Root-boring insects in Gutierrezia sarothrae
Item Type text; Thesis-Reproduction (electronic)
Authors Falkenhagen, Thomas Jay, 1952-
Publisher The University of Arizona.
Rights Copyright © is held by the author. Digital access to this material is made possible by the University Libraries, University of Arizona. Further transmission, reproduction or presentation (such as public display or performance) of protected items is prohibited except with permission of the author.
Download date 06/10/2021 09:28:59
Link to Item http://hdl.handle.net/10150/348286 ROOT-BORING INSECTS IN GUTIERREZIA SAROTHRAE'
by
Thomas Jay Falkenhagen
A Thesis Submitted to the Faculty of the
SCHOOL OF RENEWABLE NATURAL RESOURCES
In Partial Fulfillment of the Requirements For the Degree of
MASTER OF SCIENCE WITH A MAJOR IN WATERSHED MANAGEMENT
In the Graduate College
THE UNIVERSITY OF ARIZONA
1 9 7 8 STATEMENT BY AUTHOR
This thesis has been submitted in partial fulfillment of re quirements for an advanced degree at The University of Arizona and is deposited in the University Library to be made available to borrowers under rules of the Library,
Brief quotations from this thesis are allowable without special permission, provided that accurate acknowledgment of source is made, Requests for permission for extended quotation from or reproduction of this manuscript in whole or in part may be granted by the head of the major department or the Dean of the Graduate College when in his judg ment the proposed use of the material is in the interests of scholar ship. In all other instances, however, permission must be obtained from the author.
SIGNED:
APPROVAL BY THESIS DIRECTOR
This thesis has been approved on the date shown below:
( * ' ■ I i A \ AAX ^ / /\prd IZ , |. 0. KLEMMEDSON Date Professor of Range Management ACKNOWLEDGMENTS
I would like to express my sincere appreciation to Dr. Robert
Eye of the Agricultural Research Service for his support and guidance
during this project, Dr. James Klemmedson for his assistance in orga
nizing and preparing my graduate program and thesis, Dr. Malcolm
Zwolinski and Dr. William Nutting for their review of this manuscript,
and Dr. Floyd Werner for his assistance in identification of the in
sects. I would also like to thank my wife, Kathy Peters, for her
patience and assistance during my graduate work and the preparation
of this thesis.
iii TABLE OF CONTENTS
Page
LIST OF ILLUSTRATIONS ...... v
LIST OF TABLES QoooaOQOo oooooooo«o oo« VXil
ABSTRACT @
1© INTRODUG T10N © © © © © © © © © ©. © © © % © © © © © © © © * 1
2 a LITERATURE REVIEW © © © © © © © © © © © © © © © © © © © © © 3
3 & METHODS © © © © © © © © © © © © © © © © © © © © © © © © © © 10
Field Sampling Area © © © © © © © © © © © © © © © © © © 10 Plot Layout and Sampling Method © © © © © © © © © © © © 10 Inspection of Plants and Larvae © © © © © © © © © © © © 12 Rearing Field Gpllections * © © © © © © © © © © © © © 13 Oviposition by Myrm^x lineolata © © © © © © © © © © © © 15
4© RESULTS AND DISCUSSION © © © © © © © © © © » © © © © © © © 19
Identity and Description of Insects © © © © © © © © © © 19 Borers. © © © © © © © © © © © © © © © © © © © © © © 20 Other Root Inhabitants © © © © © © © © © © © © © © 29 Damage and Route of Travel © © © © © © © © © © © © © © 33 Imeplata © © © © © © © © © © © © © © © © © 33. Hi. Pponielas^ carol mens, is © © © © © © © © © © © © © © 37
° ° ° 0 0 © © © © © © © © © © © 39 Borer Associations and Combined Effects © © © © © © 41 Field Density © © © © © © © © © © © © © © © © © © © © © 42 Borer/PIant Relationships © © © © © © © © © © © © © 42 Borer/Plant Height Relationships ©©©©©©©©© 47 Role of Borers as Weed Control Agents ©©»©,©©»» 52
LIST OF REFERENCES © © © © © © © ©.© © © © © © © © © © © © 58
IV LIST OF ILLUSTRATIONS
Figure Page
1. Sampling area . 11
2„ Insectary rearing boxes 14
3o Oviposition chamber„» @ » op # » « » « » # » » -» op » # 16
4. Cup and stem arrangement (M=medium, S=stems) ...... 16
5. Myrmex lineolata (Pascoe) (== Qtldocephalus ). A, larva; B, pupa; C, adult ...... 21
6. Number of Myrmex lineolata pupating and emerging during 1976 ...... 23
7. Comparison of emergence by M„ lineolata with mean maximum soil temperatures for the sampling area from February, 1976 through January> 1977...... 25
8. Comparison of emergence by M. lineolata with mean moisture percent of G..sarothrae foots from February, 1976 through January, 1977 ...... # . 26
9. Hippomelas carolinensls (Horn). A, larva; B, adult .-. ... O . 9 . O O . 9 9 9 . . . . . # . " 9 28
10. Chrysobothris arizonica (Chamberlain). Adult ...... 30
11. Agrilus glbbicollis (Fall). A, larva; B, adult ..... 30
12. Enoclerus laetus nexus (Barr). A, larva; B, adult . . . 32
13. Mordella species. A, larva; B, adult ...... 34
14. Route of travel and damage by Myrmex lineolata in Gutierrezia roots. A, longitudinal root section; B, cross“Sections ...... 35
v vi
LIST OF ILLUSTRATIONS— Continued
Figure Page
15. Route of travel and damage by Hippomelas carolinensis in Gutierrezia roots. A, longitudinal root section; B, cross-
sections o O » O _ -O O e ff O G O o o o O ' o o o g o o o o 38
16. Route of travel and damage by Agrilus gibbicollis in Gutierrezia roots, A, longitudinal root section; B, cross- sections ...a,..,...,....,.,..,. 40
17. Percentage of plants infested with Myrmex lineolata over the sampling period ...... 43 LIST OF TABLES
Table Page
1. Number and percentage of G. sarothrae plants infested by any borer 45
2. Number and percentage of G„ sarothrae plants infested by each borer species . , ...... 45
3„ Mean number of each borer species per plant 46
4, Number and percentage of G« sarothrae plants in each of seven height classes and the number and percentage of these plants currently or formerly infested by borers ...... 48
5, Number and percentage of G„ sarothrae plants in each size class infested by M„ lineolata or H„ c^ai^^l^Ln(^^g^^^ @ @ @ * * * * * * * * 50
6. Mean and maximum numbers of M„ lineolata larvae and mean numbers of H„ carolinensiS larvae per plant in each size class of G, sarothrae . . . 51
vii ABSTRACT
The purpose of this study was to investigate the biology of root-boring insects in Gutierrezia sarothrae, a range weed. This in cluded identifying major insects infesting the roots, documenting the approximate damage caused by each type of insect, and determining each insect's approximate life history and field density. Results reflect general trends for a specific area for a limited period of time and should hot be used in other areas without more extensive research.
Plants in the field were sampled by a line transect method.
Larvae were removed from the roots and reared to adults on artificial lima bean media in the laboratory. Field and laboratory data reflected emergence patterns.
Four borer species were reared and identified. These included
Myrmex lineolata (Pascoe) (sOtidocephaius) (Curculionidae), Hippomelas carolinensjs (Horn) (Buptestidae), Chrysobothris arizonica (Chamberlain)
(Buprestidae), and Agrilus gibbicollis (Fall) (Buprestidae). Possible predators and other root inhabitants included Enoclerus laetus nexus
(Barr) (Cleridae), Rhadalus testaceus (Le Conte) (Melyridae), Mordella species (Mordellidae), and Certainiops abdominalis (Brown) (Asilidae).
