Blackwell Publishing AsiaMelbourne, AustraliaWBMWeed Biology and Management1444-61622006 Weed Science Society of JapanDecember 200664204211Original ArticleArthropods on Rosa multiflora branch tipsL. C. Jesse
et al.
Weed Biology and Management 6, 204–211 (2006)
RESEARCH PAPER
Abundance of arthropods on the branch tips of the
invasive plant, Rosa multiflora (Rosaceae)
LAURA C. JESSE,1* KIRK A. MOLONEY2 and JOHN J. OBRYCKI3
Departments of Entomology and 2Ecology, Evolution and Organismal Biology, Iowa State University, Ames, IA and
3
Department of Entomology, University of Kentucky, Lexington, KY, USA
1
Rosa multiflora Thunb. (Rosaceae) is an invasive species in the USA, where it grows in pastures
and wooded areas. A disease of unknown etiology, rose rosette disease (RRD), infects
R. multiflora and other Rosa spp. The goal of this research was to determine the effects of habitat and disease symptoms on the abundance of Phyllocoptes fructiphilus Keifer (Acari: Eriophyidae), the presumed vector of RRD, and other arthropods on R. multiflora. We collected
branch tips from healthy R. multiflora plants growing in the sun and shade, as well as RRDinfected R. multiflora growing in the sun. The samples were collected from June 2002 to April
2004 from three sites in Iowa, USA. The samples were collected approximately every 2 weeks
during the summer, monthly during the fall, and once during early spring. The fall samples
were only taken from RRD-infected plants because they retain leaves throughout the winter.
We found that P. fructiphilus was present on diseased and healthy R. multiflora growing in the
sun and on healthy R. multiflora growing as understory plants under a tree canopy (shaded), but
the greatest numbers were observed on the diseased plants growing in the sun. Several other
mite species, both predatory and phytophagous, Chaetosiphon sp. aphids, and the thrips species,
Frankliniella exigua Hood and Neohydatothrips variabilis Beach, occur in the same plant microhabitat as P. fructiphilus. This is the first study to document the presence of additional arthropods in the same microhabitat (branch tips) as P. fructiphilus. Future research needs to isolate
and identify the causal agent of RRD so it can be confirmed that, of the many arthropods
feeding on R. multiflora, only P. fructiphilus vectors RRD.
Keywords: biological control, invasive species, non-target effects, Rosa multiflora, rose rosette
disease.
INTRODUCTION
Rosa multiflora Thunb. (Rosaceae: Rosa), native to Japan,
Korea, and parts of China (Steavenson 1946), was first
introduced into the USA as an ornamental plant in the
early 1800s (Rehder 1936). In the 1940s and 1950s,
R. multiflora was planted as a living fence, for wildlife
habitat, and as a means to control soil erosion in the eastern and midwestern USA (Klimstra 1956). Living fences
*Correspondence to: Laura Jesse, Iowa State University, 104 Insectary, Ames, IA 50011, USA.
Email: ljesse@iastate.edu
Received 26 December 2005; accepted 14 June 2006
are plants grown closely together for the purpose of containing livestock. Rosa multiflora and osage orange,
Maclura pomifera (Moraceae), were promoted as affordable fencing alternatives and to make farms better habitat
for game animals by providing wildlife cover (Lewis
1941; Steavenson et al. 1943; Epstein et al. 1997). Initially, it was assumed that R. multiflora would not spread
extensively or invade pastures and wooded habitats
(Steavenson 1946); however, within a decade R. multiflora was spreading due to seed dispersal and vegetative
growth (Klimstra 1956).
Rosa multiflora currently infests ≈ 45 million acres
(18.2 million ha) in the eastern USA, reducing the value
of land for grazing and recreational purposes because
dense patches are impassable and are not consumed by
© 2006 The Authors
Journal compilation © 2006 Weed Science Society of Japan
doi:10.1111/j.1445-6664.2006.00222.x
Arthropods on Rosa multiflora branch tips
most livestock (except goats) (Loux et al. 2005). Rosa
multiflora has been declared a noxious weed in nine
states, including Iowa (Amrine & Stasny 1993). Infestations are most common in pastures and in nearby
forested areas, similar to areas where it was intentionally planted (Steavenson et al. 1943; Steavenson 1946;
Burgess 1948; Epstein et al. 1997).
