Effect of Spray Application of KN03, Urea and GrowthRegulators on the Yield of Tommy Atkins Mango

S.A. OosthuyseMerensky Technological SeNices, P.O. Box 14, Duivelskloof 0835

ABSTRACT

The effects of foliar spray application of KN03, low biuret urea, GA3 (a gibberellin), CPPU (a syntheticcytokinin) and NAA (a synthetic auxin) on fruit retention, average fruit weight and yield at harvest, andmonetary return taking currently obtained prices into account were evaluated. Applications were madewhile the trees were in flower or subsequently, just prior to the commencement of fruit drop. The effectof a foliar application of Wuxal® -boron at flowering and of panicle pruning were additionally evaluated.Of the treatments applied during flowering, KN03 application was the only treatment to noticeablyincrease fruit retention, average fruit mass, yield and monetary return. Of the treatments applied afterflowering, application of CPPU + GA3, NAA, or NAA + GA3 noticeably increased fruit retention, yieldand monetary return. Increases in return realized were in the order of 19 to 33%.

UITTREKSEL

Die effekte van KN03, lae biuret ureum, GA3 ('n gibberelien), CPPU ('n sintetiese sitikinien) en NAA ('nsintetiese ouksien), deur middel van blaar sproei toediening, op vrugval, gemiddelde vrugmassa enopbrengs met oestyd en die inkomste, met inagneming van die huidige pryse wat behaal word, isgeevalueer. Toedienings is gemaak terwyl die bome in die blomstadium was en ook net voordat vrugvalvoorgekom het. Die effek van 'n blaartoediening van Wuxal® -boor tydens blom en vrugsteel snoeiingis ook geevalueer. Van die behandelings wat tydens blom toegedien is het net KN03 toediening 'nmerkbare vermindering op vrugval gehad het en 'n vermeerdering in gemiddelde vrugmassa, opbrengsen inkomste. Van die behandelings toegedien na blomstadium en net voor die aanvang van vrugval hetdie toediening van CPPU + GA3, NAA en NAA + GA3 vrugval merkbaar verminder en opbrengs eninkomste verhoog. Verhogings in inkomste was in die omgewing van 19 tot 33%.

Trees of the mango cultivars Zill, Tommy Atkins,Haden, Kent, and Heidi notably retain less than one fruiton average per panicle, despite sufficient irrigation andfertilization and adequate control of panicle diseases.After the respective events of flowering, fruit set andinitial fruit drop, a large proportion of the panicles on theupper portion of the canopy can be observed to bedevoid of fruit. In view of the high percentage of fruitlesspanicles, especially on a particular portion of the ca-nopy, it would appear that the cultivars in question cropbeneath capacity, at least under local conditions.

Wolstenholme and Robert (1991) reviewed themango yield problem, concluding that the tree carbohy-drate status, and moisture deficits that cannot be met bysoil moisture and arising due to the advent of hot anddry conditions, are the major yield constraints in existingcultivars.

It is well known in many tree fruits, including mango,that retention of a fruit, i.e., the capacity of a fruit toprevent itself from being shed, relates positively theability of the fruit to attract nutrients, which in turn isdependent the fruit's ability to produce growth promot-ing hormones.

Pre-flowering foliar sprays of urea or KN03 have beenfound to increase the number offruit retained per panicle(Nunez-Elisea, 1985; Baghel et al., 1987). Certain reportssuggest that urea and KN03 may, directly or indirectly,function as growth regulators in certain circumstances.KN03 sprays to mango trees before flowering have been

reported to advance and increase the intensity of flower-ing (Bondad and Linsangan, 1979; Nunez-Elisea, 1985),and pre-flowering urea spray application was reportedto be effective in increasing the number of leafy inflores-cences produced by Washington navel orange trees(Lovatt, 1990).

Boron deficiency has been implicated as a factorlimiting fruit retention in mango in view of a particularneed for boron for efficient fertilization (Robbertse, et al.1988). Boron applied as a spray to Tommy Atkins mangotrees in flower was reported to increase yield (Robbertse,et al. 1990; Coetzer et al. 1991).