Borers may reduce the ability of G. sarothrae plants to compete with desirable forage plants by damaging supporting tissues in the roots and lowering plant vigor. The potential for using these root-borers for
controlling (3. sarothrae plants could not be assessed by the present
study; however, further research is warranted.
viii CHAPTER 1
INTRODUCTION
The purpose of this study Was to investigate the biology of root-boring insects of Gutierrezia sarothrae«, This included identify ing major insects infesting the roots, documenting the approximate dam age caused by each insect, and determining each insect's approximate life history and field density. Hopefully this information will allow reasonable speculation on the feasibility of using these insects as biological control agents and will suggest further avenues of research to more accurately determine the precise role of foot-borers in damage and mortality of Gutierrezia. Thus, the ultimate objective of this study is to evaluate the capacity of insects to limit populations of
G. sarothrae and to determine how these insects might be used to aug ment and sustain existing control measures.
Preliminary investigations indicated that root-boring insects in Gutierrezia. as with many insect populations, tended to aggregate on certain plants in specific areas. Some plants had high numbers of larvae, while others had few or none. Also, large plants could physically support higher numbers of larvae than could smaller plants.
These attributes of insects, the large variations in plant size, and limits of time and facilities, restricted this study to obtaining primarily qualitative information. Of necessity, sampling procedures
1 2 used here bore closer resemblance to a survey than a statistically
accurate quantitative sampling scheme. Samples reflected general
trends for a specific population, in a small area, for a relatively
short period of time. Thus, care should be exercised in extending re sults to populations of insects in,Gutierrezia sarothrae in other areas.
Such wide application requires extensive research over a wide variety of habitats and is beyond the scope of the present study. CHAPTER 2
LITERATURE REVIEW
Gutierrezia sarothrae is a woody, perennial half-shrub of the family Cbmpositae. The plant is low-growing and compact and usually does hot exceed three feet in height (Campbell"and Bomberger, 1934).
Commonly called broom snakeweed, Gutierrezia sarothrae also is referred to as matchweed, broomweed, resinweed, turpentineweed, sheepweed, or yellowweed (Benson and Darrow, 1952; Vine, 1960; Kearney and Peebles,
1973).
Snakeweed grows in a wide variety of habitats and elevations.
Kearney and Peebles (1973) noted the occurrence of Gutierrezia sarothrae from Saskatchewan to Kansas and southward into northern
Mexico and Baja California at elevations from 3000 to 8000 feet. Vine
(1960) reported the occurrence of snakeweed in West Texas, New Mexico,
Arizona, California, Utah, Montana, Idaho, Nevada, Kansas, and Sas katchewan. Gesink, Alley, and Lee (1973) also have reported
£. sarothrae in Wyoming, although the weed is not particularly abun-> dant in that state.
Gutierrezia grows as well on the plains as on slopes and irreg ular topography and occurs on a wide variety of soil types (Benson and
Darrow, 1952). Parker (1972) stated that Gutierrezia grows well on soils ranging from gravelly, shallow, immature soils to deep, sandy, well-developed loams and clayey soils.
3 4
Because of its wide adaptability to different climatic regions, elevations9 and soil types, Gutierrezia sarothrae is found growing among a variety of vegetation types„ According to Benson and Darrow
(1952) snakeweed grows in the upper creosote-bush desert, desert grass land, and oak woodland, Schmutz and Whitman (1962) noted the abundance of Gutierrezia in the oak-chaparral of Arizona while Jameson (1966) ob served that snakeweed is often found in juniper-pinyon woodland in
Arizona and New Mexico,
Unlike many range weeds, G, sarothrae reproduces by seeds only,
Lillie (1963) pointed out that the density of Gutierrezia may be main tained even when chemical control efforts are imposed because of the widespread establishment of new seedlings. According to Campbell and
Bomberger (1934) viable seeds are produced in sufficient numbers to produce numerous seedlings on rangeland during favorable years.
Snakeweed is regarded as an undesirable plant on rangeland,
A number of investigators have pointed to the aggressive and competi tive nature of the plant as being undesirable (Campbell and Bomberger,
1934; Benson and Darrow, 1952; Jameson, 1966; Jameson, 1970; Schmutz and Whitman, 1962), Snakeweed is generally unpalatable to livestock
(Campbell and Bomberger, 1934); however, Parker (1972) noted that
snakeweed may be eaten when other more desirable forage is in short supply, Schmutz, Freeman, and Reed (1974) also note that consumption of snakeweed by cattle has resulted in major economic losses in the
Southwest by poisoning.
Snakeweed may not be completely undesirable, Stoddart, Smith,
and Box (1975) noted that most plants have some value if they are not too great in number, and that even shrubs of low forage value may by their presence have beneficial effects on other vegetation. Snakeweed might provide shelter for the establishment of grass seedlings, pro vide limited forage for some livestock, and reduce erosion. However, even these attributes are of questionable value (Campbell and Bomberger
1934; Benson and Darrow, 1952; Jameson, 1966; Kearney and Peebles, 1973
Schmutz et ai„, 1974).
Most of the literature seems to reflect that large populations of G, sarothrae are undesirable to a productive rangeland ecosystem.
Thus, efforts to manipulate and control this plant might prove reward ing. Such control measures should be based on a comprehensive knowl edge of the major abiotic and biotic factors affecting growth and reproduction of snakeweed.
Information concerning how specific abiotic factors affect
G. sarothrae is very limited. Jameson (1970) noted that snakeweed populations tend to oscillate somewhat independently of climatic con ditions. Although snakeweed does respond to changes in climate,
Jameson felt that the major factors resulting in changes of snakeweed levels are the decline of older communities and a build-up of younger communities. Campbell and Bomberger (1934) observed that both black grama and snakeweed were found in a depleted condition after contin uous drought but both recovered during years with above average rainfall.
Major biotic factors influencing abundance of snakeweed include plant competition, grazing influences, and attack by insects. G. saro thrae may have a repressing effect on other plants, may grow in a balanced coexistence, or may in turn be repressed by other plants
(Campbell and Bomberger, 1934; Schmutz and Whitman, 1962; Arnold, 1964;
Jameson, 1966; Smith and Schmutz, 1975)« Feeding by livestock on snakeweed does not appear to be a significant factor influencing snake weed population levels„ Feeding by other small mammals is probably also not significant, although Campbell and Bomberger (1934) indicated that rodents cut off many small G,. sarothrae branches „ Major influ ences of grazing seem to result from a shift in competitive advantage because of grazing on other more desirable forage (Smith, 1970;
Jameson and Reid, 1965; Jameson, 1970; Smith and Schmutz, 1975),
The role that insects play in regulating range weed densities is often,ignored by range managers„ However, insects are an important component of the rangeland ecosystem. Recent studies concerned with insects affecting mesquite have shown that native insects may have a significant detrimental effect on production of mesquite pods and via ble seeds (Smith and Ueckert, 1974), Ueckert (1973) also noted several occurrences where native insects have decreased weed densities,
Furniss and Barr (1975) evaluated insects affecting major shrubs in the Northwest and pointed out that insects may significantly influence both desirable and undesirable rangeland vegetation, Rogers and Garri son (1975) noted that insects may play a key role in reversing the competitive advantage that one plant species may have over another.
Information concerned with insects specifically affecting
G. sarothrae is very limited, Campbell and Bomberger (1934) pointed out that insects clearly affected G„ sarothrae plants in 1925 and 1927,
During 1924 and 1925 Gutierrezia plants were attacked by larvae of Crossidius pulchellus (Lee) which established galleries in the woody roots and lower stems of the plants. The investigators felt that this
infestation and extremely low rainfall in 1924 resulted in the death of several Gutierrezia plants. Linsley and Chemsak (1961) reported that Crossidius pulchellus was responsible for killing large numbers of G„ sarothrae in Torrance, Guadalupe, and Lincoln Counties, New
Mexico. Campbell and Bomberger (1934) attributed the death of
Gutierrezia plants in 1927 to attack by leaf rollers. Campbell and
Bomberger found that the large mature plants of Gutierrezia were ap parently quite susceptible to insect injury. Well-established seed lings from two to four years old were more resistant to both insect and drought conditions.
Several control methods are commonly used against range weeds.
These methods include cultural, mechanical, chemical, and biological control and burning. Each of these methods are discussed by Stoddart et al. (1975). Cultural control involves mainly the manipulation of grazing intensity to favor the competitive advantage of preferred grass species. However, even proper grazing may not prevent invasion by range weeds.
