Age-Dependent Movement Patterns of Japanese Beetle andEuropean Chafer (Coleoptera: Scarabeidae) Grubs in

Soil- Turfgrass Microcosms

M, G. VILLANI ANDJ. p, NYROP

Department of Entomology, New York State Agricultural Experiment Station,Cornell University, Geneva, New York 14456

Environ.Enlomol.20(1): 241-251 (1991)ABSTRACT Movement patterns of japanese beetle, Popillia japonica Newman, and Eu-ropean chafer, Rhizotrogus (Amphimallon) majalis (Razoumowsky),grubs as influenced bygravity, host plant position, and external disturbances were studied in laboratory soil-turfgrassmicrocosms. Second through third instars just before pupation were monitored using radio-graphic techniques. Neonates were monitored using destructive sampling. Results demon-strate significantly different movement patterns between speciesand among age groups. Thedevelopment stage of the grub had a large effect on japanese beetle grub behavior and ameasurable, but lesser effect, on European chafers. All larval stagesof European chafers andall larval stages of japanese beetle, except neonates and postoverwintering third instars,displayed a downward movement in response to disturbance. Neonate japanese beetlesshowed little movement while postoverwintering. japanese beetles moved upward whendisturbed. European chafer grubs of all age classes displayed random vertical movementwith some arrestment in or near sod. Preoverwintering and postoverwintering third-instarchafers showed less dramatic arrestment behavior than other instars tested. Second-instarjapanese beetles behaved similarly to European chafers; however, third instars behaved verydifferently. All third instars except those tested in late winter and early spring showed someinnate downward movement in the soil microcosms. japanese beetles tested in late winterdisplayed random movement with some arrestment in sod, whereas those tested in earlyspring exhibited upward movement and arrestment in sod.

KEY WORDS Insecta, soil insects, movement, behavior

IN THE NORTHEASTERNUNITEDSTATES,turfgrassroots are subject to intense feeding pressure froma complex of scarab beetle grubs that includes theJapanese beetle, Popillia japonica Newman, andthe European chafer, Rhizotrogus (Amphimallon)majalis (Razoumowsky). Although grub responseto temperature has been described through fieldobservations (Hartzell & McKenna 1939, Gyriscoet a1. 1954, Tashiro & Gambrell 1963, Tashiro eta1. 1969, Fleming 1972, Tashiro 1987) and micro-cosm studies (Villani & Wright 1988), the impor-tance of grub development on species-specific be-havior of the imature stages is generally unknown,

The annual vertical migration of scarab grubsand other soil insects in the soil profile is oftenassociated with seasonal changes in soil tempera-ture (Villani & Wright 1990). Villani & Wright(1988) explored the effect of fluctuating soil tem-perature on Japanese beetle and European chafergrub movement patterns through the use of radio-graphic monitoring of soil-turfgrass microcosms.These studies indicated that third-instar Japanesebeetles were highly responsive to changes in soiltemperature, whereas European chafers were rel-atively insensitive to fluctuating temperatures.These results appear to parallel reports of fieldpopulations under similar temperature conditions.What is not clear, from field or previous laboratory

studies, is whether the seasonal movement of thesegrub species is a direct response to changes in soiltemperature, a response triggered by grub age, ora combination of both. Knowledge of the behaviorof Japanese beetle and European chafer grubs iscrucial to asking more complex questions concern-ing the interaction of these insects with their en-vironment and with potential chemical and bio-logical control agents. The study reported here wasdesigned to determine the influence of grub de-velopment on the movement of Japanese beetleand European chafer grubs in the soil profile fromegg hatch through prepupation. This was done insoil-turfgrass microcosms to avoid confounding en-vironmental heterogeneity.

Materials and Methods

Japanese beetle and European chafer grubs werecollected from untreated turfgrass in central NewYork state. Grubs were used in behavior studiessoon after collection from the field or were held insoil-filled boxes at or near soil temperatures en-countered in the field (Table 1). For age classestested immediately or soon after collection, thesetemperatures reflected those experienced by grubsimmediately preceding their introduction into themicrocosms. Overwintering grubs were stored at

0046-225Xj91j0241-0251$02.00jO © 1991 EntomologicalSocietyof America

242 ENVIRONMENTAL ENTOMOLOGY Vol. 20, no. 1

Table J. Storage conditions and dates European chaferand Japanese beetle grubs were used in soil microcosmstudies

Storage Duration of Dates grubsStage and species" temp, storage, db were used in

oc; microcosms

Neonate JB 7 Aug.Neonate EC 25 JulySecond-instar JB 14 Aug.Second-instar JB 21 Aug.Second-jnstar JB 10 7 28 Aug.Second-instar EC 21 Aug.Second-jnstar EC 10 7 28 Aug.Early third-instar JB 20 5 11 Sept.Early third-instar JB 20 12 18 Sept.Early third-instar EC 11 Sept.Early third-instar EC 20 7 18 Sept.Prewinter JB 10 5 16 Oct.Prewinter JB 10 12 23 Oct.Prewinter EC 10 5 16 Oct.Prewinter EC 10 12 23 Oct.Early-overwinter JB 10,4.5 61,21 11 Dec.Early-overwinter JB 10,4.5 61,28 18 Dec.Early-overwinter EC 10,4.5 40,21 11 Dec.Early-overwinter EC 10,4.5 40,21 18 Dec.Late-overwinter JB 10 103 22 Jan.Late-overwinter JB 10 131 29 Jan.Late-overwinter EC 10 103 22 Jan.Late-overwinter EC 10 77 29 Jan.Postwinter JB 30 MarchPostwinter EC 10 60 30 March

a EC, European chafer; JB, Japanese beetle.b _, not stored, used immediately after collection.

temperatures slightly higher than those experi-enced in the field to avoid accidental freezing. Asa result, the physiological age of overwintering grubsmay be slightly greater than that of field popula-tions at the same time period. For this reason, shiftsin behavior observed in laboratory-held overwin-tering grubs may actually anticipate similar changesin field populations. All grubs were held at 20°Cfor 24 h before being tested.

