Distribution of hesperus (Knight) (: ) on Cotton

L. T. WILSON,’ T. F. LEIGH,l D. GONZALEZ,* AND c. FORISTIERE3 University of California

J. Econ. Entomol. 77: 131.3-1319 (1984) ABSTRACT (Knight) on cotton in the field were, on the average, distrib- uted on the fifth through seventh nodes from the terminal. Most nymphs were located on squares, while the majority of adults were found on bolls In comparison, adults 011 glasshouse plants were located predominantly in the upper four mainstem nodes; most were found on leaves. In the glasshouse, L. hesperus that were on fruit structures were almost entirely on squares. The difference in distribution between L. hesperus populations in the field and glasshouse appeared to be largely due to different plant growth patterns. Glasshouse plants were spindly, with a simple canopy characterized by short branches, few fruits, and rela- tively little vegetation. Plants in the field developed a full canop) , larger and more branches, and more fruit per branch, thus providing more habitat turther down the plant for L. hesperus to feed. These results suggest that glasshouse experiments provide biased infor- mation on L. hesperus distribution and, possibly, feeding behavior.

Lygus hesperus (Knight) is a pest of cotton in Cal- periments involving confinement or habitat mod- ifornia’s San Joaquin Valley. Feeding is reported ification, such as using a cage or rearing in a to occur primarily in fruiting structures in the up- glasshouse or growth chamber, however, has been per terminal area of the plant (Jubb and Carruth limited (Latson 1972). 1971, Tugwell et al. 1976, Gutierrez et al. 1977, Here we report the results of a study done to Mauney and Henneberry 1979); young squares examine how the distribution of L. hesperus on (flower buds) are preferred (Mauney and Henne- cotton is affected by fruit availability and by berry 1979, Pack and Tugwell 1976). Tugwell et activity, and to compare results obtained in the al. (1976) estimated for two cultivars that about field and the glasshouse. 10 and 1%of the potential yield in their field trials was lost due to Lygus spp. damage to bolls. The actual amount of feeding on bolls was possibly less Methods and Materials and that on squares greater since, when older Field and glasshouse experiments were con- squares are damaged by Lygus bugs, development ducted with the cotton cv. Acala SJ-2. of the fruit often continues but results in a dam- In 1974, insect distribution on all plants within aged boll (Pack and Tugwell 1976). The amount 24 rows 1 m long (embedded within 0.5 ha of of damage to bolls is also likely to depend on the cotton) was mapped from 0800 to 1400 hours at relative phenologies of the crop and the Lygus five time periods through the season near Shafter, spp. Large early season populations probably cause Calif. On two additional sample dates, plants from more damage to squares, while populations devel- a commercial field were mapped near Bakersfield, oping later probably damage more bolls. Since bolls Calif. The information recorded included the fol- are located further from the plant terminal than lowing for each observed L. hesperus adult and are squares, the distribution of Lygus bugs on the nymph: mainstem node location (MSN)-for these plant may affect the relative availability of squares analyses, the first partially unfurled mainstem leaf and bolls as feeding sites. in the terminal equals position 1; branch node lo- Lygus adults are flighty; for this reason, most cation (BSN), and whether the L. hesperus were detailed studies of feeding behavior and damage on primary sympodial and mainstem node leaves have concentrated on nymphs (Tugwell et al. 1976) or on monopodial branches-the mainstem node or on caged nymphs or adults (Gutierrez et al. leaf equals 0,the first sympodial position equals 1; 1977, Jubb and Carruth 1971, Mauney and Hen- structural location-structures are considered as neberry 1979). The validation of results from ex- fruit (square, flower, or boll), stem, branch, or leaf (top or bottom surface). Mainstem nodes are produced successively as ’ Dept. of Entomology, Univ. of California, Davis, CA 95616. the plant grows and a single MSN leaf (position 0 * Div. of Biological Control. Dept. of Entomology, Univ. of for each branch) is produced at each node. Mono- California, Riverside, CA 92521. Previously Univ. of California, Kearney Hortic. Field Stn., podial (or vegetative) branches, which produce Parlier, CA 93648. secondary sympodial (or fruiting) branches, are

