May 2001

ABSTRACT

Since its first appearance in Florida in 1989, Metamasius callizona (Chevrolat) has spread into 15 counties, leaving a trail of dead or dying bromeliads in its wake. Because commercially-grown bromeliads are chemically treated and thus not susceptible to weevil infestation, it only threatens Florida's bromeliads in the wild. Two of Florida's 16 native bromeliads have been classified as endangered directly as a result of weevil decimation: Tillandsia fasciculata (Swartz) and T. utriculata (L.) (Frank and Larson, 2000). Therefore, the focus of this study involved wild Tillandsias found in Deer Prairie Slough of Myakka River State Park. Over the course of 15 weeks, the author visited this site once a week to record the progressive damage wrought by M. callizona so as to estimate the rate of decline. This paper provides an overview of the results of that study, observations, and possible interpretations of the available data. It is important to note that this is an ongoing study, and therefore no final or definite conclusions can be drawn from this data alone. However, the purpose of this study was to contribute to the current evaluation of the threat to Florida's bromeliads by M. callizona, and particularly to determine how much time remains until these species may be decimated by what has come to be known as the "evil weevil".

INTRODUCTION

Most of what is known about Metamasius callizona is due to the efforts of Dr. Howard Frank of the University of Florida. Dr. Frank has been tracking the weevil since 1991, two years after its arrival in Ft. Lauderdale over a decade ago. Since then, he has traced its movement throughout southern Florida. Therefore, much of the general information in this paper is based upon his research. Most of it is readily available on the internet, and therefore I will only briefly provide an overview of the weevil and the bromeliads which it attacks. The discussion will then focus on the data collected from Myakka River State Park over the course of 15 weeks. This will entail observations regarding size, location, density, and other characteristics of infested and uninfested plants, as well as the methodology used to conduct the study. Finally, I will provide an estimation of the rate of decimation by the weevil at this site. I will conclude by discussing the possible solutions to this problem, as well as the ecological significance of Florida's wild bromeliads.

THE BROMELIADS

The term "bromeliad" is used to refer to members of the family Bromeliaceae, which contains three subfamilies. All of Florida's 16 native bromeliads are members of the subfamily Tillandsioideae. Bromeliads are also known as "air plants" because many of them are epiphytes. An epiphyte is any plant that grows on another plant, but does not parasitize its host. Some bromeliads are facultative epiphytes, meaning they are also able to grow terrestrially; whereas others are obligate epiphytes and are unable to grow in the soil. The latter have a very reduced root system or none at all, and if present it is used mainly for grasping onto the branches of its host. Florida's bromeliads fall into both of these categories. One of the most common bromeliads in the state is T. usneoides (L.), or Spanish moss, and it is an obligate epiphyte. However, the two bromeliads in this study, T. utriculata and T. fasciculata are facultative epiphytes and can therefore be found both on the ground and in trees (Benzing 1980).

Figure 1. Tillandsia sp.

Because bromeliads have adapted to the extreme conditions of life in the forest canopy, they are highly specialized and capable of living under a great deal of environmental stress. Tillandsias in particular are extremely drought resistant, as they must be in order to survive Florida's dry winters and hot summers. Some examples of adaptive features are greatly reduced vegetative growth and the use of specialized cells, called trichomes, to catch and absorb water and nutrients on their leaves. Bromeliads also feature another structural adaptation that enables them to store water. The leaves of the plants grow in a tight rosette pattern from the base, and in the center of this spiral of leaves a cavity is formed where water may be trapped. This cavity is known as a phytotelma, or tank, and the size and capacity of phytotelmata vary among the species (Benzing 1980). The central cavity typical of tank bromeliads is minimal in T. utriculata and T. fasciculata. Nonetheless, they are considered true tank bromeliads, for they collect enough water in the phytotelmata and in the leaf axils to maintain a thriving aquatic ecosystem (Fish 1976). Some of these organisms have not yet been described by scientists. Following is a list of aquatic invertebrates that have been found living in Florida's bromeliads:

It is probable that some of these species live only in bromeliads, or at least depend on them for the aquatic phase of their lifecycle. There are also various species of ants and spiders which can be found living in wild Tillandsias as well. Although they do not depend on the plant itself for food, they do prey on the other insects that are drawn to it (Benzing 1980).