Soil-turfgrass microcosms were designed to pro-vide a simple experimental chamber for studyinggrub movement in which soil moisture, soil tem-perature, and matrix heterogeneity could be con-trolled. Sandy loam soil screened (0.5 by 0.5 emmesh) to ensure homogeneous soil condition wasplaced in the chamber, and sod that consisted of25% each of 'Touchdown,' 'Glade,' 'Adelphi,' and'Gnome' grass was placed at one end of the cham-ber. To balance soil moisture in the microcosms, amoist sponge was placed at the other end of eachchamber opposite the sod. The addition of spongesat the ends of the microcosms opposite from thesod resulted in nearly symmetrical gradients of in-creasing soil moisture away from the center of thearena. Movement of the first instars of both speciesand early second-instar Japanese beetles within asoil profile was studied using destructive sampling,and later instars were studied with radiographicimaging (Villani & Wright 1988). Two experimentswere conducted. The objective of both experimentswas to measure the movement of different-aged

grubs in response to a potential host (sod) and grav-ity. Responses of seven age classes of grubs werestudied.

In the first experiment, soil microcosms wereplaced in Plexiglass chambers with inside dimen-sions of 3 by 3 by 30 em. One side of each chamberwas removable and had a hole (approximately 1cm in diamter) drilled through the center. Ducttape covered the hole except when it was removedto allow placement of grubs at the midpoint of thechamber.

Chambers were filled with 250 g of soil, and thesoil was compressed from the 30-cm length of thechamber to 21 cm (about 30 psi). A piece of sodwas placed at one end of the chamber in contactwith the soil and held in place by a wooden block.The other end was plugged with a damp spongeand a wooden block. These microcosms were keptuninfested in "sod up" or "sod down" positionsovernight. The next day, grubs were introduced inthe microcosms, which were then placed in com-plete darkness at 20°C. Except for neonates andsecond instars, microcosms were radiographed insets of seven at 2, 3, 6, 12, 24, 27, 32, 48, 51, 56,and 72 h after the placement of grubs in the mi-crocosms. Each age class was studied in two sep-arate replicates 1 wk apart, with a total of 56 grubsof each species tested in each age class.

First-instar behavior was measured through de-structive sampling of the microcosms because thegrubs were too small to radiograph. One hundred"sod up" and 100 "sod down" microcosms wereeach infested with three laboratory-reared, first-instar European chafers or Japanese beetles. Themicrocosms were divided into two sets to be openedat 24 and 72 h, at which time the location of thegrubs was determined. Soil moisture was measuredvia standard gravimetric techniques; mean per-centage moisture when microcosms were set upwas 9.92 (S2 = 1.91, n = 6). Additional soil moisturemeasurements were taken 24 and 72 h after grubinfestation.

Second-instar Japanese beetles were studied us-ing destructive sampling and through the use ofradiographs, whereas second-instar European cha-fers were large enough for the use of radiographyalone. Destructive monitoring of Japanese beetleswas similar to the procedure outlined for first in-stars. One hundred microcosms were each infestedwith two second-instar Japanese beetles. Micro-cosms were placed in an upright position, half withthe sponge on top and half with the sod on top.The microcosms were divided into two sets to beopened at 24 and 72 h. In addition, 112 microcosmswere infested with one European chafer or Japa-nese beetle grub and monitored using radiographyat 3, 6, 12, 24, and 48 h. Data obtained usingdestructive sampling and radiographs were groupedfor analysis.

Patterns of grub movement in the microcosmswere analyzed in two ways. First, box plots (McGillet al. 1978) were used to portray the position of

February 1991 VILLANI & NYROP: MOVEMENT PATTERNS OF SCARABEID GRUBS 243

Table 2. Means and standard deviations (0 = 15) ofpercentage soil moisture in microcosms used in the firstset of experiments 24 h after placement of grubs

Table 3. Means and standard deviations (0 = 2) ofpercentage soil moisture in microcosms used in the secondset of experiments

• ~ 110percentile

Fig. 1. Information provided by box plots used todisplay grub positiondata.

Position of sod Position inf SDin microcosm microcosm

Up Top 11.75 2.09Up Middle 9.48 1.72Up Bottom 11.44 1.84Down Top 11.65 2.03Down Middle 9.57 1.78Down Bottom 11.59 1.74

the grub population at different points in time. Boxplots are useful because they concisely summarizea distribution of observations and facilitate com-parison of two or more such distributions. Popu-lation level patterns of movement when disturbed(escape), dispersive movement, and arrestment atfood are discernible through the use of box plots.An annotated box plot is illustrated in Fig. 1. Onlybox plots of the population distributions for 3, 24,48, and 72 h are shown. Restricting the data pre-sented to dispersive patterns at relatively uniformtime intervals is visually more straightforward andillustrates population trends well.

Although box plots are useful tools for visualizingthe distribution of a population, they provide noinformation on the movement of individual grubswithin a population. Use of the radiographic tech-nique makes it possible to monitor the movementof individual grubs over time (Villani & Wright1988). To examine individual patterns of move-ment, we plotted the rate and direction of move-ment of each grub from time t to time t + 1 as afunction of the position in the soil profile at timet. In this way, the movement patterns of individualgrubs leading to the observed population distri-bution at time t + 1 could be visualized.

