Plant and Soil 125, 159-168 (1990). © Kluwer Academic Publishers. Printed in the Netherlands.
PLSO 8301
Nutrient uptake by potato crops grown on two soils with contrasting physical properties L.A. MACKIE-DAWSON, P. MILLARD and D. ROBINSON 1 Plants Division, Macaulay Land Use Research Institute, Craigiebuckler, Aberdeen, A B 9 2Q J, UK. 1present address: Department of Physiology and Crop Production, Scottish Crop Research Institute, Invergowrie, Dundee, DD2 5DA, UK Received 19 June 1989. Revised February 1990
Key words:
nutrient uptake, potato, Solanum tuberosum, root growth, soil type
Abstract
Potatoes were grown on two contrasting soils but in adjacent sites to investigate the effect of soil type on tuber production, nutrient uptake and nutrient inflow rates (uptake rate per unit length of root). The year of the study was wetter than normal. Tuber growth, root growth and nutrient uptake were all greater on the coarse rather than the fine-textured soil. However there was no difference in nutrient inflow rates between plants growing in the two soils. Therefore, it was concluded that the crop on the finer textured soil did not have an adequate nutrient supply, particularly of N, relative to the crop on the coarser-textured soil. The reasons for the low supply of nitrogen in the fine textured soil are not clear, but it might have been due to the smaller root system or to enhanced losses of nitrogen by denitrification caused by the combination of soil physical properties and poor drainage in a wet year.
Introduction
Although much is known about the effects of soil physical conditions on water uptake by field grown crops (Hamblin, 1985), little is known about their effects on nutrient uptake. Soil physical conditions (such as bulk density, particle size distribution, pore size distribution and soil strength) might well affect nutrient uptake in a number of ways. Soil structure and pore continuity can influence the hydraulic conductivity of the soil. This in turn would affect the supply of nutrients to roots via mass flow of soil solution. The supply by diffusion might be affected by the changing tortuosity of the diffusion pathways in the aqueous phase (Nye and Tinker, 1977). Structure might also have indirect effects on root metabolism (e.g. 0 2 supply, redox potential etc.) (Russell, 1977).
Physical impedance to the penetration of roots into pores (Russell, 1977) can lead to changes in root growth which might affect the length and spatial distribution of roots in each volume of soil. Resistance to root penetration has been shown by Goss (1977) to have no effect on nutrient uptake when barley seedlings were grown in glass ballotini and nutrient solution. Although root growth was reduced by the application of a high external pressure, the uptake rate per unit length of root (inflow) apparently increased to compensate, so to that the total uptake remained the same. Similar compensatory effects were found by Shierlaw and Alston (1984) in maize and ryegrass grown in soil in a growth room. In the field, although the external pressures experienced by roots can restrict root extension, the productivity of crops is often not affected, e.g. Ellis et al. (1977). Cannell et al. (1980)
160
Mackie-Dawson et al.
found that soil physical conditions can influence nutrient uptake and yield directly but unfortunately inflow rates were not measured during this study. Compacting a soil in the field has also been shown to improve nutrient uptake by the crop (Goldberg et al., 1983; Holmes et al., 1983). The field experiment reported here was located at a site with two adjacent areas of different soil types but with similar nutrient status. This enabled direct comparisons of nutrient uptake on two soils with contrasting soil physical properties, but with similar microclimates.
Materials and methods
Site description and experimental design The site was located at Upper Oldmill Farm, Turriff, Aberdeenshire, where two distinctly different soil types are located in the same field. The experimental plots (0.48 ha each) were at an altitude of 107 m above sea level. Plot A (National Grid Ref. NJ 735463) was on the east side of the field with a north north-westerly aspect, and plot B (NJ 734463) was on the west side with a north north-easterly aspect. Each plot was subdivided into eight subplots.
Soil survey The survey was primarily undertaken to assess the nature and variability of a number of soil characteristics, notably surface horizon depth and texture, presence and intensity of an indurated subsoil, soil drainage and parent material. Inspection holes were located on a 20 × 20 m grid over the two plots and 42 soil profile descriptions were recorded.
Instrumentation Porous cup mercury manometer tensiometers, as described by Berryman et al. (1976), were installed in each subplot, and readings were made weekly. Rainfall data was collected using a tipping bucket raingauge located 13 km away at Cross of Jackston (NJ 7503039). Less detailed rainfall measurements were made on site using a funnel raingauge and results agreed well with those at the Cross of Jackston site.