As for M. callizona, it appears to be dependent upon bromeliads for at least the pupal development stage of its life cycle. Of Florida's sixteen native species, five of them have been affected by M. callizona. These are: Tillandsia utriculata (endangered), T. fasciculata (endangered), T. balbisiana Schultes (threatened), T. flexuosa Swartz (endangered), and T. paucifolia Baker. Both T. utriculata and T. fasciculata have been classified as endangered by the Florida Endangered Plant Advisory Council as a direct result of weevil damage (Frank 1999). It appears that T. fasciculata is somewhat more resistant to weevil decimation than T. utriculata, due to the fact that it can produce pups, or shoots, when the mother plant dies. T. utriculata, on the other hand, reproduces exclusively by seed (Frank 1999, Benzing 1980). Nonetheless, both species have proven susceptible to widespread decimation by the weevil in areas where the bromeliads were once abundant (Creel 1999).

The plant is killed when the larvae shred the central core of the plant to build a cocoon for pupation. The adult emerges from the cocoon, and though it does feed on the leaves, it does not kill the plant. Because the larvae need a substantial amount of plant material to form a cocoon, it will not attack smaller species of bromeliads such as T. usneoides and T. bartramii (Elliott). However, several of Florida's rarest and endangered native bromeliads could be affected by M. callizona. These include T. pruinosa Swartz (endangered), T. variabilis Schlechtendal (threatened), Catopsis floribunda Brongniart (endangered), and C. nutans Swartz (endangered) (Frank 1999). These species are found in the unique and fragile ecosystems of the Fakahatchee Strand, Big Cypress National Preserve, and the Everglades. The weevil has not yet been found in any of these sites, though it is moving closer at an alarming rate. A survey performed by Dr. Frank in March and April of 1999 found the weevil within four miles of Big Cypress National Preserve (Frank and Thomas 2001).

Figure 2. Larval mining and cocoon.

THE WEEVIL

Metamasius callizona (Chevrolat) is a member of the weevil subfamily Rhynchophorinae, and it is one of three species of the genus Metamasius found in Florida, only one of which is native. M. callizona made its first appearance in the state in 1989, at a bromeliad nursery in Ft. Lauderdale, where it is thought to have arrived in a shipment of plants imported from Veracruz, Mexico. Though the nursery was treated with pesticides, within two months the weevil was present in Broward County and Palm Beach County. As of January 2001, M. callizona has now been found in a total of sixteen Florida counties. It has spread naturally as well as by the movement of infested plants (Frank and Larson, 2000).

Following is an excellent description of how the weevil affects its host, courtesy of the Institute of Food and Agricultural Sciences of the University of Florida:

All life stages of the weevil may be present in the same plant. Adult females, which primarily feed on leaves but have been seen to feed on flowers, cut slits in host plant leaves close to their feeding sites, into which they lay eggs individually. Newly emerged larvae begin to consume leaf tissue as they move down to the base of the stem. They tunnel into the growing stem tissue, producing large holes in the base that may cause the plant to dislodge from its support structure on the tree. Larval damage is generally confined to the base of the plant but can also reach up the flower stalk. Several weevils may successfully develop on the same host plant, provided there is sufficient plant tissue. However, larger larvae will attack smaller ones upon encountering them. The weevil usually pupates in the center of the plant's stem, within a cocoon it constructs from shredded plant material. When reared in planted pineapple tops, the weevil sometimes pupates in the soil near the base of the plant. Damage is often accompanied by the production of a light brown, gelatinous material, which may be the plant's defensive reaction. This gel can be seen covering entrance holes to the weevil mines. Other symptoms of weevil damage include adult feeding marks on leaves, browning of leaves, and decomposition of the base of the central leaves, which can easily be pulled out when larval mining is significant (Frank and Larson 2000).