Micro-cosm Vertical Rotated 90" Horizontalarien·tation, f SO f SO f SO

position"

0 6.1 1.5 6.5 1.4 5.9 0.93 4.4 0.6 4.9 0.8 4.0 0.86 3.1 0.1 3.3 0.2 3.0 0.39 2.8 0.1 2.8 0.8 2.7 0.0

12 2.7 0.1 2.7 0.1 2.6 0.115 3.0 0.5 3.0 0.5 2.7 0.118 5.0 1.0 4.4 1.2 3.6 0.621 7.5 1.1 6.8 0.4 6.4 3.3

" Position in centimeters from the sod.

Movement behavior in postoverwintering grubsthat occurred in response to sod and in response togravity was differentiated in the second experimentby manipulating the position of the soil microcosmsto negate the influence of gravity. Experimentalmicrocosms similar to those used in the first ex-periment but wider (25 by 3.5 by 35 cm) were usedin this study. Turfgrass and sponges were placedon opposite ends of each chamber, as in the firstexperiment, but the microcosms were then placedin one of three positions. In the first position, mi-crocosms were vertical as in the previous set ofexperiments so that they were 35 cm high, 25 cmdeep, and 3.5 cm wide (Fig. 14). For each species,eight chambers were used, four with the sod onthe bottom of the chamber and four with the sodon the top. In the second position, the microcosmswere rotated 90° from the vertical position so thatthey were 25 cm high, 35 cm wide, and 3.5 cmdeep and the sod lay vertically along one 25-cmside (Fig. 15). For each species, eight chambers inthis position were used. In the third position, themicrocosms were placed in a horizontal position sothat they, were 3.5 cm high, 35 cm wide, and 25cm deep (Fig. 16). Microcosms were constructedas in the first experiment, except that 3.2 kg ofloamy sand soil at 4% soil moisture was added, andthe soil was compressed at 30 psi using an hydraulicpress.

Microcosms were infested with either five post-overwintering third-instar European chafers orJapanese beetles 24 h after the microcosms wereset up. One side of each of the chambers was re-moved, and five holes were made with a pencilhalfway between the top and bottom edges of thesoil approximately midway along the horizontalaxis of the chambers. One grub was placed in eachof the holes in the microcosms. Microcosms wereradiographed 24 and 144 h after infestation. Per-centage soil moisture was also determined at thesetimes by taking eight samples in each chamber at3-cm intervals from the center. Grub movementwas analyzed by plotting grub position in the mi-crocosms at 24 and 144 h and by comparing dis-tributions using log-linear models.

Medlen (50 percentile)

Lower quertlle(25 percentile)

Outlier.**

10 percentile

Upper quertllll(75 percentile) ", •••

~

244 ENVIRONMENTAL ENTOMOLOGY Vol. 20, no. 1

13.0

13.0

o

o

o

oo

-6.S 0.0 6.5

Grub Location 6 hours

o

oo

o 0.."..- - - - - - - - - - - --- ---o 00 0

Japanese Beetle (second instar) Sod Up

00o 0 000

oo 0

~.'8.. :. ------ ------ b"~""

o

European Chafer (second instar) Sod Down

2.0

b

c.2U~i5011

i~ 1.0

/l.!cE-.,.r:.EN.,~> ,0'"~OJ ·1.0::0'0:~'0.5 ·2.0

-13.0

a

Fig. 4. Rates of movement of second-instar Euro-pean chafers in "sod down" microcosms between 6 and12 h and Japanese beetles in "sod up" microcosms be-

. tween 24 and 48 h. Shaded areas indicate combinationsof rates and initial positons that could not be realized.Grubs at the shaded-unshaded interface are at the topor bottom boundaries of the microcosms.

24 72·13.0

Hours

Fig. 2. Box plots depicting the distribution of first-instar European chafers and Japanese beetles at 24 and72 h after placement in microcosms with sod at the top(sad up) or bottom (sod down).

Results and Discussion

13.0

Q.i 0

E -6.5

~ -13.0~ 13.0Q.c3 c 6.S

il.g 0

;§ -6.5

6.5

Neonate larvaeEuropeanChafer JapaneseBeetle

Percentage soil moisture within the soil profileswas virtually symmetrical from the chamber mid-points (Table 2). Therefore, the possible influenceof moisture gradients on grub movement behaviorcan be ruled out.

Neonates. Most first-instar European chafersmoved downward during the first 24 h regardlessof sod position (Fig. 2). However, after 72 h, grubsin both treatments were found predominantly inthe sod. Although position of the sod in the micro-cosms clearly influenced grub position, the highvariability in neonate location at 72 h suggests thatgrubs are sampling the microcosm or that host-finding behavior contains a strong random element.The destructive sampling of this stage precludedanalysis of individual grub movement over timethat would have provided more insight into thisbehavior.

Second instarEuropeanChater JapaneseBeetle

Hours

Fig. 3. Box plots depicting the distribution of sec-ond-instar European chafers at 3, 24, and 48 h and Jap-aneSe beetles at 3, 24, 48, and 72 h after placement inmicrocosms with sod at the top (sod up) or bottom (soddown).

13.0

9 T:6.5

Q.;l

~

~ 0

E ·6.SQ..r:. -13.0Q. 13.0OlCl !6.S

~ 0

-6.S

-13.024 48 ·72

There was little movement of neonate Japanesebeetle grubs in either treatment except for a slighttrend downward, which was similar to the initialmovement of neonate European chafers (Fig. 2).Because females normally place eggs 4-10 em deepin the soil (Fleming 1972), early first-instar Japa-nese beetles likely feed initially on organic matterand fine roots. This behavior pattern has been re-ported in other scarab species (Gray et al. 1947)_The attractive or arrestant properties of turf grassroots to Japanese beetle grubs could not be ascer-tained at this stage in our study. The tendency ofneonate Japanese beetles to be down in the soilprofile may have important ramifications on timingof insecticide treatments and sampling if first in-stars have not yet moved to the soil-sod interface.

