Viability of pollen by pollen tube germination

When the anthers were developed were observed with a magnifying glass and recorded the day when these were opened, when the pollen was fresh was taken and was grown in Petri dishes containing medium based on agar (4 g L^-1) and sucrose (0.06 mol L^-1). It was cultured in incubation at 36 °C; the number of germinated pollen tubes was observed at 3, 6, 12, 24, 48 and 72 h. The count of germinated pollen grains per box was determined to determine the percentage of germination and the time required to reach at least 1% of germination. The length of the pollen tubes was measured by using a (Carl Zeiss^®) compound microscope observing 10 to 40 diopters, in order to know the time required for germination and pollen fertility measuring. A completely randomized design was used, with three replicates per variety, each repetition consisted of a Petri dish containing the three anther pollen grains taken at random.

Viability of pollen by staining

The viability of the pollen was also evaluated by staining with an acetic-carmine acid solution. In each plant, two anthers were collected per flower (before pollen release). These were placed in one tube per plant and fixed in Farmer solution (3:1, absolute ethyl alcohol, glacial acetic acid), in solution were stored in refrigeration until evaluation. For the study, an anther was placed on a slide, cut transversely, a drop of aceto-carmine dye was added and with the blade of a scalpel the anther was gently crushed to release the pollen into the dye drop, coverslip and allowed to elapse 5 to 10 s to observe the microscope ((40x) (Carl Zeiss^®); for the pollen grain count seven preparations were made per plant, in each preparation four fields were observed in which the number of stained grains and the number of uncoloured pollen grains were evaluated. With these data the percentage of viability of pollen in each variety was determined. The study was conducted in a completely randomized experimental design with seven replications per variety.

Receptivity of stigma

Stigma receptivity was found by modifying the method of ^{Osborn et al. (1988)} which consisted of placing a drop of hydrogen peroxide (H₂O₂) of 80% of the stigmas and take a positive result bubble production, the adequacy of the method was performed because this concentration was greater reaction and stigma bubbling was more efficiently observed. This experiment was carried out in a completely randomized design with five replicates (plants) per variety, in each plant four inflorescences were taken.

Moment of pollination

Since there is no antecedent of the receptivity moment of the stigma, pollinations were carried out in three schedules of the day (8:00, 14:00, 18:00) taking as a parameter the amount of nectar present in the nectar of the hermaphrodite flowers, which was measured with a micro-pipette of 1 000 µL, being performed in the nine varieties, in four inflorescences of five plants. In addition, the test was performed ^{Osborn et al. (1988)}, to know stigmatic receptivity. The synchronization was observed between the amount of nectar and the receptivity of the stigma. A completely randomized design with five replicates (plants) was used per variety, the receptivity of the stigma and volume of the nectar were evaluated.

Type of pollination

To determine the type of pollination of the poinsettia plants, four treatments were evaluated: 1) emasculate hermaphrodite flowers and cover completely to avoid the arrival of external pollen; 2) cover the female flowers and when they were receptive pollinate with pollen from another plant of the same variety and then re-cover, remove the male flowers as they appeared; 3) cover the female flowers and when they were receptive they were pollinated with pollen from the same plant and covered again; and 4) cover the female flowers and when they were receptive they were pollinated with flower pollen from plants of another variety and covered again. The experiment was developed in a completely randomized design, with five replicates (plant) per variety, pollinating 15 flowers per replicate. The percentage of mooring of fruit per plant was evaluated.

Analysis of data

In the five experiments, data were studied by analysis of variance, mean comparison test with Tukey test (p≤ 0.05) with program SAS 9.2 (^{SAS, 2002}).

Viability of pollen by pollen tube germination

In the poinsettia varieties the gynoecium style had a length of 6 mm (Figure 2). The pollen tube germination in vitro start at 6 h after initiating the culture but was up to 72 h when it reached 6 mm needed to transport the male gametes from the stigma to the ovary.

Figure 2 A) length of poinsettia style, measured with ruler and millimeter paper; and B) anatomical cut of a female flower where the length of the style is observed. 