The panicles of mango trees are seen as being adrain on reserves and currently produced assimilates(Wolstenholme and Cutting, 1991). By pruning thepanicles during the early stages of their development,and thereby markedly limiting the extent of their devel-opment, a reduction in the drain on reserves would beanticipated, and consequently, the availability of assimi-lates to young fruitlets after flowering would be expectedto be greater. Chacko (1984) concluded from investiga-tions made of the pattern of movement and distributionof 14C-photosynthates in mango at various stages offlowering and fruit development, that the heavy drop ofmango fruitlets soon after flowering is primarily due tothe competition between them for a limited supply ofassimilates. An increase in fruit retention might thusoccur in response to panicle pruning.

The present study was initiated to evaluate the effectof foliar spray application of KN03, low biuret urea, GA3(a gibberellin), CPPU (a synthetic cytokinin) or NAA (a

) period of flowering, effects an enhanced intensity ofiNering due to the greater number of panicles thatbsequently develop per terminal shoot. An increasedain of flowering on reserves and currently produced;similates would be expected to be greater followinguning. Furthermore, it would be anticipated that the~veloping panicles removed when pruning would addI the decrement in growth substances. The prunedees were often observed to set vast numbers of fruitlitially. Excessive fruit drop soon followed, however..ccording to Chacko (1984), the immediate and mostlirect reason for the heavy drop of mango fruitlets is the;ompetition between them for a limited supply of assimi-ates. In light of the above, it might be concluded that the'eduction in fruit retention associated with pruning wasjue to a reduced availability of assimilates to the youngfruit during their early stages of development.

Poor fruit retention has been attributed to the nega-tive effect of low temperatures on pollination and fertili-zation (De Wet and Robbertse, 1986; Robbertse, et aI.,1988), as well as the negative effect of low temperaturesduring panicle development on the number of herma-phrodite as opposed to male flowers formed per panicle(Mullins, 1987). Pruning, as performed in the presentstudy, generally deferred the onset of flowering frommid-July to mid- or late August when conditions werewarmer. A positive effect on fruit set and retention mightthus have been expected.

In the opinion of the author, the results concerningfruit retention reflect a trade-off between temperaturesduring flowering and the initial stages of fruit develop-ment, and the availability of substances required for fruitgrowth. This apparent increase in number of fruit re-tained when pruning was performed on July 31 as op-posed to July 3 and 7, supports this perception.

A reduction in return of 8% is indicated followingpruning on July 31 when panicle elongation was gener-ally occurring. However, this reduction does to take in toaccount the monetary benefit associated with uniformflowering, synchrony in the time of fruit set, and unifor-mity in the stage of maturity at harvest. It might reason-ably be assumed that this benefit would more thancompensate for the reduction in return of 8%. In view ofthe results presented here, as well as those presentedat the previously (Oosthuyse, 1991a), pruning of Sensa-tion mango trees in late July might be recommended.

Panicle removal by chemical means can by em-ployed as an alternative to hand pruning. A double sprayof Chemox® (dinoseb acetate) at 300 ppm (product) wasrecently found to be effective in deblossoming Sensationmango trees. Application made when the first paniclesto emerge were fully developed (100% anthesis), fol-lowed by a repeat application three weeks later, wassuccessful in causing panicle die-back and uniformflowering from axillary buds adjacent to the points ofpanicle attachment. Phytotoxisity to the leaves was notcaused.

In view of the probable withdrawal of Chemox® forcommercial use, future trials will incorporate other pros-pective chemical deblossoming agents. Ethrel®sprayed on trees at concentrations of greater than 1200ppm (a.i.) may prove to be effective.

Gibberellin (GA3) applied as a spray before theflowering period was reported to be effective in delayingand synchronizing flowering of Keitt mango (Nunez-Eli-sea and Davenport, 1991). In an experiment recentlyperformed by the author, spray application of GA3 at 200ppm approximately three weeks before flowering, mar-kedly delayed budbreak and gave rise to shoot as op-posed to panicle development. Failure to obtain thedesired response may have been related to the time ordose of the application. This option is to receive furtherinvestigation.

