Re-description of seven predatory mite species of family Phytoseiidae (Acari: Mesostigmata) sourced from Florida citrus groves

Predatory mites in the family Phytoseiidae (Acari: Mesostigmata) are of great importance as biological control agents of pest mites and other arthropods. Correct identification of species is crucial to implement effective biological control of target pests. Here, we provide re-descriptions of seven phytoseiid mite species collected from citrus orchards in Florida. The several important morphological features including dorsal setae lengths, dorsal solenostomes, shape of calyx of spermatheca, chelicera dentition, measurements, and shape of macrosetae on legs currently used to discriminate phytoseiid species were missing in the original descriptions and re-descriptions of these species. Additionally, we observed the presence of a previously unnoted taxonomically important character on Proprioseius meridionalis Chant. Therefore, the re-description was essential for further diagnosis of this species. Accordingly, the validity of the presence/absence of this structure as a diagnostic character to separate species groups in the genus Proprioseius should be re-considered. Furthermore, Typhlodromalus peregrinus, a species for which a series of morphological variations are reported in previous descriptions, is re-described and illustrated from Clermont, Florida, a location very close (10 km) to its type location (Minneola), and the leaves of type host citrus. The macrosetae StIV was knobbed apically in all our specimens of T. peregrinus indicating invalidity of sharp-pointed or knobbed StIV to separate this species from a closely related species, T. aripo De Leon. These re-descriptions and species are important to utilizing authentic and promising candidates for biological control.


Introduction
Predatory mites in the family Phytoseiidae (Acari: Mesostigmata) are important biological control agents of many pest species that include spider mites, thrips, and whiteflies [1,2] may also attack the relatively recently introduced Asian citrus psyllid, Diaphorina citri Kuwayama (Hemiptera: Liviidae), responsible for vectoring the causal pathogens of huanglongbing or citrus greening disease that is currently spreading across Florida and other states [3,4]. The phytoseiid species such as Amblydromalus limonicus (Garman & McGregor), Amblyseius andersoni (Chant), A. swirskii Athias-Henriot, Neoseiulus barkeri Hughes, N. californicus (McGregor), N. cucumeris (Oudemans), N. fallacis (Garman), Transeius montdorensis (Schicha) and Phytoseiulus persimilis Athias-Henriot are currently reared on a commercial scale and utilized to control several pests all over the world [5]. Determination of the identity of native populations of predatory mites is of considerable importance for achieving effective local pest control. Indigenous populations or species can often provide greater control success in the specific region and environmental conditions to which they are adapted [6]. Phytoseiid mite fauna in the United States has been well documented with about 370 species [7][8][9][10][11]. A total of 140 species, including synonyms, have been reported from Florida [1,11,12]. Studies on phytoseiid taxonomy from Florida date back to the mid-1950s with a series of new species being described between 1955 to 2011 [1,[13][14][15][16][17]. However, most of these descriptions or re-descriptions did not include many crucial taxonomical characters which are currently used to discriminate phytoseiid species such as dorsal setae lengths, dorsal solenostomes, shape of calyx of spermatheca, chelicera dentition, and measurements and shape of macrosetae on legs. The inadequate descriptions and re-descriptions or the absence of redescriptions for some phytoseiid species have made accurate species identifications a challenge. This situation has often resulted in misidentifications and overall confusion among taxonomists.
We have been collecting predatory mites from multiple citrus groves in Florida, to find and identify species for biological control in citrus crops. The older descriptions or re-descriptions of some species lacked some important morphological features for species identification. Those did not include the necessary details for species separation, and therefore do not meet the current standards. We provide the re-description of seven phytoseiid mite species collected from multiple locations in Florida and important for biological control.

Mite species and locations of collection
Permission was sought and granted from the orchard owners prior to undertaking all field collecting for this study. Information on the species, their habitat, and the location of the collection are provided in Table 1.

