Entomopathogenic Potential ofVerticilliumandAcremoniumSpecies (Deuteromycotina: Hyphomycetes)

Journal of Invertebrate Pathology 73, 309–314 (1999) Article ID jipa.1998.4841, available online at http://www.idealibrary.com on

Entomopathogenic Potential of Verticillium and Acremonium Species (Deuteromycotina: Hyphomycetes) Tove Steenberg*,† and Richard A. Humber‡ *Danish Pest Infestation Laboratory, Skovbrynet 14, DK-2800 Lyngby, Denmark; †Section of Zoology, The Royal Veterinary and Agricultural University, Bulowsvej 13, DK-1870 Frederiksberg C., Denmark; and ‡USDA-ARS Collection of Entomopathogenic Fungal Cultures, Plant Protection Research Unit, US Plant, Soil and Nutrition Laboratory, Tower Road, Ithaca, New York 14853-2901 Received June 19, 1998; accepted December 18, 1998

Hyphomycetes with conidia formed in slimy heads from hyaline mycelium were isolated from a range of insect hosts. Isolation on artificial medium and microscopic examination revealed that these fungi in many cases were not Verticillium lecanii despite a superficial resemblance to this common entomopathogen. The fungi were identified as Verticillium fusisporum, Verticillium psalliotae, Verticillium lamellicola, and species of Acremonium. Isolates of these fungi were bioassayed against the sweet-potato whitefly (Bemisia tabaci) and against the housefly (Musca domestica) to examine their entomopathogenicity. A test was also conducted with a coleopteran (lesser mealworm, Alphitobius diaperinus) to further evaluate the host range for some of the fungi. V. lamellicola was not pathogenic to the two species of insects treated, while varying levels of pathogenicity were shown for the other species. In general, V. lecanii was the most pathogenic species. Immature whiteflies appeared to be more susceptible to fungal infection than adult houseflies, and the host range for several of the fungi also included lesser mealworm. r 1999 Academic Press Key Words: Acremonium; Verticillium fusisporum; V. lamellicola; V. lecanii; V. psalliotae; Alphitobius diaperinus; Bemisia tabaci; Musca domestica.

INTRODUCTION

Verticillium lecanii (Zimmermann) Vie´gas is a wellknown pathogen of arthropods. The host range of the species is wide and includes homopteran insects as well as a range of other arthropod groups. V. lecanii is considered to be a species complex that includes isolates with very variable morphological and biochemical features (Gams, 1971; Jun et al., 1991). Conidia of V. lecanii are short ovoid to cylindrical, but they are never fusiform, falcate, or curved. Among the nine entomogenous species of Verticillium listed by Gams (1971), some form conidia with more or less pointed tips, e.g., V. fusisporum W. Gams. Occasionally some of these spe-

cies are isolated from arthropod cadavers (Balazy et al., 1987; Kalsbeek et al., 1995). However, V. lecanii seems to be by far the most common Verticillium species isolated from arthropods. Information on the pathogenicity against insects of the other species of Verticillium is scarce. However, Ekbom and Åhman (1980) studied the effect of an isolate of V. fusisporum against different homopteran glasshouse pests and found that the isolate was highly pathogenic to insects. The genus Acremonium Link also is reported to contain species with pathogenicity to insects and other arthropods. Acremonium species have been isolated from arthropods (Petch, 1931, 1938; Gams, 1971; Balazy et al., 1987) but there are very few reports verifying the pathogenicity of these fungi for insects (Riba, 1983). Acremonium was monographed with Verticillium and a range of other related fungi by Gams (1971). Fungi which on first inspection resembled V. lecanii were isolated from diseased insects on several occasions. After in vitro isolation and microscopic examination it became clear that these isolates did not belong to V. lecanii, despite the morphological variation found within this species complex. In several cases the proportion of ‘non-lecanii’ isolates among the total number of fungi isolated from the diseased host population was surprisingly high. Therefore, this study was initiated in order to identify the fungi and to test their pathogenicity against insects. MATERIALS AND METHODS

Origin of Fungal Isolates Diseased insects were collected from epizootics in the sweet-potato whitefly Bemisia tabaci on poinsettia in a glasshouse and in laboratory rearing cages of B. tabaci on tobacco, Myzus persicae on pepper, and Thrips tabaci on beans. All isolates from a particular host population were made from separate cadavers attached to the same leaf. Isolates from B. tabaci were from adults; isolates from M. persicae were from adults and nymphs;

309

0022-2011/99 $30.00 Copyright r 1999 by Academic Press All rights of reproduction in any form reserved.