As a result of a large seed bank, any reduction of
R. multiflora must include long-term management plans
to destroy new seedlings as they appear (Loux et al.
2005). Currently, mechanical and herbicidal control
methods are most commonly used to suppress R. multiflora (Loux et al. 2005). These methods are expensive,
time-consuming, often ineffective, and have negative
environmental effects, including the destruction of nontarget vegetation (Loux et al. 2005).
Biological control might offer a long-term, viable solution to R. multiflora management: one macro-organism
and one disease have been identified as having the
potential to provide biological suppression of R.
multiflora. The first is Megastigmus aculeatus (Hymenoptera; Torymidae), whose larvae feed in R. multiflora seeds
(Balduf 1959). The second is rose rosette disease (RRD),
a disease with unknown etiology and uncertain classification, which reduces plant vigor and eventually kills
R. multiflora. Despite extensive study, the causative agent
of RRD has not been identified (Epstein & Hill 1999).
The symptoms of RRD (witches broom or mosaic discoloration) were first reported in an unidentified rose
species in Manitoba, Canada, in 1940 and 1941 (Conners 1941, 1942). It was reported in the USA in 1941, in
Wyoming, from an ornamental rose, Rosa rubriflora; an
unidentified native rose species also was noted to have
symptoms of RRD in the same area in 1942 (Thomas &
Scott 1953). Based on the disease symptoms and the
ability to infect healthy plants by grafting, the causal
agent of RRD was believed to be a virus (Thomas &
Scott 1953). Additionally, the symptoms are observed
when diseased plants are treated with the antibiotic, tetracycline, which kills bacterial pathogens. Evidence also
suggests that the agent is not a phytoplasma because tests
using polymerase chain reactions with primers known to
detect phytoplasma failed to detect this type of pathogen
(Epstein & Hill 1999). Thus, the causal agent, which is
not identified, is presumed to be a virus.
Rose rosette disease is presumed to be spread by an eriophyid mite, Phyllocoptes fructiphilus (Acari: Eriophyidae)
(Allington et al. 1968; Amrine et al. 1988); however,
Doudrick et al. (1986) could not demonstrate the transmission of RRD to R. multiflora by P. fructiphilus in
205
greenhouse trials. Doudrick et al. (1986) stated that
the inability to establish colonies of P. fructiphilus on
R. multiflora is the reason they failed to observe mite
transmission of RRD. On the basis of the studies conducted by Allington et al. (1968) and Amrine et al.
(1988), it is generally accepted that P. fructiphilus is the
only vector of RRD. However, only P. fructiphilus and
Tetranychus urticae, the two-spotted spider mite, have
been studied as possible vectors (Amrine et al. 1988).
Phyllocoptes fructiphilus has a four-stage eriophyid mite life
cycle: egg, protonymph, deutonymph, and adult (Kassar
& Amrine 1990). In a 4 year study conducted in Indiana,
Amrine (1996) observed higher mite populations on
R. multiflora with RRD compared to mite populations
on healthy R. multiflora (Amrine 1996). Kassar and
Amrine (1990) found P. fructiphilus wintered on
R. multiflora singly and in groups under bud scales and
loose bark in the previous year’s growth.
Previous researchers have noted an apparently lower
incidence of RRD in multiflora rose plants growing in
shaded areas as an understory plant under a tree canopy
(Epstein & Hill 1995). However, the effect of habitat on
populations of P. fructiphilus has not been examined previously. It is unknown whether this is related to physiological differences in the plants (e.g. effects of lower
light intensity on plant growth, which might influence
the disease) or if the mite prefers more open, sunny
locations (e.g. related to differences in mite dispersal
behaviors). If differences are observed in relation to
P. fructiphilus occurrence on R. multiflora plants growing
in open canopy (sunny habitats), in comparison to plants
growing in the understory (shaded habitats), this might
indicate that RRD is not an effective mortality factor of
plants in the shady habitats if RRD is only vectored by
P. fructiphilus. Many R. multiflora-infested pastures are
adjacent to wooded areas that also contain multiflora
rose. Rosa multiflora plants in these wooded areas might
remain unaffected by RRD and serve as a seed source to
reinfest pastures.