Thanks are due to Gustav van Veijeren of Constantiafor his generosity in making trees available for this study,and to Wynand Saaiman of Merensky TechnologicalServices for abstract translation.

LITERATURE CITEDCHACKO, E.K. 1984. Physiology of vegetative and repro-ductive growth in mango (Mangifera indica L.) trees.Proc. First Aust. Mango Res. Workshop, eSIRO, Mel-bourne, pp. 54-70.

DE WET, E. AND P.J. ROBBERTSE. 1986. A preliminarystudy of the pollen of Mangifera indica L. cv. Haden inSouth Africa. S. Afr. J. Plant and Soil 3: 87-89.

MULLINS, P.D.F. 1987. The effect of varying climaticregimes on floral development and behaviour in mango(Mangifera indica L.) cvs Haden and Sensation. Mastersthesis, University of Natal, Pietermaritzburg.NUNEZ-ELiSEA, R. AND T.L. DAVENPORT. 1991.Flowering of 'Keitt' mango in response to deblossomingand gibberellic acid. Proc. Fla. State Hart. Soc. 104:41-43.

OOSTHUYSE, 1991a. Mango de-blossoming to syn-chronize flowering and effect uniform fruit maturity. Inter-nal report, Merensky Holdings: Research Talks andPapers, 1991.

OOSTHUYSE, 1991b. Stages of development of themango panicle. SA Mango Growers' Assoc. Yearbook,11:59-61.

ROBBERTSE, P.J., E. DE WET, AND L.A. COETSER,1988. The influence of temperature and boron on pollentube growth in mango. SA Mango Growers' Assoc.Yearbook 8:4-6.

WOLSTENHOLME, B.N. AND J.G. M. CUTIING. 1991.Source-sink competition during the critical flowering andfruit set period - A mini-review of strategies in three fruittree architectural models. Supplement to J. Southern Afr.Soc. Hart. Sci. 2:21 (abst.).

synthetic auxin) on fruit retention, average fruit weightand yield at harvest, and monetary return taking curren-tly obtained prices into account. Composite applicationwas evaluated in most instances. Applications weremade while the trees were in flower (Experiment I), andsubsequently, just prior to the commencement of fruitdrop (Experiment II). In Experiment I, the effect of a foliarapplication of Wuxal® -boron, a nutrient formulation richin boron, and the effect of panicle pruning were addition-ally evaluated.

MATERIALS AND METHODSExperiment I

Thirty nine, 10-year-old Tommy Atkins mango treeswere selected for uniformity of size in an orchard atConstantia in the north-eastern Transvaal (grower: Gus-tav van Veijeren). To effect uniform flowering, the treeswere deblossomed on July 17 and 19. Panicle removalwas achieved by pruning the terminal shoots 5 cm be-neath the point of panicle attachment, which meantadditional removal of the terminal whorl of leaves. OnAug. 28, the trees had already begun to re-flower, thepanicles having been at the development stage of "inshoot: 5-10 cm" (Oosthuyse, 1991). To control paniclediseases, the trees were sprayed with Nustar® (5 ml/1 001H20) every 14 days during the period of flowering.Spaying commenced on Aug. 21.

The experiment comprised of the following treat-ments:

1. Control: Untreated

2. Panicle pruning: Newly developing panicles wereheaded just above the site of the second or third primarybranch from the point of panicle attachment. The remain-ing panicle branches were headed back severely (2/3removed) once they had developed somewhat. Paniclepruning commenced on Aug. 28.

3. Wuxal® -boron

4. KN03 + NAA

5. Urea + NAA

6. KN03 + CPPU

7. Urea + CPPU

8. KN03 + NAA + GA3

9. Urea + NAA + GA3

10. KN03 + CPPU + GA3

11. Urea + CPPU + GA3

12. KN03

13. Urea

Concentrations and times of application:

Wuxal® -boron: 150 ml/1001 H20; applied twice, onAug. 28 (panicles in shoot: 3-15 cm) and on Sep. 11(panicles 50 to 100% anthesis).

