FNZ 68 - Simuliidae (Insecta: Diptera) - Bionomics
Craig DA, Craig REG, Crosby TK 2012. Simuliidae (Insecta: Diptera). Fauna of New Zealand 68, 336 pages.
(
ISSN 0111-5383 (print),
ISSN 1179-7193 (online)
;
no.
68.
ISBN 978-0-478-34734-0 (print),
ISBN 978-0-478-34735-7 (online)
).
Published 29 June 2012
ZooBank: http://zoobank.org/References/9C478D54-FEB2-45E8-B61C-A3A06D4EB45D
Bionomics
Little is known about general biology of Austrosimulium. Tonnoir (1925) made little comment and Dumbleton (1973), while he dealt with Simuliidae in general, gave little for Austrosimulium. Works that do deal specifically with biology of Austrosimulium are those of Towns (1976 et seq.) for A. australense and Crosby (1974a, b) for A. tillyardianum. Crosby (1974a) provided details for A. tillyardianum immature stages that included methods for separating the nine larval instars using morphological characters and measurements, aspects of larval biology, and the changes in population structure in relation to both water levels and the territorial behaviour of larvae that determined the availability of attachment sites. Pendergrast & Cowley (1966) gave a brief, informative account of the general life cycle and the four stages of Austrosimulium, including the eggs. They provided a page of figures probably based on A. tillyardianum given the pupal gill structure and comments about larval preferences for hard substrates: however, a drawing of a larva on a rock inaccurately portrayed the labral fans as directed towards the substrate. Since these earlier works, Austrosimulium is mentioned in numerous ecological studies, e.g., Collier & Winterbourn (2000) on New Zealand stream insects, however usually briefly and unidentified to species.
Adults
For most people, it is the biting of the female adult that captures attention, but few species actually bite humans (see species descriptions for details). In the North Island it is generally A. australense that is bothersome, although A. tillyardianum is also to some extent. Neither species tends to fly far from running water. Austrosimulium longicorne, sporadically distributed in North and South Island, is not known to bite. In a major review of host location by female simuliids, Sutcliffe (1986) broke the flying behaviour of the females into 3 parts: (1) post emergence, (2) non-oriented flight driven by endogenous activities and hunger, and (3) host-location proper, where long-, middle-, and close range (including post-landing activity) behaviours are involved. These behaviours involve attraction to exhaled carbon dioxide (CO2), visual, and thermal cues. In New Zealand CO2 released slowly from a gas cylinder does not attract Austrosimulium (TKC pers. obs.). For some simuliids, notably those attracted to birds, chemical cues along with CO2 are involved (Adler et al. 2004).
The thermal cues involve infrared radiation and probably the shape of host contrasted against the background. Consequently the colour of clothing is important, with lighter shades providing some protection (see Crosby 1992; Orr 1996). The earliest record known for New Zealand is an informative newspaper article by Richard Henry (1896), who commented on the fact that light coloured clothing was less attractive — this is possibly the first written observation on this aspect of behaviour for simuliids.
Attraction to various colours by female simuliids was examined in detail by Davies (1951, 1961, 1972) who well showed that dark blue attracted the most flies and white and lighter colours the least.
Crosby (1988) noted that there was variation in numbers of A. ungulatum and A. dumbletoni attracted to moulting, individually-caged, Fiordland crested penguins. He also noted that ducklings, humans, and penguin odour were equally attractive to A. ungulatum when moulting penguins were present, but that virtually no A. dumbletoni females were attracted to these others in the presence of moulting penguins, indicating that A. dumbletoni was host specific to Fiordland crested penguins. We can report that A. dumbletoni females are attracted to humans in the absence of moulting penguins, but do not bite. Allison et al. (1978) noted that there were 2 peaks of biting by A. ungulatum on captive penguins: one peak just after dawn and the other peak before sunset. This is well known for other simuliids elsewhere (Crosskey 1990; Adler et al. 2004).
The blood ingested by a female simuliid is used to produce eggs and in general insects that blood-feed are categorised as “anautogenous”. Those that do not blood-feed are “autogenous” and use nutrients carried through from the larval stage into the adult to produce eggs: therefore the importance of optimal feeding in the larval stage (see below). More eggs can be produced if a blood meal is taken. Of considerable importance is that Austrosimulium (certainly A. tillyardianum) can process the lymph of blood rapidly and extrude that as a diuretic fluid, allowing concentration of the more nutritive blood cells (see Frontispiece). This phenomenon has not been reported for Austrosimulium previously and otherwise not for simuliids either. It is, however, well known for other blood- or fluid-feeding insects (Chapman 1998).
