Tydeoid mites (Acari: Tydeidae, Edbakerellidae, Iolinidae) occurring on Citrus in southern Africa

Tydeoids were collected from different southern African localities by beating citrus foliage and branches. Pronematus ubiquitus (McGregor) was the most frequently encountered, followed by Tydeus munsteri Meyer and Ryke. Some species appeared to be restricted to certain climate types, e.g. Parapronematus geminus Meyer and Rodriques being found only in coastal conditions of Mozambique and Triophtydeus immanis Kuznetzov only occurring in the arid lower Orange River Valley. Four new genera (Tetratriophtydeus, Orfareptydeus, Kakamasia, and Lourus) are proposed along with four new species (O. stepheni, K. cataracta, Pseudopronematulus augrabiensis, and L. citricolus). A key is presented and has been created to facilitate the identification of the tydeoids found and to encourage further research on the role of tydeoids in the citrus ecosystem in southern Africa.


Introduction
Mites of the superfamily Tydeoidea have been recorded on citrus foliage in many parts of the world and certain species are widespread. One such species is Tydeus spathulatus Oudemans (5 Tydeus californicus Banks, 1904?) that has been collected from citrus in California (McGregor 1956), Israel (Gerson 1968), Egypt (Rasmy et al. 1972), Spain (Garcia-Marí et al. 1985), and Portugal (Carmona 1970), and herein also from Zimbabwe. Pronematus ubiquitus (McGregor) is also widespread with records on citrus from California (McGregor 1956), Spain (Garcia-Marí et al. 1985), Italy (Castagnoli 1984;Vacante and Nucifora 1986), and Egypt (Rasmy et al. 1972). These species probably serve as prey for various predatory mites on citrus as has been found in other crop systems (Flaherty and Hoy 1971;Calis et al. 1988). Some tydeids are themselves predators of phytophagous mites on citrus (Muma 1965;Rasmy 1969). Only one tydeid has been considered a pest of citrus, namely Lorryia formosa Cooreman, originally described from Morocco (Smirnoff 1957(Smirnoff , 1959Cooreman 1958;Baker 1968a;Flechtmann 1973;Garcia-Marí et al. 1985). Flechtmann (1981) referred to this species, which is also widespread and common on citrus in Florida (Aguilar and Childers 2000), northeastern Mexico (Badii et al. 2001) and Spain (Garcia-Marí et al. 1985), as being predaceous, microphagous, saprophagous, and phytophagous. In Mexico, L. formosa was found to be more suited to orange leaves than to grapefruit leaves and was considered to be only a casual visitor on the latter species (Badii et al. 2001). This mite is also known to feed on sooty mould or honey-dew (Smirnoff 1957;Mendel and Gerson 1982), which may apply to other tydeids too (English-Loeb et al. 1999).
The phytoseiid mite fauna on citrus foliage in southern Africa has been fairly well studied (Grout 1994(Grout , 2001 and certain species are known to contribute to the biological control of citrus pests (Keetch 1972;Grout and Richards 1992). However, little is known about tydeoid mites on citrus in southern Africa, though they are as frequently encountered as phytoseiids. Meyer (1998) considers T. munsteri Meyer and Ryke to be a predator of Eutetranychus orientalis (Klein) in South Africa and the second author has often collected P. ubiquitus in association with Calacarus citrifolii Keifer. This study was undertaken to identify the tydeoids that are present in citrus before establishing what contribution they may be making towards the biological control of the pests, or whether any of the species are phytophagous.
Four new genera and species are proposed herein. The nomenclature of the Tydeoidea follows André and Fain (2000). Notations for dorsal body setae follow Kethley (1990) and for legs André (1981). Measurements are in micrometres (mm).

Material and methods
Most of the tydeoids were collected from citrus foliage and branches during 2002 and 2003. Some specimens were collected earlier by colleagues working in outlying areas. In all cases, mites were collected by beating branches with a length of black polyethylene irrigation tubing (25 mm diameter) over a dark blue plastic board. Usually, four or five trees were beaten for a period of about 5 min. No attempt was made to compare population densities but only to acquire enough tydeoids to serve as a representative sample. The tydeoids were transferred from the board to 75% ethyl alcohol by using a very fine paintbrush. Permanent mounts were made using either Hoyer's solution (Krantz 1978) or PVA. The first author identified all the mites. Length of body includes the gnathosoma, width was measured at the level of setae c 1-2 , setae from tip to base, and legs from articulation facet between coxa and trochanter to tip of tarsal claws.
The type and voucher material are deposited in the National Collection of Arachnida (NCA), Plant Protection Research Institute, Pretoria, South Africa.

