Milkweed Matters: Monarch Butterfly (Lepidoptera: Nymphalidae) Survival and Development on Nine Midwestern Milkweed Species.

The population of monarch butterflies east of the Rocky Mountains has experienced a significant decline over the past 20 yr. In order to increase monarch numbers in the breeding range, habitat restoration that includes planting milkweed plants is essential. Milkweeds in the genus Asclepias and Cynanchum are the only host plants for larval monarch butterflies in North America, but larval performance and survival across nine milkweeds native to the Midwest is not well documented. We examined development and survival of monarchs from first-instar larval stages to adulthood on nine milkweed species native to Iowa. The milkweeds included Asclepias exaltata (poke milkweed) (Gentianales: Apocynaceae), Asclepias hirtella (tall green milkweed) (Gentianales: Apocynaceae), Asclepias incarnata (swamp milkweed) (Gentianales: Apocynaceae), Asclepias speciosa (showy milkweed) (Gentianales: Apocynaceae), Asclepias sullivantii (prairie milkweed) (Gentianales: Apocynaceae), Asclepias syriaca (common milkweed) (Gentianales: Apocynaceae), Asclepias tuberosa (butterfly milkweed) (Gentianales: Apocynaceae), Asclepias verticillata (whorled milkweed) (Gentianales: Apocynaceae), and Cynanchum laeve (honey vine milkweed) (Gentianales: Apocynaceae). In greenhouse experiments, fewer larvae that fed on Asclepias hirtella and Asclepias sullivantii reached adulthood compared with larvae that fed on the other milkweed species. Monarch pupal width and adult dry mass differed among milkweeds, but larval duration (days), pupal duration (days), pupal mass, pupal length, and adult wet mass were not significantly different. Both the absolute and relative adult lipids were different among milkweed treatments; these differences are not fully explained by differences in adult dry mass. Monarch butterflies can survive on all nine milkweed species, but the expected survival probability varied from 30 to 75% among the nine milkweed species.

Cardenolide fingerprint of monarch butterflies reared on common milkweed,Asclepias syriaca L.

Monarch butterfly,Danaus plexippus (L.), larvae were collected during August 1983 from the common milkweed,Asclepias syriaca L., across its extensive North American range from North Dakota, east to Vermont, and south to Virginia. This confirms that the late summer distribution of breeding monarchs in eastern North America coincides with the range of this extremely abundant milkweed resource. Plant cardenolide concentrations, assayed by spectrophotometry in 158 samples from 27 collection sites, were biased towards plants with low cardenolide, and ranged from 4 to 229 µg/ 0.1 g dry weight, with a mean of 50 µg/0.1 g. Monarch larvae reared on these plants stored cardenolides logarithmically, and produced 158 adults with a normally distributed concentration range from 0 to 792 µg/0. l g dry butterfly, with a mean of 234 µg/0.1 g. Thus butterflies increased the mean plant cardenolide concentration by 4.7. The eastern plants and their resultant butterflies had higher cardenolide concentrations than those from the west, and in some areas monarchs sequestered more cardenolide from equivalent plants. Plants growing in small patches had higher cardenolide concentrations than those in larger patches, but this did not influence butterfly concentration. However, younger plants and those at habitat edges had higher cardenolide concentrations than either older, shaded, or open habitat plants, and this did influence butterfly storage. There were no apparent topographical differences reflected in the cardenolides of plants and butterflies. Twenty-eight cardenolides were recognized by thin-layer chromatography, with 27 in plants and 21 in butterflies. Butterflies stored cardenolides within the more polar 46% of the plantR d range, these being sequestered in higher relative concentrations than they occurred in the plants. By comparison with published TLC cardenolide mobilities, spots 3, 4, 9, 16, 24 or 25, 26, and 27, may be the cardenolides syrioside, uzarin, syriobioside, syriogenin, uzarigenin, labriformidin, and labriformin, respectively. Cochromatography with cardenolide standards indicated that desglucosyrioside did not occur in the plants but did occur in 70% of the butterflies, and aspecioside was in 99% of the plants and 100% of the butterflies. The polar aspecioside was the single most concentrated and diagnostic cardenolide in both plants and butterflies. ButterflyR d values were dependent on those of the plant, and both showed remarkable uniformity over the range of areas sampled. Thus contrary to previous reports,A. syriaca has a biogeographically consistent cardenolide fingerprint pattern. The ecological implications of this for understanding the monarch's annual migration cycle are significant.

Cardenolide connection between overwintering monarch butterflies from Mexico and their larval food plant,Asclepias syriaca.

