ABIOTIC POLLINATION AND INITIAL FRUIT SET OF THREE AVOCADO CULTIVARS. Abstract

ABIOTIC POLLINATION AND INITIAL FRUIT SET OF THREE AVOCADO CULTIVARS Melva Lucy Quiñones-Roldán 1 ; Ana María Castillo-González 1 ; Alejandro F. Barrientos- Priego 1 ; Edilberto Avitia-García 1 ; Gad Ish-Am 2 ; and Shmuel Gazit 2. 1 Posgrado en Horticultura, Departamento de Fitotecnia. Universidad Autónoma Chapingo. Chapingo, Estado de México 56230. México. 2 Faculty of Agriculture, Department of Horticulture. Hebrew University of Jerusalem. P.O. Box 12. Rehovot 76100. Israel. Abstract Under cultivated conditions in Coatepec Harinas Mexico, the presence of abiotic pollination in female and male stages of flower opening, as well the initial fruit set in absence of fertilization were studied. The study was carried out in the avocado cultivars Hass, Fuerte and Colín V-33, during two flowering periods (1998 and 1999). Branches were covered in four sides of the tree (N, S, E, W) one month before flowering with a cloth bag having openings of 2 mm of diameter, to prevent insects and permit air circulation. Flowers were collected 48 h after anthesis, for the localization of pollen tubes in pistils under optic and fluorescence microscope. The initial fruit set of bagged branches was determined five weeks after anthesis, and fruit set evaluation was carried out each month. Anatomical observations were done in small fruits, sampled from open pollination and bagged branches. In 1998 no pollinated flowers could be found in female and male flowers in the bagged branches. In 1999 only 1.7% of the sampled flowers were pollinated with a mean of 0.07 pollen grains per stigma. These pollen grains germinated, but none developed through the style. The majority of pollen grains found in the flowers from bagged branches were of the male opening and from the East part of the tree. In 1998 and 1999 there was

initial fruit set in the bagged branches (1.3 and 2.5 fruits per inflorescence, respectively). Nevertheless in the third month of evaluation all the fruits were abscised. According to our findings we state that the abiotic pollination in avocado is insignificant for its final fruit set, and that there is a need for insects to carry out pollination. The lack of embryo in all the sampled fruits from bagged branches, shows that an initial fruit development of Hass, Fuerte and Colín V-33 can be achieved without fertilization, but it cannot complete the fruit development. Key words: Persea americana Mill., self pollination, open pollination, wind pollination, reproductive biology, receptive stigma, pollen-tube, pollen. INTRODUCTION The avocado (Persea americana Mill.) has an unique and sophisticated flower behavior, that can be classified as protogineous with dicogamy synchronicity. The bisexual flower opens twice; were the first opening is functionally female and the second opening, usually in the next day, is functionally male and the pollen is shed, but the stigma is apparently not receptive (Nirody, 1922). The pollination occurs only during the female opening and carried out by pollinating insects mainly bees (Osuna, 1982; Avilán y Rodríguez 1995; Ish- Am y Eisikomitch, 1998). This type of cross-pollination is done by the pollen transfer between complementary cultivars classified as A and B type (Ellstrand, 1992). In the avocado they are many evidences of self-pollination. Cool nights or cloudy mornings delay the opening and closure of flowers during the morning. Such delay could result in the presence of flowers with female and male openings in the same tree during the first hours of the midday. In consequence of this overlap, close pollination could occur, mainly in solid blocks of the same cultivar, as in California (Bergh and Gustafson, 1958).