Second Instar. Second-instar European chafersin both "sod up" and "sod down" microcosms ex-hibited downward movement upon placement inmicrocosms (Fig. 3). After this initial movementdownward, European chafers in both treatmentsshowed indications of nondirected dispersive

February 1991 VILLANI & NYROP: MOVEMENT PATTERNS OF SCARABEID GRUBS 245

Early third instarEuropeanChafer JapaneseBeetle

o

13.0

o

.6.S 0.0 6.S

Grub Location 6 hours

1.5

a European Chafer (early third instar) Sod Downco~~isoil

.! ~ 0.5

~~c--;.e

EN~ ~.O.5

::s

24 48 7272

~~9~ ¥~~~~ ¥$

13.0

.13.0

~·6.5

Q. 6.5

"~ 0

E .6.5U£" -a:ga.Q)

Cl 16.5o

Fig. 6. Ratesof movementof early third-instar Eu-rop~1llchafersbetween 6 and 12 h when sodwasplacedat the top or bottom of the microcosms.

are in the lower portion of the microcosms at thistime. By comparison, grubs in the "sod up" mi-crocosms are positioned randomly in the profile at6 h and moved randomly. Examination of themovement rates for the 56-72-h interval (Fig. 7a)reveals that a majority of the chafer grubs in the"sad up" microcosms have stopped movement inthe sod at 56 h; however, the remaining grubsexhibited movement similar to those in the 6-12-hinterval.

Box plots of early third-instar Japanese beetlesindicate population distributions similar to those ofearlier instars when sad was down in the turfgrassmicrocosm but a substantially different distributionwhen sod was up (Fig. 5). In "sad down" micro-cosms, Japanese beetles moved down in the profileand remained there. Analysis of rates of movement(Fig. 7b) in the "sad up" microcosms revealed abimodal behavior; some grubs behaved very muchlike second instars (arrested in turf), whereas othergrubs exhibited an apparently innate downwardmovement when placed in the microcosm andshowed no further movement.

Studies focusing on the effect of temperature onJapanese beetle development rates (Regniere et al.1981a) suggest that Japanese beetles undergo dia-pause as overwintering grubs. Our data indicate

13.0

oo

o

.6.S 0.0 6.S

Grub Location 6 hours

oo

o

•.• ~ ....•.•.•.•............•....•.........~ •....•..............•.•.•....•......•........ -~ Poo 0 0 0

o

European Chafer (early third instar) Sod Up

1.5

·1.5·13.0

c,9c:;eisoil

!~0.5•• ala:E--:;i.eENal-i; cD ..0.5

:::!:

bHours

Fig. 5. Boxplots depicting the distribution of earlythird-instar European chafers and Japanesebeetlesat 3,24, 48, and 72 h after placement in microcosmswith sodat the top (sodup) or bottom (soddown).

movement patterns, presumably in search of food.The nondirected movement of European chafergrubs is apparent when rates of movement duringthe 6-12-h interval in "sod down" microcosms isstudied. (Fig. 4). Grubs near the sod in the "soddown" microcosms (-13 cm) showed no tendencyto move toward rather than away from the bottomof the microcosms, suggesting that contact with sodthrough random search and subsequent arrestmentis important in this grub species, Arrestment in sadoccurs in both "sod up" and "sod down" micro-cosms by the 48-h time period.

Japanese beetle grubs in both "sod up" and "saddown" microcosms exhibited initial downwardmovement (Fig. 3). In "sod down" microcosms,Japanese beetles remained at the bottom at the sod-soil interface. This behavior is in contrast to Eu-ropean chafer grubs in microcosms with similarsad placement; the Japanese beetle population isarrested in its entirety by the 24-h time period,whereas European chafer grubs are disperseo.through the microcosm and move in a randomfashion (Fig. 4). Japanese beetles in "sod up" mi-crocosms exhibited both dispersive behavior (24and 48 h) and arrestment in sod at 72 h. Rates ofmovement during the 24-48-h interval for "sadup" microcosms (Fig. 4) indicates a mass move-ment upward toward the host regardless of positionat 24 h.

Early Third Instars. European chafer grubs inboth "sad up" and "sad down" microcosms initiallyexhibited downward movement (Fig. 5). Compar-ison of box plots in "sad up" and "sod down" mi-crocosms indicates that grubs in both treatmentswere arrested in sad after 72 h, but populationdistributions were much tighter in the "sod down"microcosms. Examination of grub movement ratesbetween 6 and 12 h for both treatments (Fig. 6a,b) indicates that grubs in "sad down" microcosmsgenerally move downward in the profile regardlessof their 6-h position, although most of the grubs

246 ENVIRONMENTAL ENTOMOLOGY Vol. 20, no. 1

Fig. 7. Rates of movement of early third-instar Eu-ropean chafers and Japanese beetles between 56 and 72h when sod was placed at the top of the microcosms.

Japanese BeetleEuropean Chafer

Pre-overwinter third instar

24 72 3Hours

Fig. 8. Box plots depicting the distribution of pre-overwintering third-instar European chafers and Japa-nese beetles at 3, 24, 48, and 72 h after placement inmicrocosms with sod at the top (sod up) or bottom (soddown).

the time of oviposition or the time of egg hatchmay account for the bimodality observed .