The objectives of this study were to: (i) document the
effects of the R. multiflora habitat (open vs closed
canopy) on P. fructiphilus; (ii) examine the effects of
RRD symptoms on the abundance of P. fructiphilus on
R. multiflora growing in open habitats; and (iii) quantify
the abundance of selected phytophagous arthropods
(aphids, thrips, and additional mite species) found on the
growing tips of R. multiflora branches, the same microhabitat used by P. fructiphilus, because these other arthropods are potential disease vectors.
© 2006 The Authors
Journal compilation © 2006 Weed Science Society of Japan
206
L. C. Jesse et al.
MATERIALS AND METHODS
Two sites were located in Story County Conservation
Board parks in central Iowa (Christiansen Forest Preserve [41°54′Ν, 93°34′W] and Dakins Lake [42°10′N,
98°34′W]) and one site was located in an infested pasture
in Boone County, Iowa (42°06′N, 93°56′W). Samples
were collected from June 2002 to April 2004 from the
two sites in Story County and from November 2002 to
April 2004 at the Boone County site. The samples were
collected approximately every 2 weeks during the summer, monthly during the fall, and once during early
spring.The fall samples were only taken from the RRDinfected plants which, unlike the unsymptomatic plants,
retain leaves throughout the winter.
On each sampling date, 20 branch tips (≈ 6 cm) were cut
from each of three R. multiflora plants in open areas (no
trees within 5 m) and three plants growing as understory
plants (shaded) with no visible symptoms of RRD, and
from three R. multiflora roses also growing in open areas,
with symptoms of RRD.The 20 branch tips were placed
in plastic sandwich-type bags with a zip seal to prevent
the arthropods from moving off the branches. No plants
growing in the understory with symptoms of RRD
were observed at these three sites so we could not sample
RRD-symptomatic plants growing in the shade as an
understory plant. At the sites in Story County, 42 samples were collected from R. multiflora in both sunny and
shady sites and 57 samples were taken from plants with
RRD. At the Boone County site, R. multiflora plants
growing in the shade or sun were sampled 24 times and
RRD plants were sampled 39 times. As a result of the
relatively low number of plants in some habitats, the
same plant was often sampled multiple times.
A wash solution of tap water, 0.2% dish soap, and 2%
sodium hypochlorite was used to remove the arthropods
from the foliage (de Lillo 2001).The 20 branch tips from
a single R. multiflora were weighed and then placed
together in a 400 or 1000 mL glass beaker. The leaves
were removed and counted from each stem to expose
any arthropods in the basal attachment sites of the leaves.
The bags which held the branch tips were rinsed with
the washing solution, the contents were poured into the
beaker, and additional washing solution was added until
the leaf tissue was fully covered. A magnetic stirring bar
(Fisher Scientific, Pittsburg, PA, USA) was placed in the
beaker and the plant material and washing solution was
stirred for 5–10 min at room temperature (de Lillo
2001). The solution was filtered through four stainless
steel sieves stacked by descending mesh size; 850 µm,
180 µm, 53 µm, and 25 µm (Topac, Hingham, MA,
USA). The plant material was rinsed in tap water in the
850 mm sieve to dislodge mites and other arthropods (de
Lillo 2001). Phyllocoptes fructiphilus range in size from
168–204 mm long and 50–60 µm wide, and so the 180,
53, and 25 mm screens were examined for mites and
other arthropods. A wash bottle filled with the wash
solution was used to rinse the screen contents into an
8.5 cm diameter (56.7 cm2) Petri dish. One to two drops
of dish soap were added to the Petri dish and mixed by
carefully moving the Petri dish back and forth (de Lillo
2001). The Petri dish was allowed to sit for at least 1 min
so the mites settled to the bottom, then the mites were
counted under a dissecting microscope (Olympus SZH;
Leeds Precision Instruments, Minneapolis, MN, USA).
Subsampling of the 53 and 25 mm screen contents was
used to estimate the arthropod numbers. A subsample
consisted of counting the arthropods in 0.5 cm × 0.5 cm
squares arranged in a figure “8” shape across the Petri
dish; 38 squares were examined for a total of 9.5 cm2.
Only P. fructiphilus mites were observed in the 25 mm
Petri dish. If there were fewer than 10 P. fructiphilus in
the 53 mm dish, the 25 mm dish was not counted.
Selected arthropods were sent to the United States
Department of Agriculture Systematic Entomology Laboratory, Beltsville, MD, USA, for identification. Voucher
specimens are deposited in the Iowa State University
Entomology Department Insect Museum.
