
(Walters and Malzer, 1990a). The soil on the experimental site was an Estherville
sandy loam (Typic Hapludoll). The N1 treatment increased fertilizer use efficiency
only at the 90 kg N ha I rate when the leaching load was high. It was concluded
that incorporation of NI with moderate N rates coupled with conservative irrigation management should reduce the risk of yield loss and minimize nitrate movement to groundwater.
Results from experiments with NIs in the southeastern United States (Touchton
and Boswell, 1980; Frye et al., 1989) suggest limited benefits to corn from NIs
due to relatively high soil temperatures, which permit nitrification of fall-applied
ammonium-N during winter months, highly permeable coarse-textured soils, and
nitrate leaching from excessive winter and spring rainfall. No yield advantage
Relative Effectiveness of Dicyandiamide (DCD) and Nitrapyrin (NP) with Several Applications
and N Sources on the Corn Yield from Coarse-Textured Soils in the Midwest"."
No. of positive
"Adapted from Malzer et al. (1989).
"Data include all N rates at or below the optimum rate fertilization within each experiment (13
experimental site years).
'Spr. PP. spring preplant: SD/split, side dress or split N application.
"UAN, urea ammonium nitrate (28% N solution); AA, anhydrous ammonia.
'Significant at the 90% probability level.
with DCD was obtained in 22 comparisons in the mid-Atlantic region of the
United States (Fox and Bandel, 1989). In five comparisons there was a lowering
of corn yield with DCD only. The reduced yield in three of these was attributed to
increased NH, volatilization losses in the presence of DCD. Townsend and McRae
(1980), from Nova Scotia, Canada, also observed that except on light sandy soils,
no yield advantage was gained with NP.
Thus soil characteristics and the amount of precipitation received during the
crop-growing season may affect the response of NI. For example, Kapusta and
Varsa (1972), working on clay pan soils in Illinois known for losses due to denitrification, found a positive response in the first year which was characterized by
good precipitation, but not in the next year, which was drier. Similar results were
obtained in a 2-year study at New Delhi, India (Prasad and Turkhede, 1971).
Benefits of the NIs in corn production are therefore limited to coarse-textured soils
and in situations where excessive soil water leads to heavy N leaching.
NP or Dwell applied with urea, anhydrous ammonia, or urea ammonium nitrate
solution did not increase yield nor improve efficiency of N applied to grain sorghum during a period of 4 years (1976- 1979),even with supplementary irrigation
RAJENDRA PRASAD AND J. F. POWER
to promote leaching and/or denitrification (Westerman et al., 1981). Two tests on
grain sorghum were conducted in the coastal plain of Alabama on Dothan and
Norfolk sandy loam soils. In the first test (Touchton and Reeves, 1985), DCD
increased grain yields when applied at the 90-kg N ha-' rate in both 1982 and
1983. Yields with 90 kg N h a - ' and DCD were equal to yields with 134 kg N
ha - without DCD. In both years, conditions were favorable for N losses via
leaching and denitrification. In the second test (Frye et al., 1989), no increase in
yield was obtained with DCD. Mascagni and Helms (1989) also failed to obtain
an increase in the yield of grain sorghum with DCD or NP on a poorly drained
Sharkey sandy clay (Vertic Haplaquepts) or on a well-drained Herbert sandy loam
(Mesic Ochraqualfs) soil of Arkansas. Success with NIs on sorghum has been
Extensive studies in the United States with NP and DCD showed increased
yields of winter wheat due to NP in the Pacific Northwest (Washington, Idaho)
(Harrison et al., 1977; Papendick and Engibous, 1980). Greater yields of wheat
were obtafned with DCD in three of eight experiments in the mid-Atlantic region
(Maryland and Pennsylvania) (Fox and Bandel, 1989) and in one of four experiments in the north central states (Illinois and Indiana) (Harms, 1987). In the eastern part of the Midwest (Illinois, Kentucky, Michigan, Ohio, and Wisconsin) a
yield increase was on the order of 9.9 to 24% (Nelson and Huber, 1980; Shyilon
et al., 1984). Nitrification inhibitors were more effective in the southern part of
the Midwest due to higher rainfall and the associated nitrate leaching. In the western part of the Midwest (Nebraska, Kansas, Colorado, Minnesota), NIs were not
effective in increasing wheat yields due to the virtual absence of leaching of N
below the root zone (Hergert and Wiese, 1980). Little advantage with NP (Nelson
et al., 1977; Boswell et al., 1976) or with DCD (Frye et al., 1989) was obtained
in the southeastern states of the United States (Alabama, Virginia, Georgia, and
Increased wheat yields with NP were obtained in Alberta, Canada (Mahli and
Nyborg, 1978). Sommer and Rossig (1978) from Germany reported that injection
of NH,+-N and NP gave similar yields as obtained with a split application of N.
