THE EVOLUTION OF PARASITES FROM THEIR HOSTS - A CASE-STUDY IN THE PARASITIC RED ALGAE

Morphological similarities of many parasites and their hosts have led to speculation that some groups of plant, animal, fungal, and algal pa rasites may have evolved directly from their hosts. These parasites, w hich have been termed adelphoparasites in the botanical literature, an d more recently, agastoparasites in the insect Literature, may evolve monophyletically from one host and radiate secondarily to other hosts or, these parasites may arise polyphyletically, each arising from its own host. In this study we compare the internal transcribed spacer reg ions of the nuclear ribosomal repeats of species and formae specialis (host races) included in the red algal parasite genus Asterocolax with its hosts, which all belong to the Phycodrys group of the Delesseriac eae and with closely related nonhost taxa of the Delesseriaceae. These analyses reveal that species of Asterocolax have evolved polyphyletic ally. Asterocolax erythroglossi from the North Atlantic host Erythrogl ossum laciniatum appears to have evolved from its host, whereas taxa i ncluded in the north Pacific species Asterocolax gardneri have had two independent origins. Asterocolax gardneri from the host Polyneura lat issima probably arose directly from this host. In contrast, all other A. gardneri formae specialis appear to have originated from either Phy codrys setchellii or P. isabelliae and radiated secondarily onto other closely related taxa of the Phycodrys group, including Nienburgia and ersoniana and Anisocladella pacifica. Gamete crossing experiments conf irm that A. gardneri from each host is genetically isolated from both its host, and from other A. gardneri and their hosts. Cross-infection experiments reveal that A. gardneri develops normally only on its natu ral host, although some abberrant growth may occur on alternate hosts. The ability of red algal parasites to radiate secondarily to other re d algal taxa, where they may become isolated genetically and speciate, suggests that this process of speciation is not a ''genetic dead end' ' but one that may give rise to related clusters of parasite species.