Distribution
Only two samples were obtained from the Western and Eastern Cape provinces so it is likely that other species may occur on citrus there. The records (Table I) show that Pronematus ubiquitus is the most widespread tydeoid on citrus foliage in southern Africa and that Tydeus munsteri is the next. Parapronematus geminus Meyer and Rodriques was well distributed amongst the low-altitude localities of Mozambique with quasi-coastal conditions. Triophtydeus immanis was found only in the arid climate of the lower Orange River Valley. The remaining species were each found only at a few localities, but in some cases these localities were more than 1000 km apart. This was the case for T. grabouwi and Perafrotydeus meyerae. It is therefore likely that the latter two species will be found on citrus anywhere in the region.

Diagnosis
Adults. This genus is closely related to Pretriophtydeus André, differing from the latter in the chaetotaxy of tibiae I-IV, namely 4-2-2-2 instead of 5-3-2-2, and in having five pairs of genital and aggenital setae instead of six pairs of genital and aggenital setae.

Etymology
The prefix Tetra refers to the main distinguishing character of the genus, namely the four setae on tibia I.
The features of this monotypic genus also define the species.

Diagnosis
Adults. This genus can be recognized by the prodorsum bearing four pairs of setae, including sci trichobothria, opisthosoma with 10 pairs of setae; female with six pairs of genital setae and four pairs of aggenital setae; palp chaetotaxy 6(v)-2-2; tibia I with a solenidion, genua II, III and IV without setae, and trochanter II with a seta, femur IV not divided.

Diagnosis
Adults. This species can be recognized by the following combination of characters: dorsum incompletely reticulated except for median area of prodorsum, which is evenly reticulated, dorsal setae short and serrated except for setae sci, which are twice as long as all other dorsal setae and smooth.

Etymology
The name of this genus is an anagram of the Afro of Afrotydeus and Per of Perafrotydeus.

Diagnosis
Adults. The features of this species are the same as those characterizing the genus. The leg chaetotaxy and epimeral formula are constant in all known stages (larva not seen) and appear to be a case of neoteny.
Protonymph can be distinguished by the presence of only two pairs of aggenital setae, one pair of progenital pores, and one pair of anal setae. Paraproctal suckers small but well developed. Coxa with coxal organ.

Etymology
This species is named after the collector of many of the specimens in this study, P. R. Stephen.

Diagnosis
Adults. All dorsal setae, except for ve and sci, are lanceolate with finely serrated edges. Setae sci are long and slightly serrate, and although not mentioned in the description, setae ve may also be serrate, but are as slender as sci. Rest of features as for genus.

Diagnosis
Adults. This species can be defined by the following combination of characters: dorsum striated with a small reticulate area anteriorly on prodorsum and reticulated elements distributed randomly on opisthosoma, dorsal setae lanceolate, faintly setose and curved, striae with semi-lunar lobes and leg tarsi with empodial claws.

Brachytydeus sp.
A tritonymph of a Brachytydeus sp. was also found on citrus at Augrabies, Kakamas, Northern Cape Province together with Pronematus ubiquitus, Triophtydeus immanis, Kakamasia cataracta, and Tydeus grabouwi. It is closely related to the tritonymph of B. monticola but differs in that the striae are more prominent and smooth and the small, reticulated areas on the dorsum are absent. This single specimen does not justify a description.

Diagnosis
Adults. For an unknown reason Baker (1970) placed this species in his subgenus T. (Afrotydeus), with four pairs of genital setae, in spite of the fact that Meyer and Ryke (1959) clearly stated that T. munsteri has the same number of genital setae as T. grabouwi Meyer and Ryke, namely six. Perhaps the type specimen he studied was a tritonymph. This species is characterized by the following combination of characters: all dorsal setae, except for vi and ve, short and spatulate and pointed distally, sce also appear pointed distally in some specimens. Setae sci longest and slightly spatulate distally.

Remarks
The South African specimens resemble the redescription of T. spathulatus (André 2005) in all respects. Tydeus californicus (Banks) may be a junior synonym of T. spathulatus because it corresponds to the redescription of the latter in all respects; five pairs of lanceolate/clavate dorsal setae posteriorly, striae arrangement of genital area and blunt setae on genua III and IV. This is a new record for South Africa. Figures 20 and 21 show how the setae of T. grabouwi Meyer and Ryke, the closely related South Africa species, differ from that of T. spathulatus.