Crop growth Solanum tuberosum L. cv. Record was planted on 22 April 1985 in eight replicate plots on each soil type. Each plot was fertilised with 10 g N -2 -2 m as Nitrochalk, and 24 g P m and 12 g K -2 m as potassic superphosphate. Fertilisers were applied as basal dressings before ridging. An extra 10g N m -2 was applied to the crop at emergence. Weeds were controlled by a preemergence spray of paraquat. Blight and aphid infestations were controlled by regular sprays throughout the season. For growth analysis, four sampling areas, each consisting of 2 m of two adjacent rows, were designated in each plot. At each harvest, a subplot was selected randomly from each plot and the plants were lifted with a hand fork. Leaves and stems were sampled as described by Millard and Marshall (1986), while tuber samples were collected as described by Millard (1986). Roots The roots were studied by two methods. On one half of the subplots, roots were counted using the profile wall method (B6hm, 1979), and on the other half, root cores were collected. For the former measurements, soil pits were dug to the maximum depth of rooting at right angles to the crop row, a grid square (0.5 × 0.5 m) was placed on the smoothed face, and the number of exposed roots inside each square was recorded. Root counts were converted into the number of roots per m 2 for 50 mm depth intervals down the profile from the top of the ridge. Cores were collected using a metal corer (0.21 m 3 volume; 36.6mm internal radius, 50mm depth) with a sharpened bottom edge. Two cores were taken from along the top of a ridge in half of the subplots. One was taken adjacent to an experimental plant ('close' sample); the other ('away' sample) was adjacent to and at a random position on the outer circle. This allowed variation in root distribution with distance from the plant to be accounted for (Vos and Groenwold, 1986). Cores were collected to the maximum depth of rooting which varied with time. The soil cores were washed out by hand using a powered water jet, and a 0.5-mm mesh sieve. Root length was measured using a 'Comair' root-
Nutrient uptake by potato crops length scanning machine (Comair-Commonwealth Aircraft Corporation Ltd., Port Melbourne, Australia), which uses the line intercept principle suggested by Newman (1966). Root diameters were also measured on a subsample of twenty roots from each core using an optical microscope with micrometer eyepiece. Root lengths per core were converted to lengths per unit ground area by dividing the total root length throughout the profile by the crosssectional area of the core. These values were combined to give an average root length per unit area for each plant. This was done by weighting the values according to the proportion of the ground area around each plant sampled by the two cores; the ratio 'close': 'away' was 1:8.6. Allowance was made where necessary for the fact that the 'away' core may have contained roots from the adjacent guard plant in the row, assuming equal densities of roots of the two plants at the mid-point between them. An assumption was made that all the soil contained roots, and that the amount of soil available to each plant was inversely proportional to the planting density. Each plant would have had access to the soil below an area of 0.16 m 2.
Physical and chemical analyses Soil pH, exchangeable cations and particle size analysis (by hydrometer) was measured on dried and sieved (<2 mm) samples collected during the soil survey (Glentworth, 1954). Soil samples were also collected and analysed at sowing, midseason and in late August for N, P, K, Ca, Mg and pH. 1K mol m -3 KC1 extractions were prepared from the soil samples, using fresh soil for ammonium and nitrate analyses. Elemental contents were determined by the methods used for the plant material (below). Moisture release curves and bulk densities were also determined for the two soil types at these times. Moisture release characteristics were measured using a combination of a tension table (Childs, 1969) and a pressure plate (Richards, 1965). Soil strength was recorded at several times during the growing season using a Bush recording penetrometer (Anderson et al., 1980). Air-filled soil porosity was calculated from 1 b / s - 0 (Campbell, 1985, p. 7), where b is the dry bulk density and s is the density of soil
161
solids, assumed to be a constant value of 2.65 Mg m -3 (Marshall and Holmes, 1979, p. 10) and 0 is the volumetric water content. Total N contents of the plant material including the roots were determined by a micro-Kjeldahl method and digestion in conc. H2SO 4 after pretreatment with salicylic acid to reduce N O 3 N. For the other elements, samples were digested in conc. H2SO 4 and K measured by atomic emission spectroscopy, Ca, Na and Mg by atomic absorption spectrometry, and PO 4 was determined colorimetrically (Murphy and Riley, 1962). Soil pH was measured in soil suspensions in both H20 and CaC12 at a sample: solution ratio of 1 : 2 v/v.
