USF Libraries

Sugar preference across butterfly families

MISSING IMAGE

Material Information

Title:
Sugar preference across butterfly families
Translated Title:
Preferencia del azúcar a través de las familias de mariposas ( )
Physical Description:
Book
Language:
English
Creator:
Olsson, Juliana
Publication Date:

Subjects

Subjects / Keywords:
Butterflies--Feeds and feeding   ( lcsh )
Butterflies--Phylogeny   ( lcsh )
Mariposas--Alimentos y alimentación
Mariposas--Filogenia
Tropical Ecology 2008
Ecología Tropical 2008
Genre:
Reports   ( lcsh )
Reports

Notes

Abstract:
Adult butterflies have a carbohydrate-based diet, acquiring their sugars mostly from nectar or from rotting fruits. Fruit-eating butterflies most often encounter fructose and glucose (Omura and Honda, 2003), while nectarivorous butterflies may encounter hexose-rich nectars in short-corolla flowers or sucrose-rich nectars in long-corolla flowers (Baker and Baker, 1983). This study explores the possibility that butterfly species exhibit certain sugar preferences that reflect not only their diet but their phylogeny as well, since feeding habits are often the result of co-evolution with pollination syndromes in the case of nectar-feeding butterflies. Butterflies from as many species as possible were given a solution of either 20% sucrose by weight or a 20% solution of combined glucose and fructose. The amount imbibed was recorded for the 26 species caught. No obvious preference for either sucrose or hexose nectars were uncovered. Butterflies would drink from either solution, indicating that fruits and flowers offering either sugar would be acceptable to most butterflies and that sugar preference is not as rigidly an evolved trait as was thought. However, minor trends towards sucrose preference are compelling grounds for further study.
Abstract:
Las mariposas adultas tienen una dieta basada en carbohidratos, consiguen los azucares principalmente del néctar o de los frutos en descomposición. Las mariposas nectarívoras encuentran néctar rico en hexosa en flores con corolas cortas o néctar rico en sucosa en flores con corolas largas. Este estudio explora la posibilidad de que algunas mariposas exhiban alguna preferencia por azucares que refleje no solo su dieta pero también su filogenia, ya que los hábitos alimenticios a veces son el resultado de co-evolución con los síndromes de polinización.
Language:
Text in English.
General Note:
Born Digital

Record Information

Source Institution:
University of South Florida Library
Holding Location:
University of South Florida
Rights Management:
All applicable rights reserved by the source institution and holding location.
Resource Identifier:
usfldc doi - M39-00034
usfldc handle - m39.34
System ID:
SFS0001210:00001


This item is only available as the following downloads:


Full Text
xml version 1.0 encoding UTF-8 standalone no
record xmlns http:www.loc.govMARC21slim xmlns:xlink http:www.w3.org1999xlink xmlns:xsi http:www.w3.org2001XMLSchema-instance
leader 00000nas 2200000Ka 4500
controlfield tag 008 000000c19749999pautr p s 0 0eng d
datafield ind1 8 ind2 024
subfield code a M39-00034
040
FHM
0 041
eng
049
FHmm
1 100
Olsson, Juliana
242
Preferencia del azcar a travs de las familias de mariposas
245
Sugar preference across butterfly families
260
c 2008-05
500
Born Digital
3 520
Adult butterflies have a carbohydrate-based diet, acquiring their sugars mostly from nectar or from rotting fruits. Fruit-eating butterflies most often encounter fructose and glucose (Omura and Honda, 2003), while nectarivorous butterflies may encounter hexose-rich nectars in short-corolla flowers or sucrose-rich nectars in long-corolla flowers (Baker and Baker, 1983). This study explores the possibility that butterfly species exhibit certain sugar preferences that reflect not only their diet but their phylogeny as well, since feeding habits are often the result of co-evolution with pollination syndromes in the case of nectar-feeding butterflies. Butterflies from as many species as possible were given a solution of either 20% sucrose by weight or a 20% solution of combined glucose and fructose. The amount imbibed was recorded for the 26 species caught. No obvious preference for either sucrose or hexose nectars were uncovered. Butterflies would drink from either solution, indicating that fruits and flowers offering either sugar would be acceptable to most butterflies and that sugar preference is not as rigidly an evolved trait as was thought. However, minor trends towards sucrose preference are compelling grounds for further study.
Las mariposas adultas tienen una dieta basada en carbohidratos, consiguen los azucares principalmente del nctar o de los frutos en descomposicin. Las mariposas nectarvoras encuentran nctar rico en hexosa en flores con corolas cortas o nctar rico en sucosa en flores con corolas largas. Este estudio explora la posibilidad de que algunas mariposas exhiban alguna preferencia por azucares que refleje no solo su dieta pero tambin su filogenia, ya que los hbitos alimenticios a veces son el resultado de co-evolucin con los sndromes de polinizacin.
546
Text in English.
650
Butterflies--Feeds and feeding
Butterflies--Phylogeny
4
Mariposas--Alimentos y alimentacin
Mariposas--Filogenia
653
Tropical Ecology 2008
Ecologa Tropical 2008
655
Reports
720
CIEE
773
t Monteverde Institute : Tropical Ecology
856
u http://digital.lib.usf.edu/?m39.34



