Area-Wide Pheromone Trapping of sea and phloxiphaga (: ) in the Sacramento and San Joaquin Valleys of California

MICHAEL P. HOFFMANN,' LLOYD T. WILSON,2 AND FRANK G. ZALOM Department of Entomology, University of California, Davis, California 95616

J. Econ. Entomol. 84(3): 902-911 (1991) ABSTRACT In 1986 and 1987, traps baited with the pheromone of Hellcoverpa zea (Bod- die) were operated adjacent to fields of processing tomatoes (Lycopersicon escufentumMiller) in the Sacramento and San Joaquin valleys of California to determine the seasonal flight patterns of male H. zea and Hehothis phfoxtphaga Grote & Robinson. The latter species was attracted to the pheromone of H. zea and was commonly captured in traps. Results indicate there are at least two generations of H. zea per year in the Sacramento and San Joaquin valleys. In general, trap catches of H. zea increased rapidly starting in mid-August and peaked during late August or early September. The timing of trapping events, defined as first catch, first peak, initiation of late peak, and late peak, was similar among locations monitored. This indicated that regional monitoring of this pest may be feasible, although further effort is needed to define early-season events better. Males of H. phloxiphaga were captured at all locations; these were generally trapped in greatest numbers in May and usually exceeded the numbers of H. zea captured from February to June. The capture of H. phloxiphaga in traps intended for H. zea can make interpretation of trap catches difficult, especially early in the season. Moon phase did not appear to influence the timing of peaks in H. zea trap catch.

KEY WORDS Insecta, Noctuidae, pheromones, tomato

SEVERAL AREA-WIDE BLACKLIGHT and pheromone ever, their efforts were limited to one location in monitoring programs have been undertaken to in- the Sacramento Valley. vestigate the seasonal flight patterns of HeldothZs The density of H. zea eggs generally begins to spp. Information generated from area-wide mon- increase in early August and peaks 2-3 wk later itoring programs is useful in determining the sea- across most fields of processing tomatoes (Lyco- sonal occurrence (Goodenough et al. 1988), move- persicon esculentum Miller) in the Sacramento ment (Schneider et al. 1989), migration (see review Valley (Hoffmann et al. 1990). Roltsch & Mayse by Fitt 1989), and spatial distribution (Slosser et al. (1984) reported similar across-field trends in to- 1987) of Heliothis adults. Such information is use- matoes in Arkansas. Based on results from a single ful to regional pest management programs by help- season, it appears that oviposition by H. zea in ing to predict when and where infestations may processing tomatoes in the southern Sacramento occur or indicating when management tactics need Valley is positively correlated with pheromone trap to be implemented. Witz et al. (1985) found that catches of male H. zea during the period of late pheromone trap catches recorded early in the sea- July (initiation of late-season oviposition) through son were useful in predicting the subsequent sea- peak oviposition (unpublished data). A similar re- sonal population dynamics of lationship between pheromone trap catches and (Boddie) and Helicoverpa virescens (F.)in cotton. oviposition by Heliothis spp. in cotton has been Little effort has been made in the San Joaquin or reported by Leonard et al. (1989). Because of sim- Sacramento valleys of California to monitor adult ilarities in across-field timing of oviposition in to- male H. zea using pheromone traps on an area- matoes and because of the apparent correlation wide basis, although such data could be useful in between trap catch and oviposition, it may not be improving regional management of this pest on necessary to operate traps in every field. A few several crops. Lange & Bronson (1981) reported traps may be adequate to monitor counties or that adult male and female H. zea catches in black- regions of the state. lights increased rapidly late in the season. How- Currently, pest control advisors begin sampling for eggs of H. zea at 722 degree-days (DD) (lower threshold of 10%) after planting (Brendler et al. ' Current address: Department of Entomology, Comstock Hall, 1985). This sampling commences even though it is Cornell University, Ithaca, N.Y. 14853. Current address: Department of Entomology, Texas A&M generally only those fields harvested after mid- University, College Station, Tex. 77843. August that are likely to incur economic levels of

0022-0493/91/0902-0911$02.00/0Q 1991 Entomological Society of America

~ June 1991 HOFFMANNET AL.: AREA-WIDEPHEROMONE TRAPPING OF Heliothis 903

phloxiphaga, and if H. phloxiphaga would be cap- tured in sufficient numbers to interfere with the interpretation of H. zea catches.

