Pollination Efficiency of Ceratina cyanea (Hymenoptera: Apidae) on Helianthus annuus (Asteraceae) Flowers at Dang (Ngaoundere, Cameroon)

To evaluate the impact of Ceratina cyanea on fruit and seed yields of sunflower, experimental field was set up from June to July in 2016 and 2017 at Dang. Observations were done on 540 capitula divided in four treatments: two treatments differentiated according to the presence or absence of protection on capitula regarding Ceratina cyanea and other flowering insects’ visits; the third with capitula protected and uncovered when florets were opened, to allow C. cyanea visits and flowers bagged uncovered and rebagged without the visit of insects or any other organisms. Bee’s daily rhythm of activity, its foraging behavior on florets and its pollination efficiency were evaluated. Bee’s daily rhythm of activity, its foraging behavior on flowers and its pollination efficiency were evaluated. Results show that, C. cyanea foraged on Helianthus annuus flowers throughout its whole blooming period. Among 32 insect species recorded on H. annuus capitula, C. cyanea ranked second ac-counting for 10.79% all visits, after Apis mellifera (76.06%). On florets, individual bees intensely


Introduction
The small carpenter bees (genus Ceratina Latreille), often with metallic colors are solitary bees of small size (3 -15 mm) belonging to the family Anthophoridae [1]. They are a widespread and moderately diverse lineage of Xylocopinae with approximately 350 species [2]. They have submarginal cells with anterior wings, unlike their neighbors, the Allodapini which have only two [1]. Species typically nest in pithy stems, forming relatively simple linear nests therein [3], making them ideal for observation of intra-nest behaviors and studies of social interaction [4]. All collect pollen [1].
The sunflower, Helianthus annuus is native of eastern North America [5]. It is widely distributed in Western Canada, Northern, Central and Southern regions of the United States and Northern Mexico [6]. This plant provides the fourth oilseed consumed in the world after palm, soybean and rapeseed oils [7]. The sunflower is resistant to drought, cold and heat [8]. It is also used in food and medicine [5]. It is a plant cultivated mainly for its seeds which constitute a source of oil [5] and its meal used in human and animal food [9]. Global production of this oilseed is estimated at 18,891 million tons in 2017 and Ukraine is the leading producer [10].
In Africa in general and in Cameroon in particular, the demand in seed oil of sunflower is very highly whereas it seed yield is weak because notably of the insufficiency of knowledge on the relations between this plant and the anthophilous insects in many agro ecological zones.
The few data published at the end of in-depth studies carried out on the interactions between insects and H. annuus are those of [11] in Bulgaria, [12] and [13] in Soudan, [14] in America, [15] and [16] in Pakistan, [17] in Kenya, [18] and [19] in Turkey, [20] in Cameroon, [21] and [8] in India, [22] in Israel, [23] in France. In all these investigations, foraging and pollination activity was carried out in detail only on Apis mellifera bees. Other Apoids such as bumble bees and wild bees visit the sunflower flowers and participate in their pollination [23] [19]. In all these investigations, the foraging behavior and pollination activity was carried out in detail on A. mellifera.
The flowering insects and their impact on pollination and fruit and/or seed yields of a plant species may vary with the species of insect, time and space [3] [24]. Hence there is a need of other studies in the Adamaoua region, to supplement existing data.
The general objective of this work is to contribute to the understanding of the relationships between H. annuus and C. cyanea, for their optimal management. Specific objectives were to: 1) determine the place of Ceratina cyanea in H. an- The vegetation near the H. annuus field had various wild and cultivated species. The experimental plant material was represented by fruit of H. annuus sampled from the surrounding of the Unit of Applied Apidology. Ceratina cyanea individuals of the experimental station were recruited among the arthropods naturally present in the environment.

Sowing and Weeding
Each year from April to May 2016 and 2017, the experimental plot was delimited, ploughed and divided into 8 subplots, each measuring 8 * 4.5 m 2 . Four seeds were sown per hole on 9 lines. There were 20 holes per subplot. Holes were separated 40 cm from each other, while lines were 50 cm apart. Weeding was performed manually as necessary to maintain plots weeds-free.

