J. E. Erickson.
The Comparative Utilization of Cultivated and Weedy Umbellifer Species by Larvae of the Black Swallowtail Butterfly, Papilio polyxenes.
Psyche 82:109-130, 1975.

Full text (searchable PDF, 1632K)
Durable link: http://psyche.entclub.org/82/82-109.html

The following unprocessed text is extracted from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.

THE COMPARATIVE UTILIZATION OF
CULTIVATED AND WEEDY UMBELLIFER SPECIES
BY LARVAE OF THE BLACK
SWALLOWTAIL BUTTERFLY, PAPILIO POLYXENES BY JAMES M. ERICKSON~
Dept. of Entomology, Cornell University, Ithaca, New York 14850 For many years, much of the emphasis in agriculture and plant breeding has been laced on increasing overall ~roduction (Allard 1960). This is because man is almost totally dependent on plants 'for his food. The things he eats come directly from plants or in- directly from herbivorous animals. Plants are also the major source, directly or indirectly, of most clothing, fuel, drugs, and construction materials.
The impact of insects on ~lants cannot be overemphasized. For example, some insects can be very successful in the biological control of weeds ( Holloway 1964). insects also have a great impact on the evolution and ecology of plants through their destruction of seeds, young seedlings, or the plants themselves (Breedlove and Ehrlich 1968, Janzen 1969, 1971). The relationships between the insect and the plants that we observe today are based upon millions of years of co-evolution. During the course of this evolution, plants have developed various mechanisms to resist insect attack. The majority of plant defenses can be classified as physical or chemical defenses (Stahl 1888). Plant physical defenses may include thickened cuticle (Tanton 1962, Feeny 1970) or hairs, spines, and thorns on the epi- dermis (Johnson 1953, Pearson 1958, Bernays and Chapman 1970), which interfere with the insects feeding. The high silica content of some plants or the crystalline materials in the leaves of many conifer species add a further physical barrier to insect attack (Merz 1959, Pathak 1969).
Plants have also evolved a great array of chemical defenses, the so-called secondary substances. These include the alkaloids, glyco- sides, tannins, flavenoids, terpenoids, essential oils, and saponins, to name a few (Fraenkel 1969, Whittaker and Feeny 1971). Stahl (1888) advanced the idea that these compounds evolved in plants as Tresent address: Dept. of Biological Science, California State Univer- sity, Hayward, California 94542.
Manuscript received by the editor May 20, 1975.



================================================================================

I 10 Psyche [March
a defensive mechanism against insect and vertebrate herbivores, pathogens, and perhaps competitors. This view has been supported, as far as insects are concerned, by Dethier (1954, 1970)) Thorstein- son ( I 960)) Ehrlich and Raven ( I 965 ) , Jermy ( I 966)) Whittaker and Feeny ( 1971 ) , and Erickson and Feeny ( 1973). Much of the interaction between insects and these secondary substances is sensory in nature, and such substances may inhibit or deter feeding (Thor- steinson 1960, Gill and Lewis 197 I ), or may prove toxic to non- adapted larvae (Taylor 1959). Some plant species even synthesize and accumulate sterols which mimic insect molting secretions (Wil- liams 1970). These phyto-ecdysones have been found in numerous fern species and prevent molting in insects feeding on such species (Whittaker and Feeny 197 I ) .
These secondary chemicals are not the only chemical means that plants have evolved to protect themselves from insect attack. Plant proteins are often deficient in some individual amino acids necessary for insect growth and development (Lord 1968, Boyd 1970). The nutritional quality or adequacy of the food plant is of utmost im- portance to a phytophagous insect (Friend 1958, Legay 1958). Aphids (Auclair et al. 1957)) beetles (Allen and Selman 1955, 1957)) butterflies and moths (Hovanitz and Chang 1962, Feeny 1970), grasshoppers (House 1959, Dadd 1961)) flies (Chapman 1969)) and other insects (Gilmour 1961, Levinson 1962)) have all been shown to exhibit quite variable 'feeding responses which in turn influence larval development, mortality, fecundity and fertility, when reared on different host plant species or on the same host plants grown under differing conditions or ages. Gordon (1959) has suggested that nutrient deficiencies in plants ". . . may 'be a result of natural selection of inedibility." The interaction between the nutritive adequacy of leaves and secondary chemistry has been dem- onstrated for oak trees and oak leaf tannins (Feeny 1968, 1969, 1970).
Agriculturists and plant breeders have become aware of the prob- lems that increased yield and palatability of crop species present in terms of the plant's inherent defensive mechanisms ( Snelling I 94 I, Painter 1951, Allard 1960, Briggs and Knowles 1967). The oldest record of inherent plant resistance to insect herbivores was by Havens (1792) in which he recognized the Hession fly resistance of the Underhill variety of wheat. However, scarcely 200 papers have dealt with this subject in the 148 year period from 1792 to I 940
( Snelling I 941 ) . Generally, as plants species were domesti- cated and cultivated, various morphological and physiological changes