310

STEENBERG AND HUMBER

and isolates from T. tabaci were from larvae. Live adult houseflies (Musca domestica) were collected with a sweep net from stables at two different farms and incubated in cardboard cages for a week. Occasionally, a few specimens died with signs of fungal infection by Verticillium species, and these fungi were isolated. From each farm there was one isolate that could not readily be identified as one of the common entomopathogenic species. In all instances, isolates were cultured on 2% Sabouraud dextrose agar (SDA). A total of 33 isolates were obtained (Table 1); all isolates were from host cadavers covered by fluffy, hyaline mycelium. Preliminary inspection in a stereo microscope revealed that all isolates produced conidia in slimy heads on phialides that were placed singly, in pairs, or in whorls on the mycelium. Identification of Fungi For macromorphological studies, isolates were inoculated in the center of petri dishes with 2% SDA and incubated in constant light at 20°C. Culture characteristics assessed after 3 weeks were colony growth rate, color and height of colony, and pigmentation of the reverse. The micromorphology was studied in mycelium samples taken from 3-week-old cultures. The arrangement of the phialides on the mycelium and the shape and size of the conidia were noted for all isolates. In addition, for eight isolates that were clearly not V. lecanii, the dimensions of 50 conidia and 25 phialides were measured. For isolates that produced a mixture of spores, e.g., small ellipsoidal conidia and curved conidia, 50 conidia of each type were measured. The majority of the isolates were subsequently deposited in the USDA-ARS Collection of Entomopathogenic Fungal Cultures (ARSEF, Ithaca, NY) (Table 1). Bioassay against Bemisia tabaci All fungal isolates originating from glasshouse pests were tested in a preliminary bioassay to assess their pathogenicity. The method used was a slightly modified version of the method of Hall (1984). Spores were harvested from 3-week-old cultures with 0.05% Triton X-100 and adjusted to 106 conidia/ml. For each isolate, nine excised leaf discs (10 mm diameter) of tobacco (cv. Samsun) each with 10–20 second- to third-instar larvae of B. tabaci were immersed in a petri dish in 20 ml spore suspension for 10 s. Following treatment, excess moisture was removed by drying the leaf discs on sterile filter paper, and the leaf discs were then transferred to square incubation chambers with 25 compartments containing a layer of 2% water agar (see Hall, 1984). After 7 days at 20°C in constant light, infection was verified by isolating fungus from cadavers supporting sporulation and comparing the morphological characteristics with those of the isolate used. Fourteen isolates that had been found to be pathogenic in the preliminary test were then bioassayed against B. ta-

baci as described above. Controls were treated with sterile 0.05% Triton X-100. All treatments were replicated three times. The number of individuals per leaf disc was counted after mounting the discs in the incubation chambers. After 7 days, dead individuals were recorded. This included individuals with mycelium growing from the cadaver. Bioassay against Musca domestica Eight fungal isolates from glasshouse pests and three isolates from M. domestica were tested against adult M. domestica. Female flies (⬃4 days old) of strain DPIL 772a (field-collected in a Danish pig stable in 1989 and since reared repeatedly in the laboratory) were anesthetized with carbon dioxide, and groups of 10 flies where immersed in spore suspensions of 107 conidia/ml for 10 s. Spore suspensions were prepared with 0.05% Triton X-100; controls were treated with sterile 0.05% Triton X-100. After drying off excessive moisture on filter paper, the flies were placed in cardboard cages (9 cm diameter) supplied with 2% sucrose and dried milk powder and incubated for 24 h at 23°C in plastic bags lined with damp tissue paper. The cages where then removed from the bags and placed at 23°C, 40–60% RH, with a photoperiod of 12:12 (L:D). Each treatment was replicated five times. Dead flies were recorded daily for 21 days and were transferred to moist chambers to induce fungus sporulation. Infectivity against Alphitobius diaperinus Eleven isolates listed in Table 1 were tested for pathogenicity against larvae of lesser mealworm (Alphitobius diaperinus Panz.). These larvae were maintained as a laboratory colony and for each assay, 10 late-instar larvae of A. diaperinus were placed in petri dishes with 3-week-old sporulating fungal culture for 10 min and then incubated separately at 26°C, 70% RH for 10 days in 30-ml plastic vials with ventilated lids. The larvae were provided with water-soaked cotton plugs. Control insects were placed in petri dishes with 2% SDA for 10 min. Dead insects were removed after 7 and 14 days and placed in moist chambers for 5 days to induce fungal sporulation. Statistical Analysis Percentage mortality was arcsine transformed, and means were separated using Tukey’s test (␣ ⫽ 0.05) in the General Linear Models procedure (SAS Institute, 1996). RESULTS

