Biologists have been arguing about what species are for as long as they have been grouping organisms into species, and I don't propose to solve the problem. Fortunately, we don't necessarily have to agree with one another. The ``hybrid policy'' proposed by the U.S. Fish & Wildlife Service in 19961 notes that
The [Endangered Species] Act does not attempt to define ``species'' in biological terms, and thus allows the term to be applied according to the best current biological knowledge and understanding of evolution, speciation, and genetics.
The biological species concept2 has been the most widely accepted and influential species definition for most of the last sixty years. In the last twenty years, however, systematists are increasingly inclined to define species in phylogenetic terms, either as minimal (or at least very small) monophyletic clades (the history-based conception) or as population systems with fixed, diagnosable differences (the character-based conception).
In one sense, it might not seem to matter how we define species. In fact, many conservation biologists are now focusing on the protection of ``evolutionarily significant units''3 precisely because systematists can't agree on how to define what species are. Remember, however, that the U.S. Endangered Species Act states specifically that
the term ``species'' includes any subspecies of fish or wildlife or plants, and any distinct population segment of any species or vertebrate fish or wildlife which interbreeds when mature.
That means that ``evolutionarily significant units'' without a formal latinized name can be protected only if they are vertebrates, e.g., grizzly bear, timber wolf, and certain salmon runs in the Pacific northwest. Plants and invertebrates can be protected only
if they have latinized names that can be applied to them. Of course, it doesn't matter whether they are recognized as subspecies (or varieties in plants), so long as they are formally recognized. It does mean, however, that if you work with plants or invertebrates and identify an evolutionarily significant unit worthy of protection, you (or a systematist friend of yours) will have to put a formal name on it if you want it to receive protection under the U.S. Endangered Species Act.
Moreover, Collar  points out that the species concept we adopt could have a large impact on the conservation decisions we make. In birds, whose taxonomy is better understood than that of any other group of animals or plants, using a phylogenetic species concept instead of a biological species concept could double or triple the number of species recognized [3, p. 131]. The result would undoubtedly be a large increase in the number of bird species that we recognize as threatened, which may or may not be a good thing. Similarly, the ``Conservation Forum'' on the status of the black sea turtle that appeared in volume 13 of Conservation Biology4 makes it clear that deciding not to recognize a population as evolutionarily distinct may have important conservation consequences - that population may no longer receive special conservation attention.