================================================================================

19751 Erickson - Papilio polyxenes III
occurred (Polunin 1960). Cultivated plant species are often larger in size, have larger seeds or seed sets, have more rapid and uniform seed germination, lose defensive structures such as hairs, thorns, or spines, and display improved palatability and nutritive content com- pared to their corresponding wild relatives. The domestication of plants involves more than simply modifying the genetics of a species, because reciprocal adaptations between the domesticated (cultivated) species and the domesticator (man) are required. The domestication and cultivation of plants is in sorts a two way street; it may bring about ecological, social, and/or genetic changes in man as man has brought these changes to the plants. Selection for increased yield or increased productivity of the edible part of a crop species does not necessarily mean an increase in pri- mary production. Above a certain point, increased yield must come at the sacrifice of some other adaptive use of energy (Cody 1966). As an example, generally increasing the yield of wheat decreases the amount of straw which is a fundamental part of the plant's self- productive maintenance equipment ( Odum I 97 I ). Is it then possible, when breeding for increased palatability, yield, or nutritive content in cultivated plant species, to alter or decrease the inherent defensive mechanisms of the plants involved, be they physical, chemical, or both? The purpose of this study is to examine this question in detail, utilizing the oligophagous butterfly, Papilio polyxenes, whose larvae feed on a variety of cultivated and wild or weedy species of the carrot family, the Umbelliferae ('Chittenden 1909, Forbes 1960, and others). Many umbellifer species upon which these larvae feed, have been cultivated for centuries, primarily for spices and condiments for prepared 'foods (Buttery et al. 1968, French 1971, Kasting et al. I 972), and for medicinal or toxic drugs (Muenscher I 95 I, Kings- bury 1964). Umbellifer plants are a major source of various vita- mins (especially vitamin A) and minerals essential for proper growth in man (Lewis and Rubenstein 1971) as well as insects (Fraenkel 1953, Dadd 1957). Through studies of larval growth efficiency and food plant utilization, comparisons as to the relative adequacy of each host plant species, cultivated versus wild, can be determined. Eggs of P. polyxenes were taken from the second generation of a culture 'founded from wild insects taken near Ithaca, New York and reared in the laboratory on carrot (Daucus caroia). A minimum of 15 and a maximum of 20 eggs were placed on each of 32 umbellifer



