|Title||Factors influencing island floras in northern Lake Michigan|
|Year of Publication||1988|
|Degree||Doctor of Philosophy|
|Number of Pages||223 pp.|
|University||University of Michigan|
|City||Ann Arbor, MI|
1. Although all factors influencing the size and composition of island floras are interrelated, the most important factor on the Manitou and Fox islands is the availability of appropriate habitat. The type and amount of each habitat granted to an Island are a result of glacial and postglacial geology and the subsequent human disturbances. 2. Island area is a factor in that large islands are larger targets for dispersed species and have the potential of having more habitats. North and South Manitou are large enough to have inland lakes and thus have habitats unavailable on the Fox Islands. Wetlands would also tend to attract potential dispersers, such as ducks, to the Islands, especially the Manitous. Intuitively, island area alone is a weak influence inasmuch as North Fox has almost as many species as South Fox, an island almost four times as large. 3. The factors influencing the distribution of native species among the Islands is different from that of alien species. Habitat shows a greater correlation with the size of native floras than does island area, assuming that the number of soil series is a close approximation of environmental heterogeneity. Alien species show little relationship with the number of soil series mapping units, but show the closest relationship with island area. The occurrence of alien species on the Islands is largely determined by the degree of present and past disturbance. Past disturbance can be determined, in part, by the number of old fields which favor the persistence of alien perennials. Present disturbance allows alien annuals and biennials already in the seed pool (and native species with similar requirements) to make an appearance on the Islands. 4. The small number of islands (four) in this study precludes any determination as to which model, exponential or power function, is the best description of the species-area relations observed here. The size of the islands and their floras also prevent the testing of the MacArthur-Wilson equilibrium theory as the Islands are too large to assure an accurate resurvey of the flora in the future. The long generation times of vascular plants and their ability to remain dormant for many years increase the difficulty of observing an equilibrium between extinction and immigration compared to vertebrates. 5. The second most important factor determining the composition of the Fox and Manitou island floras, after the occurrence of appropriate habitat on an Island, is the ease of species dispersal over the water barrier surrounding the Islands. The greater proportion of sarcochores in the native floras of South Manitou, South Fox, and North Fox suggests that birds may be an important dispersal vector. Gulls of South Manitou's gull colony make logical vectors for some species found only on that island, and the modification of the dune habitat in the colony insures the persistence of these species. Similarity values among the Islands, and between the Islands and the Mainland, show that the floras of the islands are generally more similar to each other than to those on the Mainland. The local distribution of some Mainland-limited species, some with specialized habitat requirements, may impede their dispersal to, and later establishment on the Islands. Those species which overcome the water barrier generally are found on more than one island. 6. Island climate including diminished temperature extremes, frequent fog, and uniform humidity may enhance the persistence of some island species. The lower heat accumulation on the islands retards the flowering of spring- and early summer-blooming species. This delay, especially for southern species at the northern edge of their range may protect young plants from late frosts. The lower springtime temperatures and diminished summer daily temperature extremes may also enhance the survival of northern species such as Rubus parviflorus, Tanacetum huronense, and Calypso bulbosa found in the study area only on the Islands. Fog may be a positive factor abetting the persistence of Picea glacua, found only on the Islands, an may contribute to the development of a thick bryophyte layer which is often the substrate for Camptosorus rhizophylus and Asplenium viride, ferns found in the study area only on the Islands. 7. An overwhelming influence on North Manitou is the deer population, which has selectively removed several native species and drastically reduced the abundance of others. Although deer are absent only from South Manitou, their impact has been greatest on North Manitou. Signs of grazing on Juniperus communis and J. horizontalis indicate that deer are responsible for the suppression of Arctostaphylos uva-ursi and Prunus pumila on dune "heaths", where all four species were reported in 1899. Deer are probably also responsible for the elimination of Taxus from the hardwood forests. Logging and perhaps spring floods may have induced changes in the impenetrable cedar swamp described in 1904 north of Manitou Lake, but the present vegetation (open marsh with scattered woody species) probably results from deer inhibiting regeneration of saplings. Overgrazing by deer may also have caused the retarded succession at Tamarack Lake bog, and the suppression of several species common on the other Islands and the Mainland. Despite the disappearance of some species by deer, North Manitou has the largest proportion of species found only on one island perhaps owing to a diversity of habitats, especially aquatic and wetland areas. 8. The presence of landbridges from the Manitou Islands to the Mainland during Lake Chippewa (10,400-7,700 BP, at the longest) does not appear to have had a lasting effect on the flora of these islands compared to the Fox Islands which were always surrounded by water since their emergence from the ice sheet. Indeed, the Fox Islands have several heavy fruited species such as Quercus rubra and Corylus cornuta which demonstrate the over-water dispersability of such species.