Identification of Fungi The 33 isolates were identified using the keys and species descriptions in Gams (1971) (see Table 1). Fifteen isolates were identified as V. lecanii; the remain-

311

ENTOMOPATHOGENIC POTENTIAL OF VERTICILLIUM AND ACREMONIUM

TABLE 1 Fungal Isolates: Origin, Identification, and ARSEF Accession Numbers Host insect Bemisia tabaci

Location Poinsettia (in rearing cage), Lyngby

Tobacco (in glasshouse), Lyngby

Myzus persicae

Green pepper (in rearing cage), Lyngby

Thrips tabaci

Beans (in rearing cage), Lyngby

Musca domestica

a

Stable, Roskilde Stable, St. Karleby Stable, St. Karleby

Strain

Fungus

BP2 BP3 BP4 BP5 BT2 BT3 BT4 BT5 BT6 BT10 BT11 BT12 BT13 BT14 BT15 MP1 MP2 MP3 MP5 MP6 MP7 MP9 MP10 MP11 TT1 TT2 TT3 TT4 TT5 TT6 625 626 615

Acremonium sp. Acremonium sp. Acremonium sp. Acremonium sp. V. lecanii V. lecanii V. lecanii V. psalliotae V. lecanii V. lamellicola V. lecanii V. lecanii V. lecanii V. lecanii V. lecanii V. fusisporum V. lecanii V. lecanii V. lecanii V. fusisporum V. fusisporum V. fusisporum V. fusisporum V. lecanii Acremonium sp. Acremonium sp. Acremonium sp. V. lecanii Acremonium sp. Acremonium sp. V. fusisporum V. fusisporum V. lecanii

ARSEF accession number 4058 a 4059

4064 4065 5582 a 4067 4068 a 4069 4070 a 4071 4072 4074 4075 4076 4077 4078 a

4079

4062 a 5580 a 4063 a 5578 a 5579 a

Denotes isolates tested against lesser mealworm (A. diaperinus).

ing isolates included 7 of V. fusisporum, 1 each of V. psalliotae and V. lamellicola, and 9 attributed to the genus Acremonium but not identified to the species level, as we were not able to find a good fit for it in Gams (1971). Macromorphological Features Most of the 33 isolates produced a high cottony, white colony on SDA. However, the V. lecanii isolates isolated from B. tabaci (BT-isolates) produced a flat, more velvet-like colony. This was also the case for V. lecanii (615) from M. domestica. The only isolate that did not form a white colony was V. psalliotae (BT5), which produced a pale yellow mycelium. The majority of non-lecanii species were characterized by a tendancy to grow in radially irregular colonies unlike the much more uniformly circular colonies of V. lecanii isolates. Acremonium isolates conversely grew as uniformly circular colonies, but differed from V. lecanii in producing areas with brown pigmentation in the reverse of the petri dishes. Some Acremonium isolates also produced brown areas in the center of the colonies with maturity.