Nutrient inflow rates Net inflow rates (I, p m o l m ~ s ~) of N, K, P, Mg and Cu at each of the harvests were calculated as I = 1/L A. (dX/dT) (see Hunt, 1979, p. 51). L A is the total root length per unit ground area (Table 4), X is the nutrient content mol m =2) of plant material (tops, roots and tubers) per unit ground area, and T is time after crop emergence. I was derived using Hunt and Parsons' (1974) growth analysis program, and represents an instantaneous rate of nutrient uptake per unit root length. Because L A w a s based on the total length of the root system recovered from the soil, and not on any subjective assessment of the proportion of the root system that might have been active in nutrient uptake, the values of I are averages for the whole root system. This is the conventional way of calculating I for field grown crops (see Barraclough, 1986, 1989; Nye and Tinker, 1977, p. 215; Robinson, 1986).
Results and discussion
Soil physical and chemical properties The soils on both sites are cultivated, weakly podzolic soils and belong to the Foudland (A) and Ordley (B) Associations. The former is developed on drifts derived from argillaceous schists, phyllites and slates, and is freely drained with a patchy indurated subsoil horizon at about
162
Mackie-Dawson et al.
0.40m depth. The Ordley soil is developed on mixed till derived from Old Red Sandstone sediments, notably conglomerate and micaceous sandstone and argillaceous schists and slates, is imperfectly drained, with no evidence of an indurated subsoil. The detailed soil survey of the site showed that the freely drained, stonier and more variable soils of the Foudland Association to the east (site A), differs considerably from the more uniform imperfectly drained soil of the Ordley Association to the west (site B). Table 1 shows the results of analyses of characteristic soil profiles of the two soil types. The Foudland soil (sandy silt loam) has a higher percentage sand and a lower percentage clay than the Ordley soil (sandy loam). Topsoil samples were also collected from the individual subplots for textural analyses. Groupings of the sand, silt and clay were transformed using an angular transformation, mean values calculated, and a t-test performed. Soil A was significantly different from soil B (at P < 0.001). The two sites did not differ in their contents of extractable P, K, Ca and Mg. These levels were at moderate (Scottish Agricultural Colleges advisory scale MISR/SAC 1988) throughout the whole growing season. The pH (H20) was lower on site B (5.0) than site A (5.5). Bulk density was higher in both the topsoil and subsoil of site B than in site A (Table 1). The moisture release curves (not shown) were broadly similar but at the higher moisture potential soil B had greater moisture contents than soil A. Seasonal changes in soil matric potential were
greatest in soil A (Fig. 1) with the soil draining to - 1 0 kPa at 50 cm depth for most of the season. In soil B (Fig. 1), the soil (below 0.15 m) rarely dried out beyond - 5 kPa. These results emphasise the better draining ability of the coarser-textured Foudland soil compared to the finer-textured Ordley soil. Soil A was always significantly (P < 0.001) drier than soil B. 1985 was a particularly wet year and evidence suggests that denitrification losses may have been high particularly in the imperfectly drained soil (Batey and Killham, 1986). Air-filled porosity was always higher in soil A than in soil B (Table 2). While the actual rate of denitrification depends on temperature and the availabilities of nitrate and carbon substrates, as well as on oxygen supply, these data show that soil B had a greater potential for denitrification losses than soil A. Colbourn, Harper and Iqbal (1984) measured considerable fluxes of nitrous oxide from a clay soil with an air-filled porosity of around 10%. Low concentrations of oxygen in soil can also restrict root growth directly although we have no data with which to test this possibility. The effect of low 0 2 on root growth interacts strongly with that of soil strength (Rye and Tinker, 1977), but the latter cannot explain the differences in root growth between the soil types. Soil strength was not significantly different in the topsoil of the two sites. Below 0.4 m soil strength was higher in soil A, probably due to the presence of subsoil induration at site A. Since roots were not found below 0.4 m in either soil, soil strength could have had little effect on differences in root production at the two sites.