Extraction of mites from samples
The ground cover and canopy leaf samples were washed in a jar filled with 250 ml of 80% ethanol. The jar was shaken for about 30 seconds to dislodge the mites. The vegetative material from the jars was then retrieved using forcep and discarded. Leaf litter samples were collected in a four-gallon plastic bag and then processed through the Berlese funnel (Collapsible Berlese funnel, BioQuip, CA, USA) and stored in 80% ethanol. The mites were kept in 60% lactic acid for 24 hours at 55˚C.
Legs Remarks. Neoseiulus marinellus description by Muma [14] was based on the material collected from leaf litter under Citrus sp. in Minneola, Florida. Many morphological details including measurements of dorsal setae were missing in the original description and subsequent re-descriptions [26,27]. Here, we provide a complementary description of this species for the first time from specimens collected from citrus leaf litter from three locations in Florida. Morphological characters of the current specimens examined in this study are in agreement with those of type materials (Fig 2). Peritreme. Long, extending beyond setae j1. Venter ( Fig 3B). Ventral setal pattern 14:JV-3:ZV. Sternal shield smooth except few anterior striations, sclerotized with three pairs of setae (ST1, ST2, ST3), two pairs of poroids (pst1 and pst2). Distance (ST1-ST3) 60, width (ST2-ST2) 60. Metasternal setae ST4 and a pair of pores (pst3) on metasternal shields. Genital shield smooth; width at level of genital setae (ST5) 58. Ventrianal shield reticulated, bearing three pairs of pre-anal setae (JV1, JV2, and ZV2), a pair of para-anal (Pa), and a post-anal seta (Pst), and with a pair of small rounded solenostomes (gv3) posteromedian to JV2; distance gv3-gv3 25. Length of ventrianal shield 115, width at level of ZV2 90, width at level of anus 75. Setae JV4, JV5, ZV1, ZV3, and six pairs of poroids on integument surrounding ventrianal shield. Setae JV5 smooth, much longer than other ventral setae, 33 in length.
Legs ( Fig 3E). Length of legs (base of coxae to base of claws) as follows: leg I 318, leg II 233, leg III 215, leg IV 313. Genua II, III, and IV each with seven setae. A short and sharp-pointed macrosetae is present on leg IV, StIV 33 in length.
Remarks. Proprioseius meridionalis described by Chant [15] came from the material collected from Psychotria bahamensis (Rubiaceae) in Homestead, Florida. In their revision of the tribe, Kampimodromini Kolodochka, Chant & McMurtry [32] divided genus Proprioseius into three species groups based on the relative length of lateral dorsal setae and presence/absence of subcylindrical, erect structure posterior to setae j6. They characterized meridionalis species group with the absence of this erect structure. However, we observed this erect structure in all specimens that we examined in our study. To ensure that the structure identified by Chant & McMurtry [32] was the one we observed in our samples, a specimen was also mounted laterally. As a result, we confirmed the presence of the erect structure. Finally, because all other morphological characters including setae measurements concurred with the original and the re-descriptions, we consider current specimens as P. meridionalis [15,31]. We believe that this structure might have been overlooked in previous descriptions, probably due to ornamentations on the dorsal shield and poor optic material used at that time. Indeed, the presence of this structure in the type material was confirmed through communication with Dr. Ronald Ochoa based on the photos (Fig 6). Accordingly, the presence/absence of this structure in all other species known in the genus should be confirmed to determine whether it can be used as a valid diagnostic character to separate species groups in the genus. Typhlodromus (Amblyseius) aerialis [30]: 88. Female (n = 5). Dorsum (Fig 7A). Dorsal setal pattern 10A:9B (r3 and R1 off shield in all specimen except, illustrated specimen where r3 folded on the dorsal shield). Dorsal shield oval with a slight waist at level of Z1, smooth. Bearing seven pairs of rounded solenostomes (gd1, gd2, gd4, gd5,  gd6, gd8, and gd9). Chelicera (Fig 7C). Fixed digit 38 (36-40) long with 14 teeth, with pilus dentilis; movable digit 41 (40-42) long with four teeth.