================================================================================

112 Psyche [March
species: 10 species of cultivated umbellifer and 22 species of wild or weedy umbellifers found in the Central New- York area. The larvae were maintained on these species throughout larval development- Mature and uninjured leaves of the wild species were gathered in the field each day and the leaves of the cultivated umbellifers were col- lected from greenhouse reared plants. All leaves were sealed in plastic bags and offered to the larvae within 2 hours. Leaves were replaced and feces collected every 24 hours to prevent bacterial or fungal development. The larvae were reared in clear plastic boxes (9 X
12 X 4 inches) (Tri-State Plastic Molding Co., Henderson, Ky.), in a climatically controlled chamber with the following para- meters: temperature 22O day, 18O night, approximately 55% hu- midity, and a I 6-8 LD photoperiod.
For the purposes of examining larval energy and nutrient utiliza- tion, a minimum of 8 and a maximum of 10 newly molted 4th instar larvae were placed individually in glass petri dishes (Pyrex, ioomm X
15mm) lined on the bottom with a piece of Whatman No. I filter paper. The ideal utilization study would, of course, encompass the entire life cycle as the efficiency of food utilization by the early instars is certainly of interest. The nutritional adequacy of the food plant material can be judged only by its ability to support growth in successive instars. There were, however, 2 reasons for utilizing only 4th instal- larvae in this experiment. Since larvae of P. polyx- enes consume approximately m % of the total food ingested during larval development during the 1st instar, these minute quantities of food ingested and digested lead to exceptionally high error values and are therefore inaccurate. Similar results were obtained 'for the 2nd and 3rd instars where the percentage of the total food consumed during larval development was 0.6% (L2) and 2.8% (L3). The ultimate instar was not included for the purposes of the energy utilization experiments due to a pre-pupal clearing of the gut in which a larva may lose up to 40% of his maximum wet weight within a 5 minute period. Once this occurs, there is no way to estimate the maximum larval weight which is necessary for various calculations to determine larval food utilization efficiencies. During
the 4th instar, approximately 10 to 15% of the total food ingested during larval development is consumed. These larger amounts lead to more accurate weights which significantly reduce the statistical error.
All the individual larvae were placed in the same controlled tem- perature room, except for the period of time each day during which new food was offered to the larvae and the feces collected. The



================================================================================

19751 Erickson - Papilio polyxenes 113
experimental larvae were fed the same leaves as the maintained cul- tures. These randomly collected leaves were split along the midrib, one half weighed and offered to the larvae and the other half weighed and used to determine the percent dry matter in the leaf material ( Waldbauer I 960, I 964).
Besides the percent dry matter in the leaf material, the calorific and nitrogen content of the leaf material, the larvae, and the feces were determined. Calorific values of the larval 'food plants, feces, and larvae were determined by means of a Phillipson non-adiabatic microbomb calorimeter (Gentry and Wiegert Inst. Inc., Aiken, S.C. ) ( Phillipson I 964). The lyophilized leaf material, feces, and larvae were subjected to 3 replications for the determination of calorific values. The organic nitrogen content of the leaf material, the feces and the larvae, were determined either by the Kjeldahl method for total nitrogen (Williams 1964) or the microKjeldah1 method (Mc- Kenzie and Wallace 1954). A minimum of 3 replicate samples was obtained for the larvae and the feces, as well as each host plant species.
The dry weight of food ingested by the larvae was estimated following the techniques of Waldbauer (1960, 1964), and Wald- bauer and Fraenkel ( 1961 ) except that plant material was lyophilized instead of oven-dried. The dry weight of the food utilized ar assimi- lated was assumed to be the dry weight of the food ingested minus the dry weight of feces. An additional group of larvae were reared along with the experimental larvae, and these were sacrificed to determine the dry weights, and thus, the percentage of dry matter of the larvae. Indices of food utilization were determined following the methods of Waldbauer ( I 960, 1 964, I 968). Many terms have been used both by ecologists and by physiologists to describe various measures and indices of food utilization and efficiency. Relationships between many of these terms are discussed by Kozlovsky (1968) and Waldbauer ( I 968 ) .
As an index of digestibility, the ratio of the amount of food as- similated to the amount of food ingested, referred to as the 'Assimi- lation Efficiency' (Clark 1946, Odum 1957, Odum 1971 ), or the 'Coefficient of Digestibility' (Waldbauer 1964, 1968, House I 965), was used. In practice, this measure is only an approximation since the numerator (as determined by the usual gravimetric procedure) does not quite represent the amount of food actually assimilated (Waldbauer 1968). This slight error is due to the presence of metabolic wastes in addition to the undigested food in the feces (Lafon 195 I ). For this reason Waldbauer ( 1968) has suggested



