We present an analysis of the long-term survival of two cohorts of seedlings of the tropical canopy tree Ocotea whitei (Lauraceae) on a 1-ha plot of mature, lowland moist forest on Barro Colorado Island, Panamá. In 1980, we counted an even-aged cohort of seedlings that germinated in 1979, then measured and tagged survivors in 1981. We also measured and tagged a second, smaller cohort of seedlings that germinated in 1981. We followed the subsequent survival of all seedlings through 1985. Seedling mortality was phenotypically, temporally, and spatially non-random. Important correlates of non-random mortality included: (1) seedling size and age, (2) an El Niño drought, and (3) biotic neighborhood. Larger and older seedlings survived better than smaller and younger seedlings, respectively, and the El Niño-related drought of 1982-1983 was associated with elevated mortality rates. Seedling density, which was strongly correlated with the proximity to the nearest conspecific adult, increased mortality. The observed mortality patterns suggest that processes consistent with the Janzen-Connell hypothesis operate during the recruitment phase of O. whitei population dynamics. However, the processes causing the observed density- and distance-dependent mortality may vary with factors such as total seed number, seedling size, and climatic variation, making it difficult to determine whether time-integrated seedling-to-adult spacing mechanisms other than self-thinning operate on a given plant population. After 6 years in the hectare studied, survivors remained densest and most numerous underneath the adult trees. We conclude that only long-term demographic data, collected at a variety of scales on a variety of species, will ultimately answer the question: do Janzen-Connell effects contribute substantially to structuring tropical forests?

When there are two types of points in a plane, their "relative dispersion" is the tendency for one type to be located near or far from the other. We present a method for describing and testing relative dispersion. Our motivation in developing this test was the need to analyze the dispersion of juvenile plants relative to conspecific adults. To demonstrate the usefulness of the method, we simulated plant populations under randomness and under five alternative "dispersion—generating processes." The results from analyses of these populations illustrate the method and its use in the interpretation of dispersion pattern. In addition, these results provide insight into the types of patterns resulting from these dispersion—generating processes.

This study examines the influence of a selected set of determinants of contraceptive method switching in rural Sri Lanka. Of interest is the question of how change in contraceptive practice at the individual level can account for patterns observed at the aggregate level. Based on calendar data on contraceptive use over a 3-year period, collected for more than 3,000 married women in a 1986 survey, the multivariate analysis shows that women who attain all or a significant proportion of their desired fertility tend to switch to more effective methods. Women who experience method failure tend to switch methods, usually to a type that is more effective. The woman's background determinants of age and education have small but significant effects on method switching, whereas the effect of household economic well-being is not significant. There is strong indication that rural couples are practicing contraception in a nonrandom fashion, switching methods in accordance with changes in their fertility motivations and contraceptive experience.

Information from 12 studies is combined to estimate the AIDS incubation distribution with greater precision than is possible from a single study. The analysis uses a hierarchy of parametric models based on a four-parameter generalized F distribution. This general model contains four standard two-parameter distributions as special cases. The cases are the Weibull, gamma, log-logistic, lognormal distributions. These four special cases subsume three distinct asymptotic hazard behaviors. As time increases beyond the median of approximately 10 years, the hazard can increase to infinity (Weibull), can plateau at some constant level (gamma), or can decrease to zero (log-logistic and lognormal). The Weibull, gamma and 'log-logistic distributions' which represent the three distinct asymptotic hazard behaviors, all fit the data as well as the generalized F distribution at the 25 percent significance level. Hence, we conclude that incubation data is still too limited to ascertain the specific hazard assumption that should be utilized in studies of the AIDS epidemic. Accordingly, efforts to model the AIDS epidemic (e.g., back-calculation approaches) should allow the incubation distribution to take several forms to adequately represent HIV estimation uncertainty. It is recommended that, at a minimum, the specific Weibull, gamma and log-logistic distributions estimated in this meta-analysis should all be used in modeling the AIDS epidemic, to reflect this uncertainty.

No abstract