Red Bluf? Materials and Methods chlco 4 1 Princeton Traps and Locations. Scentry Heliothis traps I Arbuckle * Woodland (Scentry Inc., Buckeye, Ariz.) (after Hartstack et Davis al. 1979) were used at all locations and were at- tached to metal stakes or fence posts, with the trap entrance about 1 m above soil level. were removed from traps at least once per week. Traps were operated by U.C. farm advisors, commercial pest control advisors, or by personnel from the De- partment of Entomology, U.C. Davis. Cooperators placed captured moths in small jars containing 75% ethyl alcohol and sent them to U.C. Davis where male moths were identified to species and numbers of individuals were recorded. H. zea were distin- guished from H. phloxiphaga using characteristics described by Lange & Michelbacher (1937). Pheromone lures for H. zea consisted of rubber septa impregnated with a 3.0:0.09-mg dichloro- methane solution of Z11-16:Al and Z9-16:Al (Halfhill & McDonough 1985). Lures were pre- pared at Yakima, Wash. (USDA-ARS), and shipped Fig. 1. Map of California indicating locations at to us where they were stored at -9°C until needed. which Scentry Hellothis traps \%'ere operated in 1986 Septa were placed in the center of the large open- and 1987. In some cases, more than one trap was operated ing at the bottom of the traps and were shaded in the vicinity of the location indicated on the map. See from direct sunlight by placing them in an open- text for a complete listing of trap locations. ended section of white polyvinyl chloride (PVC) pipe (2.54 cm long, 1.91 cm diameter). In 1986 and 1987, traps were generally installed damage from N. zea (Lange & Bronson 1981). in April or May and operated until September or iiowever, if N.zea pheromone traps could be used October. Sites extended from near the southern end 10 predict or indicate the timing of activity on a of the San Joaquin Valley to near the northern end regional basis, then initiation of sampling for H. of the Sacramento Valley and encompassed a direct :pa eggs in processing tomatoes could be better north-to-south distance of about 544 kin (a range limed. from about 35" to 40" N latitude) (Fig. 1). This A potential problem in using pheromone traps investigation was originally intended to be limited for H. zcw is the unwanted capture of males of to areas where processing tomatoes were grown; Ildiothis phloxiphaga Grote & Robinson, a widely however, cooperators were available at the two (hibuted species in California (Lange & Mich- northernmost locations which were outside of the qlbacher 1937). Kaae et al. (1973) reported capture typical processing tomato production area. Traps this nonpest species in traps baited with virgin operated in 1986 or 1987 or both were located in f"de H. zea. Raina et al. (1986) determined the the following locations (km north of Arvin): Red thical constituents of the p~ieromone of H. Bluff (5.14). Chico (197), Princeton (473), College I'hlo*iphaga and further examined cross-attractan- City (426), Arbuckle (131),N.E. Yolo County (410), between these two species. Although the adult Woodland (399), E. Yolo County (396), East Davis of H. zea and H.phloxiphaga are morpho- (379), U.C. Davis (378), King Island (325), Fire- '%ically distinct (Lange 6: h4ichelbacher 1937), baugh (182), Mendota (16.i), U.C. Westsicle Field between these species becomes dif- Station at Five Points (134), Huron (114), and Arvin. ficult after moths have been in a pheromone trap The traps located at King Island and south were lor a few days. After a few days, most scales are in the San Joaquin Valley; all others to the north abraded away and identification can be accom- were in the Sacramento Valley. Traps located at p lished only by examination of male genitalia. Arbuckle and north were designated to be in the The objectives of this research were to determine northern region, and those at Firebaugh and south Ihe Seasonal flight activity of male H.zea in the were designated to be in the southern region. The Sacramentoand San Joaquin valleys of California, remaining traps were in the central region. hilar the H. zea activity was over the area Although the number and specific sites of traps the seasonal flight activity of male H. varied between years, traps were always located 904 JOURNAL OF ECONOMIC ENTOMOLOGY Vola 84, no. 3 near the locations described above. The U.C. Davis 1986 and May and June 1987, four locations during location had two traps spaced about 300 m apart; July 198i, and three locations during other months In the second analysis, the frequency of t all other locations had a single trap. Traps were rap placed within commercial agricultural areas, but catch peaks occurring during each moon phase to prevent damage from agricultural equipment, (full, last quarter, new, first quarter) were corn. they were not placed directly in crops. The agri- pared using a x* test. Unsmoothed data from boltl cultural and nonagricultural habitats at or sur- years and all locations were used. Trap catch peaks rounding the trap sites varied considerably among were defined as any observation where the numbe, sites. For example, most San Joaquin Valley traps of moths per day exceeded the number