Limiting factors of four rare plant species in Ola`a forest of Hawai`i Volcanoes National Park.

VanDeMark, Joshua R.
Pratt, Linda W.
Euaparadorn, Melody
Journal Title
Journal ISSN
Volume Title
Starting Page
Ending Page
Alternative Title
Three endangered or candidate endangered plant species native to `Ōla`a Forest (Cyrtandra giffardii, ha`iwale; Phyllostegia floribunda, a mint with no common name; and Sicyos alba, `ānunu) were studied for more than two years to determine their stand structures, short-term mortality rates, patterns of reproductive phenology, success of fruit production, seed germination rates in the greenhouse, presence of soil seed bank, and survival of both natural and planted seedlings. The role of rodents as seed predators was evaluated for S. alba using seed offerings in open and closed stations. A fourth endangered species (Cyrtandra tintinnabula) at a remote site in `Ōla`a was visited to determine its stand structure and mortality rate. Cyrtandra giffardii displayed a stable population structure with many adults and few small or very large plants; the monitored population had a mortality rate of 7.3% over three years. Mortality of plantings from 2003-2004 in a re-introduced population of Phyllostegia floribunda was 21.4%. The stand structure of C. tintinnabula indicated a relatively stable population with both small and large plants present and a short-term mortality rate between visits of 14.5-17.0%. Four groups of S. alba vines were monitored; 3 of these have persisted in place for at least 15 years. All species monitored had annual patterns of flower and fruit phenology, although male inflorescences of S. alba showed a subannual pattern. Successful transition of flowers to fruit was high for P. floribunda (51.5%), moderate for C. giffardii (23.3%), and undetermined for S. alba. High percentage viability was demonstrated for seeds of P. floribunda and S. alba (78.5-100% positive to strongly positive in tetrazolium tests), but seed viability was not tested for C. giffardii. Greenhouse germination rates were high for P. floribunda (88.0-92.0%), but variable and relatively low for C. giffardii (0-19.3%) and S. alba (4.0-11.1% in 2007 and 0 in 2008). No soil seed bank was detected for S. alba in 3 seasonal samplings, but P. floribunda was found to have a viable seed bank in April that persisted from at least the previous summer. Rodent predation of S. alba seeds was 93.3% in fruit offerings in accessible bait stations. Mortality of natural seedlings was high for both P. floribunda (90.2%) and S. alba (69.7%). Planted seedlings of P. floribunda produced flowers and fruit in their first year, and reproduction was higher in sunny plots than in shady plots. Mortality was high in both planting treatments, and 2 survival rates did not differ significantly in sun and shade (χ 2 = 0.48, df = 1, p = 0.490). Three planted seedlings of S. alba survived for 12-16 months but did not reproduce. Floral visitors were observed at C. giffardii and P. floribunda using digital video cameras and recorders. In almost 200 hours of observation, no visitors entered the flowers of C. giffardii, although one very small insect, either a micro-wasp (Hymenoptera) or fly (Diptera) was seen on the exterior of a corolla. In almost 300 hours of video observation, three floral visitors were identified at P. floribunda flowers. Honeybees (Apis mellifera) were likely pollinators, as they contacted both anthers and stigma of flowers. The mean visitation rate of honeybees was 0.003 visit/flower/hour, and visit duration ranged from 2 to 17 seconds. Fruitflies (Drosophilidae of undetermined species) crawled around flower interiors, but did not seem to forage for either nectar or pollen. Fruit fly mean visitation rate was 0.006 visit/flower/hour, and visit duration was 28 to 1,424 seconds. The third observed insect visitor was an endemic geometrid moth caterpillar (Lophoplusia giffardi), which was seen feeding on foliage and flowers of P. floribunda. In conclusion, two of the three regularly-monitored rare plant species of `Ōla`a Forest appeared to have more than one limiting factor inhibiting the natural increase in their populations, while for P. floribunda the most important factor was high seedling mortality. Most plants of the monitored C. giffardii population appeared to be hybrids, probably with the more common species C. lysiosepala. Seed germination rates were low, and natural seedlings were not observed. Pollinators were not seen in many hours of observation, indicating that cross pollination is a rare or uncommon event. The re-introduced population of P. floribunda had relatively low mortality, and reproduction was successful with high rates of fruit formation from abundant flowers. Seed germination rates were high, and a soil seed bank was detected. Natural seedling recruitment was observed, but high seedling mortality indicated that this life stage was the most vulnerable in the species. The population of S. alba was small and the vine life form precluded an accurate estimate of the number of adult plants in `Ōla`a Forest. Natural dormancy was likely a factor in the observed low rate of seed germination. No soil seed bank was detected, and alien rodents were implicated as seed predators. Natural recruitment was observed at multiple sites in `Ōla`a, but seedling mortality was high. The cause of seedling mortality was not identified.
Endangered species, seed bank, Hawaii
Geographic Location
Time Period
Related To
Table of Contents
Rights Holder
Local Contexts
Email if you need this content in ADA-compliant format.