We sampled 526 dooryard, 18 varietal block, and 784 commercial citrus trees in Florida between May 2009 and Apr 2014 for eriophyoid mites including Diptilomiopus floridanus Craemer & Amrine, Aceria sheldoni (Ewing), Aculops pelekassi (Keifer), and Phyllocoptruta oleivora (Ashmead). In total, 1,423 D. floridanus individuals were collected from dooryard citrus trees and 1 each from the lemon cultivar ‘Bearss’ and sweet lime trees from the Florida Citrus Arboretum in Winter Haven. Diptilomiopus floridanus was collected from dooryard citrus in the following counties in Florida: Broward, Collier, Dade, Indian River, Lee, Martin, Palm Beach, Pinellas, Polk, Sarasota, and St. Lucie. The mite was not observed in Charlotte, Hardee, Hendry, Highlands, Manatee, or Okeechobee counties (Florida). Percentages of the eriophyoid mite species collected from dooryard citrus trees and varietal blocks were: A. pelekassi 3.2%, A. sheldoni 9.1%, D. floridanus 17.3%, and P. oleivora 84.2%. The 6 counties with the highest percentages of D. floridanus on dooryard trees were: Collier 36%, Broward 25%, Indian River 25%, Palm Beach 25%, Martin 23%, and St. Lucie 22%. In commercial citrus, A. sheldoni was collected less than 1.0%, A. pelekassi 4.0%, and P. oleivora 75.5%, of the time. Diptilomiopus floridanus was not collected in commercial citrus orchards during this survey. Diptilomiopus floridanus had significantly greater infestation rates on lime and lemon trees compared with grapefruit, tangerine, tangelo, sweet orange, and pummelo trees. However, infestation rates on lime and lemon trees were not significantly different from those on sour orange trees.
Prior to 2008, 3 species of eriophyoid mites were known to occur in commercial citrus in Florida: the citrus rust mite, Phyllocoptruta oleivora (Ashmead), the pink citrus rust mite, Aculops pelekassi (Keifer), and the citrus bud mite, Aceria sheldoni Ewing. All 3 species are in the Eriophyidae, 1 of 3 families within the Eriophyoidea (Lindquist & Amrine 1996). The other 2 families are the Phytoptidae and Diptilomiopidae. The Eriophyoidea is a large and diverse acarine group that includes bud, blister, gall, and rust mites (Baker et al. 1996). Most eriophyoid species are not considered to be economic pests, although they all feed on plant tissues (Baker et al. 1996).
The citrus rust mite has been recognized as a pest on Florida citrus since prior to 1879 (Ashmead 1880). Over 80 years later, A. pelekassi was found in Florida citrus orchards (Denmark 1962; Burditt et al. 1963). Researchers actively looked for A. pelekassi during the early 1970s without success (Childers 1994). Muma (1975) later stated that P. oleivora was the only eriophyoid mite of importance on Florida citrus. Aculops pelekassi was found again in sweet orange and Murcott, a putative hybrid of Citrus reticulata Blanco and C. sinensis (L) Osbeck (Hodgson 1967), blocks in Collier, DeSoto, Lake, and Polk counties during 1989 and 1990, and its presence resulted in unexpected economic losses for many growers (Childers 1994). Both species of rust mite, A. pelekassi and P. oleivora, are capable of causing serious injuries to developing citrus fruits and leaves (Childers & Achor 1999). However, A. pelekassi has greater potential destructiveness due to its rapid population increase during Apr to May while citrus fruits are small. Subsequent feeding injury to the fruit resulted in high levels of russeting and subsequent retarded fruit growth (Childers & Achor 1999).
The citrus bud mite was first reported on Florida citrus by Attiah (1959). The mite is not commonly found on fruit, leaf, or twig surfaces, but rather in sheltered places, including under the calyx of fruit, under bud scales, in petiole bases next to buds, in developing blossoms, or in axillary buds (Boyce & Korsmeier 1941; Jeppson et al. 1975). Aceria sheldoni currently is not considered an economic problem for Florida citrus growers (Childers & Achor 1999). However, it is a pest of lemons in California where increases in flower and young fruit abscission can occur in the axillary buds of infested lemons (Walker et al. 1992; Phillips & Walker 1997).
