INFORMATION TO USERS This m atM iil was produced from a m icrofilm copy of the original document. While the moat advanced technological means to photograph and reproduce this document have been used, the quality is heavily dependant upon the quality of the original submitted. The following explanation of techniques is provided to help you understand markings or patterns which may appear on this reproduction. 1. The sign or "target" for pages apparently lacking from the document photographed is "Missing Paga(s)". If it was possible to obtain the missing page(s) or section, they are spliced into the film along w ith adjacent pages. This may have necessitated cutting thru an image and duplicating adjacent pages to insure you complete continuity. 2. Whan an image on the film is obliterated w ith a large round black m ark, it is an indication that the photographer suspected that the copy may have moved during exposure and thus causa a blurred image. You w ill find a good image o f the page in the adjacent frame. 3. Whan a map, drawing or chart, etc., was part o f the m aterial being photographed the photographer followed a definite method in "sectioning" the m aterial. It is customary to begin photoing at the upper le ft hand comer of a large dieet and to continue photoing from le ft to right in equal sections w ith a small overlap. If necessary, sectioning is continued again — beginning below the first row and continuing on until complete. 4. The m ajority of users indicate that the textual content is of greatest value, however, a somewhat higher quality reproduction could be made from "photographs" if essential to the understanding of the dissertation. Silver prints o f "photographs" may be ordered at additional charge by w riting the Order Departm ent, giving the catalog number, title , author and specific pages you wish reproduced. 5 . PLEASE NO TE: Some pages may have indistinct print. Filmed as received. University Microfilms International 300 North Zaab Road Ann Arbor, Michigan 48106 USA St John's Road. Tylar't Groan High Wycomba. Bucks. England HP10 8HR J I 77-11,717 SURGEONER, Gordon Allen, 1949THE LIFE HISTORY AND POPULATION DYNAMICS OF THE VARIABLE OAKLEAF CATERPILLAR, HETEROCAMPA MANTEO (DBLDY.), IN MICHIGAN:-------Michigan State University, Ph.D., 1976 Entomology Xerox University Microfilms , Ann Arbor, Michigan 49106 THE LIFE HISTORY AND POPULATION DYNAMICS OF THE VARIABLE OAKLEAF CATERPILLAR, HETEROCAMPA MANTBO (DHLDY.), IN MICHIGAN By Gordon Allen Sur^eoner A DISSERTATION Submitted to Mlchlran State University In partial fulfillment of the requirements for the decree of DOCTOR OF PHILOSOPHY Department of Entoraolopry 1976 Abstract Life History and Population Dynamics of the Variable Oakleaf Caterpillar, Heterocampa manteo (Dbldy.), in Mlchifian by Gordon Allen Surreoner The life history and population dynamics of the variable oakleaf caterpillar Heterocampa manteo (Doubleday) were in­ vestigated in mixed oak forests of the Manistee National For­ est of west central lower Michigan. Samplinp techniques were designed to estimate prepupal densities in five permanent re­ search plots. Overwinterinn* populations of H. manteo showed little mortality, but each season ca. 50% of the prepupal population remained in a prolonged diapause lasting two years or longer. Development times for prepupae and pupae were determined and adult flight activity monitored over two years. The fe­ cundity potential per female was ca. ^00 e^vs, but parasitism levels by the en*r parasltoids Telenomus sp. and Trlchonramma sp. accounted for ca. 8 5 % mortality in 1 9 7 ^ and 33^ mortality in 1975* Tbe larval instars were identified and the develop­ mental times and foliave consumption of each instar determined. Each larva consumed ca. 33^ cm of oak foliage, but larvae parasitized by Dlradops bethunel (Cresson) consumed ca. 60% less foliage. A method to predict defoliation based on rela­ tive density of early instars was devised. Seven parasitold species were found to attack larvae of H. manteo. Total larval parasitism In 197^ was ca. declined to 1 6% in 1975* but This decline was attributed in part to the poor synchronization of many of the parasitolds to the population of H. manteo which remained 2 years or longer in the prepupal stage. Predators of H. manteo were determined and a larval defense secretion containing- formic acid noted. A laboratory rearing technique using oak foliage as a nutrient source was developed for H. manteo. The effect of temperature, photoperiod, density, and soil moisture on dia­ pause, larval coloration, developmental times, and mortality was also investigated. Acknowledgements This study was made possible by the trust and fore­ sight of the taxpayers of the United States and Canada. It is the millions of ordinary citizens* who through their tax dollars have Invested in education* that merit my gratitude. I would hope that in some manner this dissertation and fu­ ture endeavors will repay that trust. Every graduate student is indebted to his major pro­ fessor for long hours of guidance* professional support and a genuine concern for his well being. Dr. William E. Wallner manifested all these qualities and more. He was not simply a major professor* but more importantly, a personal friend. To my committee members* Drs. Fred Stehr, Roland Fischer* John Hart and Wayne Myers, I am grateful for assis­ tance in organizing a meaningful graduate program* hours of research discussions* technical expertise, and editorial as­ sistance with this dissertation. I am Indebted to all mem­ bers of the Entomology Department* faculty, fellow graduate students* clerical and technical staff. Dr. James Bath has created a department providing graduate students with the finances* facilities* and above all the opportunities to expand one's potential to Its limit, both professionally and personally. Finally, I wish to express my appreciation to my wife, Shirley, who has been at ray side throughout this re­ warding period. Table of Contents List of Tables . . . . . . Page ............................. v ............ List of Figures vli Introduction ................................. . . . . . 1 Research A r e a .................. 4 Method for Estimation of Prepupal Population . . . . . . 8 Prepupal Densities and Distribution .... .......... 11 Soli Samples as a Detection S u r v e y ....................... 19 Coloration of Larvae and P r e p u p a e ....................... 22 Defensive Secretions . . . ............................. 28 Cold Temperature Mortality Studies .................... 31 Diapause and Prepupal Development 31* .................. Pupal Sexln.g and Sex R a t i o s ............................. 54 Developmental Times of P u p a e .......... 55 Flight A c t i v i t y ...................... 58 Fecundity Potential 60 . . . . . . . . . . . . . . . . . . Egg Laying H a b i t s ...................... 64 Development of E m b r y o s ............ 65 Egg Parasitism.............. 66 Larval Instars ......................................... 73 ................................. 77 Larval Feeding Habits laboratory R e a r i n g .................................... 79 Parasltoids of Larvae 82 ................................. Multiple P a r a s i t i s m ...................... * ............ 9 0 Hyperparasltlsm ....................................... Parasitism as Related to Prolonged Diapause ill . . . . . . 90 91 Plage ft^edation ................... Sampling of Larval Populations ...................... Foliage Consumption............ Predicting Defoliation 93 . 9^ 96 .............................. 101 Control Philosophy and Insecticide Trials ............ 105 Suggestions for Further Research 109 .................... Literature Cited 113 Appendix - Determination of Larval Instars of Heterocampa manteo and Reduction of Larval Head Capsule Size by the Parasltoid Diradops bethunel .............. 119 Iv List of Tables Table 1. Parse Prepupal densities of H. manteo■ Newaygo» Michigan, 1973-75 1^ 2. Prepupal densities of H. manteo. Big Star lake, Michigan, 1 9 7 3 - 7 ? .................... .. . 1*+ 3. Prepupal densities of H. manteo. Branch Pole, Branch, Michigan, 1973-76 . . . . . . 1 5 5* Prepupal densities of H. manteo. Branch Mature, Branch, Michigan, 1973-76 15 Prepupal densities of H. manteo. Dublin, Michigan, 1973-75 . 16 6. Generation Index of H. manteo prepupal popu­ lations, Michigan, 1973-75 • ................... 18 7. Prepupal color types of H, manteo as related to density, Michigan, October 1973 .............. 25 Seasonal color patterns of H. manteo prepupae, Michigan, 1973-7^ . . . . 7 . . . . . . ........ 25 8. 9. Percent mortality of I£. manteo prepupae exposed to cold temperatures for a 2^-hour time period . . 33 10. Developmental times of H. manteo prepupae .... 37 11. ftrepupal densities and carryover population of H. manteo. Michigan, 197^-75 . . . . . . . . . ^5 12. Pupation of encapsulated prepupae of H. manteo examined August 28, 1975 ......................... **8 13* Number of days to emergence of H. manteo pupae maintained at various temperatures .............. 57 1^. Number of eggs dissected from gravid females of H. manteo. Michigan, 1 9 7 ^ - 7 5 ................. 6 3 15. Egg parasitism of H. manteo and S. canlcosta by Trlchorramma sp. and Telenomus sp., Michigan, 1 9 7 ^ , ........................... 6 7 16. Egg parasitism of H. manteo by Trlchogramma sp. and Telenomus sp.,"Michigan, 1 9 ?^ . . . “ . . . . v 72 Table Pape 17. Characteristics of Instars of H. manteo larvae In HI chi p a n ............................... 7** 18. Effect of temperature on stadia of H, manteo 19. Percent parasitism of H, manteo larvae, Michigan, 1 9 7 * * ................................... 8** 20. Percent parasitism of H. manteo larvae, Michigan, 1975 ................................. 21. Rearing regimes for H. manteo larvae used In foliage consumption study ........ • 22. Consumption of foliage from two Quercus sp. by Instars of H.m a n t e o .......... • • 81 98 100 23. Effect of temperature on consumption of oak foliage by H. manteo . . . . . . • • • . , . 1 0 3 2**. Consumption of oak foliage by non-parasitlzed H. manteo larvae and those parasitized by ...10** B. b e t h u n e l .................. 25. Insecticides and dosages applied for control of H, manteo. Michigan, 1 9 7 ^ ............. 26. 107 Efficacy of Insecticides tested for control of H. m a n t e o .................................. 109 vl List of Figures Figure 1. 2. Page Location of researoh plots for the study of H. manteo populations In Michigan, 1973-76 • • . • 7 Diagram of sampling design used to estimate prepupal densities of H. manteo. Michigan, 1973-76 ............ “ .............................10 3» Bed color phase of H. mnfcao prepupae (Mag. 2.5x) ............. 13 A. Parasitized by D. bethunel B. Normal prepupa "". . . .......... ...........13 4. Green color phase of H. manteo prepupa showing tra­ ces of red coloration (Mag. 2 . 5 * ) .......... 13 5. Variance vs. mean of prepupal samples, Michigan, 1973-76 ................ ' ......................... 17 6. 7. Number of samples containing zero prepupae per 2 0 samples vs. mean density of prepupae. . . . . 21 Possible color pattern of H, manteo larva feeding In canopy (Mag. 1.5*) 24 8 . Typical color pattern of H. manteo larva completion of feeding (Mag. 2x3 9« upon T ................. 24 Opening of defense gland of H. manteo immedi­ ately prior to formic acid secretion(Mag. 13x) . 10, Defense gland of H. manteo Illustrating three chambers (Mag. 1 3 5 : ) ........................... 3 0 11, The effect of temperature on development of prepupae of H. manteo collected April 8 , 1974 12, • . The effect of temperature on development of prepupae of H. manteo collected May 17. 1974 . . . 30 39 40 13, Collection date as related to development of H, manteo prepupae maintained at 24°C ............ 4l 14, Soil temperature (°C) at 8 cm depth Big Steer lake, Michigan, 1974 . . . . . . . . . . 42 15, Pupation of H. manteo under veurlous soli moisture c o n d i t i o n s ...............................52 16, Egg mass of H, manteo two days af^er ovlposition showing rod ring characteristic of viable eggs (Mag. 5 x ) .................................. vii 59 Figure Page 17* Bgfif mass of 2* five days after ovlposltlon showing first instar head capsules T«ag. 1 O x ) ........................................ 59 18, Adult male of H, manteo showing typical color pattern of aduTts (actual size) ................. 59 19* Number of adults of H. manteo collected in black lights, Michigan, 1$7**...................... 6l 20, Number of adults of H. manteo collected in black lights, MicKigan, 1 ^ 7 5 .................. 62 21, Parasitized eggs of H, manteo illustrating adult exit holes A. Trlchogramma sp............................. 70 B, Telenomus s p ................................ 70 22, (a) (b) (c) (d) First instar head capsule of H.manteo ,. , • 76 Second instar head capsule of H. manteo ,• , * 76 Third instar head capsule of H.manteo .. . . 76 Top (left) normal fourth instar head capsule of 2, m a n t e o .......................... 76 (right) normal fifth instar head capsule of H, m a n t e o .................. 76 Bottom (left) Fourth Instar head capsule of H, mawteo parasitized by D. b e t h u n e l . . . . . . . . 76 Tright) Fifth instar head capsule of H, lBAntao parasitized by D. bethunel .................. 76 (e) larva of D. bethunel in overwintering prepupae of H. m a n t e o ................ 76 (f) Pupa of D. bethunei with H. manteo pre­ pupae in overwintering c e l l ............76 2 3 . Numbers of adult D. bethunel from Malaise traps, Big Star lake, Michigan, 197**.............. 88 24’, Percent parasitism of H, manteo larvae, Michigan, 197**-75 25* 92 Mean larval instar of H. manteo in pyrethrum spray samples. Big Star lake, Michigan,1975 • • . 97 vlil INTRODUCTION In 1973» 20,000 to U0,000 acres of* mixed oak forests of the Manistee National Forest, In the west central area of Michigan’s lower peninsula, were totally defoliated by the variable oakleaf caterpillar, Heterocampa maviteo (Double day). This research project was developed In response to that lar^e defoliation. Since Initial reported outbreaks In the late 1800*s (Hooker, 1908), this Insect has continued to cause scattered defoliations throughout the eastern and central United States. Between 1955*75* 18 states reported defoliations In the Cooperative Economic Insect Reports. Ma­ jor Infestations of over 1,000,000 acres have occurred In such diverse locations as Missouri (Kearby, 1975a)* Minnesota (Wetzel, 1972), Arkansas (Anonymous, 1971) and Virginia (Anony­ mous, 195®)* Despite these massive Infestations the Insect Is considered a minor forest pest. Defoliations occur late In the o-rowlnjff season) consequently, 2 to 3 years of consecutive defoliation are required before simlflcant tree mortality can occur. Infestations are rarely consecutive and the host trees are generally poor quality oaks, although a wide vari­ ety of trees are subject to attack (Wilson, 1971). Defolia­ tion does, however, reduce the vljror of trees and Increases susceptibility to attack by secondary pests (Kulman, 1971)* Evidence Is also mountlnr that defoliation-weakened trees are more susceptible to infection by the root rot fungus, Armlllarla mellea (Vohl.) Quel., and that the ultimate damare to 1 2 a stand from defoliation may be more serious than Is Immedi­ ately apparent (Houston and Kuntz, 1 9 6 *1-1 Parker and Houston. 