For some simuliids and other biting flies the above behaviours are obligatory, but for others they are facultative and females will take blood meals opportunistically when possible. We surmise that Austrosimulium species that blood feed are of the latter kind. Still, there are many places in New Zealand where simuliids can be commonly found in large numbers in running water, but the adults do not bite, or do so in far fewer numbers than would be expected, giving rise to the possibility that there are sub-populations that do not blood feed. Whether an Austrosimulium species is a blood feeder, or not, can be partly judged by the size of the sclerotised abdominal tergites. These are large in the known non-blood-feeders (cf Fig. 70, 72). Smaller tergites presumably allow the abdomen to expand farther. Therefore, size of tergites can be used to suggest whether a species is autogenous or not (Rubtsov 1989; Dumbleton 1973).
An adult that has not laid eggs is known as “nulliparous”. One that has laid eggs, whether or not it has blood-fed, is “parous”, and eggs laid leave follicular remnants in the ovary, so the number of times the ovarian cycle (gonotrophic cycle) has occurred can be determined (Crosskey 1990; Adler et al. 2004). Some simuliids are known to repeat the gonotrophic cycle up to 6 times and are termed “multiparous”. Little is known about the number of gonotrophic cycles for Austrosimulium, apart from those biting the Fiordland crested penguin (Eudyptes pachyrhynchus Gray), and infected with Leucocytozoon tawaki Fallis, Bisset & Allison (Craig & Crosby 2008). Transmission of Leucocytozoon requires at least a “biparous” situation — ingestion of blood with the leucocyte, production of eggs, oviposition, then another blood meal to transmit the parasite back to penguins.
The above discussion on blood feeding leads to two common questions; first, “What do simuliids bite when I am not around?”, and second, “What did simuliids feed on before humans arrived in New Zealand?” Both are interlinked.
There is now substantive evidence that a claw with a basal tooth is a plesiomorphic condition (Currie & Grimaldi 2000) and that this equates with blood feeding on birds (Malmqvist et al. 2004). If Austrosimulium conforms to this pattern then it would be expected that members of the ungulatum species-group, which in large part possess a basal tooth on the claw, would be ornithophilic. However, apart from the reports of A. dumbletoni and A. ungulatum biting the Fiordland crested penguin, (e.g., Craig & Crosby 2008), there is little definitive scientific literature to indicate that this group of simuliids is, in New Zealand, bird feeding, even on poultry flocks. Instead, A. ungulatum is the scourge of humans in parts of the South Island and Stewart Island — clear evidence that ornithophily is not obligatory, and feeding on humans is opportunistic. This is in agreement with some of the findings of Malmqvist et al. (2004) for Scandinavian simuliids. Not well documented is that A. vexans on the Auckland Islands will, if possible, bite humans — thence its name (see p. 157). Henry (1896) made some useful observations regarding simuliids and the Maori around the time Europeans arrived in Fiordland, and speculated how Maori may have protected themselves from being bitten by covering their exposed parts with grease. But, an answer to the first question is probably, in part — “A bird if one can be found, or nothing if a bird or human is unavailable”.
The second question, regarding food source before humans were available, is more substantive and primarily relates to the indigenous vertebrate fauna of New Zealand. Possible blood sources prior to arrival of humans were restricted to bats, seals, and birds. Tuatara (Sphenodon), leiopelmatid frogs, skinks, and geckos were probably not greatly involved, if at all, even when more widely spread than present day. Of note, though, is that Sphenodon uses bird burrows for shelter. Accounts of the vast number of birds originally in New Zealand and the “dawn chorus” are numerous. Worthy & Holdaway (2002) referred to New Zealand as “the land of birds”. However, earlier, Holdaway (1989), rather pointedly noted that by the time Europeans arrived in New Zealand, the avifauna was but a battered remnant of that of pre-human times. Similarly, the southern fur seal Arctocephalus forsteri (Lesson), widespread around New Zealand, was used extensively by Maori and slaughtered in vast numbers by Europeans in the 1800s and its population is now estimated to be a mere 10% of the original (Phillips 2009). Henry (1896) also considered that seals would be a major food source for simuliids. Bats (Mystacina) were also much more common than at present (Holdaway 1989). Was this sparse list of vertebrate species the blood source for Austrosimulium prior to arrival of humans? It must have been so.