During a routine survey of citrus trees on 28 Sep 2008 in the Hollywood, Florida, area, pest survey specialists with the United States Department of Agriculture (USDA), Animal and Plant Health Inspection Service (APHIS), Bureau of Cooperative Agricultural Pest Survey (CAPS), discovered an unusual looking mite on sour orange (Citrus aurantium L.; Rutaceae) in John Williams Park (personal communication, K. M. Griffiths and M. E. Meadows, USDA, APHIS, CAPS). The mite was identified as Diptilomiopus assamica Keifer (Welbourn 2008). This species was originally described from India and later reported from northern Queensland (Australia) on citrus (Keifer 1959; Knihinicki & Boczek 2002).
Questions were raised about markings on the female genital cover flap of this new species (personal communication, J. W. Amrine, West Virginia University, and C. Craemer, Biosystematics, Agricultural Research Center, Pretoria, South Africa) versus the original description of D. assamica by Keifer (1959). Diagrams of D. assamica depicted a genital cover flap lacking in markings (Keifer 1959), and this lack of markings was later confirmed by Keifer and Knorr (1978). Additional taxonomic characters differed from the original description by Keifer (1959), and a new species, D. floridanus Craemer & Amrine was described (Craemer et al., in press).
Concern about the distribution and abundance of this new rust mite species and its potential impact on commercial citrus in Florida prompted survey efforts from the original collection site outward. In addition, there are no published records of extensive surveys of the statewide mite fauna on dooryard citrus in Florida.
Materials and Methods
During 2009 and 2010, sampling for Diptilomiopus and other eriophyoid mites was restricted to the greater Clewiston, Davie, Ft. Lauderdale, Hollywood, Homestead, Loxahatchee, and Plantation areas with multiple collections taken within John Williams Park in Hollywood. The 2 closest commercial orange and grapefruit orchards to the greater Ft. Lauderdale area were in the Southern Gardens Groves, Clewiston, a former citrus grove in Loxahatchee, and commercial and dooryard lime trees were in the Homestead area. Beginning in 2009 and continuing through 2014, sample trees were randomly selected from lists of dooryard citrus locations compiled from multiple sources. Dooryard trees were sampled on the east coast between 2009 and 2012 and on the west coast during 2013. A few trees from both coastal areas were re-sampled during 2014. Because D. floridanus was originally found in an eastern coastal habitat, most of the dooryard citrus locations selected for sampling were within 10 to 15 km of either the east or west coasts. After a site was located from the list and sampled, additional citrus trees within a several block radius were searched. Usually 2 or 3 additional locations would be sampled before moving on to another known address having 1 or more dooryard trees. Homeowners were provided with a brief explanation of the project and permission was obtained prior to inspection in all but 6 instances.
SAMPLING METHODS
Many mite species rapidly leave a disturbed leaf or fruit during sampling. Therefore, rapid preservation of the fauna was required to accurately measure each sample. Dooryard trees that were sampled varied substantially in age and vigor. Most trees showed visual degrees of infection caused by the bacterial disease, citrus greening. In moderate to large citrus trees, 8 to 12 clusters of leaves and associated twigs were clipped from the tree with pruning shears and dropped individually into a 5 L bucket containing approximately 250 mL of 80% ethanol following the protocol of Childers and Ueckermann (2014). In smaller trees, i.e., 2 or 3 year-old trees or trees less than 2 meters in height, 4 to 6 leaf and associated twig samples were taken. Fewer than 15 trees of this smaller size were sampled during the survey. Occasionally, 2 to 4 fruits from a tree were included with the leaf samples or processed separately.