Because the base of the plant is destroyed, evidence of the weevil is not difficult to find. An infested area is generally littered with bromeliad leaves and even whole plants. On the other hand, I never once ecountered a live adult weevil throughout my investigation of this insect. One difficulty, of course, is that its host plants live high in the forest canopy. In Mexico, however, where both the Tillandsias and M. callizona reside, the weevils are even more scarce, and their damage to the bromeliads there is relatively insignificant. Therefore, it is speculated that there are one or more insects that prey on the weevil in its natural habitat. Dr. Frank and his colleagues are currently searching for such a predator as a possible means of biological control. A potential candidate is now being studied at the Escuela Agrícola Panamericana in Honduras. It is a parasitoid tachinid fly of the genus Admontia that has been known to parasitize other related Metamasius species. Eventually, researchers hope to use such a biological control agent to eliminate or control the weevil in the wild where chemical spraying is not feasible (Frank and Larson 2000).

CASE STUDY: MYAKKA RIVER STATE PARK

SITE SELECTION

The site was selected by Park Biologist Belinda Perry. It is located in a section of the park known as Deer Prairie Slough, a swampy oasis surrounded by a wide expanse of dry prairie, and it is home to a large number of bromeliads. The presence of M. callizona was first discovered at Myakka River State Park in September of 2000 (Frank and Thomas 2001). I was first introduced to this project in October of 2000 by Ms. Perry. We began by selecting 20 specific points along the old railroad grade that transects the swamp. Each of the 20 points was selected for certain characteristics, which I will discuss further. We also selected a single Red Maple tree, adjacent to the railroad bridge, which hosted over 100 bromeliads, some healthy and others dead or dying as a result of weevil damage.

Rather than monitor every single bromeliad at the site, which would have been extremely tedious, we elected to limit this study to those that represent the variability of the bromeliad habitat. For instance, site #10 was chosen because there were three bromeliads found on a tree surrounded by water, in order to observe whether or not these bromeliads will be less vulnerable to weevil attack due to their isolated location. Terrestrial bromeliads, fallen bromeliads, and isolated vs. crowded bromeliads were included in the study as well. We also chose sites that included a variety of host tree species.

RATING SCALE

It was necessary to establish a means of scoring the bromeliads based on their degree of health or demise. The following scale was applied to all bromeliads in the study and recorded in the data charts on a weekly basis:

It is important to note that not all visible damage to a bromeliad is necessarily caused by M. callizona. This can only be positively determined in cases where the bromeliad has fallen from its host and can be inspected. Larval mining is unmistakable. However, other types of damage, such as yellow and drooping foliage and nibbled leaf tips, may also be caused by frostbite, dehydration, or other insects. Nevertheless, the nature of bromeliad habitats is such that observations must often be made from a distance, and the criteria for the bromeliad scores were applied regardless of the cause of damage. Therefore, it may be concluded that not all 2's or even 2.5's were necessarily affected by the weevil, particularly those which retained those ratings without exhibiting further demise¹. However, it is generally safe to conclude that those bromeliads which did demonstrate a steady decline in health were attacked by the weevil². Conversely, there were a number of bromeliads which had shown little or no evidence of damage, but suddenly dropped to the ground, riddled with weevil scars³. There were also a handful of bromeliads that had flowered and thus died of natural causes. Such bromeliads were duly noted and are not included in the calculations which apply to the rate of weevil damage⁴.

It is also important to note that the process of scoring is relatively subjective, and can be influenced by such factors as time of day and weather. Towards the end of the study, I gained access to a good pair of binoculars, which greatly improved my ability to observe certain specimens, and thus increased my confidence in the data. On certain occasions when I was accompanied by Ms. Perry or another volunteer, I found a second opinion to be quite helpful. Also, as my familiarity with the tell-tale signs of weevil damage increased, so did my ability to score bromeliads appropriately.

DATA COLLECTION

As stated previously, the data was recorded on a weekly basis, over a period of 14-15 weeks. The transect data was collected for a full 15 weeks, whereas part of the Red Maple study was conducted for only 14 weeks, due to the time it took to sketch all of the 135 bromeliads in the tree. It was necessary to sketch each bromeliad in the Red Maple, as well as a couple of other points along the transect which contain multiple bromeliads, in order to make consistent observations of individual specimens.