There is now, however, evidence (Worthy et al. 2006) that there were Miocene crocodilians and mouse-sized mammals in New Zealand before what is often referred to as the “Oligocene transgression” (25 Mya), when New Zealand was in large part, or even completely, submarine (Campbell & Landis 2001). There is no reason why such animals could not have been food items for simuliids and certainly, exquisite fossils of larval simuliids from Australia are known for the Lower Cretaceous (135 Mya) (Jell & Duncan 1986). But, this may not be the case for New Zealand, since our study here indicates Austrosimulium did not arrive until well after the Oligocene transgression (p. 60, 66, 86).
Another question not commonly asked, but of considerable interest, involves the ready access of female Austrosimulium to blood meals, notably that from humans. As discussed elsewhere, blood meals are used to produce eggs, so “Are there more simuliids in New Zealand now than before humans arrived?” A superficial answer to that would be yes, but it may be more complex. It is known (see Control section p. 30) that application of insecticides to farmlands can, via runoff, significantly alter aquatic invertebrate populations in streams (Hopkins et al. 1966; Stout 1975): thus, with apparent loss of predators, simuliids reoccurred in vast numbers. There is also the expectation that with enrichment (eutrophication) of waterways from fertiliser running off from farmland filter feeders such as simuliid larvae would increase in number. So, as a result of human interaction, there may well be more simuliids in some places in New Zealand.
The term “sandflies” is well chosen, because these insects are commonly a problem on beaches and at river mouths. Is this because they are searching for sea birds and seals? Maybe, but again, apart from the three species of Austrosimulium known to bite Fiordland crested penguins (Fallis et al. 1976; Craig & Crosby 2008), no scientific literature indicates that the large flocks of birds, or even seals for that matter, that still exist along New Zealand’s coastlines are bitten by simuliids. Related to this, a major die-off of yellow-eyed penguins (Megadyptes antipodes (Hombron & Jacquinot)) on the Otago Peninsula (1989–1990) was deemed not be caused by simuliids (Gill & Darby 1993). So, it can only be assumed that birds and perhaps seals provided blood meals to female Austrosimulium in pre-human times, but this assumption can be made with a certain amount of assurance, because there was little else.
Currently, with abundant livestock from country-wide farming operations, it might be assumed that biting by simuliids would be serious, thence well investigated and reported. There are, however, few data on the effects of New Zealand Austrosimulium on introduced animals. In Australia, however, there is a small, but definitive literature mainly for the subgenus Novaustrosimulium Dumbleton. McCarthy (1961) noted that A. (N.) pestilens Mackerras & Mackerras caused deaths in macropods (e.g., wallabies) by biting around the eyes and blinding them so they dehydrated, or starved to death. He also noted, in passing, earlier reports of severe effects on humans, livestock, and pets. Lee et al. (1962) in an examination of Australian biting flies as possible disease vectors, were of the opinion that while A. pestilens was spectacular at times in its attacks on domestic animals, A. bancrofti was more likely to be a potential vector of livestock diseases.
Muller & Murray (1977) trapped A. pestilens females from live sheep, along with numerous other species of biting flies. Other adults of A. pestilens and A. bancrofti (Taylor), captured from resting places, were shown to have taken blood meals from marsupials, oxen, sheep, and goats. They commented that major attacks by A. pestilens could result in deaths of livestock and macropods.
A considerable suite of wide-ranging work on the two main Australian pest simuliids was done by Hunter & Moorhouse (1976b) and particularly Ballard (1988 et seq.) and coworkers (Ballard and others, 1988 et seq.). Apart from Ballard’s cytological and molecular work mentioned elsewhere, investigations were made into host finding behaviour of A. bancrofti females and colonisation of substrates by larvae. Of interest here is the strong probability of at least 8 sibling species comprise the currently recognised A. bancrofti. Such a number of probable sibling species probably contributed to the differences found by Ballard in host finding behaviour of various populations of A. bancrofti.
Tang et al. (1996) examined the 16S rNA gene of nine species of simuliid known to transmit Onchocerca. One was A. bancrofti, the remainder Simulium species. Little was said about A. bancrofti, except that it was shown to be sister to Simulium.