All plant material sampled, including leaves, twigs, and fruit, was vigorously agitated in the alcohol solution and then removed and discarded. The alcohol wash per sample was transferred into a labeled glass jar for processing. Samples from individual trees within each dooryard location were processed separately and placed in individual labeled jars. Random samples of leaves from each tree were collected from the inner and outer canopy areas as well as the middle, high, and low areas inside and outside of the tree canopy. All mites were removed from each sample and multiple numbers of eriophyoid mites were collectively slide-mounted on 1 or more slides in modified Berlese medium (Amrine & Manson 1996).
Sampling of commercial orchard sites in Florida between 2009 and 2012 consisted of 20, 50, or 100 mature spring flush samples, or 20 fruit samples. At Southern Gardens Citrus, 10 leaves were collected from each of 5 randomly selected trees per replicate in each block. Four fruit were collected separately and at random from the same block per replicate. Both leaf and fruit samples were replicated 3 times in each of the 9 blocks of trees at this location. Additional commercial orchards were sampled and varied from 4 to 10 fruit or leaves per tree and replicated 5 times. Numbers of leaves or fruit per tree or numbers of trees sampled during 2012 by independent scouts were not recorded. All samples were immediately washed in 80% ethanol as stated above and placed in individual labeled glass jars.
STATISTICAL ANALYSES
Population dynamics of D. floridanus was investigated, including the potential for antixenosis, the inability of a plant to serve as a host, and its effects on distribution. Causes of antixenosis may include lack of a chemical or visual cue recognized by the pest (Kogan & Ortman 1978), and degrees of antixenosis may vary among species of citrus.
To evaluate potential antixenosis, trees were coded 1 for presence of D. floridanus and 0 for absence (Excel macro program PooledIn fRate, Biggerstaff 2006) and percent infestation was calculated (95% confidence intervals). If the 95% confidence intervals did not overlap, then our determination was that a significant difference in antixenosis occurred between the citrus species. Generalized linear mixed models with means separation by Tukey honest significant difference (HSD) tests (Proc Glimmix: https://www.sas.com/content/dam/SAS/en_us/doc/other1/editorial-guidelines.pdf ; SAS Institute 2016) were used to determine the degree of antixenosis expressed by each citrus species (lemon, lime, sour orange, tangelo, and sweet orange). Mite counts were square root transformed prior to analysis. Mite counts on different cultivars were log transformed.
Results
Adults and immatures of D. floridanus are white to light brown in color, about 160 μm in length with an arched body, and dorsoventrally thicker than the other 3 eriophyoid species found on Florida citrus. The chelicerae of D. floridanus are exposed and well developed. This combined with an extended and downward slopping prodorsal shield gives the mite the appearance of having a defined head (Fig. 1). These characters are readily apparent when looking in a dish containing the mites in alcohol with 15 to 20x magnification.
A total of 526 dooryard, 18 varietal block, and 784 commercial citrus trees representing at least 18 species in the genus Citrus which were sampled including the following genera: Citrus Fortunella (kumquats) and Poncirus (Trifoliate orange) were sampled in 17 central and southern Florida counties between May 2009 and Apr 2014 (Table 1). One or more D. floridanus adults and immatures were collected in 11 of these counties (Table 2). Eggs of this species were not found. The 6 counties with the highest infestation rates of D. floridanus were Collier 36%, Broward 25%, Indian River 25%, Palm Beach 25%, Martin 23%, and St. Lucie 22%. Collier is the only west coast county among the 6, with the remaining 5 on the east coast. Diptilomiopus floridanus was not observed in Charlotte, Hardee, Hendry, Highlands, Manatee, or Okeechobee counties. A total of 1,423 D. floridanus were collected from dooryard citrus trees and 1 each from the lemon cultivar ‘Bearss’ and a sweet lime tree at the Florida Citrus Arboretum in Winter Haven (Florida). This was the 1 instance during the study where D. floridanus was collected from known pesticide-treated trees and from an inland county. The spray records for the Fruit and Spice Park (Homestead, Florida) were not available.
