Teachers Survey Orchid Bees in the Amazon

Field Notes From the Amazon

Amazon Research Initiative for Educators

Field Report: 2023 Orchid Bee Survey

Date: July 12 – 19, 2023
Location: Explorama’s Napo-Sucusari Biological Reserve, Loreto Peru
Study Site: The Amazon Conservatory for Tropical Studies (ACTS) Canopy Walkway. Platform #2. Forest floor and canopy.
Lead Researcher: Dr. Robert Naczi
Team Members: Maddie Beitler, Cesar Sevillano Guerra, Kathy Hockman, Shannon King*, Mike McAloon, Garly Aliaga Pashanaste, Roldan Hidalgo Peso Dawn Porta, Kathy Richardson* (*author of this report)


With the high level of biodiversity in the Amazon, many species are still unknown, let alone monitored for changes in these diversity patterns over time and space. Scientists often use indicator species to learn about other species’ diversity and monitor change. In the process of building a long-term inventory of orchid bee diversity of ACTS, we will test hypotheses about their distribution and species richness for the forest floor and the canopy. In this specific research project, we will be examining and monitoring orchid bee biodiversity and abundance. The long-term aspect of this project will enable us to look for environmental trends.

Materials and Methods
To attract bees, we began with cotton balls tied on a string. Then, we saturated each cotton ball with a concentrated and unique scent and placed at a minimum of two meters apart. We placed scents in alphabetical order at each site, in order to track the particular scent that attracted each bee. The scents were Benzyl Acetate, Eucalyptol, Eugenol, Methyl Salicylate, and Vanillin. We placed each set of five at the forest floor level and the canopy platform level (canopy platform #2). The stations were monitored daily by team members. At the start of each day, we saturated and hung a new set of cotton balls. On the second day, a new cotton ball was placed near the first day’s bait, leaving both baits suspended, in case the first ball maintained attractiveness. On the third day, an additional set of cotton balls were added to each bait site, leaving all cotton balls suspended. The fourth day, we removed the first and second-day baits and left the third-day bait suspended. The fifth day we replaced the third day, leaving the fourth day still suspended.

We captured orchid bees by hand in ziploc bags, which we immediately sealed and marked with the date, scent, and location. For subsequent captures, we utilized new bags, which we marked in the same manner. Start and stop times for each level were synchronized. However, the duration of time observed was not standardized between capture dates.

Figure 1. Shannon King using ziploc bags and scented cotton balls to collect bees.

Orchid bees were frozen overnight for pinning and drying the following day. The bees were placed in a Schmitt box for preservation and transportation.

Figure 2: Pinned bees in a Schmitt box.
Figure 3. Closeup example of pinned orchid bees in the Schmitt box
Figure 4: Local naturalist guides, Roldan and Garly, work on Orchid Bee ID
Figure 5. ARIE members using Celestron Micro Direct Digital Microscope to investigate the anatomy of orchid bees

We documented four genera of orchid bees during the sample timeframe. Euglossa dominated the species collected by far, but Eulaema had a notable representation as well. Exaerete did not appear until the fourth and fifth day of collection. We collected only a few specimens of Eufriesea, and no Aglae.

During three sampling days, we collected a total of 103 bee specimens. We collected 26 orchid bee specimens in the canopy and 77 specimens on the forest floor.

Table 1: Numbers of specimens collected at each bait at the two sampling locations.

Hypothesis #1:  No differences exist in orchid bee frequency between rainforest floor and rainforest canopy.  Chi Square Value: 30.  p value: 0.000001. Accordingly, we reject this hypothesis.

Figure 6:  Elevation Preferences of Orchid Bees

Hypothesis #2: No differences exist in orchid bee frequency collected from each of the five baits.
Chi Square Value: 40. p value: <0.000001. Thus, we reject this hypothesis.

Figure 7:  Frequency of Bees Collected at Each Bait

Due to varying weather conditions, time spent collecting data was highly variable. In future experiments, sample times and duration should be uniform throughout the experiment.

Another consideration would be canopy placement because the canopy platform was heavily trafficked, and some of the samples were moving quite a bit more than others because of the movement of the walkway when people passed by. Another variable may have been people stopping on the platform to birdwatch and their proximity to the baits might have influenced the bee’s behavior.

Abiotic data collection may need to be taken into account, as the wind, light, precipitation and temperature were highly variable on each day, and may have affected the collections.

Other recommendations include collectors recording the time at which each species was captured, and also the specific elevation at each collection site should be more uniform.

Next Steps
Identify the species by morphology under microscopes and DNA analysis. Most remaining identifications are for Euglossa species. Collected specimens will be deposited in the Museo de Historia Natural de Universidad Nacional Mayor de San Marcos, Lima.

For additional information or questions about this field report, please contact Dr. Lindsey Swierk (lindsey.swierk@actsperu.org), Director of Scientific Research for The Morpho Institute and Associate Director of Research for the Amazon Conservatory for Tropical Studies. 

Learn more about the Amazon Research Initiative for Educators (ARIE)

Download Syllabus

Share this with your friends: