Search

One of the essential requirements for an introduced plant species to become invasive is an ability to reproduce outside the native range, particularly when initial populations are small. If a reproductive Allee effect is operating, plants in small populations will have reduced reproductive success relative to plants in larger populations. Alternatively, if plants in small populations experience less competition for pollination than those in large populations, they may actually have higher levels of reproductive success than plants in large populations. To resolve this uncertainty, we investigated how the per capita fecundity of plants was affected by population size in three invasive milkweed species. Field surveys of seed production in natural populations of different sizes but similar densities were conducted for three pollinator-dependent invasive species, namely Asclepias curassavica, Gomphocarpus fruticosus and G. physocarpus. Additionally, supplemental hand-pollinations were performed in small and large populations in order to determine whether reproductive output was limited by pollinator activity in these populations. Reproductive Allee effects were not detected in any of the study species. Instead, plants in small populations exhibited remarkably high levels of reproductive output compared to those in large populations. Increased fruit production following supplemental hand-pollinations suggested that the lower reproductive output of naturally pollinated plants in large populations is a consequence of pollen limitation rather than limitation due to abiotic resources. This is consistent with increased intraspecific competition for pollination amongst plants in large populations. It is likely that the invasion of these milkweed species in Australia has been enhanced because plants in small founding populations experience less intraspecific competition for pollinators than those in large populations, and thus have the ability to produce copious amounts of seeds.

Constitutive and induced plant resistance against herbivores occurs throughout the plant kingdom, but little is known about the evolutionary relationship between these two types of resistances. We examined the relationships between constitutive and induced resistance to the diamondback moth, Plutella xylostella, in 11 wild species of crucifers, and analyzed the changes in volatiles associated with their expression in two species. We used larvae of P. xylostella and jasmonic acid (JA) as elicitors of the induced response. The level of resistance was estimated as the relative number of eggs laid on the plants by P. xylostella. Substantial variation in constitutive resistance was observed among the 11 crucifer species. When the plants were damaged by larvae, a negative correlation was found between constitutive and induced resistance. However, a positive correlation was detected between constitutive and induced resistance when the plants were treated by JA. The shift in resistance was associated with changes in the emission of volatiles. These results strongly suggest that (1) a trade-off occurs between constitutive and induced resistance in wild crucifers, and that (2) such a trade-off can be observed by treating the plants with a natural, but not an artificial, elicitor.

We examined the distribution of butterflies over the mostly arid and semi-arid continent of Australia and analyzed the proportion of migrant species and species diversity with respect to an array of climatic and geographic variables. On a continent-wide scale, latitude explained virtually no variance in either proportion of migrants (r 2 =0.01) or species diversity (r 2 =0.03) in Australian butterflies. These results are in marked contrast to those for temperate-zone birds from three continents where latitude explained between 82 and 98% of the variance in frequency of migrants and also accounted for much of the variance in bird species diversity. In eastern Australia where rainfall regimes are similar to those in temperate Europe and North and South America, latitude explains 78% of the variance in frequency of butterfly migrants. In both eastern and central Australia, latitude also accounts for relatively high proportions of the variance in species diversity. Rainfall patterns and especially soil moisture are negatively associated with migration frequency in Australian butterfly faunas, both alone and in combination with other climate variables. Where moisture levels are relatively high, as in eastern Australia, measures of temperature are associated with migration frequency, a result consistent with findings for temperate-zone birds, suggesting latitude is a surrogate for temperature. The ultimate causes of migration in temperate-zone birds and Australian butterflies are the uneven temporal, and in Australia also spatial, distribution of resources. Uneven distribution is brought about primarily by temperature in temperate regions and by erratic rainfall over much of arid Australia. As a key determinant of productivity, especially in the tropics and subtropics, aridity is likely to be an important determinant of the global distributions of migrants.

Mortality estimates for the immature stages of two butterfly species, Danaus plexippus and D. chrysippus, were obtained by observing the survival of egg cohorts on different sized patches of food plants (Asclepias spp.), over a one-year period. Losses were variable (0-100%) but usually high (90% and over) throughout the year for both species. Most of the losses in both species occurred in the early stages. The mortality by the third instar accounts for 86-100% of the total losses by instar V. Accordingly both species fall into Price's (1975) type A survivorship category. The size of patches of host plants affected losses. The trend was for increasing losses with increasing patch size. A full life-budget is presented for D. plexippus and implications of the observed mortality levels for competition between the two butterfly species is discussed.

Fortnightly fruit fly captures for a 2-year period at Cooloola (south-east Queensland) contained 11 species. Three species, viz., D. tryoni, D. neohumeralis and D. endiandrae predominated. The peak trap catches of 7 species corresponded with the peak fruiting times of their major hosts. There was no direct relationship between temperature and rainfall and the variations in population numbers. The host plants of some species do not grow in the Cooloola area and there is evidence that large numbers of flies migrate into the region from other breeding areas up to 100 km away. Pockets of tropical rainforest such as Cooloola could be important adult fruit fly feeding areas even in the absence of larval host plants.

Infra-red aerial photography was used on two occasions to map the abundance and dispersion of milkweeds (Asclepias spp.). The scale at which the aerial photographs could be interpreted showed that milkweeds occur in large contiguous areas or patches. These patches are abundant, the number in any size-class declining exponentially with increasing size. Analysis of the maps using various dispersion indices and spatial autocorrelation statistic showed that patches have a clumped dispersion. Large patches tend to be surrounded by smaller sized patches. The scales at which clumping occurred indicated a close association between milkweeds and the degree of disturbance of the 'natural' environment by human activities. The utility of resource mapping using aerial photography is demonstrated.

11 fruit fly species captured at 47 sites in a natural forest area at Cooloola (south-east Queensland) revealed specific patterns of spatial abundance. Although all species were collected throughout the study area, D. bryoniae, D. mayi, D. neohumeralis and D. tryoni were more prevalent (average number caught per trap) in the open Eucalypt forest than the rainforest, whereas C. aequalis, D. absonifacies and D. endiandrae were more prevalent in the rainforest. D. cacuminatus, D. choristus, D. quadratus and D. signatifrons were equally prevalent throughout both forest types. Fly numbers were not distributed randomly throughout the trap sites. The clumped dispersion patterns seemed to be species specific as assessed and summarised by Taylor's Power Law. The exponent (b) relating mean spatial abundance to its variance ranged from 1.6-5.11 for the 11 species captured. Changing patterns of trap catches from one sampling period to another were analysed using correlograms for the 6 most abundant species (D. tryoni, D. neohumeralis, D. endiandrae, C. aequalis, D. cacuminatus and D. mayi). These revealed changing patterns of relative spatial abundance which can be related, in part, to changing population abundance levels. The various spatial patterns recognised are related to each species movement, breeding and feeding behaviour. It is proposed that flies migrate into the rainforest area from distant locations and that the rainforest habitat is an important adult feeding site.