Plant and Soil 255: 333–341, 2003.© 2003 Kluwer Academic Publishers. Printed in the Netherlands.

333

Tomato root distribution, yield and fruit quality under subsurface dripirrigation

Rui M.A. Machado1,3, Maria do Rosario, G. Oliveira1 & Carlos A. M. Portas2

1Depart. de Fitotecnia, Universidade de Evora, Apartado 94 7002-554 Evora, Portugal. 2Instituto Superior deAgronomia, Tapada da Ajuda, 1349-017, Lisboa, Portugal. 3Corresponding author∗

Received 3 May 2002; accepted in revised form 18 November 2002

Key words: minirhizotron, processing tomato, root length intensity, subsurface drip irrigation

Abstract

Tomato rooting patterns were evaluated in a 2-year field trial where surface drip irrigation (R0) was compared withsubsurface drip irrigation at 20 cm (RI) and 40 cm (RII) depths. Pot-transplanted plants of two processing tomato,‘Brigade’ (C1) and ‘H3044’ (C2), were used. The behaviour of the root system in response to different irrigationtreatments was evaluated through minirhizotrons installed between two plants, in proximity of the plant row. Rootlength intensity (La), length of root per unit of minirhizotron surface area (cm cm−2) was measured at bloomingstage and at harvest. For all sampling dates the depth of the drip irrigation tube, the cultivar and the interactionbetween treatments did not significantly influence La. However differences between irrigation treatments wereobserved as root distribution along the soil profile and a large concentration of roots at the depth of the irrigationtubes was found. For both surface and subsurface drip irrigation and for both cultivars most of the root systemwas concentrated in the top 40 cm of the soil profile, where root length density ranged between 0.5 and 1.5 cmcm−3. Commercial yields (t ha−1) were 87.6 and 114.2 (R0), 107.5 and 128.1 (RI), 105.0 and 124.8 (RII), for1997 and 1998, respectively. Differences between the 2 years may be attributed to different climatic conditions.In the second year, although no significant differences were found among treatments, slightly higher values wereobserved with irrigation tubes at 20 cm depth. Fruit quality was not significantly affected by treatments or by theinteraction between irrigation tube depth and cultivar.

Abbreviations: CI – ‘Brigade’; CII – ‘H3044’; DAP – days after planting; La – root length intensity; R0 – surfacedrip irrigation; RI – irrigation tube at 20 cm depth; RII – irrigation tube at 40 cm depth;

Introduction

Drip irrigation in processing tomato is a common tech-nique in Portugal due to the Mediterranean climate,with dry and warm summers and high evapotranspir-ation rates throughout the growing season. These areconditions that make subsurface drip irrigation a suit-able alternative to the surface system. With subsurfacedrip irrigation, evaporation from the topsoil is reducedand water runoff is negligible (Phene, 1991; Pheneet al., 1992). In addition, with surface drip irrigation,roots grow preferentially around the emitter area (Oli-

∗ FAX No: +351-266-711163. E-mail: rmam@uevora.pt

veira et al., 1996), which in turn can contribute toimprove water availability to the plants when usingsubsurface drip irrigation. The purpose of the presentstudy was to compare surface vs subsurface drip irrig-ation (at two different depths) on the root distributionof two processing tomato cultivars. Knowledge ofrooting patterns is essential to irrigation and fertilisermanagement and consequently to tomato yield andquality. Besides using minirhizotrons for root systemanalysis, trenches were opened perpendicularly to theplant row to examine the root distribution along thesoil profile.

334

Table 1. Soil physical and chemical characteristics

Depth (cm)

0–40 41–74 75–100

Sand (%) 92.60 95.20 96.00

Silt (%) 1.70 2.00 1.60

Clay (%) 5.70 2.80 2.40

Bulk density (g cm−3) 1.51 1.60 1.64

Organic matter (%) 1.09 0.44 0.32

pH (H2O) 5.70 6.20 6.20

N (µg g−1) 3.45 2.05 2.18

P (µg g−1) 81.84 73.04 43.12

K (µg g−1) 69.72 68.06 78.02

Ca2+ (cmolckg−1) 0.30 0.12 0.11

Mg2+(cmolckg−1) 0.18 0.15 0.09

Table 2. Monthly rainfall and air temperature during the growingseason

Date Rainfall Temp. (◦C)

(mm) Max. Min.