of moths were operated in a mixed cropping system domi- per day recorded on the previous and follo\cjng nated by cotton and processing tomatoes. Those in observations. Another, similar analysis was per. the northern Sacramento Valley were in a mix of formed using the same data except to eliminate tree crops, processing tomatoes, rice, cucurbits, and some of the minor fluctuations in trap catch; peaks a variety of other crops. Traps at U.C. Davis were less than or equal to one per day were er- located near a variety of experimental plots, of cluded. These two data sets were also separater1 which most were tomatoes. With the exception of into two seasonal periods (May, June, July and Au. Chico, Red Bluff, and King Island, all traps were gust, September, October) to determine if the nloon located adjacent to plantings of processing toma- phase effect was more apparent during the early toes. versus the late season. Each time period enco“. Unless stated otherwise, catches of male H.zea passed three complete lunar cycles. Similar anal- ancl H. phloxiphaga were converted to trap catch yses were performed on these two data sets. per day and smoothed using a three-observation Location. To determine if seasonal patterns in running mean. Where trap catch records are pre- capture of male H. zea were related to location sented as means of several locations, the mean is (distance south to north), the timing (day of year) based on all records for the week and each point of four readily defined trapping events were re- on the graph corresponds to Wednesday. Not all gressed against distance (kilometers) from the traps commenced or terminated operation at the southernmost location at Arvin. A sitnple linear same time each year; consequently, means or pro- regression analysis was conducted using Cricket portions depicted in figures are not always based Graph 1.2 (Cricket Software 1988). The first trap- on the same number of traps throughout the year. ping event was the “first catch” of the season. This Locations at which moth catch per day never ex- was defined as the first catch of the season that was ceeded 0.5 during the season are not presented. If preceded by at least one observation without H. traps were inoperable for >2 wk (two weekly ob- zea and followed by at least two observations with servations missed), then this time interval plus the H. zea catches. The second trapping event was preceding and following observation were not used referred to as “first peak” of the season. This event in analyses. occurred on the first date of the season when the Moon Phase. Before attempting to associate pat- pattern of trap catch fit the definition of peak as terns in trap catch with the phenology of H. zea, given above. First peaks were obtained only at it was important to determine if the phase of the locations where traps were in operation by 1 May. moon influenced trap catch as clemonstrated for The third trapping event, referred to as “initiation H. virescens by Hartstack el al. (1978). Two anal- of late peak,” occurred on the date of lowest trap yses were performed to assess this influence on trap catch inimediatel) before the late-season peak or, catch In the first analysis, mean catch per trap per proceeding back in time from the late-season peak, clay (unsmoothed data) on the day of full moon the first date at which the trap catch per day did was compared with mean catch per trap per day not decrease by more than one moth per cla~,. This at new moon within each month using two-way event identibed the sudden increase in trap catch analysis of variance (ANOVA) and treating the that usually preceded the fourth trapping event, experiment as a randomized complete-block (lo- which is referred to as “late peak.” This event cation) design. The locations selected for this com- occurred on the date of the largest trap catch after parison included three at Davis and one at the U.C. 1 August. Unsmoothed data were used lo establish Westside Field Station. These locations were used dates of “first catch,” ancl all other events were because they were generally operated longer each based on smoothed data. Where regressions were year and because trap catches were recorded more not significant, a 95%confidence interval about the often (usually twice per week) than at other loca- d‘iy of !ear for the individual sites was calcrilated tions. If a trap catch was not recorded on the day for each event; i e., on average 95% of the dates of full or new moon, then catch per day was in- fell within the confidence interval. terpolated based on the catch per day at the ob- Temperature. Maximum and minirnum tem- servation just before and following the full or new perature data were obtained from 10 weather sta- moon. If any observation interval for a trap ex- tions located near the trap sites and operated for ceeded 7 d, the data for both moon phases for that the University of California Statewide Integrated month were not used. These criteria allowed us to Pest Management (UCIPM) Project. The locations use two locations for comparisons Juring October from n hich temperature data were ohtainecl in- June 1991 HOFFMANNET AL.: AREA-WIDE PHEROMONE TRAPPING OF HdiOi?hiS 905