1971). With the recent Introduction of the gypsy moth. Por- thetrla dlspar L., Into Michigan, the possibility now exists that oak trees may be defoliated twice In one growing season; an early season defoliation caused by P. dlspar and a late season defoliation of second flush leaves caused by the vari­ able oakleaf caterpillar. If such a double defoliation does occur, tree mortality may be extensive. Large acreages of forest lands are aesthetically damaged and larvae represent a nuisance to homeowners and recreational users of oak fo­ rests (Millers and Erickson, 1970)• In addition to aesthetic damage Kearby 0 975a) has reported that larvae and prepupae secrete a formic acid solution which causes skin lesions. The variable oakleaf caterpillar Is native to North America. It has been recorded from nearly all states and Canadian provinces east of a line drawn from western Ontario through eastern Texas. It belonrs to the family Notodontl- dae, and was first described by Doubleday (t8**-l). Packard (1 8 9 5 ) provided a bibliography of early taxonomic synonymy and descriptions of the egg, larval, pupal, and adult stages. Wilson (1971). based on early data, published a general des­ cription of the life history and natural controlling agents. A brief synopsis of the current Information on the Insect's biology, based primarily on Wilson's (1971) description, Is presented. This Is Intended to provide an overview of the life history, predators and parasltolds of the Insect as known before this study was Initiated. There Is one generation per year in the northern 3 part of the variable oakleaf caterpillar's range. In the area south of a line extending: from Virginia to Missouri, two generations are common (Wilson, 1971)• The insect over­ winters as prepupae In silken cells constructed In the leaf litter and soil. Pupation normally occurs the following spring although some prepupae have been reported to remain In the soil for a year or more before pupation. Adult moths In the northern range begin to emerge near the end of May or early June and lay their eggs singly on the leaves of the host (Wilson, 1971). Each female may lay as many as 500 eggs which hatch In 7 to 10 days. The first lnstars skeletonize the lower surface of the leaves, but subsequent lnstars con­ sume the entire foliage between the major veins. About mid- August the larvae cease feeding, drop to the ground and spin silken cells In which they overwinter. In the south where two generations per year occur, the moths emerge about mid-April. feed during May to early July. soil and pupate. Eggs are laid and larvae The larvae then drop to the Adult moths emerge and lay their eggs In late July, with feeding of newly hatched larvae completed by late September or early October. The fall generation then overwinters as prepupae1 with pupation occurring In April of the following year. The information concerning the predators and parasltolds of H. manteo and their Impact upon population control was meager. Wilson (1971) stated, "Several species of In­ sects of the families Tachlnldae, Ichneumonldae, and Braconldae have been reared from the closely related saddled promi­ nent and probably attack the variable oakleaf caterpillar as if well." No Information on parasltolds of H, manteo had been published. Hooker (1908) reported that In Texas the preda­ tory ground beetles Calosoma scrutator Fab. and Calosoma calldum Fab. were found on oak trees with larvae of H. manteo In their mandibles. Despite the large acreages defoliated by R. manteo and Its aesthetic impact on homeowners and recreational users, there existed no substantive study of the life history and population dynamics of this Insect. The principal objective of this study was to develop a comprehensive understanding of the life history and population dynamics of H. manteo In Michigan. Specific efforts were made to monitor population levels* parasitism rates, and the effect of the environment upon survival and development of the insect*s life stages. Attempts were also made to formulate methods for predicting defoliation. Research Area In the summer of 1973* 5 permanent research plots were established on federal land In the Manistee National Forest. Each plot was hectares In size with trees consist­ ing predominantly of northern red oak, Ouercus borealis var. maxima Mich., and white oak, Quercus alba L. In addition, some red maple, Acer rubrum L., beech, Fagus grandlfolla Ehrh., eastern white pine, Plnus strobus L., and sassafras, Sasslfras albldura (Nutt.), were present in all plots. The plots were selected on a north-south transect within the Manistee National Forest (Fig. 1). This transect also pro­ vided an opportunity to study populations of H. manteo at 5 various densities. All research sites had been severely de­ foliated by the redhumped oakworm, Symnerlsta canlcosta Franclemont, during 1970, *71, and *72 (Robertson et al. 1972 1 Hillers and Wallner, 1975). The most southerly plot, designated Newaygo, was 5 km north of Newaygo, Michigan In Newayro County, Everett twp,, T13N, R12W., Sec. 30. Trees within this plot consisted of pole-slZed red and white oaks approximately 25 cm in dbh. The Big Star lake plot was considered the epicenter of the 1973 defoliation with 100J6 of the foliage consumed. This plot was approximately 56 km north of the Newaygo plot. In lake County, lake twp., T1?N, , RUtff. , Sec. 33. Trees consis­ ted of red and white oak approximately 30 cm In dbh. Two plots were located in the vicinity of Branch, Michigan, lake County, Sweetwater twp., T18N., Rl^W., Sec. **- and 5. These plots were selected In close proximity to each other (1.2 km) to Investigate the effects of tree age on population dynamics of H. manteo. The plot, designated Branch Mature, consisted of mature oaks with an average dbh of kO cm. The Branch Pole plot was comprised of pole-sized oaks with an average dbh of 25 cm. These 2 plots were completely defoliated by larvae of H. iw^ptwo during August, 1973* located 1 km north of Dublin, Michigan. The fifth plot was The precise location was Kanlstee County, Norman twp., T21N., R14W,, Sec. 36. This area consisted of red and white oaks approximately 2 5 cm In dbh which had been severely defoliated by S. canlcosta Frclmt. during 1971 and 1972. All five plots were situated on marginal lands pos­ sessing poor, sandy soil, low in fertility ahd moisture hold- 6 Fig. 1. Location of research plots for the study of H, manteo populations In Michigan 1973-76. 1. Dublin research area 2. Branch Pole research area 3. Branch Mature research area I*. Big Star Lake research area 5. Newaygo research area ■ Manistee National Forest 7 FIG. 1. 8 lng capacity* The solid of the replon have been classified as Rubicon-Gra.ylinr (Veatch, 1953)* The area was previously covered with white pines which were harvested in the late l800*s, The organic layer of soil in all 5 plots never ex­ ceeded 1 3 crai below which mineral sand was present to an un­ known depth. A soil sample from Branch Pole indicated that an 8 cm diameter core of soil removed to a depth of 20 cm contained 75% sand, 15a clay and 10^ loam. Perhaps, the greatest value of the oak trees in this reprion is their soil conservation preperties. Method for Estimation of Prepupal Populations The samplinr techniques used to estimate insect popu­ lations have been reviewed by Morris (1955) and Southwood (1966). In this study an estimate of absolute population (no,/unit area) was most suitable for lonr term population estimates, ^he most easily sampled life stare is the pre- pupa of H. manteo. Prepupae are defined as those larvae which have entered the soll-lltter to overwinter. They dif­ fer physiologically from larvae in that they have ceased feedinr, have completely evacuated the direstive tract, and exhibit a photo-nerative response. This stare is most suited for samplinr because the duration of the prepupal stare is considerably longer than all other life stares combined. Pre­ pupae remain stationary once they have entered the litter, so samples can be taken durinr the late fall and early sprinr months when time demands are minimal. Ideally, the sampllnr procedure would satisfy the followinr criteria* 9 1) allow for a comparison of populations between plots* 2) Indicate spatial distribution of prepupae. 3) allow for a comparison of populations through time. U) Indicate areas of high or low density within plots by mapping the number of prepupae recovered per sample. 5) provide living specimens to be used In further re­ search. Prepupal samples were based on a quarter of a square meter ( 5 0 cm x 5 0 cm) of soli andlitter removed cm, to a depth of 10 Grlmble and Newell (1972a) had used a similar procedure to estimate pupal populations of Heterocampa guttlvltta (Walker). In each plot a systematic sample grid composed of 20 subplots was established. jacent subsamples (Pig. 2). Each subplot was 36 m from ad­ In October of 1973 soil litter samples from each subplot were removed. The sample unit size was standardized using a 50 cm x 50 cm aluminum frame. A stralght-bladed shovel was used to cut along the Inner mar­ gin of the frame and remove the soil and litter. Each sample was placed Into a 30 cm x 76 cm x 20 cm polyethylene bag and labelled according toplot, grid location, and sample date. Prepupae which lay on the sample perimeterand were severed by the shovel were Included within the sample. In a similar manner 3 additional sets of samples were taken 1 meter from the original grid location* These samples were timed to mea­ sure possible winter mortality, predation of prepupae In the spring, and proportion of the population which remained In the soll-lltter for a year or more. In the autumn of 197**- I .5m 1 Pig. F -Oct. sample W - March sample S-May sample A-Aug. sample Diagram of sampling design used to estimate prepupal densities of H. manteo» Michigan, 1973-6. 11 and '75 the grid was shifted 5 meters to the north In all plots to minimize the possible effect of litter removal on prepupae entering the soil at the cessation of feeding in 197** and 1975* Samples were refrigerated at U.V5 C until soil and litter were sieved. The samples were broken by hand Into small pieces and sieved through , cm wire mesh screening. Prepupae were found In curled repose similar to that of white grubs and were too large to pass through the mesh. They were readily apparent In the sieve frame because their red (Fig. 3) and green (Fig. **) colors contrasted markedly with the dark soil and litter. By examining soli which had passed through the sieve It was estimated that recovery ex­ ceeded 95%* Prepupae were segregated as to plot and recorded ac­ cording to grid location. The color of prepupaef whether green, pink, or red was also recorded, Prepupae parasitized by Dlradops bethunel (Cresson) were also determined for each sample by examining prepupal body and head capsule size (Plff* 3) (Surgeoner and Wallner, 1975)- Prepupal Densities and Distribution The results of prepupal samples are presented In Tables 1-$. The sampling technique provided a measure of population densities with a standard error approximately 20% of the mean. By plotting the variance vs. the mean (Fig, 5) for all 5 plots It was apparent that prepupae were Oi contagiously distributed (S >3?) at high densities, but that at low densities the population approached a Polsson distrl- 12 but Ion (S2 » X). The sample distributions for each date were tested using a Kolmogorov-Smirnov test* At the lower densities (ca 1.5 prepupae per .25 m ), the distributions were Polsson ( p < * 0 5 )i at densities greater than 2 prepupae . 2 5 m the distributions were determined to be negative binomial ( p < . 0 5 ). There was no clear Indication why prepupae were clustered within certain samples. In areas of high prepupal concentra­ tion, additional samples were removed 1 meter adjacent to previous samples. The variability between these samples was extremely high which Indicated clustering resulted from micro environmental factors. Grlmble and Newell (1972a) found no clear Indications of preferred pupation niches for H. guttlvltta. In a long term study of this nature. Information con­ cerning whether the population was increasing or decreasing through time was necessary. Prepupal samples present an ex­ cellent Indicator of population change. The Index of each overwintering population to that of the previous year Is pre­ sented in Table 6. The populations In all study plots have continued to decline since the defoliations of 1973. The overwintering population in 1 9 7 5 was approximately 33# of the 1973 population. Wllcoxon's matched-pairs signed ranks test showed that populations at Big Star Lake were significantly higher ( p < .01) than all other plots. This was expected since the area was considered the epicenter of defoliation. Prepupal densities for any sample date at Branch Pole were not slgnl- 13 FIG. 3 . Red color phase of Heterocampa manteo (Doubleday) (Map:. 2,5x) A) Parasitized by Dlradops bethunel (Cresson) B) FIG. Hormal prepupa Green color phase of H. manteo prepupa showing traces of red coloration (Ma,repupae/. 2 Plot 73-74 7^-75 73-75 .50 .66 .29 .2 0 5 .2 0 3.25 .72 .6 3 .45 1.90 1.20 .52 .63 .39 3.40 2 .2 5 1.20 .66 .53 .35 2.70 .75 .6 5 .28 .86 .24 Oct 73 Oct 74 Newaygo 2.55 1 .7 0 Big Star 7 .2 0 Branch !tale3.65 Branch Mature Dublin Generation Index Oct 75 19 flcantly different from those of Branch Mature which Indi­ cated tree age did not affect prepupal populations. During this study prepupal densities did not pro­ vide a reliable method for predicting defoliation. Despite an overwintering prepupal population at Big Star lake of approximately 8 prepupae/.2 5 m In 1973* there was little evidence of defoliation In 197^. This was explained in part by the large number of prepupae which remained In prolonged diapause and by extremely high levels of egg parasitism In 197^. Crimble and Newell (1972a) similarly, reported pupal densities of H. guttlvltta were poorly correlated with sub­ sequent defoliation. Prepupal densities do, however, indi­ cate where the potential for defoliation existed. Soil Samples as a Detection Survey Detection surveys measure the presence or absence of an insect population In a particular area. Ideally, the sample method should detect low density populations and yet require little time, training, man-power or equipment. The 2 use of 5» *2 5 m samples taken along a straight line with points spaced 36 meters apart provides a survey method to de­ tect prepupal populations of H. manteo. Experience has 2 shown that a single person with a shovel, .2 5 m frame and selve (.6*+ cm mesh) can complete such a survey for an area In less than 4-0 minutes. The accuracy of this survey technique was based on data collected estimating prepupal populations using 20, 2 •25m samples. At the lowest densities found In the threeyear study (ca. .5 prepupae/.2 5 m ) T2 of 20 samples contained 20 zero prepupae (Fie. 6). The probability of detecting; at least I prepupa In 5 samples at such densities Has i x x x x y|) « *95 ox 95#. At densities ca* 1 prepupae/.2 5 m , In the plot with the highest propor­ tion of zeros• 9 of the 20 samples contained zero prepupae. The probability of detecting at least 1 prepupa In 5 samples at such densities was 1 99^. x x x At densities above 3 prepupae per ,25m x • .99 or there was never an Instance when 5 or more samples contained zero pre­ pupae. Detection surveys as described can therefore be used to locate populations of prepupae at densities of .5 prepupae/ 2 .25m and above• This type of survey can be carried out in mixed oak forests from Sept. to Nov. and Apr..to July In a northern area such as Michigan. When populations are detected, a re­ searcher may expand the number of samples to 20 for an esti­ mate of prepupal populations. A prediction of defoliation cannot be based on pre­ pupal estimates. Actual prediction must be based on early Instar populations (see prediction of defoliation). This type of sampling scheme allows the researcher to Identify areas with the highest prepupal densities and therefore the highest probability for defoliation. These areas may then be sampled In A u gust to estimate early lnstar densities for possible de­ foliation* A series of detection surveys was conducted In mixed oak forests throughout the state. Samples were taken during June, 1975* In the vicinity of Ellis Lake, Ludlngton, Flier City, Cadillac, Muskegon, Greenville and Brethren, Michigan. 21 (/) Ui -J Q. 2 < CO 6 U. o C> z 2 3 MEAN PIG. 6 4 5 6 7 8 9 DENSITY (per .25m2) Number of samples containing zero prepupae per 20 samples vs. mean density of prepupae. 10 22 Only at Brethren* Michigan* which was nearest to the research 2 plots* were prepupae detected (.** prepupae/.2 5 m ). Coloration of Larvae and Prepupae The variable oakleaf caterpillar was so named be­ cause of the tremendous variation of colors and pattern found In the larvae and prepupae. larvae and prepupae vary from solid red (Pig. 3) to nearly complete green (Pig. **)• This color and pattern variability has been reported from all areas In which H. manteo Is distributed (Wilson* 1971). Prepupae collected In Oct.* 1973* varied through all color types. The prepupae In samples were categorized as red, pink or green. In the highest density plot* Big Star lake, the red color predominated* and as density declined the percentage of red colored prepupae also declined* Table 7. By comparing the percentage of red prepupae In the fall and subsequent spring samples* It was evident that the num­ bers of red prepupae declined* Table 8 . In April* 1976, no red prepupae were found In soil and litter samples* These declines In red prepupae were explained because individual prepupae actually change color. The red pigment gradually fades to pink and eventually to green with small traces of pink still evident (Pig. *0. All prepupae observed In the laboratory were bright green immediately before pupation. Packard (1 8 9 5 ) based on information supplied by C.V, Riley reported that all prepupae became a unifora parls green de­ spite the fact that several color types were placed Into the soil. In feeding trials* larvae of H. manteo were capable of 23 changing colors. Green larvae with a red hand on the dorsum, the thorax and first abdominal segments (Pig1. 7) changed to a dull, pink color (Pig:. 8) when feeding? ceased. color change occurred within Zk hours. The complete Klots (1967) and Dyar (1891) had previously observed dramatic color changes In the final instars of Heterocamna when feeding: ceased. Klots (1 9 6 7 ) found that fifth lnstars of H. pulverea (Grote and Robinson) turned a brilliant pink when feeding? ceased, but that this color faded to a pale green Just before pupa­ tion. The color change of H. manteo to a greyish-pink may Increase prepupal survival on the forest floor since the green larvae are readily apparent against the background of fallen leaves. This advantage, if any, is considered minor since all prepupae observed burrow directly into the litter. The possible significance and determining: factors of H. manteo coloration are questionable. Since prepupae of high density populations possess the greatest red color, color­ ation may be density dependent (Iwao, 1 9 6 8 ), Other possible explanations are that color is characteristic of population quality (Wellinrton, 196*0 or that color is dependent upon diapause Induction (Tauber et al., 1970), or the nutritional quality of the host (Balmer and Knight, 192*0. Fuzeaw-Breasch (1972) has reviewed the factors which can Induce color changes. Klots (1 9 6 7 ) found that sibling larvae of H. pulversa show a marked dimorphism. He concluded that larval dimorph­ ism was not linked with the rate of development, sex or any discernible adult characteristic. Tests were conducted to determine if larval density Fig. 7 Possible color pattern of H. manteo larva feeding In canopy (Mag. 1.5 Fig. 8 Typical color pattern of H, manteo larva upon cessation of feeding (Mar. 2.5x). 25 Tkble 7. Prepupal color types of H. manteo as related to density, Michigan, Oct. 1974. O Density/.2 to Location % Green % Pink % Red Newaygo 2 .5 5 + 2 .6 18 **5 37 Big Star Lake 7.20 ± 5 . 1 15 22 63 Branch Mature 3 .**o ± 3 . 2 19 29 52 Branch Pole 3.65 ± 2 . 7 26 25 49 Dublin 2.70 + 2.7 26 48 26 Table 8. Seasonal color patterns of H. manteo prepupae Michigan, 1973-7**. Season % Prepupae Green % Prepupae Pink f Prepupae Red Pall, 1973 20.8 3 3 .6 4 5 .8 Winter, 197** 28.6 35.0 36 .2 Spring, 197** 36.8 3 9 .