1997

May 81.0 22.7 11.7

June 36.4 24.4 13.3

July 18.9 29.7 15.7

Aug. 0 25.7 14.9

1998

May 95.6 23.5 11.4

June 25.6 27.2 13.3

July 0 31.4 15.0

Aug. 0 33.0 15.6

Materials and methods

Experimental site

The experiment was conducted in 1997 and 1998 ona Regosol Soil (Typic Quarzipsamments) at the An-tónio Teixeira Research Station, in Coruche, Portugal.Soil characteristics and meteorological data duringthe experiment are summarised in Tables 1 and 2,respectively.

Experimental design and treatments

Three drip irrigation depths: (surface (R0), subsurfaceat 20 cm depth (RI) and subsurface at 40 cm depth

Table 3. Crop fertilisation (kg ha−1)

1997 1998

Preplant

N 30.0 33.3

P 22.0 20.4

K 172.6 141.9

Ca 95.9 95.8

Mg 6.2 6.2

S 38.4 33.12

Fertigation

N 102.0 91.3

P 31.7 26.0

K 216.1 191.7

Ca 35.1 36.9

(RII)) and two processing tomato cv, ‘Brigade’ (CI)and ‘H3044’ (CII), were arranged in a split-plot ex-perimental design with four replications. Drip depthwas defined as the main factor and the cultivar as thesecondary factor. Plot sizes were 1.5 × 10 m2, eachwith seven rows.

The daily volume of applied water was estimatedfrom ETm (minus rainfall) measured the day beforeirrigation. When rainfall exceeded the ETm value, ir-rigation was suspended and the exceeding water wasconsidered in the calculation of the subsequent ir-rigation volumes. ETm was estimated using the cropcoefficient (Kc) and the Penman Montheith referenceevapotranspiration (ETo) data from a nearby weatherstation (ETm = Kc·ETo). The crop coefficients used inthis work were average values established by Dooren-bos and Kassam (1986) for the following crop stages:0.75 for the development stage (from transplanting tobeginning of fruit set); 1.15 for the mid-season stage(from the beginning of fruit set to blooming) and 0.88for the late-season stage (from blooming to fruit ripen-ing, when 75% of the fruits were red or orange). Tominimise the effect of different irrigation treatmentson plantlets establishment, all plants were sprinklerirrigated at transplanting. Drip irrigation was started11 days after transplanting and ended when 75% offruits were red or orange. The total amount of waterapplied to the crop was 476.0 mm in 1997 and 523.4mm in 1998.

Soil preparation, preceding the installation of theirrigation tubes, consisted of a 40–50 cm deep mould-board ploughing followed by two 10–15 cm disc

335

harrow operations. Fertilisers were applied (Table 3)before transplanting (in a 15 cm band) directly belowthe row and by fertigation, starting on the third weekafter transplanting. Ca (NO3)2, KNO3 and H3PO4,

were applied three times per week via fertigation, ac-cording to the absorption rates estimated by Phene etal. (1986, 1987). Fertiliser concentrations in the irrig-ation water and the injection rate were calculated inorder to ensure that the electrical conductivity of thewater never exceeded 2.5 mS cm−1. RAM emitters(2.3 l h−1) were placed 40 cm apart.

Forty-day-old tomato seedlings were transplantedat 20 cm within the row and 150 cm between rows, fora total of 33 333 plants ha−1.