The majority (85%) of 394 monarch butterflies sampled from overwintering sites in Mexico contain the same epoxy cardenolide glycosides, including most conspicuously a novel polar glycoside with a single genin-sugar bridge (aspecioside), as occur in the milkweedsAsclepias speciosa andA. syriaca. This cardenolide commonality was established by isolating aspecioside and syriobioside from the wings of overwintering monarchs and the two plant species, and comparing Chromatographie and NMR spectrometric characteristics of the isolates. When combined with the migratory pattern of monarchs and the distribution of these two milkweed species, this chemical evidence lends strong support to the hypothesis thatA. syriaca is the major late summer food plant of monarchs in eastern North America. This finding may be of ecological importance, forA. syriaca contributes less cardenolide and cardenolides of lower emetic potency to monarchs than most milkweeds studied to date.

A natural toxic defense system: cardenolides in butterflies versus birds.

We have verified that wild birds can become conditioned to reject naturally toxic insects either visually (experiment 1) or by taste (experiment 2). We have also verified, however, that unconditioned taste rejection of noxious chemicals by wild birds also occurs (experiment 3). Such unconditioned responses to the aposematic visual and taste cues of many insects, in fact, often appear to be as important as, or more important than, conditioned responses. In a large number of laboratory feeding experiments with wild birds as predators of aposematic insects, initial and/or long-term rejection occurs without prior laboratory conditioning experience. Although in some experiments the birds may have previously been exposed to (and therefore perhaps conditioned by) the aposematic prey in the wild, other experiments have used naive birds or insects whose ranges do not overlap those of the birds. Wiklund and Jarvi, for example, tested the response of 47 naive hand-raised birds of four species to five aposematic insect species, and found that 69/136 (51%) insects were rejected visually without even tasting, while 63 were tasted and then rejected. Only four of the insects were actually ingested. Similarly, in Bowers' study of the response of Massachusetts blue jays to aposematic western U.S. Euphydryas butterflies, several blue jays consistently rejected the butterflies visually or by taste without having eaten any. While these studies were not designed to separate neophobic effects from innate visual and/or taste aversions, they do differentiate between conditioned and unconditioned responses. Since both conditioned and unconditioned rejections can be demonstrated in the lab by insectivorous birds, and our available field evidence does not yet let us distinguish the mechanisms behind the observed patterns, our initial question, of the relative importance of conditioned versus unconditioned rejection mechanisms in different natural situations, is not yet answerable. The most important requirement of a food-rejection strategy is that it prevents both poisoning and starvation. We have shown, however, that rejection of a noxious insect by a bird can take place at four distinct levels (visual, non-destructive taste sampling, destructive taste sampling, or post-ingestional physiological rejection), the first three of which may be either unconditioned or conditioned by a physiological reaction to ingestion.(ABSTRACT TRUNCATED AT 400 WORDS)

Plant-determined variation in cardenolide content and thin-layer chromatography profiles of monarch butterflies,Danaus plexippus reared on milkweed plants in California : 3. Asclepias californica.

Variation in gross cardenolide concentration of the mature leaves of 85Asclepias californica plants collected in four different areas of California is a positively skewed distribution ranging from 9 to 199 µg of cardenolide per 0.1 g dry weight with a mean of 66 µg/0.1 g. Butterflies reared individually on these plants in their native habitats contained a normal distribution of cardenolide ranging from 59 to 410 µg of cardenolide per 0.1 g dry weight with a mean of 234 µg. Cardenolide uptake by the butterflies was a logarithmic function of plant concentration. Total cardenolide per butterfly ranged from 143 to 823 µg with a mean of 441 µg and also was normally distributed. Populational variation of plant cardenolide concentrations occurs within subspecies, but the northern subspeciesA. c. greenei does not differ significantly from the southernA. c. californica. Generally higher concentrations occur in butterflies from northern populations and in females. No evidence was adduced that cardenolides in the plants adversely affected the butterflies. Low cardenolide concentrations in the leaves and the absence of cardenolides in the latex characterize bothA. californica andA. speciosa, but notA. eriocarpa. Thin-layer chromatography in two solvent systems isolated 24 cardenolide spots in the plants, of which 18 are stored by the butterflies. There was a minor difference in the cardenolide spot patterns due to geographic origin of the plants, but as in our previous studies, none in the sexes of the butterflies. UnlikeA. eriocarpa andA. speciosa, A. californica plants lack cardenolides withRf values greater than digitoxigenin. Overall, the cardenolides of bothA. californica andA. speciosa are more polar than those inA. eriocarpa. A. californica plants contain cardenolides of the calotropagenin series including calotropin, calactin, and uscharidin, and the latter is metabolically transformed by monarch larvae to calactin and calotropin. Cardenolides of this series also occur inA. vestita, andA. cordifolia from California, the neotropicalA. curassavica, and the AfricanCalotropis procera, Gomphocarpus spp., andPergularia extenso; they therefore cross established taxonomic lines.A. californica is the predominant early season milkweed in California and may be important in providing chemical protection to the spring generation of monarchs in the western United States.A. speciosa, A. eriocarpa, andA. californica each imparts distinctive cardenolide fingerprints to the butterflies, so that ecological predictions are amenable to testing.