Stoddart et al. (1975) emphasized that the use of fire on brushlands is highly complex and variable because of great variations among different species, field techniques, and environments. These authors note that, because of the decreased occurrence of fire on rangeland, mesquite as well as associated species such as oak, juniper burrpweed, cholla, and snakeweed have increased. They also state that mesquite and other shrubs may be decreased by fire with the effects lasting up to 15 years» However, fire may not be applicable in areas where insufficient grass cover is left to carry the flame, on irregular
terrain where containment is difficult, or in areas where social or
legislative restrictions prevent its use, Schmutz and Whitman (1962)
further pointed out that chaparral species are unusually well adapted
to overcome the injurious effects of fire. Thus, burning often may not
be applicable or justified for control of snakeweed.
Mechanical control is usually associated with control of larger weed species such as mesquite, with control of snakeweed being inciden
tal, Schmutz and Whitman (1962) point out that mechanical control in
chaparral in Arizona is often difficult and unsuccessful because of
shallow rocky soils and steep slopes, Mechanical control is also very
costly.
Chemical control has generally been the preferred method for
controlling snakeweed, A number of herbicides including picloram,
2,4-D, 2,4,5-T, and silvex have been found to be effective in con
trolling G, sarothrae by several investigators (Schmutz and Whitman,
1962; Lillie, 1963; Jameson, 1966 and 1970; Schmutz and Little, 1970;
Gesink et al,, 1973), Most of these investigators report substantial
increases in desirable forage production following control. However,
Smith and Ueckert (1974) point out that chemical control methods as well
as mechanical control methods are often unsatisfactory in eliminating
range weeds because of inadequate kill, high cost, and the need for
repetition of treatment, Harris, Peschken, and Milroy (1969) empha
sized that chemical control of the Hypericum perforatum in British
Columbia was costly and ineffective. Cost of control varied widely depending on topography, accessibility, and the chemical used and was always high when compared with the yield from the land. Weeds also reappeared soon after the chemicals dissipated. Although these com ments were not about snakeweed* the same criticisms could justifiably be levied against chemical control of snakeweed, Ueckert, Polk, and
Ward (1971) also note that public concern over environmental pollution has curtailed the use of some herbicides.
Biological control of G. sarothrae has not yet been attempted; however, several researchers have determined that use of native insects to control native weeds holds high potential. Andres and Angalet
(1963), in reference to puncturevine, stated that rangeland weeds that were difficult to suppress, that occur extensively on low-value land, and where the cost of chemical or mechanical control would be uneco nomical, might prove to be good candidates for biological control,
Andres (1971) suggested the conservation of weed-feeding insects as a method of controlling weeds by allowing native insects to increase in numbers and by utilizing these insects in controlling native weeds.
Smith and Ueckert (1974) have pointed out that many advances have been made in managing and controlling insect pests of economic plants.
These principles may in time be adapted to the management of beneficial insects, such as those that attack noxious plants. Even though the potential for biological control of range weeds is high, range managers have as yet given little emphasis to these opportunities» Thus, ex tensive research regarding insects and microorganisms affecting range weeds is needed before biological control becomes a practical range management tool. CHAPTER 3
METHODS
Field Sampling Area
A study area was established approximately 20 miles north of
Sasabe, Arizona, on the Buenos Aires Ranch. This site was chosen be cause Gutierrezia sarothrae plants were fairly dense and uniformly distributed and supported a population of root-boring larvae. In addi tion to G. sarothrae, the site was interspersed with scattered young mesquite trees and a cover of naturally occurring and seeded range grasses. The site was easily accessible, yet isolated enough to be protected from human interference.
Plot Layout and Sampling Method
A line transect method was used to sample G. sarothrae plants for insect larvae. Although not widely used in insect sampling, this method is commonly used for sampling rangeland plants. The method was found to be rapid and simple, yet gave an accurate picture of what was happening in the field.
Sampling involved stretching a 200-foot line perpendicular to a row of stakes from a randomly selected stake and. removing all G, sarothrae plants growing under or within one foot on each Side of the line (Figure 1). Transects could be laid out in a matter of minutes.
Preliminary investigations with shorter transects yielded an
10 11
Line
^ 8 ft". irxVervaU b etw een 5)4-ake*
Figure 1. Sampling area insufficient number of plants to reflect general trends with a single transect, A single 200-foot transect provided between 15 and 30 plants which yielded enough larvae to reflect general trends.
The line of stakes was started approximately 50 yards back from the main highway. One hundred stakes were placed eight feet apart and were arranged roughly parallel to a ranch road coming off the main highway. Enough plants remained between the samples to maintain site integrity and prevent initial samples from affecting subsequent sam ples. Stakes were set back from both the highway and ranch road to eliminate possible edge effects and to permit more uniform sampling.
Samples were taken weekly during initial sampling and during active development and movement of larvae. Frequency of sampling was reduced to once every two weeks during periods of inactivity in the winter. Daily soil temperatures (6 cm depth) and weekly rainfall records were kept on the site.
Inspection of Plants and Larvae
After removal from the sampling area, plants were taken to the laboratory where larvae were exposed by stripping away the bark and splitting the wood. Extremely careful searching was required, espe cially with smaller, relatively obscure larvae (1 mm in length).
Data accumulated from each sample included both plant and in sect information. Plant data included size and condition of foliage and roots, root moisture content (dry weight basis), and whether or not the plant had been previously attacked by insects. Insect data included species, approximate size, stage (larva, pupa, adult), number 13
per plant, and location and orientation in the plant. These observa
tions were used to coordinate plant size with infestation levels as well
as reflect seasonal movements3 size variations9 and field densities of
the insects, Emergence characteristics were coordinated with soil tem
perature, rainfall, and root moisture content. Approximate damage
caused by each insect and their movements through the plant also were
documented.
Rearing Field Collections
All larvae collected from the field were reared on a modified
USDA lima bean diet (Patana, 1969), Free water content of the basic diet was high when compared with the natural situation encountered in the roots and initially resulted in high mortality. The diet proved much more acceptable when some of the free water was removed from the medium. Water was removed by hand squeezing the medium in paper towels,
Plastic creamer cups (9/16 oz) were half-filled with the dried medium,
A larva was placed head first into a hole scraped into the medium about the diameter and length of the larva. Most larvae immediately began to feed and usually burrowed into the medium by the next day. Cups were covered with plastic lids to slow drying of the medium, but it had to be changed at approximately one month intervals when obvious drying occurred.
Larvae were reared in an environment that hopefully approxi mated the situation in the field. Cups were placed in plastic shoe boxes half-filled with sand and covered with a ventilated lid (Figure 2),
These boxes were placed in a screened open-air insectary. Boxes were 14
Screen
Sand
Figure 2. Insectary rearing boxes 15
shaded at all times and sand temperatures were recorded. Sand tem
peratures in the insectary were approximately 3° C cooler than field
temperatures <, Larvae were inspected almost daily during periods of high activity (summer) and only infrequently during periods of inactiv
ity (winter). Changes in size, stages, and orientation were observed, recorded, and compared with field activity.
Larvae reared to adulthood were retained for identification.
Adults of one species (family Curculionidae) were obtained in sufficient numbers from field collections and rearing for further study of be havior and life habits.
Oviposition by Myrmex lineolata
Myrmex lineolata adults were the only borers reared in suffi cient quantities for further studies of mating, oviposition habits, and development of eggs. This phase of study required development of a microcosm that paralleled the natural habitat closely enough to allow weevils to freely mate and oviposit (Figure 3). As in insectary rear ing of the larvae, plastic shoe boxes with ventilated lids were employed.
Two galvanized standards were placed in the bottom of the box and about
3 cm of sand were poured around them in order to make a level surface.
The sand allowed free movement of adults over the entire area. The
standards were approximately 15 by 15 cm and had five holes of suffi cient size to hold 9/16-oz plastic creamer cups.
Cups were placed in each hole and contained either Gutierrezia
stems or lima bean media. Three cups in each standard contained moistened pieces of sponge the same size as the cups. Sections of 16
4— ---k 4— — —k
TT^Dnxzr-T?
