4. Aen, Sor Hom Sa 123(9:791-797 1998 GA, Application Alters Flowering Phenology of ‘Hass’ Avocado Samuel Salazar-Garcia'and Carol J. Lovatt® Department of Botany and Plant Sciences, University of California, Riverside, CA 92507-0124 Abonionat 1x WwoRDs. Persea americana, determinate inflorescence, lowering inhibit inflorescence development Awsract: The objectives of the present research were to quanti 1) the contribution that vegetative shoots produced in ‘the summer vs. fall and indeterminate vs. determinate inflorescences make to yield und 2) the effects of GA, on lowering ‘expression and inflorescence phenology of summer and fall shoots of “Hass” avocado (Persea americana Mill.) under field ‘conditions. Anthesis started earlier on fall than summer shoots of 10-year-old ‘Hass’ avocadotrees; however, no difference im the date of full bloom was observed. Indeterminate inflorescences that underwent early anthesis set more fruit than those with delayed anthesis, conversely, determinate inflorescences with delayed anthesis set more fruit. Indeterminate inflorescences comprised 90% of total inflorescences and contributed 73% of total fruit yield, but individual determinate inflorescences were at last three times more productive than the indeterminate ones. Summer and full shoots were sprayed with 0, $0, 100, or 1000 mgcL~' GAs in November, December or January. GA, stimulated apical growth ofall shoots. If secondary axes of an inflorescence bud were differentiated at the time of GA, application, the inflorescence developed inadvance uf inflorescences on branches nottreated with GA, Inaddition,GA. caused precocious evelopment of the vegetative shoot of indeterminate inflorescences relative tothe flowers in the same inflorescence and relative tothe vegetative shoot of indcterminate inflorescences from untreated branches. Stimulition of vegetative growth at the inflorescence apex by GA, inhibited growth of axillary buds. GA, at 0 mL. had noeffeet on the number of determinate or indeterminate inflorescences produced by either summer or fall shoots. Higher concentrations of GA, increased the indeterminate inflorescence, ‘number of vegetative sheots and inactive buds produced by both shoot types. Avocado trees generally produce two orthree vegetative flushes curing the year (ie., spring, summer, and fall flushes). Not all the branches contribute toeach flush (Davenport, 1982; Scholefield et al., 1985; Venning and Lincola, 1958). Spring flush shoots can produce either summer or fall flush vegetative shoots. Shoots produced early in the summer can contribute tothe fall vegetative flush, but late summer shoots typically produce inflorescences «during the winter-spring flowering period (Thorp et al,, 1993), Fall shoots that developed early (September in the Northern hemisphere) may flower immediately during the winter-spring bloom periods other fall flush shoots contribute to the spring vegetative flush. The proportion of summer and fall vegetative apices that continue the growth of the shoot or develop into inflorescences is unknown, Hass’ avocado trees have a profuse and extended flowering period. Flowering, which may start early in the winter (January). ‘continues throughout most ofthe spring season, with peak blooms ‘occurring between March and May under Southern California conditions (Lovatt, 1990). Asbloom progresses, competition with the developing spring flush increases. This competition is intense ‘nindeterminate inflorescences, in which the vegetative apex starts Received forpublication 16Sept 1997. Accepted for publistion 4 May 1998. This reseuch was supported in pt the Citrus Research Center abd Agricultial Experiment Station ofthe Univ of California, Rivenide and by the California ‘Avocado Commission. Thspapesrepresensapieonof the destin suited by SSG. paral ulimen of reqiements fora PHD a Botany the Un, af California 8.8.6. acknowledges he financial upprtof CONACYT, INIFAP tnd SNI of Mesicoand tke University of Califomia, Riverside, We tank Chale W. Copsins Ir for is advice onthe selection of GA, concentration aa for his ical review of the manuseep. We hank Abbott Labostones for donating the Propbband Dick Whiley foeatcesso his avocado orchards in Ranch Clio. ‘The cos of polishing this pper wes defrayed in part the payment of pase charges, Under postal regulations, this paper thaefore must be hereby marked ‘aderiement soley w indicate this at "Caren address: Canpo Experiment Santiago Ixuinls-INIFAP, Apdo, Postal 100, Sanage Iulia, NAY 63300, Mexico: to whom epi requests shuld be dase “Provesor of pnt physiology 4, Awe, Soc. Hort. ct. 123(5):791-797. 