KN03: 4 kg/1001 H20; applied twice, on Aug. 28(panicles in shoot: 3-15 cm) and on Sep. 11 (panicles 50to 100% anthesis).

Urea: 1 kg low biuret urea/1 001 H20; applied twice, onAug. 28 (panicles in shoot: 3-15 cm) and on Sep. 11(panicles 50 to 100% anthesis).

NAA: 88.9 ml Planofix®/1001 H20 (40 ppm a.i.); appliedon Sep. 18 (panicles at small fruit set).

Tt J j r T X

I 1 ± f- I

1 ;;; CONTROL2 := PANICtE PRLtlING.) ::: WUXAI.4=KN03+NMS=\KA+NAA6 :: KN03 + CPPU7 ::: lHA + CPPlJ8 =KN03+ NAA+GA9=lMA+NAA+CA

10 == KN03 + CPPU + GA11 = lJR[A + CPPU+ GA

12::: KN03

13=lftA

Fig. 1 Average number of fruit retained per tree (verticalbars: + SE).

400

380

3360

:c 340tr>

~ 320

=s 300..::280vtr>

~ 260

~ 240

aT

-

tn

a drr

1 = CONTROL2::: PA1'fCt.E PRltllNG:5 =: WUXAL4=KH03+NM5 ::: UREA + NAA6 ::: KN03 + CPPU7 = UREA + CPPU8=KN03+NAA+GA9=UREA+NAA+CA

10:::: KN03 + CPPU + GA11 =UREA+CPPU+GA

12=KN0313::: UREA

Fig. 2 Average fruit weight at harvest (vertical bars: +SE).

70r-

601 = CONTROl2 ::: PANIQ.r PRlt.e.IG3 :; WUXALA :::: KN03 + NAA

5 = UR[A+ HAA

6 = KN03 + CPPU7 ::: UR£A + CPPU8=I<N03+NAA+GA9=UREA+NAA+GA

10 = KN03 .•. CPPU .•. GA11 = UREA + CPPU + CiA12 = KNOJ

13 = UREA

Fig. 3 Average weight of the fruit harvested per tree(vertical bars: + SE).

I

It T1

± trr drI

CPPU: 10 ml, 10% CPPU/1001 H20 (10 ppm a.i.); ap-plied on Sep. 18 (panicles at small fruit set).

GA3: 138.9 ml ProGibb®/1001 H20 (40 ppm a.i.); appliedon Sep. 18 (panicles at small fruit set).

When more than one growth regulator was applied(Sep. 18), the growth regulators were mixed in the sametank. In every instance, the trees were sprayed to run-off.

1 2 3 4 5 6 7 8 910111213Treatment

-. T

1 T

-

[ ± 1 - drI

In-

1 _ CO,.TROL2 ~ PAJtCl.( PRlJ,':NG:5 - WUXAl."-KN03 +HM5 - UREA + NAA6 .,. KNOJ •. CPPU7 •• URO •. QlPUa_KNO~+NAA+CA9-URO. .•. NAA .•. CA

10 = KN03 .•..CPPU 't" GA11 ~ UREA .•.. CPPIJ .•. GA12"'" KHOJlJ=UREA

Fig. 4 Average return for the fruit harvested per tree(vertical bars: + Sf).

" 160co

!!:-140

~ 100a.

c 80.2~'" 60E~ 406 70

60 150140

g 30~ 20c~ 10~'" 0c.- -10~~ -20.~ -30

-40-50-60

1 •• CONIIKll

2 '" PJoJ.ICl[ Pfl\Jt_1G:5 - WUXAl.• '" KNOJ .•..NAA5 =~+N.AA

6_KNOJ+CPP\J

7 '" ~[A + CPPU8 ;ktlOj +NAA+ GA9 = URE.4 .•.. /iliA .•.. GA

10'" "'N03 .•.. epP\J .•..CAl' '" lR[A .•• CPPU .•. CA12;;;: IiN03

U=~EA

Fig. 5 Average percentage increases in return relative tothe average return for the control trees.