Figure 3. Oviposit ion chamber
Figure 4. Cup and stem arrangement (M=medium, S=sterns) 17
Gutierrezia stems about 5 cm in length and 0*5 cm in diameter were in serted radially around the perimeter of the sponges (Figure 4), Stem sections were cut from terminal portions of Gutierrezia plants and small branches and foliage were left attached. Early attempts indi cated weevils preferred stems with foliage rather than stripped stems.
These also were preferred to pine dowels used in early attempts. The other two cups in each standard were filled with unsqueezed lima medium to provide a moisture source for adult weevils as well as another pos sible oviposition site. Cups containing medium were changed when it dried, while cups containing stems were inspected and replaced every other day.
Mold was a slight problem on both the sponge surface and the branches themselves. However, this did not seem to adversely affect the weevils, especially since the stems were changed every other day.
Fifteen adult weevils were initially placed into the chamber on June 7, 1976, The chamber was kept at a constant temperature of
27° C with a 12 hour alternating light/dark regime. Adults were added to the chamber as they were reared out or collected from the field and reached a maximum density of 47 on July 12, Because of mortality and decreasing replacement of weevils, this phase was terminated on
August 99 1976, when the level of adult weevils declined to three.
Adults added to the chamber preferred aerial parts of the stems.
Copulation was almost immediate and was a frequent occurrence. Even tually 25 eggs were recovered during a period from July 1, 1976, to
August 9,' 1976, 18
Eggs were placed on the lima bean medium described in the rear ing section. Eggs and hatched larvae were reared at 27° C with an alternating 12 hour light/dark regime. Although no eggs were reared to adulthood, six reached sizes ranging from 9 to 11 mm in length when this phase was terminated on December 30, 1976„ CHAPTER 4
RESULTS AND DISCUSSION
Identity and Description of Insects
Larvae found in Gutierrezia roots and reared to adulthood re vealed that several species of boring beetles as well as some possible
predators inhabit the plant. Three wood-bo’rers and one bark-borer were
identified. Myrmex lineolata (Pascoe) (gQtidocephalus), a wood-borer
of the family Cur.culionidae, was the most prevalent borer. Other wood-
borers, Hippomelas carolinensis (Horn) and Chrys obothr is arizonica
(Chamberlain), belonged to the family Buprestidae. The bark-borer was
Agrilus glbbicollis (Fall), also of the: family Buprestidae.
Other reared larvae included several possible predators and un
classified insects. Possible predatory Coleoptera included Enoclerus
laetus nexus (Barr) (family Cleridae), Rhadalus testaceus (Le Conte)
(family Melyridae), and an unidentified member of the genus Mordella
(family Mordellidae). One dipterous predator, Certainiops abdominalis
(Brown) (family Asilidae) was also reared, as well as an unidentified
parasitic hymenopteran. One lepidopteran inhabiting the root area also
was reared. However, insufficient specimens were available to assess
the food source of the larvae or to identify the adults.
19 20
Borers
Myrmex lineolata0 Since Myrmex lineolata (Pascoe) was the most numerous species sampled, information concerning it was most complete»
Eggs were not obtained from the field; however, some adult specimens did oviposit in the laboratory. Both field and laboratory reared in sects provided the basis of descriptions of larvae, pupae, and adults
(Figure 5),
Eggs were laid in the laboratory from July 1, 1976, through
August 9, 1976» Eggs were cream-colored and oblong-shaped with diam eters approximately 0,5 by 1 mm, M, lineolata adults laid their eggs . singly on or within the lower G, sarothrae stems. For the 18 eggs which hatched, mean hatching time was 12 days. Because eggs were in cubated under a constant temperature df 27° C in the laboratory, hatching time may have differed substantially from hatching time in the field.
Larvae were white, cylindrical, legless, and crescent-shaped ‘ ; ■ • (Figure 5A), Sizes of larvae varied widely regardless of the time when they were sampled. Length ranged from 1 to 14 mm in both laboratory reared and field collected larvae. Diameter of the large anterior end varied from 0,5 to 4 mm. Gallery diameter corresponded closely with the diameter of the anterior portion of the larvae; this will be dis-? cussed in the Damage and Route of Travel section.
Pupae (Figure 5B) were distinctly weevil-like as evidenced by prominent snout development. Specimens reared from the laboratory and those collected in the field did not differ markedly in appearance.
Size was fairly constant and varied from about 7 to 10 mm in length. /V B C
Figure 5 Myrmex lineolata (Pascoe) (=Otidocephalus), A, larva; B, pupa; C, adult 22
Mean pupation time of insectary reared larvae during early summer was
16 days and varied from 6 to 34 days. The mean pupation time of three larvae that pupated in early winter averaged 44 days and ranged from 28 to 51 days. Because soil temperatures were lower in the in sectary than in the field, these values may differ from actual field pupation time.
The adult stage did not vary markedly between laboratory reared and field collected specimens (Figure 5C)» Colors of the elytra were alternating black and white longitudinal stripes; the pro- notum and head regions were black. These regions were covered with a pubescence, noticeable by careful scrutiny with the unaided eye.
Length varied from 6 to 19 mm.
Adults tended to climb toward aerial portions of their en vironment. This was particularly noticeable in the oviposition chamber where the majority of adults crowded onto the terminal portions of stem sections as well as to the inside of the oviposit ion chamber lid. When disturbed, weevils would frequently fall to the ground and remain mo tionless, apparently feigning death.
A total of 32 Mo lineolata larvae pupated and emerged during the 1976 season. The remainder of collected larvae died or did not emerge until the following.year. Most larvae reared in the insectary pupated in May and June and emerged in June and July (Figure 6).
Sporadic emergence also occurred from October through December. Most of the larvae collected from February 8 to June 4, 1976, emerged during the same year. However, all but two of the larvae collected after
June 4, 1976, did not emerge until 1977. Because these larvae did not 23
lOi
15-
\0-
5 -
merqence
S
Mon4k s
Figure 6. Number of Myrmex lineolata pupating and emerging during 1976 24 emerge during 1976, most of the 1976 emergence was probably completed prior to June, Emergence of larvae reared in the insectary was de layed approximately one month* Since rearing temperatures averaged
3° C lower than field temperatures, emergence might be expected to be delayed. Rearing on the artificial diet also might have delayed emergence.
An evaluation of climatic data provided further information regarding time of emergence of adult weevils* Average maximum monthly soil temperatures (6 cm depth) in the field were compared with the emergence pattern (Figure 7), The emergence pattern of insectary reared larvae was shifted one month to the left in Figures 7 and 8 to approximate the earlier emergence of field populations* May emer^ gence corresponded to rising average maximum temperatures beginning in April and leveling in June /through August*
Moisture content of G* sarothrae roots might provide a clue to emergence patterns * Percent moisture content of Gutierrezia roots
(dry weight basis) varied inversely with soil temperatures because roots were more subject to drying with rising temperatures in the ab sence of rainfall* May emergence appeared to be associated with root moisture content that began declining in April and did not rise again until July when precipitation resumed (Figure 8), If low moisture is a requirement for emergence of larvae, then delayed emergence in in- sectary reared larvae might be partially explained by the high moisture content of the lima bean medium*
Soil temperature and root moisture content seemed to correlate well with emergence patterns of Myrmex lineolata. However, these 25
2 5 i
c n
20 ELme r^cnce Soil 'Vemperasures•35
-20
I
F M A M T J A SO NDJ M om4ti s
Figure 7. Comparison of emergence by M. lineolata with mean maximum soil temperatures for the sampling area from February, 1976 through January, 1977 26
G ~ > CA(H cy -to u Emergence V £ Roo4 Moi&4ure “50 L d 15-
■MO u V *> io- -30
-10 i- v \ -to -a £ RocV Moi &4ure Percer»4 5 — i— — i— -V "r- — r —I- M T J Av 0 KI J M on + Ks
Figure 8. Comparison of emergence by M. lineolata with mean moisture percent of G. sarothrae roots from February, 1976 through January, 1977 27
studies were far from being statistically conclusive. These associa
tions would have to be validated by several seasons of documentation
of emergence trends in the field coupled with laboratory rearing
trials where moisture and temperature were controlled and monitored,
Hippomelas carolinensis, Hippomelas carolinensis larvae were
found in considerably lower density than M» lineolata. As a result,
information concerning this borer is not as complete. No information
regarding egg. form or size is included because no eggs were found in
the field and not enough reared adults were obtained for oviposition
studies. Pupal descriptions also are omitted for this species. Adult
stages were not found in the field, thus descriptions refer only to
laboratory reared beetles which may or may not resemble field specimens,
Hippomelas larvae (Figure 9A) were white, long, and slender with the anterior and characteristically expanded and flattened.