In opposition to the general accepted concept that the self-pollination can occur between flowers of the same tree or cultivar when there is an overlap of female and male openings. Studies in Florida, USA have demonstrated that the main form of pollen transfer was by real self pollination within the flowers. Also it was found, that the self-pollination in the male opening is the main way of pollination of commercial cultivars in Florida and that there is no need of pollinating large insects. Concluding that the avocado flowers are mainly pollinated by wind, gravity and possibly by trips (Davenport et al., 1994). The self-pollination in the male opening has been observed in Israel under dry Mediterranean climate, but under this conditions the pollen tubes don t grow (Davenport and Lahav, 1992). Sedgley (1977a) also observed self-pollination in avocado flowers in the male opening without pollen tubes reaching the ovary. Gazit et al. (1995) found that in most of the cases, the pollen germinates well when is applied to the stigma in the male opening, nevertheless, the pollen tubes do not penetrate more that the upper part of the style. The same authors indicate that after three weeks of hand pollination of flowers in the male opening of Ettinger, Reed and Gween, no normal fruits could be found; demonstrating that the pollination in the male opening in Israel is not effective. It has been observed that many of the fruit set is carried out in the end of the flowering period, where the day temperatures are between 15 and 16 C. This while the density of insects doesn t fluctuate that much and that this doesn t explain the fruit set pattern (Visscher and Gavin, 1988). In contrast Robbertze et al. (1998) indicates that the bees play an important role in avocado pollination and with out them the fruit set is very low. The wind is an important factor for avocado fruit production, because it can cause alterations in fruit set and in fruit quality. Winds higher than 10 km h -1 limits the insect flight, and if the wind is dry and warm it can negatively influence the fertilization. Also if there is cold wind, fruit set can be reduced by the restriction of bee flight and pollen tube growth (Cautín, 1996). Other studies have demonstrated the benefits of windbreak tree barriers of avocado pollinazers; this could be because the movement of pollen by bees, and

probably by wind (Cautín, 1996). According to the literature the avocado pollen is heavy and sticky to be transported by wind. Nevertheless, some studies show the benefit of wind movement in an avocado pollinizer (Bekey, 1992). In the avocado the pollination and fertilization are indispensable for the set and development of fruit (Free, 1993). In Fuerte and Ettinger a high correlation between the percentage of successful pollination (pollen tubes reaching the ovary) and the yield of both cultivars (Gazit y Shoval, 1995). In Mexico the commercial orchards are established with only one cultivar, and there is a good fruit set compared with other countries (Osuna, 1982). In Atlixco, Mexico it was found that the stigma of cv. Fuerte is receptive in the female and male openings and that probably the period of the stigma receptivity is regulated by ambient conditions (Osuna et al., 1985). The dicogamy is not well expressed because the predominant climatic conditions and it is frequent the overlap of the two flower openings in some cultivars; thus enhancing the self-pollination, pistil receptivity in different stages of flower opening and finally fruit set (Papademetriou, 1975; Osuna, 1982; Davenport et al., 1994). It is also possible that the high initial fruit set that is reached in some orchards could be caused mainly by the pollinating activity of insects and probably by other abiotic factors (Bekey, 1992; Free, 1993; Robbertze et al., 1988; Schroeder and Hofshi, 1998). The factors that affect pollination and fruit set have been studied under California, Florida and Israel conditions. In contrast under climatic conditions of Mexico there is almost no knowledge about the subject. Based on this problem the a study was carried out to evaluate if abiotic pollination is present in avocado and determinate if there is initial fruit set in absence of fertilization. MATERIALS AND METHODS The study carried out between October 1997 and April 1999, in the Experimental Station La Cruz belonging to the Fundación Salvador Sánchez Colín-CICTAMEX, S.C. Which

is situated in the municipality of Coatepec Harinas, State of Mexico at 18 46 30 North latitude and 99 46 38 West latitude, with altitude of 2,140 m. Vegetative Material Hass, Fuerte and Colín V-33 threes were used established at 5x5, 10x10 and 5x5 m, with ages of 6, 31 and 19 years old, respectively. All grafted on Mexican race rootstock seedlings. Experimental Design A complete random experimental design was used with a factorial treatment arrangement 2x3x3, being the factors year (1998 and 1999), cultivar ( Hass, Fuerte and Colín-V-33 ) and flower opening (female and male). In all cases four replicates were used, where the tree was used as experimental unit. A variance analysis was carried out using the ANOVA procedure using SAS software and for mean separation the Tukey s test (P<0.05) was used. Spontaneous Abiotic Pollination The presence of abiotic pollination suggested by Davenport (1986) was studied. For this four trees of each cultivar were selected and four branches, one in each side of the tree (N, S, W and E side), were bagged 1 m long with cheesecloth having 2 mm of opening diameters. The bags permitted air to circulate freely inside the bag and preventing most of the major insects of going inside. Flowers were collected inside the bags, 160 female opening flowers and 160 male opening flower (10 per branch) were collected. The flowers were fixed in a solution of ethanol 96%: acetic acid 100% (2:1, v/v) and the rate of pollination were determined. Each month the number of fruits of five inflorescence in bagged branches per tree were registered and a single sample of 40 fruits per cultivar were fixed in a solution of ethanol (96%) + acetic acid (100%) + formaldehyde (37%) + water (50% + 5% + 10 % + 35, respectively in volume). After two days, the fixed fruits were rinsed with water, placed in ethanol 50% for