A completely randomized design was employedwhere single trees served as plots. Each treatment wasapplied to three trees. The fruit were individually weighedon the general harvest date, determined on the basis ofthe initiation ofthetransition in pulp colouration. Averageprices received on European markets in 1992 forTommyAtkins fruit of the various counts after deduction forpacking and shipping costs were obtained from HansMerensky Exporting Company. A price was accordinglydesignated to each fruit. Due to pronounced varianceheterogeneity relating to treatment, analysis of variancewas not performed. The data (number of fruit, averageweight of fruit, and total weight of fruit per tree at harvest;and monetary return per tree) were interpreted in termsof the treatment means and their standard errors.

Experiment II

On Sep. 20, before the commencement of fruit dropand when fruitlets were generally "marble" to "goen" insize, 65 trees were selected for uniformity of size in theaforementioned orchard. Immediately after selection,the following spray applications were made:

1. Control: Untreated

2. CPPU

3. CPPU + GA3

4. CPPU + KN03

5. CPPU + Urea

6. CPPU + GA3 + KN03

7. CPPU + GA3 + Urea8. NAA

9. NAA + GA3

10. NAA + KN03

11. NAA + Urea

12. NAA + GA3 + KN03

13. NAA + GA3 + Urea

Concentrations:CPPU: 10 ppm CPPU (10 ml, 10% CPPU/1 001H20)

GA3: 30 ppm GA3 (104.2 ml ProGibb@/1001 H20)

NAA: 40 ppm NAA (88.9 ml Planofix@/1001 H20)

KN03: 2% (2 kg KN03/1001 H20)Urea: 1% (1 kg low biuret urea/1 001H20)

When more than one substance was applied, both wereplaced in the same tank. Spraying was performed torun-off.

A completely randomized design was employedwhere single trees served as plots, each treatment hav-ing been randomly designated to five trees. The fruit

-

T ±m fIT ta

[ -

1 _ CONIROL2 ...• O"P\JJ "'CFPU+CA.4. :: CPf'U + KNO..s5 = CPP\J + UREA6 = CPPU .•.. GA .•.. KNO..s7 = CPf'U + CA - lM::.8;;; HA..6

9=HAA+GA10 = N .•••.•• KNO.511 _ NAL .• LH.6.12 =HAA t GA tKIl0313 _ NAJ. • GA j UR[.A

Fig. 6 Average number of fruit retained per tree (verticalbars: + Sf).

"•.~ 250~'"::: 200~"-(; 150

~ 100E"z

--

-

1 = CONTROl2 = CJ'f'U:5 = 0'f'U'" GA.• = CPPJ + KNO..s

5 = CPf\I + LilIA6=Cf'f\J+G"'~KNO'}]=CPf\J .•. CA. .•. UR(A8 ~ NAil9=NA.A+GA

10 ~ NAA + KNOj1 1 = NAA ~. uPE.6

12 ~ NAA + GA t KN03U=NAA4-GA+lP.(A

Fig. 7 Average fruit weight at harvest (vertical bars: +Sf).

'5.: 360v

~ 340•><l

450:§

r sqr = 9%p (sig. level) = 0.015Y = 398 - 0.1 24x

-§,~ 400

250 "'"o 50 100 150 200 250 300 350 40(' 450 500Number of Fruit

Fig. 8 Relationship between fruit number and averagefruit weight.

3001------- -- -----

" I T§ 250~

1 _ CONT

2;;;0'I'U

~ ~ O'I'U •. GI...t-CPf'Uf-KH03

5 = CPr'U •. lIl[A6-Cf'PJ+GA+K"i037 =CPJ\I+CA .•. UR(A8=NAA9::;: tfAil .•• GA

10::; NA.I\'" KH0511 =.NAA +-UREA12 = NAA" Go:. •. KIID.!\j ~ tlAA .•. GA .•. UR[A

-

II

~d r± 1~ td

.L~a. 150c,~

IY tOOEa••. 50co

Fig. 10 Average return for the fruit harvested per tree(vertical bars: + SE).

were individually weighed on the general harvest date,determined on the basis of the initiation of the transitionin pulp colouration. Average prices received on Euro-pean markets for Tommy Atkins fruit of the variouscounts in 1992 after deduction for packing and shippingcosts were obtained from Hans Merensky ExportingCompany. A price was accordingly designated to eachfruit. Due to pronounced variance heterogeneity relatingto treatment, analysis of variance was not performed.The data (number of fruit, average weight of fruit, andtotal weight of fruit per tree at harvest; and monetaryreturn per tree) were interpreted in terms of the treatmentmeans and their standard errors.

