A cryptic species complex in the ant parasitoid Apocephalus paraponerae (Diptera: Phoridae)

by Shellee Morehead, Jon Seger, Don Feener and Brian Brown
Cryptic species are distinct in all the usual ways except that they look the same even to experts. Prior to about 20 years ago cryptic species were thought to be uncommon, for the obvious reason that they were not easy to recognize. But the increasingly widespread application of molecular genetic methods to studies of population structure and systematics has revealed that many long-established "species" are really two or more fully isolated and ecologically differentiated forms. We recently discovered that a well known and well described ant-decapitating parasitoid fly, "Apocephalus paraponerae" is in fact a complex of at least four species. Even though these sister species are virtually identical morphologically, they appear to have been separated from each other by as much as seven million years. This finding supports the emerging view that species diversities may be much higher than previously believed, at least in certain taxonomic groups.
Apocephalus paraponerae males and females mate, feed and oviposit on injured worker ants. Larvae develop inside the ant, migrate to the head and pupate there, eventually causing the head to separate from the thorax. The flies seen here have been attracted to an injured Paraponera clavata, but populations of this "species" are also attracted to ants in the genera Ectatomma and Pachycondyla.

poster figure
We began to suspect that flies attracted to Paraponera and Ectatomma might belong to different species when we noticed that individuals collected from Ectatomma workers did not respond to 4-methyl-3-heptanone and 4-methyl-3-heptanol. These compounds are produced in the mandibular glands of P. clavata, and we had previously shown that they serve as long-range olfactory cues for flies attracted to P. clavata. Additional behavioral experiments at study sites in Costa Rica and Panama confirmed that flies associated with the two host ant species respond to different chemical cues, and a careful morphometric study showed that flies associated with E. tuberculatum are slightly but significantly smaller, on average, than those associated with P. clavata, even though both ants occur and attract "A. paraponerae" at both study sites.

To test the implication that there are actually two distinct populations of flies, each specialized on a different host ant, we sequenced a portion of the mitochondrial COI gene from several individuals attracted to each ant species at each study site, and also from individuals attracted to Paraponera in Colombia and Ecuador, and to Pachycondyla in Colombia. We were not surprised to find substantial genetic differences between the flies attracted to the three different ants, but we were very surprised to find differences just as large between some populations of flies attracted to the same host ant, as summarized in the figure below.
Relationships and average genetic distances of "A. paraponerae" collected from three species of host ants at various locations in Central and South America. Several individuals from each collection were sequenced, and they never showed more than minor differences. The scale of distance (Ks) is synonymous substitutions per synonymous nucleotide position.

poster figure
The four populations of Paraponera flies form a monophyletic group, but the Central American populations appear to be well separated from the South American populations. The biggest surprise by far was finding that Ectatomma flies in Costa Rica are only distantly related to Ectatomma flies in Panama, which are more closely related to all of the Paraponera flies and to the Colombian population associated with Pachycondyla. The genetic distances among these apparently distinct populations suggest that they have been separated for times on the order of one to seven million years, based on calibrations of the COI molecular clock for other Diptera. Many recognized species show distances no greater than these, so we infer that these populations of "A. paraponerae" represent at least four different species.

In retrospect, it makes sense that Apocephalus should readily speciate, because their ecology includes all of the features believed to promote host-race formation: they meet, mate, and reproduce at a well defined host resource, and they use highly specific chemical cues to locate the host. Thus a host switch can give rise immediately to strong reproductive isolation and then to selection for improved performance on the new host. It will be interesting to see how many more cryptic species of Apocephalus there really are!

Morehead SA, Seger J, Feener DH Jr, Brown BV (2001) Evidence for a cryptic species complex in the ant parasitoid Apocephalus paraponerae (Diptera: Phoridae). Evolutionary Ecology Research 3:273-284
Related poster
Do island endemics suffer increased genetic loads?
by Kevin Johnson and Jon Seger

To Jon Seger's research interests
To Donald H. Feener Jr.'s research interests

To a list of all posters in the session

Directory | Calendar | BioNews | Faculty | Graduate Study | Undergraduate Study
Programs & Facilities | About the Department | Biology Home