Lewis and Stefanson (1975) obtained no yield advantage with NP under field conditions in Australia. In a field experiment on a sandy loam soil at New Delhi
(Singh and Prasad, 1992). wheat yield with 80 kg N ha-' + DCD was greater
than that obtained with 120 kg N ha I without DCD (Fig. 4). Application of DCD
beyond 15% of N as DCD reduced wheat yield.
An increase of 4- 12% in grain yield of wheat due to neem cake-coated urea
compared to urea was obtained in India at Kanpur (Agarwal et al., 1980), Hissar
K g N ha1-*/. D C D - N
Figure 4. Effect of DCD on wheat grain yield. Total N applied (fertilizer and DCD). Adapted
from Singh and Prasad (1992).
(Bhatia et al., 1985), and Pusa (Prasad et al., 1986; Mishra et al., 1991). Success
with NIs in wheat in the United States has been mixed. Nitrification inhibitors are
effective in increasing wheat yields in the Pacific Northwest and the southern Midwest but not in the southeastern states and the western Midwest. The data from
other parts of the world are too limited.
In a 2-year study at New Delhi, India, Parashar ef al. (1980) found a significant
increase in cane yield with neem cake-coated or mixed urea at 75 kg N ha - I and
with NP applied with 150 kg N ha I . Furthermore, there was a significant residual
effect on a ratoon crop and 75 kg N ha I as NP-treated urea or neem cake-coated
urea produced almost the same yield as 150 kg N ha - applied as prilled urea
(Sharma et al., 1981). Singh er al. (1987) also found an increased cane yield with
neem cake-coated urea. Nitrification inhibitors could have a place in sugarcane
culture, but more field data are needed before a definite conclusion can be drawn.
On sandy loam soils of Michigan, no yield advantage was obtained with NP,
while the yield and number of marketable tubers increased with NP on Idaho soils
(Potter etal., 1971). Broadcast application of N,as urea, with spraying of inhibitor
RAJENDRA PRASAD AND J. F. POWER
(NP on terazole) followed by thoroughly mixing the compounds with the soil gave
some potato yield increases on Washington soils, whereas no effect was found
with band (row) application (Roberts, 1979). Hendrickson et al. (1978) found a
yield reduction and a decreased quality of tubers with up to 4.4 kg ha-l NP applied with ammonium sulfate and diammonium phosphate.
On a Plummer fine sand (Grossarenic Paleudults) at Hastings, Florida, application of 5-6 kg ha I DCD significantly increased the tuber yield in 1983 but not
in 1984 (Frye et al., 1989). Also, no increase in tuber yield was recorded due to
DCD at Gainsville, Florida.
On alluvial soils in Ludhiana (India), ammonium sulfate and calcium ammonium nitrate are superior to urea in the absence of NP, but urea treated with NP is
comparable to ammonium sulfate and is better than calcium ammonium nitrate
(Sahota and Singh, 1984). Treatment with NP increased N uptake and N recovery
by potato and decreased the optimum dose by 1 1-40 kg N ha - I. Increased potato
yields with neem cake-coated urea were found at Simla (Sharma etal., 1980) and
Palampur (Sharma et al., 1986).
Reeves and Touchton (1989) in pot culture studies found that cotton was sensitive to DCD. Although significant reductions in plant growth did not occur unless DCD exceeded rates normally applied, their results suggest a need for caution
when applying DCD to cotton. On a Norfolk sandy loam (Rhodic Paleudult) in
Alabama there was a tendency for yields to decline with DCD, while on a Decatur
silt loam in the same state and on a BeulahBosket very fine sandy loam (Typic
Dystrochrepthlollic Hapludalf) in Mississippi there was no significant increase
in cotton yield (Frye et al., 1989). However, in India, an increased yield of cotton
due to neem cake coating of urea was reported by several workers (Seshadri and
Prasad, 1979; Jain et al., 1982). Cotton seems to be sensitive to DCD and therefore
this NI should not be used for cotton.