RESULTSExperiment IFruit retention

The only treatments that resulted in a noticeableincrease in the number of fruit retained were Treatment8 (KN03 + NAA + GA3) and Treatment 12 (KN03)(Fig. 1). Low biuret urea, when applied solely or incor-porated, had a pronounced negative effect on fruit reten-tion.

Average fruit weight

In general, the increases in fruit retention were eachaccompanied by decreases in average fruit weight, andvise versa (Fig. 2). However, in Treatment 2 (KN03), theeffect of an increase in retention was also accompanied

130

120

110

~ 1000>

0 901Jm 80;,:~ 70~I-

60

50

40

30

~ 1-IT IT

11-

-

±

1 ;:; corlfROl..

2 _ O'!'V5_CWU+GI.4. ••. O'PU .•. KH05

5 = CN'I.J + If(A6 = Cf'f\J .•. C" .,. K"I03

7 :; CPf\I + CA .•. UIl[AII - NAA9 •• HAl; .•. GA

10 ::::NAA+KHOJ11 .:. N,l.A •. UREA

12.:.N.A.A.+G4+KNO.iD = NA,A .•• GA .•• lJR(A

Fig. 9 Average weight of the fruit harvested per tree(vertical bars: + SE).

50

4030g

~ 20,~ 10>~

IY 0c.~-10

~ -20~ -30

1 _ CONI2 = cPf'U

J=cPF'U+-C..• = CPPU +- KNO.55 = CPPU .,. LilIA6 = cPPU .•. GA .•. I(N037 = CPJ\I .•. CA .•.. U~A

8 = NAA9=NAA+GA

10 =N.u.-KN0311=NA.A+VPEA12 = HAt.. .•. GA. I 1<t10313 =tIAATGA+l.P[A

Fig. 11 Average percentage increases in return relativeto the average return for the control trees.

by an increase in average fruit weight. Panicle pruning(Treatment 2), although not having had an effect on fruitretention, resulted in a noticeable increase in averagefruit weight.

Yield

In considering the total weight of fruit harvested pertree, the only treatment to effect a noticeable increasewas Treatment 12 (KN03) (Fig. 3). This might have beenanticipated in view of this treatment having caused anincrease in fruit retention as well as an increase in aver-age fruit weight. Although Treatment 8 (KN03 + NAA +GA3) gave rise to an increase in the number of fruitretained, a resultant positive effect on yield was offset bythe associated reduction in average fruit weight. Theeffect of panicle pruning on yield was not marked, des-pite the positive effect of this treatment on average fruitweight.

Return

In considering return based on prices received dur-ing the 1991/92 export season, Treatment 12 (KN03)was the only treatment that resulted in a noticeableincrease in return (Fig 4). A 33% average increase inreturn relative the average return for the control treeswas realized (Fig. 5).

Phytotoxisity

The Urea and KN03 applications were slightly phto-toxic. Burn and at the tips of the older leaves and margins

leaves on new flush growth was apparent. Curling of theyoung leaves was apparent, as well as blackening (ne-crosis) of young fruitlets on the panicles. In view of thelatter effect, KN03 applications at a lower concentrationof 2%, for example, may prove to be more beneficial thanthe 4% applications.