PAGE 1

Sugar Preference across Butterfly Families Juliana Olsson Department of Integrative Biology, University of Ca lifornia at Berkeley ABSTRACT Adult butterflies have a carbohydrate-based diet, a cquiring their sugars mostly from nectar or from ro tting fruits. Fruit-eating butterflies most often encount er fructose and glucose (Omura and Honda, 2003), wh ile nectarivorous butterflies may encounter hexose-rich nectars in short-corolla flowers or sucrose-rich n ectars in long-corolla flowers (Baker and Baker, 1983). Th is study explores the possibility that butterfly sp ecies exhibit certain sugar preferences that reflect not only their diet but their phylogeny as well, since feeding habits are often the result of co-evolution with po llination syndromes in the case of nectar-feeding butterflies. Butterflies from as many species as po ssible were given a solution of either 20% sucrose by weight or a 20% solution of combined glucose and fr uctose. The amount imbibed was recorded for the 26 species caught. No obvious preference for either su crose or hexose nectars were uncovered. Butterflies would drink from either solution, indicating that f ruits and flowers offering either sugar would be acceptable to most butterflies and that sugar prefe rence is not as rigidly an evolved trait as was tho ught. However, minor trends towards sucrose preference ar e compelling grounds for further study. RESUMEN Las mariposas tienen una dieta basada en carbohidra tos, consiguen los azucares principalmente en la forma fructosa y glucosa, de nectar o de frutos en decomposicin. Las mariposas nectarvoras encuentran nctar rico e n hexosa en flores con corolas cortas o nctar rico en sucrosa en flores con corolas larg as. Este estudio explora la posibilidad de que algunas mariposas exhiban alguna preferencia por azucares que refleje no solo su dieta pero tambin su filogenia, ya que los habitos alimenticios a veces son el resultado de co-evolucin con los sindromes de polinizacin. Mariposas de varias especies fueron alimentadas con sucrosa al 20% o una solucin al 20 % de glucosa/fruxtosa. La cantidad consumida fue medida para 26 especies. No se detect o ninguna preferencia. Esto indica que las flores y frutos ofreciendo cualquiera de lo s azucares sern aceptados por la mayora de las mariposas, y que la preferencia de a zcar no es un carcter evolutivo tan rigido con se pensaba. INTRODUCTION Most adult butterflies depend on sugar solutions fo r their diet, which they get either from nectars (Romeis and Wackers, 2002) or juices of rot ting fruit (Omura and Honda, 2003). The nectar-feeding butterflies are important pollin ators of the plants they depend on, and certain pollination syndromes have evolved to ensur e a tighter relationship between flower and butterfly (Baker and Baker, 1983). While flower shape and color are agreed to be important components of pollination syndromes, t he importance of the content of the nectar is less certain. Baker and Baker (1983) found that plants are fairl y constant within a species in terms of sucrose-dominance or hexose-dominance of t heir nectars, regardless of variation