Root measurements

Root distribution was estimated at blooming (82 and75 DAP, for 1997 and 1998 respectively) and harvest(114 and 104 DAP, for 1997 and 1998, respectively)using 1.5 m long minirhizotrons with 5.2 cm diameter.One tube per treatment was installed, parallel to anddistant 10 cm from the plant row, between two plants,at an angle of 30 ◦ to the vertical. At 10 cm incrementsalong the tube, roots intersecting the minirhizotronwall were recorded with a 35 mm camera adapted toan endoscope. Root length intensity (La) was estim-ated (Tennant, 1975) and converted into root lengthdensity (Table 4) using the regressions defined in thecalibration procedure described in Machado and Oli-veira (2001). For statistical analysis values of rootlength intensity (cm cm−2) were transformed usingthe equation y = √

(x + 1). This transformation isrecommended when zero values are common in theoriginal data set (Underwood, 1981), as in the presentstudy.

Crop yield and quality

For yield and quality evaluation, all the fruits of plantsgrown in a 12 m2 area were hand harvested (114 and104 days after transplanting, in 1997 and 1998, re-spectively) when approximately 80% of the fruits werered or orange. Fruits were weighed after been sortedinto mature, green and rotten sub-groups for commer-cial yield evaluation. From the mature fruits subset,a sample of 2.5 kg was taken and passed through a0.8 mm mesh sieve, to separate seeds and epidermisfrom the juice. Soluble solids (◦Brix) and pH weremeasured in the homogenised juice.

Results and discussion

Root parameters

La data obtained in 1997 and 1998 only for somedepths were significantly different respect to the po-sition of the drip irrigation tube, cultivar or interac-tion between depth of the irrigation tube and cultivar(Tables 5–8). This outcome was probably associatedto the high variability of La values, as confirmedby the variation coefficients of the ANOVA, whichranged between 10 and 80%. Great spatial root vari-ability is associated with interaction between rootsystem structure and soil conditions and has been doc-umented by many authors (Brown and Scott, 1984;Hamblin, 1985; Oliveira et al., 2000; Upchurch,1987; Van Noordwijk, 1993; Zobel, 1991). Vari-ability may also be associated to the method used.Vos and Groenwold (1987) reported a coefficient ofvariation of 1.5–1.7 times greater for minirhizotronsthan for core sampling. However, the minirhizotronmethod allows a greater number of replications andis less labour-intensive than core sampling. Bar-Yosefet al. (1991), using soil coring, obtained similar res-ults when comparing subsurface versus surface dripirrigation and they also could not find an unequivocalrelationship between treatments and root distribution.Despite the lack of significant differences betweentreatments based on minirhizotron analysis, the distri-bution of the root system, in relation to the placementof the drip irrigation tube (Figure 1), observed on soilprofiles opened perpendicularly to the plant row, showdifferent rooting patterns among irrigation treatments.A large concentration of roots at the depth of the ir-rigation tubes was found, which is in agreement withthe observations reported for tomato under subsurfacedrip irrigation by Bar-Yosef et al. (1991), for maize byMitchell (1981) and Phene et al. (1991) and for cottonby Plaut et al. (1996). By using drip irrigation tubes at40 cm depth (RII) the vertically growing seminal rootsquickly reached the moist soil area near the emitter,where a large number of fine roots were found (Figure1). This result is consistent with the rooting patternsdescribed for processing tomato by Portas (1984), whospecifically identified a two-phase response for rootgrowth. In the first phase, vertical growth of seminalroots, with support and reserve functions, occurs. Sub-sequently (phase II) fine roots (Ø < 1 mm) develop toensure water and nutrient uptake.

The need to insert minirhizotrons at 10 cm from therow, in order not to damage the drip irrigation tubes,

336

Table 4. Relation between root length density (RLD) and root length intensity (La) (Machado and Oliveira, 2001)

Depth Equations n R2 r Sig.

(cm)

0–40 RLD = 0.4820 + 0.0113 La – 0.00003 La2 – 2 × 10−8La

3 24 0.576 0.76 ∗∗40–100 RLD = 0.1874 + 0.0045 La – 0.0002 La

2+ 9.5 ×10−7 La3 36 0.772 0.88 ∗∗∗

(∗∗ , ∗∗∗ Significant at P < 0.01 and 0.001 levels, respectively).