4 hours and latter in ethanol 70% for 4 hours. After this they were conserved in glycerol + ethanol 100% + water (25%+50%+25%, respectively in volume). Pollen-tube Development Fixed flower pistils were placed in a solution of NaOH 8N for 18 hours, this permitted to soften and clear them. After the treatment they were rinsed with distillated water and the dyeing was done with solutions of lacmoid or aniline blue. The dye with 0.25% of lacmoid in ethanol 30% was applied in pistils previously divided in four parts; stigma, upper part of the style, lower part of the style and ovary. The pistil parts were placed on a glass slide and two drops of lacmoid was applied, then a glass cover was placed on top and a gentle squash with a pencil eraser was carried out. The pistil parts were observed in a light microscope Galean III, Leica) at 10x40x. In the case when aniline blue was used, entire pistils were placed on a glass slide and two drops of aniline blue were applied. A glass cover was placed on top and a gentle squash was done with the help of a pencil eraser. The pistils were observed under fluorescence light ( = 395.440) microscope equipped with a mercury lamp (Axiophot Carl Zeiss). In all the pistils the number of pollen grains, number of germinated pollen grains, number of pollen-tubes in the upper and lower part of the style, number of pollen-tubes penetrating the ovary, ovule and embryonic sac. Of the total of pistils, 75% were stained with lacmoid and 25% with aniline blue. RESULTS AND DISCUSION Abiotic pollination In both years no differences (P< 0.05) were found regarding to number of pollen grains per stigma, percentage of germinated per grain in the stigma and style, even between cultivars

and also in female and male openings. No differences were found between the four sides of the tree. Also no interaction were detected between the evaluated factors. No pollen grains were detected on the stigmas belonging to bagged flower in the female and male openings from the three cultivars evaluated during the first year (1998). In the second year only a very low number of pollen grains on the stigmas were found (0.7 pollen grains per stigma and 1.77% of the flower had germinated pollen grains. None of the germinated pollen grains developed a pollen tube throughout the stigma (Table 1). All cultivars showed inadequate pollination, this because a very low number of pollen grains per stigma. Hass, Fuerte and Colín V-33 had a mean of 0.01, 0.05 and 0.05 pollen grains per stigma, respectively. Also in the three cultivars a low percentage of flowers had pollen tubes on the stigma. In Fuerte only 1.56% of the evaluated flowers showed pollen tubes in the stigma. In Hass and Colín V-33 the percentage was lower; 0.47% and 0.63%, respectively. No development of the pollen tubes was observed in the upper part of the style (Table 2). About this concern, Papademetriou (1975) mentioned that inadequate pollination (one or very few pollen grains on the stigma) causes variation in production. In those cases the pollen tubes could have a limited development and can t reach the ovule, caused probably by a slow growth o maybe by a delay in pollen germination. Also, by the time it reaches the ovule it is possible that the egg cell is not in good condition for fertilization. In the three evaluated cultivars, most of the pistils that showed pollen grains came from flowers in the male opening (Table 3). In addition to this last, all the flowers came from flower from the East part of the tree canopy (Table 4). It is possible that the wind that comes from the East in the area where the study was carried on, shacked the bagged inflorescenses causing the transfer by gravity of the pollen to the stigma within the flower or to a stigma of a flower on a lower position. It was also observed that flowers with male opening did not started their pollen shed at the same time and it was frequent the overlap of male opening without pollen shed and with pollen shed in the three cultivars. So we could