1998, growing after inflorescence elongation and the vegetative shoot continues expanding during anthesis and fruit set (Bower and Cutting, 192; Cutting and Bower, 1990; Whiley, 1990; Zilkah et al, 1987)-Singularly and in combination, these factors may reduce fruit set. Manipulation of avocado flowering to change theso relationships temporally may help to increase fruit set and yield. :Kogenous application of gibbercliins (especially GA.) might provide a tool to manipulate time and intensity of flowering. GAy application has been shown to inhibit flower initiation in both deciduous (Sedgley, 1990) and subtropical fruit crops, such as Jemon (Nir et al. 1972), mandarins (Iwahori and Oohata, 1981) and orange (Guardiolaet al, 1977). Similar responses to GA; have been found in tropical fruit crops. The time of GA, application influences the response. For example, application of GA, to apical ‘buds of mango (Mangifera indica L. ‘Dushehari’) before floral initiation stimulated vegetative growth in 75% of shoots. How- ever, once floral meristems were present, GA, did not inhibit flowering (Kachru ct al., 1972). In another case, GAy treatment repressed floral bud initiation of “Keitt” mango and delayed flowering by >4 weeks (Nunez-Elisea and Davenport, 1991). Porlingis and Boynton (1961) found a dual effect of GA; on strawberry (Fragaria chiloensis) flowering. GA, accclerated the ‘appearance of flowers that had differentiated at the time of appli cation but inhibited the initiation of new flowers under inductive conditions. In coffee (Coffea arabica L.), the effect of GA, (100 ‘mg:L~*) was also dependent on the stage of floral bud development (ehuch etal, 1990), ‘The goal ofthe present research was to determine whether Gs applied to the canopy of *Hass” avocado might have potential ‘management tool to inerease yield, Thus, this research haat the following objectives: 1) to determine the time of flowering of vegetative shoots produced during the summer vs, fall fu ‘examine ifthe time of anthesis affects fruit set; 3) to quantify which type of inflorescence (determinate vs. indeterminate) contributes more to yield, and 4) 10 quantify the effects of dose and time of application of GA, on inflorescence number and phenology. v1Type shat “Tagged shoots Mar. Is Dec. Van 151m, n, valuation dats (1993-91), Feb Tig |, Nun of sums and fal ‘His’ avocado shoot a the tage of anthesis ‘fit lowers Each oie is amion +s of 0 reer and TO sho0 per bee Materials and Methods ‘The research was conducted in a commercial ‘Hass’ avocado orchard in Rancho California (Southern California, lat. 33° N) from 1993 to 1995. Avocado tres planted at spacing of 7x 7m ona south oriented hill were used, Water and nutritional manage ‘ment were standard grower practices and no visual symptoms of ‘water stress oF nutritional disorders were observed ‘Tit OF ANTHESIS IN 1994 FOR APICAL INFLORESCENCES OF 1998, SUMMER AND FALL FLUSH SHoors. Twenty 10-year-old “Hass” avocado {ees that set no fruit during spring bloom 1993 were selected (block 1). In each tree, 10 shoots that developed during ‘Summer 1993 ane 10 that developed during Fall 1993 weretagged. For cach shoot, the number of inflorescences with flowers at nnesis were quantified throughout the 1994 flowering season. ENVECT OF GA, ON SLOWERING OF 1993 SUMMER AND FALL FLUSH sstoors. Thiny 10-year-old Hass’ avocado trees were used during 1993-98 (block 2). To avoid a possible effect of fruit load on flowering, banches, each at similar phenological tage and with no fruit sci from the 1993 spring bloom were chosen in September 1993. On the south side of each tree (developmentally the most sxlvanced and uniform, four branches I min length were selected snd their shoots tagged according tothe vegetative flush in which they developed (summer oral) Foreach shoot, the apical bud and is five immediate axillary buds were tagged. For GA, treatments, «afresh solution of GA, prepared from Progibb 4% (Abbot Labo- ratories, North Chicago) plus 1 mL:L Triton X-100 (Sigma Chemicals, St. Louis) in water (pH 5.5) was sprayed onto selected ‘branches until runoff, Treaiments consisted ofa single application ‘of one of four GAsconcentrations (0,50. 