Larvae were substantially larger than the other borers. Length of
collected and reared larvae was as long as 40 mm. Head width reached
6 mm with thickness as much as 4 mm. Gallery diameter was often
larger than the head size and will be considered further in the Damage
and Route of Travel section.
The adult stage (Figure 9B) was metallic black with no apparent
pubescence. Size varied from 10 to 17 mm in length,
Chrysobothris arizonica. Information on Chrysobothris .
arizonica is incomplete because only two adults were reared from col
lected larvae. Larvae were similar to H, carolinensis larvae; however,
£» arizonica larvae were more tapered and slightly darker. A B
Figure 9 Hippomelas carolinensis (Horn), A, larva; B, adult 29
£<, arizoriica adults (Figure 10) were also similar to H, carolinensis in color and general appearance9 but C c arizonica adults were smaller.
Both adult specimens of C 0 arizonica were about 8 mm long,
Agrilus eibbicollis. Information on Agrilus gibbicollis also is incomplete because of low field density. Eggs were not found in the field and not enough adults were available for oviposit ion studies.
Adult stages obtained from both laboratory reared larvae and field col lected adults did not appear different.
Larvae (Figure 11A) were white and more slender than previously described forms. Length reached 25 mm. Gallery diameter was closely aligned with head contours and size. The head region was typically flattened and enlarged.
Adults (Figure 11B) were metallic brown with a single grey stripe extending longitudinally about half-way down each side of the pronotum to the posterior end of each elytron. Body form was much narrower relative to the other Buprestidae described. A. gibbicollis adults were the smallest of the borers and averaged ;between 5 to
6 mm in length.
Other Root Inhabitants
Enoclerus laetus nexus. The only probably predator found in habiting G. sarothrae roots in any quantity was Enoclerus laetus. Ac cording to Borror, DeLong, and Triplehorn (1976), the majority of
Gleridae are predaceous. Both larvae and adults of many types prey on wood-boring larvae. Figure 10. Chrysobothris arizonica (Chamberlain). Adult
Figure 11. Asrilus &ibbicollis (Fall). A, larva; B, adult 31
Larvae of Enoclerus laetus (Figure 12A) were pink with a very
small head on a body which tapered anteriorly. This larval form pos
sessed legs and was highly motile and quite fast when disturbed. Max
imum length of larvae was 8 mm,
Enoclerus larvae always were found within borer galleries. In
14 of 16 instances where these larvae were found in Gutierrezia roots,
M, lineolata larvae also were found. This was the only borer found
associated with Enoclerus larvae. Although actual feeding on M,
lineolata by Enoclerus larvae was not witnessed, the association
strongly suggested that Enoclerus larvae may have been predators of
M» lineolata larvae.
Pupation by jE, laetus also occurred within the borer galleries,
Pupae of this species could readily be distinguished from other species
by their pinkish color which was slightly paler than larval color, but
still apparent,.
Combined field and laboratory data indicated emergence pri
marily occurred from June through August, This corresponded closely
with emergence patterns of M, lineolata. Such synchrony may further
suggest predation by E, laetus or at least simplify predation if it
occurs,
Adults (Figure 12B) were characteristic of the "checkered
beetles," Head color varied from red to black. Elytra were black with
two white transverse bands across each elytron and a white spot at each
anterior junction. Heavy pubescence was readily noticeable over the
entire body. Length varied from 6 to 8 mm. 32
A B
Figure 12. Enoclerus laetus nexus (Barr). A, larva; B, adult 33
Horde11a species. The Horde!la species that were reared from
field collected larvae were not believed to be predators in the larval
stage, even though some members of this genus are known to be preda
ceous, Larvae always were found inhabiting roots in an advanced state < of decay. Although Hyrmex lineolata was found in 11 of 16 instances where Horde11a species were found, H, lineolata larvae were found sole
ly in the undecayed portions of the roots. This may indicate that
secondary infestation by Hordella species follows infestation of
G„ sarothrae roots by borers such as H, lineolata.
Hordella larvae (Figure 13A) were white and similar in size to
H, lineolata larvae. However, Hordella larvae were more flattened on
the ventral side than H. lineolata larvae and tapered to a point on
the posterior end. Length ranged from 1 to 7 mm.
Adults (Figure 13B) were satiny black and were distinctly
humpbacked, The abdomen extended past the elytra to. a tapered point.
Length ranged from 4 to 6 mm.
Other possible predators were not obtained in sufficient quan
tities to speculate upon their role in affecting the borers. However,
these insects warrant further research concerning their role in limit
ing borer populations.
Damage and Route of Travel .
Hyrmex lineolata
Determinations of the route of travel of H» lineolata and the
tissues they ingest were made by dissecting the roots and tracing the
galleries excavated by the larvae (Figure 14). Galleries were packed A B
Figure 13. Mordella species. A, larva; B, adult 35
Rou4-e df Travel Routeo i Travel i Gallery Diamc'fer Cl Entrance Ho!e( 0-1 wvn)
4" \. 0 mm G r o u n d Level t 5. Omm l | Z > ^ Crown H 2.0 mm
4 1.5 mm 4 f 3.0mm
l6» mm
«jr 1.0 mm f 1.0 mm
\\ mm Heertwoo^
(Xnactwe Xylem)
Sapwooj (Active Xylem )
4 mm A B
Figure 14. Route of travel and damage by Myrmex lineolata in Gutierrezia roots. A, longitudinal root section: B, cross-sections 36 tightly with sawdust-like frass arranged in arclike layers. Trails were approximately circular (Figure 14B); however, this simple pattern was distorted when upward and downward paths merged. Because of merg ing of trails some galleries appeared to be similar to galleries of the Buprestidae. Even when galleries were distorted, the identity of larvae making the galleries could be determined by careful examination of the frass.
Site of oviposition was usually about 3 cm above the root crown
(Figure 14A), but field and laboratory data indicated that oviposition may occur anywhere in the lower region of the stems. Oviposition need not occur in branch axes. Eggs may be inserted in roughened bark or attached to the outside of the stem. Eggs also may be placed inside the stems. This seems to indicate that M. lineolata has some appara- tus for boring oviposition holes.
Route of travel began with entrance by the larva into the plant at the site of oviposition. Entrance holes were less than 1 mm in diameter and were generally between 0.5 and 0.7 mm. The developing larva bored down through the stem, past the root crown, and into the roots. Downward travel continued 3 to 8 cm below the root crown, but did not continue to the root tip. Most boring activity was restricted to the upper third of the roots.
When the terminal point was reached by the larva it turned and began boring upward by a route which was distinct from the initial route or closely parallel to the initial boring. These routes of travel were direct rather than winding and gradually increased in 37 diameter. Diameter of boring was as large as 4 mm when the larva ascended to the root crown.
Pupation and development into an adult occurred within the roots at or slightly below the root crown. This region was below the soil surface or in the litter surrounding the plant base. Here the adult excavated an oval emergence hole through the root approximately
1 mm in length.
Ingestion of plant tissues during vertical travel was restrict ed to the Xylem or wood, except for entrance and exit holes where move ment was primarily horizontal. Damage was specifically restricted to the heartwood . of the roots. In some samples the entire area of the heartwood of smaller roots was eaten away at some point along the route by a single larva. However, in larger roots the combined action of several larvae was necessary to completely destroy a portion of the heartwood area. Since heartwood is a supporting rather than a con ducting tissue, boring by larvae probably did not influence root trans port mechanisms. However, boring activity may have structurally weakened the plant and made It more vulnerable to attack by other organisms.