Material examined
South Africa: Northern Cape Province: one male from Citrus sp., Augrabies near Kakamas,

Diagnosis
Adults. This species can be recognized by having four pairs of lanceolate/clavate setae, e 2 (28) (Figure 21), f 1-2 , and h 1 , the latter located caudally on opisthosoma. Seta d 1 (18) is depicted in Figure 20. Setae sci are simple and not expanded distally as in T. africanus Baker.

Diagnosis
Adults. This new genus is closely related to Pronematulus Baker, Pronematus Canestrini, and Homeopronematus André. It differs from them all in that femur IV is divided (however, femur IV may be divided in some species of Pronematulus), and in the setal formula of the tarsi, namely 8(v)-6(v)-6-5, as opposed to 8(v)-7(v)-6-6 in Pronematulus, 8(v)-6(v)-5-5 in Pronematus, and 8(v)-6(v)-6-6 in Homeopronematus. It further differs from Pronematus and Homeopronematus in that genu II has three setae instead of two. Femur IV of both Pronematus and Homeopronematus bears two setae, instead of one seta as in Kakamasia.

Etymology
The genus name refers to the town close to the locality where this species was collected and its gender is feminine.

Diagnosis
The diagnostic characters of this species are the same as those defining the genus. Venter ( Figure 23): epimeral formula 3-1-4-2. Genital area with four pairs of aggenital setae. Eugenital and genital setae absent but three pairs of anal setae (ps) present, only ps 3 situated ventrally. Anal opening extends posterior to ps 3 to form a lobe. Ventral cupule ih close to setae ps 2 .

Etymology
The species name cataracta is derived from the Greek word kataraktes (5 waterfall), which refers to the Augrabies Waterfall close to the collection site.

Genus Pseudopronematulus Fan and Li
Pseudopronematulus Fan and Li 1992, p 396.

Diagnosis
Adults. This genus can be defined as follows: prodorsum with three pairs of setae and a pair of trichobothria (sci), setae ve posteromedially to sci; opisthosoma with 10 pairs of setae; setal formula of genua 3-3-2-1; tarsus I without apotele and femur IV divided.

Diagnosis
Adults. This species can be recognized by setae vi (16) which are shorter than half the length of sci (47), ve (25) almost half its length, and sce (38) longer than half its length.

Remarks
This species is closely related to the type species, P. acus, which differs from P. augrabiensis in that vi, ve, and sce are shorter than half the length of sci while only setae vi are shorter than half the length of sci in P. augrabiensis. Most opisthosomal setae of P. augrabiensis are relatively longer than those of P. acus. Apparently only setae sci are setose whereas the rest of the dorsal setae are smooth in P. acus. Unfortunately, the description of P. acus lacks enough detail, which renders a thorough comparison of these species impossible. Pseudopronematulus augrabiensis is the second species described for this genus.

Diagnosis
Adults. Parapronematus can be recognized by seta ve, which is present (minute), or absent on prodorsum, vi posteromedially to sci, tarsus I without apotele distally, presence of forked setae on femora III and IV and femur IV entire.

Diagnosis
Adults. This species can be distinguished by the following combination of characters: tarsus I as long as or longer than tibia I, two of terminal setae on tarsus I longer than segment and all terminal setae serrated along entire length; members of ventral setae half as long as distance between them and longitudinally aligned.

Remarks
The identification of the South African specimens as P. ubiquitus was based on reference material identified as such by the late Dr Smith Meyer. She studied the type specimens of P. ubiquitus in 1970 and identified the South African specimens accordingly. At present the condition of the type material renders an examination impossible (Dr Ron Ochoa, personal communication). This species appears to be the dominant species in citrus orchards in southern Africa.

Diagnosis
Adults. This new genus is closely related to Neopronematus Panou et al. but differs from the latter in that setae h 2 and ps 2 are present. The leg chaetotaxy of both genera is the same. Panou et al. (2000) based Neopronematus on two of Kuznetzov's (1972) species, namely Pronematus rapidus and P. neglectus, with the former as type species, and a new species they described from Greece. They ignored P. testatus Kuznetzov which, along with L. citricolus sp. n., differs from Neopronematus in bearing dorsal seta h 2 and anal seta ps 2 .

Etymology
The genus name is derived from the Latin word lobus meaning lobe and the Greek word oura meaning tail and refers to the posterior lobe of the opisthosoma of L. citricolus; L. testatus also have a small lobe. The gender is masculine.  Figure 30): prodorsum with four pairs of setae, sci a trichobothrium, opisthosoma with 10 pairs of setae and three cupules. All setae slightly setose. Striae very fine, longitudinal on propodosoma, transverse medially on opisthosoma. Posterior margin of opisthosoma with a distinct lobe.

Etymology
The species name is derived from the Latin words Citrea for Citrus and colus meaning dwelling in.