100, and 1000 me:L.') and three spray dates, the 13th day of either November 1993, December 1993, or January 1994. All treatments were sprayed before budbreak, Control branches received only water plus Triton X-100. To determine the stage of inflorescence bud development ‘teach spray date, the apical bud of one summer and one fall shoot per replication were collected. Buds were immediately fixed in FAA GS formalin : 3 acetic acid = 90 ethanol, by volume), dehy~ rated, infiltrated, embedded, sectioned, and stained as described by Salazar-Garcia otal. (1998). The type of growth produced by apical and sxillary buds, and inflorescence phenological stage were recorded for individual shoots. Developmental sages were ‘evaluated avcording to the macro. and microscopic analysis of Salazar-Garefactal. (1998). A randomized complete block Jesig With 10 individual tree replications was used. Percentages of the total number of shoots were transformed by aresin of the square root ofthe percentage prior tostatistical analysis (Steel and Torri 1980). For means comparison, the Duncan's multiple range test ‘wasused. When atwo means comparison wasneeded,an Ftestwas performed before the / test [FRUIT SET ON EARLY AND LATE SEASON INFLORFSCENCES (1994), Fifly-nine 10-yeur-old “Hass’ avocado tres (block 3) with regular fruit load (30 to 80 kg/tree) were used. In each tree, two groups of 50) apical inflorescences at Stage 8 of development (cauliflower stage) (Salazar-Garefa et al., 1998) were tagged in an uniform manner around the tree. The first cobvrt (early inflorescences) was tagged on 12 Feb, 1994. The second one (late inflorescences) was selected a month later (12 Mar.). Fruit set by determinate and indeterminate inflorescences was recorded alter the June feuit drop period on 8 Aug, 1994. A second evaluation wax done on 8 Dec. 1994 when fruit reached physiological maturity and were close to harvest, Experimental layoutand data analysis forthisand the next stuuly were performed in the same way. A randomizedblock design With $9 individual tre replications was used. Before statistical analysis the data were transformed by the square root of the “observation plus one. PRODUCTIVITY OF DETERMINATE AND INDETERMINATE INFLORES* ‘cexers i999). This rescarch was conducted in block 3 used above ‘At the beginning of this evaluation (spring 1995) all trees were in their “off” crop year and were bearing little to no fruit. Flowering ‘Table 1. Stage of inflorescence development of apical buds of summer and fall shoots of “Hass” avocado atthe time of GA, treatment tage of pee ett inflorescence T3 Now. 199) Tan, 1994 development’ Sommer Fall Summer Buds Stage 100 80 30 stage 3 20 0 2» Stage a 20 2 x0 Stage 5 30 Stage 6 30 80 0 "Stages of inflorevceave development Salazar Gareraetal, 1998), T= Convex primary axis meristem with baci One ortwo secondary mis meso in the axl of bracts. 2= Flat primary axis meristem with Separated bracts. One to three secondary axis menstems inthe axils of brits, 3 inflorescence meristems. 4 = Flat primary axis meristem, Ten secondary axis inflorescence meristems. longation of oldest secondary axisin wit tertiary axis meristems present. Initial development of perianth of termi tertiary axes. 6=Flongation of youngest sccondary axis meristems Okdest primary axis meristem. Four secondary ax have complete perianth, and anthers with sporogenous issue ane present lowers of secondary and condary axes are completely formed, including eyme of flowers. Flowers ynoecium is at early loeule formation. Obtained from median longitudinal sections of 10 buds per shoot type and 10 tree replications foreach sampling dae. 2 4. Avie Soe. Hoar, Se. 123(5):791-797, 1998,Table 2. Apical growth produced by “Hass’ avocado branches sprayed with Gaon 13 Nov. 1993, Shoot ype and Total shoots (9) tong Inflorescences Vegetative shoors Tnactive buds Samer shoots ° 10.0.0 006 0a 50 90.8 ab 926 00a 100 54.2 be 33h i254 1008) Me 6338 25a Fall shoots 0 29a sb laa 0 96.33 37h 008 100 714 ab 3b 3a 1000 4156 5008 25a * Mean separation Ta coTums Tor each shoot type by Duncan's mulipe range test, P= 005, \was abundant since it was the beginning ofthe “on” cycle. Ineach tree, five branches I'm long distributed around the tee were tagged. The total number of determinate and indeterminate inflo- Fescences prevent in each branch was counted a the anthesis stage (11 Apr. 1995), In addition, the number of flowers per inflores- ‘cence was counted fora sample of fourinflorescences periinflores- ‘cence type per tee. Because our 1994 study of fruit set showed no significant differences between the evaluationsdone after the June fruit drop period (8 Aug.) and final fruit set (8 Dec.) only fruit set at physiological maturity was quantified for this study (6 Dec. 1995), For data analysis atthe branch level, the average of branches per tree was used, Results ‘Tinte OF ANTHESIS IV 1994 FOR APICAL INFLORESCENCES OF 1998, SUMMER AND FALL FLUSH SHOOTS. Anthesis started in January and \wax first observed on fall shoots. By 19 Feb., cight out of ten summer and fall shoots had inflorescences at anthesis (Fig. 1), Inflorescences borne on fall shoots were mostly indeterminate in the apical position and few shoots produced axillary inflores- cenees, Summer shoots had a more intense production of apical and axillary inflorescences that reached anthesis almost