The arthropod abundance data were summarized per
weight (gram) of the 20 branch tips collected from each
plant. The data were log-transformed prior to statistical
analysis, so the median values are presented. Analysis of
variance was used to analyze site (three sites), habitat
(sun, shade, RRD plants in the sun), and site by habitat
effects. Within an effect, the least squares means were
compared with the Tukey test (SAS 9.1 for Windows;
SAS Institute, Cary, NC, USA).
RESULTS
Plants with the symptoms of RRD had the most leaves
(Fig. 1), which was expected because one of the symptoms of RRD is increased leaf production. However, we
observed no correlation between the number of leaves
and the number of P. fructiphilus (linear R2 = 0.0004). We
observed aphids, thrips, eriophyid mites, and other mite
species on healthy and diseased R. multiflora growing in
the sun and on healthy R. multiflora growing in shaded
habitats (Fig. 2). The most commonly observed arthropods were eriophyid mites on RRD-symptomatic plants.
The median number of eriophyid mites found on
R. multiflora was different depending on the habitat the
plant was growing in (sun or shade) and whether or not
© 2006 The Authors
Journal compilation © 2006 Weed Science Society of Japan
160
(a)
140
300
Average arthropods g–1
Average
number of leaves
Arthropods on Rosa multiflora branch tips
120
100
80
60
40
250
200
150
100
50
20
0
207
0
Shade
Sun
Sun + RRD
Shade Sun
Location of plant
Aphids
Fig. 1. Average number of leaves (±SE) on 20 Rosa multiflora branch tip samples over a 2 year period in three field
sites. RRD, rose rosette disease.
Non-eriophyid mites found in the R. multiflora samples
included predatory mites, fungivorous mites, and phytophagous mites (Table 1). The non-eriophyid mite
species were most abundant during the summer. More
non-eriophyid mites were found on R. multiflora growing in the shade (P < 0.0001). In the sunny habitat,
similar numbers of mites were observed on RRDsymptomatic plants and apparently healthy plants
(P = 0.7).The number of mites at the three field sites was
also significantly different (P = 0.0006), with Dakins
having the most and Boone having the fewest mites.
The thrips species observed on R. multiflora included
Frankliniella exigua Hood and Neohydatothrips variabilis
Beach (Table 1). The highest number of thrips was
Sun
Sun & Shade
RRD
Eriophyid mites
Sun
Sun & Shade
RRD
Other mites
Sun
Sun &
RRD
Thrips
(b)
Average arthropods g–1
700
600
500
400
300
200
100
0
Shade
Sun
Aphids
Sun & Shade
RRD
Sun
Sun & Shade
RRD
Eriophyid mites
Sun
Sun & Shade
RRD
Other mites
Sun
Sun &
RRD
Thrips
(c)
500
Average arthropods g–1
the plant was diseased (Fig. 3) (P < 0.006). The greatest
number of eriophyid mites was found on the plants
showing symptoms of RRD (P < 0.0001) and plants in
the sun had more eriophyid mites than plants growing in
the shade (P = 0.006). At the three sites, the number of
mites tended to be highest during the summer months,
although eriophyid mites were present on the RRD
plants on all sampling dates. There were significant differences among the number of eriophyid mites at the
three field sites (P < 0.0001). The number of mites at
Dakins and Christiansen was similar (P = 0.34), but
Boone was different from both Dakins and Christiansen
(P < 0.0001). This is probably related to the different
sampling times; Boone was not sampled during the summer of 2002. Even though the sampling times differed,
the number at the Boone site was generally lower. Most
of the eriophyid mites were identified as P. fructiphilus;
however, two mites from Dakins were identified as Phyllocoptes adalius Keifer (Table 1). Phyllocoptes adalius is not
known to be a vector of RRD (Amrine et al. 1994).
Sun & Shade
RRD
450
400
350
300
250
200
150
100
50
0
Shade
Sun
Aphids
Sun & Shade
RRD
Sun
Sun & Shade
RRD
Eriophyid mites
Sun
Sun & Shade
RRD
Other mites
Sun
Sun &
RRD
Thrips
Insects
Fig. 2. Average number of arthropods per gram (±SE) on
the branch tips from three healthy Rosa multiflora plants
growing in the shade and in the sun, and from three
R. multiflora plants exhibiting symptoms of rose rosette disease (RRD) at three locations in central Iowa: (a) Christiansen Park, Story County, (b) Dakins Park, Story County
and (c) Boone County.
observed at Dakins Park (P < 0.0001). Christiansen Park
and the Boone site had similar numbers of thrips
(P = 0.7). The habitat (shady or sunny) of the
R. multiflora and the disease status did not have an effect
on the number of thrips present (P = 0.08).