================================================================================

114 Psyche
[March
'Approximate Digestibility' as a less ambiguous term to describe this measure. However, Hiratsuka (1920) and Waldbauer (1964) point out that the uric acid content of the phytophagous insects is relatively low and that the difference between true and measured assimilation efficiencies is negligible.
The efficiency with which ingested food is converted to biomass is calculated by dividing the dry weight of food ingested into the dry weight gained by the larva during the instar. This index, referred to by the physiologists as the 'Efficiency of Conversion of Ingested Matter' (Waldbauer 1968) and by ecologists as the 'Ecological Growth Efficiency' (Gerking I 962, Odum I 97 I ) , is an overall measure of an animal's ability to utilize for growth the food in- gested.
The efficiency with which digested food is converted to biomass is calculated by dividing the dry weight of food assimilated into the dry weight gained by the larva during the instar. This index, re-
ferred to by Waldbauer (1968) as the 'Efficiency of conversion of Digested Matter' and by Gerking ( 1962) and Odum ( 1971 ) as the 'Tissue Growth Efficiency', decreases as the proportion of digested food metabolized for energy and maintenance of physiological func- tions increases ( Waldbauer I 968).
In his classic work on accessary growth factors, Hopkins ( 1912) pointed out that absolute quantities cannot be used to compare the intake of food by animals growing at different rates. Valid com- parisons could only be made on the basis of the rate of intake relative to the mean weight of the animal during the feeding period. Wald- bauer ( 1964, 1968) working on this basis proposed the 'Consump- tion Index' calculated in this experiment as : dry weight of food ingested
duration of
x
mean dry weight of the
feeding period
animal during feeding period
The mean weight of the animal is most accurately calculated from the area under its growth curve as determined by integration. A weighted average of daily weights will give an almost identical value if the growth curve is smooth (Waldbauer 1964). Such three vari- able equations are at times difficult to discuss, being included in the present work only for later comparison with other insect species (see Waldbauer 1968). From the equation, if 2 individual larvae ingest the same total amount of food, with the larva on the less nutritiously adequate food plant taking a great deal of time and gaining little biomass and the other larva on a more acceptable food plant gaining



================================================================================

19751 Erickson - Papilio polyxenes 1.1 5 a great deal of biomass in a short time, the respective consumption indices may in fact be identical.
A more useful index for the growth rates of an individual larva is the mean dry weight added to larval biomass per day. This
measure gives the investigator a much more accurate estimation as to the growth potential of the various umbellifer food plants. For the purposes of food utilization and efficiency determinations, the experiment was concluded when the larvae molted into the ulti- mate instar. The larvae were then reared through to the adult stage on the same experimental plants that they fed upon before and dur- ing the utilization experiments.
All resulting adult females were
then utilized in various host plant selection experiments. The data are generally presented as a mean and standard error for the larvae in any particular host plant treatment group. The various experimental parameters were subjected to one way analysis of vari- ance (Guenther 1965, Snedecor and Cochran 1967) to determine differences among the various treatment groups. A typical T-test for 2 independent samples of unequal sizes utilizing a pooled variance (Guenther 1965) was used for analysis of the differences between cultivated and wild umbellifer species. Linear regression analyses were performed and the significance of the correlation coefficients was tested using a table of critical 'r' values (Snedecor and Cochran 1967). All statistical procedures were completed with the aid of a programable calculator (Olivetti programa 101 or microcomputer Various plant parameters differed greatly between cultivated (do- mesticated) and wild (weedy) species of Umbelliferae offered to the swallowtail larvae (Table I).
The dry matter content of the leaf
material was significantly lower (P < 0.01 ) at a mean of approx- imately 12.25% for the cultivated umbellifers than the wild umbel- lifer species which had a mean of about 21% dry matter. In terms of caloric content, no significant difference (P < .3) was found between the cultivated species, at a mean of about 4.11 cal/mg dry weight, and the wildly occurring species, at a mean of about 4.13 cal/mg dry weight. The nitrogen content of the leaf material was significantly higher ( P < 0.05 ) in the cultivated species, averaging approximately I % higher than the wild species in terms of total nitrogen. This value becomes significant when converted to protein content, as I % nitrogen equals approximately 6.25 % protein (Lord 1968).