Diptilomiopus floridanus was the second most abundant eriophyoid mite collected from dooryard and varietal citrus blocks with 91 of 526 trees infested. Infestation rates of the 4 eriophyoid mites on these trees were as follows: A. pelekassi 3.2%, A. sheldoni 9.1%, D. floridanus 17.3%, and P. oleivora 84.2% (Table 1). Thirty-eight percent of lemon, lime, and sour orange trees were infested with 1 or more Diptilomiopus on the east coast compared to 27% of the same varieties on the west coast. Diptilomiopus floridanus was most commonly found along coastal areas within 10 to 15 km of the ocean on dooryard citrus trees. A total of 238 D. floridanus and 889 P. oleivora were collected from 1 lemon tree in the community of Redington Shores, Pinellas county (Florida) on 19 Jun 2013. The tree was less than 1 km from the ocean and the northern most coastal collection site for this new mite species on either coast. Twenty-five of the 91 trees infested with D. floridanus had higher numbers of this species versus P. oleivora and included 9 cultivar ‘Tahiti’ lime, 5 Key lime, 1 sweet lime, 4 lemon, 3 sour orange, and 3 sweet orange trees.
Fig. 1.
Photographs taken with a scanning electron microscope of the new species of Diptilomiopus floridanus Craemer & Amrine on Florida citrus. (A) Dorsal view of prodorsum, legs, and well developed chelicerae. (B) Dorsal view of the mite. (C) Lateral view of the mite. (D) Dorso—lateral view of the mite with extended, downward gnathosome.
Table 1.
Eriophyoid mites on citrus and related Rutaceae collected from dooryard and varietal collection trees in Florida (2009–2014).
Table 2.
Number and infestation rate of Diptilomiopus floridanus Craemer & Amrine collected from dooryard citrus trees and varietal blocks in Florida (2009–2014).
Table 3.
Comparative infestation rates of Diptilomiopus floridanus Craemer & Amrine on the major citrus species sampled in Florida between 2009 and 2014.
Lime and lemon trees had significantly greater infestation rates of D floridanus compared with infestation on grapefruit, tangerine, tangelo, sweet orange, and pummelo trees. However, infestation rates were not significantly different from those on sour orange (Table 3). The infestation rate of the mite on sour orange was not significantly different from infestation on pummelo trees. In situations where species of citrus trees occurred together with 1 or more lime, lemon, or sour orange trees, then they were removed from a second analysis and different results were obtained (Table 4). However, the analysis of infestation rates of lime, lemon, and sour orange remained unchanged from the other citrus species.
The numbers of D. floridanus occurring were compared between the major citrus species (Table 5). Sufficient degrees of freedom were available to compare numbers of mites only on sweet orange, lemon, lime, tangelo, and sour orange trees, and no significant differences were found with this analysis.
Table 4.
Comparative infestation rates of Diptilomiopus floridanus Craemer & Amrine where positive trees were removed from the analysis if they were from multi-tree dooryard locations having lemon, lime, or sour orange present.
Diptilomiopus floridanus was not found among the 784 commercial citrus trees sampled in Dade, Hardee, Hendry, Highlands, Indian River, Okeechobee, Palm Beach, Polk, or St. Lucie counties between 2009 and 2012 (Table 6). A total of 120 commercial citrus orchards were surveyed for eriophyoid mites between 1986 and 1999 and D. floridanus was not found among the 64,887 mites identified (Childers & Achor 1999). That survey included at least 26 commercial citrus orchard sites within 50 km or less of coastal areas in Florida. However, dooryard citrus trees were not sampled in that survey.
Discussion
Insecticide fogging for mosquito control in John Williams Park (Hollywood, Florida) occurred prior to some of the sampling dates. The insecticides used and dates of application were not available. Mosquito problems were quite evident in the park. However, they were not a problem during 1 or 2 sampling dates. We suspect presence or complete absence of motile stages of this mite at this location were due to the mosquito fogging.