Table 1. Some soil properties of sites A and B Horizon
Horizon Depth
Soil separates (%) Clay
Sand
Silt
pH (H20)
0-19 19-36 36-68
6.4 6.1 4.1
43.9 57.0 56.9
43.2 36.9 39.0
5.48 5.79 5.68
0.95 1.05 -
0-32 32-48÷
13.0 13.0
45.0 48.0
38.0 39.0
5.00 4.82
1.11 1.51
(cm)
Bulk density (g cm -3)
Foudland ( A ) Ap B(s) B(x)
Ordley ( B ) Ap B(g)
Nutrient uptake by potato crops •
I
nl
lC]I
D D
i l
0
Do
"C0o
D% []
5 0 I-
[]
,.o ° oD
D o*,
/•o l
D
100
150
o•
e~
n
~
°
oD
o
o°°~
0D o •BD
,+o~,.
o ao •
.~.
O 0
D aO
o%°~Q
•D
oo
D
D
D
0
0
D
D
o
D
0
D~
o
o
o
o
Oo
o
o o
o o o c
°o
o
• gD
oo
o
15cm SITE A
• • •
50 cm
o
15 c m
[] o
SITE B 50 cm
0 D
o
o
o
~
o•
<~eD
o
• • • I] ~ 0 •
go
II0
c
• • D [3 ~
oo
o@DO D 0
•
•
"° tOOOO i o • q:
o . ~-
o
o
250
I l II
liD
o
200
•
oO~ o o D
o eae°
MATRIC POTENTIAL
•
OD
I
t
(cm H20 )
I i • 0 •
i
163
RAINFALL ( c m ) 5
o
4
300 DD
3
350
2
400
0
1
!~
14
O"
14
28
42
56
70
lY4
D A Y S A F T E R EMERGENCE
Fig 1. Matric potential at 2 depths at sites A and B.
Table 2. Air filled porosities (%) of soils A and B at two depths Soil depth
(m)
Time (days after emergence) 24
38
52
80
16 12
27 18
17 6
3 5
Soil A 0.10-0.15 0.25-0.30
Soil B O. 10-0.15 0.25-0.30 a
7 0a
23 6
4 0a
3 0+
25% of the canopy dry matter, predominantly leaf material, had been lost. Tuber dry matter yields were greater on soil A than soil B at each harvest, while canopy growth was little affected by soil type (Table 3). Total dry matter production by crop A was greater than by crop B, throughout the season, but these differences were only found to be statistically significant at 38 and 52 DAE respectively.
Root growth
Soil saturated.
Crop growth Both crops reached 50% emergence on 17 June, and all harvests are subsequently referred to by the number of days from this date (Days after emergence, DAE). Tuber initiation had occurred by 24 DAE in both crops, while the maximum canopy dry weight was recorded 52 DAE (Table 3). Thereafter, the canopy senesced until the final harvest (80 DAE), by which time some
Considerably more roots were found in soil A than in soil B at all depths down the profile (Fig. 2) especially at the earlier harvests. This occurred at all sampling occasions throughout the season. The peak numbers of roots were between 0.05 and 0.15 m in both soil types. Root lengths and dry weights obtained from the soil cores were summed to give values per unit ground area (L A and L w respectively). Although L w and L A were generally higher in soil A, this was only significantly so at 53 DAE only (Table 4). Values recorded here are at the low end of
164 Table
Mu&k-Dawson
et al.
3. Dry weights of canopy, tubers and the total crop (excluding roots (pm-‘)
at sites A and B
Time (days after emergence)
Site
Canopy
Tuber
Total
24
A B SE (7df) Significance
111 103 8.0 NS
14 3 1.2 **
125 106 8.2 NS
38
A B SE (7 df) Significance
291 243 17.9 *
146 84 12.8 **
437 327 23.0 **
52
A B SE (7df) Significance
301 303 18.6 NS
652 576 16.2 **
954 879 27.9 *
80
A B SE (7df) Significance
230 224 16.0 NS
766 695 28.0 *
997 919 33.9 NS
** P
* P < 0.05, NS not significant. ROOT
NUMBER
I
100
SO.CM
DEPTH
Fig
2. Root distribution
assessed by the profile wall method at 31 days after emergence; * P 0.05, ** P 0.01. Soil A -O-,
Soil B
-a-. Tuble
4. Root dry weight (L,)
and length (L,),
Time (days after emergence)