simulta- neously (data not shown), [EFFECT OF Ga, ON FLOWERING OF 1998 SUMMER AND FALL. FLUSH SHOOTS: STAGE OF INFLORESCENCE DEVELOPMENT OF APICAL BUDS OF SUMMER AND FALL SHOOTS AT THE TIME OF GAs TREATMENT. By [3 Nov. 1993, macroscopic development of apical buds on summer fush shoots showed pointed buds with closed bud scales, which corresponded to Stage 2 described by Salazar-Garefa etal. (1998), Microscopie analysis of buds at this stage revealed a flat primary axis meristem with separated bracts in which one to three second- ary axis meristems had already formed (Table 1). Organogenesis, ‘was monitored from November through January. On 13 Nov. ‘apical buds on fall flush shoots had all reached Stage 2, but 20% had already progressed to Stage 4, by which time all 10 secondary axis meristems ofthe inflorescence had formed. By 13 Dee. apical buds of summer and fall shoots presenteda wide range of develop- ‘mental stages with 30% ofthe apical buds on summer lush shoots advanced to Stage 6. Externally, the buds were rounded with only bases of outermost scales remaining such that bracts enclosing the inflorescence were visible, Microscopie analysis showed thatthe ‘youngest secondary axis meristems were elongated and that the ‘oldest basal secondary axes were completely formed, including the terminal eyme of flowers. The flowers had a complete perianth and anthers with sporogenous tissue evident. The gynoecium was at 4. Aver, Soe. Hor. Sex. 123(5):791-797, 1998, the stage of early locule formation (Salazar-Gareiaetal., 1998). By 13 Jan,, >70% of the apical buds on shoots of either age were at Stage 6 (Table 1) and thus had not reached budbreak Re sroxst OF INFLORESCENCE RUDS TO GA; TREATMENT. Increas- ing concentrations of GA, applied to summer and fall flush shoots in November reduced the production of inflorescences with a ‘concomitant inerease in vegetative shoot production (Table 2) Control summer shoots produced only inflorescences (100% of total shoots), whereas summer shoots sprayed with either 100 oF 1000 mg:L”' GA, in November had fewer inflorescences, In contrast, GA, at 1000 mg:L" decreased the production of inflores- ences and increased the production of vegetative shoots from apical buds on fll shoots (Table 2). Later GA, sprays in December or January did not affect the type of growth produced (vegetative or reproductive) by apical buds of either summer and fall shoots In addition, atthe later application dates, GA, sprays did not have ‘consistent effect on the proportion of apical buds that remained inactive. This proportion, compared to control, was significantly increased (P = 0.10) only by GAs at 1000 mg-L" in December, causing 8.3% of the apical buds of summer shoots to remain inactive (data not shown), LEFPECT OF GA; ON THE PRODUCTION OF DETERMINATE: AND INDE= THRMIINATE INFLORESCENCES BY APICAL BUDS, Untreated apical buds (control) on both summer and fall Mush shoots produced more indeterminate than determinate inflorescences in the two blocksin this study. Production of indeterminate inflorescences by control summer shoots ranged from 50% to 58% of total shoots, whereas ‘determinate inflorescences ranged from 39% to 50% (Table 3). Fall shoots always produced a higher proportion of indeterminate inflorescences (50% to 70%) and a lower proportion of determi nate inflorescences (24% to 40%). For summer shoots, the produc tion of determinate inflorescences was significantly reduced when 100 or 1000 mg-L°' GA, was applied in either November or December (Table 3). Number of determinate inflorescences on fll shoots was reduced only when GA, was applied in November. In, general, GA; application, even at the highest concentration was without significant effect on the production of indeterminate inflorescences for most treatment dates, except for December and. January for which significant increases in the number of indeter- ‘minate inflorescences resulted (Table 3). EFHECT OF GA, OY AXILLARY BUDS. GA, treatment consistently, ‘caused a decrease in the production of inflorescences and an increase of the number of inactive buds borne in axillary position fon summer shoots. This effect increased with increasing GAs concentrations. For summer shoots, growth of axillary buds was totaly inhibited only with GA, at 1000 mg:L-" applied in Novem- 793‘Table 3. Apical production of determinate and indeterminate faflorescences as.afTected by shoot type and prebloom GA. sprays tothe “Has” avocado. Spray date and typeof inflorescence Shoot type and T3Now. 1993 13 Dev. 1998 1B Jan, 199) GA Cet) Detcrminate Indeterminate Determinate __Indotenninate__‘Deerivinate Indeterminate Total shoots (%) Sumumer shoots 0 4962 soda 3040 SRILA 4580 sa2a 50, 4068 s02a 387 ab 363A 1194 6670 100 10.0 4420 Bb BAB 3698 583.0 1000 oon H2a doc 5338 880 3128 Fall shoots 0 ALA 68.8 ab 300 7008 4000 S008 50 82AB 80a 53a 943A 008 76.7 AB 100 168 Bub 166 TI8AB 008 100.0. 