Genetic variation of introduced Hawaiian and native Costa Rican populations of an invasive tropical shrub, Clidemia hirta (Melastomataceae)

Clidemia hirta is one of the most common woody invasive plants in mesic to wet forests in Hawaii, where it was introduced around 1940. The species is relatively uncommon by comparison in its native range of Central and South America and some Caribbean Islands. We examined genetic variation in allozymes of 20 C. hirta populations on four Hawaiian Islands to determine the introduction history. For comparison, we measured genetic variation in 20 native populations across Costa Rica. Mean levels of genetic variation in Hawaiian and Costa Rican populations were low compared to other woody or introduced plants (11.5-12.5% polymorphic loci, 2.05-2.50 alleles per polymorphic locus, and 0.045-0.063 expected heterozygosity). Most genetic diversity was held within rather than among populations in both areas (G(ST) = 0.120 and 0.271 in Hawaii and Costa Rica, respectively). Hawaiian populations had a high degree of genetic similarity, and no genetic differentiation was found among the four Hawaiian Islands sampled. These patterns of genetic variation in Hawaii suggest that no intraspecific hybridization of genotypes from different parts of the native range has occurred and that introductions to the different islands came from the same or similar source populations. The low levels of genetic diversity in parts of both the native and introduced ranges suggest that genetic variation is unrelated to invasiveness in C. hirta.     

Getting here from there: testing the genetic paradigm underpinning introduction histories and invasion success

Aim To explore the potential of genetic processes and mating systems to influence successful plant invasions, we compared genetic diversity of the highly invasive tropical treelet, Miconia calvescens, in nine invasive populations and three native range populations. Specifically, we tested how genetic diversity is partitioned in native and invaded regions, which have different invasion histories (multiple vs. single introductions). Lastly, we infer how levels of inbreeding in different regions impact invasion success. Location Invaded ranges in the Pacific (Hawaii, Tahiti, New Caledonia) and Australia and native range in Costa Rica. Methods Genetic diversity was inferred by analysing variation at nine microsatellite loci in 273 individuals from 13 populations of M. calvescens. Genetic structure was assessed using amova , isolation by distance (IBD) within regions, a Bayesian clustering approach, and principal coordinates analysis. Results Microsatellite analysis revealed that invaded regions exhibit low levels of allelic richness and genetic diversity with few private alleles. To the contrary, in the native range, we observed high levels of allelic richness, high heterozygosity and 78% of all private alleles. Surprisingly, despite evident genetic bottlenecks in all invasive regions, similarly high levels of inbreeding were detected in both invasive and native ranges (F_IS: 0.345 and 0.399, respectively). Bayesian clustering analysis showed a lack of geographical structure in the Pacific and evidence of differing invasion histories between the Pacific and Australia. While Pacific populations are derived from a single introduction to the region, multiple introductions have taken place in Australia from different source regions. Main conclusions Multiple introductions have not resulted in increased genetic diversity for M. calvescens invasions. Moreover, similar inbreeding levels between native and invaded ranges suggests that there is no correlation between levels of inbreeding and levels of standing genetic diversity for M. calvescens. Overall, our results show that neither inbreeding nor low genetic diversity is an impediment to invasion success.     

The classification, evolution and biology of the Costa Rican species of Cryptophion (Hymenoptera: Ichneumonidae)

The six known Costa Rican species of the campoplegine ichneumonid genus Cryptophion Viereck are described and keyed. The distribution of species throughout Costa Rica is detailed based on data gathered by an intensive Malaise trap survey of the ichneumonids of the country. Five new species are recognized: C. espinozai. C. guillermoi, C. manueli. C. moragai and C. tickelli , and a sixth, C. inaequalipes (Cresson) redescribed. The monophyly of the genus is demonstrated and the phylogeny of the Costa Rican species reconstructed. Host relationships have been established for all species in Costa Rica; they develop as koinobiont endoparasitoids of first to third instar larvae of Sphingidae or Saturniidae. Analysis of the host relationships from a phylogenetic perspective suggested that the genus first evolved using macroglossine sphingid larvae feeding on rubiaceous understorey plants as hosts, and subsequently diversified to utilize sphingine sphingids and ceratocampine saturniids feeding on a variety of food-plants. Most species appear to be monophagous and oligophagy is apparently a derived feature of one sister-species pair, C. espinozai and C. manueli . In Santa Rosa National Park, Costa Rica, only a small proportion of the species of Saturniidae and Sphingidae present are used as hosts by Cryptophion species. No one host species is parasitized by more than one Cryptophion species. No species of Cryptophion is known to parasitize more than one host species feeding on any one plant species. The monophagous species C. inaequalipes is only known to develop in its sphingid host when that host is feeding on one of its two alternative food-plants.     