PAGE 2

in the environment. Plants with the same pollinator type exhibit similarities in sugar ratios even without being taxonomically related. Th is strongly suggests that pollinators, like butterflies, have important sugar preferences that can shape the evolution of nectars. Most nectars contain sucrose, glucose and fructose; only a few have only one detectable sugar, and none of these contain fructose alone, or sucrose and fructose without glucose. In general, Percival (1961) noted that families wit h deep-tubed flowers tend to be sucrose-rich while shallow-tubed or cupped flowers are most often hexose-rich. Baker and Baker (1983) distinguish butterfly-visited flow ers into two groups: those with deep narrow corollas characteristically rich in sucrose and visited primarily by butterflies, and those smaller short-tubed hexose-rich flowers visit ed equally by bees and butterflies. Sugar preference studies have been done in birds ( Martinez del Rio et al., 1992), showing that nectar-feeding hummingbirds have speci alized on sucrose, but that most fruit-eating passerines cannot digest sucrose and i nstead show a preference for fruits containing the monosaccharide sugars fructose and g lucose. Plants pollinated and dispersed by different birds thus use different typ es of sugars in their nectars and fruit pulps as rewards. If birds exhibit such diet-relate d preferences for sugar types, one would expect fruit-eating butterflies to possibly special ize on fructose and glucose, while nectarfeeding butterflies of at least long-corolla flower s may prefer sucrose. Baker and Baker (1983) did a survey of plant famil ies and found that pollinator syndrome is the most important component of which s ugar type a flower offers. Hummingbird pollinated flowers are sucrose rich and hummingbirds prefer sucrose. The case for butterflies and their flowers is still unk nown. Butterflies may exhibit sugar preferences that reflect their diet, on fruit or ei ther type of nectar. These patterns of preference may fall along taxonomic lines if mechan isms like the enzyme deficiencies reported in birds (Martinez del Rio et al., 1992) e xist in butterflies. Another possibility, that most butterflies have the enzymatic ability to use sucrose and hexose sugars, would explain why butterfly flowers supply either sugar t ype. Some studies have been done to determine sugar pre ference in individual butterfly species, but one does not get an idea of big-pictur e trends from them. Watt et al. (1974) found no preference in Colias sp., as it was found that this genus has enzymes t o process both hexose and sucrose sugars. Pieris brassicae a species of Pierid, was found to prefer sucrose to fructose (Romeis and Wackers, 2000), as Battus philenor in Erhardt’s 1991 feeding experiments. Even though fructose ands gluc ose were the major sugars found in the fruits eaten by Nymphalis xanthomelas Kanisca canace and Vanessa indica sucrose was the most effective feeding stimulant in these s pecies, with fructose following close behind (Omura and Honda 2003). Ithomiines specializ e on members of Asteraceae and Boraginaceae (Baker and Baker, 1983), which have he xose-rich nectars, so it is hypothesized that they prefer hexose sugars to sucr ose. Besides these findings, few studies have been devoted to discovering sugar pref erences in butterfly species let alone in multiple butterfly families. This experiment aims to see if a relationship betw een sugar preference and butterfly family phylogenies exists. Butterflies in this study were fed a solution containing only the sugars of interest, without pre senting flower options. Though other monosaccharide and disaccharide sugars may be prese nt in nectars, they occur in minute amounts, which is why this study focuses on sucrose glucose and fructose in terms of sugar preference in butterflies.

PAGE 3

MATERIALS AND METHODS Study Site The experiment was conducted at the Selvatura butte rfly garden and the Monteverde Butterfly Garden, both in Monteverde. Butterflies w ere collected from inside the gardens, while feedings were conducted in a separate room at the Monteverde Butterfly Garden, and in an eclosion chamber at Selvatura. Preparation of Sugar Solutions The sucrose solution consisted of 20% sucrose by we ight. The hexose mix consisted of 10% fructose by weight and 10% glucose by weight. B oth solutions were stored in capped containers and remade every other day. Catching and Storing Butterflies were caught using a simple butterfly ne t, and then transferred to a 1 m tall cylindrical holding net, where they were kept for a t least two hours before feeding, or overnight in the case of most of the morphos and he liconiines. This delay ensured that the butterflies would be less active and more inclined to eat the sugar solution presented to them without trying to escape. Butterflies were cau ght in the late morning between 8 and 11 a.m. when they are most active, and were identif ied using The Butterflies of Costa Rica and their Natural History (DeVries, 1987). Feeding Butterflies were transferred individually to the fe eding room, where they were placed on wax paper in front of a bead of 100 m L of sugar solution. While a nearby light bulb shon e on the butterfly in order to heat it and induce it to feed, the butterfly was held lightly by the wings and its proboscis was unrolled with an un bent paperclip to place the tip of the proboscis in the liquid. At this point, if the butt erfly immediately started to feed, I would release its wings and let it feed until it was read y to fly away; otherwise, I would continue keeping its proboscis extended in the sugar solutio n until it kept it extended on its own for at least three seconds. Every butterfly that di d not start feeding immediately was given three chances to eat. At the end of a feeding session, when the butterfly was satiated or had used up its three chances, the rema ining sugar solution was sucked up by 20 m L microcapillary tubes and the volume measured. The wax paper would then be cleaned, a new bead of 100 m L would be applied by blowing sugar solution out of a microcapilllary tube, and a new butterfly would be pulled from the holding net. This same feeding process was used for sucrose feedings and hexose feedings, though they were conducted on separate pieces of wax paper and with separate microcapillary tubes in order to avoid contamination. RESULTS