1313 1314 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 77, no. 5

ber and December 1981 to obtain data on the feeding behavior of adult L. hesperus and their distribution on the plant. Those bugs were collect- ed from at the University of California, Kearney Agricultural Center. To avoid cage ef- fects on L. hesperus behavior, experiments were conducted to test methods to prevent bugs from flying from plant to plant within the glasshouse. In the first test, a small dot of Testors paint was applied to the cuneus area of the wings (Benedict 1975). The bugs spent most of their time grooming L and ca. 67% had freed their wings within 24 hours. In a second test, the cuneus was clipped from the hemelytra with surgical scissors (Wilson et al. 1983a). This method worked quite satisfactorily. Care had to be taken during removal of the cu- neus, however, to ensure that the posterior end of the abdomen was not inadvertently removed. Two Fig. 1. Branching pattern of cotton. (1) Mainstem; trials were then conducted using the cuneus-re- (2) mainstem node leaf locations (position 0, represented moval technique. In each trial, 25 adult L. hes- by petioles); (3) monopodial or vegetative branch; (4) perus females were placed, one per plant, on plants sympodial branch node leaf (at position 1);and (5) fruit with no fruiting structures older than squares, and structure positions, either retained (-), or abscised on 25 plants having a full complement of fruiting (- - -). structures including full-size green bolls. The cot- ton was grown, one plant per 6-liter pot, in a mix- ture of peat and sandy loam soil. Water was ap- often formed at the lower mainstem nodes, and plied as often as every day and the temperature primary sympodial branches are usually produced in the glasshouse was maintained between 29 and higher up the plant (Fig. 1). Tharp (1960) and 35°C. Hoagland’s solution (Hoagland and Arnon Mauney (1984) present detailed descriptions of the 1950) was added in the water once a week. morphology of the cotton plant. For the first trial, observations were made once Glasshouse experiments were done in Novem- or twice per day for a total of 7 days. For the

Table 1. Mean number of mainstem nodes through the season and the average number of nodes, distant from the terminal, where Lygus hesperus nymphs and adults were distributed

Sampling period 18-28 8-12 26-29 17-20 6-1 1 June July July Aug. Sept. Nymphs Nonvegetal ivd ab 2.92 4.57 4.91 5.60 3.50 SD 3.35 3.14 3.25 3.56 2.97 n 13 138 139 37 12 All data f 4.63 5.36 5.83 7.21 8.24 SD 4.79 3.92 4.58 5.48 8.17 n 19 154 169 614 17 Adults Nonvegetative f - 4.25 5.16 3.60 - SD - 2.30 3.10 2.07 - - n - 12 19 5 All data f - 4.25 6.09 3.60 - SD - 2.30 3.94 2.07 - n - 12 22 5 - Mainstem nodes f 11 14.77 17.84 21.33 22.88

a Primary sympodial branches plus mainstem node leaves. The first node is identified as having any part of its upper surface visible. October 1984 WILSON ET AL.: DISTRIBUTION OF L. hesperus ON COTTON 1315

Table 2. Cumulative proportional distribution of Lygus hesperus from the first partially unfurled leaf in the Cotton terminal; field data summed for all sampling dates

Nymph9 Adultsb Mainstem node NonvegetativeC Vegetative Nonvegetative Vegetative Total branches branches branches branches Total 1 0.16 0.08 0.15 0.12 0.13 0.12 2 0.24 0.11 0.22 0.18 0.13 0.18 3 0.32 0.14 0.30 0.26 0.13 0.24- -. 4 0.43 0.14 0.39 0.38 0.13 0.35 5 0.53 0.15 0.48 0.53 0.13 0.49 6 0.63 0.17 0.57 0.61 0.13 0.55 7 0.72 0.20 0.65 0.74 0.13 0.68 8 0.80 0.24 0.73 0.79 0.13 0.72 9 0.87 0.28 0.79 0.88 0.25 0.81 10 0.92 0.30 0.84 0.91 0.25 0.84 11 0.94 0.34 0.86 0.95 0.25 0.88 12 0.96 0.41 0.89 0.97 0.37 0.91 13 0.97 0.45 0.91 0.98 0.37 0.92 14 0.98 0.50 0.92 0.98 0.50 0.93 15 0.99 0.64 0.94 0.98 0.75 0.96 16 0.99 0.70 0.95 1.0 0.75 0.97 17 -1.0 0.81 0.97 1.0 1.0 18 -1.0 0.85 0.98 19 1.0 0.89 0.98 20 0.92 0.99 21 0.96 0.99 22 0.98 -1.0 23 0.99 -1.0 24 0.99 -1.0 25 1.0 1.0