say that the very few pollinated flower probably were many by self pollination in the male opening, something similar to was observed by Davenport et al. (1994), but we must keep in mind that this was very limited and in a very low proportion. The results show that probably there was no pollen transportation by wind between adjacent threes of the same or by other cultivars. This because in most of the female opening flower stigmas examined had no pollen grains in spite of the presence of complementary flower type cultivars in close plots. About this respect, Kirk and Sherman (1998) mentioned that the avocado pollen grains are big and sticky, this is why the insects carry it and not the wind, and when the insects are excluded the fruit set is very low. Initial Fruit Set in Bagged Inflorescences The initial fruit set of the bagged inflorescences showed differences between years, as well in cultivars (Table 5 and 7). An interaction between years and cultivars was also found. The canopy part of the tree was not significant in all the cultivars evaluated (Table 6). During the two years of evaluation Hass showed more initial fruit set (1.63 fruits per inflorescence) that Fuerte and Colín V-33 in which a mean of 0.30 and 0.14 fruits per inflorescence was found, respectively (Table 5). No differences were found between the four parts of the tree canopy in respect to initial fruit set (Table 6). Nevertheless, the East part was slightly higher that the other three and it was the side where pollinated flowers were found (Table 4). In both year of evaluation the initial fruit set was low and 1999 it was slightly higher (Table 7). In the two years a gradual fruit abscission was present and only in 1998 0.16% survived after 3 months, in 1999 none of the fruits survived. The low percentage of initial fruit set and abscission in all cultivars was directly related to the inadequate pollination and absence of fertilization in all the sampled flowers from the

bagged inflorescences, because the set and normal development of fruit depends of a good pollination as a successful fertilization. It is also known that a poor pollination could result in a low fruit set, small fruits and abortion of ovule (Gillaspy et al., 1993). Robbertze et al. (1998) found that covered trees with bees inside are more productive that without them. Anatomic cuts showed that from 29 small fruits from open pollination; 3 had embryo and/or endosperm, 10 embryo, 3 endosperm and 6 nucellus. In contrast 60 examined fruit from bagged inflorescences showed that only two had embryo in globular state, with the presence of endosperm and nucellus (Table 8). The diameter of fruit from open pollination and bagged inflorescences was similar; 3.2 and 3.0 mm, respectively (Table 9). It is possible that the fruit from bagged inflorescences that showed embryo could be derived from self-pollination when an overlap of female and male flower opening probably happened, just like Bergh and Gustafson (1958) suggests. In 1999, 120 fruits (40 per cultivar) from bagged inflorescences were examined without finding any embryo, endosperm and nucellus. The absence of pollinated flowers in the first year and the inadequate pollination of some flowers in the second year explain the lack of embryos in almost all the examined fruits in both years. It was also demonstrated that initial fruit growth could occur in absence of fertilization. The lack of embryo was the main cause of fruit abscission up to the 3 months from anthesis. Our results are similar to the report of Lovatt (1990), who mentioned that the pollination could initially stimulate the development of the ovary and the initial fruit set, but this stimulus ends in 2 or 4 weeks after pollination and after this time there is fruit abscission. The pollen during its germination, tube growth and during or after nucleus fusion produces the main stimulus for fruit growth. The factors involved in fruit growth stimulus by pollen probably include plant hormones like auxins and gibberellins (Gillaspy et al., 1993). A secondary stimulus comes from the seed in development and mainly from the endosperm, which starts to grow fast and immediately after fertilization, producing high rates of auxins (Lee, 1987).

Our results demonstrate that for Hass, Fuerte and Colín V-33, growing under Coatepec Harinas, Mexico the abiotic pollination is insignificant on fruit set, thus there is a need of presence and activity of pollinating insects. And that the initial fruit set in bagged inflorescences that lack of embryo don t develop and have abscission after some months after anthesis. References Avilan, L.; and M. Rodríguez. 1995. Epoca de floración y cosecha del aguacate en la región norte de Venezuela. Agronomía Trop. 45(1): 35-50. Avitia G., E. 1996. Anatomía precigótica y postzigótica en relación al aborto de óvulos y semillas en Spondias purpurea L. Tesis de Doctor en Ciencias. Programa de Botánica. Instituto de Recursos Naturales, Colegio de Posgraduados. Montecillo, México. 118 p. Bekey, R. 1992. To be or not to be pollination of avocados. California Grower 1: 30-32. Bergh, B. O. and C. D. Gustafson. 1958. Fuerte fruit set as influenced by cross-pollination. California Avocado Soc. Yrbk. 42: 64-66. Cautín M., R. 1996. Nuevos antecedentes sobre requerimientos de polinización y variedades. Universidad de Chile; Chile. Publicaciones Miscelaneas Agrícolas Nº 45: 15-29. Davenport, T.L. 1986. Avocado flowering Hort. Rev. 8: 257-289. Davenport, T. L.; and E. Lahav.1992. Is a pollinator required to maximize avocado production?, pp. 667-668. In: C. J. Lovatt (ed.). Proc. 2nd. World Avocado Congr., Orange, California.

