Ecological Archives M077-003-A1

Wayne P. Sousa, Peter G. Kennedy, Betsy J. Mitchell, and Benjamín M. Ordóñez L. 2007. Supply-side ecology in mangroves: do propagule dispersal and seedling establishment explain forest structure? Ecological Monographs 77:53–76.

Appendix A. Methods and results of an experiment testing the effects of paint marking and salinity on buoyancy of mangrove propagules.


To determine if our marking technique affected propagule buoyancy, possibly biasing estimates of dispersal distance, we compared the rates and duration of flotation for painted and unpainted propagules of each mangrove species. Further, in our study areas, water salinity varies markedly in space and time as a function of tidal stage and rainfall, so propagules are exposed to a variety of salinities as they disperse. We therefore tested for the independent and interactive effects of paint marking and salinity on patterns of propagule flotation. Specifically, we used a two-way factorial design to test for the effects of marking with fluorescent paint (three levels: red ServiStar®, pink Plasti-kote®, or none) and water salinity (two levels: fresh and brackish). The six treatment combinations were replicated twice for each species. Groups of painted or unpainted propagules were floated in water-filled plastic tubs. Tubs used for trials with Avicennia and Laguncularia measured 20 cm in diameter and 15 cm deep, while those used for Rhizophora were 30 cm in diameter and 20 cm deep. The numbers of propagules per replicate tub was 20 for Avicennia, 18 for Rhizophora, and 15 for Laguncularia. For each species, the size-distribution of propagules was the same in all treatments and replicates, and matched that used in the field dispersal trials. Water in the tubs was changed at each sampling; the fresh water tubs received rainwater (0 ppt salinity), while the brackish water tubs received water of 16 ppt salinity. Evaporation caused the salinity of the brackish water treatment to rise as high as 20 ppt between sampling times.

The Avicennia and Rhizophora trials began on 21 July 1995 and ran for 29 d, while that for Laguncularia started on 9 September 1995 and ran for 16 d. The lengths of these laboratory trials were very comparable to the periods over which movements of marked propagules of the respective species were monitored in the field. The condition of the propagules was monitored within 1–2 h of the start of a trial, and every 2–5 d for its duration. For Avicennia and Laguncularia, propagules were dichotomously scored as floating at the surface or sunk to the bottom. Rhizophora propagules were scored as either (1) floating horizontally at the surface, (2) floating at an angle with the stipule end at the surface and the radicle end submerged, but not touching the bottom, or (3) floating vertically in the water column, radicle end downward, or at an acute angle from the water surface with the radicle end of the propagule touching the bottom. These represent successive stages in the natural progression of positions assumed by dispersing Rhizophora propagules following abscission from the parent tree. Freshly released propagules generally float in a horizontal position, but gradually assume a vertical orientation as the radicle end of the propagule selectively imbibes water and increases in mass (Davis 1940, Rabinowitz 1978).

We used two-way ANOVA to examine the effects of paint and salinity treatments on the mean proportion of floating propagules. In the case of Rhizophora, we compared the proportions of propagules floating horizontally, the position most conducive to lateral movement by flow. At any sampling date, very few propagules were found in the intermediate floating position, so analysis of the proportion of propagules floating horizontally was de facto also an analysis of the proportion floating vertically. Variances of proportions were homogeneous among treatments for all tests (range of Cochran’s C: 0.375 – 0.743, P > 0.05), so no transformation was necessary to meet ANOVA assumptions.


We detected no effect of marking with fluorescent paint on patterns of propagule flotation. With one exception, all Avicennia propagules regardless of treatment floated for the entire 29-d trial (an unpainted propagule rotted and sank between days 6 and 11). Over the 16-d Laguncularia trial, the paint treatment also did not have a detectable effect on the proportion of propagules floating (e.g., day 16: F2,6 = 1.087, P = 0.395). Beginning on day 5 and persisting until the end of the trial, a higher proportion of Laguncularia propagules floated in brackish than fresh water (e.g., day 5: F1,6 = 21.235, P = 0.004; day 16: F1,6 = 10.573, P = 0.017), but there was no evidence of an interaction between the effects of paint and salinity (e.g., day 16: F2,6 = 0.000, P = 1.000). At 16 d, an average of 74.4 % of Laguncularia propagules was still floating in the tubs containing brackish water as compared to 37.8 % in those containing fresh water. On none of the sampling dates of the 29 d Rhizophora trial was there a statistically significant effect of paint (e.g., day 29: F2,6 = 1.000, P = 0.422). On the day the Rhizophora trials were initiated, the mean percentage of propagules that floated horizontally was higher in brackish (71.0%) than fresh (41.0%) water (F1,6 = 9.981, P = 0.020), however, this salinity effect did not persist (salinity day 29: F1,6 = 1.000, P = 0.356) and there was no interaction between paint and salinity (e.g., day 29: F2,6 = 1.000, P = 0.422). By the end of the trial, 98.6 % of the Rhizophora propagules were floating vertically.

In summary, while propagules of two of the three species tended to be more buoyant in more saline water, there was no evidence that the fluorescent paint marks influenced the pattern of floating. Therefore, it is unlikely that paint marks had any significant effect on patterns of propagule dispersal in the field.


Davis, J. H. J. 1940. The ecology and geologic role of mangroves in Florida. Publication 517, Carnegie Institute, Washington D.C., USA

Rabinowitz, D. 1978. Dispersal properties of mangrove propagules. Biotropica 10:47–57.

[Back to M077-003]