For New Zealand, the first known listing of arthropods that affected humans and livestock was that by Helson (1956). Only A. australense and A. ungulatum were recorded as biting humans, and none were recorded as biting livestock of any type! Johnstone et al. (1992), however, investigated a series of incidents involving dermatitis in domestic cats in Nelson, and suggested that it was probably the result of biting by A. australense. However, while that may be so, it is more likely to have been bites from A. tillyardianum, larvae of which occur in astronomical numbers in the Maitai River (Tonnoir 1925; DAC pers. obs.) and may have been one reason for the early attempts to control simuliids in Nelson (see below p. 32). Johnstone et al. (1992) also noted that the cat dermatitis was quite similar to that on sheep on the West Coast after exposure to high densities of A. ungulatum. Orr (1996), a veterinarian, in a small publication on New Zealand Austrosimulium, noted that adults fed on owls, cranes, raptors, and small birds, such as starlings and thrushes — feeding on nestlings as well as older birds. She also noted that cattle, horse, sheep, and goats were all bitten, sometimes badly. There is, unfortunately, no attribution for this information. Henry (1896), on the other hand, was quite definite regarding the depredation of simuliid females on his dog, the impetus for his ingenious control method for the adults. More recently only A. australense, dumbletoni, and ungulatum are listed as ectoparasites of New Zealand vertebrates (Tenquist & Charleston 2001) and then, again, only for Fiordland crested penguins.
Still, biting by simuliids is of concern, in part because of the numbers biting humans, and especially their greater effects on those not previously bitten — tourists. TKC has often recorded biting rates equivalent to over 1000 per hour in many localities, especially just before sunset or before rain: on one occasion in February 1977 at Jackson Bay, South Westland he collected 360 specimens (nearly all A. ungulatum) in the act of biting his exposed arms and legs in 5 minutes, giving an equivalent biting rate of about 4000 per hour. Mackereth et al. (2007) in a review of potential vectors and vector-borne diseases in New Zealand listed A. australense and A. ungulatum as “medium threat”, because the flies could harbour and transmit protozoans and filaroids. They noted, however, that A. ungulatum had been shown to not be the vector of Whataroa virus, found in south Westland, now known (Derraik & Maguire 2005) to be transmitted by endemic mosquitoes. An outbreak of psittacinepoxvirus in imported rosellas (Platycerus sp.) was blamed on simuliids, as was an outbreak of poxvirus in shore plovers (Thinornis novaeseelandiae (Gmelin)) at Pukaha Mount Bruce (Gartrell et al. 2003).
The ability of simuliids to transmit disease causing organisms was investigated when myxomatosis of rabbits was introduced unsuccessfully into New Zealand in the early 1950s (Miller 1952b) and was further considered for later attempts with other haemorrhagic rabbit diseases. Gurr (1953) speculated in his review that simuliids and mosquitoes could possibly act as mechanical vectors. This brief review was prepared by Gurr using his own personal observations and the Entomological Research Station files of Dumbleton. It might be expected that Dumbleton would have written this review, but in May 1952 Dumbleton left Nelson, where he had been based as entomologist since 1929, to take up a 3-year post with the South Pacific Commission, Noumea, as Plant and Animal Quarantine Officer. The trials to establish myxomatosis proved unsuccessful as there were no appropriate vectors, as was the case in the 1980s when it was rumoured that myxoma virus had been illegally imported (Parkes et al. 2002). Crosby & McLennan (1996) when considering potential vectors of rabbit haemorrhagic disease virus (RHDV) were of the opinion that simuliids would not be involved. Parkes et al. (2001) in an examination of the epidemiology of RHDV concurred, but felt that Crosby & McLennan underestimated the possibility of transmission.
Blood meals taken are not used directly for flight energy; that is derived from sugars. There is no information on sugar sources used by either female or male New Zealand Austrosimulium. However, an intriguing observation was made by Baughan Wisely in the Murchison Mountains area of Fiordland in early 1953 (pers. comm. to TKC 1969): at a campsite he observed “sandflies” appearing to feed on a young shoot of a fern, suggesting that plant sap might be used as a source of nutrients as has been recorded for some simuliid species (Crosskey 1990). Of relevance here, however, are a series of studies by Hunter and co-workers on Canadian simuliids. Burgin & Hunter (1977a, b, c) showed that simuliid females and males fed both on floral nectar and homopteran honeydew secretion. The honeydew feeding was based on the crop contents containing melezitose and stachyose sugars that are unique to honeydew. Flowers and nectar are not particularly abundant in New Zealand, but in the South Island there is plentiful honeydew on southern beech (Nothofagus Blume) produced by two species of Ultracoelostoma Cockerell, the sooty beech scale, which is widespread in Nothofagus forest (Kelly et al. 1992). There is, however, no record of New Zealand Austrosimulium making use of this sugar source. Stanfield & Hunter (2010) showed that honeydew sugars affected flight performance in Canadian female simuliids. Honeydew-fed adults flew farther, but the flight speed of about 0.18 m/sec was not affected by the type of sugar meal; this speed is much slower than speeds given by Hocking (1953) and Crosskey (1990).