Plant-determined variation in the cardenolide content, thin-layer chromatography profiles, and emetic potency of monarch butterflies,Danaus plexippus L. Reared on milkweed plants in California: 2.Asclepias speciosa.

The pattern of variation in gross cardenolide concentration of 111Asclepias speciosa plants collected in six different areas of California is a positively skewed distribution which ranges from 19 to 344 µg of cardenolide per 0.1 g dry weight with a mean of 90 µg per 0.1 g. Butterflies reared individually on these plants in their native habitats ranged from 41 to 547 µg of cardenolide per 0.1 g dry weight with a mean of 179 µg. Total cardenolide per butterfly ranged from 54 to 1279 µg with a mean of 319 µg. Differences in concentrations and total cardenolide contents in the butterflies from the six geographic areas appeared minor, and there were no differences between the males and the females, although the males did weigh significantly more than females. The uptake of cardenolide by the butterflies was found to be a logarithmic function of the plant concentration. This results in regulation: larvae which feed on low-concentration plants produce butterflies with increased cardenolide concentrations relative to those of the plants, and those which feed on high-concentration plants produce butterflies with decreased concentrations. No evidence was adduced that high concentrations of cardenolides in the plants affected the fitness of the butterflies. The mean emetic potencies of the powdered plant and butterfly material were 5.62 and 5.25 blue jay emetic dose fifty units per milligram of cardenolide and the number of ED50 units per butterfly ranged from 0.28 to 6.7 with a mean of 1.67. Monarchs reared onA. speciosa, on average, are only about one tenth as emetic as those reared onA. eriocarpa. UnlikeA. eriocarpa which is limited to California,A. speciosa ranges from California to the Great Plains and is replaced eastwards byA. syriaca L. These two latter milkweed species appear to have a similar array of chemically identical cardenolides, and therefore both must produce butterflies of relatively low emetic potency to birds, with important ecological implications. About 80% of the lower emetic potency of monarchs reared on A. speciosa compared to those reared onA. eriocarpa appears attributable to the higher polarity of the cardenolides inA. speciosa. Thin-layer Chromatographie separation of the cardenolides in two different solvent systems showed that there are 23 cardenolides in theA. speciosa plants of which 20 are stored by the butterflies. There were no differences in the cardenolide spot patterns due either to geographic origin or the sex of the butterflies. As when reared onA. eriocarpa, the butterflies did not store the plant cardenolides withR f values greater than digitoxigenin. However, metabolic transformation of the cardenolides by the larvae appeared minor in comparison to when they were reared onA. eriocarpa. AlthoughA. eriocarpa andA. speciosa contain similar numbers of cardenolides and both contain desglucosyrioside, the cardenolides ofA. speciosa overall are more polar. ThusA. speciosa has no or only small amounts of the nonpolar labriformin and labriformidin, whereas both occur in high concentrations inA. eriocarpa. A. speciosa plants and butterflies also contain uzarigen, syriogenin, and possibly other polar cardenolides withR f values lower than digitoxin. The cardenolide concentration in the leaves is not only considerably less than inA. eriocarpa, but the latex has little to immeasurable cardenolide, whereas that ofA. eriocarpa has very high concentrations of several cardenolides. Quantitative analysis ofR f values of the cardenolide spots, their intensities, and their probabilities of occurrence in the chloroform-methanol-formamide TLC system produced a cardenolide fingerprint pattern very different from that previously established for monarchs reared onA. eriocarpa. This dispels recently published skepticism about the predictibility of chemical fingerprints based upon ingested secondary plant chemicals.

Cardenolide sequestration by the dogbane tiger moth (Cycnia tenera; Arctiidae).