V. fusisporum (625 and 626) from houseflies produced a brownish pigment in the reverse; this was not seen in V. fusisporum from M. persicae in which the reverse for all five isolates was without any appreciable pigmentation. V. lamellicola (BT10) likewise did not produce a pigment but was clearly distinguished from the remaining isolates by a very low growth rate. V. psalliotae also grew less abundantly than the other isolates and, furthermore, produced a red-brown pigment that quickly colored the agar in the whole dish. It, furthermore, produced a distinct, mouldy smell. Micromorphological Features The Acremonium isolates (see Table 2) were distinguished by producing long and slender phialides that in some cases were placed at right angles on the mycelium. In contrast to the Verticillium species, these phialides never occurred in pairs or in whorls. The conidia of the Acremonium isolates were rather small and subglobose to ellipsoidal. The ‘non-lecanii’ isolates of Verticillium produced a mixture of conidial forms, in which one part of the conidia was relatively small and

312

STEENBERG AND HUMBER

TABLE 2 Measurements of Conidia and Phialides for Different Species of Verticillium and Acremonium Phialides Small conidia Fungus V. psalliotae V. lamellicola V. fusisporum V. fusisporum Acremonium sp. Acremonium sp. Acremonium sp. Acremonium sp.

Large conidia

Mean Mean Mean Strain length Range width Range length BT5 BT10 MP6 MP1 TT5 TT3 BP2 BP4

4.7 2.9 3.6 4.5 2.9 2.9 3.2 3.0

3.6–5.6 2.4–4.0 3.2–4.8 3.2–4.8 2.4–4.4 2.4–3.6 2.8–4.0 2.4–4.0

2.4 1.4 1.6 1.6 1.6 1.6 1.7 1.6

2.4–2.8 0.8–1.6 1.6–2.0 NC NC 1.2–2.0 1.6–2.4 2.4–4.0

11.9 6.5 8.0 10.6

Range 8.8–15.2 4.8–8.8 6.4–10.4 7.2–15.2

Mean Mean Range Mean width Range L/W L/W length 2.4 1.3 1.8 2.0

1.6–3.2 1.2–1.6 1.6–5.4 1.6–2.4

5.0 5.0 4.5 5.2

4.3–6.5 3.5–6.7 3.0–5.5 3.7–7.6

31.3 18.8 16.6 20.8 33.7 53.7 30.0 30.6

Range 25.6–36.0 13.6–22.4 11.2–23.2 12.8–27.2 15.6–46.4 32.0–78.4 19.2–41.6 20.8–46.7

Mean basal width Range 2.1 0.9 1.5 1.6 0.9 1.6 1.4 1.4

1.6–2.4 0.8–1.2 1.2–1.6 1.6–2.0 0.8–1.2 NC 1.2–1.6 NC

Note. For each isolate, 25 phialides and 50 conidia were measured at 400⫻ magnification. For isolates which produced conidia with different shapes (e.g., small ellipsoidal and large curved conidia), 50 conidia of each group were measured. All data are given in µm. NC (not calculated) denotes that all values were identical.

ellipsoidal, and another part was large and fusiform or curved. In some cases, the conidial dimensions differed from those given in the species description; e.g., isolate MP1 was identified as V. fusisporum although the fusiform conidia were longer than stated in Gams (1971). In contrast, V. lecanii isolates produced only cylindrical or ellipsoidal conidia. Bioassay against Bemisia tabaci In the preliminary test, 29 of 30 isolates isolated from glasshouse pests were pathogenic to B. tabaci

immatures. Fungus-killed immatures did not always support fungus outgrowth 7 days after treatment but, based on the white body color, they could readily be distinguished from live individuals and from those that died from other causes. Only V. lamellicola (BT10) did not cause mortality. In the bioassay, there were highly significant differences between treatments (F ⫽ 12.64, df ⫽ 14, P ⬎ 0.001) (Table 3). All isolates caused mortalities that differed significantly from those in the control. The overall control mortality on day 7 was 6.2%.

TABLE 3 Bioassays against Bemisia tabaci Immatures and Adult Musca domestica

Fungus

Strain

Acremonium sp. Acremonium sp. Acremonium sp. Acremonium sp. V. lecanii Acremonium sp. V. lecanii V. psalliotae V. lecanii V. lecanii V. lecanii V. lecanii V. lecanii V. lecanii V. lecanii V. fusisporum V. lecanii V. lecanii * V. fusisporum V. fusisporum V. lecanii Control