Table 5. La at different depths (82 DAP – blooming stage) (1997). data have been transformed

Depth (cm)

Treat. 0 – 10 10 – 20 20 – 30 30 – 40 40 – 50 50 – 60 60 – 70 70 – 80

Depth (R)

R0 (Surface) 8.11 8.49a 11.40a 8.18ab 12.17a 3.91 4.73 2. 10b

RI (20 cm) 6.17 6.68b 9.30c 12.05a 12.83a 6.75 5.84 3.12ab

RII (40 cm) 4.45 7.38ab 10.30b 7.56b 7.81b 5.02 4.33 2.16b

Cult. (C)

CI (‘Brigade’) 4.45 4.91 9.66 7.94 10.72 5.38 4.23 3.27

‘CII (‘H3044’) 8.03 10.12 10.88 10.58 9.80 5.07 5.71 3.42

F (R) 0.41NS 12.20∗ 50.31∗∗∗ 11.57∗ 30.95∗∗ 3.93NS 1.59NS 8.59∗F (C) 9.70NS 3.21NS 0.64NS 1.83NS 0.09NS 0.05NS 1.55NS 0.02NS

F (R×C) 1.94NS 0.67NS 1.59NS 0.42NS 0.16NS 0.22NS 0.26NS 0.03NS

Within each column means with different letters are significantly different. ∗, ∗∗, ∗∗∗ Significant at P < 0.05, 0.01 and 0.001 levels,respectively (LSD).

Table 6. La at different depths (75DAP – blooming stage) (1998): data have been transformed

Depth (cm)

Treat. 0 – 10 10 – 20 20 – 30 30 – 40 40 – 50 50 – 60 60 – 70

Depth (R)

R0 (Surface) 5.36a 4.60 5.11a 4.96 4.11 4.05 1.34

RI (20 cm) 2.67a 3.53 5.68a 7.40 4.15 6.90 5.18

RII (40 cm) 1.00b 2.35 1.00b 4.09 7.46 8.69 3.39

Cult. (C)

CI (‘Brigade’) 5.02a 5.98a 4.72 6.48 5.84 7.96 3.89

CII (‘H3044’) 1.00b 1.00b 3.13 4.48 4.64 5.13 2.73

F (R) 16.41∗ 5.31NS 30.36∗∗ 1.06NS 4.95NS 1.77NS 3.88NS

F (C) 30.18∗∗ 74.61∗∗∗ 1.25NS 1.23NS 0.43NS 4.93NS 0.37NS

F (R×C) 12.03∗∗ 5.08NS 1.23NS 1.42NS 0.08NS 2.39NS 0.44NS

Within each column means with different letters are significantly different. ∗, ∗∗, ∗∗∗Significant at P < 0.05, 0.01 and 0.001 levels,respectively (LSD).

could have hidden differences in root system beha-viour. At the same time, with subsurface drip irrigationsome roots growing along drip irrigation tubes wereobserved, taking advantage from the lower soil res-istance at the soil–tube interface and the water whichprobably flowed along it.

Root length density (RLD) along the soil profile,estimated from the regression equations defined us-ing the minirhizotron calibration reported in Machadoand Oliveira (2001) is displayed in Figures 2 and 3.Data for the interaction between the irrigation treat-ments and cultivars are only shown when there was asignificant effect on RLD. For both surface and sub-

337

Figure 1. Root distribution with subsurface drip irrigation at 20 cm (RI) and 40 cm (RII) depth.

Figure 2. Root length density at blooming stage for different irrigation treatments (Surface – R0, 20 cm - RI, 40 cm - RII, ‘Brigade’- C1,‘H3044’- CII). (RLD for irrigation × cultivar interaction are only shown when significant).

Figure 3. Root length density at harvest for different irrigation treatments (Surface – R0, 20 cm - RI, 40 cm - RII).