The distance flown by female Austrosimulium in search of a blood meal is of some interest, given that female adults of some species can be found long distances away from suitable breeding sites. For A. australense, though, the females move little laterally from the immediate vicinity of the breeding stream. Within some tens of metres from running water the rate of biting adults decreases markedly, a behaviour also noted by Dumbleton. This is in full agreement with a series of studies by Cumber (e.g., 1962) on insects associated with crops in the Rangitikei and Manawatu districts, and elsewhere. Of the thousands of insects collected, only one was a simuliid, even though there were suitable breeding sites nearby. Dumbleton (1973) remarked upon A. ungulatum accumulating on sea and lake beaches, with the assumption being made that they return to higher elevations to oviposit. Tonnoir (1925) never discovered the immature stages of A. ungulatum. It was Dumbleton who showed that the larvae and pupae were found at low density in small, cold-water, heavily-shaded streams, and we have found this too. The huge numbers of adults that occur at times indicates an ability to aggregate. Hence, we suggest that for A. ungulatum, and perhaps other species, there is a 3-part life-history strategy: 1 — production of adults, at low densities possibly, but from multitudinous localities; 2 — dispersal, usually downstream; 3 — aggregation at lower altitude, preferably at river mouths and beaches where there is likely to be a blood source from birds and perhaps seals. This scenario requires the blood-fed female then to find a suitable running water habitat in which to oviposit. There have been some suggestions (e.g., Rothfels 1981) that simuliids return to the natal locality to oviposit. Experiments by Hunter & Jain (2000), suggested that at least for the Canadian simuliid species they investigated, there was no return to the natal site and that any suitable waterway was used. Nothing, however, is known about this for Austrosimulium.
Related to the above is that the emergence of simuliids elsewhere is known to be mainly diurnal (Crosskey 1990). But again, little is known about that for New Zealand Austrosimulium, however, Colbo (1977) recorded that Australian A. bancrofti emerged mainly at sunset. Still, the informative article by Henry (1896) hinted at such periodicity. He found that after trapping almost astronomical numbers of female simuliids, the numbers during the rest of the day were much reduced, returning to high numbers the next day — was this emergence and aggregation?
The time between blood meals is dictated by rate of digestion of the meal and that, in part, is determined by temperature. Fallis et al. (1976) and Allison et al. (1978) indicated that at ca 16°C, digestion took 10 days. Downes (1958) noted that for most species of simuliid it is assumed a blood meal is necessary for full ovarian development; but if a blood meal is not taken, some oocytes still develop fully.
Crosby (1974a, b) from his work at Wainui, Banks Peninsula, noted that female adults of A. tillyardianum attracted to humans were almost all parous as indicated by follicular remnants. No males were ever collected, even though the sex ratio of reared adults was approximately 1 : 1, as in all species reared to adult in the present study. Reared females showed no ovarian development after 3–4 days, suggesting obligatory anautogeny for A. tillyardianum.
Of relevance here is that while simuliids generally are considered not to enter structures, Adler et al. (2004) recorded that some species enter nest boxes of birds. They feed on the birds and then remain to digest the meal. Austrosimulium ungulatum certainly enters enclosed spaces. Craig (2007, 2009) recorded incidents where large numbers of A. ungulatum females entered a van, fed on the occupants during the night, and then attempted to escape in the morning. That experience was substantiated a number of times, with A. ungulatum congregating in the wheel wells of the vehicle and biting the occupants’ legs, even while the vehicle was in motion (Craig, pers. obs.): a phenomenon well-known to many New Zealand residents who have stopped their vehicles in areas with simuliids. Once again, Henry (1896) was quite clear that simuliid females (without doubt A. ungulatum) entered his tent in large numbers and bit. We also note here for the first time that the large collection of A. australense (as caecutiens) made by Sinclair in 1844 was likely to have been from a tent or similar structure as they were only slightly damaged and not blood-fed, and such numbers of this species cannot be collected by sweepnetting.