Host specificity testing and suitability of the parasitoidMicroctonus hyperodae (Hym.: Braconidae,Euphorinae) as a biological control agent ofListronotus bonariensis (Col.: Curculionidae) in New Zealand

The behaviour of the parasitoidMicroctonus hyperodae Loan was studied under quarantine conditions to determine its likely host range in New Zealand. The species was imported from South America as a potential biological control agent of Argentine stem weevil,Listronotus bonariensis (Kuschel). The study involved systematic evaluation of the parasitoid's behaviour when exposed to 24 non-host weevil species; all but three of these were native to New Zealand. Of those tested, four were found to sustain someM. hyperodae development. However, further examination showed that in all but one species,Irenimus aequalis (Broun), parasitoid development was impeded, with up to 50% of the larvae becoming encapsulated. Overall, those weevil species that were attacked produced only 19% of the parasitoids derived fromL. bonariensis controls. As an adjunct to this quarantine study, a review of the habitats of the native weevil and target pest populations indicated that refugia would probably exist for native alpine species. I. aequalis was not considered to be threatened byM. hyperodae as this weevil has benefited from the advent of European agricultural systems to the extent that it is now recognised as a minor pest. In view of its relatively oligophagous behaviour, the parasitoid was recommended as suitable for release.      

Reproductive trade-offs in a specialized plant/pollinator system involving Asplundia uncinata Harling (Cyclanthaceae) and a derelomine flower weevil (Coleoptera: Curculionidae)

The reproductive interactions of a specialized plant/pollinator system involving Asplundia uncinata Harling (Cyclanthaceae) and a derelomine flower weevil (Coleoptera: Curculionidae: Derelomini) were studied at La Selva, Costa Rica. The inflorescences of A. uncinata exhibit a suite of cantharophilous characters including a precisely synchronized anthesis, protogyny, thermogenesis, olfactory attraction of visitors via modified staminodes, and narrow interfloral entrances. The weevil pollinators use the inflorescences for feeding, mating, and oviposition. The larvae are detritivorous and develop either in the detaching staminate flowers or, at a more favorable rate, in the rotting infructescences. The rate of infructescence abortion was high and caused by low levels of pollination. Manual pollination treatments yielded significantly higher seed counts than obtained under natural conditions, and furthermore demonstrated the inflorescences' ability to reproduce via geitonogamy. In the longer term, the reproductive benefits of maintaining low levels of pollination may shift away from the weevils and towards the plants via an increase in the size of the pollinator population.     

Biomass allocation,growth, and photosynthesis of genotypes from native and introduced ranges of the tropical shrub<Emphasis Type="Italic"> Clidemia hirta</Emphasis>

We tested the hypothesis that the tropical shrub Clidemia hirta appears more shade tolerant and is more abundant in its introduced than native range because of genetic differences in resource acquisition, allocation, and phenotypic plasticity between native and introduced genotypes. We examined growth, biomass allocation, and photosynthetic parameters of C. hirta grown in a greenhouse from seed collected from four populations in part of its native range (Costa Rica) and four populations in part of its introduced range (Hawaiian Islands). Six-month-old seedlings were placed in high (10.3–13.9 mol m^–2 day^–1) or low (1.4–4.5 mol m^–2 day^–1) light treatments and grown for an additional 6 months. Our study provided little evidence that Hawaiian genotypes of C. hirta differed genetically from Costa Rican genotypes in ways that would contribute to differences in habitat distribution or abundance. Some of the genetic differences that were apparent, such as greater allocation to stems and leaf area relative to whole plant biomass in Costa Rican genotypes and greater allocation to roots in Hawaiian genotypes, were contrary to predictions that genotypes from the introduced range would allocate more biomass to growth and less to storage than those from the native range. Hawaiian and Costa Rican genotypes displayed no significant differences in relative growth rates, maximal photosynthetic rates, or specific leaf areas in either light treatment. In the high light environment, however, Hawaiian genotypes allocated more biomass to reproductive parts than Costa Rican genotypes. Phenotypic plasticity for only 1 of 12 morphological and photosynthetic variables was greater for Hawaiian than Costa Rican genotypes. We conclude that genetic shifts in resource use, resource allocation, or plasticity do not contribute to differences in habitat distribution and abundance between the native and introduced ranges of C. hirta. Electronic Supplementary Material Supplementary material is available in the online version of this article at . Clidemia hirta individuals grown in a common garden in high light for 10 months from Hawaiian (left) or Costa Rica (right) seed sources     

Insectary production and synopsis of Fopius caudatus (Hymenoptera: Braconidae), parasitoid of tephritid fruit flies indigenous to Africa