For nymphs 775; 675 on nonvegetative branches. For adults 74; 68 on nonvegetative branches. C Primary sympodial branches plus mainstem node leaves. second trial, observations were made two to four recorded. A total of 1,165 observations were re- times per day for a total of 8 days. During each corded for living L. hesperus. observation, the location for' each bug for each Means and standard deviation estimates were plant age was recorded as in the field studies. When obtained for the field experiments. Glasshouse data, L. hesperus were found on leaves, we recorded expressed as the proportion of observations in each their location if they were on veins or on inter- cell of the matrix of MSN x BSN x structural veinal tissues, and whether they were feeding. location x activity x plant age, were analyzed by Numbers of dead and missing individuals were a nonreplicated factorial analysis of variance (Steel

Table 3. Structural distribution of Lygus hesperus nymphs and adults on branches of plants in the field, summed for all dates

Nymphsa Adultsb Branch Leaf surface Leaf surface node Squares Bolls Squares Bolls TOD Bottom Top Bottom oc 23 49 2 4 1 13 42 117 80 0 3 7 18 2 5 16 54 32 1 2 4 8 3 4 27 11 1 1 4 1 3 20 4 1 5 1 8 Auxiliary branches 4 9 10 3 1 Monopodial (vegetative) branches 4 27 41 15 1 2 1 3 Total 51 150 277 145 4 12 14 30

a Twenty-four Lygus nymphs were also on the stem tissues. Two Lygus adults were also on the stem tissues. c Position 0 represents the mainstem leaf which subtends the branch 1316 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 77, no. 5

Table 4. The structural distribution of Lygus hesperus nymphs and adults in the field through the season

Nymphs Adults Sampling Leaf surface Fruit Leaf surface Fruit period Stems Stems Top Bottom Squares Bolls Top Bottom Squares Bolls 18-28 June 1 4 8-12 July 2 3 19 65 15 1 4 3 26-29 July 1 2 19 59 28 2 2 10 17-20 Aug 2 2 11 17 11 1 1 3 6-11 Sept. 3 7 5 1 Total 5 11 60 146 55 4 7 16 and Torrie 1960). Only one of the interactions with ing on older fruit structures. Very few were plant stage as a factor was significant; therefore, observed on stem tissue. Table 4 shows the struc- all plant-stage interactions were pooled with the tural distribution of nymphs and adults through residual variation to give a better estimate of the the season at the Shafter site. The relative numbers population random error (Zar 1974). In reporting of both on bolls increased as the season progressed, subsequent results of glasshouse experiments, we reflecting the relatively greater abundance of bolls therefore ignore plant stage as a factor except when later in the season. discussing age of fruit structures fed upon for both plant stages. Glasshouse Lygus Distribution