It is rare to obtain a male adult in the field. They are occasionally collected at car headlights (NZAC label data), or while light trapping for Lepidoptera. Sometimes they have been collected by sweeping, once in a Malaise trap, and once attracted to headlamps deep in a cave in the Mt Arthur area. We only have one record of an A. australense male collected without lights, nets, or traps (St Johns, Auckland, in sunshine while inactive on a leaf of a shrub, S. E. Thorpe, 7 Oct 2011). Normally, they are obtained by rearing pupae, thence the importance of collecting and keeping live pupae.
Oviposition and mating
The number of eggs produced by a female simuliid is variable (150–600 eggs per gonotrophic cycle) and depends in part on the gonotrophy of the fly — blood meals allow larger numbers of eggs (Adler et al. 2004). Eggs of Simuliidae are markedly triangular in shape, a feature unique to the family (Tonnoir 1925; Craig & Crosby 2008). Tonnoir (1925) recorded dimension of eggs of A. tillyardianum as 0.126 × 0.189 mm, and Crosby (1974a, b) gave the dimensions for this species as falling in the ranges 0.12–0.15 × 0.20–0.24 mm.
Crosskey (1990) categorised simuliid oviposition behaviour into 4 types: scattering, dabbing, stringing, and layering. Colbo & Moorhouse (1974; 1979) noted that A. pestilens scattered eggs over the water, usually during a receding flood. Austrosimulium bancrofti was also thought to have similar behaviour since no egg masses had ever been found. Whether any New Zealand Austrosimulium species scatter eggs is unknown.
Crosby (1974a, b) recorded that A. tillyardianum females laid between 250 and 330 eggs in a single mass 6–8 × 3–5 mm on the downstream side of a protruding rock in a riffle area, about 15–100 mm below the water surface; these observations were in full agreement with Tonnoir (1925), and would be categorised as layering using Crosskey’s definition. Scarsbrook (2000), based on Towns (1976), stated that A. australense eggs were laid on the surface of rocks and organic material below water in batches of 200–500 eggs; TKC has normally found layered egg masses on the trailing vegetation in the water. Crosby (1974a) had eggs hatching in 6–8 days at 16–19°C; this time was shorter than reported by Tonnoir (1925) who stated that eggs of A. tillyardianum took 14 days to embryonate and hatch, but the difference in length to hatching could be related to the time of year the observations were made. Tonnoir (1925) described the changes eggs undergo as they matured — white immediately upon oviposition, then turning yellowish, and eventually light brown when fully mature, as in other simuliids (Crosskey 1990).
Nothing is known about male Austrosimulium precopulatory behaviour. This is similar to other simuliids, where also little is known (Crosskey 1990). Nothing, again, is known about the mating habits of normal adults of New Zealand Austrosimulium, in agreement with Adler et al. (2004) who commented that a least-observed facet of simuliid behaviour is that of mating. Crosby (1973a), however, while commenting on a single gynandromorph specimen of A. australense, noted that normal females of this species, attempting to obtain blood meals and then to escape were all mated, based on spermatozoa in their spermathecae. Copulation is known to elicit blood feeding in simuliids and other dipterans (Downes 1958; Wenk 1988). Moorhouse & Colbo (1973) observed mating swarms of A. pestilens to be specifically associated with riparian bushes of Callistemon R. Br. (bottlebrush). They noted that the preponderance of females captured had mated and that blood-fed females captured elsewhere also contained spermatozoa. Therefore, the assumption which can be made for New Zealand Austrosimulium females that bite, is they have already copulated, as found in other simuliids.
Where mating takes place is currently unknown for New Zealand simuliids. No mating swarms or mating on the ground (Crosskey 1990) have ever been reported. An assumption might then be made that since adults seem to leave the natal stream immediately (Crosby 1974a), mate searching and copulation takes place in local vegetation as in A. pestilens. It has been suggested that riparian vegetation is a habitat requirement for some New Zealand Austrosimulium species (Dumbleton 1973).
A mated, blood-fed female will develop eggs then needs to find running water in which to oviposit. It is known that blood-fed simuliid females do not always return to the natal stream to oviposit, and appear to be opportunistic (Lake & Burger 1983; Hunter & Jain 2000), but again nothing about this is known for Austrosimulium.
Larvae
As Adler et al. (2004) pointed out, flowing water is the life blood of all simuliids and the larvae have exploited virtually every conceivable flowing water habitat. Austrosimulium is no exception and larvae can be found on stones in high-altitude, low-flow trickles, to low-altitude, large, braided rivers, with trailing vegetation rather than stones being the normal substrate for some species. There is even one instance (Dumbleton 1973; this work p. 40) where larvae of A. stewartense and A. extendorum were found at high tide level where a stream flowed into the sea (Fig. 466).