Dry weight (g m-‘) LW Soil A
Soil B
4 28 38 52 80
38.5 40.4 35.6 38.7 16.3
36.6 37.7 29.4 20.7 19.4
** P cO.01, NS not significant.
per unit ground area in soils A and B Length (km m-‘) LA NS NS NS ** NS
Soil A
Soil B
2.3 3.3 2.8 3.7 2.5
2.6 2.1 2.1 2.3 1.8
NS NS NS ** NS
Nutrient uptake by potato crops ROOT 5 r
DEPTH
165
DENSITY (Kin m' ' 3 ) 10 i
15
(m)
0.1
0.2
j //
. / //
0.3 ~
/ ~ ;
/ :S
L
0.4
Fig 3. Mean seasonal values of root density (km m -3) assessed by root coring. Soil A - 0 - - , soil B - 0 - , plus and minus 1 standard error.
the range of values reported by Vos and Groenwold (1986) (3.4 to 7.1 km m-2), but they were studying a much deeper rooting crop. Root densities (Lv) in the upper 0.15m of soil were always greater (P < 0.05) in soil A than in soil B (Fig. 3). At lower depths, L v was similar in both soils. The greatest lengths of roots were found between 0.05 and 0.15 m below the top of the ridge, as found by Vos and Groenwold (1986). Their root len§th densities varied between 10 and 20km m- in the hill. These values are smaller than results given by Lesczynski and Tanner (1976) and Asfary et al. (1983), where values of mean densities of greater than 30 km m - 3 were reported. The lower root density in this work may be attributable to differences in the potato variety used. Measurements of root diameter showed that, on the first two sampling dates, root diameter was very similar for plants grown on both soils (Table 5). These mean values of diameter agree well with these reported by Vos and Groenwold (1986), where most of the roots were reported to have diameters less than 0.24 mm. At harvests
three and four, root diameter was significantly ( P < 0 . 0 5 ) greater on soil B than soil A. This may reflect the higher bulk density in site B compared to site A (Table 1). Diameters of axes and laterals of ryegrass and maize were reported by Shierlaw and Alston (1984) to increase when the soil in pots was compacted and the bulk -3 density increased from 1.2 to 1.75 g cm
Nutrient uptake The maximum nutrient uptake by both crops was measured at 52 DAE, when the uptake of N, P, K and Ca was significantly greater by crops grown on soil type A than on soil B (Table 6). After 52 DAE the amount of N, P and Mg in the crops remained constant, while there was a net loss of K and Ca (data not shown). Between 53 DAE and final harvest the nutrient content of the canopy of both crops decreased due to both leaf abscission during canopy senescence and the remobilisation of nutrients for tuber growth (Millard and MacKerron, 1986).
166
Mackie-Dawson et al.
Table 5. Root diameter (mm) at four harvest dates Soil A
Soil B
Significance
0.26 0.073
0.27 0.024
NS
SEM (7df)
0.23 0.017
0.26 0.010
NS
SEM (7dr)
SEM (7dr)
0.25 0.014
0.32 0.028
SEM (7df)
0.26 0.053
0.43 0.040
Time (days after emergence) 24
38
52
80 * P < 0 . 0 5 , NS not significant.
Table 6. Maximum uptake (tops, tubers and roots), g m -2, at 52 DAE, measured for crops grown on soil types A and B Nutrient
Soil A
Soil B
S.E.D.
Significance
N P K Ca Mg
23.7 1.8 32.9 8.9 1.4
14.5 1.5 22.4 5.4 1.0
1.76 0.11 2.47 1.06 0.18
* * * * NS
* P < 0.05, NS not significant.
Nutrient inflow rates
There were no differences in the net inflow rates of N, K, P, Ca and Mg between plants growing on soils A and B at any of the harvests (Fig. 4). The seasonal pattern of nutrient inflow was the same for each nutrient. Maximum values of I were measured 38 DAE apart for N, where on site A maximum inflow rates were measured at 52 DAE. This mid-summer peak in I is similar to that found for potatoes by Asfary et al. (1983), for winter wheat by Gregory et al. (1979) and Barraclough (1986) and for oilseed rape by Barraclough (1989). In contrast, Mengel and Barber (1984) found that in maize I declined gradually during the season. It has been shown previously that the roots of a potato crop retain a considerable physiological capacity to take up N late on in the season
(Robinson and Millard, 1987). The reduction in I in the latter half of the season cannot be related only to the ageing of the root system. It is likely also to be a reflection of the rapid depletion of available nutrients in the upper layers of the soil (e.g. Armstrong et al., 1986; Millard and Robinson 1990). In the present work, I was found to decline below zero at 64 DAE. This does not imply that nutrient ions were being lost from roots in an efflux process, but that they were being lost from the crop as a whole by the end of the experiment, due to leaf and root senescence. This, together with the small changes in the sizes of the root systems that were measured, led to the generation of negative values of I at the final harvest. Similar patterns have been reported by Mengel and Barber (1974) and by Gregory et al. (1979).