** 24.0 Summer, 597** 55*8 37.2 6.8 % Change Pall-Summer +35*0 +3,6 -39 % Prepupae Red Par. 1976 26 affected the color patterns of H. manteo. Twelve larvae from a single egg mass were reared at different densities! In one Instance 6 larvae were reared Individually each being fed In a single pint-sized container. The other 6 larvae were reared together within a single pint container. Both groups of larvae were maintained at 24°C with a 16 -hour photoperiod, larvae reared Individually never consumed all the food pro­ vided! whereas» during the final lnstarv larvae reared toge­ ther consistently ate all food provided such that in each 2*fhour period approximately 6 hours were spent without food. There was no significant difference in color or pattern of individual or group reared larvae. This Indicated that lar­ val color was not density dependent or determined by a short­ age of foliage. I concluded from rearing over 200 larvae that color was not dependent upon photoperiodt larvae reared In complete darkness maintained the same color and pattern as siblings reared under a f6-hour photoperiod. Siblings reared at 26°C, 24°Ct 20°C and 15°C did not differ significantly in color. By rearing over 200 larvae through to adults I concluded that larval or prepupal coloration was not sex-linked nor linked to any discernible adult coloration. As previously noted the color of prepupae gradually faded from red to green as they neared pupation. A study was conducted to ascertain If field-collected green prepupae pupated significantly sooner than red prepupae. Prepupae were collected from Big Star lake in March of 197^. They were segregated by color and placed Individually Into 50 27 Jars filled with soil and litter (soil moisture* 18JO and maintained at 24°C, and 16-hour photoperiod* The mean number of days required for pupation by the preen prepupae Has 29.2 + 12.57 days compared to 3 6 . 8 + 12.5 days for red prepupae. This was significantly different using Wllcoxon#s matchedpalrs signed ranks test (p<.05). The color of prepupae ap­ peared to reflect the physiological age of the Individual with green indicating a more mature condition. This knowledge may prove valuable when prepupal sam­ ples are made in an area. If fall or early spring samples are predominantly red (Fig. 3) s. researcher can assume that a major component of the prepupae are from the current year* If, however, the majority of the prepupae are green to dull pink (Fig. U) one can assume that a majority of the prepupae are from a population which has remained In the soil a year or longer (see diapause section). For example, during the growing season of 1975» larvae of H. manteo were extremely rare at Big Star Lake, Michigan. Pyrethrum spraying of ma­ ture oak trees showed an average of less than 20 larvae per tree (see larval sampling). During late August, 1975* 3 ex­ perienced collectors spent 6 man-hours searching for larvae on the foliage and discovered only 12 Individuals. However, soil samples In the fall and spring showed an average of 2*5 prepupae/.25m . The majority of these prepupae must have carried over from populations in 1973 and 197^. This was strongly Indicated because no red prepupae were found in sam­ ples during Oct., 1975 or April 1.976| whereas. In Oct., 1973, approximately 69% of the prepupae were red In color. 28 Defensive SeoretIona Prepupae of H. manteo frequently spray a fine mist which produces an extremely pungent odor similar to that of formic acid. This secretion causes a burning sensation to cuts but none to normal skin* When tasted the substance caused a severe burning sensation to the tongue, but no le­ sions. In Missouri the secretions caused veslcatlons and skin lesions to those handling variable oakleaf caterpillars (Kearby, 1975a)• The reaction to collectors in Michigan was not as severe since no blistering occurred. The opening to the gland producing the secretion is a narrow slit on the median of the ventral prothoracic seg­ ment (Pig. 9)* The position of the gland is Identical to that of Schlzura coclnna (Abbot and Smith) described by Detwller (1922). The gland opening Is everted at the time of spray ejection (Pig. 9) which apparently allows some direc­ tional control of the spray. The secretion was observed to be expelled several centimeters. Several prepupae were dis­ sected, which revealed that the gland producing the secretion extended posteriorly Into the meso and meta-thoraclc segments occupying a large volume of the thoracic cavity. The gland consists of two small anterior sacs and a single large pos­ terior sac (Pig. 10). Detwiler (1925) had previously de­ scribed a similar Internal gland for several notodontld lar­ vae. The substance secreted was analyzed by Dr. M. Zablk, biochemist, Dept, of Entomology, Michigan State University, and was found to be comprised of formic acid and probably 29 some long chain fatty acids. Poulton (1886), Packard (1886), and Herrlok and Detwller (1919) have all reported formic acid secretions by larvae of Notodontldae. The other chemical compound associated with the secretion of H, manteo has been reported to be acylic ketones (Eisner et al. 1972). When early instars were handled In the field collec­ tors discovered that formic acid secretions were ejected by second and subsequent Instars. observed to eject any fluid. First Instar larvae were never A carabld adult, Plnocedera sp., was observed to approach a fifth Instar. The larva sprayed the adult beetle which Immediately retreated, cleaning Its eyes and antennae. A similar repelling action was noted by Eisner et al. (1972) In tests conducted with Lvcosoma sp. The acylic ketones of H. manteo repelled formic acid-produc­ ing ants (Eisner et al., 1972). Kearby (1975a) reported that no birds were observed to feed on larvae In Missouriy similarly, no birds were observed feeding on active larvae In the canopy In Michigan. Kearby (1975a) suggested that the lack of bird predation may be a result of these defensive secre­ tions. There was little predation of prepupae In the soil. This was determined by comparing overwintering populations to populations the following years. Wllcoxon’s matched- palrs signed ranks tests showed no significant difference be­ tween fall populations In 1.973-75 to those the following May and April ( p < .05). Shrews, deer mice, Peromyscus mlnlcula- tus balrdll (Hoy & Kenlcott), white-footed mice, Peromyscus leucopus (Rafinesque), have been reported as predators of 30 f Fig. 9 . Fig. 10. Openlnr of defense rland of H. manteo immediately prior to formic acid ejection (War. 13x). Defense e-land of H. manteo showing three chambers (Mag. 13x). 31. Insect populations causing significant population mortality (Holllng, 1959i Campbell, 1975)* These vertebrates were pre­ sent In the research plots, but apparently rarely fed on pre­ pupae. The prepupae maintain a relatively large supply of defensive secretion. Eisner et al. (1972) discovered 10, 13» and 22 mg. of secretions In the three sacs of H. manteo fifth instars. Although overwintering prepupae do not eject any fluid, the supply of formic acid and ketones may deter pre­ dation by vertebrates which have previously attempted to con­ sume prepupae. The red and green colors of larvae may serve to warn potential predators. Cold Temperature Mortality Studies In Minnesota soil samples collected in 1 9 ^ Indicated 2 an average of 6 prepupae per ft under Infested trees, but In 19^5* no larvae of H. manteo were found in the canopy (Wetzel, 1972). In Virginia (Anonymous, 1958)• where several million acres were defoliated In 1956 there was little or no adult emergence despite the fact that prepupae were abundant In the soli during 1956-57. Declines In populations of H. manteo may have resulted from predation or cold-temperature mortal­ ity of overwintering prepupae. A study was therefore conduc­ ted to determine cold-temperature tolerances of prepupae. To assure cold-hardiness, prepupae were collected from Big Star Lake during February of 197^. Prepupae sieved from samples were stored In moist litter at 4.^°C until coldtolerance trials were Initiated. exceeded 1 week. The storage period never Prepupae removed from storage were placed 32 In 2 3 cm long, 9 mm glass tubes where they were separated by nylon screening. Six prepupae were placed In each tube which was furnished with a few drops of water at the bottom, before sealing the top and bottom with rubber plugs. The tubes containing prepupae were placed In ethylene glycolwater baths maintained at constant temperatures as described by Casagrande (1975). Rrepupae remained In the cold temperature baths for a 24-hour period at which time the tubes were removed from the baths. They were then held at 20°C for 4 hours after which mortality was assessed. Prepupae were examined under a steromicroscope and the mouthparts touched with a metal probe. If no response to the probe was noted, the prepupa was consid­ ered dead. The results of cold temperature trials are presen­ ted in Table 9. Results Indicated that under these conditions pre­ pupae were unable to withstand temperatures below -6.8°C and that significant mortality occurred bfclow 0°C. Prepupal sam­ ples (Table 1-5)* however, indicated that there was no sig­ nificant mortality during the winter months of 1973* 1974 or 1975* Wllcoxon's raatched-palr signed ranks tests were used to compare the fall and spring samples for the years 1973-76. In no Instance was there a significant reduction in prepupal numbers (p<.05). Soli temperatures were recorded during the winter of 1974-75 using a Vfeathervane (Model T-6 0 3 -1 6 ) threepoint, 31-day recording thermograph. The temperatures were monitored at the surface of the litter, at the lltter-soll interface (7.6 cm deep), and within the mineral sand approxl- 33 Table 9. Percent mortality of mantao prepupae exposed to cold temperatures for a 2*4— hour time period. Temperature -6.7°C % Mortality 100 No. of larvae 32 -5.5°C 9^.1 17 -3.9°C 53.7 5* -2.2°C 29.^ 17 -1.1°C 2 3 .2 56 11.1 72 0°C 3* mately 3 0 cm In depth. The thermograph showed that tempera­ tures at the surface remained constant at 0°C throughout the winter months and +2.2°C at the lltter-soll Interface where prepupae are found. The temperature In the mineral soil was approximately 1,1.°C throughout the winter. There was little or no fluctuation of these temperatures from mid-November, 197**, until early April 1975» because snow cover acted to In­ sulate diurnal fluctuations. Prepupae were never exposed to lethal temperatures during the winter of 197^-75* During the 3 winters of this study, the litter or soli was never found to be frozen. Prepupae apparently seek out a habitat In which winter temperatures do not reach lethal levels. Such a behavioral survival mechanism was termed frost avoidance (Salt, 1 9 6 1 ). The digestive systems of prepupae were always totally evacuated, another mechanism to minimize the possi­ bilities of mortality by freezing (Salt, 1953) ot fungal and bacterial Infection, Diapause and Prepupal Development Diapause Is defined "as a state In which a reduction of growth processes or maturation occurs which Is not neces­ sarily caused by Immediate environmental Influence, does not depend for Its continuation on unsuitable conditions, and Is not easily altered by change to a more favorable condition" (Slmmonds, 19^8). Diapause may be of two types, either obli­ gate or facultative (Andrewartha, 1952). Obligate diapause Is defined as Inherent and seems to occur Independent of any normal variation In the environment. Facultative diapause 35 occurs as a response to the appropriate stimulus from the environment. Prepupae of H. manteo drop from the canopy to the forest floor upon the cessation of feeding. This was con­ firmed by placing inverted cone traps with a basal area of 2 ,25m under infested treesi large numbers of prepupae were discovered within these traps which Indicates that caterpil­ lars drop to the forest floor rather than crawl down the tree bole. Once on the forest floor the Insect then burrows Into the leaf litter and soil, overwinters In a silken cell approx­ imately 3 cm In diameter. The cells are generally found at a depth of 8 cm where the humus and mineral soil interface* although some prepupae (5%) can be found within the leaf lit­ ter. The prepupae diapause within the cell throughout the winter and early spring before pupating and emerging as adults In late July. Prepupae collected in October* 1973* were placed in­ dividually Into 50 ml glass jars filled with 3 0 gm of fieldcollected litter and soil containing approximately 18# HgO by weight. These Jars were than placed In Sherer environ­ mental chambers from 20 Oct., 1973 to 28 Dec.* 1973l a per­ iod of 69 days. Consequently* prepupae were held at k,k>°C for 3 months before studies were conducted to assess the effect of temperature on prepupal development. In addition, prepupae were collected in the months of March* April and May after successfully overwintering. Prepupae collected In the spring were also placed Individually Into 5 0 ml glass jars 3 6 filled with 30 gm of field-collected litter and soil. The jars were labelled* sealed with lids and placed in Sherer environmental chambers. One chamber was controlled at 2^f°C with a l6-hour light and 8-hour dark photoperiod1 additional chambers were maintained at 20°C* 15°C* 10°C, and ^,^°C with complete darkness* The results of prepupal rearings are presented in Table 10. Sex of prepupae did not significantly affect de­ velopmental time (p<*05). There was no development of pre­ pupae at temperatures of 10°C or lower* but development did occur at 15°C, The minimum termperature for pupation was ap­ proximately 12*5°C. Prepupae were found to remain viable for up to 7 months when stored at temperatures between ^.4°C and 10°C. The time required for pupation showed extreme vari­ ability. This variability was explained in part because pre­ pupae were of different physiological ages. The prepupal coloration served as an indicator of these ages (see H. man­ teo coloration). The intensity of diapause also explained a major component of the variability in developmental times. Tauber and Tauber (1976) stated that "diapause is largely a dynamic state* i.e. as the season progresses* diapause depth or intensity decreases* and the animal's response to dia­ pause-maintaining factors diminish," The date on which pre­ pupae were collected from the field was correlated with the time necessary for pupation. The time necessary for development of prepupae col- 37 Table 10. Developmental times of H. manteo prepupae. Temperature of Environmental Chamber Mean No. Days Until Pupation fremale Male 2 ^°C 23.1+11.7 23.lt> ± 1 1 . 2 2 0 °C 3b. 8 ± 1 9 . 2 3 6 . 8 ± 20.7 I5°c 38.2 ± 17.8 39.b ± l b .8 10°C b.b°C No pupation No pupation 38 looted In the field on 8 April, 1974 le presented in Pis. Ilf whereas the time necessary for pupation of prepupae collected on 17 May Is presented In Pig. 12. The physiological age of prepupae on different collection dates strongly affected the rate of prepupal development. Prepupae collected 8 April, 1974 showed 50% pupation at 24°C after 26 days as compared to 1.6 days for prepupae collected 17 May, 1974 maintained at the same temperature. The conclusion that physiological ages affected development time was reinforced by comparing develop­ ment of prepupae reared at 24°C and 16-hour light, 8-hour dark photoperiod (Fig. 13). The longer the time spent in the soil during the spring months, the shorter the developmental time. The mean number of days for pupation of prepupae collected 17 May, 1974 was 13.4 ± 3.2 days, as compared to 30*23 ± 8 . 1 7 for those collected 15 April, 1974 and 34.5 ± 12.9 for pre­ pupae collected 14 March, 1974. Soil temperatures of the litter soil Interface at Big Star lake were recorded for the spring of 1975 using a Veathervane (Model T-603-16) 31-day thermograph and are pre­ sented in Fig. 14. 10 May, 1975* Temperatures did not exceed 12,8°C until Pupation does not occur at temperatures below 12.8°C, but diapause intensity Is reduced as prepupae re­ mained In the soil. The prepupal diapause was considered obligatory since all prepupae tested under "nonnal environmental stimuli'* (i.e. those field-collected from litter and soil), exhibited some Intensity of diapause. The diapause, however, was due to 39 99 90 Percent Pupation 60 45 30 e---e 24*C • • 20#C <►*•-o 15* C 7 14 21 28 35 42 49 56 Days PIG. 11. The effect of temperature on development of prepupae of H. manteo collected April 8 t 1974. bo 99 90 75 60 Percent Pupation 4 5 30 7 Days PIG. 12, The effect of temperature on development of prepupae of H, nmntao collected Way 17, 19?b, *fl /> * 99 90 75 Percent 60 Pupation 45 30 • 17.V.74 o— o|5.