© 2006 The Authors
Journal compilation © 2006 Weed Science Society of Japan
L. C. Jesse et al.
208
(a)
Average no. of Eriophyidae
18 000
16 000
14 000
12 000
10000
8000
6000
4000
2000
9/17/2003
10/17/2003
11/17/2003
12/17/2003
1/17/2004
2/17/2004
3/17/2004
4/17/2004
9/13/2003
11/13/2003
12/13/2003
1/13/2004
2/13/2004
3/13/2004
4/13/2004
8/17/2003
8/13/2003
10/13/2003
7/17/2003
7/13/2003
5/17/2003
6/17/2003
4/17/2003
3/17/2003
2/17/2003
1/17/2003
12/17/2002
11/17/2002
10/17/2002
8/17/2002
9/17/2002
7/17/2002
6/17/2002
0
Date
(b)
Average no. of Eriophyidae
18 000
16 000
14 000
12 000
10 000
8000
6000
4000
2000
6/13/2003
5/13/2003
4/13/2003
3/13/2003
2/13/2003
1/13/2003
12/13/2002
11/13/2002
10/13/2002
8/13/2002
9/13/2002
7/13/2002
6/13/2002
0
Date
(c)
Average no. of Eriophyidae
18 000
16 000
14 000
12 000
10000
8000
6000
4000
2000
Date
The aphids identified from R. multiflora were all Chaetosiphon spp. (Table 1). The aphids were abundant during
the summer at the three field sites. There was a significant difference in the numbers of aphids at the three field
4/22/2004
3/22/2004
2/22/2004
1/22/2004
12/22/2003
11/22/2003
10/22/2003
9/22/2003
8/22/2003
7/22/2003
6/22/2003
5/22/2003
4/22/2003
3/22/2003
2/22/2003
1/22/2003
12/22/2002
11/22/2002
0
Fig. 3. Average number of eriophyid mites per gram over time on
the branch tips from three healthy
Rosa multiflora plants growing in the
shade and in the sun, and from three
R. multiflora plants exhibiting symptoms of rose rosette disease (RRD)
at three locations in central Iowa:
(a) Christiansen Park, Story County,
(b) Dakins Park, Story County and
(c) Boone County. (), sun; (), sun
and RRD; (), shade.
sites (P = 0.01), with the highest number at Boone and
the lowest number at Christiansen Park. There also was
a significant difference in the number of aphids in each
habitat (P = 0.048). Rosa multiflora with RRD had the
© 2006 The Authors
Journal compilation © 2006 Weed Science Society of Japan
Arthropods on Rosa multiflora branch tips
209
Table 1. Identification of selected arthropods on the
branch tips of Rosa multiflora
that mite populations on symptomatic R. multiflora were
14-fold higher than those on unsymptomatic plants.
Taxa
The lowest number of P. fructiphilus was observed on the
R. multiflora growing in the shade as understory plants.
Epstein and Hill (1995) found a lower number of RRDsymptomatic R. multiflora in shaded areas and a slower
increase of disease incidence over a 6 year period in
shaded areas compared to plants growing in full sun.
They hypothesized that this could be related to lower
ambient air temperature affecting the mites or some other
unknown factor that reduced the mites’ capacity to vector
RRD.We confirmed that there are fewer P. fructiphilus on
plants growing in full shade, but we hypothesize this could
be a result of the absence of RRD, not the cause. Plants
showing symptoms associated with RRD often have
greatly increased density of leaf growth, thus providing
many more leaf axil microhabitats for P. fructiphilus mite
populations. This interpretation is consistent with the
observation that there are fewer P. fructiphilus on plants in
the sun with no symptoms of RRD.
Arachnida
Acari†
Eriophyidae
Phyllocoptes fructiphilus Keifer
Phyllocoptes adalius Keifer
Phytoseiidae
Neoseiulus spp.
Phytoseius spp.
Stigmaeidae
Cheylostigmaeus spp.