lo'

80-

40 -

Bo- 40 -

10 -

"i'PMAMJ J A 801 Month Month a0

10

0 4

a

0

a

1

c

E

-.FM A M J J A 8 0 N l vi Month Figs. 2a-c. Seasonal trap catches of male H.zea by location in California in 1986. Full moon phase indicated at top of figure. cluded Red Bluff, Chico, Princeton, College City, (DD) were calculated for each location from these Woodland, U.C. Davis, King Island, Mendota, U.C. temperature records using the UCIPM computer Westside Field Station, and Arvin. Some stations system. The lower and upper developmental represented more than one trap site. Degree-days thresholds of H. zea of 13.9 and 34.5"C,respec- JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 84, no. 3 : 906

0 0 FMAMJ J ASON FMAMJJASON Month Month

1987 H. tea 987 ~.zeaO O O O O O N. UCD :m 8. UCD

E. Dad

ia

0 King Iilnad

01 ' . FMAMJJA8,ON , Month

Month Figs. 3a-d. Seasonal trap catches of male H. zeu by location in California in 1987. Full moon phase indicated at top of figure. June 1991 HOFFMANNET AL.: AREA-WIDE PHEROMONE TRAPPING OF fkblhiS 907

1986 1987 '(")I a b A First Catch ( f I 128.2. CI f 22.3) n I 6 A Firstcatch (126)n- 1 0 0 First Peak ( I 160.8, CI f 14.7) n I 9 First Peak (TI136, CI f 19.7) n I 6 175 t - t I

0 AA A 0 A 125 - 8 A A G 4 u: I.1.t.I. 1. 0 o Late Peak (fI 241.4, CI f 16.3) n = I5 0 Late Peak( T1243.6, CI f 18.2) n I 15 b 275 A hit.Late Pk. (fI 212.2, CI f 24.7) n I 14 A Init. Late Pk. (VI217.6, CI f 22.1) n I 14 fa 0