1000 508 25b 00a, 83.3AB 19a 72.8 AB “Rican separation in columns foreach shoot by Duncan's malliple ange fet, P=0.05 (lowercase Fetters) and 0.10 (uppercase eters i ber and December (Fig. 2). Axillary buds on untreated fll shoots {Control did not reach budbroak; this situation was not changed by GA, treatment (data not shown). Due to the greater number of anillary inflorescences produced by summer shoots, flowering ‘was more intense than that of fall shoots. AIl axillary buds that developed produced only determinate inflorescences TBP#tcT OF GA) ON THE RST OF INFLORESCENCE BUD DEVELO svt. By 13 Jun, control branches had no inflorescences atthe cauliflower stage (Stage 8; Salazar-Gareta et al., 1998), which is easily distinguished by elongation of secondary axes and the presence of small closed flowers on the tertiary ates, but branches Sprayed with 50 mg:L ' GA; in November or December had 8% and 18% of Wola inflorescences atthe cauliflower stag, respec tively (Fig. 3), Amore advanced flowering was obtained with GA) at 100 me:L. "applied in November or December, where 71% and 256 of otal inflorescences were athe cauliflower tageby 13 Jan. By the same evaluation dat, branches treated with GA. at 1000, mg-L in either November or December had ~80% of total inflorescences at this stage (Fig. 3) Inflorescence development was also inereased withthe January GA; application, the effect of which was leary observed only 14 d later. By 27 Jan..only 23% of total inflorescences produced from the apical bud of summer and all shoots (pooled data) not treated with GA; had reached the aulitiower stage, whereas 80% of those shoots treated with 50 or 10Kimg-L~'GAon 13 Jan, had reached this stage, Al shoots treated with Ga reached this stage earlier than the control. Earlier spray tes like Novernber and December, and higher concentrations had the most striking ffoets (Fi The date by which 50% of total inflorescences reached the cauliflower stage was calculated for control and GA, treatments (data not shovin).GAyat.50 mg-L* sprayed ether in December ot January caused inflorescence development tobe 23d ahead ofthe ‘ontral. GA, al 100 mg;L~ in either November or December caused the cauliflower stage to occur 37 and 23 d earlier than the control, respectively. Cauliflower stage also was advanced 37 d with GAvat 1000 mg L~ applied in November or December and23 d when applied in Jannary. Although GA;-treated shoots reached the cauliflower stage earlier, anthesis was not earlier. TENUCtS OF Gas ON VEGETATIVE SOOT DEVELOPMENT OF INDE ‘TERMINATE INFLORESCENCES, For indeterminate inflorescences de- ‘eloping at the apex of either untreated (control) summer or fall flush shoots, vegetative shoot growth and subsequent leaf expan- sion were delayed relative o elongation ofthe secondary axes of the inflorescence (Fig. 4). Thus, by 5 Mar, 1994, when almost 704 100F% of total inflorescences of the consol had reached the éauiower stage, or ster stage of development, Fe. 3), only 115 hd inated the fist leaves oftheir vegetative shoot apex (Fig. ).GA, sprays altered norma inflorescence phenology and faused the vegstative shoot to develop precocious and thus, Simultanecosly with elongation ofthe secondary and tertiary axes of th inflorescence (Fig). An evaluation done on 5 Mar. 1998, Showed that GA, at 100 and 1000 mg applied in November, December or Hn case significant precocious devclopmento the vegetative shoot (Fig 3). Late aplication dates December and January) provided a greater response for all GA; concentrations, including50mgL" GAstreatmentwith 1000 1000 mL" ineither November or December caused shoot development tobe initiated 37 earlier than inthe contol (data not shown). RELATIVE PRODUCTIVITY OF FARLY VS. LATE SEASON DETER ‘ST 0 IOHTERDIINATE AFLORESCENCR (994, 195). Esl info festences({hote that reached the cauliflower stage by 12 Feb, 1994; Slage 8 of the scale of Salazar Gara etal, 1998) were predominany located in the quadrant of the tee most exposed to the sin (SW) Lite inflorescences, wich reached the [> cauliflower stage by 12 March, were lo- | nar ated in the NE quadrant. The number of 100 0 total shoots L FHSS Dee, 1993 Nov. 1993, GAs (mg L") and application date Tig, 2. Flowering respon of ailry buds of “lass evecado summer shoot Sprayed with GA, at iffrent dix, Evasion dat: 26 Mar. 1994 All gew [growth contd of determinate inflorescences. 