Tarsonemidae
Tarsonemus waitei Banks
Tetranychidae
Panonychus ulmi Koch
Tetranychus spp.
Tydeidae
Pronematus spp.
Tydeus spp.
Insecta
Hemiptera‡
Aphididae
Chaetosiphon spp.
Thysanoptera§
Thripidae
Frankliniella exigua Hood
Neohydatothrips variabilis Beach
Number of
individuals
identified
15
2
1
1
3
1
1
4
Use of Phyllocoptes fructiphilus as a disease vector
3
4
20
3
5
† Identified by Ronald Ochoa, USDA-SEL, Beltsville, MD, USA;
‡ identified by Gary L. Miller, USDA-SEL; § identified by Sueo
Nakahara, USDA-SEL.
most aphids, whereas healthy R. multiflora (growing in
the sun or shade) had similar numbers of aphids.
DISCUSSION
This study is the first to document the presence of other
possible arthropod vectors on R. multiflora with RRD, as
well as evaluating the effects of habitat on populations of
P. fructiphilus. Phyllocoptes fructiphilus populations were
highest on the R. multiflora with RRD symptoms. These
results are similar to those reported by Epstein and Hill
(1999), who observed much higher numbers of mites on
symptomatic R. multiflora compared to unsymptomatic
plants sampled from May to September from 1990 to
1992 at two sites in Iowa. Amrine (1996) also reported
Rose rosette disease has the potential to be an important
tool in managing R. multiflora. Epstein et al. (1997) demonstrated RRD will readily spread to healthy plants in
plots where RRD is augmented. Rose rosette disease
reduces the flowering and seed set of infected plants and
results in the death of the plant within ≈ 5 years (Epstein
et al. 1993). As the presumed vector of RRD, it is
important to understand the occurrence of P. fructiphilus
on R. multiflora. The augmentation of the disease can be
successfully accomplished by grafting buds from diseased
plants onto healthy plants (Epstein 1995); however, this
process is too labor-intensive to be used in heavily
infested areas. As an alternative, land managers cut the
branches of diseased plants and place them on healthy
plants with the expectation that P. fructiphilus present on
the branches will move onto the healthy plants. There is
no evidence that the mites will move onto the healthy
plants, but this study clearly shows that higher mite populations are consistently found on R. multiflora with the
symptoms of RRD.
There are two major concerns in the manipulation of
P. fructiphilus and RRD for the biological control of
R. multiflora. The first is that ornamental roses are susceptible to RRD, particularly hybrid tea roses (Epstein
& Hill 1999). Although the movement of the mite onto
ornamental roses, causing infection, appears to be limited, the risk of RRD is a concern to rose-growers and
the rose industry (Dettmann & Pagliai 1993; Harwood
1994). Unfortunately, because RRD naturally occurs in
© 2006 The Authors
Journal compilation © 2006 Weed Science Society of Japan
210
L. C. Jesse et al.
R. multiflora populations, there will always be a source of
inoculum with or without augmentation. A better
understanding of the temporal and spatial dynamics of
mite populations throughout the year will help determine the ways to limit the potential spread of RRD
from multiflora rose to ornamental roses.
The second problem with the augmentation of RRD is
that the causative agent of RRD has not been isolated
and identified (Di et al. 1990; Epstein & Hill 1999;
Rohozinski et al. 2001). Many attempts have been made
to isolate the causal agent, but none have been successful
(Rohozinski et al. 2001). Di et al. (1990) found doublestranded RNA associated with diseased tissue, indicating
the causal agent could be a cryptic virus. Rohozinski
et al. (2001) successfully transmitted a virus-like agent
from R. multiflora to Nicotiana spp., but observed no
virus-like particles.
Arthropod transmission studies of RRD have only been
conducted with P. fructiphilus and T. urticae, the twospotted spider mite, primarily because T. urticae commonly infests greenhouse-grown R. multiflora used in
experiments. Tetranychus urticae was not observed to
transmit RRD (Amrine et al. 1988). Our current study
documented that other known vectors of plant pathogens commonly occur on R. multiflora. Aphids, thrips,
and some mites are known vectors of plant viruses (Maramorosch 1963), but it is unknown if they are involved
in the transmission of RRD.