Fopius caudatus (Szépligeti) is an endophagous koinobiont egg-larval parasitoid native to Africa. It has recently been noted as a candidate for augmentative biological control of several Dacinae fruit fly pests (Diptera: Tephritidae), due to its ability to parasitize the egg stage. Previous attempts to establish this parasitoid in Hawaii, Guatemala, and Costa Rica were unsuccessful due to inability to maintain parasitoid colonies under laboratory conditions. A cohort of F. caudatus collected from Kenyan fruit flies infesting Coffea arabica was successfully colonized in Hawaii at 28 °C and 60–80% RH, resulting in the development of a laboratory-adapted colony amenable for mass production. The parasitoid was successfully developed from eggs of Ceratitis capitata and Bactrocera latifrons as a factitious host. The wasps were propagated for 15 weeks until the rearing stabilized, at which point >10,500 adults were produced with an overall sex ratio of 0.52 females and a mean host parasitism rate of 17.3%. It could parasitize Medfly eggs in fruits other than coffee, including papaya, mango, pear, squash, and sweet pepper. Female F. caudatus oviposited mainly in 24–48 h old Medfly eggs, although occasionally a few individuals eclosed when first instar fly larvae were exposed. Mean developmental time from egg to adult was 19.8 d for males and 21.5 d for females. Mean longevity was 5.2 d for males and 14.2 d for host-deprived females. This study enabled us to maintain a colony of F. caudatus for research and redistribution to other countries for biocontrol programs against Medfly.     

Hydrilla stems and tubers as hosts for three Bagous species: two introduced biological control agents (Bagous hydrillae and B. affinis) and one native species (B. restrictus)

Field observations suggested that the introduced Hydrilla verticillata (L.f.) Royle biological control agent, a stem weevil, Bagous hydrillae O'Brien, would feed on hydrilla tubers and stems, and a native species, Bagous restrictus LeConte, would feed on hydrilla stems. In choice tests, B. hydrillae readily oviposited in hydrilla tubers. Larval development of B. hydrillae in hydrilla tubers was similar to that in stems; greater adult biomass was attained and less time was needed to complete development when the larvae were fed tubers. Larvae of the hydrilla tuber weevil, B. affinis Hustache, did not complete development in hydrilla stems. Larvae of B. affinis completed development more rapidly when fed new compared with old hydrilla tubers. The native B. restrictus successfully completed development in hydrilla stems, although the larvae required slightly more time compared with the biocontrol agent, B. hydrillae. These findings indicated that feeding on tubers by B. hydrillae may benefit the species particularly when hydrilla stems are seasonably absent or unsuitable especially in more northern climates. In terms of hydrilla control, damage to tubers by this species constitutes a reduction in future infestations of hydrilla propagated by tubers. Finally, hydrilla is suitable to the native weevil, B. restrictus, because larvae completed development in hydrilla stems.     

Effect of temperature on the termination of prolonged larval diapause in the chestnut weevil Curculio sikkimensis (Coleoptera: Curculionidae)

Adults of the chestnut weevil Curculio sikkimensis emerged over a 3-year period under laboratory and quasi-field conditions due to a prolonged diapause that occurred at the mature larval stage. Variable proportions of the larvae remained in diapause after a single cold (5 degrees C) treatment of 120 days. Extension of the chilling period to as long as 540 days did not increase the percentage of diapause termination, and excessively long chilling actually reduced the percentage. Chilling was not indispensable to the termination of larval diapause. Diapause intensity was very high and variable, and more than 1000 days at 20 degrees C was necessary to reactivate all diapause larvae. When the diapause larvae were exposed to cycles of low (5 degrees C for 120 days) and high (20 degrees C for 240 days) temperatures, the percentage of diapause termination reached 100% after two or three such cycles. Thus, the prolonged diapause of C. sikkimensis has characteristics similar to the common short winter diapause in other insects, but has unique characteristics that ensure polymodal reactivation over several years.     

Six new species and two new records of (Pteridophyta) from Costa Rica

Six new species of Pteridophyta from Costa Rica are described: Dryopteris flaccisquama A. Rojas, Hypolepis lellingeri A. Rojas, H. moraniana A. Rojas, Melpomene alan-smithii A. Rojas, Selaginella osa?nsis A. Rojas and Terpsichore esquiveliana A. Rojas. Blechnum stoloniferum (Mett. ex E. Fourn.) C. Chr. and Trichomanes micayense Hieron. are new records for the country. Only D. flaccisquama and M. alan-smithii are present outside Costa Rica.     

