Reference: Biol. Bull-174: 109-125. (April. 1988) An Assessment of Poecilogony in Marine Invertebrates: Phenomenon or Fantasy? K. ELAINE HOAGLAND* AND ROBERT ROBERTSON Department of Malacology. The Academy of Natural Sciences, Philadelphia, Pennsylvania 19103 Abstract. Poecilogony, denned as more than one mode of reproduction within a single species, has been reported in various invertebrates, including mollusks and poly-chaetes. Many cases that have been described involve planktotrophic and non-planktotrophic development in allopatric populations, or instances of planktonic larval stages and benthic juveniles being found together, but not associated with adults. There is always the possibility of mis-assignment of larvae to adults of the wrong spe-cies. Most cases that offer these kinds of evidence are now known to involve cryptic species, not poecilogony. There are a few species in which release of young oc-curs both at metamorphosis and a day or so before. There may be cases in which extracellular yolk or nurse-egg production is variable and allows the release of larvae at different stages, but no actual instance is known. A few instances are known of allopatric populations with different modes of development and other differences in reproductive characters that lack reproductive isolation when brought into the laboratory. The polychaetes Streblospio benedicti, Cirriformia tentaculata, Boccardia proboscidea, and the opisthobranch Elysia chlorotica are in this category. All examples of poecilogony require fur-ther genetic substantiation. Despite the scarcity of proven examples of poecilo-gony, the presence of more than one mode of reproduc-tion within a genus is the rule in most invertebrate phyla. The evolutionary and ecological significance of these patterns is discussed. Introduction There have been numerous reports of more than one pathway of development within a single species of ma-rine invertebrate. Quotes such as these are common: "It Received 25 June 1987; accepted 25 January 1988. * Person to whom reprint requests should be addressed. is an established fact that some animal forms have a different mode of development under different biological conditions" (Mortensen, 1921, p. 241). "Studies . . . cast further doubt on an already-failing dogma that a particular larval mode is a set and inflexible species char-acteristic" (Hadfield, 1972). The term poecilogony was invented to describe the phenomenon. ". . . poecilo-gony was first mentioned by Giard (1904) [sic], who gives a number of examples of its occurrence . . . which, as far as the echinoderms are concerned, have proved to be wrong. . . . That the ability exists is, however, a fact." (Thorson, 1950, pp. 29-30). If poecilogony exists, it is of interest to evolutionary biologists. However, further investigation has often revealed that cryptic species were responsible for the supposed poecilogony. The word poecilogony derives from the Greek, poikil-, various, and goneia, reproduction. Even Webster's [un-abridged] Third New International Dictionary ( 1 97 1 ) ex-pressed doubt as to whether poecilogony occurs: "A sup-posed method of development occurring in invertebrate animals when in the same species there are two kinds of young although the adults are exactly alike" (italics ours). In this paper, we first define the phenomena we mean to include under the term poecilogony. We then review reported cases of poecilogony known to us in ma-rine invertebrates, particularly mollusks and poly-chaetes. We also demonstrate that congeneric species in many invertebrate groups differ in type of larval develop-ment, hence the mode of larval development has a com-plex history within most invertebrate phyla. The likeli-hood that poecilogony is real is assessed and theoretical implications are discussed. We suggest ways of testing the hypothesis of poecilogony in some cases where the data are insufficient for a judgment to be made. Materials and Methods We searched the published literature for references to poecilogony. We categorized the data in each paper as to 109
BioStor