4, Asi, Se: Hoar, Set. 123(5):791-797, 1998.100 i? ae 57 aa x oes i, | aN 5) am im toe g 3 é & 3 3 : g g : 27, 19Feb. 5 Mar, Evaluation dates (1994) 26 Mar. Fig. 3 above) Percent spcal inflorescences atthe cauliflower stage (Stage 8 Salzar Garcia et a, 1998) in Has” avocado shoots (pol of summer and al shoot) a affected by spray date and GA, concentrations, days from the cauliflower stage to anthesis was longer for early than for the late cohort of inflorescences. Early inflorescences that ‘were mainly borne on fall shoots underwent anthesis on =12 Mat, Late inflorescences took only three weeks to advance from the cauliflower stage to anthesis, with anthesis occurring 2 Apr. 1994. No significant differences on fruit set due to time of anthesis were detected at the end ofthe June drop period or at physiological ‘maturity (Table 4). However, fruit set by early or late developing determinate vs. indeterminate inflorescences was significant! affected by time of anthesis. Determinate inflorescences were ‘more productive when they developed later in the season (Table); ‘whereas indeterminate inflorescences set more fruit when they developed earlier in the season, ‘The results ofthe second year study showed that indeterminate inflorescences were predominant, representing =90% of all inflo- rescences produced (Table 6) and yielded 73% of total fruit harvested. However, individual determinate inflorescences were three times more productive than the indeterminate ones. The number of flowers per determinate vs. indeterminate inflores- ccences was not significantly different (Table 6) Discussion GA, sprays caused precocious development of inflorescence buds and vegetative apices of indeterminate inflorescences in “Hass’ avocado, which resulted in early budbreak and inflores- cence elongation. Even at a low concentration (50:mg-L"), when GA, was applied in December or January, 50% of the inflores- cences reached the cauliflower stage 23 dearlir than thenon-GA;- sprayed controls. Higher GA, concentrations further accelerated the rate at which buds reached the cauliflower stage, but the date Of anthesis was not significantly affected. This response is not undesirable, however, early budbreak might present the risk of Fig. 4 (below). Precious vegetative shoot development of ndterminat inflorescences on Has’ avocado summer shots sprayed with GA in Doce 1093. From lft tright0 50, 10, and 10X0mg LGA. Pastre taken om ar J. Amen. Soc. Hower. Set. 123(5):791-797, 1998, 798ial erminate inlloressences wih eaves 15 Noy, 1998 13 Doe. 1998 13 Jan, 1994 ‘Spy dates Fig, . Precocious vegetative shoot development of indetrminateinleescencesof "Hse av shots (pol of summer und all soos) a aaenced By GA, concentation and spray te. Falaton dat: 3 Mar. 19, injury from low temperatures. GA, application in November, when buds hadonly one three secondary axis inflorescence meristems without apical bracts, resulted in production of vegetative shoots or partially formed inflorescences (with fewer secondary axes). Stimulation of the growth of the vegetative primary axis meristem by GAy arrested the growth of secondary axis meristems adjacent to it, and the response was stronger at higher GA, concentrations, However, flowering was not fully iibited, demonstrating that GAs did not revert secondary axis meristems into vegetative ones. Aborted meristems could be observed after inflorescence elongation as, scars with small inflorescence bracts, In December, when buds with up to 10 secondary axis meristems (Stage 4; Salazar-Garcia, eral, 1998), some of which had one pair of bracts formed, were present, application of GAs resulted in faster development of ‘complete inflorescences. The response of ‘Hass’ avocado to earlier GA, sprays deserves further investigation, since no GAs treatment ‘was applied before the beginning of inlorescence initiation in this, sudy No signs of toxicity were observed for any GA, concentration evaluated GA, at 50 and 100 mg-L~! had no negative morphologi- caleffects. However, GA, at 1000 mg-L” applied forall treatment dates caused a remarkable elongationofinflorescenceanes, which in general appeared to0 weak to support setting fruit. Despite this axis elongation, individual flowers produced with 1000 mg-L" GA, were not elongated but rather they were smaller than control flowers. The most often reported effects of GAson the physiology of Flowering in deciduous fruite-ops (Sedgley, 1990),cirus (Daven- por, 1990), and mango (Kachru et al., 1972; Nunez-Elisea end Davenport, 1991) are inhibition of flower initiation and delay of flowering. However, in strawberry and coffee GA, also acceler- ated the appearance of lowers that were differentiated at he time of application, resulting in early anthesis (Poriingis and Boynton, 1961;Schuch etal, 1990). Forthe GA; spray dates and ates tesied in our study, GA, never delayed flowering, but hastened the