Results and Discussion Only 48 of the 1,165 observations were recorded as far as or farther than MSN 5 from the terminal, Field Lygus Distribution while no L. hesperus were recorded beyond the first node of a sympodial branch (Table 5).Summed Mainstem Node Distribution. Table 1 shows the number of mainstem nodes through the season and across all other factors, a L. hesperus was observed the average node of distribution from the terminal most commonly on a vein on the top surface of for the Shafter data. As the plants aged and pro- the third mainstem node leaf which subtends that duced more mainstem nodes, the L. hesperus were node, and not feeding. main effects and the interactions were sig- found increasingly farther away from the plant All terminal. On a proportional basis, however, the nificant (P < 0.05). The MSN x BSN interaction was significant 0.05) only because the higher mean node about which they were distributed was (P < mainstem nodes (1 and 2) have only one node on approximately 36% from the terminal for all the each branch, with the branches also considerably data, or 25% from the terminal when only insects reduced in size. Likewise, the significance of the on primary sympodial branches and mainstem node leaves were counted. The preference shown by L. hesperus for the terminal area was apparent, although a large proportion was found several Table 5. Distribution of Lygus hesperus adults by nodes distant from the terminal. mainstem node, branch node, structure, and by activity Table 2 presents the cumulative mainstem node on glasshouse plants distribution for nymphs and adults summed for Mainstem nodea Branch node the season. Both stages were similarly distributed with the means at five to seven nodes from the 12345678 01 terminal when averaged for the whole season. 129 340 496 152 29 13 4 2 819 298 Those L. hesperus on vegetative branches were Structural location consistently on branches originating further from the mainstem terminal than those on primary Leaf ~ ____~ sympodial branches and mainstem node leaves. Top surface Bottom surface Fruit Stem However, L. hesperus on vegetative branches were Vein Between Vein Between usually on the ends of these branches, which bend veins veins upwards; this placed them in a vertical position 304 279 108 57 137 232 similar to that observed for the remaining insects. Structural Distribution. Most L. hesperus were Behavior found located on fruit structures (Table 3). About Feeding Not feeding twice as many nymphs were on squares as on bolls, 284 833 while the reverse was observed for adults. An ap- parent increase in preference shown by adults for Mainstem counts for all data (1,165 points); remaining counts bolls may parallel increased preferences for feed- ignore data from fifth node and lower on the plant. October 1984 WILSON ET AL.: DISTRIBUTION OF L. hesperus ON COTTON 1317

K?

a - I23 45-81 2 3 45-81 2 3 45-8

MAINSTEM NODE FROM TERMINAL Fig. 2 A-C. Structural distribution of adult Lygus on plants in the glasshouse.

2: 302: BSN x L interaction was largely due to the fact 00 5 IO 15 20 25 that fruit is rarely associated with the mainstem MAINSTEM NODE FROM TERMINAL node leaf. Such fruits usually only occur on older Fig. 3. The cumulative distribution of Lygus on plants and develop from normally dormant second plants in the field and glasshouse. axillary buds (see Tharp [1960] and Mauney [ 19841). MSN x L and MSN x A Interactions. Lygus tion feeding was little affected by dorsal or ventral hesperus near the plant terminal were most often location on the leaf. Thirty-three percent of the L. on the top surface of leaves, in particular on vein hesperus were found feeding on vein tissue com- tissues; L. hesperus on the bottom surface of leaves pared with 14% on intervein tissue. This last dif- were most abundant on the third node and least ference reflects repeated field observations that we abundant on the first node and on lower nodes have made of other hemipterans probing leaf veins, (Fig. 2A). Fig. 2B indicates that, for L. hesperus and rarely probing intervein tissue when feeding found on leaves, vein tissues are preferred higher on leaves. Latson (1972) commonly observed Ly- on the plant and nonvein tissues are equally pre- gus lineoluris (Palisot De Beauvois) fe'eding on the ferred lower on the plant. When leaves are small, nectaries arising from these leaf veins. as on the top of the plant, a greater proportion of Feeding on Flower Buds Versus Bolls. Of the the tissue is veins; as leaves grow, the proportion 591 L. hesperus observed on the plants without of intervein tissue increases. bolls, 58 were found feeding on squares and 93 The proportion of L. hesperus on fruit struc- were found feeding on vegetative tissue. Of the tures (squares and bolls) and stems was greatest for 574 L. hesperus observed on plants with bolls, 41 the lower mainstem nodes (Fig. 2C). The signifi- were feeding on squares, and 97 were feeding on cance of the MSN x A interaction seems to result vegetative tissue. Six were observed on bolls, but from the large proportion of feeding on the unex- none were feeding. These results contrast sharply panded terminal vegetative tissues, and the de- with those in the field. creasing tendency to feed on the lower nodes. The marked increase of L. hesperus on fruit structures Field/Glasshouse Comparison lower on the plant and the slight decrease on the lowest nodes may reflect the greater numbers of Glasshouse experiments with a major focus on fruit structures in this region and, possibly, a great- the interaction of L. hesperus feeding behavior er preference for these structures, which would and their distribution in the cotton plant were done mainly be medium-sized and large squares. to provide supportive information for our field re- BSN x A and L x A Interactions. Twenty and sults. In the glasshouse, L. hesperus were much 40% of the L. hesperus were observed feeding more clumped in the top of the plant than were when found on position 0 and on the first branch those in the field (Fig. 3). Similarly, several re- node, respectively, compared to 2% feeding when searchers who have studied the feeding pattern of on stems. In contrast, 71% of the bugs on fruit Lygus spp. on cotton grown under artificial con- structures were observed to be feeding. The inter- ditions have reported that the Lygus spp. prefer actions partially explain themselves because the the top of the plant. This apparent bias is probably large majority of fruit structures are on position 1 due io several factors, the most obvious being the of glasshouse cotton plants and because a greater simpler structure of glasshouse plants. The glass- proportion of these L. hesperus are feeding while house plants appear spindly compared with plants on fruit structures. The resultant effect is a greater in the field. In addition, very few branches pro- amount of feeding on position 1. While most bugs duced more than four nodes; most had two or few- on leaves were on the upper surface, the propor- er. A less complicated branching structure would 1318 JOURNAL OF ECONOMIC ENTOMOLOGY Vol. 77, no. 5 I