338

Table 7. La at different depths (114 DAP – at harvest) (1997): data have been transformed

Depth (cm)

Treat. 0 – 10 10 – 20 20 – 30 30 – 40 40 – 50 50 – 60 60 – 70 70 – 80 80 – 90 90 –100

Depth (R)

R0 (Surface) 7.01 9.13 10.30 7.89 11.45 4.28 5.82 5.42 5.49 3.70

RI (20 cm) 7.05 7.15 9.83 9.47 8.81 4.63 4.64 4.97 4.54 6.33

RII (40 cm) 6.45 8.96 9.55 8.99 7.59 5.57 3.44 3.84 3.80 3.42

Cult. (C)

CI (‘Brigade’) 6.29 7.55 10.28 7.93 10.76 4.81 4.42 4.62 4.78 4.65

CII (‘H3044’) 7.38 9.27 9.50 9.65 7.81 4.85 4.85 4.86 4.44 4.31

F (R) 0.06NS 0.62NS 0.08NS 1.84NS 0.98NS 0.36NS 1.99NS 0.72NS 1.27NS 4.36NS

F (C) 0.32NS 0.80NS 0.17NS 1.26NS 2.08NS 0.01NS 0.15NS 0.04NS 0.09NS 0.06NS

F (R×C) 0.36NS 1.54NS 1.05NS 2.97NS 1.94NS 0.24NS 0.25NS 0.99NS 0.41NS 0.58NS

Within each column means with different letters are significantly different (LSD).

Table 8. La at different depths (104 DAP – at harvest) (1998): data have been transformed

Depth (cm)

Treat. 0 – 10 10 – 20 20 – 30 30 – 40 40 – 50 50 – 60 60 – 70 70 – 80 80 – 90 90 – 100

Depth (R)

R0 (Surface) 2.79 6.78 6.99 5.42 4.27 7.87 8.35 3.58 2.79b 3.36b

RI (20 cm) 4.70 3.89 5.82 6.21 3.35 3.23 7.25 2.95 6.61a 7.04a

RII (40 cm) 4.84 3.37 5.72 5.44 6.74 4.61 4.74 5.74 8.18a 7.88a

Cult. (C)

CI (‘Brigade’) 5.64 5.90 8.49a 9.49a 8.57a 6.05 6.97 4.66 7.10a 5.91

CII (‘H3044’) 2.57 3.40 3.33b 1.90b 1.00b 4.38 6.58 3.52 4.62b 6.27

F (R) 2.26NS 3.12NS 0.06NS 0.11NS 2.25NS 4.52NS 2.21NS 0.48NS 7.06∗ 12.80∗F (C) 2.47NS 5.30NS 28.90∗∗ 20.74∗∗ 27.15∗∗ 2.64NS 0.03NS 2.47NS 6.63∗ 0.07NS

F (R×C) 2.05NS 3.07NS 0.05NS 1.61NS 1.93NS 10.39∗ 2.87NS 4.16NS 16.18∗∗ 22.74∗∗

Within each column means with different letters are significantly different. ∗, ∗∗Significant at P < 0.05 and 0.01 levels, respectively (LSD).

surface drip irrigation and for both cultivars most ofthe root system (Figure 2) was concentrated withinthe top 40 cm of the soil profile where root lengthdensity reached 0.5–1.5 cm cm−3. These results in-dicate that root growth occurs preferentially in the0–40-cm soil layer and is independent of the drip ir-rigation depth used. Similar results have been reportedby other authors for processing tomato under surfacedrip irrigation, using direct sowing (Bar-Yosef, 1977;Bar-Yosef et al., 1980; Maynard et al., 1980; Oliveiraet al., 1996; Sanders et al., 1989; West et al., 1979) ortransplanted plantlets (Machado et al., 2000).

