Jess K. Zimmerman. 1991. Ecological correlates of labile sex expression in the orchid Catasetum viridiflavum. Ecology 72:597-608.
Supplements
Supplement 1: Contingency tables used to analyze ecological correlates of labile sex expression in the orchid Catasetum viridiflavum. Tables of infloresence numbers by environmental conditions.
Ecological Archives E072-001-S1 (ESPS 9002).
Description
INTRODUCTION
During 1985 and 1986 I followed the flowering of a Panamanian population of the unisexual, epiphytic orchid Catasetum viridiflavum Hook. Data in the contingency tables presented below were collected to address two questions related to the evolution of labile sex expression in this orchid: 1) What is the association of habitat (substrate type and light availability), plant size, and inflorescence production, and how are habitat and plant size associated with plant gender? analyses of these tables and their interpretation appear in my article "Ecological correlates of labile sex expression in the orchid Catasetum viridiflavuin" (Ecology, vol. 72E31).
METHODS
Catasetum viridiflavum were located along approximately 14 km of trail and 4 km of shoreline on Barro Colorado Island, Panama (9009'N, 79051'W). Locations of 248 plants were recorded during the 2 yr study period. At intervals of 2-4 wks between 15 June and 8 Dec. 1985 and 15 May - 15 Dec., 1986 I recorded the number and sex of inflorescences produced by each plant. Complete flowering records for both years were obtained for only 186 plants as a result of mortality or because plants became obscured by vegetation. In October 1985 I recorded the type of substrate on which each plant occurred (standing dead tree, dead limb of living tree, or living tree), and subjectively estimatedtheir relative plant size (four categories) and relative canopy openness (four categories). These classifications were used for all analyses (similar results were obtained when I repeated the classifications the following year). Eight plants were excluded from some analyses because large changes in the canopy around these plants (e.g., adjacent tree fall) prevented me from assigning them to a single canopy openness category over both years.
Data for flowering frequency and sex expression in relation to habitat characteristics (substrat~, canopy openness) and plant size were studied using contingency tables (Tables 1-7) analyzed using log-linear models (Fienberg 1980, Proc Catmod, SAS 1987). The tables were created by combining categories to provide adequate cell sizes. Fienberg (1980) suggests that an average cell size of 4-5 is "adequate" for log-linear analysis (as long as the observations are not confined to only a few cells in the table). Cells registering a zero frequency were adjusted to a value of lE-20 before analysis, as suggested in Proc Catmod (SAS 1987).
Relative canopy openness and relative plant size were combined into two categories (two lowest categories vs. two greatest categories), as was substrate (dead vs. live wood). Analyses of flowering frequency in individual years (Tables 1,2) were conducted using two categories: nonflowering vs. flowering plants. Similarly, sex expression in individual years (Tables 4,5) was analyzed comparing female and male plants. Hermaphrodites, those plants producing a mixture of male and female flowers, were excluded from these tables because they were infrequent within years. For plants observed for two yrs, total inflorescence production (Table 3) was adjusted to three categories: no flowering, 1-2 inflorescences, and 3-7 inflorescences. Sex expression in the combined data set (Table 6) was also adjusted to three categories, i.e., male, female, and hermaphrodite. An additional analysis of plant size and sex expression was conducted by recombining size classes into the smallest vs. the three largest categories (Table 7) and eliminating the substrate distinction. This was done to more finely distinguish the potential effects of plant size on sex expression.
LITERATURE CITED
Fienberg, S.E. 1980. The analysis of cross-classified categorical data. Second Edition. Massachusetts Institute of Technology Press, Cambridge, Massachusetts, USA.
SAS Institute, Inc. 1987. SAS/STAT guide for personal computers, Version 6 edition. Cary, North Carolina, USA.
The original paper or microfiche copies of supplements submitted prior to 1998 are on file in the Publications Office of the Ecological Society of America.
ESA Publications | Ecological Archives | Permissions | Citation | Contacts