The plumule morphology of species of Ceratophyllum was studied for purposes of: 1) determining extent of variability in forking of first epicotyl leaves, and 2) evaluating plumule features as taxonomic characters by assessing interspecific and intraspecific (interand intrapopulational) variation. Embryos from specimens collected throughout ranges of eight of species (Ceratophyllum australe, C. demersum, C. Ilerenae, C. muricatum, C. oryzetorum, C. platyacanthum, C. submersum, and C. tanaiticum) exhibited no variation in morphology of first epicotyl leaves. all instances leaves were simple. addition, no intrapopulational variation in was detected in C. demersum, C. llerenae, or C. submersum (rarity of material prohibited study of intrapopulational variation in other five species). contrast, embryos from specimens collected throughout range of ninth species, C. echinatum, possessed forked leaves at all nodes of plumule. C. echinatum, interpopulational variability was observed in number of segments produced by forked plumule leaves at first node. The segment number is not correlated geographically, and combinations of segment number can be observed within a single population. The unusual plumule morphology of C. is distinctive within genus and has probably resulted from phylogenetic reduction of lower plumule axis. These observations support recognition of C. as a distinct species. The embryo morphology of Ceratophyllum has been interpreted variously since early studies of its structure in 18th century. Gaertner (1788) provided a preliminary description of embryo, but regarded two cotyledons as a separate organ, vitellus (Aboy 1936; Schleiden 1837). Richard (1808) described embryo of Ceratophyllum with four cotyledons rather than two. Brongniart (1827) considered possibility that in Ceratophyllum, cotyledons represent outgrowths of radicle, in which case first foliage leaves constitute true cotyledons. Schleiden (1837, 1838) was first to identify correctly two cotyledons of Ceratophyllum based upon his discovery of epicotyledonary internodes in some mature seeds. Most botanists eventually recognized embryo of Ceratophyllum with two cotyledons, above which a pair of simple, decussate plumule (epicotyl) leaves arise. Arber (1920) further remarked that forked leaves do not occur in plumule until fourth node above cotyledons, but are preceded by juvenile (simple, linear) leaves. 1936, however, an important observation was made by Aboy (1936) who studied Ceratophyllum demersum and poorly known eastern American species C. echinatum. She reported a difference in leaf morphology at first plumule node between two species (Aboy 1936, p. 25): In young plant of C. demersum var. echinatum [=C. echinatum], the first pair of leaves may fork once or twice .... young plant of C. demersum, . . . first pair of leaves never fork[s]. Aboy based her conclusion upon examination of only one collection of fruits of C. echinatum, so a fundamental distinction between plumule-leaf forking characters of individuals of C. demersum and C. throughout their ranges could only be implied. Evidence supporting Aboy's work appeared soon afterwards in a paper by her mentor, W. C. Muenscher (1940), who concluded that not only did seedlings of C. demersum and C. exhibit different plumule morphologies at first node above cotyledons (as Aboy described), but at second node as well. At that level, C. demersum consistently produced simple leaves only, whereas seedlings of C. produced forked leaves at all plumule nodes. As Arber (1920) had already observed, he found that forked leaves did not occur in C. demersum until third or fourth node above cotyledons. Muenscher (1940, p. 232) regarded these seedling differences to be most constant character with which to separate C. demersum and C. echinatum,