PAGE 4

Thirty three species from the families Pieridae, Pa pilionidae and Nymphalidae were tested. Of these, 26 species of three families and five nymphalid subfamilies were surveyed for both sugar solutions. Mean hexose volu mes were subtracted from mean sucrose volumes for all species that had representa tives from both feedings, giving positive or negative values. Differences of <5% wer e listed as zero and discounted. The result of the Sign Test pointed to 14 cases where a species preferred sucrose and nine cases where a species preferred the fructose-glucos e mixture. In three cases there was little or no difference. Overall, butterflies do no t show a preference for sucrose or hexoserich sugars (Sign test, n=14 plus, 9 negative, p> 0 .05). For all three Papilionid species tested, mean sucr ose consumption was greater than mean hexose consumption, though in one trial, the difference was just 7 m L. Three of four Pierid species seemed to prefer sucrose to hexose. In these cases, there were too few species to test trends statistically. With eight negative and eight positive points for the Sign Test, no trend could be found in sugar preference, either in Nymphalidae as a family, or within the subfamilies Ithomiinae, Morphinae, Heliconiinae and Nymphalinae Further Observations On the whole, the Pierid butterflies were very fini cky eaters, and it was difficult to get them to take up even a little of either sugar solut ion. The Papilionids and certain members of Heliconiines were hard to feed as well, but to a lesser extent. Ithomiines and Morphos would readily eat, but whereas the smaller glass-wi ng Ithomiines took half an hour to eat about 30 m L, the Morphos would eat the entire hundred m L to the point where I had to start feeding them more until they were satiated. Parides individuals and most Pierid individuals were extra ordinarily active when caught, but could not be stored overnight since the y would die, so they were only stored between two and four hours. TABLE 1. Sugar consumption in three butterfly fami lies. Members of the family Nymphalidae are shown in red, Papilionidae in green and Pieridae in blue. Butterfly individuals were collected from the Monteverde Butt erfly Garden and the butterfly garden at Selvatura. Species Sucrose Sucrose Hexose Hexose Number fed Mean volume eaten ( m L) Number fed Mean volume eaten ( m L) Greta oto 5 30.682 5 26.788 Ithomia heraldica 1 36.05 1 36.8 Pteronymia fumida 1 40.26 0 — Morpho granadensis 4 58.2875 3 62.544 Morpho peleides 5 83.364 5 78.474 Caligo eurilochus sulanus 1 33.4 1 400 Caligo memnon memnon 1 346.3 2 344.2

PAGE 5

Danaus plexippus 1 89.7 1 76.6 Dione moneta poeyii 3 55.089 2 44.737 Dryadula phaetusa 6 30.2183 1 49.5 Dryas iulia 2 40.791 3 58.3 Eueides isabella 4 31.91 5 34.788 Heliconius charitonius 3 29.123 3 31.929 Heliconius hecale zuleika 2 70.13 3 65.878 Heliconius sapho leuces 4 23.51 4 26.117 Heliconius sara fulgidus 2 23.29 2 24.08 Heliconius erato 2 35.79 2 26.449 Catonephele numilia esite 3 56.228 3 31.053 Myscelia cyaniris cyaniris 3 30.263 3 36.404 Siproeta steleres 2 103.553 2 59.6 Battus polydamas polydamas 4 49.212 4 24.5 Parides lycimenes lycimenes 2 77.434 1 59.145 Parides arcas mylotes 0 — 1 12.1 Parides iphidamas iphidamas 0 — 1 68.4 Parides erithalion 0 — 1 11.3 Papilio astyalus pallas 1 89.47 1 81.3 Papilio polyxenes 1 36.3 0 — Papilio thoas nealces 0 — 1 177.6 Ascia limona 1 43.16 1 59.2 Phoebis sennae 2 46.316 3 23.86 Phoebis philea philea 1 59.5 1 55.5 Anteos clorine 0 — 1 92.6 Appias drusilla 1 42.1 2 18.421

PAGE 6

FIGURE 1. Preference of sugar types by Papilionidae Bars show results of a Sign Test where the mean volume of hexose solution eaten was subtracted from the mean volume of sucrose eaten. Bars pointing toward sucrose pref erence represent positive results. Negative results indicate hexose preference. FIGURE 2. Preference of sugar types by Pieridae. P ositive results of a sucrose-minushexose Sign Test are shown as bars pointing to sucr ose preference, while negative results point to hexose preference.