Preoverwintering Third Instar. European cha-fer grubs at this stage show initial movement downin both "sod up" and "sod down" microcosms (Fig.8). Subsequently, these grubs move upward andare arrested in sod in both the "sod up" microcosmsand "sod down" microcosms, although the vari-ability in the distributions is greater in the "soddown" system. Rates of movement of grubs in "soddown" microcosms during the 56-72-h interval (Fig.9) appear to have been random, a pattern similarto those of previous life stages. Note, however, thatby 56 h, most of the grubs were in the upper port!onof the chamber. Our data suggest that chafer grubmovement can be characterized by an initial movedownward to the bottom of the chamber followedby a dispersive pattern leading to arrestment in sodwhen contact is made. Box and rate plot patternsat this stage indicate that movement patterns areconsistent with early stages but that chafer grubsmay not be arresting in sod in the "sod down"microcosm, whereas arrestment in "sod up" mi-crocosms seems to persist. This suggests that Eu-ropean chafer behavior (escape, dispersion, arrest-ment) is the product of age-specific response to avariety of external cues (disturbance, food, physicalproperties of soil, etc.). The relative importance ofthe external factors appears to shift as grubs age,and the setup of the experimental microcosms maycause these various cues to either reinforce or ne-gate one another. Although the propensity of pre-overwintering chafer grubs to arrest at a food sourceis not eliminated in the "sod up" microcosms, itsrole appears diminished, whereas the escape anddispersive behavior appears to remain strong. It isimpossible to separate the strength of the externalcue (sod) from the internal response (propensity toarrest); however, sod condition was held constantthroughout the experiment, whereas grub behaviorin both species shifted as grubs aged.

13.0

6.5

-13.0

0-j 0

E -6.5

£ -13.0~ 13.0is.~ ,,6.5S 0

~-6.5

o13.0

13.0

o

-6.5

Grub Location 56 hours

oo

European Chafer (early third instar) Sod Up

Japanese Beetle (early third instar) Sod Up1.5

<:.2'0 1.5~i5011;~Ci a 0.50:]

C'"••E"•....~~

:::E "-0.5

ii:J...'S;:a.:

c~••Eo~ ~.o.s~u

ii:J...-;;;:a.:

b<:.2U~i5011•• 0.5••OJa:

a

that third-instar Japanese beetles exhibit behavioralpatterns consistent with diapause. These grubs werecollected in the field when soil temperatures weredropping in the field. The bimodal behavior maybe the result of testing grubs that were not all at astage of physiological development that was re-sponsive to this environmental cue. Alternatively,some grubs may have been at a depth in the soilthat had not yet experienced low soil temperatures(soil cools first at the surface layers as air temper-ature drops in the fall [Villani & Wright 1990]).

It is also possible that third-ins tar Japanese bee-tles have evolved to "anticipate" the changing ofthe season by moving downward at a particularstage of their development. Third-instar Japanesebeetles accidentally introduced to the Azores fromthe temperate United States move downward intothe soil profile at approximately the same time asthey do in the United States, even though soil tem-peratures in the Azores do not match those in theUnited States (M. Klein, personal communication).This suggests a physiological age-based mechanismfor initiating changes in third-ins tar behavior. Ifchanges in the behavior of Japanese beetle grubsare driven by the physiological age of the insect,

February 1991 VILLANI & NYROP: MOVEMENT PATTERNS OF SCARABEID GRUBS 247

o

130

13.0

o

oo

o 00

°0 0 00 'be. 0 -- Q .•..•..•.•..•.•..•..•..•. -- .•..•..•..•..•. ----:-0

o 0

o

EurC?peanChafer (early-winter) Sod Up

European Chafer (early-winter) Sod Down15

c.2Uf!o o011 0

::'! 0.5 0

1i~a:: £~.cEN~ to;- ·0.5

~:giii:I....; ·15:c .13.0.E

b

a

Japanese beetle grubs at this stage were all foundat the bottom of the microcosms regardless of theposition of the sod. This parallels their behaviorunder field conditions when a mid-winter thawoccurs. Under such conditions, overwintering Jap-anese beetle grubs do not respond to temporarilywarming surface temperatures as do Europeanchafers. Instead, they remain at their overwinter-ing depth until spring (Fleming 1972).

Late-Winter Third Inslar. As with earlier in-stars, the initial movement for both species wasdownward into the soil profile. Both grub speciesaggregated near or in the sod in the microcosmsregardless of its placement (Fig. 12), a pattern sim-ilar to that exhibited by second and early thirdinstars. Patterns of movement rates were nearlyidentical in both species (Fig. 13).

More late-winter third-instar European chaferswere arrested in sod than in previous stages. Thisdata reinforces our hypothesis that European cha-fer grubs exhibit a shifting response to hosts whilemaintaining consistent movement patterns in thesoil profile.

The positive response of late-winter Japanesebeetle grubs to host plants may signal the breakingof diapause brought about by the development ofthe grubs and by the relatively warm soil temper-

·65 0.0

Grub Location 56 hours

Fig. II. Rates of movement of early-winter third-instar European chafers between 56 and 72 h when sodwas placed at the top or bottom of the microcosms.

o

o

24 48 72

Japanese Beetle

o 80...... __ .__ _-_ _-~ ~.~~_.

o o~

European Chafer (pre-overwinter) Sod Down. ..

European Chafer13.0

~8~g. 6.5

~ 0

e- ·6.5 ¢~ ·13 .•.&: 13.0

5.Q) 6.5C 1 .

~ -6.5

·13.0

iii:I'tI.;:g.5 .6.5 0.0 6.5 13.0

Grub Location 56 hours

Fig. 9. Ratesof movement of preoverwintering third-instar European chafers between 56 and 72 h when sodwas placed at the bottom of the microcosms.