|V.74 14.111.74 7 14 21 28 35 42 49 56 No. of Days PIG. 13. Collection date as related to development of H. manteo prepupae maintained at 2^°C, 20 Temperature (°C)of Soil 8cm in Depth 21 31 March 10 20 30 April 10 20 30 May DATE PIG, 1**. Soli temperatures (°C) at 8 cm depth, Big Star Lake, Michigan, 197^. ^3 some environmental factor since larvae reared In the labora­ tory under conditions of total darkness and temperatures of 20° C and 15°C pupated within 3 weeks of feeding cessation. Similarly, larvae reared at 24° C under a 16 -hour light and 8-hour darkness photoperiod did not diapause. The environ­ mental stimuli which Initiated diapause was not examined* but I was of the opinion that diapause was under photoperlodlc control. A short day cycle, I.e. 12 hours of light and 12 hours of darkness has been reported to Initiate diapause In a large number of insects which overwinter In temperate cli­ mates (Dewilde, 1972). In the south 2 generations of H. man- teo occuri larvae completing development in late June or early July do not exhibit diapause (Wilson, 1971). This reinforces the hypothesis that diapause Is initiated by a Hshort-dayH photoperiod. During the spring of 197^, populations of prepupae were estimated by soil sampling. In all plots there appeared to be minimal overwintering mortality. Cone traps with a 2 basal area of .2 5 m were placed on grid locations of all plots to monitor adult emergence. Adult emergence appeared considerably lower than expected, based on successful over­ wintering levels. Emergence was not, however, accurately de­ termined because of extensive vandalism and wind damage to cone traps. After adult flight (as measured by black lights) prepupae existed within the soil directly beneath the cone traps. This presented evidence that a porportion of the pre­ pupae did not pupate during 1 9 7 ^, but remained as viable pre­ kb pupae In the soil. Craighead (1950) and Wilson (1971) have both reported that some prepupae of H. manteo may remain dor­ mant In the soil throughout an entire year. Twenty soil samples were taken in each plot to de­ termine the percentage of the prepupal population which re­ mained In prolonged diapause. critical to ensure that* The timing of the samples was 1 )prepupae In the soil would not emerge soon after the samples were taken, and 2)that pre­ pupae discovered were from the previous year rather than newly dropped larvae. Samples to estimate the percentage of the population In diapause were made during mid-August. At this time, adult emergence as measured by black lights had ceased and no vi­ able pupae existed In the solli only prepupae. In addition, when prepupal samples were token, fifth lnstars were not pre­ sent within the canopy. Throughout July, August and Septem­ ber larval development was monitored In the field using pyrethrum samples of entire oak trees (see larval development). It was assumed that If the mean larval Instar did not exceed 4, then larvae which had completed fifth Instar feeding had not existed within the canopy. Mean larval lnstar was deter­ mined as the average age of larvae taken from pyrethrum sam­ pled trees. For example. If a spray resulted In a collection of 100 larvae of which 5 0 were third lnstars and 50 fourth lnstars, the mean larval lnstar was considered to be 3 The results of soli samples taken In mid-August are presented In Table 11, Approximately 50% of the prepupal Table 11. Prepupal densities and carry over population of H. manteo. Michigan, 197^75. Spring Densities August Densities f 197^” Carry Over Spring Densities August Densities T I M ? Carry Over Big Star lake 8.35 2.85 32.^# 5.35 2.55 ^7.9# Newaygo 2M 1.60 61.5* 1.75 1.05 60.0# Branch Mature 3.10 1.55 *9.6* 1.70 1.20 70.6# Branch Pole 2.25 1 M 5b M 2.05 1.35 65.0# Dublin 1.^0 58.6% 1.^5 .^0 27.5# Location .85 46 population did not pupate during summer of 19 7^ 1 and 5 5 % did not pupate during the summer of 1975* This phenomenon Is probably not unique to Michigan as evidenced by Anonymous (1958) who stated that In Virginia* "For reasons not wholly known* there was little or no emergence of adult Insects des­ pite the fact that prepupae were abundant In the soli during 1 9 5 6 - 5 6 .** Problems In the determination of prepupal develop­ ment are therefore additionally compounded because In any sample of prepupae the possibility exists that some have been In the soli less than a year, 2 years* 3 years or perhaps longer. No way to accurately differentiate the different aged prepupae was discovered. During August of 1974 prepupae collected at Big Star lake and Branch Pole plots were placed Into plastic MavJetR Injection capsules filled with litter and soil* and then bur­ led to a depth of 5 cm. These prepupae were assumed to have originated from the 1 9 7 3 defoliation by H, manteo since no fifth lnstars had yet been found In the canopy. In a similar manner* prepupae were placed Into Injection capsules during spring of 1975* On 28 August* 1975 the capsules* In all plots* were checked to determine the number of prepupae which had pupated. The results of this study are presented In Table 12. Some prepupae Implanted In capsules during August of 1974 had still not pupated by August of 1975* These prepupae origin­ ated from the 1 9 7 3 defoliation and remained viable for over­ wintering to 1976. The percentage of prepupae remaining In prolonged diapause was approximately 1 0 ^ as determined by **7 those which dlapauaed In the capsules. This was misleading since the litter In most capsules had greater soil moisture and higher relative humidities than adjacent litter. Pre­ pupae In dry capsules (equivalent to surrounding litter) were those which had not pupated. In September 1975* 232 prepupae of known life his­ tories (either reared in the laboratory Or In field sleeve cages) were placed In open-ended soft-drlnk and beer cans* The cans were burled midway Into the soli and duff and pre­ pupae dropped onto the surface of the litter within the cans at the cessation of feeding. This allowed them to overwinter at their chosen depth and hopefully avoided the excess soli moisture experienced with Injection capsules. In the spring, netting was placed over can tops and secured with rubber bands to hold emerging adults during the summer. The cans will be checked by students currectly on the project at the end of adult emergence In 1976 to determine the number of prepupae remaining In diapause. If necessary, cans will be monitored In future years to determine how long diapause may last under field conditions* The diapause exhibited by prepupae In the fall after cessation of feeding Is obligatory since the entire popula­ tion manifests some diapause Intensity. The diapause ter­ mination In the spring is facultative since not all prepupae behave alike. Prepupae collected In the early spring always pupated In the laboratory! but In the field approximately $0% of the population remained In the soil for a year or longer. Table 12, Pupation of encapsulated prepupae of H. manteo examined 28 August, 1975* Mortality Cause Site Date of Implantation No. Implanted No Pupation Pupated 32 Fungus * Mortallt Farasltolds Ants 8 1 2 10.2* 76 6 3 - 10.2* 15.7* Big Star lake Aug. 21/7^ *1-8 5(10.5*) Big Star lake May 20/75 88 3(3.W Branch Pole June 6/75 32 10(31*) 17 5 - - Branch Pole Aug. 28/7** 25 0* 23 2 - - Newaygo May 11/75 32 2(6.3*) 26 6 - - 8* 18.8* 49 The factors which caused the termination of diapause were not totally apparent. The prepupae have to be exposed to a threshold temperature greater than 12.5°C before pupa­ tion occurs9 but some additional stimulus(1) Is required to terminate diapause. This stimulus(l) must be applicable only during certain time periods. There exist "windows in time" during which some secondary stlmulus(l) is required to Initiate pupation. May and June. This apparently occurs In Michigan during If the stimulus Is not received during this time period the prepupal diapause again intensifies and secon­ dary stimuli will thereafter not Initiate pupation until dia­ pause Intensity again declines. This would explain why pupa­ tion never occurs during July to October even though soil temperatures are above the threshold. The Intensity of dia­ pause was not as high for prepupae remaining In the litter and soil a second year. Prepupae collected In October of 1974 (a major portion of which were believed derived from the 1973 larval population) did pupate in the soil without being exposed to several months of cold temperature. These pre­ pupae ( however* required at least 40 days to pupate when held at 24°C. Two critical questions concerning this theory of dia­ pause intensity were not satisfactorily answered* what are the secondary stimuli which are necessary to terminate dia­ pause and* what factors cause diapause Intensity to again In­ tensify? Several examples of Lepldoptera pupae and prepupae which remain In the soil for 2 years or longer have been cited by Powell (1974). He suggested that soli moisture was pro­ 50 bably the critical factor which terminated diapause In those he studied. With prepupae of the pink bollworm, adult emer­ gence and diapause cessation are strongly correlated with high levels of soli moisture following rains of ,$ Inches or db more (Brazzell and Martin, 1959)* Prepupae In Mavjetf0' In­ jection capsules, which showed high levels of moisture, con­ sistently pupated while those In capsules sealed such that soli moisture did not Increase, did not pupate. This sug­ gests that soil moisture or high degrees of relative humidity are secondary stimuli resulting In diapause termination. The soils of the research area are extremely sandy {75%) with low water holding capacity. The leaf litter and associated humus acts to conserve most water reaching the soli surface. In years subsequent to major defoliation the leaf litter Is greatly reduced and the water holding capacity Is therefore reduced. This reduction in water holding cap­ acity would increase the probability that prepupae do not achieve the critical secondary stimuli for diapause termina­ tion. Thus, In the years following major defoliations a large percentage of the prepupal population remains In pro­ longed diapause. This would prevent massive population out­ breaks and thus consecutive defoliations. The Insect would benefit because a complete collapse of the population due to starvation does not occur. In addition, the insect population Is able to escape a large number of parasltold species which have been Increasing with the Increase of the population of H. manteo (see larval parasitism as related to prolonger dia­ pause ). 51 Studies were conducted to determine If soli moisture affected diapause termination. Prepupae collected from Big Star lake and Branch plots In April, 1976, were placed Indi­ vidually Into 50 ml Jars. dried soil. Bach Jar contained JO gm of oven- The soil was dried at 100°C for *4-8 hours to en­ sure that all moisture was removed. Distilled water was ad­ ded to create various percentage soil moistures by weight. The Jars were then sealed with lids and maintained at 20°C and complete darkness for a period of 33 days. pupae were tested under each moisture condition. Twenty pre­ The results of the soil moisture experiment are presented In Pig. 15* When the experiment was terminated, at the higher moisture levels of 22#, 20#, and 1 5 #t over 9 0 # of the prepupae had pupated. At the lower soil moisture levels of 12.5#, and 10# only 70# of the prepupae had pupated. Most of those prepupae which did not pupate at the lower moisture levels were dead because of desiccation. The experiment did not clearly demon­ strate that high levels of soli moisture were a necessary stimulus for pupation. In the laboratory prepupae In sealed Jars consistent­ ly pupated. The only Instance where prolonged diapause occur­ red was when prepupae were placed In a large 2m x 1m x 2m field cage. The bottom of the cage was lined to a depth of 10 cm with field-collected litter and soil. Prepupae collec­ ted, April 1976, were placed In the cage and allowed to form cells In the soil. The cage was maintained at 20°C with ca. 12-hour photo-period. tered. Every 3 days the soil surface was wa­ lOOr 90 80 70 a - et% • - to% o - 15% I t . 5% A - • - 10 % PUPATION 60 50 40 VA N 30 20 10 6 FIG, 15* Pupation of H. manteo under various soil moisture conditions* 53 Adult emergence began after 1^ days and continued for another month. and tabulated. When adults emerged they were removed When adult emergence was completed the cage was left In place for an additional month, after which the soli and litter were removed and selved. Si* of the 18 pre­ pupae placed in the cage remained viable, but did not pupate. The environmental factors which allowed some prepupae to pu­ pate and yet allowed to remain in prolonged diapause were not determined. Several possible factors are suggested. Prolonged diapause did not occur In those larvae which were held In sealed 50 ml jars. It did occur In a large cage perhaps be­ cause prepupae produce a compound which Inhibits pupation of neighboring prepupae. This suggestion does not seem likely since precentage pupation does not appear to be dependent on prepupal densities (Table 11). C02 could initiate pupation. Possibly, Increased levels of In the sealed jars the C02 Is allowed to accumulatei whereas, In open cages the C02 can dis­ sipate. The relative humidity of the air which prepupae in­ take may affect pupation. In sealed jars Including those with low soil moisture, condensation could be seen on lids indica­ ting high levels of humidity. In field cages relative humidi­ ty was assumed to be less. The factor which caused prepupal diapause to Inten­ sify In July and August was not determined. A decreasing photoperiod In July and August may intensify diapause. Pre­ pupae In the soil receive little light, but Dewllde (1962) reported that extremely low levels of light intensity could 5** cause diapause Initiation. The probability of high moisture levels remaining for several days during July and August Is remote. Diapause Intensity could Increase with shortening photoperiod and the probability of secondary stimuli also de­ creased as the summer season progressed. pause for H. mpntflo is speculative. The theory of dia­ A great deal of experi­ mentation Is required to confirm or deny It. The Insect, be­ cause of Its size, abundance, ease of rearing and prolonged diapause presents an Ideal organism for further Investiga­ tions concerning diapause. Pupal Saxlng and Sex Ratios Attempts were made to determine the sex of larvae and prepupae of H. manteo. No reliable method to sex larvae ex­ cept by dissection for developing gonads or ovaries was dis­ covered. Allen and Grlmble (1.970) were able to sex larvae of H. guttlvltta In a similar manner. Separation of sexes prior to emergence as adults has practical applications. It may be Invaluable for workers Isolating sex pheromones and popu­ lation ecologists determining sex ratios, particularly If adults of one sex are more difficult to monitor. The pupae of H. manteo can be sexed by visible exter­ nal characters found on the ventral side of the caudal seg­ ments. The genital opening Is on the eighth abdominal seg­ ment In females and on the ninth In males. Packard (1 6 9 5 ) and Ehrllch et al. (19^9) have provided Illustrations for sexual differentiation of pupae. A sex ratio of 1,12 to 1.00 males 55 to females was determined by examining and sexlng 3 9 6 pupae• TOiese were then reared until adult emergence. In all cases sex determination by pupal examination proved accurate. Developmental Times of Pupae Prepupae were examined dally to determine the date upon which pupation occurred. Once pupation had occurred* pupae were removed from 50 ml glass Jars. They were placed on the soil and litter surface filling a ^0 cm x 2 5 cm por­ celain dissecting tray. Six pupae were spaced on each tray. A large block of wood was set beside each* and a 1000 ml beaker was Inverted over the pupa and the wood. Every 3 days the litter surface was wetted with distilled water to main­ tain soil moisture. The Inverted beaker was used to hold newly emerging adults and maintain higher levels of humidity. The relative humidity under Inverted Jars was approximately 70jtf. Newly emerged adults climbed onto the wooden blocks for drying and expansion of wings. Using this technique adults with excellent wing formation emerged over 95% of the time. In several Instances pupae were left In silken cells In the 50 ml Jars to observe adult emergence. The adults emerged within the cells and Immediately worked their way to the soil surface cm) before wings began to expand. Once the surface was reached they climbed any vertical object and remained quiescent as the wings expanded and dried. The wing coloration Is similar to that of tree bark (Fig. 18) and adults are difficult to observe on tree boles* upon which they climb In nature. 56 The effect of temperature on pupal development le presented In Table 13* The minimal temperature required for adult emergence was approximately 12.5°C since no emergence occurred at 10°Cf but emergence did occur at 15°C. Pupae could be stored at temperatures of 4.4°C for periods up to 2 months without significant mortality. The duration of the pupal stage was not significantly affected by the sex of pupae. The variability associated with pupal development was extremely small when compared to prepupal development# for apparently no type of diapause existed within the pupae. Each pupa was the same physiological age at the beginning of ex­ periments! whereas the Intensity of prepupal diapause varied among Individuals. The date of adult emergence could be pre­ dicted with precision by knowing the date of pupation and the temperature regime to which the pupae were exposed. This knowledge proved useful In providing large numbers of male and female adults for pheromone studies. at Pupae were stored Immediately after pupation* until the desired number of pupae was achieved. By regulating the temperature after removal from storage the desired date for adult emergence was achieved with reasonable accuracy. In the laboratory the time necessary for adult emer­ gence varied from 2 to 4 weeks depending upon temperature. Soil and litter temperatures were recorded during late Hay* June, and July of 1975 using a Weathervane® 3-polnt 31-day recording thermograph. The temperatures fluctuated between 16,6°C and 32°C throughout this time period which Indicated 57 Table 13. Number of days to emergence for H. manteo prepupae maintained at various temperatures. Temperature of Environmental Chamber Mean No. Days until Emergence Female Male 2 ^°C 1 5 .1 20°C 20.3 ± 3*69 15°C 29.6 10°C ± 3.07 ± 5.9 lfc.8 ± k.e 20.5 ± 3.96 29.6 ± 5.28 No emergence 58 that the pupal stage lasted 2 to b weeks In the field* In 2 Instances newly formed pupae were monitored in the field and the pupal stage lasted 17 and 22 days. The peak adult emer­ gence of H. manteo as measured by black light traps occurred about 25 July, 1973* *7b and *75* This Indicated that most pupation in the field occurred about the first of July. Flight Activity Adult moths (Fig. 18) are cryptically colored similar to the bark of tree boles upon which they rest during the day. The adult stage has been described In detail by Backard (1 8 9 5 ). Hales can be readily distinguished from the females on the basis of antennal pectination* the males possessing the more pectinate antennae. Adults are short-lived1 in the laboratory they sur­ vive b to 7 days. They were kept in large screen cages and supplied with a water-molasses (5 0 *5 0 ) solution upon which they were observed to feed. Females* however* readily depo­ sited eggs without feeding on the molasses solution. The adult moths are active during the night. The nocturnal activity of the adults was monitored by black lights during 197^ and 1975* Records of the number of moths o b ­ tained In black lights were kept from three locations* Star lake* Newaygo* and Dublin. Big In both years adult emergence began about the first of July with peak emergence occurring about 2 5 July and adult activity ceasing by mid-August (Fig. 19* 20). Female catches were rare In black light traps* with 59 Pig. 16. Egg mass of H. manteo two days after ovlposltlon showing red ring characteristic of viable eggs (Mag. 5x). Fig. 17. Egg mass of H. manteo five days after ovlposltlon showing first lnstar head capsules (Mag. lOx). Pig. 18. Adult male of H. manteo showing typical color pattern of adults (ac tual size). 