~ probably result in fewer Lygus spp. moving down the use of their facilities. We also thank P. Kadan for I to these branches, resulting in a greater concentra- proofreading and typing the manuscript. tion in the upper nodes. Lygus spp. apparently prefer particular areas of the cotton plant, as has References Cited been reported for other cotton (Fye et al. 1969, Fye 1972, Hillhouse and Pitre 1976, Wil- Benedict, J. H. 1975. Ecological and behavioral stud- son and Gutierrez 1980, Wilson et al. 1980, 1982, ies of the damsel bugs, Nabis alternatus Parshley 1983b) and confirmed by our field results. How- and N. amerfcoferus Carayon (Hemiptera: Nabidae) in Northern California hay alfalfa. Ph.D. disserta- ever, the preferred mainstem node distribution is tion, University of California, Davis. probably further from the terminal than previ- Fye, R. E. 1972. Preliminary investigation of vertical ously thought. Since the hemelytra were removed distributions of fruiting forms and insects on cotton from the L. hesperus in the glasshouse, the possi- plants. J. Econ. Entomol. 65: 1410-1414. bility of a bias exists from this factor. Fye, R. E., R. 0. Kuehl, and C. D. Bonham. 1969. The observed bias of L. hesperus towards the Distribution of insect pests in cotton fields. U.S. Dep terminal (96% of 1,165 observations with bugs on Agric., Agric. Res. Serv. Misc. Publ. No. 1140. the top four nodes) in the glasshouse resulted in Cutierrez, A. P., T. F. Leigh, Y. Wang, and R. D. Cave. fewer L. hesperus being associated with bolls than 1977. An analysis of cotton production in Califor- nia: Lygus hesperus (Heteroptera: Miridae) injury- in the field. This difference may be relevant from an evaluation. Can. Entomol. 109: 1375-1386. an economic perspective. Mauney and Henneber- Hillhouse, T. L., and H. N. Pitre. 1976. Oviposition ry (1979) and Pack and Tugwell (1976) are prob- by Heliothfs on soybeans and cotton. J. Econ. Ento- ably correct in reporting that Lygus spp. prefer mol. 69: 144-146. squares; this assumption is shared by most re- Hoagland, D. R., and D. I. Arnon. 1950. The water searchers who have worked with Lygus spp. How- culture method for growing plants without soil. Cal- ever, the results of our field experiments suggest if. Agric. Exp. Stn. Circ. No. 347. that the relative preference for bolls may be great- Jubb, C. L., and L. A. Carruth. 1971. Growth and er than previously thought. In California, Lygus yield of caged cotton plants infested with nymphs and adults of Lygus hesperus. J. Econ. Entomol. 64: spp. are considered to be a potential economic pest 1229-1236. of cotton for a period starting 2 weeks after the Latson, L. N. 1972. Behavior studies of the tarnished appearance of squares greater than 5 mm in di- plant bug, Lygus lineolaris (Palijot De Beauvois), on ameter until the peak of squaring. Possibly, this cotton, hirsutum L., and horseweed, susceptibility period should be slightly extended so Erigeron canadensis L. Ph.D. dissertation, Mississip- that fewer young soft bolls are subject to economic pi State University, Mississippi State. damage. During most of the susceptibility period, Mauney, J. R. 1984. Anatomy and morphology of when squares are more abundant than bolls, most cultivated cotton, pp. 59-80. In R. J. Kohel and C. damage would be to squares unless the preference F. Lewis [eds.], Cotton. American Society of Agron. Madison, Wis. 605 pp. for bolls is considerably higher than presently es- Mauney, J. R., and T. J. Henneberry. 