================================================================================

Psyche
[March
Table I. Dry weight, calorific values and nitrogen content of various species of Umbelliferae offered to the larvae of the eastern black swat lowtail, Papilio polyxenes, during the 4th instar. Plant species
Mean calories
Mean percent
per mi l I i gram Mean percent
dry material
dry weight of of nitrogen
in leaves
leaf material in leaf material
5 SE 1 v 2 2 ~ ~ ~ 9 3 2 S E ~ , ~
Cu I t i vated species
Anethum graveolens
Di l I
Ap i urn grave0 I ens
Celery
Carurn carvi
--
Caraway
Coriandrum sat i vum
Coriander
Daucus carota
--
Carrot
Foen icu 1 um vu l qare
Fennel
Liqusticurn scothicum
Scotch lovage
Past i naca sat i va
--
Parsnip
Petrosel i num cri spum
Parsley
Pimpinella anisurn
An i se
Wild species
Aegopodium podagraria
Goutweed
Aeqopod i urn var i qatum
Goutweed
Aethusa cynapium
Fools parsley




================================================================================

197 51 Erickson - Papilio polyxenes 117
Table I. Continued.
Mean calories
Mean percent
per milligram Mean percent
dry material
dry weight of of nitrogen
in leaves
leaf material in leaf material
Plant spec i es å SE'*~ 2 S E ~ , ~ å SE~,-* Angelica atropurpurea 25.22 5 0.41
Angel i ca
N = 25
Anqelica iriquinata 18.49 2 0.65
Angel ica
N = 36
Cicuta bulbifera
--
18.17 å 0.42
Bulb bearing water hemlock N = 17
C icuta macu lata
--
Water hemlock
Coelopleurum lucidum
Conium maculatum
--
Po i son hem l oc k
Cryptotaenia canadensis
Honewort
Daucus carota
--
Carrot
Heracleum maximum
--
Cow parsnip
Heracleum sphondylium
Cow parsnip
Imperator ia ostruth ium
Masterwort
Levisticum officinale
Lovage
Pastinaca sativa
--
Parsnip
Pseudotaenidia montana
Mountain pimpernel
S i um suave
--
Water parsnip




================================================================================

118
Table I. Continued.
Psyche
[March
Mean calories
Mean percent per mi I l igram
Mean percent
dry material dry weight of
of nitrogen
in leaves leaf material
in leaf material
Plant species å S E ~ , ~ sE1p3 å SEl,1* Taenid ia interqerrima 24.31 k 0.64 3.97 2 0.13 2.81 2 0.20 Ye1 low pimpernel
N = 30
Thaspium barbinode 30.79 k 0.58 3.93 k 0.23 2.22 2 0.19 Meadow parsnip N = 28
Zizia aptera
- 19.80 2 0.73 4.31 k 0.17 3.21 2 0.22
Heart shaped a lexanders
N = 27
Zizia aurea
-- 23.53k0.16 4.25k0.19 4.2320.11
Golden alexanders N= 18
he significance values of the IT1 statistic with 30 df are: 0.05 = 2.042, 0.01 = 2.750.
2~ = 5.666
'T = 0.422
4~ = 2.349
The larvae reared on the 32 umbellifer species all ingested ap- proximately the same total amount of [food, averaging about 144 dry weight mg, during the 4th instar (P < .3) (Table 2). Generally, all larvae completed the 4th instar in about 3 days (P < .3), thus the rate of food consumed also did not vary significantly (T = 0.247, P < .4). The proportion of ingested food which was digested and assimilated ('Approximate Digestibility') averaged approximately 53% for the larvae reared on the cultivated umbellifer species and about 47% for the larvae reared on the wild umbellifer species ( P <o.o5). The efficiency with which ingested food was converted to larval biomass ranged from approximately 25% 'for the larvae reared on the cultivated umbellifer species to approximately 19% for the larvae reared on the wild umbellifmer species (P < 0.05). The efficiency of conversion of digested food into larval biomass did not vary significantly between the two groups of plants tested