60 males outnumbering females by 33 >1 • monitoring nocturnal activity* T"o evenings were spent At Big Star lake during the evenings of 28, 2 9 July, 1975» black light catches were checked every 2 hours from 8 p.m. to 6 a.m. Adult moths were found In the 12 p.m., 2 a.m., and 4- a.m. checks, which Indi­ cated the activity of adults was nocturnal rather than cre­ puscular • Fecundity* Potential Upon emergence adult females were found to be provlgenlc, the abdomen being completely filled with eggs, labora­ tory females reared from prepupae collected In field samples were dissected to determine the potential number of eggs which could be laid by each female. The females were segregated by plot to determine If fecundity potential was dependent upon population density, I.e. prepupal density. Females were stored in 70% alcohol until the ovaries were dissected under a stereomlcroscope. The results of female dissections are presented In Table 14. The average number of eggs which could potentially be laid per female was 4-06 which Is in agreement with Wilson (1971) who reported that each female may lay as many as 5 0 0 eggs. There was no significant difference In fecundity poten­ tial of females from various plots. The females from Big Star lake which had the highest larval and prepupal densities did, however, produce the fewest number of eggs. Individual gravid females were monitored In the labo- 73 t Ntwoygo No. of Adults Big Star 21 1 11 21 July PIG. 19. 31 10 20 30 2 Aug Number of adults of H. manteo collected In black lights 1 Michigan, 197^. 107 103 75 t Newaygo Dublin Z 20 21 11 21 July FIG. 20. 31 20 10 30 2 Aug Number of adults of H. manteo collected In black lights * Michigan, 1975* 63 Table 14. Number of eggs dissected from gravid females of H. gggnteo, Michigan, 1974-75. Plot Location r No. Eaos/Female Newaygo 377.6 ± 136.3 N=15 Big Star lake 363.5 ± 1^1.9 N»23 Branch Foie 426.3 ± 168.3 N=12 Branch Mature 403.0 ± It8 Dublin 461.6 + 126 8I .5 Nsll 6b ratory to determine the number of viable eggs produced. Only 5 females were observed to lay their complete egg complement« but the average number of viable eggs produced was 388.t ± 86.71 which was closely correlated to that predicted by dis­ section. Fecundity mentioned above was determined under lab­ oratory conditions, and most likely would be reduced under field conditions because of premature adult mortality. Egg Laying Habits Wilson (1-971) based on descriptions provided by Pack­ ard (1895) reported that eggs of H. manteo were laid singly. Females in Michigan deposit eggs In large hemispherical masses (Fig. 17) on the underside of host foliage red or white oak. Despite careful observations no eggs were found on other tree species. To verify that egg masses were indeed those of H, manteo. larvae reared from eggs were sent to Dr. Franclemont, Cornell University* for Identification, and adults reared from egg masses were identified by John Newman, Michigan State University. as H. manteo.. In both cases larvae and adults were identified The earlier reports of eggs laid singly are be­ lieved to be due to mlsldentlflcation of larvae and adults. They may have been those of H. guttlvltta which does deposit eggs singly (Grimble and Newell, 1972b* Allen* 1973), During the course of this study, 120 separate egg masses were examined in the field. The egg masses ranged in size from 10-350 eggs with a mean of $b.5 ± bb»7 eggs. This suggested that each female could lay about eight egg masses 65 during her adult life# A blmodal distribution of egg mass sizes existed with a large group of egg masses of a hundred eggs or more and another group of 20-30 eggs per mass. This suggests that females lay 1 or 2 large masses in addition to several small egg masses• Development of Etebryos The eggs of Heterocampa manteo are light green (Fig. 1 6 ) when Initially laid. Within 2 days a reddish band (Fig. 1 6 ) is evident upon the upper surface of the viable eggs. The eggs are hemispherical in shape (Fig. 21) with a smooth surface. The eggs have been described as begin **about 8 mm in diameter, hemispherical, shining under high power, irregu­ larly hexagonally sculptured, the sculptures consisting of raised lines" (Packard, 1895). No evidence of hexagonal sculptures was observed on eggs in this study. The time for embryo development lasts ^ to 9 days de­ pending upon temperature. Embryos generally develop within 5 days at 24°C, 20°C, and 15°C| once the first Instar head capsule develops (Fig. 17) ecloslon occurs within 2b hours. The egg stage is short in duration. This confers a distinct evolutionary advantage since eggs are heavily parasitized by Trlchogranrma sp. and Telenomus sp. The short duration of the egg stage Increases the probability that eggs will not be parasitized, since searching time of the parasltolds is re­ duced. 66 Egg Parasitism Tiro species of Insects* Trio ho gramma sp. (Trl cho grammat ldae ) and Telenomus sp. (Scelionldae) were found to para­ sitize the eggs of H, manteo and the red-humped oakworm Symmerista canlcosta Franclemont. Actual specific Identifica­ tion of these parasltolds proved Impossible* Specimens were sent to personnel at both the U.S. National Museum and the Canadian National Collection. Tn both cases identification could not be made at specific levels. Percentage parasitism was extremely high during 197**as evidenced in Table 15* The majority of the egg masses were collected at ground level, while few egg masses from the up­ per canopy were examined due to the difficulty of sampling. The parasitism of egg masses from the upper canopy did not appear to be significantly different from those of the lower canopy. Allen (1,972) reported that egg parasitism of H. gut- tlvltta by Telenomus coelodasldls Ashm. and Trlchogramma mlRiley averaged 7 to Q% higher in the upper canopy of sugar maple and beech. Tlcehurst and Allen (1972) have re­ viewed the biology of T. coelodasldls Ashm., the egg paras1told of J|. guttlvltta. In New York the aggregate parasitism of H. guttlvltta by T, mlnutum and T. coelodasldls was **-0 to 78J6 (Allen* 1972). The extremely high levels of egg parasi­ tism by Trlchogramma sp. and Telenomus sp. in this study are therefore not unusual. Hanson et al. (1976) suggested that these parasltolds were responsible for the collapse of dam­ aging populations of the red-humped oakworm and orange-humped 67 Table 1 5 . Site Ess parasitism of H. manteo and S. canlcosta by Trl ohogrrmnma sp. and Telenomus sp.» Michigan, 1975. s.. canlcosta No. Masses TotaT Telenomus # Hasses # Eggs % Par. Newaygo 2 2 97.1* 9 8 68 9 2 .6 * 2 2 20 1033 9 1 .6 * 18 12 1*9 2910 95.***. 1*2 32 2 92 12 568 2 Big Star Dublin Branch Pole Branch Mature Parasitized Trlchogramma 100# H. manteo Newaygo Branch Pole 2 100 11 557 Branch Mature. 1 32 Big Star 29 2851 6 376 Dublin 90* 2 0 75.2* 5 9 100* l 1 8 bf 20 21 5 4 82.6* 68 mapleworm, Symmerlsta leucltys Franclemont. Kearby (1975b) has sugarssted that egg parasitism was responsible for the collapse of H. manteo populations In Missouri. In Michigan* extremely high levels of egg parasitism contributed to the collapse of populations. Where adequate numbers of eggs were examined, para­ sitism of S, canlcosta eggs was consistently higher than para­ sitism of H, manteo eggs. This phenomenon was explained by the egg laying habits of the 2 species. S. canlcosta lays Its eggs on a single horizontal plane and when egg masses were examined they were generally totally parasitized as all eggs were accessible to adult parasltolds. In contrast, H. manteo deposits Its eggs in large hemispherical masses (Fig. 1 6 ), eggs plied on top of each other. This proved significant In reducing egg parasitism since the Inner core of eggs In large masses escapes parasitism as It Is Inaccessible to the adult parasltolds. The Inner eggs hatch successfully, the outer eggs being parasitized. This phenomenon produces many egg masses which show a donut-shaped pattern as the larvae In the centers of the egg masses emerge. During the course of this study, most small egg masses were completely parasitized» whereas. In large egg masses ca. t.0% of the eggs escaped para­ sitism (see egg parasitism) because the Inner core of eggs was Inaccessible to the adult parasltolds. There appeared to be a strong selection pressure against production of small egg masses or egg masses laid as a single layer. In many Instances, egg masses were found after para­ sltolds or larvae had emerged. It was therefore important to 69 differentiate eggs from which parasltolds or larvae had emerged* Eggs from whtbh larvae successfully emerge are clear and generally broken In half. Parasitized eggs are always dark In color and the emergence holes left by the adult parasltolds determine which species has caused the para­ sitism* Trlchogramma sp. leaves a small smooth-edged circu­ lar exit hole (Fig. 21A) and on several occasions 2 to 3 adult parasltolds were noted to have emerged from a single egg. Telenomus sp. produces an elliptical emergence hole* about twice the size of Trlchogramma sp., about which strips of shredded chorion are found (Fig. 21B). In all cases Tele­ nomus sp. proved to be a solitary parasltold. Parasitism levels of eggs appeared to be about equally distributed between Trlchogramma sp. and Telenomus sp. ple parasitism was not observed. Multi­ The adults are apparently able to distinguish eggs previously parasitized by their own species or other parasltolds. After completing oviposition the females of both species would drag their ovipositors over the egg surface apparently "marking" the eggs. Before females oviposit Into the eggs they constantly tap the surface with their antennae apparently to detect previously parasitized sggs (Tlcehursb and Allen, 1973)* Adults of Telenomus farlal Lima can differentiate between parasitized and nonparasitized eggs (Rabinovich, 1970). In 197^* the total egg parasitism for all 5 plots was 85# yet this figure dropped to in 1975 (Table 1 6 ). dramatic decline in parasitism was evident in all plots. This The decline was apparently related to overwintering success of the 70 Fig. 21. Parasitized eggs of H. marteo Illustrating adult exit holes. A. Tr1chogramma sp. B. Telenomus sp. 71 parasltold populations# Those parasltolds probably overwin­ ter In alternate hosts although Kearby (1973) reported that Telenomus sp. parasitizing H. manteo overwintered as adults under the loose bark of trees. An unknown species of Tele­ nomus does overwinter In the eggs of the white marked tussock moth Hemerocampa leucostlgma (J.E. Smith) (Baker, 1972), a moth which was relatively common In the vicinity of the re­ search plots. Members of the genus Trl cho gramma In general have a very broad host ranpe which suggests the parasltold of H, manteo may overwinter In some alternate host. Massive population outbreaks of H. manteo are often associated with population Increases of the red-humped oak­ worm, S, canlcosta. and the orange striped oak-worm, Anlsota sanatoria (J.E. Smith) (Beach, 1972). The life cycles of these are similar to that of H. manteo with the latter 2 spe­ cies overwintering In the duff as pupae. There appears to be no prolonged diapause In these 2 species since pupae were never found In August soli samples. The egg parasltolds which attacked H. manteo eggs also caused high levels of parasitism In eggs of A. sanatoria and S. canlcosta. Egg parasitism ap­ pears to have been a major factor In the decline of popula­ tions of these defoliators. Removal of this parasltold pres­ sure may permit population Increases. Generally, when 1 spec­ ies of fall defoliator Increases dramatically other species also become far more abundant (Beach, 1972). The parasltold decline could be caused by a reduction In numbers of overwin­ tering hosts (I.e. Hemerocampa sp. Malacosoma sp.). The popu­ lations of the fall defoliators may In fact be Interrelated 72 Table 16 . Egg parasitism of H. manteo by Trlchogrammft sp. and Telenomus sp.* Michigan, 1975• Site Number of Examlrmd Percent Parasitism .* Newaygo 2*253 36 1 Branch Pole 1 .3^8 31 .8* Branch Mature Big Star Dublin None 52U 1*711 38.7* 23 .8* 73 to population levels of spring defoliators. During the outbreak of 1973* 13 masses from Big Star Lake were collected by Thomas Ellis, Michigan State Uni­ versity. Twelve of the egg masses were parasitized by Tele­ nomus sp. with a combined parasitism of 80#. sp. parasitized the other egg mass with parasitized. Trlchogramma of the eggs being In 1973* when major defoliation occurred, egg parasltolds, particularly Telenomus sp., were extremely abun­ dant, causing heavy mortality. These levels of parasitism were not sufficient to prevent defoliation. larval Instars Packard (1 8 9 5 ) described 5 Instars of H. manteo but these descriptions did not accurately match the Instars found in Michigan. Determination of lnstars could not be based on larval coloration nor head capsule setal patterns because of extreme variability. Determination was therefore based on head capsule size (Surgeoner and Wallner, 1975) (see Appendix). A description of the larval lnstars Is presented In Table 1? and figures of the various Instar head capsules are presented In Fig. 22. Head capsule measurements showed 5 lnstars which was confirmed by rearing larvae from egg to adult In the lab­ oratory. Other species of the genus Heterocamoa have also been reported to develop through 5 lnstars (Klots, 19^7 1 Allen and Grlmble, 1970). First lnstars of H. manteo do not possess large prothoraclc horns which is characteristic of other spec­ ies of Heterocamna (Packard, I8 9 5 ). Table 17. Characteristics of lnstars of H. manteo larvae In Michigan, Character 1st Instar 2nd Instar 3rd Instar 4-th Instar 5th Instar HEAD CAPSULE WIDTH (mm) .^0-.55 mm .75-.95 mm 1 .20-1 ,^5 mm 1 .65-2.25 mm 3.05-3.50 mm 3-6 mm 5-9 mm 8-15 mm 15-27 mm 28-37 mm BODY LENGTH DORSAL CHALAZAE pro thorax only HEAD CAPSULE COLORATION Dark discolorations at the base of P2 * Pj, L, 0 and A 3 setae discoloration uniting AP setae forming inverted V, prothorax, abdominal segments 1 and 8 Dark area forming between P2 and L setae. absent Strong black band Joining P2 » L, A 3 and Ai setae. 75 Pig. 22. (a) First instar head capsule of H. manteo (b) Second instar head capsule of H. manteo (c) Third instar head capsule of H. manteo (d) Top row (left) Normal fourth instar head capsule of H. manteo (right) Normal fifth Instar head capsule of H. wflntftQ Bottom row (left) Fourth instar head cap­ sule of H, manteo parasitized by D. bethunei (right) Fifth instar head capsule of H. nmwtfto parasitized by D. bethunei (e) Larva of D. bethunei in overwintering pre­ pupa Of H. TnantftQ (f) Pupa of D. bethunei with H. manteo prepupa in overwinterlng cell 3 mm 77 Larval Feeding Habits First lnstars of H. manteo are gregarious remaining close to the egg mass from which they hatch. They skeleton­ ize the lower surface of the oak leaves on which the egg mas­ ses are found* Larvae feed In a straight line, all larvae touching each other and facing the same direction* This re­ sults In large areas of total skeletonization, rather than random patches skeletonized on the lower leaf surface. The presumed advantage of this behavior Is to prevent fecal con­ tamination of the host foliage* when larvae feed hanging up­ side down the feces always drops away from the leaf surface. In 1975* when parasitism of eggs was reduced the leaves on which these first lnstars fed were readily apparent In the canopy. These "flag" leaves w-re a dried brown color and contrasted markedly with the green canopy. The coloration of the early lnstars Is also variable, the overall body color being a light green with the dorsum generally having some red coloration. A common color type found In the canopy was for the larva to be overall green with a red stripe running along the dorsum of the thoracic and first abdominal segments. When disturbed the larvae readily drop from the leaves, although early lnstars generally remain attached by a silken thread. If the early lnstars were to drop to the forest floor a large proportion would likely never reach the foliage again. By contrast, fourth and fifth lnstars drop directly to the forest floor, rarely spinning a silken strand. They are mobile and recllmb tree boles to 78 seek out new leaves. The larvae of £. manteo are not consi­ dered to he highly dispersive. There Is no evidence of de­ liberate migration, or wind dispersion as In the gypsy moth (Leonard, 1971). In 197** and 1975 when larval populations were scarce, fifth lnstars could often be found In the vicin­ ity of leaves where eggs had been laid. If the foliage sup­ ply Is ample the larvae do not disperse, but remain feeding in the same area. It appeared from general field observation that the larval population in the lower canopies and on small shrub-sized oaks did Increase as the season progressed. This was apparently due to larvae dropping from the higher leaves when disturbed by predators, wind, etc. and recllmblng to resume feeding. Prior to molting larvae attach themselves to the un­ derside of the leaf by producing a silken mat. These larvae are extremely vulnerable to predation as they can not drop from the leaves when threatened. In nature, the time spent attached to the leaves was about 2k hours for each molting period. Prom dissections of field-collected larvae I discov­ ered that most Individuals consume their exuvlum after each ecdysls. The second and later lnstars are solitary and feed along the leaf margins. jor veins Is consumed. The entire leaf except for a few ma­ As mentioned the later lnstars are often found feeding In the vicinity of old egg massest but 2 late Instar individuals were never found feeding on the same leaf. When collecting larvae In the field, If large numbers were placed In collection containers mortality of a 79 large percentage ensued within 2 to 3 hours. In these con­ tainers the odor of formic acid (see defensive secretions) was potent. Apparently the large lnstars were spraying for­ mic acid when disturbed by other larvae and thereby contamin­ ating the food supply and possibly killing larvae dlredtly, since mortality after 2 hours was high. The solitary feeding habits of the later lnstars appeared necessary to prevent mor­ tality of sibling larvae by spraying each other. laboratory Rearing The rearing of Insects In the laboratory Is desirous since a great deal of basic biology may be determined from laboratory culture. In addition, laboratory rearing tech­ niques are often necessary to create populations for control projects. I.e. sterile male, pheromone studies, parasltold production, A technique was therefore developed to rear popu­ lations of H. manteo In the laboratory. The mortality associ­ ated with this rearing procedure was extremely low (5%), The disadvantages of the procedure are that it Is time consuming and fresh oak foliage Is required. In the spring months prepupae were collected from the research plots by removing quarter meters of soli and litter. These prepupae were then placed Individually Into 50 ml Jars and maintained at a constant 2**-°C until pupation occurred. Pupae were kept In Jars until 5 days before expected emergence. Five male and 5 female pupae expected to emerge on the same date were placed in moist (18£) soil and litter lining the 80 bottom of a ^5 cm x 30 cm £ 30 cm screened cage. were burled In the litter to a depth of 3 cm. The pupae The cage was provided with a cotton swab moistened with a 50*50 molasseswater solution and several sprigs of white oak foliage were placed In the cage. Shortly after emergence adults mated and egg masses were laid on the oak foliage, but more commonly the sides of the cage. The eggs were left In the cages until they were determined to be viable by the development of a red ring (Pig. 16 ). Eggs were then removed from the cage and placed In 50 mm x 20 mm petrl plates lined with moist paper totalling. The eggs were held at 24°C until ecloslon. The time necessary for ecloslon was generally 3 days after the red ring was ob­ served. When larvae hatched they were placed Individually Into 50 mm x 20 ram petrl dishes for lnstars 1 . 