goo O A0 'lo OO 2501226 O0 O0 A A AA A A A A 1751.t.*.1 .**'*1*1 '1. '.*SI- 0 100 200 300 400 600 100 200 300 400 600 Kilometers Fig. 4. Day of year of occurrence for four trapping ev-ents observed in 1986 (a) and 1987 (b), relative to clistance from south (0 km = Arvin) to north. Mean day of year (Y)for each event and 95% confidence interval (CI) about the day of year at single sites. tively, were used (Wilson & Barnett 1983). Accu- zea infestations in processing tomatoes (Brendler mulated degree-days for the period 1 May-30 Sep- et al. 1985), especially in the Sacramento Valley, tember were regressed against distance south to where much of the acreage is not harvested until north to determine if there was a gradient in ac- September. Processing tomatoes grown in the San cumulated degree-days over the area monitored Joaquin Valley are generally harvested early enough with H. =ea traps. This time increment was selected to avoid these infestations. Using blacklight traps, because it encoinpassed the majority of the season Lange & Bronson (1981) reported similar seasonal during which H. zea adults are active. patterns of capture of H. zea over a 10-yr period To determine if the observed trapping patterns at Davis. The number of H. zea captured per clay reflected the number of generations of H. zea, ac- varied considerably among locations. The maxi- cumulated degree-days were calculated between mum catch per clay for the season ranged from <3 first catch or first peak and subsequent events. For at Huron to near 100 at Davis. rxample, the degree-days accumulated between first The number of early-season trapping events ob- catch and late peak were calculated and compared served at some locations was relatively low (Fig. with the expected degree-days required to com- 1) because traps were started late or because trap ~hteone or more generations. A t test was used to catch patterns did not fit the events as defined lest nhether the calculated mean number of de- above. Early-season information (first catch, first gree-days was significantly different from the ex- peak) from the southern and central San Joaquin I'ected number of degree-days for one or more Valley is limited to that from the U.C. Westside generations of H. zea. The duration of a generation Field Station. Six dates of first catch were observed (egg to egg) \\.as considered to be 538 DD (Wilson in 1986. The earliest date of first catch occurred ti Barnett 1983). on 25 April and the latest on 16 May. The mean date of first catch in 1986 was about 8 May (mean day of year rounded to nearest whole number and Results and Discussion converted to calendar date). A single date of first Welicoverpa zea. In 1986 and 1987, 15 and 17 catch was recorded on 6 May 1987. It appears !hat trapping sites, respectively, produced adequate these first-catch observations reflect the emergence data. hlore sites were established each year than of adults from diapause. Blanchard (1942) reportetl are reported here, but because of failure of coop- overwinter survival of H. =ea pupae in areas north eratofs to adequately maintain traps or because of the 40th parallel, indicating that diapause is '"Sufficient data were collected from sites, they are possible over the entire area encompassed in this not included. The largest number of H. zea males trapping study. At Davis, Velez (1970) reported generally captured in August and September emergence of adult H.zea from overwintered P11- (Fig. 2a-c, 3a-d). This corresponds to the time of pae as early as 4 April to as late as 1 Jtlne. Eight Year with greatest potential for damage from H. of the nine first peaks observed in 1986 were re- 908 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 84, no. 3 corded in the Sacramento Valley. Of these eight 2000 - peaks, the earliest occurred on 28 May and the n Q 1986 @=12615, CEt320.0) latest on 6 June. The other first-peak record was 1750 1987 @=1337.6, CM93.1) on 16 June at the U.C. Westside Field Station. In $3 1987, five first peaks were recorded. The earliest 8 !gJ 6 15OOlC PI first peak was on 9 May and the latest was on 28 n8 CI ** *+ May. The mean date of first peak occurred on 31 a 1250 - QD om May in 1986 and 16 May in 1987. $3 Fourteen and fifteen dates of initiation of late 4 r( 1000- L peak and late peak, respectively, were observed each year. The first date of initiation of late peak 3 750 2 I I t 1 I occurred on 3 July and the last on 20 August in ' 1986. In 1987, the first and last date of initiation of late peak was on 16 July and 26 August, re- spectively. The mean date of this event in 1986 and 1987 was about 31 July and 5 August, respec- tively. The mean date of late peak was 29 August in 1986 and 1 September in 1987, respectively. Late peaks occurred as early as 15 and 22 August and as late as 8 and 28 September in 1986 and 1987, respectively. The regression analyses of the day of year of first H. zea catch (1986 only), first peak, initiation of late peak, and late peak against distance of the trapping site north of Arvin were not significant in either year (P > 0.05) (Fig. 4). These results in- dicated little relationship between latitude and sea- sonal timing of male H. zea as reflected in pher- omone trap catches. Goodenough et al. (1988) reported a significant correlation between latitude and early-season trapping of H. xea and H.vires- cens. However, their investigation encompassed a much greater south-to-north distance. In general, 95% confidence intervals estimated for single sites and for each event are relatively broad. However, within subregions, variation in timing of events is relatively small. For example, first peak in 1986 and late peak in 1987 in the Sacramento Valley (King Island and north) all occurred within an 11-d period. Like the trapping events disciissecl above, the regression analyses of degree-days accumulated (1 May-30 September) against distance south to north was not significant (P > 0.05) (Fig. 5). The two southernmost locations were consistently warmer than the other sites, and King Island and Davis were consistently cooler. The King Island and Da- vis locations were most likely cooler because of the influence of ocean breezes. Altitude, which ranged from 4.0 rn at King Island to 115 m at Arvin, may haye accounted for some of the variation in ac- cumulated degree-days among locations. Considering that 538 DD are required per gen- eration, these results show that there could be at least two complete generations of H. zea occurring each season in the central and northern San Joaquin and Sacramento valleys. In addition, there would also be sufficient time (220 DD) for eggs laid during the late-season peak to develop to the pupal stage (Wilson & Barnett 1983) and to overwinter. The degree-day accumulations shown in Fig. 5 are for 1 May-30 September. Depending on the location, -

