Sexy smells

Danielle holding a male junco. Notice the white tail feathers.

Danielle holding a male junco. Notice the white tail feathers.

The activities are as follows:

Animals collect information about each other and the rest of the world using multiple senses, including sight, sound, and smell. They use this information to decide what to eat, where to live, and who to pick as a mate. Choosing a mate is an important decision that requires a lot of information, such as how healthy a potential partner is, and information about their genes. Mate quality can affect how many offspring an animal has and if their genes will get passed on to the next generation.

Danielle removing preen oil from a junco.

Danielle removing preen oil from a junco.

Many male birds have brightly colored feathers that are attractive to females. For example, the peacock has bright and elaborate tail feathers, called ornaments, which are thought to communicate a male’s quality. Besides using their sense of sight to see ornaments, female birds may use their other senses to gather information about potential mates as well. Danielle, a biologist, wanted to figure out if birds use vision and their other senses, such as smell, to determine the quality of potential mates.

Danielle decided to research how dark-eyed juncos communicate through their sense of sight and smell. Dark-eyed juncos, a type of sparrow, are not colorful birds like peacocks, but they have bright white feathers in their tails. Male dark-eyed juncos have more tail-white than females. Females may use the amount of white in a male’s tail to determine whether he is a high quality mate. Danielle was also interested in several chemical compounds found in junco preen oil, which birds spread on their feathers. This preen oil contains compounds that give birds their odor. Danielle found that males and females have different odors! Just as males have more white in their tail feathers, they also produce more of a chemical called 2-pentadecanone. Danielle wanted to test whether this chemical might be a signal of mate quality.

A preen gland where birds produce preen oil.

A preen gland where birds produce preen oil.

To test her two alternative hypotheses, Danielle captured male juncos at Mountain Lake Biological Station in Virginia. She measured their amount of tail-white by estimating the proportion of each tail feather that was white, and adding up the values from each feather. She also took preen oil samples and measured the percent of each sample that was made up of 2-pentadecanone. She followed these birds for one breeding season to find out how many offspring they had. If females pick mates based on visual ornaments, then she predicted males with more tail-white would have more offspring. If females pick mates based on smell, then she predicted males with more 2-pentadecanone would have more offspring.

Featured scientist: Danielle Whittaker from Michigan State University

Flesch–Kincaid Reading Grade Level = 9.4

To learn more about Danielle’s work with juncos, see her blog posts “The sweet smell of (reproductive) success” and “Deciphering avian aromas” on the BEACON website. To learn more about Danielle and her research, check out this episode from the PBS/NOVA webseries “The Secret Life of Scientists and Engineers” where she was featured.

There is a scientific paper associated with the data in this Data Nugget. The citation and PDF for the paper is below.

Whittaker, D., N.M. Gerlach, H.A. Soinic, M.V. Novotnyc, and E.D. Ketterson (2013) Bird odour predicts reproductive success. Animal Behaviour 86(4): 697-703

Average Rating


  1. Maira Goytia says:

    I teach Animal Behavior, a Junior-Senior level class, in the Biology department at Spelman College.
    We used successfully this case study during our section on Mating Behaviors. We used the case study in a 50-minute class-period. The students were highly engaged. We used the opportunity to discuss deeply on how to determine whether a correlation is significant, how to interpret both types coefficient: the Pearson product-moment correlation coefficient (rho) and the coefficient of determination (R^2). The students understood how to determine statistically whether the data would support a correlation. The other aspect was that students were not expecting the visual cue not to correlate. Thus they were exposed to getting a result that did not match their prediction, and were forced to go a little further and think about the causes and describe future experiments.
    The last question on the study was done individually, but I think that giving the students the opportunity to discuss it in groups would have been more productive as I got answers that went all over the place.
    Finally, it is very well prepared for the instructors. Minimal time put in to prepare the class. Reading the original article allowed me to get some more information and share it with the students that seemed interested in going deeper in the topic.


Speak Your Mind