Hippomeias carolinensis
Damage and route of travel by H. carolinensis was similar to
M. lineolata. but galleries were substantially larger and sausage- shaped (Figure 15B) in cross-section. After hatching from the eggs, larvae bored down through the lower stems, past the root crown, and into the foots (Figure ISA), Boring by H. carolinensis larvae was much more extensive than boring by M, lineolata and continued almost 38
En4-r»nce
Gallery Dimensions Rou+e 4 I.S mm X 5mm T 3 tnm X 6mm i S. mm X 4 mm t J. hnm X 4 mm A B Figure 15. Route of travel and damage by Hippomelas carolinensis in Gutierrezia roots. A, longitudinal root section; B, cross-sections 39 to the root tip before the return boring began. Depth of boring often exceeded 20 cm which was in marked contrast to the relatively shallow borings of M. lineolata. Upon reaching the terminal boring point, the larvae turned and began moving upward. Because of the larger size of these larvae, the return trail often intersected the initial boring, resulting in one large zone of destruction. Initial borings were only about 1 by 2 mm, while terminal borings often were as large as 4 by 25 mm in diameter. H. carolinensis larvae ingested the same tissue areas as M„ lineolata (Figure 15B). . Borings by a single H. carolinensis larva often included the entire heartwood area of some roots. Agrilus gibbicollis Boring by A. gibbicollis larvae began up to 8 cm above the root crown after hatching of eggs deposited in the branch axes. Frass trails were sausage-shaped (Figure 16B) as with H, carolinensis, but were con siderably smaller. Route of travel extended almost to the root tip be fore return boring began (Figure 16A). Initial boring and the return route were parallel yet distinct. Both routes were direct and did not wind or spiral. Adults exited in the area of the root crown. Initial borings were only about 0.5 by 1 mm, while terminal borings were as large as 2 by 3 mm. A. gibbicollis larvae attacked different tissues than M, lineo lata, or the other two buprestids. Agrilus larvae restricted their boring to the bark tissue (Figure 16B) and more specifically to phloem tissue. Destruction of this tissue sometimes interfered with movement of metabolic products, but since a single larva could not girdle a root. 40 er RooV Crown m m m A B Figure 16. Route of travel and damage by Agrilus gibbicollis in Gutierrezia roots. A, longitudinal root section; B, cross-sections 41 several larvae would probably need to inhabit a single root to cause significant damage. Borer Associations and Combined Effects M. lineolata, H, carolinensis, and A. gibbicollis larvae were found boring singly within a specific plant or in conjunction with each other, M, lineolata larvae were frequently observed inhabiting the same plant or root with H„ carolinensis or A, gibbicollis. H. carolin- ensis and A. gibbicollis larvae were not observed in the same plant, but since densities of both larvae were low these observations were not regarded as an indication of competitive exclusion. Since they fed on different root parts, they would not be expected to interfere with each other. These larvae sometimes inhabited the same plants in areas where larvae were more numerous. Because H, carolinensis and M. lineolata both exploited the same food supply, they are potential competitors. However, competition did not appear to be important in this study. Per plant densities of larvae were not high enough to exploit the entire food source. In the root crown and slightly below. Where consumption of plant tissue was maximal, root diameter was also maximal. By restricting their boring to the upper third of the roots M, lineolata larvae did not experience the competition with H. carolinensis which might be expected if their galleries extended to the lower narrow portions of the roots. Destruction of the roots eventually resulted in the collapse of aerial portions of G, sarothrae plants. This external manifestation of internal borer activity was not apparent until the action of bacteria. fungi, and secondary insects resulted in advanced deterioration of the root system. At this stage, roots and borer galleries were invaded by ants and termites and borers were no longer a part of the insect spectrum. Results of this study suggest that the effects these borers have on plant health and mortality may be dependent on environmental factors. In years with optimum rainfall and temperature conditions even substantial tissue damage may be minimized by the ability of the plant to send out new roots. However, damage by borers may be sig- nificant only when adverse climatic conditions such as drought or excessive soil moisture lower the resistance of G. sarothrae. Damage may range in severity from limiting plant growth to actual plant death. Damage may result directly from borer action or by a sequence of attack by insects and microorganisms which eventually result in plant mortality. Field Densit Abundance of root-boring larvae remained fairly constant throughout the year. Only M. lineolata larvae were collected in suf ficient quantities to detect periodic variation in density (Figure 17). Slightly lower densities were apparent for this species prior to emer gence in April and May, while slightly higher densities were evident from August through January. Borer/Plant Relationships G. sarothrae plants were classified into three categories: Those infested with root-boring larvae, previously infested as cen 70- 20 10 30- 50 - - - iue 7 Pretg o pat ifse ihMre lnoaa vr h smln period sampling the over lineolata Myrmex with infested plants of Percentage 17. Figure —i —i —i —i i— i— i— i— i— i— i— — n -3 W -W N S-13 -31 t - s 5-11 f l V fc l . y - S t l - s -16 ---- t c3 T1 > T-1 fc-30 » ' . ( lt r « —i —i —i —i —i —< —i —i « i « i i— i— i— i— <— i— i— i— i— i— i— i— i— i— i— 1— 1 $ >11 1 - 1 % t-M VH % " M H 1-1 1-1 H M " % VH t-M % s k e e W 1 9 S -3 io-K » - 1 1-23 'S 9 * ----- 1 « "-' *l M ll-l Iflt t U U»l-4 - l » U U t t l f I l - l l 'M l -* H '5 - " '« » • —i —i —i —i —i— i— i— i— i— i— i— i— i— 1— 4> LO 44 evidenced by abandoned galleries, or never infested by root-borers (Table 1). These data indicate not only a high resident borer popu lation; they also indicate that another 20% of the plants had been previously infested. Less than one-third of the plants were unaffected by root-borers. Percentage of plants infested by borers becomes more meaningful when broken down by species (Table 2). M, lineolata accounted for most of the present infestation. H. carolinensjs accounted for only 17% of the present infestation; however, a single H» carolinensjs larva con sumed as much plant tissue as several M. lineolata larvae. Thus, lower densities of H, carolinensjs larvae may inflict as much damage on JS. sarothrae plants as higher densities of M„ lineolata. A. gibbicollis and C. arizonica were found in such low abundance that they were considered relatively unimportant components of the Gutierrezia insect population. However, this, may not be the case in "... -N ' " other areas. Samples taken from another site about 30 miles north ap peared to sustain a substantially higher density of A. gibbicollis. Moreover, these same plants did not appear to be infested with M. lineolata. Thus, results of the present, study are valid only for this particular area at this particular time. M„ lineolata also was the species with the highest mean number of larvae per plant (Table 3). However, these figures did not give a very meaningful indication of how the larvae were distributed in the field. Although many plants had only one larva of M. lineolata, num bers per plant often reached levels as high as 14. Thus, the average of 2,8 for this species was not particularly meaningful. On the other 45 Table 1, Number and percentage of G, sarothrae plants infested by any borer Type.of Infestation Number of Plants t Currently Infested 298 49.9 Current or Past Infested 425 71.3 Not Infested 171 28.7 Table 2„ Number and percentage of G„ sarothrae plants infested by each borer species* Type of Larva Number of % of % of Currently Plants All Plants Infested Plants M. lineolata 280 47.0 94.0 H. carolinensis 51 8.6 17.1 A. gibbicollis 13 2.2 4.4 £. arizonica 2 0.3 0.1 * Some plants were infested with more than one species of insect larvae 46 Table 3„ Mean number of each borer species per plant Type of Larva Mean Standard Deviation M, lineolata 2,84 2.42 H„ carolinensis 1,16 0.37 A. gibbicollis 1.00 0.00 C. arizonica 1.00 0.00 47 hand, the value of 1,2 for H, carolinensis gave an excellent approxi mation of field conditions. Most plants contained only one larva, per plant, while the maximum number found in any plant was only two. Since densities of A. gibbicollis and £, arizonica were so low, mean numbers of each species per plant were not considered reliable. Particularly unreliable was the estimate of A. gibbicollis density since preliminary samples in other areas showed levels as high as seven