HOFFMANNET AL.: AREA-W IDE PHEROMONE TRAPPING OF ffeliOthiS 909

1OC Helicoverpa zea 1986 Full Moon 80 T

60

40

20

0 1987

60 T

40 T

20 I T

0 May June July August Sept. Oet. Month Fig. 6. Male H.zea trap catch in pheromone traps relative to either full or new moon phase in 1986 and 1987. VWical bars represent standard error of the mean. late-season activity commences and thereby better during the new moon phase, whereas in 1987 the determine when sampling for H.zea eggs should opposite results were obtained. Results from the begin. second analysis (~2)showed that the proportion of Moon Phase. Results of the first analysis of the all peaks (n = 163) occurring during the full, first- influence of moon phase on trap catch showed that quarter, new, and second-quarter moon phase were the mean catch per day at full and new moon 0.25G, 0.287, 0,220, and 0.238, respectively. When Phases did not differ significantly within months peaks equal to or less than one moth per day were except during September (Fig. 6). In September removed, the distribution of peaks (n 114) was 1986, mean catch per day was significantly higher 0.246,0.272,0.237, and 0.316, respective!y. In nei- 910 JOURNAL OF ECONOMIC ENTOMOLOGY Vola 84, no. 3

9 1.0 1086 1986 - Northern 0.8 - Northern ...... _---_ SoulhernCaatrnl

OA 9b 0.4

L..B 0.2 9 2 0.0 19w I 0.8 fi 3- n - Northern C"tnl E i3 ...... subhem g 0.6 2. f 0.4