Crop yield and fruit quality

The number of plants per ha was not significantly af-fected by the treatments or their interaction (Tables 9and 10). The commercial yield for both years (Tables9 and 10) was higher using subsurface compared tosurface irrigation. Commercial yields (t ha−1) were87.6 and 114.2 (R0), 107.5 and 128.1 (RI), 105.0 and124.8 (RII) for 1997 and 1998, respectively. Differ-ences between the 2 years can be attributed to differentclimatic conditions. These results are in agreementwith those obtained for tomato by other authors (Bar-Yosef et al., 1991; Davis et al., 1985; El-Gindy andEl-Araby, 1996; Hanson et al., 1997; Hutmacher etal., 1985; Phene et al., 1985, 1987). Camp (1998) re-

339

Table 9. Yield, ◦Brix and pH (1997)

Treatment Plants ha−1 Total yield Commercial ◦Brix pH

(t ha−1) yield (t ha−1)a

Depth (R)

R0 (surface) 31500 102.72 b 87.59 b 4.29 4.07

RI (20 cm) 31583 123.91 a 107.50 a 4.55 4.10

RII (40 cm) 31417 121.83 a 105.00 a 4.59 4.07

Cult. (C)

CI (‘Brigade’) 31444 111.71 95.83 4.76 4.06

CII (‘H3044’) 31556 120.59 104.23 4.20 4.10

F (R) 0.01NS 7.79∗ 5.17∗ 0.65NS 0.57NS

F (C) 0.03NS 3.57 NS 3.78 NS 4.48NS 2.75NS

F (R×C) 0.27NS 1.96 NS 2.14 NS 0.20NS 0.17NS

Within each column means with different letters are significantly different. ∗Significant at P < 0.05 (LSD).a Red and orange fruits.

Table 10. Yield, ◦Brix and pH (1998)

Treatment Plants ha−1 Total yield Commercial ◦Brix pH

(t ha−1) yield (t ha−1)a

Depth (R)

R0 (surface) 31691 128.12 114.16 5.36 4.38

RI (20 cm) 31274 136.87 128.10 5.09 4.36

RII (40 cm) 32500 133.80 124.80 5.25 4.37

Cult. (C)

CI (‘Brigade’) 31913 122.74b 114.93b 5.57 a 4.38

CII (‘H3044’) 31730 143.12a 129.76a 4.90 b 4.36

F (R) 2.15NS 2.23NS 2.30NS 0.37NS 0.30NS

F (C) 0.15NS 13.77∗∗ 10.62∗∗ 5.30∗ 1.12NS

F (R×C) 2.14NS 1.96NS 3.15NS 0.26NS 0.64NS

Within each column means with different letters are significantly different. ∗, ∗∗Significant at P < 0.05 and 0.01 levels, respectively (LSD).aRed and orange fruits.

ported that subsurface drip irrigation may enhance thecommercial yield compared to surface drip irrigation.

During 1997, with superficial irrigation (R0) thecrop often showed a phosphorus deficiency (purplecoloration on the leaf undersides) which could havecontributed to a lower yield in this year. Geisenbergand Stewart (1986) reported that with low soil temper-ature phosphorus absorption decreases and only highlevels of this nutrient near the root area can meet plantneeds. With subsurface drip irrigation phosphorus ap-plied deep in the profile enable the crop to utilise phos-phorus more efficiently as it was reported by Phene etal. (1986). These observations have to be tested furtherapplying different amounts of phosphorous.

The irrigation treatments did not significantly af-fect ◦Brix and pH of tomato fruits (Tables 9 and 10).This is in agreement with the observations of Daviset al. (1985), Phene et al. (1986, 1987) and Bar-Yosef et al. (1991), for the ◦Brix and of Davis et al.(1985) for the pH. The tomato juice pH was similar forthe two cultivars. For both years the ‘Brigade’ ◦Brixwas higher than ‘H3044’. ◦Brix values were higherin 1998, which can be attributed to better climaticconditions. According to Grierson and Kader (1986),sugar content is closely correlated with solar radiationduring fruit growth. Fruit dry matter content decreasesat low temperatures and low radiation levels (Castilla,1985).

340

Conclusions

Results of this 2-year field study indicate that the to-mato root system of the two cultivars, at differentirrigation depths, had the same behaviour. Roots con-centrated preferentially around the emitter area. So,using subsurface drip irrigation systems an accuratemanagement of irrigation and fertilisation is essentialto prevent high variations of water and nutrients in thesoil which could affect crop yield. At 10 cm from theplant row the depth of the irrigation tube did not affectroot length intensity.