2 and 3« The larvae were provided with fresh oak foliage each day and reared at a variety of temperatures and photoperlods. The effect of temperature on the stadia Is presented In Table 18, The duration of each Instar was related to temperature, es­ pecially at temperatures below 20°C, The minimal temperature for development was approximated 12.5°C. First Instar larvae kept at this temperature remained alive for 22 days. but never molted to the second Instar. ably longer than other stadia. The fifth stadium was consider­ When larvae developed to the fourth Instar they were placed Individually into one pint plas­ tic containers where they were fed dally with red or white oak foliage. Up to 6 larvae could be reared In each container, but the mortality associated with rearing greater numbers Table 18. Effect of temperature on stadia of H. manteo. Instar and Duration (days) Temperature 1st 2nd 4th 5th Total 5.11 ± .89 10.65 ± 2.55 25.67 ± 2.55 3.56 ± .87 3.01 3.30 ± .54 23.9°CB 3.55 ± .91 3.38 + .88 3.94 ± .89 5.30 ± .89 10.32 ± 2.22 26.94 ± 2.65 21.1°^ 5.09 ± .79 4.17 + .81 6.21 ± .66 6.84 ±1.73 12.04 ± 1.73 3^.55 + 2.50 1 5.5°^ 7.43 + 3.56 7.77 ±1.33 7.90 ±1.33 10.40 +1.82 18.47 ± 4.30 51.94 + 5.11 -H 26.7°^ 0 00• 3rd A - complete darkness B - 16-hour llprht, 8-hour dark photoperiod 82 together was significant. When larvae were collected In the field from oak foliage and subsequently reared In the laboratory significant mortality occurred because of disease. Larvae Infected were lethargic and excreted extremely watery feces. The causal organisms of the disease was not determined although It was not of fungal origin. When the disease developed In the lab­ oratory cultures* Infected larvae were removed and the rear­ ing chambers and working area* used when feeding larvae* sterilized. A strong bleach was used for sterilization. The disease did not appear to cause significant mortality In na­ ture, but was highly contagious In confined rearing chambers. The disease never originated in those colonies Initiated from overwintering prepupae. Parasltolds of Larvae Seven Insect species were found to parasitize larvae of H. mA.nfcao. These species included 3 tachlnlds* 3 lchneumo- nlds and 1 braconld. The Intensity of larval parasitism In 197^ was approximately 85#, with parasitism levels reaching 90# In the population epicenter at Big Star Lake. A brief description of the parasltolds and their relative significance Is presented. Tachlnldae Wlnthemla datanae (Townsend), larvae of H. manteo were collected In all plots either by searching for the cater­ pillars on the foliage or by the use of pyrethrum samples 83 (see larval sampling). Parasitism rates In 19?^ by V. da ta­ nas were less than 10# (Table 19) whereas In 1975 parasitism levels were less than 3% (Table 20). The adult parasltold attacks the fourth and fifth Instar larvae, depositing eggs directly onto the host* One or 2 eggs are generally laid, often on the dorsum of the thorax directly behind the hostfs head. This parasltold Is not host specific and has been re­ ported to attack a wide variety of large Lepldoptera (Gulmaraes, 1972). Canada. It Is found throughout the U.S., Mexico and It overwinters as a larva In the prepupa of H. manteo. Pupation occurs the following spring during the period of normal pupation by H. manteo. Despite the fact that 2 eggs are often noted on the lntegumen of larvae only 1 parasltold per host emerges. Archvtas atterlmus (Rob,-Desv.). Parasitism by this species was less than 5# in 197^ (Table 19) and 1975 (Table 20) and It was therefore considered a minor parasltold. The adults larvlposlt on the leaf foliage, the larvae being at­ tached to the substrate. They wait for a suitable host to pass, at which time the parasltold larvae attach themselves to the Integument of the host and burrow Into the larva* The reader Is referred to Ravlln (1975) for a complete review of the taxonomy and biology of this genus. This species at­ tacks a wide variety of Lepldoptera larvae Including many species of Noctuldae and Notodontldae. out the continental U.S.A. It Is found through­ Those which parasitized H. manteo overwinter as larva within the prepupa. When host pupation occurs the A. atterlmus larva kills the host and punctures Table 19, Percent parasitism of H. manteo larvae, Michigan, 197^. % Sample Location Newaygo N % larvae Non-parasitlzed Parasitized D. bethunei % Parasitized Phobocampe sp, % % Parasitized Parasitized P. schlzurae by Tachlnlds lk% 20% 18% 35% lk% kl% 15% 10* 20% 20% 20% ko% 10 20? 20% 30% 10% 20% 1^ 2k% 18? 18% 6% kl% 100 26% Big Star 100 10% Branch Mature 10 Branch Pole Dublin 85 Table 20. Site Percent parasitism of H. manteo larvae, Michigan, 1975* Number Overall of Para­ H. manteo sitism _____________ Parasltold D. bethunei L. schlzurae Othei 253 13.8£ 7.5% 5.1% Big Star 67 22. h% 18.6% v. 00• Branch Mature 60 18.3# 9.2% 9.2% —— Branch Pole 232 11.6% b.3% b.3% 3.0% Dublin 59^ 1 6 .6% 5.0% 10.0% 1 .6% Newaygo 1.2% — 86 the Integument at the caudal end of the pupa. The puparlum Is then formed within the remains of the pupa and the adults emerge through the caudal openings made earlier by the larvae. Only 1 parasltold per host Is produced. Lesoesla schlzurae (Townsend). This parasltold caused less than 5% parasitism (Table 19) In 1974 and In 1975 parasitism rates were similarly about 5% (Table 20). adults larviposit In chorion onto the host. The A wide variety of lepldopterous larvae throughout North America have been reported parasitized by this species. The parasltold over­ winters as an early Instar within the host and exits from the prepupa the following spring when normal pupation occurs. It was discovered, by rearing prepupae collected In August that this parasltold remains in a prolonged diapause with the host. Evidently, If the prepupae of £. manteo remain In pr&longed diapause so does the early Instar larvae of L. sohlzurae. Ichneumonl dae Barvlypa sp. Parasitism rates by this lchneumonid were less than 1% In 1974 and 1975* Members of this genus oviposit Into a variety of lepldopterous larvae. The parasl­ told overwinters in H. manteo prepupae and emerges from the pupa. It Is considered a minor parasltold of H, manteo. Phobocampe sp. Specimens of this parasltold were sent to the Canadian National Collection and Dr. H. Townes, American Entomological Institute. was possible. No species Identification Parasitism rates ranged from 14 to 30% in 1974, but were less than 1% In 1975* This insect therefore appears to be of major importance in reducing larval populations In 87 years after defoliation, tout is protoatoly poorly adapted for long term population control, particularly at low densities. The genus Photoocfflwp^ Is reported to attack a wide variety of lepldopterous hosts Including memtoers of the genera Hemerocampa. and Schlzura (Meuesbeck et al., 1951)* The Insect is a solitary endoparasltold which apparently ovi­ posits into early instar larvae. The larva of the parasltold emerges from the early fifth instar and pupates on the foliage toy the larval remains. The pupa often drops to the forest floor soon afteri If not, then does so when leaves drop in autumn. floor. It therefore overwinters as a pupa on the forest This Insect destroys the host In the early fifth In­ star, and therefore helps to reduce levels of defoliation since over of the foliage Is consumed in the fifth stadium. D1radons toethunel (Cresson). This insect was con­ sidered the major larval parasltold with parasitism rates ranging from 18£ to 35^ in 197** (Table 19) and from 5% to 19# In 1975 (Table 20). It is apparently host specific to H. mantao and found throughout the eastern and central United States. The parasltold oviposits Into first and second In­ star larvae and overwinters as an early instar (Pig. 22e) within the prepupae. Adult activity of D. toethunel as mea­ sured toy number of adults collected In Malaise traps is pre­ sented In Pig. 23. Adult activity corresponded to the time when peak numbers of first and second lnstars of H. manteo were found. In the laboratory adult females readily attack first and second lnstars and viable colonies of D. toethunel can be reared in the laboratory by maintaining host larvae. No. of D. bethunei PIG, 23. 15 15 15 (5 Moy June July August Sept. Numbers of adult S. bethunei from Malaise traps, Big Star lake, Michigan, 197^. 89 The prepupae parasitized by fi. bethunel can be segregated on the basis of head capsule size (Pig. 23d) (Surgeoner and Wallner, 1975) and reduced body size (Fig. 3)« In the au­ tumn of 1973 and the sprinpr of 1 9 7 ** ca. \&% of the prepupae were parasitized by D. bethunel. In samples taken In mid- August of 197** 1356 of the prepupae remained parasitized. This parasltold had adapted to the prolonged diapause of the host. The parasltold larva remains In the host and does not develop to maturity and pupate (Pig. 2 3 f) unless It receives some stimulus from the host. This stimulus Is probably a hormone secreted by the prepupae which Initiates pupation, larvae parasitized by D. bethunel consume 6 3 # less foliage than non-parasltlzed larvae thus accounting for the reduced body and head capsule size. The parasltold Is of benefit, therefore, In not only reducing larval numbers but also In reducing the amount of foliage consumed. Braconldae Protomlcroplltls schlzurae (Meus,). This Insect was classified as Mlcrogaster schlzurae by Nuesebeck et al. (1.951)» tJUt was identified as Protomlcroplltls schlzurae (Mues.) by W.R. Mason, Research Branch, Blosystematlcs Re­ search Institute, Ottawa, Canada. The reported hosts Include several other genera of Notodontldae. told producing 1 to 1** larvae per host. This Is a endoparasl­ The mean number of parasltolds per host was 6 at Big Star lake and Newaygo In 197**. Parasitism rates In 197** ranged from 6% to U>1% and the Insect was considered a major mortality agent of H, manteo. In 1975# however, the parasitism rates were less than The 90 parasltold attacks the early Instars and remains In the host until the prepupae drops to the soli* Once the prepupa forms the overwinterIn* cell, It Is killed, the parasltold larvae exit, and spin silken cocoons In which they overwinter. The species Is synchronized to a one-year life cycle since adults emerved, from all pupae examined, the following: sprin*. Multiple Parasitism In 197** approximately of the larvae dissected were parasitized by more than 1 parasltold species. No 1 species, however, was consistently found In multiple-parasltlzed larvae. The hl*h level of multiple parasitism was ap­ parently a consequence of extreme competition for H. manteo larvae since were parasitized. Pew hosts existed which had not previously been parasitized. In 1975* "by contrast, when only 1 5 ^ of the larvae were parasltlzdd, less than IJf of the larvae were parasitized by more than 1 species. Hyperparas1t1sm An ichneumonld hyperparasltold Mesochorus vlttator (Zetterstedt) was reared from pupae of Phobocampe sp. and Protomlcroplltls schlzurae. The specimens were Identified by Dr. C. Loan, Research Branch, Blosystematlcs Research Insti­ tute, Ottawa, Canada. Accurate estimates of the level of hyperparasltlsra proved Impossible because M. vlttator larvae develop within parasltold larvae. Thus, dissection of H. manteo larvae does not reveal hyperparasltold larvae since they remain within the larvae of the primary parasltold. The 2 species which M. vlttator attacked were extremely rare In 1975. M. vlttator may have contributed to the decline of 91 these parasltold populations* Adults of Mesochorus sp. were never reared from other parasltold species although species In the genus commonly parasitize a wide variety of parasltolds (Muesebeck et aL» 1951)* Parasitism as Related to Prolonged Diapause The percentage parasitism of H. manteo larvae In 1974 (84)0 and 1975 (l6/0 Is presented In Pig. 24. It was evident that not only was there a dramatic decline In parasitism rates9 but that only 2 parasltold species caused any signifi­ cant levels of parasitism in 1975* These 2 parasitolds were D1 radons bethunel (Cress) and Lespesla schlzurae (Townsend), both of which could successfully remain in prolonged diapause within their hosts. For example» when prepupae In prolonged diapause were collected In mid-August only these 2 parasitolds were reared from the prepupae. Other parasitolds are apparent­ ly synchronized to a one-year life cycle. The majority of the prepupae which overwintered in 1974-75 were derived from the 1973 defoliation, as larval popu­ lations were extremely low In 197*** Only D. bethunel and P. schlzurae were able to successfully maintain themselves In the dlapauslng population while the other parasitolds were forced to overwinter on the small numbers of larvae developing In 1974. In the spring of 1975 the other parasitolds were few In number while the 2 parasitolds synchronized to the prepupae emerging after 2 years did remain In numbers sufficient to cause measurable parasitism. § 50 PARASITISM 40 - Q. bethunel [TT| - Phobocompe $p. ^ - B echlzuroe E “ L. Khteuroe f" | - Other tochmids 30 \0 20 (V) 10 zM 1974 FlO. 2k* 1975 Percent parasitism of H. manteo larvae, Michigan, 197^-75* 93 The parasitolds synchronized to a one-year life cycle contributed to the collapse of the defoliating population. These parasitolds» because of their poor synchronization to the two-year life cycle, do not cause high levels of parasi­ tism In low density populations. The two parasitolds, D. bethunel and L. schlzurae which may remain 2 years or longer within their host should be considered most suited for long term population regulation at low densities. ftredatlon No Insects were observed to feed on the eggs of H. manteo,, but a number of species were observed to feed on the larvae. The early lnstars appeared most susceptible to pre­ dation as later lnstars were either too large or able to bet­ ter defend themselves with formic acid secretions. Species of the following families were observed to feed on early ln­ stars of H. manteo t 1) Adults and nymphs of Reduvlldae 2) Adults of the family Clerldae 3) Adults and nymphs of Fentatomldae U) larvae of Chrysopldae 5) Adults and larvae of Carabldae (In particular, Calosoma sp. and Plnacodera s p . )* Only the large predaceous ground beetles Calosoma scrutator Fab. and C. frlgldum Kirby were observed to feed on late lnstars. Hooker (1 9 0 8 ) reported that In Texas adults of Calosoma were found to feed on larvae of H. manteo. The 9b Impact of insect predators on populations of H. manteo was not determined. I believe that perhaps 15% to 2$% of the early lnstars are destroyed by the predators and less than $% of the later lnstars. These percentages were based on general field observations over a three-year period and not on ex­ perimental evidence. There was little evidence of vertebrate predation of H. manteo larvae. This may have been due to the large quan­ tities of formic acid found In the defense gland of the larvae and prepupae. In the autumn and spring of 197** and 1975 acres of the forest floor showed signs of digging and scratching by the wild turkey Meleagrls gallopavo Vlelllot. Consequently, In March of 1975* 12 turkeys shot during the local hunting season In lake County were dissected. These birds were brought to the Department of Natural Resources check In center at Baldwin, Michigan. With the permission of the hunters, the crops and gizzards were removed and their contents examined under a stereomlcroscope. In 1 individual the remains of a single prepupae was diseovered. The turkeys did feed on the prepupae, but the numbers consumed was believed minimal. Sampling of Larval Populations A sample technique was required to detexmlne relative larval densities and to monitor lnstar development. The can­ opy of entire trees was sprayed to determine the number of larvae per tree and the age distribution of larvae. By spray­ ing complete trees one avoided the difficulty of collecting 95 caterpillars In the upper canopy. The undergrowth beneath trees which were to be sampled was removed with the aid of 2 an axe. A 225 ft tarp was then placed beneath the canopy. The tarp was constructed in two equal sections which snapped together along a mid-seam. Pour stakes elevated the corners of the tarp and created a concave surface. Thus all Insects which dropped to the tarp remained on the surface and were collected and stored In pint Jars containing 70% ethyl alco­ hol. The trees were sprayed from atop a 12.15 ® ladder mounted on a truck. The trees were sprayed using 4,5 liters of spray solution containing .141 of 0.5^ (AX) pyrethrum ap­ plied using a KWH S- 6 6 m knapsack sprayer. The canopy of the tree was completely covered with spray solution using this technique. The tarp was left In place beneath the trees for £ hour after spraying at which time the Insect fauna were col­ lected. Wallner (1971) has described a similar spray tech­ nique for evaluating insecticide efficacy. The larvae collected using this technique were clas­ sified as to Instar and dissected to determine the parasitism levels of the population. An extremely high variability be­ tween trees existed and the technique could not be used to accurately define densities per plot. The method did show the relative difference of densities between years. The popula­ tion at Big Star Lake In August 1973 averaged 1083 ± 516.4 larvae per treet by contrast the populations in August of 1974 and *75 were 1 6 . 