The analysis of variance showed highly significant differences in germination of the pollen tube of the nine varieties of poinsettia. The poinsettia of sun had the highest germination of pollen tube (>90%), four of the shade cultivars had 80-84% of viable pollen grains; Nutcraker white presented the lowest viability of pollen (Table 1). The above indicated that the varieties of sun could be good males, the opposite happened with Nutcracker white. These values were higher germination 30 and 17% observed by ^{Lyra et al. (2011)} in the study conducted with Jatropha molissima and J. ribifolia.^{Narbona et al. (2005)} observed an average of 31.8% of pollen viability in E. boetica, which was also lower than what was observed in the present investigation.

Table 1 Viability of pollen by germination of pollen tube at 36 °C and by staining with acetic acid-carmine in nine varieties of poinsettia. 

Valores con la misma letra son estadísticamente iguales de acuerdo con la prueba de Tukey (p≤ 0.05).

As for the time required for germination of pollen tube, ^{Cavalari (2010)} conducted a study in Euphorbia with Codiaeum variegatum, and observed that pollen grains of this genus reached maturity at 65 h in the medium of germination compared to the 72 h required by the poinsettia varieties of this research. This indicated that it was required about 72 h after pollination for fertilization to occur.

Viability of pollen by staining

In the preparations made to observe under the microscope two types of stains were presented, the grains of pollen that showed intense coloration were the viable grains, while the almost transparent pollen grains were the sterile grains (Figure 3). The viability of pollen by staining was higher (91 to 93%) in the sun varieties; four shade varieties had 80-85% viable pollen, while Prestige and Nutcraker white showed the lowest values (Table 1).

Figure 3 A) viable pollen grain on poinsettia; and B) nonviable pollen grains, observed at 10 diopters. 

The viability of pollen by carmine staining was very similar to that obtained with the germination of the pollen tube; therefore, either methodologies were reliable to evaluate the male gametes (Table 1); given the above, it was considered that the assessment of viability by staining pollen grains was adequate because it is a simpler methodology.

Receptivity of stigma

The sunflower varieties Bethlehem, John Paul and Dawn Christmas had a higher percentage of flowers (43, 43.7 and 46.3) that presented bubbling, which showed the receptivity of the stigma (Figures 4 and 5) without differences between them. In contrast, shade varieties had a lower percentage of receptive flowers, of which Sonora White glitter was the one with the least receptivity (24%), whereas Prestige and Nutcraker white did not show bubbling in the stigma, indicating absence of receptivity.

Figure 4 Receptivity of the stigma on poinsettia. A and B) bubbling in the stigma. ^{Osborn method (1988)}.  

Figure 5 Receptivity of the stigma in seven varieties of poinsettia at 8:00, 14:00 and 18:00 h.  

As for the time when the stigmas were receptive, it was observed that with the exception of Nutcraker white and Prestige, all the varieties had more receptive flowers at 14:00 h, in the majority of the varieties the less receptivity was to the 8:00 in the morning.

The results suggest that poinsettia was receptive at the time of day when the weather was warmer (14:00 h) (Figure 5), which could indicate that in hotter places, stigmatic receptivity could occur before the hour of the highest receptivity observed in this research, where the average annual temperature is 21.5 °C, according to the above, ^{Campos (2013)} observed that the best time to perform the pollination at Christmas in Zacatepec, Morelos (24 °C annual mean temperature ) were at 11:00 am as they obtained the highest number of fruits per plant (1.85) and seeds per fruit (2.56) compared to the pollination at 12:00 of the day, in which 1.1 fruits per plant were harvested and 0.83 seeds per fruit.

Considering the results obtained by ^{Lozada and Herrero (2006)}, who mentioned that stigmatic receptivity directly affected reproductive success, the Prestige and Nutcracker white varieties that did not show receptivity to the stigma were not able to bear fruit, on the contrary, poinsettia of sun presenting a higher percentage of receptive flowers could form more capsules. Based on the above results, the sun varieties are the most viable candidates for being females in a genetic improvement program at poinsettia.