V. PHYTOTOXtCITY OF NIs
Some of the results obtained in field experiments could be due to phytotoxicity
of NIs, although obvious symptoms may not have appeared under field conditions
at dosages used. Joseph (1992) reported that wheat benefited when DCD was applied at a 10% N level, while yield was reduced when the level of DCD-N was
raised to 20%. Reeves and Touchton (1989) applied DCD at 0, 2.5,5, 10, 15, and
20 mg DCD N kg ~I soil along with urea or sodium nitrate at 50 mg N kg I soil
in a pot culture study with Norfolk sandy loam (Typic Paleudult). Six days after
application of DCD at 15 or 20 mg kg ’ soil, cotton leaves developed mottled
chlorosis. After 20 days, mottled chlorosis developed on leaves of all plants
treated with DCD. The chlorosis intensified with DCD rates and progressed to
necrosis with DCD-N rates of 20 mg kg I soil. Symptoms were similar for cotton
treated with both N sources. Reductions in leaf dry weight and foliar toxicity
symptoms suggested that the primary site of phytotoxicity of DCD was in leaf
tissue and not in root tissue. DCD linearly increased the leaf tissue concentrations
of N, P, and K and lowered concentrations of Ca” and Mg”. Lack of DCD x N
source interaction suggested that reduced Ca’ ’- and Mg uptake resulted from
direct effects of DCD and not from indirect effects caused by the inhibition of
nitrification and an increased NH,+ uptake. It was suggested that when banded N
applications are made or root growth is restricted due to compaction, phytotoxicity from DCD-N concentrations at 5 mg kg I in the root zone of cotton
might diminish any potential benefits derived from increased N efficiency gained
through the inhibition of nitrification.
In a greenhouse study with Cherry Belle radish, Feng and Barker (1989) found
that as the concentration of NP or Captan in the medium with NH,+-N increased,
growth of roots and shoots in radish was restricted and leaves were stunted, showing interveinal chlorosis, marginal necrosis, and upward cupping. The roots were
stunted and twisted and failed to expand properly. Ca” and Mg2‘ contents in
shoots, 4 weeks after seeding, were considerably lowered when NP was applied
with ammonium sulfate or urea; on the other hand, K contents were increased.
Many reports (Kirkby, 1968; Wilcox ef a/., 1973) using various plants have shown
that acidity of the medium and deficiencies of K + , Ca”, and Mg” are major
reasons for toxic effects of NH,’. However, Goyal et a/. ( 1982) observed that even
though the pH of the nutrient solution was regulated at or near neutrality, toxicity
persisted in radish plants; large amounts of K and Ca2+ in the solution did not
correct the toxicity.
Plants grown with ammonium fertilizer and NI usually contain lower concentrations of Ca” and Mg2+(English et al., 1980; Mathers et al., 1982). This tendency is attributed to competitive absorption between NH,’ and other cations.
English et al. (1980) suggested that chemical inhibition affects the permeability
of plant cell membranes by altering their integrity or activity. Ca2’ and Mg”
concentrations are correlated negatively to the residual NH,+ in the medium but
are correlated positively to residual nitrate. Plant weight is also negatively correlated with the residual NH,’.
Yield reductions and phytotoxicity from use of DCD have been reported by a
number of researchers (Cowie, 1918; Maftoun and Sheibany, 1979). Symptoms
of DCD phytotoxicity developed in the greenhouse within 3 to 20 days after application of DCD, depending on the crop and DCD rate (Reeves and Touchton,
1986). Symptoms expressed on corn and sorghum were chlorosis and necrosis that
RAJENDRA PRASAD AND J. F. POWER
began at the leaf tips and progressed down the leaf margin. Symptoms on other
crops were mottled interveinal chlorosis and leaf margin chlorosis and necrosis.
Based on visual symptoms, sorghum and cotton are more sensitive to DCD than
corn (Reeves and Touchton, 1986).
Concentrations as low as 2.5 mg DCD-N kg - I increased the stomatal conductance of water in cotton plants grown in the greenhouse (Reeves et al., 1988). This
effect was noted under conditions of high transpirational demand in the afternoon.