3 ± 11. .7 and 19.8 + 13*5 larvae per tree re­ spectively. This sampling technique did show the massive 96 collapse of the population between 1973 end 197**-. By deter­ mining the mean larval lnstar In each sample It was also possible to monitor larval development through the growing season (Fig. 25). Foliage ConsunrotIon by H. manteo Larvae Overwintering densities of prepupae do not provide an accurate method for predicting defoliation because a sig­ nificant portion of the prepupae may remain longer than 1 year in the litter. It is also difficult to predict defoli­ ation by egg mass densities since egg parasitism by the para­ sitic hymenoptera TrlchogrwiHnfr sp. and Telenomus sp. has pro­ ven to be as high as 90% In Michigan. The density of early lnstars on foliage Is the most reliable parameter for pre­ dicting defoliations. Estimation of defoliation Is essential when formulating control strategies and must be based upon foliage consumption by Individual larvae. A study was there­ fore conducted to determine the amount of foliage consumed by various lnstars and the effects of host plant* temperature and parasitism upon foliage consumption were also Investigated. Instars were reared according to previously outlined laboratory rearing procedures* The foliage used for feeding trials was taken from several red and white oaks situated on the Michigan State campus. The environmental conditions used in this study are presented In Table 21. The foliage fed to Individual larvae was changed dally and before adding new foliage the leaf outline was traced on 2 5 mm grid paper. LARVAL INSTAR 97 y « .0 6 6 2 9 * 1.17 MEAN r2 * .9 5 JULY PIG. 25. AUG. SEPT. Mean larval Instar of H. manteo In pyrethrum spray samples. Big Star Lake, Michigan* 1975* Table 21. Rearing regimes for H. manteo used in study. No. of larvae Reared Temperature a. alba Q. borealis Parasitized R.H. Photoperlod 26.?°C 16 12 5 60% complete darkness 23.9°C 25 25 Ik 60% 16-hour light 21,2°C 11 20 3 60% complete darkness 15.5°C 11 10 0 60% complete darkness 99 When foliage was removed the outline of the remaining leaf area was subtracted from that of the original and dally con­ sumption for each Instar was tabulated. Small fra&nents of foliage eaten from leaves> but not consumed were considered part of total consumption. When predicting defoliation these fragments represent leaf area which is removed from the tree thus Increasing the level of defoliation. Approximately lOjtf of the foliage consisted of this type which Is In close agree­ ment for that determined for the gypsy moth P. dlsoar (Brah­ man, 1975)* larvae used to Investigate the effect of parasitism were placed as first lnstars Into cm x 3 0 cm cages along with 2 male and 2 female D 1 radons bethunel (Cress). After 3 days larvae were removed and reared In the same manner as nonparasltlzed larvae. The foliage consumption of first lnstars psorasltlzed by p. bethunel was assumed to be that of nonparasltlzed larvae, since eggs of D. bethunel do not hatch until larvae reach second Instar. Parasitism was confirmed by reduced head capsule size of fifth lnstars (Surgeoner and Wallner, 1975). The effect of host plant on foliage consumption Is presented In Table 22. Analysis of variance showed that there appeared to be no significant difference in consumption of larvae reared at 21°C, 24°C and 27°C| because of this, larvae reared at these temperatures were pooled for this comparison. There was no significant difference In total foliage consumed whether northern red oak or white oak| early lnstars consumed more white than red oak foliage which Is not surprising since Table 22. Consumption of foliage from two Quercus sp. by lnstars of H. manteo. 2 Total Consumption (cm )/Instar _______________ 1st________2rA_________ 2rd___________frth .52 + .17 1.55 ±**0 6.86 ± 1.79 36.13 ± 9-55 Q. alba .70 ± .29 1.88 ±.53 9.31 ± 3.08 ^ 5.09 ± 12.0if 288.3 ± 65.03 333.* ± 67.8^ 277.5 ± ^0.87 33^.5 ± ^2.51 100 Q. borealis 5th_________ Total 101 white oak is the preferred host (Wilson, 1971). The effect of temperature on foliage consumption Is presented In Iteble 2 3 . Because foliage type did not signifi­ cantly affect consumption, larvae reared on white oak or red oak were pooled for this analysis. Wo significant difference occurred In foliage consumption at temperatures of 21°C, 24°C or 27°C| however, larvae reared at 15°C consumed significant­ ly (P. <.05) less (b2,6%) foliage. Raham (1970) showed that larvae of Pierls rauae con­ sumed significantly more food when reared at 2 2 ,5 ° as compared to 2^.3°C. This was apparently due to the Increased duration of feeding. The fifth stadium of H. manteo was significantly longer (P<.05) at 1 5 °C than other temperatures (Table 1 8 ), but dally consumption was 73 % hower than that of higher tem­ peratures. larvae reared at 1$°C were capable of normal de­ velopment although adults were reduced In size. Mukerjl and Guppy (1970) have shown that when rate of food Intake Is low In Pseudaletla unlouncta (Haw.) there was a reduction In fe­ cundity. Preliminary studies Indicated that fecundity of H. manteo reared at 1 5 °C was lower than other temperatures. The effect of parasitism by D, bethunel on larval con­ sumption Is presented In Table 2Jf. Parasitized larvae con­ sumed 61.3^ less foliage than non-parasitized (P< .0 1 ). This reduced consumption Is correlated with lower weight and smal­ ler head capsule size (Surgeoner and Wallner, 1975). Predicting Defoliation The ultimate objective of the feeding trials was to 102 devise a method to predict defoliation* ods are presented. Two possible meth­ The first Involves the amount of foliage consumed by first lnstars. Adult females deposit eggs In large clusters of Jt^JOO placed on the underside of oak leaves. The first lnstars feed gregariously, skeletonizing the lower leaf surface. Leaves, from which non-parasltlzed egg masses have developed, are readily apparent since the lower leaf surface Is completely skeletonized. These "flag" leaves may be seen 2 0 m Into the canopy and verification of H. manteo feeding is made by the presence of hatched eggs which adhere to the leaf. From the data provided In Table 22 first Instar feeding on white oak represented .0 0 2 1 of the total consumptioni whereas on red oak first Instar feeding represented .0015 of the total consumption. It Is possible to predict defoliation by the level of first Instar skeletoni­ zation. If 0,1% of the white oak foliage Is skeletonized the 1^ expected defoliation would be or **7*6% whereas If ,1% of the red oak foliage Is skeletonized expected defoliation 1fOUld ** or 66%, This method for predicting defolia­ tion assumes that all first lnstars survive to complete de­ velopment and that all first Instar skeletonization Is de­ tectable • Over 85% of the foliage consumed by H. manteo is by fifth lnstars. It Is possible to predict the ultimate levels of defoliation based on relative density of larvae/leaves. A In the research areas the total foliage consumed (i.e. 33^ cm ) was converted to number of leaves consumed. Based on JO leaf Table 23. Effect of temperature on consumption of oak foliage by lnstars of H. manteo. 2 Temperature and foliar consumed (cm )/Instar 21°C 2^°C 27°C Instar _ _15°C. 1 *59 t .28 .52 ± .16 .63 ± .23 .66 ± .22 2 1.89 ± .67 1.66 i .*»6 1.73 ± .**7 1.55 ± .61 3 9.29 ± 3.**9 7.58 ± 2.19 7.**5 ± 2.19 6.60 ± ^.78 ± 11.66 37.*H ± 10.07 *M),25 + 12.75 30.93 ± 11.58 286.59 ± ^.70 270.70 ± *1-7.09 297.06 ± *f7.22 153.03 ± ^7.30* 3^3.1 *1-6.57 317.8 3**7.1 192.53 ± 53.3* 5 Total ± ♦Significant at the .05 level. ± *1-8.05 ±52.89 1.5 Table 24. Consumption of oak foliage by non-parasItlzed larvae of H. manteo and those parasitized by D. bethunel« o foliage consumed (cm )/Instar 1st 2nd 3rd 1.76 ± .53 8.11 ± 2.62 1*0.8 ± 11.94 1.36 ± .38 5.25 + I .87 19.86 ± 4.85* 103.16 + 35.83* 4th 5th Total .58 ± .22 284.71* ± 1*8.73 335.96 + 49.32 Parasitized .58 + .22 ♦Significant at the .01 level from non-parasltlzed larvae. 130.79 ± 36.83* 1701 Non-parasltlzed 105 samples the average white oak was 48.4 cm 2 2 2 ± 7.7 cm 2 average red oak leaf was 64,8 cm f 1 6 . 2 cm , and the In these stud­ ies the "average" larva consumed 5 * 1 6 red oak or 6 , 9 1 white oak leaves. By taking branch samples from various areas of the canopy It Is possible to determine the relative density of larvae per leaf, IT early lnstar densities average 1 lar­ va / 1 0 leaves one would expect approximately 5 0 # defoliation whereas If densities were 1 larva/ 5 leaves one would expect approximately 100# defoliation. Branch samples taken during mid-August should be used to predict defoliation. The mean larval lnstar during this time from past studies was ca, 3 * If larval densities are greater than 1 per 10 leaves one would expect noticeable defoliation. will vary between sites and years, The "average” oak leaf When predicting defolia­ tion the "average" leaf for each tree species should be asses­ sed for each location. In 1974, ca. 40# of the larval population was para­ sitized by D. bethunel. High parasitism levels reduce the expected degree of defoliation. larvae parasitized by p. bethunel consume ca. 6 0 # less foliage than normal larvae. Thus If parasitism levels were 40# one would expect a 40 x . 6 » 24# reduction In expected defoliation. Control Philosophy and Insecticide Trials This Insect Is considered a recreational and urban rather than forest pest. Control of vast acreages of forest land is not warranted because the Insect has little impact on 106 the trees and defoliation rarely lasts more than 1 year* In Michigan, consecutive defoliation does not occur because of large numbers of prepupae remaining In prolonged diapause and a complex of parasitolds and predators attacking the egg and larval populations. The insect has not caused consecutive defoliations in other regions of the U.S. as well, with lit­ tle tree mortality ever reported. A large scale spray operation may In fact perpetuate continuous defoliation. In any year a relatively large por­ tion of the population remains In prolonged diapause In the soil. This reservoir population Is considered non-susceptlble to foliar application of insecticides because prepupae do not feed and are burled ca* 8 cm into the soli and litter. By contrast, most predator and parasltold populations will be re­ duced by Insecticide application. from* The reduction would result 1 ) direct insecticide mortality, 2 ) reduction of host populations and 3 ) possible reduction of alternate hosts. Widespread application of Insecticides will therefore create a situation In which a large reservoir population emerges in the year following spray application with few natu­ ral controlling agents to supress this population. Star Lake in 197^ ca. 2,5 prepupae/.25m and soil. At Big remained in the litter This population was sufficient to again initiate defoliation, particularly If predator and parasltold popula­ tions were reduced by spraying. In areas where human Interaction and usage of oak for­ ests are Intense i.e. urban areas, cottage sites, heavy use recreational sites, and along roadsides, spraying may be 107 Justified. Individual home and cottage owners, and resort owners require some form of control to protect their property from defoliation and large numbers of larvae. The acreages involved in spray programs of this type will be small when compared to the overall extent of the defoliation. Conse­ quently, in years subsequent to application of insecticides predators and paras itoids may disperse into sprayed areas from adjacent nonsprayed areas. At the present time, no insecticides are registered for control of the variable oakleaf caterpillar. In 197** several insecticides were evaluated for control. The Insec­ ticides tested and rates of application are listed in Table 25. Table 2 5 . Insecticides and dosages applied for control of H. manteo. Michigan, 197**. Insecticide 80# SevlnRW.P. Treatment application rates/ 1 0 0 gals, water 1.5 lbs. 39# OyloxR L.S. 2 pts. 22.7# ZectranR E.C. 2 pts. 1.3 lb/gal OrtheneR .75 lb/AI .6 9 # WiuracldeR A.S. 2 qts . 3.2# DipelR W.P. 1 lb. Each treatment consisted of 3 randomly selected white or red oak trees, 10 to 15 m in height. A Kiekens K.W.H. backpack mist blower was used to apply 1 gal of spray mix per tree. The application was made from atop a 12 m ladder mount- ed on a 3/**-ton pickup. Three 100 ft polyethylene tarps 108 were placed beneath trees and phytophagous as well as bene­ ficial Insects were oollected. Insecticides were applied In the vicinity of Big Star Lake on the morning of 15 Aug* 1974. Temperatures on that date were 25*5°C and wind speed less than 7 km per hour. Insects which dropped to the tarp were collected 24 hours after spray treatment. These were consi­ dered the population killed by the Insecticide. The mean larval lnstar during the Insecticide trial was 3 * 2 and no fifth lnstars were found during spray application. Tarps were left in place beneath the trees and 7 days after spray appli­ cation a 0,5% spray (pyrethrum) was applied to the canopy of each treatment and surviving organisms were collected and counted. The percent survival was determined by the number of larvae collected from the pyrethrum samples divided by the total number of larvae collected from each tree. The results of Insecticides trials are presented In Table 26. Insecticide trials did not show significant difference between treatments. This was a consequence of poor experimen­ tal design (too few samples) and extremely high Inter-tree variability of populations, lhe tests did Indicate that a number of compounds proved satisfactory in reducing H, manteo populations. These included Sevii5^# Orthene®, and Zectrar^. The use of pyrethrum Is also suggested for control of larvae because of low mammalian toxicity and excellent efficacy In removing larvae from trees. The effect of Insecticides tested on non-target organisms was reported by Wallner and Surgeoner (1974). To adequately test the efficacy of Insecticides two 109 prerequisites are required* 1 ) populations sufficient to cause defollatlon, and 2 ) because of high variability approxi­ mately 2 0 trees samples for each insecticide tested. In 1975* populations of H. manteo were too low to adequately evaluate Insecticides. Table 26, Treatment Efficacy of Insecticides tested for control of H, manteo. Mean No. of Larvae Surviving/Taro* % Survival Thuracide 1 ,3 a 36 Dlpel 2 .0 a 75 Sevin 3.3a 15 Zectran 3.7a 21 Orthene ^.7a 28 Dylox 6 .0 a 55 Untreated 13.0 97 1Column means followed by same letter not significantly different at the . 0 5 level by analysis of covariance. Suggestions for Further Research This research study was designed to determine the life history and population dynamics of the variable oakleaf cater­ pillar. The broad nature of the program suggested many aven­ ues of research which should be investigated to present a comprehensive understanding of the biology of this Insect, Due to time constraints and resource limitations, I was not able to investigate many of these research areas. Many ques­ tions remain unanswered and are here suggested for further 110 research. I would hope that in all Invest1 gatlone the re­ searcher would ask "Does this Insect and the researoh sug­ gested warrant expenditure of tax dollars or oould the re­ sources be better utilized In the study of some other prob­ lem?" The damage caused by the variable oakleaf caterpillar is minimal and researoh should not be related to economic damage. The population dynamics of this insect are similar to many forest and agricultural pests which remain at low densities for long periods of time, yet occasionally produce epidemic populations of short duration. It is hoped that the study of this insect will produce principles of Insect popu­ lation dynamics which can be utilized In the study of other economic pests. For a more comprehensive study of the biology of this insect the following areas of needed research are sug­ gested. During the 3 years of this study populations of H. manteo have declined since the major defoliation of 1 9 7 3 . We as entomologists and scientists in general become involved In research efforts only when a problem exists. As entomologists we Investigate high density rather than low density populations. This investigation has suggested factors which prevent con­ secutive defoliation and bring about the reduction in popula­ tion levels. The factors which led to the 1973 defoliation were not studied. This would require a long term research pro­ ject of populations at low densities. The researcher is pre­ sented with major problems In sampling and funding support when studying low density populations. The Insect should be viewed as an ideal research organism in the study of insect Ill population dynamics* The accessibility of the research area* sampling techniques devised during this study• and Informa­ tion concerning population levels In 1 9 7 3 - 7 6 should aid In any future research efforts. densities should be attempted. A study of the Insect at low At the very least each August and September prepupal samples could be taken to monitor the prepupal population which remains In prolonged diapause and the overwintering population each year. This population monitoring will indicate when populations again begin to in­ crease and If the percentage prepupae remaining In diapause changes dramatically In any year. Prepupae provide an excellent research organism for the study of Insect diapause. They exhibit* X ) prolonged diapause lasting 1 * 2 , 3 years and perhaps longer, 2 ) are relatively abundant and easy to sample, 3 ) are larcce and can be readily handled and stored, and 40 laboratory rearing pro­ cedures have been developed. The following questions may be answered In future re­ search* 1. How to differentiate between the various aged prepupae. Some chemical waste product. I.e. uric acid may Increase In prepupae as they remain In the soil. This chemical may be identified and quantified to determine age of prepupae. 