Estimation of the Frequency of Ceratina cyanea Visits on
Helianthus annuus Capitula Specimens of all insect taxa were caught using insect net on unlabeled flowers and conserved in 70% ethanol, excluding butterflies that were preserved dry [29], for subsequent taxonomic identification.

Study of the Activity of Ceratina cyanea on Helianthus annuus Florets
In addition to the determination of the flower visiting insect frequency, direct observation of the foraging activity of Ceratina cyanea on florets was made in the experimental field. The floral products (nectar or pollen) harvested by C. cyanea during each floret visit were registered based on its foraging behavior.
Nectar foragers were seen extending their proboscis on to the base of the corolla and the stigma, while pollen gatherers scratched the anthers using their mandibles and their legs [29] [30].
In the morning of each sampling day, the number of opened florets carried by each labelled capitula was counted. During the same days as for the frequency of visits, the duration of individual floret visits was recorded (using a stopwatch) according to six times: 7 -8 h, 9 -10 h, 11 -12 h, 13 -14 h, 15 -16 h and 17 -18 h. Moreover, the number of pollinating visits which was defined as visits with contact between the bees and stigma [31], the abundance of foragers (highest number of individuals foraging simultaneously per floret, per capitula and per 1000 florets) [20] and the foraging speed (number of florets visited by individual bee per minute [31]) were recorded during the same dates and daily periods as the registration of the duration of visits. The abundance of foragers per floret and per capitula was noted following the direct counting. For the abundance per 1000 florets (A 1000 ), the number of individuals of C. cyanea was counted on a known number of florets at the moment x. The abundance per 1000 florets is calculated using formula: where F x and A x are respectively the number of bloomed florets and the number of active foragers at the moment x [28]. The foraging speed (F s ) was calculated using the formula: , where d u is the duration (sec) given by a stopwatch, and F l , the number of florets visited during d u [28].

Evaluation of the Effect of Insects Including Ceratina cyanea on Helianthus annuus Production
Parallel to the installation of treatments 1, and 5, 840 capitula with florets at bud stage were protected in 2016 and 2017, to form three treatments: • treatment Y [treatment 2 in 2016 or 6 in 2017]: 240 capitula protected using gauze bags net to prevent insect visitors; • treatment C [treatment 3 in 2016 or 7 in 2017]: 400 capitula protected using gauze bag nets to prevent insect or any other organism visits and destined to be visited exclusively by Ceratina cyanea. As soon as the first florets were opened on each capitulum of treatments 3 and 7, the gauze bag was gently removed and this capitulum was observed for up to 10 minutes; the visited capitula was marked and protected again; • treatment Z [treatment 4 in 2016 and 8 in 2017]: 200 capitula protected using gauze bag nets and destined to be uncovered then rebagged without the visit of insects or any other organism. As soon as the first florets were opened on each capitulum of treatments 4 and 8, the gauze bag was carefully removed and this capitulum observed for up to 10 minutes, while avoiding insect or any other organism visits. At the maturity, achenes were harvested and counted from each treatment. The fruiting rate, the percentage of fruits with seed and the percentage of normal (well developed) seeds of unprotected capitula were then determined for each treatment.
The evaluation of the effect of insects including C. cyanea on H. annuus production was based on the impact of flowering insects on pollination, the impact of pollination on H. annuus fruiting, and the comparison of fruiting rate, the percentage of fruits with seed and the percentage of normal seeds of treatments 1, 2, 4, 5, 6 and 8.
For each year, the fruiting rate due to the foraging insects including Ceratina cyanea (Fri) was calculated using the following formula: where FX and FY are the fruiting rates in treatment X (capitula left for free pollination) and treatment Y (capitula protected from insect visits), and Eg the effect of the gauze bag net which can be calculated using the formula Eg FZ FY = − , where FZ is the fruiting rate in treatment Z (capitula protected then unbagged and rebagged without insect or any other organism visit). Finally, where b is the number of achenes formed and a the number of viable florets initially set [20].
The impact of flower visiting insects including C. cyanea on fruits with seed and normal seeds were evaluated using the same method as mentioned above for the fruiting rate.