BP2 BP3 BP4 TT3 TT4 TT5 BT3 BT5 BT12 BT13 BT14 BT15 MP2 MP3 MP5 MP6 MP11 VL10 625 626 615

Bemisia tabaci, % mortality Day 7 71.6 b,c,d 77.2 a,b,c 41.3 h

64.6 c,d,e,f 46.5 f,g,h 94.1 a,b 87.3 a,b 50.7 e,f,g,h 88.2 a 29.7 h 46.8 c,d,e,f,g 68.8 c,d,e 60.5 c,d,e,f,g 55.2 d,e,f,g

6.2 i

Musca domestica, % mortality Day 21

Musca domestica, % sporulation

Day 7

Day 14

27.8 a

42.6 d

66.7 b

50.0 b

25.4 a 21.8 a 15.7 a

42.4 d 50.9 b,c,d 39.2 d

64.4 b,c 70.9 b 66.7 b

47.5 b,c 34.5 b 29.4 b

22.0 a 32.0 a

42.0 c,d 82.0 a,b,c

66.0 b 98.0 a

30.0 b 88.0 a

12.0 a

16.0 d

26.0 d

22.0 d

28.0 a 18.0 a 22.0 a 32.0 a 14.0 a

92.0 30.0 d 36.0 d 84.0 a,b 16.0 d

100.0 a 44.0 b,c,d 44.0 b,c,d 96.0 a 22.0 d

94.0 a 18.0 b,c,d 24.0 b,c,d 82.0 a 0.0 d

Note. Letters within each column indicate Tukey groupings (␣ ⫽ 0.05). * Denotes an isolate from an unidentified aphid. Total % sporulation for M. domestica was measured 26 days after inoculation.

ENTOMOPATHOGENIC POTENTIAL OF VERTICILLIUM AND ACREMONIUM

Bioassays against Musca domestica The mortality caused by the 10 isolates tested against houseflies did not differ significantly at day 7 postinoculation. By day 14 and day 21 the differences between isolates were highly significant (day 14: F ⫽ 10,24, df ⫽ 11, P ⬎ 0.001; day 21: F ⫽ 23,09, df ⫽ 11, P ⬎ 0.001). Likewise, the difference in percentage sporulation between isolates was highly significant (F ⫽ 24,78, df ⫽ 11, P ⬎ 0.001). At day 21 there were no differences in the mortalities of the three isolates of V. fusisporum and the control mortality (Table 3). The other isolates were pathogenic, including 3 isolates of Acremonium, V. psalliotae, and V. fusisporum. In general, the isolates of V. lecanii were more pathogenic than those of the other species. By day 21 a group of isolates with pathogenicity intermediate to that of houseflies could be distinguished. This group included V. psalliotae (BT5) and three isolates of Acremonium (TT3, TT5, and BP2). Infectivity against Alphitobius diaperinus Patent infection of A. diaperinus larvae was obtained with the three Acremonium isolates tested, with one out of three isolates of V. fusisporum (BP2), and with the three isolates of V. lecanii tested. V. psallioae and V. lamellicola were not pathogenic to this host. DISCUSSION

Fungi Isolated from Diseased Insects Fifteen of the 33 isolates studied here are identified as V. lecanii using the information in Gams (1971). These identifications, however, must be regarded as provisional since Gams (1971) treatment of V. lecanii includes a wide range of morphological variability and this taxon is now widely agreed to be an unresolved species complex. Molecular studies of a wide range of isolates of Verticillium from insects (Humber, unpublished) indicate that the use of V. lecanii will have to be restricted to a narrow range of fungi with a distinctive set of molecular profiles coming primarily from Southeast Asia (Zimmermann described this fungus from coffee green scale collected in Bogor, Indonesia). While the great majority of collections of V. lecanii as recognized by Gams (1971) fall outside of V. lecanii in this narrowly restricted sense, the V. lecanii species complex does contain several distinctive taxa defined by their molecular profiles and traditional taxonomic characters that will be recognized as separate species once it can be determined whether they must bear names currently treated as synonyms of V. lecanii or be described as entirely new taxa (Humber, unpublished). In the meantime, the only reasonable course of action is to still use the name V. lecanii in the broad sense of Gams (1971), as has been done here. A high proportion of the fungal isolates isolated from