In New Zealand, Austrosimulium larvae are almost ubiquitous in running water. However, cascades and vertical flows are one habitat that Austrosimulium has not managed to colonise, unlike the major radiation into this habitat by the Polynesian Simulium Latreille species in the subgenus Inseliellum Rubtsov (Craig 2003).
Simuliid larvae attach themselves to the substrate using a pad of salivary silk into which the hooks on the posterior proleg are embedded. This was illustrated for Austrosimulium by Tonnoir (1923a, b). The functioning of the circlet of hooks and the anal sclerite of simuliids was examined in detail by Grenier (1949) and Barr (1984). Basically, a larva bends over and secretes a pad of salivary silk onto the substrate, grips it with the hooks of the prothoracic proleg and then loops the posterior proleg hooks over and onto the pad. This is the normal method of movement. Larvae can also move, or disperse, by secreting a filament of silk into the water and then releasing from the substrate. The influence of a trailing silk filament to resettlement of the larva was investigated by Fingerut et al. (2006), who showed that the filament helped to grab the substrate from a greater height than otherwise would be possible in turbulent flow, and unexpectedly helped passage through pools.
Crosby (1974a) noted that first instar larvae of A. till-yardianum drifted downstream from the oviposition site to other slower velocity localities, drifting to higher velocity positions as second and later instars. Colbo & Moorhouse (1979) also showed that size of instar of A. bancrofti determined the level of velocity tolerated. Fonseca & Hart (1996) fully substantiated that dispersal of simuliid larvae was mediated by flow. In the A. tillyardianum study by Crosby (1974a) the number of larvae per square metre was influenced mainly by water level; floods reduced the population though movement of substrate and lower water level concentrated larvae on stones in higher velocity. Sustained low water allowed growths of algae and diatoms as well as sediment to accumulate on stones, and thence made them unsuitable for colonisation by simuliid larvae. Towns (1981a) in his study on benthic invertebrates in the Waitakere Ranges, which included A. australense, the only simuliid, noted conspicuous periphyton growths, even in forested reaches during spring and summer (November–February). Dense periphyton is inimical to simuliid larvae, and Towns’ observation that a low density of simuliid larvae occurred at peaks in periphyton growth is in accord with studies elsewhere. Casual observations by DAC and REGC of A. australense in the Glen Esk Stream, Piha, which also drains the Waitakere Ranges, follow those of Towns. Larvae are at maximum density in November, but by February are reduced, concomitant with lower flows and greater growth of periphyton. In part this relates to the predilection of A. australense larvae for trailing vegetation. Lower flows mean that bankside vegetation is not near the water, and that which is trailing in the water is covered with periphyton, unless velocity is high. Related too is the study by Death (1996) who showed that an unidentified simuliid was one of the dominant taxa in unstable streams that had clear substrate, but that this did not apply to stable streams.
In one of his studies, Towns (1981b) used a canopy to exclude light and inhibit periphyton growth. This provided a sufficiently clean substrate that a good population of A. australense persisted into summer, while populations elsewhere declined as normal. His observations are in full agreement with those of Quinn (2000), Crosby (1974a), and earlier observations from elsewhere on simuliids (e.g., Chutter 1968; Casey & Ladle 1976).
An unusual study by Death (1989) examined invertebrate communities outside and inside a short cave at Cave Stream in the foothills of Canterbury. The density of A. till-yardianum larvae was lower in the cave and larvae drifted more. As filter feeders, though, simuliid larvae should not be as affected as browsers which rely on periphyton that is reduced in the dark. Larvae of Austrosimulium have previously been reported from caves, in low numbers, but with no comment made (May 1963).
A more recent detailed study by Suren & Jowett (2006) on the Waipara River, Canterbury, was basically in agreement with the above, although again the species of Austrosimulium was not identified. They showed a moderate (ca 25%) decrease in populations after floods and major increases with larvae concentrated (145%) during low flow.
Quinn (2000) noted that simuliid larvae were sensitive to UV as originally substantiated by Donahue & Schindler (1998) and others later (e.g., Kelly et al., 2001; 2003). It had been suggested earlier by Craig (1997) that heavy pigmentation in larvae of Simulium (Inseliellum) was protection against insolation. We too have observed this for Austrosimulium. Larvae at high altitude will inhabit the undersides of perched stones (Fig. 487) when possible, and tend to be pale in colour (Fig. 317). If on the upper sides of stones (Fig. 474), they tend to be darker in pigmentation (Fig. 319). Larvae at lower altitudes tend to be paler (Fig. 305), but that is not always so (e.g., Fig. 314). Factors influencing colour of Simulium vittatum Zetterstedt larvae have been investigated by Zettler et al. (1998) and colour was shown to be under environmental control.