with levels of four to five more common than single infestations. Because of its similarity to H. carolinensis. densities of £. arizonica (one per plant) may be valid but this could not be confirmed with only two infested plants, Borer/Plant Height. Relationships The type of G. sarothrae plants preferred by borers might be reflected by relating borer infestation levels to.plant size. Plant height gave a better indication of plant size than root length. Gutierrezia plants were divided into seven height classes (Table 4). The greatest percentage of plants fell within the 40-50 cm class (30,9%), Approximately 56% of all plants fell into the 40-60 cm range, while 93% fell within the 30-70 cm range. By analyzing the percentage of plants in each height class that were attacked by any borer, the varying susceptibility of the plants to infestation became apparent (Table 4). As plants grew taller, suscep tibility to borer infestation increased dramatically. In plants less than 40 cm high, the average percentage infestation was less than half the average infestation of 71% for all plants. In plants 40-50 cm high the percentage infestation of 68% approached average. Plants over 48 Table 4, Number and percentage of G. sarothrae plants in each of seven height classes and the number and percentage of these plants currently or formerly infested by borers Plant Height Plants Plants Currently or Formerly invested . Number % Number % cm 4 3 0 24 4.0 8 33.3 30-40 114 19.1 38 33.3 40-50 184 30.9 126 68.5 50-60 148 24.8 137 92.6 60-70 88 14.8 80 90.9 70-80 29 4.9 27 93.1 > 8 0 9 1.5 9 100.0 49 50 cm high showed a marked increase in infestation ranging from 90-100%, This trend might be explained in several ways. Higher infestation levels are possible in the larger plants since the larger plants have corre spondingly larger root diameters which would encourage survival and support higher levels of infestation. Also, taller plants are exposed to possible infestation for longer periods of time than smaller plants. In addition, taller plants occupy a larger surface, put out more flowers and foliage, and may provide a more attractive target to ovi positing females. An evaluation of the number and percentage of G. sarothrae plants in each size class currently infested by M. lineolata and H, carolinensis larvae reveals a wide range of attack by both species (Table 5), Because of their larger size H, carolinensis larvae were originally thought to be restricted to the larger plants; however,. this was not confirmed by the data. Although M. lineolata larvae in habited all size plants, occurrence of larvae increased dramatically in plants over 40 cm tall. The occurrence of H, carolinensis larvae also increased in plants taller than 40 cm but declined again in plants taller than 70 cm, H. carolinensis larvae were not found in plants over 80 cm tall. The mean number of each borer species per plant was presented earlier. By relating mean numbers of M, lineolata and H, carolinensis larvae per plant to each plant size class, these data might be more meaningful (Table 6), Data for both species did not vary substantially over the range of size classes although averages for M. lineolata did increase slightly with larger plants. 50 Table 5» Number and percentage of G» sarothrae plants in each size class infested by M* lineolata or H„ carolinensis M, lineolata H. carolinensis Plant Height (cm) No. of Plants % No. of Plants % <■30 2 8.3 1 4.1 30-40 14 12.2 4 3.5 40-50 79 42.9 23 12.5 50-60 93 62.8 16 10.8 60—70 62 70.4 6 6.8 70-80 24 82.8 1 3.4 > 8 0 6 66.7 0 0.0 51 f Table 6„ Mean and maximum numbers of M„ lineolata larvae and mean numbers of H„ carolinensis larvae per plant in each size class of G. sarothrae M„ lineolata •H. carolinensis Plant Height Mean S.D. Maximum Mean S.D. < 3 0 1.5 0.7 2 1.0 0.0 30-40 1.9 1.1 5 1.0 0.0 40-50 2.7 4.3 7 1.2 0.4 50-60 3.1 2.5 13 1.2 0.4 60-70 3.7 6.2 12 1.0 0.0 70-80 3.9 3.3 14 1.0 0.0 > 8 0 3.2 1.5 5 0.0 0.0 52 When maximum numbers of M 0 lineolata larvae were noted in each plant size class, results were more significant. These data (Table 6 ) indicate a higher infestation potential by larvae with increasing sizes of plants. Multiple oviposition in the larger plants as well as in creased survival in these plants may account for this increase in larval numbers. Role of Borers as Weed Control Agents Several investigators have cited the potential of using native. insects to control populations of native weeds (Andres, 1971; Ueckert et al., 1971; Ueckert, 1973). The present study did not confirm that native borers could be used to reduce and limit populations of G 0 sarothrae. However, the findings do permit reasonable speculation on the potential of using root-boring beetle larvae for biological con trol of G. sarothrae populations. M» lineolata and H. carolinensis inflicted the most damage to Gutierrezia roots. Both insects caused considerable destruction of the heartwood of the roots, Stoddart, Smith, and Box (1975) pointed out that competition between plants is largely dependent upon rooting habits, Thus, by weakening the structure of the roots, the ability of G, saro thrae plants to compete with other plants may have been reduced, Al^ though G. arizonica and A. gibbicollis did not appear to be important in the present study, they may have a more significant effect in other areas. Stoddart et al. (1975) pointed out that the action of a bio logical agent may destroy the plant host directly or weaken the plant. 53 making it npn^competitive with other plants or subject to attack by other insects and pathogens» Although the activity of borers studied here could not be linked to decreased competitiveness or mortality of G„ sarothrae^ other root-boring larvae have been linked to mortality of Go sarothrae and other weeds„ Campbell and Bomberger (1934) noted that the action of root- boring cerambycid larvae resulted in the death of several Gutierrezia plants* Linsley and Ghemsak (1961) also attributed mortality of large populations of G, sarothrae to these same root-boring larvae. Root- boring Curculionidae have been shown to adversely affect several weed and non-weed plant species* Andres and Angalet (1963) noted that tun neling by Weevil larvae lessened overall vigor of puncture vine. Lar val feeding by weevils in alfalfa tap roots has caused significant damage to these plants and resulted in increased rates of decline of the stand (Pesho, 1975)* Bertwell and Blocker (1975) noted that wee vils may be a limiting factor in the distribution of forbs and forb species in the tallgrass prairie* Buprestid larvae have also been shown to cause damage to weed species* Root-boring Agrilus larvae have been found to be effective in destroying or weakening Klamath weed (Harris et al*, 1969)* Other insects and pathogenic microorganisms may contribute to the impact that root-borers have on plants * Insects that feed on other plant parts would probably augment the impact of root-borers by lessen ing overall plant vigor* Although this study did not address these other insects, Campbell and Bomberger (1934) noted that attack by leaf- rollers resulted in the death of G* sarothrae plants* Microorganisms 54 also may augment the decline of G. sarothrae plants. The present study indicated that roots of plants in an advanced state of decline were decayed by root-rot fungi. Although a direct link between root-borers and fungi was not established, decayed roots invariably showed evi dence of extensive borer activity. Infestations of root-boring wee vils have been linked to the increased Incidence of several bacterial and fungal plant diseases in alfalfa (Pesho, 1975). Although pathogens may not have actually been transmitted by weevil larvae, Peshb pointed out that feeding by larvae provided an entry pathway for plant diseases. Insects and microorganisms may also lessen the impact that root- boring larvae have on G. sarothrae stands. Several possible predators and one parasite were recovered in the galleries of root-boring larvae. How these insects limit populations of root-boring larvae needs to be assessed before using borers in a biological control program. The im pact of microorganisms on root-borers should also be assessed. Harris et al. (1969) noted that Agrilus larvae employed in the control of Klamath weed were particularly subject to fungal attack on damp sites. The impact of root-boring beetle larvae, other insects, and microorganisms on G. sarothrae plants could also be modified by the resistance of G.