Other applicable information regarding site characteristics were also recorded. These include the species of the host tree (except for terrestrial bromeliads, which were duly noted), the number of bromeliads at that particular site, distance from the ground and the road, and whether the plant had an inflorescence. Bromeliads that were particularly difficult to observe due to an obstructed view were also indicated on the chart with an asterisk. Each transect point was numbered 1-20, and every plant within each point was labeled A-Z, corresponding to the number of plants. Similarly, the Red Maple was labeled with a number for each branch, and a letter for each bromeliad on the branch.

OBSERVATIONS

I observed a couple of significant patterns throughout the course of this study. First, however, I must again state that this study is ongoing and therefore none of the following interpretations are conclusive. The monitoring of the bromeliads at this site will be continued by another volunteer at the park, who will record data in the same manner, until the bromeliads are either wiped out completely or a biological control agent is introduced. Meanwhile, certain patterns in the progression of weevil damage through the site are worth noting.

First of all, terrestrial and low-lying bromeliads seem to have an advantage. Virtually all of the bromeliads found on the ground and on fallen branches maintained a rating of 1, indicating optimum health. Unfortunately, this probably only implies a temporary immunity, for other places with many terrestrial bromeliads, such as the Hidden Forest in Broward County, have been destroyed indiscriminately by the weevil (Creel 1999). It is possible that these bromeliads are less favorable to the weevil, and therefore they may at least be spared as long as others are available in the trees.

Another possible advantage is isolation. Solitary bromeliads appear to be more successful. One seemingly exceptional specimen is site number 18, where a single bromeliad, only eight feet high on the trunk of a Laurel Oak, was killed by M. callizona. This was not, however, a completely isolated bromeliad, for although it was the only plant on that particular tree, there were others in the vicinity that were not monitored but that were damaged. Another significant example is site number 10, where there are three bromeliads on a Red Maple that is surrounded by water. All of these plants maintained optimum ratings throughout the study. There are two other sites, numbers 1 and 15, which are located on the opposite side of the narrow waterway that runs parallel to the road. All of the bromeliads at these two sites also maintained a rating of 1. It is believed that the weevil can only fly for short distances, and so must fly from branch to branch and tree to tree in order to travel any distance (Frank and Thomas 1992). If that is the case, this may be another advantage to bromeliads in isolated locations. Again, this simply may deter the weevil only as long as there are more accessible bromeliads available. As the bromeliad population steadily declines, time will tell how much longer these specimens will remain immune to weevil attack.

A parallel trend which can be easily observed in the Red Maple is the vulnerability of densely crowded bromeliads. There are 135 specimens in this tree that are included in the study, as well as many more plants that were already dead when this study began. Therefore, many of the branches (particularly Branch 5 and 6) are extremely crowded. As a result, once the weevil infests one plant, its neighbors are soon to follow. I believe this is the reason there were more losses due to weevil damage in the Red Maple than in the transect along the railroad grade, although the same pattern does occur in certain transect sites as well. For example, site number 3 contained sixteen bromeliads in four trees within close proximity, all of which were healthy at the beginning of the study⁵. No weevil damage was recorded until Week 11, and since then four bromeliads, which had been fairly healthy, suddenly appeared dead on the ground, at the rate of one per week over the next four weeks. It is reasonable to expect that more will soon follow suit now that the weevil is present.