Chaetosiphon spp. aphids were common on the
R. multiflora and this genus transmits plant viruses. For
example, Chaetosiphon fragaefoili (Cockerell), the strawberry aphid, and Chaetosiphon jacobi both transmit strawberry viruses (Frazier 1968; Krczal 1979). One of the
thrips species (Frankliniella sp.) observed on R. multiflora
is from a genus known to transmit plant tospoviruses
(Maramorosch 1963). If these arthropods transmit RRD,
they could spread the disease among R. multiflora plants
and also to non-target ornamental roses.
Future research needs to isolate and identify the causal
agent of RRD so it can be confirmed that, of the many
arthropods feeding on R. multiflora, only P. fructiphilus
vectors RRD. Until this is done, it is impossible to confirm that large-scale augmentation of RRD to control
R. multiflora can be implemented without increasing
non-target RRD infections.
Use of eriophyid mites in weed biological control
Phytophagous mites are good candidates for successful
weed biological control programs because of their lim-
ited host range; it is estimated that 80% of eriophyids are
monophagous (Briese & Cullen 2001). In addition,
mites have a high rate of population increase and often
have good dispersal ability. Eriophyid mites have been
used in biological control programs against invasive
weeds, including St John’s wort, rush skeletonweed, and
field bindweed (Briese & Cullen 2001). Many eriophyids cause galling in plants, altering resource allocation
and reducing plant vigor. Non-gall forming eriophyids
also affect plant growth. For example, Aculus hyperici,
feeding in the growing tips of St John’s wort, causes a
decrease in leaf size and shortening of the internodal distances (Briese & Cullen 2001). Currently, P. fructiphilus is
the only eriophyid mite that is being considered for use
as a disease vector in biological control (Briese & Cullen
2001).
ACKNOWLEDGMENTS
We would like to thank the Story County Conservation
Board for letting us conduct this survey in their parks
and Norma Neyva-Estrada, Statistics Department, Iowa
State University, for assistance with statistical analysis,
and the USDA-SEL, Beltsville, MD, USA, for the identification of our arthropods. This research was partially
funded by a grant from the College of Agriculture at
Iowa State University.This is a Journal Paper of the Iowa
Agriculture and Home Economics Experiment Station,
Ames, Iowa, Project No. 6628, supported by the Hatch
Act and State of Iowa funds.
REFERENCES
Allington W.B., Staples R. and Viehmeyer G. 1968. Transmission of rose
rosette virus by the eriophyid mite Phyllocoptes fructiphilus. J. Econ. Ent.
61, 1137–1140.
Amrine J.W. 1996. Phyllocoptes fructiphilus and biological control of
multiflora rose. In: Eriophyid Mites, World Crop Pests, Vol. 6 (ed. by
Lindquist E.E., Sabelis M.W. and Bruin J.). Elsevier Science B. V.
Amsterdam, 741–749.
Amrine J.W., Hindal D.F., Stasny T.A., Williams R.L. and Coffman C.C.
1988. Transmission of the rose rosette disease agent to Rosa multiflora by
Phyllocoptes fructiphilus (Acari: Eriophyidae). Ent. News 99, 239–252.
Amrine J.W., Kassar A. and Stasny T.A. 1994. Phyllocoptes fructiphilus
(Acari: Eriophyidae), the vector of rose rosette disease, taxonomy,
biology and distribution. In: Epstein A.H. and Hill J.H., eds. Rose
Rosette Symposium 1994 – Iowa State University (Ames, IA, USA, 19–21
May 1994). Iowa State University Press, Ames, IA, 61–66.
Amrine J.W. and Stasny T.A. 1993. Biocontrol of multiflora rose. In:
Biological Pollution: the Control and Impact of Invasive Exotic Species (ed. by
McKnight B.N.). Indiana Academy of Science, Indianapolis, IN, 9–21.
Balduf W.V. 1959. Obligatory and facultative insects in rosehips, their
recognition and bionomics. Ill. Biol. Monogr. 26, 1–194.
Briese D.T. and Cullen J.M. 2001. The use and usefulness of mites
in biological control of weeds. In: Acarology: Proceedings of the 10th
International Congress (Canberra, Australia, 5–10 July 1998). CSIRO
Publishing, Melbourne, 453–463.
Burgess C.H. 1948. Living fence. Farm Q. 3, 50–54.
