A lot of info about this subject has already hit the newsstand, but I'll
add a few comments based on what others have said. I studied Gambusia and
many of its relatives when I lived in Florida.

I don't think there's much evidence that any kind of swallow can control
mosquitos. I think the stories about Purple Martins in the east were made
up by martin lovers. Mostly, I don't think mosquitos *can* be controlled
by vertebrate predators except in rather small bodies of water where
specific ones (such as mosquitofish) occur without *their* predators.
Drainage canals were dug through huge areas of Florida salt marshes to
allow mosquitofish and others of their family access to mosquitos that were
otherwise in fishless shallow waters, and it worked to some degree, but not
entirely. They also did massive spraying.

Mosquitos are likely to be less common in ponds that have a good complement
of both fish and invertebrate predators than they are in shallow temporary
waters that are relatively predator-free. How to make this work for the
wetlands under discussion?

Boxes for swallows around wetlands should be for Tree Swallows, as they
nest over water and are much more characteristic of wetlands than are
Violet-greens.

I think it's in fact illegal to release mosquitofish anywhere in Washington
state. At least that's what I was told once.

Gambusias are not a problem in Florida; they're native there, and all
through the southeast west to Texas, where a second species occurs. But
there is a *tremendous* problem with introduced fish in Florida, as Don
Baccus wrote, and in many parts of the Everglades all you see are
introduced species when you look into the water! This was happening when I
left there in the 60s, and it's gotten much worse since. It broke my heart
when I visited in the 80s.

For some reason, introduced fish have the potential to have MUCH greater
effects on ecosystems than do introduced birds, something about aquatic
systems, I guess. Horror stories abound about fish introductions, while
for the most part, birds have actually had rather little effect on native
species or ecosystems. So mosquitofish are in fact potentially much worse
than starlings.

Mosquitofish are surface feeders, which is why they're great for mosquito
control, as the mosquito larvae spend a lot of time in the open at the
surface. Many of the native fish species are midwater and bottom feeders
and wouldn't be as likely to be such effective predators on mosquitos.

I think Gambusia can be called killifish, as they are in a family
Poeciliidae that is closely related to the family Cyprinodontidae, the
official "killifish" family. The Poeciliidae are also called "topminnows."
The two, together with another small family, are in a distinctive order
Cyprinodontiformes, which occurs widely in tropical America and Africa.
Gambusia and other topminnows are indeed live-bearers, with copulation and
internal fertilization (they seem to have made an advance beyond the bird
condition :-).

Finally, I agree wholeheartedly with Jon Anderson's assessment of the
situation, especially the part about how inappropriate it would be to
introduce an *exotic* species to control a *native* species. Having grown
up in southern Florida, I am almost apoplectically against introduced
species and introducing species. There is never justification to do so,
except for very well understood host-specific herbivores or parasites on
plants and insects that are of great economic importance and can't be
controlled in any other more benign way. A generalized predator such as
Gambusia will eat any animal food it can find, and if it is a superior
competitor (and it certainly has a high reproductive potential) to some
small native fish, then by introducing it we've made still another boo-boo.

And we've got enough boo-boos in Congress with which to contend.

Dennis Paulson, Director phone: (206) 756-3798
Slater Museum of Natural History fax: (206) 756-3352
University of Puget Sound e-mail: dpaulson at ups.edu
Tacoma, WA 98416