Davenport, T. L.; P. Parnitzi; S. Fricke; and S. Hugues. 1994. Evidence and significance of self-pollination of avocados in Florida. J. Amer. Soc. Hort. Sci. 119(6): 1200-1207. Ellstrand, N. 1992. Sex and the single variety. Calif. Grower 1: 22-23. Free, J. B. 1993. Insect Pollination of Crops. 2nd. edition. Academic Press. Cambridge, Great Britain. 679 p. Gazit, S.; and S. Shoval. 1995. The relationship between pollination rate and yield in Fuerte and Etinger avocado. World Avocado Congress III. Program and book of abstracts. Tel Aviv, Israel. p. 32. Gazit, S.; S. Shoval; E. Katz; and D. Eisenstein.1995. Can commercial avocado cultivars in Israel set fruit after pollination at the male opening stage? World Avocado Congress III. Program and book of abstracts. Tel Aviv, Israel. p. 34. Gillaspy, G.; Ben-David, H. and Gruissem, W. 1993. Fruits: A developmental perspective. Plant Cell 5: 1439-1451. Ish-Am, G.; and D. Eisikowich. 1998. Quantitative approach to avocado pollination. Proceedings of World Avocado Congress III 1995. Tel Aviv, Israel. p. 46-51. Kirk, V. P.; and G. Sherman. 1998. Insect visitors to avocado flowers. Subtropical Fruit News 6(1): 7-10. Lovatt, C. J. 1990. Factors affecting fruit set/early fruit drop in avocado. Calif. Avocado Soc. Yrbk. 74: 193-199. Nirody, B. S. 1922. Investigations in avocado breeding. California Avocado Assn. Yrbk. 6: 65-78.

Osuna E., T. 1982. Estudio de la diferenciación floral y la expresión de la dicogámia en la variedad Fuerte de aguacate (Persea americana Mill.) en la región de Atlixco, Puebla. Tesis de Maestría en Ciencias, Programa de Fruticultura, Colegio de Postgraduados. Chapingo, México. 89 p. Osuna E., T.; García V., A.; and E. Pimienta B. 1985. Expresión de la dicogamia en la variedad Fuerte de aguacate (Persea americana Mill.) en la región de Atlixco, Puebla. Agrociencia 62: 69-77. Papademetriou, K. M. 1975b. Pollen tube growth in avocados (Persea americana Mill.). Calif. Avocado Soc. Yrbk. pp. 99-102. Robbertze, P. J.; M. F. Johannsmeier; and T. M. Morudu. 1998. Pollination of Hass avocados. South African Avocado Growers Yearbook 21: 63-68. Salazar G., S. 2000. Fisiología reproductiva del aguacate. pp. 55-83 In: El Aguacate y su Manejo Integrado. Téliz O., D. (Coordinador). Ediciones Mundi - Prensa, México, D.F. Schroeder, C. A.; and R. Hofshi. 1998. Some aspects of pollination and fertilization in subtropical fruit species. Subtropical Fruit News 6(1): 2-5. Sedgley, M. 1977a. Reduced pollen tube growth and the presence of callosa in the pistil of the male floral stage of the avocado. Scientia Horticulturae 7: 27-36. Visscher, K. P.; and S. Gavin. 1998. Insect visitors to avocado flowers. Subtropical Fruit 6(1): 7-10.