Assessment of the Pollination Efficiency of Ceratina cyanea on Helianthus annuus
The contribution of C. cyanea in the fruiting rate and the percentage of normal achenes was calculated using data of treatments 3 and 4 in 2016 and those of treatments 7 and 8 in 2017.
For each year, the contribution of C. cyanea in fruiting rate (Frc) was calculated using the formula where FC is the fruiting rate in treatment C (capitula visited exclusively by C. cyanea).

Data Analysis
Data were analyzed using descriptive statistics, student's t-test for the comparison of means of two samples, Pearson correlation coefficient (r) for the study of the association between two variables, chi-square (χ 2 ) for the comparison of two percentages, using Microsoft Excel 2010 software and R-commander, version 2.13.0.

Frequency of Ceratina cyanea Visits on Helianthus annuus Flowers
Among the 2744 and 8756 visits of 21 and 31 insect species recorded on H. annuus flowers in 2016 and 2017 respectively, Ceratina cyanea ranked second with 8.12% and 11.60% of visits. The first place was occupied by Apis mellifera both years (74.06% and 76.53%) ( Table 1). The difference between the percentages of C. cyanea visits in 2016 as well as in 2017 is highly significant (χ 2 = 26.56; df = 1; P < 0.001).
The weak frequency of visit of C. cyanea on H. annuus capitula compare to that of A. mellifera could be explained by the strategies adopted by this social bee that consist of recruiting a great number of workers for the exploitation of an interesting food source [32]. Consequently, there might be a limitation of the number of individuals C. cyanea (1241) on H. annuus capitula due to the occupation of the majority of open florets by A. mellifera workers (8747) [20].
The presence of other plants species with bloomed flowers able to attract Ceratina cyanea could also explained the weak frequency of this solidary bee on H. annuus florets. The significant difference between the percentages of Ceratina cyanea visit in 2016 and 2017, could be explained by the presence of a significant number of its nests (10 nests) in 2016 than in 2017 (6 nests) close to the experimental plot.

Floral Products Harvested
Individuals of Ceratina cyanea were seen collecting nectar ( Figure 1) and pollen ( Figure 2) on H. annuus florets. Nectar collection was regular and intensive whereas pollen collection was less intensive. On 639 visits recorded in 2016, 545 (85.29%) were devoted to exclusive nectar harvest and 94 (14.71%) to pollen harvest. In 2017, on 526 visits, 411 (78.14%) were devoted to exclusive nectar harvest and 115 (21.86%) to pollen collection. For the two cumulated years of 1165 visits recorded, 1075 (82.06%) were devoted to exclusive nectar harvest and 209 (17.94%) to pollen harvest. Nectar and pollen were harvested during all scheduled time frames.

Daily Rhythm of Visits
Ceratina cyanea was active on H. annuus florets from 8 am to 5 pm in 2016 and in 2017, with a peak of visits between 12 and 13 pm in 2016 as well as in 2017 ( Figure 2). In 2016, the correlation was significant between the number of Ceratina cyanea visits and the temperature (r = 0.81; df = 4; P < 0.05) and not significant between the same number of visits and relative humidity (r = −0.71; df = 4; P > 0.05) (Figure 3(a)). In 2017, the correlation was not significant between the number of C. cyanea visits and the temperature (r = 0.75; df = 4; P < 0.05) and between the number of these visits and relative humidity (r = −0.76; df = 4; P < 0.05) (Figure 3(b)). ( )

Rhythm of Visits According to the Flowering Stages
Ceratina cyanea visits were numerous on the H. annuus flowers when the number of opened florets was highest (Figure 4). The correlation between the number of C. cyanea visits and the number of H. annuus opened florets was not significant (r