PAGE 7

FIGURE 3. Preference of sugar types by Nymphalidae. Subfamily Ithomiinae shown in red, Morphinae in blue, Danainae in black, Heliconi inae in green and Nymphalinae in purple. Positive results of a Sign Test indicate su crose preference; negative results indicate hexose preference. Neutral preference indi cates <5% difference in mean consumed volumes. DISCUSSION Even though the Sign Test showed no overall sugar p reference across butterfly species, this is to be expected. The butterflies examined in clude three familes and several subfamilies within Nymphalidae, some of which feed on fruits and some of whom have longer and shorter proboscides. What is perhaps mos t surprising is that, despite possible preferences, all species readily took both sugar ty pes, suggesting that both would serve as an acceptable reward for any of the species in the survey. Within the broader pattern, there were some trends worth investigating further. For example, all three Papilionidae species fed exh ibited a preference for sucrose. Indeed, this finding supports Erhardt’s (1991) study of Battus philenor where individuals preferred both sucrose and fructose to glucose, but preferred sucrose to fructose. Confirmation of sucrose preference would also tie i n to DeVries’ observations (1987) that

PAGE 8

the genus Parides feeds on flowers of the families Balsamaceae and R ubiaceae, which have fairly long tubes and therefore are probably s ucrose-rich. Three of the four genera of Pierid butterflies fed in this study also preferred sucrose to hexose, so it would be interesting to se e if further studies corroborate a preference for sucrose over hexose, since there cur rently are conflicting lines of evidence for Pierid sugar preferences. Romeis and Wackers’ 2 000 study of Pieris brassicae revealed a preference for sucrose over fructose, bu t interestingly enough, Romeis and Wackers (2002) also found that Pieris brassicae fecundity is compromised by sucrose, and glucose is the only sugar with a positive effec t on both longevity and fecundity. Furthermore, according to DeVries (1987), Pierids f eed on red flowers in general, and as many red flowers are hummingbird pollinated, one wo uld expect them to be sucrose-rich. Pierids also feed on shallow-tubed Lantana camara however, which according to Percival (1961) points toward a hexose-rich nectar. If hexoses are good for Pierids, we might expect to see a hexose preference, but there is a possibility that the Pieris brassicae preference for sucrose can be expanded to the whole family. Again, further study is necessary to determine if there is a family trend f or sucrose preference. The Nymphalidae did not show a strong preference f or either sucrose or hexose, but even if future studies corroborate this lack of feeding preference pattern, it would not be entirely unexpected, since certain subfamilies o f Nymphalids have become specialized on different diets. For example, adults of the subf amily Morphinae do not visit flowers, and instead feed exclusively on plant sap and the j uices of rotted fruits and fungi (DeVries, 1987); meanwhile, Heliconiines and Ithomi ines eat nectar, often from Lantana camara There might then be patterns within smaller clade s, but the family as a whole would not have a single evolved preference. Ithomiines have specialized on members of the fami lies Boraginaceae and Asteraceae, whose nectars are predominantly hexoserich (Baker and Baker, 1983). This corresponds with Percival’s (1961) observations tha t shallow flowers, usually found in inflorescences, tend to be hexose dominant. However those observations do not correspond with this study’s findings, wherein Greta nero ate more sucrose and Ithomia heraldica had no discernable preference. Clearly more feedin g experiments need to be carried out. Morphinae species are specialized fruit eaters, so one would expect them to prefer fructose and glucose, as these sugars are the most common sugars in fruit (Omura and Honda, 2003); however, this is not what was found. Two of the four Morphinae species studied ( M. peliedes and C. eurilochus ) preferred the hexose solution, while M. granadensis on average ate more sucrose solution and C. memnon showed no discrimination between sugar solutions at all. Look ing again at Omura and Honda’s 2003 study, where they found that sucrose was the prefer red sugar but that fructose was also a very good feeding stimulant and occurred in higher quantities, perhaps concentration plays a role in sugar feeding preferences. Part of the problem could also be that Morphos have a tendency to eat the entirety of the volume o f solution they are presented with, which made it hard to find a pattern in terms of pr eference. One way to possibly overcome this would be to feed them until they are fully satiated, which would require one to put down far more than 100 m L at a time. Future experiments on Morphinae sugar preference might consider feeding them solutions of fructose, a hexose mixture, and sucrose, each with the necessary ethanol and acetic acid compounds to entice feeding.