flowering process of *Hass' avocado. This is the first report of advanced flowering caused by GAs on the avocado. Under the cimatie conditions of California, vegetative growth and leaf expansion of indeterminate inflorescences of “Hass’ avocados delayed relative to the elongationof the secondary axes, of the inflorescence, GA, application dramatically ineteased the 796, rate of development ofthe vegetative shoot atthe apex of indeter- ‘minate inflorescences relative to elongation ofthe sevondary axes of theinflorescence, The degree of precociousness was highes with both higher GA, concentrations and later application dates, which was probably related to presence of warmer day temperatures during the late winter-carly spring. A common hypothesis in the Titerature to explain low fruit set in avocado is that there may be a ‘competition between the vegetative shoot that is developing atthe time that the flowers of indeterminate inflorescences are setting fruit (Bower and Cutting, 1992; Cutting and Bower, 1990; Whiley, 1990; Zilkab et al. 1987). The precocious vegetative shoot de- velopment in indeterminate inflorescences caused by GA; sprays ‘might result in more successful inflorescences in terms of fruitset and yield by eliminating the competition between reproductive and vegetative growth. Leaves of the precocious vegetative shoot of indeterminate inflorescences treated withGA, were sufficiently ‘mature to be sources (based on net CO, assimilation rates) at the ‘Summer anc fall flush shoots had a similar rate of inflorese development and although some fal flash shoots started anthesis, carlicr than summer shoots, the date for fall bloom was not differcat, Similar results for “Hass’ avocado in Australia were reported by ‘Thorp et al. (1993). The absence of significant differ- encesin the typeof growth produced by apical buds of summer and fall flush shoots in response to GA, sprays, suggests that a uniform, response to GA; application to entire trees might be achieved regardless of the relative size of the summer vs. fall Tushes. However, responses to whole-tree application of GAy must be ‘quantified hefore this growth regulator can be recommended on a commercial scale in avocado, In particular, the possible effect of GA, on excessive shoot elongation should be evaluated. ‘A pool of determinate and indeterminate inflorescencesshowed that fruitset was not influenced by the time of flowering. Nonethe- less, fruit set was better on early flowering indeterminate inflores cences and better on late flowering determinate inflorescences, ‘These results reflect the Fact that carly flowering inflorescences were mainly indeterminate, whereas late ones were mainly deter- ‘minatc. Thus, compared to indeterminate inflorescences, fruit set ‘on determinate inflorescences occurs under warmer conditions, ‘Table 4. Fruit set by a pool of determinate and indeterminate *Hass’ ‘vocado inflorescences during the early and late part of the 1994 flowering season (inflorescences were wigedat the culiawerstage). Evaluation time ‘Time oF bloom After Tune drop At physiological maturity (cauliflower stage) _(8 Aug 1994) (8 Doc, 1994) Fruit sei/50 inflorescences IE Early (12 Feb. 1994) 0.914018 086 £0172 Late (12 Mar. 1994) 0.95 +0.16 « 088 £0168 om, “Rigan separation Tw cotumns by Few, P=¢ ‘Table 5. Time of flowering and final fruit set by detersinateandindeter tminate Hiss’ avocado inflorescences forthe 1994 lowering season “Type of inflorescence eierminate Indeterminate ‘Fruit set inflorescences ESF Early (12 Feb. 1999) 0390.13 5" 04940.104 Late (12 Mar. 1994) 05920.128 0.29008 ‘ican separation in columns by Text, P=O-10 J. Amen. Soc: Horr. Sct. 123(5):791-797. 1998.“able 6. Production of determinate and indeterminate inflorescences and conteibuton ofeach type of inflorescence to yield of Hass’ avocado tees ‘during the 1995 flowering season, Characteristic evaluated Deierminate Indeterminate Numer of inflorescences/T-nvTong branch (11 Ape 5) Te=0aTD" TSS#I Ia Percontage tol invTorescences 1026 807 4 [Number of flowervinflorescence 1204267 a 10522624 ‘Number of mature fri/|-m-long branch 032008» 80.124 ‘Number of mature rut/109 inflorescences 158a 5b Percentage fruit set (fruivflower) ou7a 0.05 Contribution to yield (%) 2236 nya "Mean -5e (where shown) separation in rows by est, which are more conducive for fruitsetand initial fruitdevelopment (Lovatt 1990). This may indicate that the greater productivity of determinate inflorescences of ‘lass’ avocadois notonly duetothe lack of competition from adeveloping vegetative apex, but alsoto their development in warmer, more favorable temperatures. The higher proportion of indeterminate (90%) vs. determinate (10%) inflorescences found inthis study