1979. Identi- timated. During routine field monitoring, our fication of damage symptoms and patterns of feeding common observation that L. hesperus adults feed of plant bugs in cotton. J. Econ. Entomol. 72: 496- at nectar glands on both leaves and fruit structures 501. may imply that the preference for bolls by adults Pack, T. M., and P. Tugwell. 1976. Clouded and may not always result in damage to these bolls. tarnished plant bugs on cotton: a comparison of in- Latson (1972) also found nectar glands to be a key jury symptoms and damage on fruit parts. Ark. Agric. feeding site. Exp. Stn. Rep. Ser. 226. Steel, R. C. D., and J. H. Torrie. 1960. Principles The greater proportion of L. hesperus on leaves and procedures of statistics. McGraw-Hill. in the glasshouse could likewise be explained by Tharp, W. H. 1960. The cotton plant, how it grows the relatively greater amount of leaf mass for and why its growth varies. U.S. Dep. Agric. Misc. glasshouse plants. The greater proportion of insects Publ. No. 178. on the bottom of leaves was observed for both the Tugwell, P., S. C. Young, B. A. Dumas, and J. R. Phil- glasshouse and field results and is probably due to lips. 1976. Plant bugs in cotton: importance of the nectar glands on the bottom surface. infestation time, types of cotton injury, and signifi- These results demonstrate the problems inher- cance of wild hosts near cotton. Ark. Agric. Exp. Stn. ent in extrapolating from laboratory or glasshouse Rep. Ser. 227. Wilson, L. T., and A. P. Gutierrez. 1980. Within- results to the field. Our studies do not provide an plant distribution of predators on cotton: comments estimate of L. hesperus feeding preferences for on sampling and predator efficiencies. Hilgardia 48: squares and bolls; however, they do suggest that 3-11. such estimates must be obtained through experi- Wilson, L. T., D. R. Booth, and R. Morton. 1983a. mentation in the field. The behavioural activity and vertical distribution of Harlequin bugs Tectocoris diopthalmus (Thunberg) Acknowledgment (Heteroptera: Pentatomidae) on cotton plants in a glasshouse. J. Aust. Entomol. SOC.22: 311-317. We thank the USDA Shafter Cotton Research Center Wilson, L. T., A. P. Cutierrez, and D. B. Hogg. 1982. and the Univ. of California, Kearney Agric. Center, for Withipplant distribution of cabbage looper, Trtcho- October 1984 WILSON ET AL.: DISTRIBUTION OF L. hesperus ON COTTON 1319

plusia ni (Hiibner) on cotton: development of a sam- distribution of spider mites (Acari: Tetranychidae) pling plan for eggs. Environ. Entomol. 11: 251-254. on cotton: a developing implementable monitoring Wilson, L. T., A. P. Cutierrez, and T. F. Leigh. 1980. program. Environ. Entomol. 12: 128-134. Within-plant distribution of the immatures of He- Zar, J. H. 1974. Biostatistical analysis. Prentice-Hall. hothis zea (Boddie) on cotton. Hilgardia 48: 12-23. Wilson, L. T., D. Gonzalez, T. F. Leigh, V. Ma&, C. Receioed for publication 22 February 1984; accepted Foristiere, and P. Coodell. 1983b. Within-plant 5 June 1984.