1. ,/ 0.2 ,a' 0.0 \ -*.I\ FMAMJJASON FMAMJJASON Month Month Fig. 7. Mean seasonal trap catches of male H. phlox- Fig. 8. Male H. phbxiphaga trap catch expressed as iphagn by region of state. proportion of total (Heliothis phloxiphaga plus Helico. oerpa zea) captured in 1986 and 1987 by region of Gal. ifornia. ther case did these distributions differ significantly from random (P > 0.05). The distributions of peaks phloxiphaga was present as early as February ant1 during the early (May, June, July) and late (August, that numbers caught gradually increased until ear- September, October) periods using all data or after ly May. Lange & Michelbacher (1937) reported H. removal of small peaks were also random. phloxiphaga was most abundant early in the season These results indicate that moon phase did not in central California. No other discernible peaks in influence the capture of H. zea males in pheromone trap catch were apparent except at U.C.Westside traps. Dent & Pawar (1988) reported similar results Field Station where, in early August, the number when trapping Helicouerpa nrmtgera (Hubner) of male H. phloxiphaga trapped per day was near with pheromone traps. Hartstack et al. (1978), how- that recorded in May and June. ever, reported a direct relationship between cap- Heliothis phloxiphaga adults can be distin- ture of H. vfrescens in pheromone traps and phase guished from those of H.zeu by differences in color of the moon. Trap catches were higher during the patterns on the wings and because of the smaller full moon phase. In both of these investigations, size of H. phloxiphaga. However, after they are traps were checked hourly or daily. If, in our in- captured in pheromone traps, distinguishing be- vestigation, traps had been checked more frequent- tween these species becomes difficult and generally ly, then possibly the influence of moon phase, if requires the dissection of male genitalia. The po- real, would have been inore apparent. Because we tential for H. phloxiphaga catches to interfere with did not know which peaks were associated with interpretation of H. zea trap catches appears great- the phenology of H. zea, these analyses do not show est in May, June, and possibly July, when both the extent to which such peaks were modulated by species are present. Frequently, >50% of moths moon phase. captured during this time are H. phloxiphaga (Fig. Heliothis phloxiphaga. Heliothis phloxiphaga 8). males were captured at all locations in both years These results indicate that care should be taken (Fig. 7).indicating that H. zea and H. phloxiphaga to identify moth species captured in traps baited are sympatric throughout the region trapped. Raina with the pheromone of H. zea, especially early in et al. (1986) estimated that cross-attractancy of H. the season in the San Joaquin and Sacramento Val- zea males by H. phloxiphaga virgin females was leys of California. Because of the time that may about 5%. However, because of declining H.phlox- be required to differentiate these two species (i.e., iphagn populations, those authors were unable to once the wing scales have been removed), com- test the response of H. phloxiphaga to N.zea virgin mercial use of pheromone traps for early-season females or pheromone. monitoring of H. tea may be limited. This would Heliothis phloxiphaga catches were generally be especially critical if the dates of first catch or greatest in late May and early June (Fig. 7) and first peak (both of which occur when H. phloxi- consequently were temporally separated from the phnga are most abundant) were ultimately used as period of highest H. zea catches. Because traps points of reference for predictive purposes. In other were generally not operational until late April or geographic areas, the identification of Noctuidae early May, little information was gathered on ac- captured in pheromone traps baited for H.zea ma?' tivity earlier in the year except at U.C. Davis in be important. H. zea and H. phloxiphaga are sym- 1987. Records from one trap operational at this patric over much of the United States. Chmb (19%) location beginning 5 January indicated that H. reported that the distribution of H. phloxiplragfl June 1991 HOFFMANNET .4L.:A REA-WIDE PHEROMONE TRAPPING OF Neblh$s 91 1