PAGE 9

Two of the three Nymphalinae species preferred suc rose, and five Heliconiinae species preferred sucrose as opposed to three helic oniines that preferred hexose. The genus Heliconius eat pollen from Gurania costaricense (Gilbert, 1983), but nothing is mentioned about the sugars they get from nectar, an d any conjecture is made more complicated by the fact that hummingbirds may also visit G. costaricense I was unable to find any literature on the feeding habits of adu lt Nymphalinae species that would point to either a sucrose or hexose preference Again, mor e studies should be carried out to find out if Heliconiines and other Nymphalid butterflies exhibit preferences for specific sugar types. Overall, data presented here cannot conclusively s ay that butterflies as a whole appear to favor one sugar type over the other even though there were more positive (i.e., sucrose preference) results from the Sign Test than negative results. A Omura and Honda (2003) point out, much of the existing literature c oncerning butterfly feeding habits points to a trend of sucrose preference over fructo se, and fructose preference over the remaining sugars, but there are not enough studies of multiple butterfly families for one to really get a sense of a pattern. Successful attraction of pollinators is complex, r elying on many factors other than nectar sugar type. Pollinators will ignore one flow er if other flowers with more desirable nectar are present (Vansell et al., 1942), and will feed on less desirable flowers of attractive competitors are not present (von Frisch, 1950). It could very well be that butterflies will eat sugars from whatever flowers a re available to them, and their sugar “preference” will change as the composition of flow er species in their environment changes. If butterflies also have the ability to di gest multiple types of sugars as Watt et al. found (1974), sugar preferences would be further co nfounded. In this case, future studies would end up with similarly inconclusive results in terms of sucrose vs. hexose preference, and no pattern would be found relating butterfly phylogeny to adult dietary preference. In any event, butters flies do not seem constrained to one or few sugar types as was found for some frugivorous and nectarivorous birds (Martinez del Rio et al., 1992). The ecological implications of these findings for nectar flowers are rather curious. Despite the strong patterns shown by Baker and Bake r (1983) with regards to nectar composition, there seems to be no cause for specifi ed sugar compositions in butterflypollinated flowers. Sugar preference works really w ell for hummingbirds, as they really strongly prefer sucrose and the hummingbird flowers are all sucrose, but if butterflies don’t really have a sugar preference, why would but terfly-pollinated flowers be constrained to sugar types within families and poll ination syndromes? Why should longcorolla flowers have sucrose-rich nectar and short corolla flowers have hexose-rich nectar? One explanation could be that the long-coro lla flowers being pollinated by butterflies are also pollinated by hummingbirds, ne cessitating the production of a sucrose-rich nectar. Another similar explanation ha s to do with bee sugar preference. Apparently long-tongued bees tend to be rewarded wi th sucrose rich nectar, while shorttongued bees rarely are rewarded with sucrose (Bake r and Baker, 1983). It could well be that certain flowers contain either sucrose or hexo se in order to reward bee pollinators, and butterflies have jumped on the existing pollina tion-syndrome wagon. However, this

PAGE 10

hypothesis is not without problems: many of the lon g-tongued bee flowers come from a few set families characterized by sucrose-rich nect ar while the short-tongued bee-flowers are from hexose-rich families like Asteraceae, so i t is hard to say if the flowers are producing specific nectars to conform to bee prefer ence, or the other way around. Finally, one might want to explore the costs of producing su crose as opposed to hexose nectars, as this might give insight into which pollinators are dictating the composition of sugar nectars. One major limitation of the study was that all of the butterflies had to be caught individually, which was time consuming and gave an uneven distribution of representatives from certain families (the Heliconi ines are much more numerous than the Papilionids). The caught individuals were also much more energetic and resistant to being force fed, so I had to institute a waiting pe riod before feeding in order to let them calm down and get hungry. The Parides and Pierid individuals in particular were difficult to feed since they could only be containe d for two to four hours without food, which was not enough time for them to become hungry and less agitated. A way to improve this study would be to acquire at least twenty individuals of each species as pupae, feed them one of the solutio ns after they hatched, contain them for a day or two, and then feed them the other solution This would greatly increase the evenness of representation between species. Such ex periments would be incredibly valuable, since to my knowledge, no other studies h ave been done looking at adult feeding preferences across butterfly families, rela ting dietary preference to phylogeny or ecology. Further studies could be undertaken to loo k at the biology of butterfly digestion that may also shed light on evolved dietary prefere nces. ACKNOWLEDGEMENTS I would like to thank the staff at the Monteverde B utterfly Garden and the garden at Selvatura for allowing me to do my experiment in th eir gardens, use their butterflies, and for helping me identify individuals. A very special thanks goes to Alan Masters for guiding me towards the idea of this project after a ll the others failed. I can’t thank you enough. Thanks also to Taegan and Pablo for running to the supply closet for me time after time, and Pablo especially for help translati ng the abstract. Literature Cited BAKER, H. G. AND I. BAKER. 1983. Floral nectar sugar constituents in relatio n to pollinator type. In C. Eugene Jones and R. John Little (Eds.). Handboo k of Experimental Pollination Biology. pp. 118-139. Van Nostrand Reinhold Publishing, Ontario, Canada. DEVRIES, P. J. 1987. The Butterflies of Costa Rica and their Natural His tory Princeton University Press, Princeton, New Jersey. ERHARDT, A. 1991. Nectar sugar and amino acid preferences of Battus philenor (Lepidoptera, Papilionidae). Ecological Entomology 16(4): 425-434. VON FRISCH, K. 1950. Bees— Their Vision, Chemical Senses and Language Cornell University Press, Ithaca, New York. In Baker, H.G. and I. Baker. 1983. Floral nectar sugar constituents in relation to pollinato r type. In C. Eugene Jones and R.