2. What environmental factors Initiate diapause In prepupae. 3» What environmental factors terminate diapause and why do some pupate and others remain in diapause. 4-. What environmental factors prevent pupation of prepupae from the first of August to October when soil temperatures 112 are above the developmental threshold. The prepupae parasitized by £• bethunel are apparent due to their reduced body and head capsule size. sltold Is able to remain In prolonged diapause. The para­ Because the prepupae parasitized by D. bethunel are readily Identifiable there exists an excellent opportunity to study how the syn­ chronization of diapause between parasltold and host Is ac­ complished. Other research objectives should Include* 1. Specific Identification of egg parasitolds Telenomus sp. and Trlchogramma sp. and larval parasltold Phobocftmny sp. 2. Identification of alternate hosts of the Insect parasi­ tolds and predators* particularly those on which egg parasitolds overwinter. 3. Factors either Intrinsic or extrinsic which affect larval and pupal coloration and the possible use of prepupal color Intensity as an Indicator of prepupal age. 113 Literature Cited Alien* D.C. 1972. Insect parasites of the saddled promi­ nent* Heterocampa guttlvltta (Lepldopterai Notodontldae) In trie northieastern United States* Can. Entomol. 1 Oki 1 609-22. ________ 1973* Fecundity of the saddled prominent, Heterocampa guttlvltta. Ann. Entomol. Soc. Amer. 6 6 i 1181-3. Allen, D.C. and D.G. Grlrable. 1970. Identification of the larval lnstars of Heterocampa guttlvltta with notes on their feeding behavior. J. Econ. Entomol. 6 3 « 1201-3. Andrewartha, H.G. 1952. logy of Insects. Anonymous* 1958. 81 *1-37 . ________1971. Diapause In relation to the eco­ Biol, Rev. 27 t 50-107. U.S.D.A. Coop. Econ. Insect Rept. U.S.D.A. Coop. Econ. Insect Rept. 21* 281. Baker* W.L. 1972. Eastern forest Insects. U.S. Dept. Agric. For. Serv, Misc. Publ. 1175* 6U2 PP« Beach* G, 1972. Forest pest report Minnesota, 1972, Pest Rept. Minn. Dept. Agrlc. 1972 1 17-20. For. Brahman, R.R. 1975. Consumption of red oak foliage by gypsy moth larvae, M, Sc. Thesis University of Michigan, ^8 pp. Brazzel* J.R. and D.R. Martin. 1959. Winter survival a^d time of ene^rence of dlapauslng pink bollworms In Texas. J. Econ. Entomol. 53* 305-8. Campell* R.W. 1975* The gypsy moth and Its natural ene­ mies. Agr. Infor. Bull. No. 381 . U.S. Dept. Agrlc. For. Serv. 10 pp. Casegrande* R.A. 1975* Behavior and survival of the adult cereal leaf beetle, Oulema melanopus (L.) Ph.D. Thesis* Michigan State University. 171 pp. Craighead* F.C. 1950. Insect enemies of eastern forests. U.S. Dept, Agr, Misc. Pub. 657. 679 pp. Detwller, J.D. 1922. The ventral prothoraclc gland of the red-humped apple caterpillar CSchlzura coneInna Smith and Abbot). Can. Entomol. 5^* 1 76-91. Ill* 192*5. Further studies of the ventral thoracic pland of certain Notodontld caterpillars. Can. Entomol. 57i 266-71. de Wilde, J. 1962. Photoperlodism In Insects and mites. Ann. Rev, Entomol. 7* 1-26. Doubleday, E. 1841. Characters of three new genera of Notodontldae, from North America. The Entomol,, 1» 5 5 -6 0 . Dyar, H.G, 1891. Preparatory stapres of Hetercampa unlcolor Pack, Psyche. 6* 95-6. Ehrllch, F.T., R,T. Franklin, and R.N, Coulson. 1969. Characters for determining sex of pupae of the oakworras Anlsota sanatoria. A. stlffma and A. vlrvlnlensls and the yellow-necked caterpillar Datana mlnlstra. Ann. Entomol. Soc. Amer. 62* 931-2• Eisner, T, , A.F. Klu*re, J.C, Carrel, and J. Melnwald. 1972. Defense mechanisms of arthropods, XXXIV, Formic acid and acyclic Ketones In the spray of a caterpillar. Ann. Entomol. Soc. Amer. 6 5 * 765-6. Fuzeau-Braesch, S. 1972. Pigments and color changes. Rev, Entomol. 17* 403-24. Ann. Grlmble, D.G. and R.G. Newell. 1972a. Saddled prominent pupal samplinp and related studies in New York State, AFBI res. Rep. 11, SUNY Coll, Environ. Scl. and For. Syracuse, 28 pp. 1972b. Saddled prominent ovipositional studies In New York and adjacent states. APRI res. Rep. 12, SUNY Coll. Environ. Scl. and For. Syracuse, 3 8 PP. Guimaraes, J.H. 1972. A revision of the prenus Wlnthemla Roblneau-Desvoldy in America north of Mexico (blpterai Tachinldae). Acq. Zool. S. Paulo, 22% 27-112. Hanson, J.B., W.H, Hoffard, and P.W. Orr, 1976. Forest pest conditions In the northeast, 1975, U.S. Dept. Affr. For. Serv. Bull. 24 pp. Herrick, G.W. and J.D, Detwller. 1919. Notes on the repuprnatorlal ylands of certain Notodontld caterpillars. Ann. Entomol. Soc. Amer. 12* 44-8. Holllny, C.S. 1959. The components of predation as revealed by a study of small mammal predators of the Euro­ pean pine sawfly. Can. Entomol, 91* 293-320. 115 Hooker* W.A. 1 9 0 8 . Injury to oak forests In Texas by Heterocanma manteo (Dbldy.) (Lepldopterai Notodontldae). Proc. Entomol. Soc. Wash. IO i 8-9* Houston* D.R. and J.E. Kuntz. 1964. Fart III* Pathogens associated with maple blight. Int Studies of maple blight* Univ. of Wise* Res. Bull. 250* 1 2 9 pp. Twao* S, 1968, Some effects of grouping In lepldopterous Insects. Colloq. Int, Centre Nat. Rech, Scl. 173* 1 8 5 -2 1 0 . Kearby, W.H, 1973* In. Tlcehurst* M. and D.C. Allen* Notes on the biology of Telenomus coleodasldls (Hymenoptera* Scellonldae) and Its relationship to the saddled prominent* Petercampa guttlvltta (Lepldopterai Notodontldae)• Can.Entomol. 105 1 1133-^3• 1975a. Variable oakleaf caterpillar larvae secrete formic acid that causes skin lesions, (Lepldopterai Notodontldae). J. Kan. Entomol. Soc. 4 8 1 280-2 . ________ 1975b. Personal communication. Klots* A.B, 1967. Larval dimorphism and other characters of Heterocampa pulverea (Grote and Robinson) (Lepldoptera1 Notodontldae). J. N.Y. Entomol. Soc. 75* 62-7. Kulman, H.M. 1971. Effects of insect defoliation on growth and mortality of trees. Ann, Rev. Entomol, I61 289-324. Leonard* D.E. 1971. Alr-bome dispersal of larvae of the gypsy moth and Its influence on concepts of con­ trol. J. Econ. Ent. 64* 638-41. Millers, I, and G. Erickson. 1970. Oak defoliation in Manistee National Forest, 1970. Unpublished re­ port on file In Field Office Northeastern area State and Private Forestry, U.S. For. Serv. St. Paul, Minn. Millers, I. and W.E. Wallner. 1975* The red-humped oakworm, U.S. For. Serv. Pest Leafl. 153* 6 pp. Morris, R.F. 1955* The development of sampling techniques for forest Insect defoliators with particular reference to the spruce budworm. Can. J, Zool. 33* 225-94. 116 Muesebeck, C.F.W., K.V. Krombein, and H.K, Townes. 1951* Hymenoptera of America north of Mexico. Synoptic catalog, Agricultural Monogr. 2. 1*1-20 pp. Mukerjl, M.K, and J.C. Guppy. 1970. A quantitative study of food consumption and growth of Pseudaletla unlouncta (Lepldopterai NoctuidaeK Can. Entomol., 1021 1178 -8 8 . Packard, A.S, 1886. The fluid ejected by Notodontlan caterpillars. Amer. Naturalist 2 0 1 811-12. 1895. Monograph of the Bombyclnae moths of Ameri­ ca north of Mexico. Natl. Acad. Scl. Mem. 7* 1-390. Palmer, L.S. and H.H. Knight. 1924. Carotin - the princi­ pal cause of the red and yellow colors In Perlllus bloculatus (Fab.) and Its biological origin from the haemolymph of Leptlnotarsa decemllneata Say. J. Biol. Chem. 5 9 1 ^ 3 .^, Parker, J. and D.R. Houston. 1971. Effects of repeated defoliation on root and root collar extractives of sugar maple trees. For. Scl, 17* 91-5. Poulton, E.B. 1 8 8 6 , Notes In I 8 8 5 upon lepldopterous larvae and pupae. Including an account of the loss of weight in freshly formed lepldopterous pupa. Trans. Entomol. Soc. London, 34 1 137-79. Powell, J.A. 1971. Occurrence of prolonged diapause In Ethmlld moths. Pan-Paclflc Entomol. 50* 220-5. Rabinovich, J.E. 1970. Population dynamics of Telenomus facial (Hymenopterai Scellonldae) a parasite of Chagas* disease vectors, VI., Ability to discrimin­ ate between parasitized and non-parasltlzed hosts. Acta. Clent. Venezolana 21* 154-6. Rahman, M. 1970. Effect of parasitism on food consump­ tion of Plerls rapas larvae. J. Econ. Entomol. 63* 820-1. Ravelin, F.W. 1975* A revision of the genus Aroh.vtas Jaennlcke (Dlptera* Tachlnldae) for America north of Mexico. M, Sc, Thesis, Michigan State Univer­ sity 8 0 pp. Robertson, J.L., R.F, Lyon, F.L, Shon, and N.L. Gillette. 1972. Contact toxicity to twenty Insecticides applied to Symmerlsta oanlcosta. Franclemont. J. Econ. Entomol, 65* 15^0-2. 117 Salt9 R.W, 1953* The Influence of food on cold hardiness of insects* Can. Entomol. 85* 261-7* 1961. Principles of Insect cold-hardiness. Ann* Rev. Entomol* 61 55-7^* Slmmonds, P.J. 1 9 *fB. The Influence of maternal physiology on the incidence of diapause. Trans. Roy. Soc. London (Series B) 233* 385-^1 Southwood, T.R.E* 1 9 6 6 . Ecological methods with particu­ lar reference to the study of Insect populations. Methuen. London. 391 pp. Surgeoner, G.A. and W.E. Wallner. 1975* Determination of the larval Instars of Heterocampa manteo and re­ duction of larval head capsule size by the paras1toid D1radons bethunei. Ann. Entomol. Soc. Amer. ---------1061-57 Tauber, M.J. and C.A. Tauber. 1976. Insect seasonality* diapause maintenance, termination, and post dia­ pause development. Ann. Rev. Entomol« 21* 81109. Tauber, M.J., C.A. Tauber, and C.J. Denys. 1970. Adult diapause in Chrysopa carnea* photoperlodlc con­ trol of duration and color. J. Insect Physiol. 1 6 * 9**9-55* Tlcehurst, M. and D. C. Allen. 1973. Notes on the biology of Telenomus coleodasldls (Hymenoptera* ScelionldaeJand Its relationship to the saddled prominent, Heterocampa /ruttlvltta (Lepidoptera* Notodontldae). Can. Entomol. 1 0 5 * 1133-^3* Veatch, J.0, 1953. Soil map of Michigan, Michigan State College Press, East Lansing, Michigan. 1 p. Wallner, W.E. 1971. Suppression of four hardwood defolia­ tors by helicopter application of concentrate and dilute chemical and biological sprays. J. Econ. Entomol. 6U-t 1^87-90. Wallner, W.E. and G.A. Surgeoner. 197^* Control of variable oakleaf caterpillar, Heterocampa manteo, and the impact of controls on non-target organisms. 197^, Pesticide Field Evaluation Research, Dept, of Entomol. Michigan State Univ. unpubl, mlmeo, Z|>2—6 . Wellington, W.G. 1 9 6 ^. Qualitative changes in populations in unstable environments. Can. Entomol. 961 ^36-51. 118 Wetzel* B.W. 1972. History and parasite study of the variable oakleaf caterpillar* Heterocampa man­ teo . and the red-humped oakworm Symmerlsta canlcosta In Minnesota. Unpublished Coop. Hept. Minn. Dept. Agr. and Univ. of Minn. 8 pp. Wilson* L.P. 1971. Variable oakleaf caterpillar. Forest Serv. Pest Leafl. 67 . ^ pp. U.S. 119 APPENDIX Determination of Larval Inatan of Hataroemmpa nuaitmo1 and Reduction of Larval Head Capsule Sise by the Parasitoid Diradop* betkunai** GORDON A. SU RGEON ER and W ILLIA M E. W ALLNER Department of Entomology* Michigan State University, East Lansing 48824 ABSTRACT Head capsule measurements were used to distinguish (Cresson) resulted in a reduction of head capsule sise. the 5 larval stages of the variable oak leaf caterpillar Percentage parasitism of overwintering prepupae of H . Heterocampa manteo (Doubleday). Parasitization of H . manteo was accurately determined by head capsule manteo larvae by the khneumonid D iradopt betktmei measurements. Dyar (1893). He reported head capsule measure­ ments between 4.0 and 4.3 m m for 5th instars, whereas we found them to be between 3.0 and 3.5 m m . An accurate comparison between our study and Dyar’s is difficult since he measured only the 5th instar and did not specify how many specimens were examined. Our head capsule measurements were made from one ocellus to the other; however, even at the widest point, the head capsule did not reach 4.0 mm. Differences may be ascribed to geographic variation of the population, the quantity and quality of nutrition, genetic variation or other possible factors. Head capsule widths of the first 3 instars did not overlap but ca. 20% of the 4th and 5th instars had head capsules smaller than anticipated. Dissection revealed that larvae with the smaller head capsule contained an ichneumonid larva. While there was some overlap between normal and parasitized 4th in­ star head capsules, parasitized 5th instars exhibited distinctly different widths (Fig. 1). This could lead to some confusion in determining the number of lar­ val instars. The ichneumonid Olesicampe sp. nr. nem atorm n (Tshek) caused a similar reduction in the larch sawfly, P ris tip k o ra e ric k s o n ii (Hartig) MATERIALS AND M ETHODS (Muldrew 1967). The parasitoid larvae affecting head capsule size Larvae of H . m anteo were collected in Michigan from mixed oak trees July-September 1973. Over­ were determined as D . bethunei. It was never reared wintering prepupae were collected by sifting samples from larvae with normal-sized head capsules and was always reared from larvae possessing the reduced of litter. head capsules. Rates of parasitism were 17% in RESULTS 1973 when H . manteo prepupal density was 8.3/0.25 m Head capsule measurements of 200 larvae of each *. and 40% in 1974 when prepupal density was 5.6/ instar confirmed the presence of 5 instars (Fig. 1). 0-25 m*. The following species of parasitoids were These measurements agreed closely with those deter­ reared from 5th instars of H . m anteo having normal mined by Allen and Grimble (1970) for H . g u ttiv itta head capsule size: P ro to m ic ro p litis skisurae (Muesbut were less than those given for H . m anteo by beck) (Hymenoptera:Braconidae), Pkobocam pe sp., B a ry ly p a sp. (Hymenoptera:Ichneumnnidae), W in them ia datanae (Townsend), Lespesia skisurae 1 U f M o p t tn : NotodoatMw. (Townsend) and A rc h y ta s a te rrim n s (Rob.-Desv.) ■Kni(twpicn: lAMOwnidu. * Michigan Agric. Exp. Stn. Journal A rticle Number 7237. (Diptera:Tachinidae). Only B a ry ly p a sp., L . s k i* * Received lor publication A pril 32, 1915. The variable oak leaf caterpillar, H eterocam pa (Doubleday), is a late summer defoliator of numerous deciduous trees found throughout eastern United States and Canada (Packard 1895, Wilson 1961). Outbreaks of H . manteo seldom last longer than 2 to 3 years. Because parasitism contributes to this population decline, accurate determination of prepupa1 mortality is essential in formulating popu­ lation studies and control strategies. Head capsule measurements have been used to determine the number of instars of various Lepidoptera (Dyar 1890). Utilizing this method Allen and Grimble (1970) reported that the saddled prominent, H eterocam pa g u ttiv itta (Walker), had 5 larval in­ stars. Packard (1895) presented detailed descriptions of the 5 larval instars of H . manteo but did not give accurate head capsule measurements. We found Packard’s descriptions unreliable for determining in­ stars of H . m anteo in Michigan. Here we present an accurate means of determining the instars of H . manteo by head capsule measurement and the level of parasitism of H . manteo prepupae by D ira d o p s bethunei (Cresson) without larval rearing or dissec­ tion. manteo 1061 120 1062 A m m a l * o r t h e E n t o m o l o g ic a l S o c ie t y o r A m e e ic a [Vol. 68,no.6 and A . a ie rrim u t overwintered in H . manteo pre­ pupae and had a combined parasitism rate of less than 10%. ra t --MUtlTUIB S-S ACKNOWLEDGMENT The authors wish to thank Dr. W. Masner, Dr. C. Loan, Dr. H. Wood, and Dr. M. Ivanochko of the Biosystematies Research Institute, Agriculture Can­ ada, Dr. H. K. Townes of the American Entomologi­ cal Institute, and Mr. W. Ravlin and Mr. J. Newman of Michigan State University for aid in insect iden­ tification. «.T 14 l-l 1.1 I N ■ IN STA ft nr F ig . I .— Reduction o f head capsule liie o f bjr the parasitoid D iradopt bethunei. V H . manteo REFERENCES CITED Allen, D. C, and D. G. Grimble. 1970. Identification of the larval instars of Heterocampa g u ttivitta with notes on their feeding behavior. J. Econ. Entomol. 63: 1201-3. Dyar, H. G. 1090. The number of molts of lepidopterous larvae. Psyche S: 420-2. 1893. On the differences between the larvae of Cecrita bilineata and Heterocampa maateo. Entomol. News: 262-3. Muldrew, J. A. 1967. Biology and initial dispersal of Olesicampe (H olocrem nut) sp. nr. nemotorvm (Hymenoptera: Ichneumonidae). a parasite of the larch sawfly recently established in Manitoba. Can. Entomol. 99: 312-21. Packard, A. S. 1895. Monograph of the Bombycinae moths of America north of Mexico. Part I, Fam­ ily I, Notodontldae. Men. Natl. Acad. Sri., pp. 224-230. Wilson, L. F. 1961. Variable oak leaf caterpillar. U.S. Forest Serv., Pest. Leafi. 77. 4 pp. Reprinted from the A n n a ls o r th e E n t o m o l o g i c a l S o c i e t y o r A m e b ic a