313

epizootics in populations of glasshouse pests proved not to be V. lecanii despite a superficial resemblance to this entomopathogen (Table 1). From M. persicae, five out of nine isolates were V. fusisporum. One group of isolates from B. tabaci were all Acremonium species (BPisolates), while another group consisted not only of V. lecanii but also contained V. psalliotae and V. lamellicola (BT-isolates). At two different farms the only hyphomycetes found to affect houseflies were V. fusisporum and V. lecanii. In contrast to the epizootics in glasshouse pests, only a few of the live-collected houseflies were killed by fungi. These results show the need for thorough examination of the micromorphology of entomopathogenic fungi in order not to overlook less well-known species. Species of Acremonium have previously been associated with mortality in insects and have occasionally been listed in studies on the natural occurrence of entomopathogenic fungi (Petch, 1931, 1938; Balazy et al., 1987; Sanchez-Pen˜a, 1990). However, Gams (1971) regarded only A. larvarum (Petch) W. Gams to be an entomopathogen. Nevertheless, the infectivity assays in this study clearly demonstrate that species of Acremonium are entomopathogens with some potential as biocontrol agents, at least against B. tabaci immatures. V. lamellicola was isolated from B. tabaci but did not cause mortality when tested against immatures of this species or against A. diaperinus. Balazy et al. (1987) likewise reported an isolate of V. lamellicola (isolated from a mite) that had no pathogenicity for bark beetle larvae. The fungus is known to be a parasite of mushroom cultures and fruit bodies of some ascomycetes and basidiomycetes; the recovery of this species on dead arthropods probably reflects merely its capability as a saprophyte. This species has straight fusiform conidia like those of V. fusisporum, but the conidial tips of V. lamellicola appear more pointed, and the ratio between the length and width is often greater. A high proportion of the conidia produced by V. lamellicola were small and ellipsoidal rather than long and narrow fusiform, and this tendency to produce a mixture of conidia with different shapes and dimensions was found in V. psalliotae as well as in the three isolates of V. fusisporum. That such variable conidial shapes were also found in single-spore isolates made from these cultures (Steenberg, unpublished data) tended to disprove the possibility that these cultures were mixed isolates. Such a phenomenon was also described for V. psalliotae by Gams (1971). In some isolates, the small and ellipsoidal conidia dominate completely, and unless an isolate produces a strong pigment, a fungus may well be misidentified as V. lecanii by virtue of the large amount of morphological variability within this species complex. V. psalliotae is mainly known as a parasite of mushrooms and rust fungi but has also been isolated from dead mites and insects (Domsch et al., 1980; Balazy et

314

STEENBERG AND HUMBER

al., 1987). The isolate BT5 matches the characterization of this species by Gams (1971), although the sickle-shaped conidia are larger than those noted by Gams. Furthermore, the isolate produces a distinct, mouldy smell, a characteristic that does not seem to be common in this species (Gams, 1971; Domsch et al., 1980). V. fusisporum was described by W. Gams (1971) from isolates from soil and dead leaves. Ekbom and Åhman (1980) isolated it from whiteflies and showed that the isolate was pathogenic to several homopteran insects. According to these references, V. fusisporum produces a red pigment in the medium. When grown on SDA in this study, only the two V. fusisporum isolates from houseflies produced a pigment. However, the pigment was brown and colored the reverse but not the agar. Thus, this macroscopic character may be of little use in the discrimination between the closely related species of entomopathogenic Verticillium since the observed differences in pigment production may depend entirely upon the growth medium used. The conidial dimensions for V. fusisporum in this study were larger than described previously for this species. Although V. fusisporum, V. psalliotae, and V. lamellicola were never observed to produce more than one conidial type within the categories sickle-shaped, narrow fusiform, and broad fusiform, the deviations of the morphologies of these fungi from the descriptions in Gams (1971) shows the need for a revision of the entomopathogenic Verticillium species.

successfully colonized saprophytes. A high proportion of the cadavers that had been treated with the three nonpathogenic V. fusisporum isolates as live flies developed sporulating fungus growth anyway. Thus the fungus is capable of growing as a saprophyte on fly cadavers. However, V. fusisporum isolate MP6 from B. tabaci was pathogenic to other B. tabaci individuals and also to lesser mealworm and must be considered a true entomopathogen even though the host range of this isolate does not appear to include M. domestica. In conclusion, the results of this study indicate that Verticillium species other than V. lecanii and even species of Acremonium can act as primary pathogens in nature and their pathogenicity for a range of insects can be confirmed in bioassay tests. It does appear that among the entomopathogenic species of Verticillium and Acremonium the greatest potential for use in applied microbial biocontrol is to be found within the V. lecanii species complex, but the other taxa within these genera found here can also show appreciable pathogenicity and should not necessarily be ruled out for use in applied fungal control against some insect pests.