Two D. floridanus were collected on lime and lemon trees at the Citrus Arboretum in Winter Haven (Florida) on 11 Jul 2012. The following pesticides were applied to those trees during 2012: 30 Mar — Danitol® (Valent USA Corp., Walnut Creek, California) + Citru-film® (Helena Holding Co., Wilmington, Delaware); 10 May — Kocide® (E. I. DuPont de Nemours & Co., Inc., Wilmington, Delaware); and 11 Jun — Danitol® + Citru-film®. The ability of this mite to survive after pesticide treatments raises concerns about pesticide resistance and the potential of the mite becoming established in commercial citrus orchards in Florida. This could be especially important in lime and lemon cultivars if and when numbers of pesticide applications can be reduced for controlling the Asian citrus psyllid (Rogers & Dewdney 2015). Multiple insecticide and acaricide applications are currently applied to control the insect vector of citrus greening, as well as to suppress eriophyoid mite pests in commercial citrus orchards. Some of the dooryard tree locations appeared to have been sprayed with pesticides based on the condition of the trees and presence of clean fruit, and foliage. Other dooryard trees had varying numbers of predacious mites, especially Phytoseiidae.
Table 5.
Results of statistical analysis testing differences in Diptilomiopus floridanus Craemer & Amrine abundance on different citrus cultivars using the model Host = log(abundance). Zero values were excluded in the analysis.
Whether D. floridanus is a long-established or a more recent introduction into Florida citrus remains unknown. There are no published area-wide survey reports on dooryard citrus. Citrus preference (lemon, lime, and sour orange) and coastal distribution may have played a role in delaying detection. Additionally, more extensive surveys of dooryard and commercial citrus in neighboring countries in the Caribbean should be conducted to determine the extent of this mite species distribution. Species within the genus Diptilomiopus have been reported as lower leaf surface feeders occurring in low numbers with no apparent injury to their host plants (Chen et al. 2004; Huang 2005; Huang & Chen 2005; Song et al. 2008). However, extensive samplings of these species are lacking.
Chakrabarti and Mondal (1983) reported that Diptilomiopus bengalensis Chakrabarti and Mondal caused yellowing to browning of leaves, and normal leaf growth was affected with heavy infestations on Gardenia jasminoides J. Ellis (Rubiaciae). Furthermore, Mohanasundarum (1981) reported that Diptilomiopus artocarpae Mohanasundarum caused drying of the twigs of Artocarpus integer (Thunb.) Merr. (Moraceae). Both reports are based on anecdotal observations. Diptilomiopus species have not been reported as economic pests on citrus or other crops. Essentially, all research conducted so far has dealt with descriptions of species. Information on feeding injury, host susceptibility, and effects of population density are lacking. Future studies should focus on developing culturing methods for this species to determine potential feeding injury and resultant economic impact on difference citrus species.
Acknowledgments
This survey would not have been possible without the tremendous assistance of Karolyn M. Griffiths, Florida Area East Director, USDA, APHIS, Plant Protection and Quarantine (PPQ); Andrew Derksen, Florida Cooperative Agriculture Pest Survey, Miami; David Hall, USDA, Agricultural Research Service (ARS), Ft. Pierce; J. W. Amrine Jr., Division of Plant and Soil Sciences, College of Agriculture and Forestry, West Virginia University; Theresa M. Badurek, Pinellas County Extension; and Rita Duncan and Holly Glen, University of Florida, Tropical Research & Education Center, Homestead. Appreciation also is extended to Adrian Hunsberger, Miami—Dade County Extension; Thomas Pospical, Mosaic Company; Tim Gast, Southern Gardens Citrus; Mark DuBois, Grove Callery—Judge LP; Ray Wambles, Wheeler Farms, Lake Wales; Renee Richards, Hollywood Parks and Recreation Department; and Michael Kesinger, Florida Department of Agriculture & Consumer Services. Appreciation is extended to Joseph G. Morse, David G. Hall, Beverly Gerdeman, and Lukasz Stelinski for reviewing this paper.