Preoverwintering Japanese beetle grubs movedto and remained in the bottom of the microcosmsregardless of sod placement (Fig. 8). They behavedin a fashion identical to that segment of the bimodalpopulation distribution in the previous life stagethat did not contact sod in the "sod up" micro-cosms. This supports the hypothesis that Japanesebeetle movement down in the soil profile in thefall is regulated, at least in part, by the physiolog-ical state of the organism and is not merely a directresponse to changes in the environment.

Early- Winter Third Instars. The importance ofarrestment in the sod regardless of its placementin the microcosms reappeared for European cha-fers in the early overwintering stage (Fig. 10). Ex-amination of the movement rate data (Fig. 11)indicates population dispersion patterns are the re-sult of continued active movement of individualsrather than a wide distribution of stationary grubs.This is most clearly seen in the data for the "sodup" microcosms.

Early·winterthird instar

Hours

Fig. 10. Boxplotsdepicting the distribution of early-winter third-instar European chafers and Japanese bee-tles at 3, 24, 48, and 72 h after placement in microcosmswith sod at the top (sod up) or bottom (sod down).

c.2 1.5

~o011•••.! ~0.5<0 •':1:]<: .•••ec-.. ,..~ ~ -0.5::E

248 ENVIRONMENTAL ENTOMOLOGY Vol. 20, no. 1

late-winter third instar a European Chafer (late-winter) Sod Up

13.0

6.5

~~ 0

E -6.5U~ ·13.0_ 13.0a.Q)o 6.5

1 0

~ -6.5

·13.0

European Chafer Japanese Beetle

Fig. 13. Rates of movement of late-winter third-instar European chafers and Japanese beetles between56 and 72 h when sod was placed at the top of themicrocosms.

lower half of the microcosms exceeded the per-centage in the upper half at 24 and 144 h; however,95% confidence intervals for the estimated per-centages overlapped at 144 h, and the differenceswere therefore not considered significant. The grubsshowed no propensity to arrest in sod,

By comparison, Japanese beetle grubs were foundprimarily at the top of the microcosms, and theposition of the sod strongly influenced the binarydistributional pattern (at 24 h, X' = 5.78; df = 1;P = 0.016; at 144 h, X' = 15.48; df = 1; P = 0.00).When the sod was on the bottom of the microcosms,some Japanese beetle grubs (26.3% at 24 hand57.9% at 144 h) were found in the lower half;however, this never occurred when the sod was onthe top (Fig. 14). There was a trend for more grubsto be in the lower half of the microcosm at 144 h,even though a X' test indicated there was no effectof time on the distribution patterns (x' = 3.6; df =4; P = 0.45).

When the microcosms were rotated 900 from thevertical position (Fig. 15), the additive effect ofgravity and sod on the distribution of grubs couldbe examined because the distribution of the grubscould now be described using four categories (up,

13.0

o

o

oo

Japanese Beetle (late-winter) Sod Up

'l,

o.•..•..•. Do.. _0._ ~ .

o 0

Grub Locallon 56 hours

c:.2 15Ueooll.,ii~ '= 05lD ••

a:]1~.t:EN~ "';" ~O.5~:gIi:J...'> -1.5:g -JM.:

b

Hours

Fig. 12. Box plots depicting the distribution of late-winter third-instar European chafers and Japanese bee-tles at 3, 24, 48, and 72 h after placement in microcosmswith sad at the top (sad up) or bottom (sad down).

atures in the microcosms. The lack of response ofearlier third-instar stages to this warm environmentsuggests the existence of internal processes that pre-vent premature breaking of diapause. The inter-active effect of development stage and temperatureon life processes of grubs has been reported byother researchers. Ludwig (1928), working withJapanese beetle grubs, suggested that different stagesof the life cycle are not affected to the same degreeby temperature and that third instars could be par-titioned into distinct slow- and fast-developing cat-egories (Ludwig 1928, 1932); however, this couldnot be confirmed in later studies by Regniere et al.(1981a,b).

Studies with the grub Anomala cuprea (Hope)in Japan indicate that changes in soil temperaturecan modify development (Fujiyama & Takahashi1973a, 1977) as well as induce and terminate dia-pause (Fujiyama & Takahashi 1973b, Fujiyama1983). Wightman (1974) indicated that chilling isneeded to induce pupation of Costelytra zealan-dica (White), but larvae must reach some devel-opmental threshold before inducement will occur.Furthermore, temperatures must rise and exceedsome minimum threshold in the spring before theprocesses that lead to pupation will commence.

Postoverwintering Third Instars. Movement ofpostoverwintering third·-instar European chafers inthe vertical microcosms was downward to the bot-tom of the microcosms regardless of the placementof the sod (Fig. 14). The dispersion of the grubsdid not differ at 24 or 144 h when the animals wereclassified as being either in the upper or lower halfof the microcosms (X' = 2.71; df = 4; P = 0.607).However, when examined visually (Fig. 14), thedistribution of grubs was more diffuse at 144 h.The position of the sod had no effect on the binary(upper or lower half of the microcosms) classifi-cation of grub position at either time (at 24 h, thedistributions were identical; at 144 h, X' = 2.57; df= 1; P = 0.103). The percentage of grubs in the

February 1991 VILLANI & NYROP: MOVEMENT PATTERNS OF SCARABEID GRUBS 249

EUROPEAN CHAFER

• ••••••••••••••••••• •I_ Grub Po.llion I

• -:0..- • •, •· ••· }·0• •· • ~• .: • • #

JAPANESE BEETLE

••

,••

•• •,.•

••.:••• •• •

: .•.j.~ .

• 0

~,,

• ••••

: r• .e :...~.• 0... ~:.:

o.