Entomopathogenic Potential of the Different Fungi

Balazy, S., Wisniewski, J., and Kaczmarek, S. 1987. Some noteworthy fungi occurring on mites. Bull. Acad. Pol. Biol. 35, 197–224. Domsch, K. H., Gams, W., and Anderson, T.-H. 1980. ‘‘Compendium of Soil Fungi,’’ Vol. 1 and 2. Academic Press, New York. Ekbom, B., and Åhman, I. 1980. The fungus Verticillium fusisporum as an insect pathogen. J. Invertebr. Pathol. 36, 136–138. Gams, W. 1971. ‘‘Cephalosporium-artige Schimmelpilze (Hyphomycetes).’’ Gustav Fischer Verlag, Stuttgart. Hall, R. A. 1984. Epizootic potential for aphids of different isolates of the fungus, Verticillium lecanii. Entomophaga 29, 311–321. Jun, Y., Bridge, P. D., and Evans, H. C. 1991. An integrated approach to the taxonomy of the genus Verticillium. J. Gen. Microbiol. 137, 1437–1444. Kalsbeek, V., Frandsen, F., and Steenberg, T. 1995. Entomopathogenic fungi associated with Ixodes ricinus ticks. Exp. Appl. Acar. 19, 45–51. Kristiansen, K., and Skovmand, O. 1985. A method for the study of population size and survival rate of houseflies. Entomol. Exp. Appl. 38, 145–150. Petch, T. 1931. Notes on entomogenous fungi. Trans. Brit. Mycol. Soc. 16, 55–75. Petch, T. 1938. Notes on entomogenous fungi. Trans. Brit. Mycol. Soc. 21, 34–67. Riba, G. 1983. Lutte microbiologique contre les aleurodes serres a` l’aide d’hyphomyce`tes entomopathoge`nes. Faune et Flore auxiliares an Agriculture, ACTA. Paris, 4-5 Mai 1983. SAS Institute. 1996. Release 6.12. SAS Institute Inc., Cary, NC, USA. Sanchez-Pen˜a, S. R. 1990. Some insect and spider pathogenic fungi from Mexico with data on their host ranges. Fla. Entomol. 73, 517–522.

The bioassay against B. tabaci showed that, except for V. lamellicola, all isolates tested were pathogenic to this host (Table 3). Among the four most pathogenic isolates were three of V. lecanii from B. tabaci and one isolate of Acremonium also from the homologous host. However, V. lecanii isolates were also among the least pathogenic cultures; so, this species may not necessarily be the best choice for control of Bemisia immatures and, as with nearly all fungal entomopathogens, the expected level of control of a particular host can be highly dependent on the isolate used. In the bioassay against adult houseflies, the mortality in general occurred very late, and none of the fungi seem to have any control potential against adult houseflies, a pest that has an estimated mean life span of 3–4 days in stables (Rasmussen and Skovmand, 1984). Interestingly, none of the three isolates of V. fusisporum, including the two from houseflies, caused mortality that differed statistically from that in the control. While V. fusisporum MP6 was pathogenic to lesser mealworm larvae, the two isolates from houseflies were not pathogenic. Because no pathogenicity for insects was shown for these isolates it is, therefore, probable that they were present as contaminants on the cuticle or in the guts of the houseflies from whose cadavers they were isolated as

ACKNOWLEDGMENTS This study was funded by the Danish Veterinary and Agricultural Research Council and by the Danish Ministry of Food, Agriculture, and Fisheries. Minna Wernegreen, Henriette Hansen, and Arne Kirkeby-Thomsen are thanked for their technical assistance and Dr. Annie Enkegaard for supplying Bemisia tabaci. REFERENCES