jrdudes 11 states (Illinois to California and south) Hartstack, A. W., J. A. Witz & D. R. Duck. 1979. plus British Columbia and Alberta, Canada. Im- Moth traps for the tobacco budworm. J. Eoon. En- proving the specificity of pheromone traps baited tord 72: 519-522. for H. zea by the addition of an H. phloxiphaga Hoffmenn, M. Y., L. T. Wilson, F. C. Zdom & R. J. repellent or by other means would be a valuable Hilton. 1990. Parasitism of Hellothis zea (Lepi- doptera: Noctuidae) eggs: effect on pest management goal. decision rules for processing tomatoes in the Sacra- mento Valley of California. Environ. Entomol. 19: ,753-763. Acknowledgment Kaae, R. S., If. If. Shorey, S. U. McFarland & I*, K. The authors acknowledge Les Ehler and two anony- Caston. 1973. Sex pheromones of Lepidoptera. mous reviewers for their critical review of an earlier XXXVII. Role of sex pheromones and other factors draft of this manuscript; Les McDonough (USDA-ARS, in reproductive isolation among (en species of Noc- Yakima, Wash.) for supplying H. zea pheromone lures; tuidae. Ann. Entomol. SOC.Am. 66: 444-448.. and Tom Eichlin (California Department of Food and Lange, W. if. & L. Rronson. 1981. pests of Agriculture, Sacramento) for confirming identification of tomatoes. Annu. Rev. Entomol. 26: 345-371. H. phloxiphaga. We also sincerely thank the U.C.Co - Lange, W. H. & A. E. Michelbacher. 1937. TWO operative Extension Farm Advisors Pete Goodell, Rich closely related species of Heliothis found in tomato Coviello, hlike Murray, and Don Rough and the several fields of central California, pp. 320-325. In California pest control advisors who operated traps for this project. Department Agriculture Bulletin 26, Sacramento. This research was supported in part by the University Leonard, B. R., J. B. Craves, E. Burris, A. M. Pavloff of California Statewide IPM program. & G. Church. 1989. Heliothis spp. (Lepidoptera: Noctuidae) captures in pheromone traps: species composition and relationship to oviposition in cotton. References Cited J. Econ. Entomol. 82: 574-579. Raina, A. K., J. A. Klun, J. D. Lopez & B. A. Leonhardt. Blanchard, R. A. 1942. Hibernation of the corn ear- 1986. Female sex pheromone of Heliothis phloxi- worm in the central and northeastern parts of the phaga (Lepidoptera: Noctuidae): chemical identifi- United States. US Dep. Agric. Tech. Bull. 838. cation, male behavioral response in flight tunnel, and Brendler, R. A., B. B. Fischer, D. H. Hall, D. M. May field tests. Environ. Entomol. 15: 931-935. & N. C. Toscano. 1985. Integrated pest manage- Roltsch, W. J. & M. A. Mayse. 1984. Population stud- ment for tomatoes. University of California Statewide ies of Hehothis spp. (Lepidoptera: Noctuidae) on to- Integrated Pest Management Project, Division of Ag- mato and corn in southeast Arkansas. Environ. En- riculture & Natural Resources Publication 3274. tomol. 13: 292-299. Cricket Software. 1988. Cricket graph, Version 1.2. Schneider, J. C., R. T. Roush, W. F. Kitten & M. C. Malvern, Pa. Laster, 1989. Movement of Heliothis virescens Crumb, S. E. 1956. The larvae of the Phalaenidae. (Lepidoptera: Noctuidae) in Mississippi in the spring: US Dep. Agric. Tech. Bull. 1135. implications for area-wide management. Environ. Dent, D. R. & C. S. Pawar. 1988. The influence of Entomol. 18: 438-446. moonlight and weather on catches of Helicoverpa Slosser, J. E., J. A. Witz, G. J. Puterka, J. R. Price 81 urmigeru (Hiibner) (Lepidoptera: Soctuidae) in light A. W. Hartstack. 1987. Seasonal changes in boll- and pheromone traps. Bull. Entomol. Res. 78: 365- worm (Lepidoptera: Noctuidae) moth catches in 377. pheromone traps in a large area. Environ. Entomol. Fitt, C. P. 1989. The ecology of Heliothis species in 16: 1296-1301. relation to agroecosystems. Annu. Rev. Entomol. 34: Velez, C. R. 1970. Ecology of the corn earworm, 17-52. Hellothis zeu (Boddie), in northern California: life Coodcnough, J. L., J. A. Witz, J. D. Lopez & A. w. table studies and some aspects of pupal diapause. Hartstack. 1988. Patterns of occurrence of He- Ph.D. dissertation, University of California, Davis. liothis spp. (Lepidoptera: Noctuidae), 1983-85. J. Wilson, L. T. & W. W. Barnett. 1983. Degree-days: Econ. Entomol. 81: 1624-1630. an aid in crop and pest management. Calif. Agric. 11alfhil1, J. E. & L. M. McDonough. 1985. Hpliothis 37( 1-2): 4-7. zru (Roddie): Formulation parameters for its sex Witz, J. A., A. W. Hartstack, E. G. King, W. A. Dick- pheromone in rubber septa. Southwest. Entomol. 10: erson & J. R. Phillips. 1985. Monitoring and pre- 176- 180. diction of Hehothis spp. Southwest EntOmoI. SUPPI. Ilartstack, A. W., J. P. Hollingsworth, J. A. Witz & D. 8: 56-70. R. Buck. 1978. Relation of tobacco budworm catches in pheromone baited traps to field popula- Received for publication 14 December 196'9; acceptPd tions. Southwest. Entomol. 3: 43-31. 24 January 1991.

-- I