Table 1. Pollinic tube development in pistils of bagged flowers from avocado Hass, Fuerte y Colín V-33 in to year of evaluation. Year Pollen grains Percentage of flowers with pollinic tube in: (No.) y Stigma Upper part of the style 1998 0.00 a z 0.00 a 0 1999 0.07 a 1.77 a 0 LSD w 0.11 2.47 0 z Mean separation within columns by Tukey s test, P 0.05. y Mean of 960 flowers. w Least significant difference. Table 2. Pollen tube development in pistils of bagged flowers of avocado. Cultivar Pollen grains Percentage of flowers with pollinic tube in: (No.) y Stigma x Upper part of the style Hass 0.01 a z 0.47 a 0 Fuerte 0.05 a 1.56 a 0 Colín V-33 0.05 a 0.63 a 0 LSD w 0.16 3.43 0 z Mean separation within columns by Tukey s test, P 0.05. y Mean of 640 flowers. x Germination of pollen grains: 100 % w Least significant difference. Table 3. Development of pollen tube in pistils of bagged flowers from avocado Hass, Fuerte y Colín V-33, during two flower opening states. Opening Pollen grains Percentage of flowers with pollen tube in: (No.) y Stigma x Upper part of the style Female 0.003 a z 0.21 a 0 Male 0.069 a 1.56 a 0 LSD w 0.11 2.47 0 z Mean separation within columns by Tukey s test, P 0.05. y Mean of 960 flowers. x Germination of pollen grains: 100 % w Least significant difference.

Table 4. Development of pollen tube in pistils of bagged flowers from avocado Hass, Fuerte y Colín V-33 by canopy part. Canopy part Pollen grains Percentage of flowers with pollen tube in: (No.) y Stigma x Upper part of the style East 0.15 a z 3.54 a 0 West 0.00 a 0.00 a 0 North 0.00 a 0.00 a 0 South 0.00 a 0.00 a 0 LSD w 0.19 4.30 0 z Mean separation within columns by Tukey s test, P 0.05. y Mean of 240 flowers. x Germination of pollen grains: 100 % w Least significant difference. Table 5. Initial fruit set in bagged inflorescences of avocado Hass, Fuerte y Colín V-33 in two consecutive years of flowering. Cultivar Fruits per inflorescence (No.) y Hass 1.63 a z Fuerte 0.30 b Colín V-33 0.14 b LSD x 0.74 z Mean separation within columns by Tukey s test, P 0.05. y Mean of 160 inflorescences. x Least significant difference. Table 6. Initial fruit set in bagged inflorescences of avocado Hass, Fuerte y Colín V-33 by canopy part. Canopy part Fruits per inflorescence (No.) y (%) East 0.934 a z 31.293 a West 0.462 a 30.387 a North 0.514 a 31.049 a South 0.841 a 30.981 a LSD w 0.919 5.383 z Mean separation within columns by Tukey s test, P 0.05. y Mean of 120 flowers. x Least significant difference.

Table 7. Initial fruit set in bagged inflorescences of avocado Hass, Fuerte y Colín V-33 in two consecutive years of flowering. Years of evaluation Fruits per inflorescence (No.) y (%) 1 w (1998) 1.382 ab z 66.667 b 2 v (1998) 0.033 c 3.658 d 3 u (1998) 0.013 c 0.167 d 4 w (1999) 2.510 a 100.000 a 5 v (1999) 0.189 bc 15.073 c 6 u (1999) 0.000 c 0.000 d LSD x 1.224 7.169 z Mean separation within columns by Tukey s test, P 0.05. y Mean of 240 flowers. x Least significant difference. w One month after anthesis. v Two months after anthesis. u Three months after anthesis. Table 8. Anatomic examination of Hass avocado fruits from bagged and open pollinated inflorescences (1998). Inflorescences Fruits Fruits with Examined embryo and/or Normal fruits with: (No.) Endosperm (No.) Embryo (No.) Endosperm (No.) Nucellus (No.) Open pollination 29 13 10 3 6 Bagged 60 2 2 2 2 Table 9. Fruit diameters of Hass avocado coming from bagged and open pollination inflorescences. Inflorescences Fruit diameter (mm) Mean Range Open pollination 3.2 2.7 a 4.0 Bagged 3.0 2.5 a 3.2 The least diameter of a fruit without embryo and/or endosperm was 2.7 mm. The least diameter of a fruit with a normal developing seed was 2.7 mm.