„ sarothrae plants to attack and by the ability of plants to compensate for damage, Results of the present study indicated that susceptibility to attack by root-boring larvae was rather low in young plants and increased with plant height and age. Approximately two- thirds of plants under 40 cm tall escaped borer injury, while those over 40 cm tall sustained increasingly higher frequencies of attack. Campbell and Bomberger (1934) noted these same trends. They found that 55 well established seedlings between two and four years old were fairly resistant to insect injury, while large mature plants were much more susceptible. Older plants in the present study often sustained re peated attacks by borers over, several years. These repeated attacks by borers may have resulted in the increased impact on older plants, Pesho (1975) noted that weevil injury to alfalfa roots was also cumula tive and damage increased in severity and incidence as plant age increased. The potential of root-borers for limiting sarothrae stands may be decreased by the ability of plants to compensate for damage, Campbell and Bomberger (1934) noted that G, sarothrae seedlings es tablish a deep taproot during the first season and develop abundant lateral roots as they mature. Observations of plants suggest that plants may put out abundant lateral roots after periods of rainfall. Production of these lateral roots by Gutierrezia plants may compensate for the reduced stability of plants infested by borers, Reproduction by G, sarothrae plants may maintain th$ stand even if established plants are destroyed by borers, Lillie (1963) noted that even when G, sarothrae plants were destroyed by herbicides, wide spread establishment of seedlings maintained the density of the stand. If root-boring insects were used in a biological control program, re production of G, sarothrae plants would probably also have to be de creased by seed insects or traditional control methods. Climatic influences may modify the impact of borers on G o sarothrae plants. Climatic conditions which adversely affect growth and development of Gutierrezia might intensify the effects of borers. 56 Campbell and Bomberger (1934) noted that borer infestations coupled with low rainfall resulted in mortality of G, sarothrae plants. Op timum moisture and temperature conditions might decrease impact of borers by increasing plant vigor, encouraging root development, and favoring reproduction. Climate also may affect the root-boring larvae. Graham (1924) noted that subcortical temperatures may affect borers by determining rate of development, distribution, and in many cases the percentage mortality. Larvae in the present study showed substantial differences in development and emergence patterns. Emergence was delayed by approxi mately one month in larvae reared in an environment that was more moist and 3° C cooler than field conditions. If root-boring larvae are used in a biological weed control program, their impact on G. sarothrae plants will need to be increased by manipulation by man, Ueckert et al„ (1971) noted that a particular insect species may be augmented by mass rearing, by transporting adults to new areas free of natural enemies, or by control of predators and parasites. The. borers in the present study were amenable to rearing on artificial media and M. lineolata mated and oviposited in the laboratory situation. Thus, these borers might be desirable subjects for a mass rearing program. The geographical range and ecological requirements, of the borers warrant further research before introductions could be at tempted in other areas. Further studies concerning Which predators and parasites affect the borers and to what degree these insects affect the borers would be needed to assess the potential of augmentation of borer populations by predator and parasite control. 57 Traditional range management or grazing practices may also affect borer populations and might be modified to augment impact of borers on G. sarothrae plants. Smith (1970) found that Coleoptera were severely reduced on overgrazed prairie in Oklahoma. However, Smith sampled mainly leaf and seed feeders and neglected the Buprestidae and root-boring Curculionidae, In addition to grazing, burning may affect insect populations. Bertwel! and Blocker (1975) found that num bers of weevils were highest on burned, grazed pasture treatments. Chemical treatment of plants with herbicides also affects borer popu lations, Ueckert and Wright (1974) found that wood-boring larvae in creased in honey mesquite trees that were top-killed by fire or drought, but not in trees top-killed with 2,4,5-T sprays. Girdling and basal treatment of trees with diesel oil or diesel oil plus 2,4,5-T also resulted in high borer activity. Considerable research will be needed before the potential for using root-boring insects in controlling G» sarothrae can be completely assessed. Even if borers are never used in an applied program the present study indicates that borers are important components in the rangeland ecosystem and may help to limit G, sarothrae populations. LIST OF REFERENCES Andres, L. A„ 1971. The suppression of weeds with insects. Proc. Tall Timbers Conf. on Ecol. Anim. Control by Habitat Manage. No. 3:185^195. Andres, L. A. and G. W, Angalet. 1963. Notes on the ecology and host specificity of Mierolarinus lareynii and M. lypriformis (Coleoptera: Curculionidae) and the biological control of puncture vine, Tribulus terrestris. J . Econ. Entomo1 . 56: 333-340. Arnold, J. F. 1964. Zonation of vegetation around a juniper tree. J. Range Manage. 17:41-42. Benson, L. and R, A. Darrow. 1952. The Trees and Shrubs of the Southwestern Deserts. University of Arizona Press. Tucson, pp. 333-336. Bertwell, R. L. and H. D. Blocker. 1975. Curculionidae from different grassland treatments near Manhattan, Kansas. J. Kansas Entom. Soc. 48:319-326. Borror, D. J., D. M. DeLong, and C. A. Triplehorn. 1976. An Intro duction to the Study of Insects. Holt, Rinehart, and Winston. New York, p. 400. Campbell, R. S. and E. H. Bomberger. 1934. The occurrence of Gutierrezia sarothrae on Bouteloua eriopoda ranges in southern New Mexico. Ecology. 15:49-61. Furniss, M. M. and W. F. Barr.' 1975. Insects affecting important native shrubs of the northwestern United States. USDA Forest Serv. Gen. Tech. Report INT-19. Ogden, Utah. 64 pp. Gesink, R. W., H. P. Alley, and G. A. Lee. 1973. Vegetative response to chemical Control of broom snakeweed on blue grama range. J. Range Manage. 26:139=143. Graham, S, A. 1924. Temperature as a limiting factor in the life of subcortical insects, J. Econ. Entdmol. 17:377-383, Harris, P., D. Peschken, and J. Milroy. 1969. The status of biological control of the weed Hypericum perforatum in British Columbia. Can, Entomol. 101:1-15. 58 59 Jameson, D„ A„ 1966. Competition in a blue grama-broom snakeweed- actinea community and responses to selective herbicides. J. Range Manage. 19:121-124. Jameson, D. A. 1970. Value of broom snakeweed as a range condition indicator. J . Range Manage. 23:302-304. 1 Jameson, D. A. and E. H. Reid. 1965, The pinyon-juniper type of Arizona. J. Range Manage, 18:152-153. Kearney, T„ H„ arid R. H, Peebles, 1973. Arizona Flora, University of California Press. Lbs Angeles, pp. 852-853, Lillie, D. T, 1963, Control of Arizona chaparral with 2,4,5-T and Silves. J. Range Manage, 16:195-199. Linsley, E. G. and J, A, Chemsak. 1961. A distributional and taxonomic study of the genus Crossidius (Coleoptera: Cerambycidae), Misc. Publ, Entomo1. Soc. Amer. 3:26-64. Parker, Kittie F , 1972. An Illustrated Guide to Arizona Weeds, The University of Arizona Press, Tucson, pp. 292-293. Patana, R. 1969. Rearing cotton insects in the laboratory. USDA/ARS Prod. Res. Rep. 108, 6 pp. Pesho, ( ?, R, 1975. Clover root curculio: estimates of larval injury to alfalfa tap roots. J. Econ. Entomof. 68:61-65, Rogers, C. E, arid J. C, Garrison, 1975. Seed destruction in indigo- bush amorpha by a seed beetle, J . Range Manage. 28:241-242. Schmutz, D. M,, B, N, Freeman, and R. E. Reed, 1974. Livestock Poisoning Plants of Arizona. The University of Arizona Press. Tucson, p. 40. Schmutz, E, M, and D. E, Little. 1970, Effects of 2,4,5-T and picloram on broom snakeweed in Arizona. J. Range Manage. 23:354-357. . Schmutz. . E„ M. and D. W, Whitman. 1962. Shrub control studies in the oak-chaparral of Arizona. J. Range Manage. 15:61-67. Smith, ( C . 1970, The effect of overgrazing and erosion upon the biota of the mixed-grass prairie of Oklahoma. Ecology. 21: 381-397. Smith, 1). A, and E. M. Schmutz. 1975. . Vegetative changes on protected versus grazed desert grassland ranges in Arizona. J . Range Manage, 28:453-458. 60 Smith, L. L, and D. N. Ueckert. 1974. Influences of insects on mesquite seed production. J. Range Manage. 27:61-65. Stoddart, L. A., A. D, Smith, and T. W. Box. 1975. Range Management. McGraw-Hill Book Company. New York, 532 pp. Ueckert, D. N. 1973. Effects of leaf-footed bugs on mesquite repro duction. J. Range Manage. 26:227-229. Ueckert, D. N., K. L. Polk, and C. R. Ward. 1971. Mesquite twig girdler: A possible means of mesquite control. J. Range Manage. 24:116-118. Ueckert, D. N. and H. A. Wright. 1974. Wood boring insect infesta tions in relation to mesquite control practices. J . Range Manage. 27:383-386. Vine, R. A. 1960. Trees. Shrubs, and Woody Vines of the Southwest. University of Texas Press. Austin, pp. 994-995.