Abundance of Ceratina cyanea
In 2016, the highest mean number of Ceratina cyanea individuals simultaneous in activity was 1 per floret, 1.54 per capitula and 14.55 per 1000 florets (Table 2).
In 2017, the corresponding figures where 1 per floret, 1.55 per capitula and 12.70 per 1000 florets ( Table 2). For the two cumulative years the mean number of C. cyanea individuals was 13.86 per 1000 florets. The difference between the mean number of this bee per capitula was not significant (t = 0.22; df = 394; P > 0.05) and 1000 florets (t = 1.84; df = 254; P > 0.05) in 2016 and 2017.
The high abundance of C. cyanea individuals per 1000 florets proves the good attractiveness of H. annuus nectar and/or pollen for C. cyanea. The attractiveness for sunflower nectar and pollen could be partially explained by the highest availability and the accessibility of these products.

Duration of Visits per Floret
The mean duration of a Ceratina cyanea visit per H. annuus floret varied significantly according to floral product harvested (Table 3).
In 2016, the mean duration of a floret visit for nectar harvest was 3.68 sec (n = 545; s = 2.20) and for pollen collection it was 9.63 sec (n = 94; s = 8.16). In 2017, the corresponding figures were 3.55 sec (n = 411; s = 1.82) for nectar and 9.53 sec (n = 115; s = 3.92) for pollen. For the two cumulative years‚ the mean duration of visit on a floret was 3.62 sec (n = 956; s = 2.01) for nectar collection and 9.58 sec (n = 209; s = 6.04) for pollen harvest. The difference between these two means was highly significant (t = 14.06; df = 1163; P < 0.001).
The significant difference observed between the mean duration of a pollen harvest and that of nectar harvest could be explained by the accessibility of each  of these floral products. Pollen is produced by the anthers, which are on the top of the stamens, whereas nectar is between the base of the style and stamens. Under these conditions, an individual bee must spend much more time on a floret to obtain its nectar load, compared to the time needed for the collection of pollen.

Influence of Fauna
Individuals of Ceratina cyanea were disturbed in their foraging activity by other individuals of the same bee species or those from other species, which were the competitor for H. annuus nectar and/or pollen. In 2016, of 838 visits, 10 (1.19%) were interrupted by A. mellifera and 2 (0.24%) by individuals of C. cyanea whereas in 2017, of 619 visits, 7 (1.13%) were interrupted by A. mellifera and 1 (0.16%) by individuals of C. cyanea. In order to obtain their nectar or pollen load, individuals of C. cyanea who suffered such disturbances were forced to visit more florets and/or capitula during the corresponding foraging trip. For the pollen foragers, these disturbances resulted in partial loss of carried pollen.

Pollination Efficiency of Ceratina cyanea on Helianthus annuus
During pollen and/or nectar harvest in sunflower's florets, individuals of C. cyanea were regularly in contact with anthers and stigma, increasing there by possibility of H. annuus pollination. The comparison of the fruiting rate (Table 4) showed that the differences observed were highly significant between treatments 1 and 2 (χ 2 = 8432.36; df = 1; P < 0.0001) and treatments 3 and 4 (χ 2 = 7352.85; df = 1; P < 0.0001).
Hence, in 2016 and 2017, the fruiting rate, the percentage of fruits with seed and the percentage of normal seeds of capitula protected and visited exclusively by C. cyanea were higher than that of capitula protected, opened and closed without a single visit.
In 2016, the numeric contribution of C. cyanea on the fruiting rate, the percentage of fruits with seed and the percentage of normal seeds via a single capitula visit were 69.75%, 63.94% and 87.17% respectively. In 2017, the corresponding figures were 29.04%, 57.18% and 64.70%% respectively. For the two cumulated years, the corresponding figures were 49.40%, 60.56% and 75.87% respectively.
On the same plant, regarding A. mellifera, [8] in India and [20] in Cameroon have revealed that the percentage of seed setting (86.9% and 62.21% respectively) due this Apidae through its pollination efficiency was significantly higher over the pollination without insects.
Thus in Dang, in view of the published works, H. annuus appears to be a plant species which benefits enormously from pollination by insects among which Ceratina cyanea is the most important and harvested nectar and pollen.

Recommendation
The protection of Ceratina cyanea nests at the vicinity of sunflower fields to in-