The results also indicate that subsurface drip ir-rigation can contribute to increase the commercialproduction, as occurred during the first year, withoutaffecting fruit quality. The hypothesis of a differentcrop response to the level of water and phosphorusapplied with subsurface drip irrigation has to be tested.

References

Bar-Yosef B 1977 Trickle irrigation and fertilization of tomatoes insand dunes: water, N and P distributions in the soil and uptakeby plants. Agron. J. 69, 486–491.

Bar-Yosef B, Stammers C and Sagiv B 1980 Growth of trickle-irrigated tomato as related to rooting volume and uptake of Nand water. Agron. J. 72, 815–822.

Bar-Yosef B, Martinez H J J, Sagiv B, Levkovitch I, Markovitchand Phene C J 1991 Processing tomato response to surface andsubsurface drip phosphorus fertigation. Bard Project ScientificReport. pp. 175–191. Bet Dagan, Israel.

Brown D A and Scott H D 1984 Dependence of crop growthand yield on root development and activity. In Roots, Nutrientand Water Influx and Plant Growth. Eds. S A Barber and D RBouldin. pp. 101–135. ASA Special Publication 49.

Camp C R 1998 Subsurface drip irrigation: a review. Trans. ASAE41(5), 1353–1367.

Castilla N 1995 Manejo del cultivo intensivo con suelo. In El cultivodel tomate. Ed. F Nuez. pp. 190–225. Mundi-Prensa, Madrid.

Davis K R, Phene C J, McCormick R L, Hutmacher R B and MeekD W 1985 Trickle frequency and installation depth effects ontomatoes. Proc. Third Int. Drip/Trickle Irrigation Congress. pp.896–901. Fresno, CA.

Doorenbos J and Kassam A H 1986 Yield response to water. FAO,Irrigation and Drainage Paper, 33, Rome.

El-Gindy A M and El-Araby A M 1996 Vegetable crops to re-sponse to surface and subsurface drip under calcareous soil. Proc.Int. Conf. on Evapotranspiration and Irrigation Scheduling. pp.1021–1028. St. Joseph.

Geisenberg C and Stewart K 1986 Field crop management. In TheTomato Crop. Eds. J G Atherton and J Rudich. pp. 527–592.Chapman and Hall, New York.

Grierson D and Kader A A 1986 Fruit ripening and quality. In TheTomato Crop. Eds. J G Atherton and J Rudich. pp. 241–280.Chapman and Hall, New York.

Hamblin A 1985 The influence of soil structure on water movement,crop root growth, and water uptake. Adv. Agron. 38, 95–158.

Hanson B R, Schwankl L J, Schulbach K F and Pettygrove G S 1997A comparison of furrow, surface drip and subsurface drip irriga-tion on lettuce yield and applied water. Agric. Water Manage. 33(23), 139–157.

Hutmacher R B, Vail S S, Muthamia J G, Mwaja V and Liu R C1985 Effect of trickle irrigation frequency and installation depthon tomato growth and water status. Proc. Third Int. Drip/TrickleIrrigation Congress. pp. 798–803. Fresno, CA.

Machado R M A and Oliveira M R G 2001 The root spatial vari-ability effect on calibration of minirhizotron readings data inprocessing tomato. Proc. The 6th Symposium of the InternationalSociety for Root Research. pp. 538–539. Nagoya, Japan.

Machado R M A, Oliveira M R G and Portas C A M 2000 Effect ofdrip irrigation and fertilization on tomato rooting patterns. ActaHort. 537, 313–320.

Maynard D N, Lorenz O A and Magnifico V 1980 Growth andpotassium partitioning in tomato. J. Am. Soc. Hort. Sci. 105,79–82.

Mitchell W H 1981 Subsurface irrigation and fertilization of fieldcorn. Agron. J. 73 (6), 913–916.