EUROPEAN CHAFER

•••

•••

"-UIa: ••:z:..• ••~ • •

UIa::z:..•'"

ftt/I.••_0

o

oo 0

0000 000

o 0

o

JAPANESE BEETLE

• tP· 0 o•• .,~oo <b

o 0oo 0--.---- "- .

o 0

o

•

o

•

o•

o• •o

••.•••••••••• 0- ••• - ••••o 0 rf'•• ,. 0....

-&

o~ o.. ...._• .a ••• _•..•.••

• ••• •• 8 •"::>000· ·1>0oOoo'·Ab

a::z:..•:!

Ula::z:..•'"

Fig. 16. Position of individual postwinter third-in-star European chafers and Japanese beetles 24 and 144h after placement of grubs in horizontal microcosms.

down; sod, no sod). European chafer grubs wereindifferent to hosts and were positively geotropic.There was no difference in the distribution of cha-fer grubs at 24 and 144 h (X2 = 0.44; df = 3; P =0.93). At 144 h, 12.5% of the grubs were found ineach of the upper quadrats of the microcosms (withand without sod), and 37.5% were found in eachof the lower quadrats (with and without sod). The95% confidence intervals for these multinomial classprobabilities overlapped. If, however, a binary dis-tribution is considered, there were significantlymore grubs at 144 h in the lower half (75%) thanin the upper half of the microcosms.

Japanese beetle grubs responded negatively togravity and positively to sod by aggregating in theupper quadrats with sod at both 24 and 144 h.There was no difference in the distributions at thesetwo times (X2 = 4.07; df = 3; P = 0.254). At 144h, 75% of the Japanese beetle grubs were in theupper quadrat with sod, 15% were in the upperquadrat without sod, 10% were in the lower quad-rat with sod, and none was found in the lowerquadrat with no sod. These multinomial class per-centages suggest that a response to gravity and tosod combined in a positive manner to determinethe distribution of grubs. These results are not un-equivocal because the 95% confidence intervals forall but the largest class percentage overlap.

By placing the microcosms in a horizontal po-sition, the innate response of grubs to gravity ob-served in the vertical orientation and that rotated90" was physicallly negated because the sod andsponges were in the same plane (Fig. 16). European

I. Grub Position ISoo

•

.:eoooo

•• 0 ••.-:·. :..:o

• Grub Position 'Sod up'o Grub Position 'Sod down'

JAPANESE BEETLE

·• 0o·........................ ~_...............•.. ~-

•••

SOD

SPoNGE

EUROPEAN CHAFER

.;-..:

III ••~. ..:oo:t -.-------- ••. -~- •••• ---- •••• _....• . ,:•• •.1. • :•• ~ ••

• •• II. "

:.UI •a: ••:z: • 0 •

'It' •• _._-_ •• , •• _~.-: ••• __ ••••

('II. :_ ••..~ :...'...- :..," ..

Fig. 15. Positionof individual postwinter and third-instar European chafers and Japanese beetles 24 and 144h after placement of grubs in microcosms rotated 90".

SODFig. 14. Position of individual postwinter third-in-

star European chafers and Japanese beetles 24 and 144h after placement in vertical microcosms.

250 ENVIRONMENTAL ENTOMOLOGY Vol. 20, no. 1

chafers exhibited apparently random movement atboth 24 and 144 h, resulting in approximately equalnumbers of grubs on either side of the microcosmwith no indication of arrestment in sod. In theabsence of gravity, Japanese beetle grubs werefound predominantly in sod at 24 h (77.5%) and144 h (84.6%). Elimination of the influence of grav-ity on Japanese beetles still resulted in rapid ar-restment in the sod.

The initial movement downward by Japanesebeetles that was evident in prior stages was notapparent in this experiment. It is possible that grubmovement upward in the soil profile rather thandownward is adaptive in bringing grubs into theroot zone in spring. The data shown in Fig. 14suggest that Japanese beetles in this growth stageshow an innate negative geotropic response witharrestment when they contact the sod. In addition,when the grubs do not locate sod, they commencea more random search behavior throughout the soilprofile.

In contrast, European chafer grubs showed noresponse to the sod and exhibited only a positivegeotropic response. The most likely explanation forthis is that, before the experiment, the grubs hadmatured to the point where location of pupationsites was the primary factor driving movement pat-terns. By this time, European chafer grubs haveceased feeding and sod would have no influenceon behavior.

Soil moisture has been shown to influence themovement patterns of grubs (Villani & Wright1988); however, it is improbable that soil moistureplayed a role in these experiments. Percentage soilmoisture within the soil profile was virtually sym-metrical from the chamber midpoints (Table 3),although the end of the chambers containing thesod were somewhat drier than the side of the cham-bers containing sponges (range, 0.5-2.6%). Re-gardless of the treatment, grubs did not respondprimarily to moisture levels. Villani & Wright (1988)showed that Japanese beetles moved to areas ofhigher soil moisture, the opposite of what we ob-served. Clearly, Japanese beetle movement pat-terns at this stage were directed primarily by grav-ity and food source, whereas European chafers weredirected primarily by gravity.

Miller & Strickler (1984) asked what mechanismsand strategies shaped the patterns of resource useby insects in a heterogeneous world. They sug-gested that finding and accepting host plants resultsfrom a shifting balance of negative and positiveinternal and external inputs. The shifting balanceanalogy is applicable to the behavior of Europeanchafer and Japanese beetle grubs. Stage of devel-opment or prior experience shapes the response ofgrubs of both species to the various external stimulipresent in the microcosms. Internal and externalfactors may work in concert. For example, innatemovement downward in the profile upon distur-bance, the propensity to arrest in sod, and theplacement of sod at the bottom of microcosms pro-

duce behaviors with apparent directed movement.By contrast, conflicting internal and external cuesmay cause apparent nondirected movement andhighly dispersed populations as with innate down-ward movement, weak arrestment in the sod, andsod placement at the top of the microcosms.

