Seed Variation and Seedling Establishment in Manicaria saccifera

Julien Ross, Mathew Krehbiel, and Benjamin Iten

Meet these Students

ABSTRACT
Manicaria saccifera is known to have fruits containing one, two, or three seeds. In an attempt to understand this phenomena, 301 fruits were collected from a random sample of 30 palms at Cano Palma Biological Station and the frequency of each type of fruit was calculated. The weights of these seeds were also recorded to determine if there is a trade-off between the number of seeds per fruit and the weight of each seed. To evaluate the relationship between seed type and viability, 255 freshly sprouted seedlings were examined to determine if they originated from seeds of single, double, or triple-seeded fruits. Fruits containing two seeds were about twice as abundant as single and triple seeded fruits. As the number of seeds per fruit increased from single to triple, seed weight decreased. The relationship between the type of seed and viability was complicated by the observation that more than one seedling almost never germinated from the same fruit.

Manicaria saccifera is an arboreal palm that is patchily distributed on the Atlantic coast of Central America and northern South America. The habitat is lowland inundated areas near the sea. Leaves are very large, two to nine meters long, and strongly plicated. The leaves have been used extensively for thatch roofs in settlements such as Tortuguero village in Costa Rica. These strong ties to human activities place it in the category of plants which should be studied for maximum viability and a future method of sustainable harvesting. One area of variation which could influence viability of progeny in M. saccifera is the number of seeds per fruit. Each fruit contains one, two, or three seeds in a spiny covering.

In order to analyze this variation in seed number per fruit, the following objectives were established: 1) To determine the variation in the number of seeds per fruit and to determine whether this variation existed among and/or within individuals; 2) To determine whether this variation involved a trade-off between seed weight and number of seeds as has been shown for other plants; 3) To determine whether the variation in seed number per fruit is correlated with the success of seedling establishment.

METHODS
This study was carried out on March 7, 1997, at Caño Palma Biological Station (about 83°32’ W; 10°45’ N), located in the Tortuguero area of coastal northeast Costa Rica.

In order to determine variation in seed number among fruits of Manicaria, 301 fruits were collected from 30 separate adult palms, about ten fruits per palm. These fruits were bagged and numbered and later counted to determine the frequency of single, double, or triple seeded fruits.
In order to quantify trade-off between seed size and the number of seeds per fruit, each fruit was weighed and the fruits with two seeds and three seeds were divided by two and three, respectively, resulting in a weight per seed value. To determine the correlation between fruit seed number and viable seedling establishment, 255 seedlings were sampled. The seed, which remains attached to the seedling for a short while after germination, was examined without damaging the seedling, and it was recorded whether the seedling originated from a single, double, or triple-seeded fruit.

The data from the palm fruits and the seedlings were analyzed using a Chi-square test and Fully Factorial ANOVA test to determine any significant correlations using the computer program, Systat®.

RESULTS
Our first result indicates that there was a significant variation in the number of seeds among individual fruits. Fruits containing two seeds were about twice as abundant as the single and triple seeded fruits (Table 1), and the calculated Chi Square value (167.32, P < .001) supports the fact that the frequency of the two seed fruits is significantly greater.

The data from the weighed fruits indicates a significant difference in the weight per seed (F = 18.03, P < .001). Seed weight decreases as the number of seeds per fruit increases (Table 1). The ANOVA test shows that 10.8 % of the variation in seed weight is explained by the number of seeds in the fruit (r² = .108). It turns out that there is also a significant difference in the weight of seeds depending on which tree they are from (F = 26.95, P < .001) and this a much better predictor of the weight of the seed (r² = .752).

The seedling data indicates that seed variation in the fruits correlates directly with seedling establishment (Fig. 1). The frequency of seedlings germinated from single, double, and triple seeds matched the frequency found in the collected fruits, with the double seed dominating.

DISCUSSION
There is a significant variation in the number of seeds per fruit in M. saccifera at Caño Palma. Fruits containing two seeds are twice as abundant and this variation correlates directly with the number of established seedlings. This striking difference from an even distribution suggests likely evolutionary selection where the two seeded fruits are the optimum number for yielding offspring. However, with more than one seed existing per fruit, a trade-off exists in the weight of each seed, thus, reducing the weight of the seeds in the two and three seeded fruits. Since the weight of a plant seed has been correlated with its success in germination and succeeding growth (Howe & Richter 1982, Kang et al. 1982), there appears to exist an optimum balance between producing enough seeds to satiate predators and producing large enough seeds that they are rigorous enough to survive.

It is also important to note that only one seedling germinated from each fruit, no matter how many seeds it had. Additional seeds were found attached to the seed that produced the seedling, but these showed no evidence of having sprouted. If we assume that only one seed per fruit is successful for whatever reason (seed predators, seed abortion, competition between seedlings) then the fruits with one seed actually have a higher proportion of seeds produced that survive. This follows the argument that the heavier one seed fruits are more viable than the lighter fruits, however, this study shows the double seed fruits have achieved a balance of more optimum viability.

It is not known whether the individual variation in seed number, seed weight, and seedling establishment is due to genetic or environmental factors. To find the answer requires further study where environmental conditions could be measured and observed to determine any relationship with seed and seedling variation. Another effective method to test seedling establishment would be to plant seeds under controlled conditions., with some seeds planted as a whole and others separated first and then planted. Observing which seeds are viable could further explain the seed variation.

It is undeniable that there is significant variation in the production of one, two, and three-seeded fruits in Manicaria saccifera and that this variation is such that two seeded fruits are much more common and viable. What remains to be shown is whether this variation is heritable and in fact optimal, or whether it is based on environmental factors. Understanding and maximizing the viability of the seeds is crucial if the Manicaria saccifera is to be considered a candidate for sustainable development.

ACKNOWLEDGEMENTS
I would like to sincerely thank Mauricio Quesada for providing ideas, advice, and his invaluable knowledge and Alfredo Cascante for helping us with data collection, data analysis, and his overall biological understanding of the rainforest.

LITERATURE CITED
HENDERSON, A.., G. GLORIANA, AND R. BERNAL. 1995. Palms of the Americas. Princeton University Press, Princeton, NJ.

HOWE H.F., AND W.M. RICHTER. 1982. Effects of seed size on seedling size in Virola surinamensis; a within and between tree analysis. Oceologia 53: 347-351.

KANG, H. 1992. Variation in seed and seedling traits in Pithecellobium pedicellare, a tropical rainforest tree. Oceologia 91: 239-244.

Meet these Students
Return to Projects Spring 1997