PAGE 11

John Little (Eds.). Handbook of Experimental Polli nation Biology. pp. 118-139. Van Nostrand Reinhold Publishing, Ontario, Canada. JANZ, N. AND S. NYLIN. 1998. Butterflies and plants: a phylogenetic stud y. Evolution 52(2): 486-502. MARTINEZ DEL RIO, C., H. G. BAKER AND I. BAKER. 1992. Ecological and evolutionary implications of digestive processes: Bird preferen ces and the sugar constituents of floral nectar and fruit pulp. Cellular and Molecul ar Life Sciences. 48(6): 544-551. PERCIVAL, M. S. 1961. Types of nectar in angiosperms. New P hytology. 60: 235-281. In Baker, H.G. and I. Baker. 1983. Floral nectar suga r constituents in relation to pollinator type. In C. Eugene Jones and R. John Little (Eds.). Handboo k of Experimental Pollination Biology. pp. 118-139. Van Nostrand Reinhold Publishing, Ontario, Canada. OMURA, H. AND K. HONDA. 2003. Feeding responses of adult butteries, Nymphalis xanthomelas Kaniska canace and Vanessa indica to components in tree sap and rotting fruits: synergistic effects of ethanol and acetic acid on sugar responsiveness. Journal of Insect Physiology. 49: 1031-1038. ROMEIS, J. AND F. L. WACKERS. 2000. Feeding responses by female Pieris brassicae butterflies to carbohydrates and amino acids. Phys iological Entomology. 25(3): 247-253. -----, AND ----. 2002. Nurtitional suitability of individual carbohydrates and amino acids for adult Pieris brassicae Physiological Entomology. 27(2): 148-156. VANCELL, G. H., W. G. WATKINS and R. K. BISHOP. 1942. Orange nectar and pollen in relation to bee activity. Journal of Econ. Entomol ogy. 35: 321-323. In Baker, H.G. and I. Baker. 1983. Floral nectar sugar const ituents in relation to pollinator type. In C. Eugene Jones and R. John Little (Eds.). Handboo k of Experimental Pollination Biology. pp. 118-139. Van Nostrand Rei nhold Publishing, Ontario, Canada. AMBROSE III, H. W., P. AMBROSE, D. J. EMLEN AND K. L. BRIGHT (Eds.). 2002. A Handbook of Biological Investigation 6th Ed., pp. 78-80. Hunter Textbooks, Inc. Winston-Salem, North Carolina. GILBERT, L. E. Anguria and Gurania (Rain-forest cucumber) 1983. In Janzen, D. (Ed.). Costa Rican Natural History, pp. 190-191. WATT, W. B., P. C. HOTCH, AND S. G. MILLS. 1974. Nectar resource use by Colias butterflies. Oecologia. 14: 353-374. In Baker, H.G. and I. Baker. 1983. Floral nectar sugar constituents in relation to pollinato r type. In C. Eugene Jones and R. John Little (Eds.). Handbook of Experimental Polli nation Biology. pp. 118-139. Van Nostrand Reinhold Publishing, Ontario, Canada.

PAGE 12

APPENDIX APPENDIX 1. Phylogeny of Lepidoptera from Janz and Nylin (1998).