Tahiti's native flora endangered by the invasion of Miconia calvescens DC. (Melastomataceae)

Abstract. The native flora of tropical oceanic islands is known to be particularly susceptible both to displacement and extinction, following the invasion of alien organisms. Miconia calvescens DC. (Melastomataceae), first introduced to Tahiti (French Polynesia, South Pacific Ocean) in 1937 as an ornamental plant, now covers over two-thirds of the Island. As it forms dense monotypic stands which have progressively overwhelmed the native forests, this plant pest is a direct threat to the rich Tahitian indigenous flora. Between 40 and 50 species of the 107 species endemic to Tahiti are thought to be on the verge of extinction. M. calvescens was finally declared a'noxious species in French Polynesia'in 1990. Without efficient control efforts and effective endangered plant conservation and protection legislation, M. calvescens could cause Tahiti and all the high islands of French Polynesia to become ecological deserts.     

The Central American species of Centaurium (Gentianaceae)

A revision ofCentaurium based on field, herbarium, and greenhouse studies shows that five species are native to Central America: C.guitense (H.B.K.) B. L. Robinson (n = 21, 22),C. pauciflorum (Mart. & Gal.) B. L. Robinson (n =36),C. brachycalyx Standley & L. O. Williams (n = ca. 36),C. setaceum (Benth.) B. L. Robinson (n = 21), andC. martinii Broome (n = 21), a new species. One population of a European species,C. minus Moench subsp.minus has been discovered in Costa Rica.     

Rediscovery of Hyalinobatrachium chirripoi (Anura: Centrolenidae) in southeastern Costa Rica

The Suretka glass frog, Hyalinobatrachium chirripoi, has been recently rediscovered in the southeastern region of Costa Rica. This species was last reported in Costa Rica in the 1950's. H. chirripoi is distinguished from H. colymbiphyllum, which appears to be its most closely Costa Rican related taxon, by having extensive webbing between fingers II-III.     

Reports in Pteridaceae (Filacales) family in the Neotropics

Several contributions to the Neotropical pteridophyte flora are made in this paper with the range extensions for four species of Adiantum, two species of Doryopteris and one species of Eriosorus. The species concerned are: A. polyphyllum Willd. for Nicaragua and Costa Rica; A. trichochlaenum Mickel et Beitel for Guatemala and Costa Rica; A. villosissimum Mett. ex Kuhn for Costa Rica; A. wilesianum Hook. for Nicaragua; D. nobilis (T. Moore) C. Chr. for Costa Rica, Colombia and Venezuela; D. redivida Fée for Panama; and E. hirtus (Kunth) Copel. for Costa Rica. The concept of D. pedata (L.) Fée is redefined.     

Dipteran parasitoidism on larvae of Caligo atreus (Lepidoptera: Nymphalidae) in Cartago, Costa Rica

Parasitoids on larvae of Caligo atreus were studied at the Estación de Biologia Tropical in Rio Macho, Cartago, Costa Rica. (1 600 masl), from March through July 2000. Fifth instar larvae of C. atreus were placed on Heliconia tortuosa Griggs var. Red Twist (Heliconiaceae) host plants at a mean temperature of 16.7 degrees C. The parasitoids obtained belong to an unidentified species of the genus Winthemia (Diptera: Tachinidae). Most flies emerge some 40 days after the eggs were laid (maximum 68 days). They make an orifice on the upper ventral part of the lepidopteran pupa. Winthemia is used commercially as biological control of cotton and banana.     

Curculionid beetles in aborted flower buds and immature fruits of <Emphasis Type="Italic">Ceiba</Emphasis><Emphasis Type="Italic">pentandra</Emphasis> (Bombacaceae)

We determined the incidence of curculionid beetles of the genus Lonchophorus on aborted and not aborted flower buds and developing fruits of the tree Ceiba pentandra in southeastern Costa Rica. Beetle larvae were found in reproductive parts of all trees sampled trees. The frequency of beetle larvae was greater in aborted buds and immature fruits. A positive correlation between larvae development and flower bud development indicates that female oviposition occurred in an early flower developmental stage and time until bud abortion is variable. Weevil herbivory could be considered as one of the main factors that cause flower bud and fruit abortion in C. pentandra.

---

Resumen Determinamos la incidencia de curculiónidos del género Lonchophorus en botones florales y frutos inmaduros tanto abortados como no abortados en el árbol Ceiba pentandra en el SE de Costa Rica. Las larvas de curculiónidos fueron encontradas en todos los árboles muestreados. La frecuencia de larvas fue mayor en botones y frutos abortados. Una correlación positiva entre el desarrollo larval y el desarrollo del botón indica que la oviposición ocurre en una fase temprana de este y que el tiempo hasta el aborto es variable. La herbivoría por larvas de curculiónidos puede ser uno de los factores más importantes en el aborto de botones y frutos inmaduros en C. pentandra.