Previous studies of European chafer and Japa-nese beetle grub movement in response to changesin the soil environment have focused on the directresponse of these species to changes in soil moistureand temperature. Data presented here indicate thatthe development stage of the grub has a large effecton Japanese beetle grub behavior and a measur-able, but lesser, effect on European chafers. Thisstage-related effect does not preclude a direct effectof the environment on grub behavior of both spe-cies should soil conditions dictate their need toescape noncyclical deterioration of their environ-ments (soil moisture or sudden and unseasonablechanges in soil temperature) (Regniere et al. 1981c).Understanding the fundamental differences in be-havioral response to soil factors among species ofsoil insects will enhance our ability to predict thestress each species can inflict on a crop and topredict the outcome of interactions between grubsand biological and chemical control agents.

Acknowledgment

We thank F. Consolie,L. Preston-Wilsey, Nancy Con-solie, Marylou Hessney, and Roseanne Consolie for tire-less technical support. P. Schroeder and C. Eckenrodemade numerous helpful suggestionsfor improving earlydrafts of this manuscript. We are grateful for partialsupport for this project through grants from the NewYork State Integrated Pest Management Program, theNew York State Turfgrass Association, Northeast Re-gional Project 169, and the Office for Research. CornellUniversity Experiment Station, Ithaca, N.Y.

References Cited

Fleming, W. E. 1972. Biologyof the Japanese beetle.USDA Technical Bulletin 1449.

Fujiyama, S. 1983. The larval diapause of three scar-abaeid beetles and its function in their life cycles, pp.149-166. In V. K. Brown & I. Hodek [eds.l,Diapauseand life cycle strategies in insects. The Hague.

Fujiyama, S. & F. Takahashi. 1977. The survival rateand the duration of development in Amomala cupTeaunder the thermal changes along with its develop-mental stage. Environ. Control BioI. 15: 19-25.

1973a. Studies on the self-regulation of the life cyclein Amomala cupTea. I. The effects of constant tem-perature on the developmental stages. Mem. Agric.Kyoto Univ. 104: 23-30.

1973b. Studies on the self-regulation of the life cyclein Amomala cuprea. II. The effects of low temper-ature and photoperiod on the induction and termi-nation of the larval diapause. Mem. Coli.Agric.KyotoUniv. 104: 31-39.

Gray, R. A. H., W. V. Peet & J. P. Rogerson. 1947.Observations on the chafer grub problem in the lakedistrict. Bull Entomol. Res. 37: 455-468.

Gyrisco, G. G., W. H. Whitcomb, R. H. Burrage, C.

February 1991 VILLANI & NYROP: MOVEMENT PATTERNS OF SCARABEID GRUBS 251

Logolhelis & H. H. Schwardt. 1954. Biology ofEuropean chafer, Amphimallon majalis Razou-mowsky (Scarabaeidae). Cornell University Agricul-tural Experiment Station Memoir 328, Ithaca, N.Y.

Hartzell, A. & G. F. McKenna. 1939. Vertical mi-gration of Japanese beetle larvae. Contrib. BoyceThompson Inst. 11: 87-9l.

Ludwig, D. 1928. The effects of temperature on thedevelopment of an insect (Popillia japonica). Physiol.Zool. 1: 358-389.

1932. The effect of temperature on the growth curvesof the Japanese beetle (Popillia japonica Newman).Physiol. Zool. 5: 431-447.

McGill, R., J. W. Tukey & W. A. Larsen. 1978. Vari-ations of box plots. Am. Stat. 32: 12-16.

Miller, J. R. & K. L. Strickler. 1984. Finding andaccepting host plants, pp. 127-157. In W. J. Bell &R. T. Carde [eds.],Chemical ecology of insects. Chap-man & Hall, New York.

Regniere, J., R. L. Rabb & R. E. Stinner. 1981a.Popillia japonica: seasonal history and associatedScarabaeidae in eastern North Carolina. Environ. En-tomol. 10: 297-300.

1981b. Popillia japonica: simulation of temperature-dependent development of immatures, and predic-tion of adult emergence. Environ. Entomol. 10: 290-296.

1981c. Popillia japonica: effect of soil moisture and

texture on survival and development of eggs and firstinstar grubs. Environ. Entomol. 10: 654-660.

Tashiro, H. 1987. Turfgrass insects of the UnitedStates and Canada. Cornell University Press, Ithaca,N.Y.

Tashiro, H. & F. L. Gambrell. 1963. Correlation ofEuropean chafer development with the flowering pe-riod of common plants. Ann. Entomol. Soc. Am. 56:239-243.

Tashiro, H., G. G. Gyrisco, F. L. Gambrell, B. J. Fiori& H. Breitfeld. 1969. Biology of European chafer,Amphimallon majalis Razoumowsky (Scarabaeidae)in northeastern United States. New York AgriculturalExperiment Station Bulletin 1366, Geneva.

Villani, M. G. & R. J. Wright. 1988. Use of radiog-raphy in behavioral studies of turfgrass-infesting scar-ab grub species (Coleoptera: Scarabaeidae). Bull. En-tomol. Soc. Am. 34(3): 132-144.

1990. Environmental influences on soil macro-arthro-pod behavior in agricultural systems. Annu. Rev. En-tomol. 35: 249-269.

Wightman, J. A. 1974. Influence of low temperatureon pupation induction in Costelytra zealandica (Co-leoptera: Scarabaeidae). New Zealand Journal of Zo-ology, 1: 503-507.

Received for publication 11 May 1990; accepted 16August 1990.