Moment of pollination

The receptivity of the stigma was lower at 8:00 am in the seven varieties that showed receptivity of the female organ, while it was higher at 2:00 pm, indicating that this time corresponded to the right moment to perform the pollination; the three varieties of poinsettia of sun were the ones with the highest percentage of receptive flowers (43 to 46%) (Figure 5). Prestige and Nutcracker white did not show stigma receptivity in any of the hour in which the evaluation was done, indicating that these two varieties could not function as female or male progenitors in a breeding program, since they also had the lowest percentage of pollen viability.

In all varieties, except for Nutcracker white and Prestige, it was observed that nectar began to flow from the nectaries in male flowers after pollen dehiscence; while the stigmatic receptivity test was positive three days after pollen dehiscence (Figure 6); the above indicated absence of synchronization in the maturity of male and female structures. This may be the reason why poinsettia is cross-pollinated, as ^{Augspurger (1983)} mentions reproductive success is influenced by the degree of synchrony of flowering events.

Figure 6 Floral asynchrony on poinsettia. Male flowers with dehiscence of pollen and nectaries, immature hermaphrodite flower, still does not present stigmatic receptivity  

The amount of nectar was higher in sunshine varieties > 38 µL, while in the shade varieties it was 13.5-16.2 µL lower (Table 2). This characteristic was related to the receptivity of the stigma, which was also higher in the sun varieties (Figure 6). The volume of the nectar could be the parameter indicating the moment of pollination. However, the pollen dehiscence is more visible and the receptivity of the hermaphrodite flowers was observed three days after the release of the male gametes, pollen dehiscence being a better indicator to identify the right moment for pollination. In this regard, ^{Vargas et al. (1999)} noted that in E. heterophylla pollen release was greater with increasing temperature increase occurring after noon or late afternoon, this in Viçosa, Brasil.

Table 2 Volume of nectar in nine varieties of poinsettia.  

Valores con la misma letra son estadísticamente iguales de acuerdo a la prueba de Tukey (p≤ 0.05).

The adequate time for pollination is variable depending on the place, as the dehiscence of pollen grains was affected by temperature and relative humidity of cultivation, as reported by ^{Lisci et al. (1994)} who pointed out that the time of day for anther dehiscence varied over time and appears to be influenced by temperature and relative humidity.

During the development of the experiment, several pollinating insects were observed in the sun varieties between 12:00 and 16:00 h, when the nectaries had on average 35 µL of nectar; orders of insects that were observed were hymenoptera, diptera (family Tachinidae), coleoptera and lepidoptera monarch butterfly (Danaus plexippus) (Figure 7) the latter was the only observed in the varieties of shade.

Figure 7 Poinsettia pollinating insects: A) hymenoptera; B) diptera; C) coleoptera; and D) lepidoptera.  

Type of pollination

The treatment 1 that consisted of emasculating flowers and covering them to avoid the arrival of foreign pollen, there was no fruit formation, which indicated that poinsettia did not produce seeds by apomixis.

In treatment 2 that consisted of eliminating the male flowers as they appeared and pollinated the female flowers with pollen from another plant of the same variety, more than two fruits, having highly significant differences with the other varieties, which generated on average less than one fruit, which reinforced that poinsettia is a cross-pollinated plant.

In treatment 3 where the female flowers were covered and pollinated with flower pollen from the same plant and covered again, no fruit formation was observed which indicated that it was possible that on poinsettia there would be auto incompatibility of gametes.

Finally, in treatment 4 where the female flowers were covered and pollinated with flower pollen from another poinsettia variety, there was fruit formation, indicating that the type of pollination of the poinsettia was crossed, since most fruits were formed of the varieties, with the exception of Prestige and Nutcracker white. The highest number of fruits was observed in the sun, especially in Belen (2.6 fruits on average), while the varieties Freedom, Burgundy and Ice punch formed 0.85 fruits on average (Figure 8), the fruits were formed in the flowers hermaphrodites.

Figure 8 Fruiting on nine poinsettia varieties pollinated with pollen from another variety (cross-pollinating).