One more observation worth mentioning is the size of the affected bromeliads. According to the sources I consulted, M. callizona generally favors large, mature bromeliads (Frank 1999). A close inspection of this phenomenon in the Red Maple somewhat contradicts previous observations. Many of the bromeliads which died as a result of weevil damage were quite small. An interesting example can be found in Branch 2, where Specimen A is very large and healthy, whereas two much smaller bromeliads nearby, Specimens B and E, were killed. Similar examples can be found throughout this tree. With the exception of Specimen A, however, most of the larger bromeliads that were spared were either located higher up in the tree, and/or were relatively isolated. Since weevil cocoons require substantial plant material, it seems advantageous for the larvae to favor the larger bromeliads. Studies have shown that pupae found in smaller plants develop into smaller adults (Frank and Thomas 2001). Perhaps in this location where the host plants are densely packed, small adult size is less of a disadvantage, since they do not have to fly any distance to find more host plants to feed on. As the crowded groups of bromeliads are eliminated, this pattern may shift as adult weevils are forced to travel further to find live plants on which to lay their eggs. However, this hypothesis assumes that larger weevils are stronger fliers, and as of yet there is no evidence to support or negate this.

RESULTS

A major concern of this study, and to all who seek a solution to the weevil invasion, is how much time remains before all of Florida's T. utriculatas and T. fasciculatas are wiped out. Therefore, I calculated the rate of decimation in this sample population of bromeliads for the duration of this study. In a matter of 15 weeks, 18% of the specimens, or 14 out of 78, on the transect were killed as a direct result of weevil damage. Over a period of 14 weeks, 39% of the specimens, or 53 out of 135, in the Red Maple were also killed by M. callizona. Therefore, at the current rates of decline in these two samples, it can be estimated that all of the bromeliads in the Red Maple will be eliminated in approximately 36 weeks, while those on the transect will last approximately 84 weeks. This is a conservative estimate, since this study was conducted during the coldest months of the year. It is probable that as the seasons change and temperatures rise, weevil activity will increase significantly (Frank and Thomas 2001).

Time is a formidable enemy in the search for a biological control agent for M. callizona. Dr. Frank and his colleagues are doing their best with the available resources, for funding is a significant limitation in this project. Currently, all funding is provided through the Florida Council of Bromeliad Societies (FCBS). On August 28, 2000 the FCBS signed a contract with the Division of Plant Industry (DPI), Florida Department of Agriculture and Consumer Services. DPI, through the state legislature, will provide funds in 2000-2002 for research and implementation of a plan to control the pest weevil M. callizona. In addition to the research of biological control, the FCBS is also sponsoring a plan to collect endangered bromeliad seeds. In the event that all wild bromeliads are wiped out, it is hoped that the weevil will also disappear for lack of a host plant. The plan is to then grow out seeds under controlled conditions and eventually transplant them to natural settings.

Although my participation in this study has terminated, the volunteers and park biologists continue to closely monitor the weevil's progression through the park. Myakka River State Park is considered the prime site for release of a biological control agent. Many unanswered questions remain, such as whether the introduction of yet another exotic insect could have further ecological ramifications. Still, there is no other known means of eradicating the weevil in the wild.

WHY SHOULD WE CARE?

Part of the problem in finding a solution to the weevil crisis in Florida is that M. callizona poses little or no economic threat, at least not directly. Although it may not threaten commercially-grown plants or major food crops, it definitely could impact Florida's thriving tourism. Not everyone comes to Florida to visit the theme parks. The State Parks rely heavily on visitors who come to appreciate Florida's unique subtropical ecosystems, which are greatly enhanced by the lush epiphytic growth throughout the hammocks and riverine areas. Such rare and unique sites as the Fakahatchee Strand, Big Cypress Natural Preserve, and Everglades National Park are especially vulnerable to weevil devastation, and the ecological impact could be severe.

In ecological terms, bromeliads are prized biodiversity enhancers. The aquatic microcosms which thrive year-round in their tanks can be likened to aerial ponds. They also increase the structural complexity of an ecosystem, another means of contributing to biodiversity (Richardson 1999). Although there may be little immediate human concern for the vast array of insects and organisms that rely on these plants for food, shelter, and reproduction, their importance to the greater ecosystem cannot be underestimated. The very nature of ecosystems entails a complex web of relationships and interdependence, and when any single component is disrupted or removed, the consequences are exponential. If Florida's bromeliads do become extinct, we can only begin to speculate on what the long-term effects would be.