Characterization of predatory mites
Remarks. Amblyseius aerialis description of Muma [13] was based on the material collected from leaves of Citrus sp. in Lucerne Park, Florida. Many morphological characters including setae measurements were not included in that original description. These measurements were provided for the specimens of different populations collected from many South American countries such as  [33], and Peru [42], but illustrations are absent in most papers except some partial drawings in De Leon [33] and Muma [13]. In addition, a full set of illustrations is available in Muma & Denmark [26] and Denmark & Muma [35]. However, the bulbous (enlarged) atrium of spermatheca is not clearly illustrated in the available literature [26,33,35,43]. Therefore, we re-described this species based on the specimens collected from its type host Citrus sp. in Fort Pierce, Vero Beach, and Clermont, all close to its type location in Lucerne Park, Florida (170, 133, and 64 km, respectively) to provide evidence for additional diagnosis including a detailed illustration with drawings of the spermatheca and other characters. Some morphological characters and measurements of the specimens that we observed are very close to those provided by others [33,35,36,40], nevertheless, there were several noticeable variations. For example, Muma [13] mentioned a smooth movable digit in the original description. However, our observation on the type materials confirmed four teeth on that cheliceral digit as mentioned in other redescriptions. The shape of the calyx of spermatheca also fits well with that of the holotype specimen (Fig 8).  Fig 9A). Dorsal setal pattern 10A:9B (r3 and R1 off shield). Dorsal shield oval with a slight waist at level of Z1, smooth. Bearing seven pairs of rounded solenostomes (gd1, gd2,  gd4, gd5, gd6, gd8 and gd9). Muscle-marks (sigilla) visible mostly on podosoma, length of
Spermatheca (Fig 9D). Calyx saccular flaring distally, 23 (22)(23) in length; atrium incorporated in the base of calyx, a very small fork is visible in some specimens, major duct long and narrow.
Legs (Fig 13E). Remarks. Tyhlodromalus peregrinus was described from Minneola, Florida, based on the material collected from orange leaves by Muma [13]. While setae measurements are absent in the original description, these measurements were provided for specimens of different populations collected from South American countries such as Argentina, Brazil, Colombia, Dominican Republic, Martinique, Guadeloupe, Guatemala, and Peru, as well as for type materials [52]. Typhlodromalus aripo De Leon [53] described from Trinidad and also reported from many South American countries shows a close affinity to T. peregrinus. Moraes & Mesa [34] pointed out that setae z2 is 50% shorter than z4 in T. aripo while it is only 20% shorter than z4 in T. peregrinus. Furthermore, they also reported macrosetae StIV as setaceous and knobbed or blunt in T. peregrinus and T. aripo, respectively [34]. The StIV was also depicted as sharppointed in T. peregrinus by Chant & McMurtry [46]. Here, we provided for the first time a complementary description of T. peregrinus based on the materials collected from Clermont, a location very close (10 km) to its type location. As stated by Moraes & Mesa [34], we observed setae z2 is 20% shorter than z4 in most or both setae at subequal length. However, all specimens examined in this study have macrosetae StIV knobbed apically. Kreiter et al. [52] examined type materials of both T. peregrinus and T. aripo and suspected T. aripo as a junior synonym of T. peregrinus. The lengths of setae z2 and z4 are listed among the apomorphic characters for some genus (e.g. Amblyseius, Transeius) in the subfamily Amblyseiinae [38]. Hence, we believe that further molecular studies or crossbreeding experiments are essential to conclude whether these species are conspecific.

Implications for biological control
We re-described seven species of predatory mites from the family Phytoseiidae, which lacked the use of important morphological features or information in the previous descriptions or redescriptions.
There is a lack of knowledge regarding the food habits and predation efficiency of a large number of phytoseiid species. Little is known about the biology, ecology, behavior, food habits, and predation of the species redescribed here, except for T. peregrinus and A. aerialis. Typhlodromalus peregrinus is a prevalent species in citrus tree canopy and ground cover plants in Florida citrus orchards [54]. According to McMurtry et al. [55], it is a generalist predator with type III feeding habit. It has the ability to feed on a wide range of food sources including mites such as Tetranychus urticae Koch and Panonychus citri (McGregor) (Acari: Tetranychidae) and can be artificially reared using pollen of plant species such as Malephora crocea (Jaquin) (Family: Aizoaceae), Quercus virginiana Miller (Family: Fagaceae), and Typha latifolia (L.) (Family: Typhaceae) [56]. Peña [57] reported that T. peregrinus also feeds on the citrus rust mite, Phyllocoptruta oleivora (Ashmead) (Acari: Eriophyidae) but prefers the broad mite, Polyphagotarsonemus latus (Banks) (Acari: Tarsenomidae) in laboratory and glasshouse trials when both species are offered. Typhlodromalus peregrinus was reported as a dominant phytoseiid species in Alabama citrus orchards, but densities were too low to provide effective control of P. citri [58,59].
The other five species considered here were mainly collected from ground cover and leaf litter. Amblyseius curiosus and P. carolinianus could be classified as type III-e generalist predators [2], as they were mainly found in leaf litter. Neoseiulus marinellus and N. planatus are also generalist predators living in confined spaces on monocotyledonous plants [2], as most of the specimens were collected from mixed ground cover including grasses from the family Poaceae. There is no information on the feeding habits and other biological parameters (development time, reproduction, or survivorship) of these species. Therefore, research is needed to assess the potential of phytoseiids found in these specific microhabitats as biological control agents. Neoseiulus baraki and N. paspalivorus which are also generalist predators found in monocotyledonous plants have been found in association with the coconut mite, Aceria guerreronis (Eriophyidae) [2]. These species were able to feed and reproduce on their prey in laboratory conditions [65,66]. Moreover, N. cucumeris a commercially available predatory mite extensively used in the past decade is a species originally found in the soil/litter habitats [2]. Therefore, it is likely that the provision of refugees and augmentative releases of N. marinellus and N. planatus may help to enhance their potential to target Eriophyidae species in citrus groves.