© 2006 The Authors
Journal compilation © 2006 Weed Science Society of Japan
Arthropods on Rosa multiflora branch tips
Conners I.L. 1941. Twentieth Annual Report of the Canadian Plant Disease
Survey 1940. Domain of Canada Department of Agriculture Science
Service, Division of Botany and Plant Pathology, Ottawa.
Conners I.L. 1942. Twenty-First Annual Report of the Canadian Plant Disease
Survey 1941. Domain of Canada Department of Agriculture Science
Service, Division of Botany and Plant Pathology, Ottawa.
Dettmann E. and Pagliai E.J. 1993. Rose rosette disease: too many
unanswered questions. Iowa Hortic. 9, 8.
Di R., Hill J.H. and Epstein A.H. 1990. Double-stranded RNA
associated with the rose rosette disease of multiflora rose. Plant Dis.
74, 56–58.
Doudrick R.L., Enns W.R., Brown M.F. and Millikan D.F. 1986.
Characteristics and role of the mite, Phyllocoptes fructiphilus (Acari:
Eriophyidae) in the etiology of rose rosette. Ent. News 97, 163–168.
Epstein A.H. 1995. Biological Control of Multiflora Rose with Rose Rosette
Disease. University extension PM-1615. Iowa State University, Ames,
IA.
Epstein A.H. and Hill J.H. 1995. The biology of rose rosette disease: a
mite-associated disease of uncertain aetiology. J. Phytopathol. 143, 353–
360.
Epstein A.H. and Hill J.H. 1999. Status of rose rosette disease as a
biological control for multiflora rose. Plant Dis. 83, 92–101.
Epstein A.H., Hill J.H. and Nutter F.W. 1997. Augmentation of rose
rosette disease for biocontrol of multiflora rose (Rosa multiflora). Weed
Sci. 45, 172–178.
Epstein A.H., Hill J.H. and Obrycki J.J. 1993. Rose Rosette Disease.
University extension PM-1532. Iowa State University, Ames, IA.
Frazier N.W. 1968. Transmission of strawberry crinkle virus by the
dark strawberry aphid Chaetosiphon jacobi. Phytopathology 58,
165–172.
211
Harwood C. 1994. Rose industry concerns regarding rose rosette disease.
In: Epstein A.H., Hill J.H., eds. Proceedings of the International
Symposium: Rose Rosette and other Eriophyid Mite-transmitted Plant Disease
Agents of Uncertain Etiology (Ames, IA, USA, 19–21 May 1994). Iowa
State University, Ames, IA, 79–80.
Kassar A. and Amrine J.W. 1990. Rearing and development of Phyllocoptes
fructiphilus (Acari: Eriophyidae). Ent. News 101, 276–282.
Klimstra W.D. 1956. Problems in the use of multiflora rose. Ill. Acad. Sci.
Trans. 58, 66–72.
Krczal H. 1979. Transmission of the strawberry mild yellow edge and
strawberry crinkle virus by the strawberry aphid Chaetosiphon fragaefolii.
Acta Hortic. 95, 23–30.
Lewis L. 1941. Prairie Farmer – its beginnings. Prairie Farmer 113, 5–17.
de Lillo E. 2001. A modified method for eriophyoid mite extraction
(Acari, Eriophyoidea). Int. J. Acarol. 26, 67–70.
Loux M.M., Underwood J.F., Amrine J.W., Bryan W.B. and Chandran
R. 2005. Multiflora rose control. Ohio State Univ. Bull. 857, 1–16.
Maramorosch K. 1963. Arthropod transmission of plant viruses. Annu.
Rev. Ent. 8, 369–414.
Rehder A. 1936. On the history of the introduction of woody plants into
North America. Natl Hortic. Mag. 15, 245–257.
Rohozinski J., Epstein A.H. and Hill J.H. 2001. Probable mechanical
transmission of a virus-like agent from rose rosette disease-infected
multiflora rose to Nicotiana species. Ann. Appl. Biol. 138, 181–186.
Steavenson H.A. 1946. Multiflora rose for farm hedges. J. Wildl. Manage.
10, 227–234.
Steavenson H.A., Gearhart H.E. and Curtis R.L. 1943. Living fences and
supplies of fence posts. J. Wildl. Manage. 7, 257–261.
Thomas H.E. and Scott C.E. 1953. Rosette of rose. Phytopathology 43,
218–219.
© 2006 The Authors
Journal compilation © 2006 Weed Science Society of Japan