Nutrient uptake by potato crops
167
40 N
K
P
Ca
Mg
200
30 P ,Ca AND Mg
N AND K INFLOW
20INFLOW
RATE
RATE
1OO
10 ( p m o l
( p m o l m .I IS- i )
24 38 52
~1~
24 3 8
52
~0
24 38
52
80
2 4 3 8 52
\~0
24 38 52
m" l s " l
80 --10
-
100 -20
TIME
( DAYS AFTER
EMERGENCE
)
Fig. 4. Inflow rates (pmol m-~s -1) of N, P, K, Ca and Mg at site A --O- and site B ---0-, plus the standard error of the difference between two means.
Conclusions
Acknowledgements
Tuber growth, root growth and nutrient uptake were all affected by soil type, and all were greater in soil A than in soil B. However, nutrient inflows were not affected by soil type. Soil type probably restricted nutrient uptake from soil B by reducing root growth. Plants on soil B did not compensate by increasing inflow rates as was noted by Shierlaw and Alston (1984) and Goss (1977); in these studies modification of root form had no effect on nutrient uptake, provided that an adequate supply of nutrients was available. In this experiment, although both soils contained similar moderate levels of extractable major nutrients throughout the year, the plants at site B were not able to compensate for a modified root system by increasing inflow rates. Plant uptake of N was less on soil B than soil A. This could have been due to losses of N through denitrification, as it was a particularly wet year and soil B had a lower air-filled porosity throughout the experiment than soil A, demonstrating a greater potential for denitrification losses. In addition, the reduction in root growth in soil B could have been due to the lower air-filled porosity producing anaerobic zones and so reducing root development.
The authors thank A J Nolan for his soil survey and description of the trial plot site and to the Department of Mineral Soils for characterisation data, Mr H Aykroyd for permission to set the experiment up in his potato crop, Miss J Cooper, SASS for her help with the statistics and L Webster and C Wilkinson for assistance with the experiment.
References Anderson G, Pidgeon J D, Spencer H B and Parks R 1980 A new hand-held recording penetrometer for soil studies. J. Soil Sci. 32, 279-296. Armstrong M J, Milford G F J, Pocock T O, Last P J and Day W 1986 The dynamics of nitrogen uptake and its remobilisation during the growth of sugar beet. J. Agric. Sci. 197, 145-154. Asfary A F, Wild A and Harris P M 1983 Growth, mineral nutrition and water use by potato crops. J. Agric. Sci. 100, 87-101. Batey T and Killham K 1986 Field evidence on nitrogen losses by denitrification. Soil Use Management 2, 83-86. Barraclough P B 1986 Growth and activity of winter wheat roots in the field: Nutrient inflows of high-yielding crops. J. Agric. Sci. 106, 53-59. Barraclough P B 1989 Root growth, macro-nutrient uptake
168
Nutrient uptake by potato crops
dynamics and soil fertility requirements of a high yield winter oilseed rape crop. Plant and Soil 119, 59-70. Berryman C, Thorburn A A and Trafford B D 1976 The use of water tensiometers. FDEU Tech Bull, 76/7. B r h m W 1979 Methods of studying root systems. Ecol. Stud. 33. Springer, Berlin. Campbell G S 1985 Soil Physics with BASIC: Transport Models for Soil-plant Systems. Elsevier, Amsterdam. Cannell R Q, Ellis F B, Christian D G, Graham J P and Douglas J T 1980 The growth and yield of winter cereals after direct drilling, shallow cultivation and ploughing on non-calcareous clay soils, 1974-8. J. Agric. Sci. Camb. 94, 345-359. Colbourn P, Harper I W and Iqbal M M 1984 Denitrification losses from 15N-labelled calcium nitrate fertilizer in a clay soil in the field. J. Soil Sci. 35, 539-547. Childs E C 1969 An Introduction to the Physical Basis of Soil Water Phenomena. J. Wiley, London. Ellis F B, Elliott J G, Barnes B T and Howse K R 1977 Comparison of direct drilling, reduced cultivation and ploughing on the growth of cereals. 