Cycnia tenera adults, reared as larvae onAsclepias humistrata, had 10 times higher cardenolide concentrations, and contained 15 times more total cardenolide, than did moths reared onA. tuberosa. Thin-layer chromatography confirmed that each individual cardenolide visualized in the adult moths reared on the former host plant corresponds to one present in the plant, thus demonstrating that the insects' cardenolides are indeed derived from the larval food. Adult weights were significantly greater when the larvae had been fed upon the higher cardenolide plant species,A. humistrata. Similar results for other milkweed-feeding insects have been interpreted by some authors as evidence against a metabolic cost of handling cardenolides. However, such interpretations confound cardenolide differences among milkweed species with other differences in plant primary and secondary chemistry that affect insect growth and development. While the cooccurrence inC. tenera of other noxious chemicals (e.g., alkaloids) is not precluded, cardenolides sequestered from larval host plants have probably contributed to the evolution of visual and auditory aposematism in this species. As the eggs are laid in large clutches and larvae are gregarious, such aposematism may have evolved via kin selection.

Plant-determined variation in the cardenolide content, thin-layer chromatography profiles, and emetic potency of monarch butterflies,Danaus plexippus reared on the milkweed,Asclepias eriocarpa in California.

This paper is the first in a series on cardenolide fingerprinting of the monarch butterfly. New methodologies are presented which allow both qualitative and quantitative descriptions of the constituent cardenolides which these insects derive in the wild from specificAsclepias foodplants. Analyses of thin-layer Chromatographic profiles ofAsclepias eriocarpa cardenolides in 85 individual plant-butterfly pairs collected at six widely separate localities in California indicate a relatively invariant pattern of 16-20 distinct cardenolides which we here define as theAsclepias eriocarpa cardenolide fingerprint profile. Cardenolide concentrations vary widely in the plant samples, but monarchs appear able to regulate total storage by increasing their concentrations relative to their larval host plant when reared on plants containing low concentrations, and vice versa. Forced-feeding of blue jays with powdered butterfly and plant material and with one of the constituent plant cardenolides, labriformin, established that theA. eriocarpa cardenolides are extremely emetic, and that monarchs which have fed on this plant contain an average of 16 emetic-dose fifty (ED50) units. The relatively nonpolar labriformin and labriformidin in the plant are not stored by the monarch but are metabolized in vivo to desglucosyrioside which the butterfly does store. This is chemically analogous to the way in which monarchs and grasshoppers metabolize another series of milkweed cardenolides, those found inA. curassavica. It appears that the sugar moiety through functionality at C-3' determines which cardenolides are metabolized and which are stored. The monarch also appears able to store several lowR f cardenolides fromA. eriocarpa without altering them. Differences in the sequestering process in monarchs and milkweed bugs (Oncopeltus) may be less than emphasized in the literature. The monarch is seen as a central organism involved in a coevolutionary triad simultaneously affecting and affected by both its avian predators and the secondary chemistry of the milkweeds with which it is intimately involved.

Seasonal and intraplant variation of cardenolide content in the California milkweed,Asclepias eriocarpa, and implications for plant defense.

Root, stem, leaf, and latex samples ofAsclepias eriocarpa collected from three plots in one population at 12 monthly intervals were assayed for total cardenolide content by spectroassay and for individual cardenolides by thin-layer chromatography. From May to September mean milligram equivalents of digitoxin per gram of dried plant were: latices, 56.8 ≫ stems, 6.12 > leaves, 4.0 > roots, 2.5. With the exception of the roots, significant changes in gross cardenolide content occurred for each sample type with time of collection during the growing season, whereas variation within this population was found to be small. Labriformin, a nitrogen-containing cardenolide of low polarity, predominated in the latices. Leaf samples contained labriformin, labriformidin, desglucosyrioside, and other unidentified cardenolides. In addition to most of the same cardenolides as the leaves, the stems also contained uzarigenin. The roots contained desglucosyrioside and several polar cardenolides. The results are compared with those for other cardenolide-containing plants, and discussed in relation to anti-herbivore defense based on plant cardenolide content. Arguments are advanced for a central role of the latex in cardenolide storage and deployment which maximizes the defensive qualities of the cardenolides while preventing toxicity to the plant.

Mortality of the Monarch Butterfly (Danaus plexippus L.): Avian Predation at Five Overwintering Sites in Mexico.

Analyses of predated butterflies on the forest floor at five monarch overwintering sites in Mexico and observations of birds foraging in mixed flocks indicate that individual birds of several species have learned to penetrate the monarch's cardenolide-based chemical defense. Predation is inversely proportional to colony size and appears to be one evolutionary explanation of the dense aggregations.

Localization of heart poisons in the monarch butterfly.