Concentrations of 5 - 10 mg DCD-N kg I increased responsiveness of stomata
to decreasing soil water content over the entire range of available soil water. The
effect of DCD on stomatal conductance was believed to be a direct effect of the
compound and not directly due to soil water availability. When soil water is limited, DCD might increase water stress and decrease yield (Frye et al., 1989).Use
of NI had a deleterious effect on the tuber grade in potato (Hendrickson er al.,
1978). Although total tuber yield increased, the percentage of grade A-USDA
tubers was reduced 2.4% with NP and 5. I % with DCD (Malzer et al., 1989).
The studies referred to earlier indicate the following: ( I ) Some field crops such
as cotton are sensitive to some NIs; and (2) phytotoxicity symptoms observed
could be due to direct or indirect effects of NIs; the indirect effects being the result
of higher than normal NH,+-N concentrations.
VI. HEALTH AND NITRATES
NIs may possibly play a role in human and animal health by reducing the NO,
content in drinking water, food, feed, and forage.
The well-known problem associated with NO3- /NO,- toxicity in humans is
methemoglobinemia or “blue-baby syndrome. It generally occurs when infants
under the age of 4 months consume too much nitrate (Rosenfield and Huston,
1950). Microbes in the stomach reduce nitrate to nitrite. When nitrites reach the
bloodstream, they convert ferrous ions in the hemoglobin to the ferric form and
produce methemoglobin (MHb), which has no oxygen-carrying capacity. Very
young children are susceptible because their hemoglobin has a greater affinity
for nitrite than hemoglobin of older children and adults. Methemoglobinemia
resulting from high nitrate concentrations in drinking water was first recognized
by Comly ( I 945) at the University of Iowa. Associated symptoms are diarrhea
and vomiting, and the child’s complexion becomes slate blue (Ewing and MayonWhite, 1951). In addition to drinking water, the incidence of methemoglobi”
nemia has occurred in young children fed unrefrigerated spinach or high nitratecontaining fruitjuices (WHO, 1978; Keating etal., 1973). In a survey of Nebraska
physicians, doctors reported 15 infants with suspected nitrate-induced methemoglobinemia (Grant, 1981). In addition to water and vegetable products, infant methemoglobinemia can occur when infant foods are prepared with nitratecontaminated water (Johnson et al., 1987). This may also happen in older
individuals who have genetically impaired enzyme systems for the reduction
of methemoglobin. The largest outbreak was reported in Hungary (Deak, 1985)
where 1353 cases occurred between 1976 and 1982.
Nitrite produced from NO,- could react in the stomach with secondary amines
resulting from the breakdown of meat and fish forming N-nitroso compounds,
which can cause stomach cancer (Fritsch and de Saint Blanquat, 1985; Saul et al.,
1981). However, it should be mentioned that nitrites which are a potential health
hazard are widely used as a preservative in salted meat and sausages (Davis, 1990)
where they prevent the growth of Clostridium botulinum, the organism that causes
botulism (WHO, 1978). Thus the risk of stomach cancer may not be closely linked
with the nitrate content in drinking water. In addition to methemoglobinemia and
stomach cancer, other health disorders reported due to the large ingestion of NO,
in drinking water are hypertension (Malberg et al., 1978), increased infant mortality (Super et al., 1981), central nervous system birth defects (Dorsch et al.,
1984), and non-Hodgkins lymphoma (Weisenburger, 199 1); nevertheless, none of
these have been conclusively proved to be due to NO,- ingestion (Spalding and
Exner, 1993). Normally, in humans only about 20% of their NO,- intake comes
from liquids and drinking water (Table V) (Isermann, 1983). In addition, overfertilization, heavy manuring, or irrigation with high NO3- water can also result in
Nitrate Uptake through Food and Drinks"
person - day
total daily intake
Meat and meat products
Oils and fats
Milk and dairy products
Vegetable\ (155 g day - I )
Drinks and water (2.75 liter day
"Adapted from Iserrnann (1983).
2 I .o
RAJENDRA PRASAD AND J. F. POWER
large NO,- accumulations in many vegetables, which increases the human nitrate load.
AND ANIMAL HEALTH
Nitrate or nitrite poisoning is also reported in animals and is again due to MHb
formation in blood with consequent asphyxiation. The conversion of nitrate to
nitrite is carried out by bacteria in the rumen and ruminants are therefore especially vulnerable to nitrate poisoning. Goats, especially Angora, may be more susceptible to NO,- poisoning than either sheep or cattle (Schneider et ul., 1990).