2. Spring barley on a sandy loam soil: Physical conditions and root growth. J. Agric. Sci. Camb. 89, 631-642. Glentworth R 1954 The soils of the country round Banff, Huntly and Turriff: Memoirs of the soil survey of Great Britain. HMSO. Goldberg S P, Smith K A and Holmes J C 1983 The effect of soil compaction, form of nitrogen fertilizer and fertilizer placement on the availability of manganese to barley J. Sc. Fd Agric. 34, 657-670. Goss M J 1977 Effects of mechanical impedance on root growth in barley (Hordeum vulgare L.). 1. The effects on the elongation and branching of seminar root axes. J. Exp. Bot. 28, 96-111. Gregory P J, Crawford D V and McGowan M 1979 Nutrient relations of winter wheat. 2. Movement of nutrients to the root and their uptake. J. Agric. Sci. 93, 495-504. Hamblin A P 1985 The influence of soil structure on water movement, crop root growth and water uptake. Adv. Agron. 38, 95-158. Holmes J C, Donald A H, Chapman W , Lang R W, Smith K A and Franklin M F 1983 Effects of soil compaction, seed depth, form of nitrogen fertilizer, fertilizer placement and manganese availability on barley. J. Sci. Fd Agric. 34, 671-684. Hunt R 1978 Plant Growth analysis. Edward Arnold, London. Hunt R and Parsons I T 1974 A computer program for deriving growth functions in plant growth analysis. J. Appl. Ecol. 11, 297-307.
Lesczynski D B and Tanner C B 1976 Seasonal variation of root distribution of irrigated, field grown Russel Burbank potato. Am. Potato J. 53, 69-78. Marshall T J and Holmes J W 1979 Soil Physics. Cambridge University Press, Cambridge. Macaulay Institute for Soil Research and Scottish Agricultural Colleges 1985 Advisory soil analysis and interpretation. Bulletin 1. Mengel D B and Barber S A 1974 Rate of nutrient uptake per unit of corn root under field conditions. Agron. J. 66, 399-402. Millard P 1986 The nitrogen content of potato (Solanum tuberosurn L.) tubers in relation to nitrogen application: The effect on amino acid composition and yields. J Sci. Fd Agric. 37, 107-114. Millard P and MacKerron D K L 1986 The effects of nitrogen application on growth and nitrogen distribution within the potato canopy. Ann. Appl. Biol. 109, 427-437. Millard P and Marshall B 1986 Growth, nitrogen uptake and partitioning within the potato (Solanum tuberosum L.) crop in relation to nitrogen application. J. Agric. Sci. 107, 421-429. Millard P and Robinson D 1990 Effect of the timing and rate of nitrogen fertilizer on the growth and recovery of fertilizer nitrogen with the potato (Solanum tuberosum L.) Crop. Fert. Res. 21, 133-140. Murphy J and Riley J P 1962 A modified single solution method for the determination of phosphate in natural waters. Anal. Chim. Acta 27, 31-36. Newman E I 1966 A method of estimating the total length of root in a sample. J. Ecol. 3, 139-145. Nye P H and Tinker P B 1977 Solute Movement in the Soil-Root System. Blackwell, Oxford. Richards L A 1965 Physical conduction of water in soil. In Methods of Soil Analysis. Eds. C Black, D Evans, J White, L Ensminger and F Clark. pp 128-152. Agronomy 9, Madison, WI. Robinson D and Millard P 1987 Short-term uptake rates and partitioning of nitrogen in a potato crop. J. Exp. Bot. 38, 841-848. Robinson D 1986 Limits to nutrient inflow rates in roots and root systems. Physiol. Plant. 68, 551-559. Russell R S 1977 Plant Root Systems: Their Function and Interactions with the Soil. McGraw-Hill, Maidenhead. Shierlaw J and Alston A M 1984 Effect of soil compaction on root growth and uptake of phosphorus. Plant and Soil 77, 15-28. Vos J and Groenwold J 1986 Root growth of potato crops on a marine-clay soil. Plant and Soil 94, 17-33.