of ‘Hass’ avocado (Table 6)agrees with the report by Thorp etal. (1994) forthesame cultivar ‘growing in Australia (65% indeterminate inflorescences). In addi- tion, our resuits were in the range reported for Schroeder (1944) for ‘most avocado cultivars studied in California (80% to 95% of total inflorescences were indeterminate). In our study, determinate inflorescenceshad a much higher percentage of fruit sct per flower (0.17%) than indeterminate (0.05%) inflorescences. The percent age of fruit set was close to the range previously reported for avocado (0.02% to 0.14) (Chandler, 1958; Bergh, 1967). Given the predominance of indeterminate inflorescenves and their low percent fruit set, eliminating the competition between vegetative and reproductive growth in these inflorescences by foliar applica tionof GA; mightprovide amanagement strategy toincrease yield. In light of this possibility and duc to the absence of any ebserved. negative effects, foliar Gy application warrants further investiga tion as tool to inerease fruit setand yield ina.commercial avocado orchard. Literature Cited Bergh, BLO. 1967. Reasons for low yields of avocados. Calif. Avo. Soe Yebk. Si:16t Bower. JP. andJ.G.M. Cutting. 1992, The effet of selective pruning on ‘veld and fruit quality in "Hass" avocado. Acta Hort. 296:55-58. Chandler, WAH, 1958. Evergreen orchards, 2nd ed. Lea and Febiger, Philadelphia. p. 205-228. Cutting, LG.M, and J.P, Bower, 1990, Relationship hetween auxin transport and calcium allocation in vegetative and reproductive flushes in avocado, Acta Hort. 275469475, Davenport. 1.L. 1982. Avocado growth and development, Proc. la, State Hort Soe. 95:92-96, Davenport. TL. 1990, Ciews flowering. Hor. Rev. 12:349-408, Guardiola,TL.. M. AgustandF. Gareia-Mart 1977, Gibberellcacidand lower bud development in sweet orange. Proc. Int, Soc. Ciricult 2:696-668 Iahori, S. ond 1.1. Oohata. 1981. Control of flow tsuma ‘mandarin (Citrus washiu Mare) with gibbetllin, Pros. In, Soe. Cit- cult 1:247-249, J. Am, Soe, Howr. Sc. 123(5):791-797, 1998 105. Number of floworsnflorescence was analyzed withthe Mann-Whitney rank som Tes, Kachrs,R.B, RN. Singh, and EK Chacko. 1972. Inhibition of Nowering i Mangifera indica L. by gibberellic acid. Acta Hort. 24:206-209, Lovatt, CJ. 1990, Factors affecting frit seviearly trait drop in avocado. Calif. Avo, Soe, Yrbk, 74:193- 198, 1L,R. Goren, and B, Lesher, 1972. Ffecisof water stress, gibberellic acid and 2-chioroetyliimed ylammoniunichlocide (CCC) on flower differentiation in Eureka’ lemon tees. I. Amer Soe. Hort, Se. 97-774 78. Nunez-Elisea,R, and'T-L, Davenport, 1991. Floweting of *Keit’ mango in response to deblossoming and gibberelic acid, Proc. Fla. State Hor. Soc. 1041-13 Porligis, LC. and D. Boynton. 1961. Growth sponses orthe sawvberry plant, Fragaria chiloensis var. ananassa, to Gibherelic acid and 10 enviroamertal conditions, Proc. Amer. Soe. Hort, Sc. 18:261-269 Salazar-Garva, S., EM. Lond, and C.J, Lovalt 1998, Inflorescence and flower development of the ‘Ilass’ avocado (Persea anzericana Mill) during “on” and “of cxop years. | Amer. Soe. Hort Sci. (in pes). Scholeticl, P.B., M. Sedgley. and D.Me-P. Alexander. 1988. Carbohy- rate cycling i relation to showx growth, floral initiation and devetop- ‘rent and yield in the avocedo. Scientia fort. 25:99-110, Schroeder, C.A. 1944. The avocado inflorescence. Calif. Avo. Soc. Yebk 24:39-40. Schuch, U-K., LH. Fuchigami, and M.A, Nagao. 1990. Gibberellic acid ‘causes earlier flowering and synchronizes fruit ripening in coffee. Plant Growth Regulat 959-64, Sealey, M. 1990. Flowering of deciduous perennial fruit crops. Hort, Roy. 12:233-264, Steel, RG.D. and JH, Tore, 1980, Priaciples and Procedures of Statis- ties, A Biometrical approach. 2nd ed. MoGra-Hil, ‘Thorp, .G.,D, Aspinall,and M. Sedgley. 1993 Influence shootaye on floral development and early fui set in avocado (Persea americana Mill) ev. “Hass' J. Hort. Soi 68:645 651 Thorp, T.G.,D. Aspinall and M. Sedgley. 1984. Preformation of node umber in vegetative and reproductive proleptic shoot modules of Persea (Lauraceae). Ann Bot. 78:13-22. ‘Venning,F.0.ndF B. Lincoln, 1958, Developmental mombology ofthe vegelaliveaxisof avocado (Persea americana.) ands signiicance spacing, pruning practices, ane yields ofthe grove, Pros, Fla Stare Hort ne. THASU-336, Whiley, A.W. 1990. CO, assimilation of developing Feuiing shoots of ey Hass avocado (Persea americana Mill) —A preliminary repon. S. Aie ‘Avo, Grow. Assn. Yeh 13-28-10. Zilkab,S. I. Klein, S. Feigenbaum, and S.A. Weiehaum. 1987. Transl. ‘cation of foliar applied urea "N to reproductive and vegetative sinks of| ‘avocado and its effects on intial fut set. J. Amer. Soe, Hor, Sei 112 1061-1065, Ni 797