Online resource 1 Apidologie Introducing nests of the oil-collecting bee Centris analis (Hymenoptera: Apidae: Centridini) for pollination of acerola (Malpighia emarginata) increases yield

1. 
   Universidade Federal do Ceará- UFC, Departamento de Zootecnia – CCA, Campus Universitário do Pici, Bloco 808. 60.356-000 Fortaleza – CE, Brazil.
   

The experiment was carried out from October to December 2010 in the Amway Nutrilite do Brasil farm situated in the Ibiapaba plateau, country of Ubajara, state of Ceará, Brazil, at 3º 51' 16"S and 40º 55' 16"W, 847.5 m above sea level (IPECE, 2010). In the county of Ubajara the climate is humid, average mean temperature around 25ºC and average mean rainfall is 1,483,5mm per year (IPECE, 2010).

Cultivated area in the Amway Nutrilite do Brasil farm reached 319.5ha, of which 255 ha are planted with 55 acerola varieties from which we chose two organic plantations (A and B) situated 2km apart from each other and bearing 13 year-old trees (Figure 1). We chose these plantations for their similarity in many aspects (same varieties, same orchard layout, same agricultural practices, etc.), similar surroundings and fruit productivity.

Plantation A was limited to the East and South by another acerola plantation but of distinct varieties, to the West by the native caatinga (scrub vegetation) and to the North by a 6 ha cashew (Anarcardiun occidentale L.) plantation. Plantation B was limited to the East by a 35 ha coconut (Cocos nucifera L.) plantation, to the West and South by other acerola plantation of distinct varieties and to the North by the native caatinga (scrub vegetation) (Figure 1).

Each plantation (A and B) comprised 15ha of acerola sub-divided in 22 orchards of 0.68 ha each planted with four acerola varieties: AC71 (8 orchards in each plantation), FP19 (6 orchards), AC69 (4 orchards) and AC13/2 (4 orchards) (Figure 1).

Figure 1 – Satellite image showing the spatial arrangements of acerola (Malpighia emarginata) plantations A and B, their orchards and distance between them. Ubajara, Ceará, Brazil. Source: GoogleEarth, 2012.

Figure 2 – One of the 22 organic orchards of plantation A showing the six acerola (Malpighia emarginata) lines, windbreak tree line and manual weeding. Ubajara, Ceará, Brazil. Source: Amway Nutrilite do Brasil.

i.) Learning about the diversity of floral visitors in acerola plantations

Diversity of floral visitors was assessed using sweep nets to sample any visitor to acerola flowers in each plantation (A and B). An orchard per plantation was randomly chosen and a 50m-long transect was set in each orchard. Then, these transects were walked at every 3 hours, from 6:00 to 18:00 h, and all floral visitors seen were captured. This procedure totalized five collects per day in each orchard and was carried out for six non-consecutive days in October 2010.

Insects captured were killed in a lethal chamber containing ethyl acetate and preserved in 70% ethylic alcohol. Later, they were pinned and dried out in a oven at Bee Laboratory of the Universidade Federal do Ceará, before sending for identification by Dr. Fernando Zanella in the Universidade Federal de Campina Grande, state of Paraíba, Brazil.

ii) Using acerola plantations to multiply the population of oil-colleting bees

Based on the results of phase I (see the results session), we distributed 14 empty trap-nest blocks all over the farm to be colonized by Centridini bees before acerola blooming in plantations A and B.

Trap nests were made by 8 x 150mm grooves excavated in pieces of wood (140 x 150 mm) which were placed one on the top of the other to form a block up to seven stacks high. Each nesting block (a, b, c … up to n) had 96 trap nests distributed in 16 columns and six lines. The 14 blocks had a total of 1,344 trap nests (Figure 3).

After being colonized, inhabited nests were introduced to plantation A. Trap-nest blocks were distributed along a straight line in the shade of the windbreak trees between cashew and acerola plantations. Each block was placed 1.5 m high and 25 m apart from each other, in a total of 14 trap nest blocks distributed along a 360 m windbreak line in the direction East-West (Figure 4).

Figure 3 – Schematic drawing of the trap-nest blocks used, highlighting the grooves excavated in pieces of wood being placed one on the top of the other to form a block up to seven stacks high.

Data were collected monitoring the bee activity all-day long and recording the foundation of new nests in the trap-nest blocks.

Figure 4 – Satellite image of plantation A showing the positioning of the 14 tap-nest blocks (a to n) and windbreak trees (w).

To learn whether Centris bees could be used to increase acerola yield, we compared fruit yield from the 22 orchards where inhabited trap nests were introduced (Plantation A), to that of the 22 orchards where no trap nests were placed (Plantation B). During the experiment, fruits were manually harvested in a daily bases and the individual tree fruit production was recorded to allow comparisons per orchard, variety and area between plantations.