Experiment IIFruit retention

The treatments that resulted in a noticeable increasein the number of fruit retained were Treatment 3 (CPPU+ GA3), Treatment 8 (NAA) , and Treatment 9 (NAA +GA3) (Fig. 6).Average fruit weight

The increases in fruit retention were each accompa-nied by a decrease in average fruit weight, and vise versa(Fig. 7). In considering the relationship between averagefruit weight and the number of fruit retained per tree, asignificant negative relationship was found (Fig. 8). How-ever, the relationship was weak, as indicated by theslope of the fitted line, and imprecise, as indicated by thelow coefficient of correlation (~). Interestingly, this rela-tionship suggests that, in Tommy Atkins at least, fruit sizecannot be markedly influenced by fruit thinning.

Yield

In considering the total weight of fruit haNested pertree, the increases obseNed were a direct reflection ofthe effect of treatment on fruit retention (Fig. 9). Notice-able increases in yield were obseNed for Treatment 3(CPPU + GA3), Treatment 8 (NAA) and Treatment 9(NAA + GA3).Return

Return, based on prices received during the 1992export season, was commensurate with yield (Fig. 10).Noticeable increases in return were obseNed for Treat-ment 3 (CPPU + GA3), Treatment 8 (NAA) and Treatment9 (NAA + GA3). An average increase in return of 26%,29%, and 19% was obseNed for Treatment 3, Treatment8, and Treatment 9, respectively, relative to the averagereturn for the control trees (Fig. 11).

In view of the preliminary nature of the reSUlts, it wasfelt that a full discussion of the effects of the treatmentsemployed is not warranted. It would appear, however,that spray applications of KN03 during the period offlowering, or spray application of CPPU + GA3, NAA, orNAA + GA3 just prior to the commencement of fruit drop,can be employed to improve the yield of Tommy Atkinsmango. Productivity increases in terms of return realizedwere in the order of 19 to 33%. The foregoing treatmentsmay also prove to be affective when applied to trees ofother mango cultivars grown that retain an average ofless than one fruit per panicle.

Future trials will be aimed at (1) confirming the resultsof the present study, (2) testing the promising sprayapplications on a number of the other mango cultivarsgrown locally for export, and (3) investigating the effectof other concentrations ofthe promising applications, forexample, 2% applications of KN03 during the floweringperiod.

ACKNOWLEDGEMENTSThanks are due to Gustav van Veijeren of Constantia

for his generosity in making trees available for this study.Thanks are also due to Wynand Saaiman of MerenskyTechnological SeNices for abstract translation.

LITERATURE CITEDBAGHEL, B.S., R.K. SHARMA, AND P.K.R. NAIR. 1987.Influence of preflowering spray of urea and NAA on fruitretention of mango (Mangifera indica L.). Prog. Hort.19:200-202.

BONDAD, N.D. and E. L1NSANGAN. 1979. Flowering inmango induced with potassium nitrate. Hortscience,14:527 -528.

LOVATT, C.J. 1990. The role of nitrogen in citrus flower-ing and fruit set. Lecture presented in March 1990 at theUniversity Cooperative Extension Class, University ofCalifornia, Riverside.

OOSTHUYSE, SA 1991. Stages of development of themango panicle. SA Mango Growers' Assoc. Yearbook11:59-63.NUNEZ-ELISEA, R. 1985. Flowering and fruit set ofmonoembryonic and polyembryonic mango as in-fluenced by potassium nitrate sprays and shoot decapi-tation. Proc. Fla. State Hort. Soc. 98: 179-183.

ROBBERTSE, P.J., E. DE WET, AND L.A. COETZER,1988. The influence of temperature and boron on pollentube growth in mango. SA Mango Growers' Assoc.Yearbook 8:4-6.ROBBERTSE, P.J., E. DE WET, AND L.A. COETZER.1990. The influence of boron application on fruit set andyield of mangoes. SA Mango Growers' Assoc. Yearbook10:24-27.

WOLSTENHOLME, B.N. AND J.G. M. CUTTING. 1991.Source-sink competition during the critical flowering andfruit set period - A mini-review of strategies in three fruittree architectural models. Supplement to J. Southern Afr.Soc. Hort. Sci. 2:21 (abst.).

WOLSTENHOLME, B.N. AND J.P. ROBERT. 1991. Somehorticultural aspects of the mango yield problem andopportunities for research. South African Mango Gro-wers' Assoc. Yearbook 11: 11-16.