radiation within the ambient range. Gross 

 photosynthesis per unit leaf area was 

 greater at the top of the tree canopy than 

 at the bottom, although the middle levels 

 had the greatest production. 



Miller (1972) concluded by suggesting 

 that the canopy distribution of red man- 

 grove leaves is nearly optimal for ef- 

 ficient water utilization rather than 

 production. This indicates that the cano- 

 py is adapted to maximizing production 

 under conditions of saturated water sup- 

 ply. 



The mangrove ecosystem model reported 

 by Lugo et al. (1976) provides hypotheses 

 on succession, time to arrive at steady 

 state conditions (see section 3.2), and 

 several aspects of productivity. The 

 model output suggests that the relative 

 amount of tidal amplitude does not affect 

 GPP significantly; instead, GPP appears to 

 be extremely sensitive to inputs of ter- 

 restrial nutrients. It follows that loca- 

 tions with large amounts of nutrient input 

 from terrestrial sources (riverine man- 

 grove communities) have high rates of 

 mangrove production (see section 3.3). 

 All simulation model -generated hypotheses 

 need to be field tested with a particular- 

 ly critical eye, since the simplifying 

 assumptions that are made in constructing 

 the model can lead to overly simplistic 

 answers. 



Mangrove productivity research re- 

 mains in an embryonic stage. Certain 

 preliminary tendencies or hypotheses have 

 been identified, but much work must be 

 done before we can conclude that these 

 hypotheses cannot be falsified. 



2.6 HERBIVORY 



Direct herbivory of mangrove leaves, 

 leaf buds, and propagules is moderately 

 low, but highly variable from one site to 

 the next. Identified grazers of living 

 plant parts (other than wood) include the 

 white-tailed deer, Odocoileus virginianus , 

 the mangrove tree crab, Aratus pi soni i , 

 and insects including beetles, larvae of 



lepidopterans (moths and butterflies), and 

 orthopterans (grasshoppers and crickets). 



Heald (1969) estimated a mean grazing 

 effect on North River red mangrove leaves 

 of 5.1% of the total leaf area; values 

 from leaf to leaf were highly variable 

 ranging from to 18%. Beever et al. 

 (1979) presented a detailed study of 

 grazing by the mangrove tree crab. This 

 arboreal grapsid crab feeds on numerous 

 items including beetles, crickets, cater- 

 pillars, littoral algae, and dead animal 

 matter. In Florida, red mangrove leaves 

 form an important component of the diet. 

 Beever et al. (1979) measured tree crab 

 grazing ranging from 0.4% of the total 

 leaf area for a Florida Keys overwash 

 forest to 7.1% for a fringing forest at 

 Pine Island, Lee County, Florida. The 

 researchers also found that tree crab 

 grazing rates are related to crab density. 

 Low densities (one crab/m ) resulted in 

 low leaf area damage (less than 1% of 

 total leaf area). High densities (four 

 crabs/m ) were accompanied by leaf area 

 damage ranging from 4% to 6% (see section 

 6.2). 



Onuf et al. (1977) investigated in- 

 sect herbivory in fringing and overwash 

 red mangrove forests in the Indian River 

 estuary near Ft. Pierce, Florida. They 

 found six major herbivorous insect 

 species, five lepidopteran larvae and a 

 beetle. Comparisons were made at a high 

 nutrient site (input from a bird rookery) 

 and a low nutrient site. Both red man- 

 grove production and leaf nitrogen were 

 significantly higher at the high nutrient 

 site. This resulted in a four-fold 

 greater loss to herbivores (26% of total 

 leaf area lost to grazing); this increased 

 grazing rate more than offset the in- 

 creased leaf production due to nutrient 

 input. 



Calculations of leaf area damage may 

 underestimate the impact of herbivores on 

 mangroves. For example, the larvae of the 

 olethreutid moth, Ecdytol opha sp., 

 develops within red mangrove leaf buds and 

 causes the loss of entire leaves. All 

 stages of the beetle, Poeci 1 i p s 



23 