Mature single-stomach animals (except horses) are more resistant to nitrate toxicosis. Other than lack of oxygen, dilation of blood vessels is another secondary
effect of nitrate poisoning. Abdominal pain and diarrhea are also reported. Other
effects of nitratehitrite poisoning in animals include poor growth rates, reduced
milk production, increased susceptibility to infections, and even abortions late in
pregnancy (Schneider et al., 1990).
Nitratehitrite poisoning symptoms appear when MHb concentrations reach
20-30% of total hemoglobin, and death due to asphyxia may occur when the
MHb level exceeds 75% of total hemoglobin. Blood containing MHb usually has
a chocolate brown color.
Feed/forage with nitrate concentrations exceeding 2.25 g kg - I NO, - -N (1 0 g
kg - NO,-) have a high risk of causing acute nitrate poisoning in ruminants;
about half of this concentration should not be exceeded in the diets of pregnant
beef cows. Drinking water for young livestock should contain less than 35 mg
liter - I NO,--N. Nitratehitrite poisoning in adult animals is likely when the
N 0 3 - - N concentration in water is more than 100 mg liter-' (Schneider et ul.,
CONTENT OF DRINKING
There is growing concern regarding NO,- content in drinking water and the
World Health Organization (WHO) has set a maximum limit of 100 mg NO,liter-' (22.6 mg NO,--N liter-!) and a recommended limit of 50 mg N 0 3 - - N
liter I ( 1 1.3 mg N liter - I); the latter limit is also fixed by the Council of European
Groundwater is the source of domestic water for almost 90% of the rural population of the United States and for about 50% of the total population (Power and
Schepers, 1989). In Denmark, West Germany, The Netherlands, and Great Britain
the use of groundwater accounts for 99,73,70, and 30%, respectively, of the total
water consumption (Strebel et ul., 1989). Groundwater forms a substantial part
of the drinking water in other parts of the world also. In addition, groundwater
contributes substantially toward irrigation; estimates for the United States are
75-80% of the total water used for irrigation (Power and Schepers, 1989). Maintenance of groundwater quality is thus of major concern.
Nitrates in groundwater can originate from geological sources, precipitation,
cultivation, animal waste, niineralization of organic N, and fertilization. Data from
the U.S. Geological Survey and the Texas Department of Natural Resources over
a period of 25 years showed that states where 9% or more of groundwater samples
contained 10 mg N03--N liter-' (45 mg NO,- liter I ) or more were Arizona,
California, Delaware, Kansas, Minnesota, Nebraska, New York, Oklahoma, Rhode
Island, and Texas (Madison and Brunett, 1985). After a careful examination of the
U.S. Environmental Protection Agency's National Pesticide Survey (NPS), the
Monsanto Company's National Alachlor Well Water Survey, and state-wide surveys in Iowa, Kansas, Nebraska, North Carolina, Ohio, Texas, Arkansas, California, Delaware, Pennsylvania, Washington, Minnesota, and South Dakota, Spalding and Exner (1993) concluded that the highest incidence of contamination
occurs in groundwater in the middle of the contiguous United States where
NO, -N levels in ~ 2 0 %
or more of sampled wells in Iowa, Nebraska, and Kansas exceeded 10 mg liter I; in contrast, the contamination was lower in Texas,
North Carolina, and Ohio (Fig. 5). Power and Schepers (1989) observed that use
of high rates of fertilizer N may be a major source of nitrates in wells in the potatoproducing area of northern Maine. The high density of septic tanks, along with
application of fertilizers and manures on agricultural lands, probably contributed
to high NO3- on Long Island. Intensive dairy operations with associated problems
of manure disposal may be a primary source of nitrates in wells in southeast Penn~
Figure 5. Incidence of' NO,--N contamination in large selected surveys (number of counties
surveyed is in parentheses). IA, Iowa; KS, Kansas: NE. Nebraska: NC, North Carolina: OH, Ohio;
TX, Texas: AR, Arkansas: CA, California; DE. Delaware; PA. Pennsylvania; WA, Washington; MN.
Minnesota; SD, South Dakota. From Spalding and Exnrr (1993).