The cardiac glycosides that monarch butterflies sequester from milkweed plants during the larval stage differ remarkably in their emetic potency and are concentrated to different degrees in the various parts of the body as well as in the two sexes (Fig. 1). The very high concentrations of these compounds in the wings probably facilitate learned taste rejection in predators and account for the relatively high frequency of Danaid butterflies with beak-marked wings in natural populations. The cardiac glycosides in the abdomen have a much higher emetic potency than those in the rest of the body. Consequently, naive, extremely hungry, or forgetful birds which capture and peck off the wings but eat the abdomen discard the least emetic glycosides and ingest the most emetic, and thus again experience emesis. The nonrandom distribution of cardenolides in the wings, abdomen, and thorax, together with the fact that monarch males not only contain lower concentrations of cardiac glycosides than females but also contain cardenolides that are overall less emetic than those in females, is interpreted as evidence that these poisons are incorporated at a physiological cost. This cost, balanced against the benefits of protection from predation, provides a selective basis for the occurrence of both emetic and nonemetic individuals in natural populations. Since birds can discriminate emetic from nonemetic monarchs on the basis of taste, it is not necessary to invoke theories of kind of group selection to explain the evolution of this kind of unpalatability.

Theoretical investigations of automimicry: multiple trial learning and the palatability spectrum.

We previously explored automimicry assuming that a species of prey was so unpalatable as to promote conditioned avoidance for a period of time after a predator encountered a single individual (Case 1). In this paper, we assume that the prey is less noxious and that two encounters are required. Case 2 allows the two encounters with unpalatables to be separated by any number of palatables, while in Case 3 the predator must encounter two unpalatables, consecutively.The general relationships in the three cases are similar, but the automimetic advantage is reduced moderately in Case 2 and greatly in Case 3. To attain the same automimetic advantage as in Case 1 requires an increase in the proportion of unpalatables, or in the induced rejection period, or both. Consequently, selection will tend to increase the unpalatability so that Cases 2 and 3 converge to Case 1.Species that are uniformly and highly unpalatable can afford to be more dispersed than automimetic species. Case-2 and -3 automimetic species will benefit greatly from gregariousness, while in Case-1 automimicry situations this is less important.

Variation in cardiac glycoside content of monarch butterflies from natural populations in eastern North America.

A new spectrophotometric assay has been used to determine the gross concentration of cardiac glycoside in individual monarch butterflies. Adults sampled during the fall migration in four areas of eastern North America exhibited a wide variation in cardiac glycoside concentration. The correlation between spectrophotometrically measured concentrations and emetic dose determinations supports the existence of a broad palatability spectrum in wild monarch butterflies. The cardiac gylcoside concentration is greater in females than in males and is independent of the dry weight of the butterflies; contrary to prediction, both the concentration mean and variance decrease southward. The defensive advantage of incorporating cardiac glycosides may be balanced by detrimental effects on individual viability.

Theoretical investigations of automimicry, I. Single trial learning.

The theory of automimicry is explored mathematically on the assumption that predators can learn to avoid noxious prey by sight for some finite period after a single noxious experience. Automimetic advantage is an inevitable consequence of the evolution of an unpalatability dimorphism. An established automimetic situation is analogous to an established perfect Batesian mimicry situation, although the evolutionary bases of the two phenomena are different. In both situations, the mimetic advantage depends upon the proportion of unpalatable prey, the memory span of the predators, and the abundance of the prey relative to the predators. Automimetic advantage is maximal when the prey are neither too common nor too rare. Remarkably low proportions of unpalatable prey can confer very substantial immunity to the population. A surprising prediction of the model is that the evolution of unpalatability will not occur in rare prey species unless they first become Batesian mimics. This in turn could lead to the evolution of mimicry complexes containing many species forming a whole spectrum of unpalatability.

Ecological chemistry and the palatability spectrum.

A new bioassay for comparing the palatability to avian predators of monarch butterflies reared on various asclepiadaceous food plants containing cardiac glycosides indicates a palatability spectrum. The monarchs reared on one plant species are six times as emetic as those fed another, while those raised on an asclepiad which lacks cardiac glycosides are not emetic at all.

Major components in the exocrine secretion of a male butterfly (lycorea).

Extracts of the extrusible secretion-disseminating organs ("hairpencils") of the male of the danaid butterfly, Lycorea ceres ceres, from Trinidad, contain a pyrrolizidine and two aliphatic esters. An odorous component, present in trace amounts, remains unidentified. Judging from the function of "hairpencils" in a related species, the secretion may play a mediating role in courtship.