PAGE 13

APPENDIX 2. Sign Test Results Subfamily Species Sucrose mean volume ( m L) Hexose mean volume ( m L) Sign of S-H Ithomiinae Greta oto 30.68 26.79 + Ithomia heraldica 36.05 36.8 0 Morphinae Morpho granadensis 58.29 62.54 — Morpho peleides 83.36 78.47 + Caligo eurilochus sulanus 33.4 400 — Caligo memnon memnon 346.3 344.2 0 Danainae Danaus plexippus 89.7 76.6 + Heliconiinae Dione moneta poeyii 55.09 44.74 + Dryadula phasetusa 30.22 49.5 — Dryas iulia 40.79 58.3 — Eueides isabella 31.91 34.79 — Heliconius charitonius 29.12 31.93 — Heliconius hecale zuleika 70.13 65.88 + Heliconius sapho leuces 23.51 26.12 — Heliconius sara fulgidus 23.29 24.08 0 Heliconius erato 35.79 26.45 + Nymphalinae Catonephele numilia esite 56.23 31.05 + Myscelia cyaniris cyaniris 30.26 36.4 — Siproeta steleres 103.55 59.6 + Papilioninae Battus polydamas polydamas 49.21 24.5 + Parides lycimenes lycimenes 77.43 59.15 + Papilio astylaus pallas 89.47 81.3 + Pierinae Ascia limona 43.19 59.2 — Appias drusilla 42.1 18.42 + Coliadinae Phoebis sennae 46.32 23.86 + Phoebis philea philea 59.5 55.5 + Total positive: 14 Total negative: 9 Family Pieridae Positive: 3 Negative: 1 Family Papilionidae Positive: 3 Negative: 0 Family Nymphalidae Positive: 8 Negative: 8

PAGE 14

APPENDIX 3. Sucrose Feeding Data nrrn r!"#n$ %r!"! &&n %r!"! ''n() *"r +r &n(' )n$ *"-.(n#' *!")n# / &#n$ *"rn(' +r $n# *r0n# *r &n&) +r )n( *"-.(n)( *!"n( *"rn&# *r0$'n( *r $n& / n( / $)n / )$n) *!"$n *!"$&n) ""r )n& %r!"! n$ +r n$ %r!"! $ %r!"! ''n() %r!"0r''n() r!")n$ %r!"0r(n( !!#'n( "/r0 !&&n&) r#n'& 1!''n( 1!n& !&n(' %r!"0r)n'& r!"#n'& r!"$n& %r!"0r $& r!"n) r!"#n ,!!r n$

PAGE 15

+r (n$ r #n'& r'n()#$ 1!(n()#$$ !2!!#'n()#$$ r(n#'()# !"'n)$&(#' 0r"))n$&) !&#n'()# !!2)n)$&(#'( %rr$n)$&(#'( %rr)n$&)$ "/#n)$&(#' !"$n&)$&# 0)n)$&(#'& r)n)$&(#'& %rr$(n)#$& "/$n)#$& !rr$)$n&(#'( !")(n)#$& !rr(&n&)$&(' 1!'n$&) "/!"!"&'n()#$ APPENDIX 4. Glucose-Fructose Feeding Data n*2*"rn# / n) *"r)(n) / )n#' *r )&n&) *"-. +r $(n$ *!"$n# r!")n) +r $n& +r )n$ *"-.n( *!"))n# +r n&) +r $n$ *r0#n'& *!"$&n(#'()# !!#$n)$&(#'( %r!"0r()n'()# "/n#$& rn&)$&(' *"r)$n)$&(#'( %r!"0rn()#$$

PAGE 16

r$n)$&(#'( r&)n'()# *r0$'n$&) *!")#n$&(#'( /n()#$$ /)n#$&) *"-.'n#'()# *r$(n)#$& %r!"!n)$&(#'( %r!"! $ %r!"0r#n'()# r!"'n()#$ / ) %r!"! $ %r!"! $ %r!"!#'n()#$$ ,r'n)$&(#'& 1!n)$&(#'( 1!'#n'()# ,!!r$&n(#'()# ,!!r$n&)$&# !))n$&(#'( !&n)$&(#'( &'n$&) !")n&)$&(' rn()#$ 0r" !"$((n)$&(#' 1!#(n)#$& %rr'n()#$ %rr&n&)$&(' !" r&'n()#$$ 0)'$n&(#'( "/&n)$&(#' rr$n$&)$ %rr)n$&) "/$'n()#$ r!"!"#n$&) !rr#(n)$&(#'& !2!!(n&(#'()( rr"$$n)$&(#'( !"n)$&(#'& 0'n#$&) 1!)n'()#$ !rr)$n&(#'()( "/!"!"&&n&)$&(' ""r)n#$&