One does not need a deep understanding of biology or ecology, however, to appreciate bromeliads in the wild. Their aesthetic value is immeasurable. Florida's natural wilderness areas would lose much of their tropical flavor without the lush epiphytes that grace the branches of its trees. Public awareness is an important factor in eliminating this pest. The Sarasota Herald-Tribune has run a couple of articles describing the plight of Florida's bromeliads. Bromeliad and native plant enthusiasts alike have been closely following this issue. Many volunteers in the State Parks are working to educate the public through interpretive tours. In addition, Dr. Frank continues to persevere in his efforts to stop or at least control the spread of this exotic weevil. Meanwhile, we must not take these unique plants for granted.

PERSONAL RELEVANCE

As a student of Environmental Studies at New College of the University of South Florida, this study was very influential to me. For one thing, I had the opportunity to contribute to the current monitoring of two endangered species and to the understanding of the cause of their decline. As an environmentalist who is also interested in botany, it was also an opportunity to further my understanding of members of the unique family Bromeliaceae and their ecological function. In addition, I was also able to forge a relationship with the Park Biologist at Myakka River State Park. Belinda Perry is very active in the community and dedicated to her work at the park, and I am certain that she will prove to be a useful reference in the future.

I believe the relevance of this study to the sciences, and in particular the environmental sciences, is quite evident. According to Dr. Frank, this is the only study that has been conducted thus far addressing the rate of decline in a population of Tillandsia spp. infested by M. callizona. A copy of this paper was requested by and submitted to both Dr. Frank and Myakka River State Park. I think it is important for the university to demonstrate a commitment to the ecological preservation of Florida's unique ecosystems and their rare epiphytes, particularly in cases such as these where the species do not necessarily represent a direct economic value.

BIBLIOGRAPHY

Benzing, David H. 1980. The Biology of the Bromeliads. Mad River Press, Eureka, CA

Creel, Olan Ray 1999/2000. "The Evil Weevil: What Will Florida Lose?" The Palmetto 19(4): 10-11, 14-16

Fish, Durland 1976. "Structure and composition of the aquatic invertebrate community inhabiting epiphytic bromeliads in South Florida and the discovery of an insectivorous bromeliad". Dissertation for University of Florida

Frank, J.H. 1999/2000. "Florida's native bromeliads imperiled by exotic evil weevil". The Palmetto 19(4): 6-9, 12.

Frank, J. H., Larson, Barbra. Oct. 2000. "Featured Creatures: Metamasius callizona" Published by Institute for Food and Agricultural Sciences, University of Florida on the WWW at http://creatures.ifas.ufl.edu/orn/m_callizona.htm

Frank, J.H., Lounibos, L.P. 1987. "Phytotelmata: swamps or islands?" Florida Entomologist 70: 14-20.

Frank, J. H., Thomas, M. C. 1992. "Metamasius callizona in Four Counties in South Florida". Journal of the Bromeliad Societies 42: 128

Frank, J.H., Thomas, M.C. 1994. "Metamasius callizona (Chevrolat) (Coleoptera: Curculionidae), an immigrant pest, destroys bromeliads in Florida". Canadian Entomologist 126: 673-682.

Frank, J. H., Thomas, M. C. 2001. "Weevils that eat bromeliads". Published on WWW at http://bromeliadbiota.ifas.ufl.edu/wvbrom.htm

Richardson, Barbara A., 1999. "The bromeliad microcosm and the assessment of faunal diversity in a Neotropical forest". Biotropica 31(2): 321-336

Zavortink, T.J. and O'Meara, G.F. 1999. "Culex (Micraedes) biscaynensis n. sp. from Florida (Diptera: Culicidae)". Journal of the American Mosquito Control Association 15: 263-270.

Appendix A
Appendix B

END NOTES

¹ See Appendix A, specimens 6A, 11nD, and 16E
² See Appendix B, specimens 3G, 4H and 6K
³ See Appendix A, specimens 3I, 18 and 20C
⁴ See Appendix A, specimens 3B, 8A, 11nA, and 12D
⁵ with the exception of Specimens 3B and 3J, which flowered and died naturally