Oliveira M R G, Calado A M and Portas C A M 1996 Tomato rootdistribution under drip irrigation. J. Am. Soc. Hort. Sci. 121(4),644–648

Oliveira M R G, Van Noordwijk M, Gaze S R, Brouwer G, Bona S,Mosca G and Hairiah K 2000 Auger sampling, ingrowth coresand pinboard methods. In Roots Methods: A handbook. Eds. AL Smit, A G Bengough, C Engles, M Van Noordwijk, S Pellerinand S C Van de Geijn. pp. 175–210. Springer, New York.

Phene C J 1991 Advances in irrigation under water shortage con-ditions. Proc. Conference on Collaborative Research and Devel-opment Applications in Arid Lands. pp. 93–110. Santa Barbara,California.

Phene C J, Bar-Yosef B, Hutmacher R B, Patton S H, Davis K R andMcCormick R L 1986 Fertilization of high yielding subsurfacetrickle irrigated tomatoes. Proc. 34th Annual California FertilizerConference. pp. 33–43. Fresno, CA.

Phene C J, Davis K R, Hutmacher R B and McCormick R L 1987Advantages of subsurface irrigation for processing tomatoes.Acta Hort. 200, 101–113.

Phene C J, Davis K R, Hutmacher R B, Bar-Yosef B, Meek D W andMisaki J 1991 Effect of high frequency surface and subsurfacedrip irrigation on root distribution of sweet corn. Irrigation Sci.12 (2), 135–140.

Phene C J, Hutmacher R B, Ayars J E, Davis K R, Mead R M andSchoneman R A 1992 Maximizing water use efficiency with sub-surface drip irrigation. International Summer Meeting, Charlotte,20 p.

Phene C J, McCormick R L, Miyamoto J M, Meek D W and DavisK R 1985 Evapotranspiration and crop coefficient of trickle irrig-ated tomatoes. Proc. Third Int. Drip/Trickle Irrigation Congress.pp. 823–831. Fresno, CA.

Plaut Z, Carmi A and Grava A 1996 Cotton root and shoot responsesto subsurface drip irrigation and partial wetting of the upper soilprofile. Irrigation Sci. 16 (3), 107–113.

Portas C A M 1984 Estudos del sistema radical de cultivos hortic-olas. XXIV Congresso Brasileiro e I Reunião Latino-Americanade Olericultura F.C.A.V. pp. 16–50. Jabotical, Brasil.

Sanders D C, Howell T A, Hile M M S, Hodges L and Phene CJ 1989 Tomato root development affected by travelling trickleirrigation rate. HortScience 24(6), 930–933.

Silva W L C, Giordano B, Marqueli W A, Fontes R R and Gornat B1998 Response of six processing tomato cultivars to subsurfacedrip fertigation. Acta Hort. 487, 569–573.

341

Tennant D 1975 A test of a modified line intersect method ofestimating root length. J. Ecology. 63, 995–1001.

Underwood A J 1981 Techniques of analysis of variance in exper-imental marine biology and ecology. Oceonagr. Mar. Biol. Ann.Ver. 19, 513–605.

Upchurch D R 1987 Conversion of minirhizotron-root intersec-tions to root length density. In Minirhizotron Observation Tubes:Methods and Applications for Measuring Rhizosphere Dynam-ics. Ed. H M Taylor. pp. 51–65. ASA Special Publication50.

Van Noordwijk M 1993 Roots: length, biomass, production andmortality. Methods for root research. In Tropical Soil and Bio-logy and Fertility, a handbook of methods. Eds. J M Andersonand J S I Ingram. pp. 132–144. CAB International, Wallingford.

Vos J and Groenwold J 1987 The relation between root growth alongobservation tubes and in bulk soil. In Minirhizotron ObservationTubes: Methods and Applications for Measuring RhizosphereDynamics. Ed. H M Taylor. pp. 39–49. ASA Special Publication50.

West D W, Merrigan I F, Jaylor J A and Collins G M 1979 Salinitygradients and growth of tomato plants under drip irrigation. SoilSci. 127(5), 281–291.

Zobel W R 1991 Genetic control of root systems. In Plant RootSystems. The hidden half. Eds. Y Waisel, A Eshel and U Kafkafi.pp. 27–33. New York.