Tonnoir (1925) commented on the interaction between a larva and others that might drift within contact range. Resident larvae would attempt to dislodge the newcomer by biting its posterior. Such behaviour by simuliids has also been observed by DAC (pers. obs.). Crosby (1974a) also noted that larger larvae were territorial and kept smaller larvae from settling nearby, this behaviour having an effect on the demography of any one population on a stone. The result of this type of behaviour is well illustrated by Serra-Tosio (1967) who found that late instar larvae kept the substrate around them clear of periphyton and other larvae, and would attack chironomid larvae. At another level, Harding & Colbo (1981) showed that both intraspecific and interspecific competition between simuliid larvae could alter distribution of larvae on the substrate. This type of phenomenon has been examined by McAuliffe (1984) and others for benthic organisms in North American streams.
Such behaviours also involve hydrodynamics of flow around the larva. This is well understood for Simuliidae (Craig & Chance 1982; Chance & Craig 1986; Currie & Craig 1988; Craig & Galloway 1988; Ciborowski & Craig 1989; Eymann 1991; Lacoursière & Craig 1993) and has been examined in considerable detail by Hart (1986, 1988), Hart & Latta (1986), Finelli et al. (2002) and others. Hart et al. (1996) even measured velocity accurately within millimeters of the larval head. They showed that there is adaptive significance to territoriality by larvae. Apart from browsing the substrate, food acquisition by larvae is by the pair of elegant labral fans supported by the head which filter waterborne particles. Larvae pay particular attention to flow around the body and through the fans, and will position themselves in relation to others to achieve optimal flow conditions. Two particular patterns result. One pattern (e.g., Fig. 455, A. australense) is a diamond-shaped distribution where larvae are close to one body length away from others — the pattern shown by Serra-Tosio (1969). There is evidence (Thomson et al. 2004) that the downstream distance from one larva to the next allows the amplitude of turbulence from the upstream larva to dampen and velocity to recover, thence the upstream larva is less of an influence on flow through the fans of the downstream larva. Another pattern is for larvae to form a row with larvae side-by-side (Chance & Craig 1986; Ciborowski & Craig 1989; Eymann 1991). This is known to increase the velocity of water between the larvae, and thence through the fans, would increase particle flux (Ciborowski & Craig 1989), and was suggested as a strategy for use when food levels were low. Although not illustrated in this work, that is a common pattern for larvae of Austrosimulium species, such as A. unicorne and A. bicorne, which occur at low density under stones (DAC & TKC pers. obs.). Both these species occur at high altitude in pristine waters, where concentrations of particulate matter appear to be low.
Colbo & Moorhouse (1979) noted that at high population levels larvae of A. bancrofti and A. pestilens formed clumps on stones for the former and on vegetation for the latter, echoing earlier comments on those two species by Mackerras & Mackerras (1948). All other species formed spaced distribution. In general, microdistribution of A. bancrofti larvae depended on age of larva, water velocity, and nature of the substrate. Horne et al. (1992) showed that larvae of A. furiosum were normally found in velocities of 0.2–0.3 m/s, whereas Simulium ornatipes Skuse in the same habitat in velocities of 0.9–1.3 m/s. Austrosimulium furiosum (Skuse) larvae were particular about the choice of microhabitat and these velocity preferences were taken as evidence for microhabitat partitioning — this difference is important when benthic sampling to ensure accurate results are obtained. For New Zealand Austrosimulium, even for dense populations, such as occur at times with A. tillyardianum, larvae are always spaced.
Where velocities could be determined during this study, larvae were rarely taken at velocities below 0.3 m/sec or above 0.8 m/sec for stone-loving species (A. tillyardianum, A. multicorne) and there was a sharp cut-off below and above these limits (see Appendix 1). The upper limit appears to relate to shear stress of the water on the substrate since that was where even a thin layer of periphyton was scoured off the substrate. This can be seen in Fig. 483 (Kowhai River, Kaikoura) where larvae were only taken from the narrow dark band of stones along the edge — those with a thin layer of periphyton. Higher velocities were recorded for species on vegetation, but it was not possible to determine the exact velocity immediately proximate to the larvae. Nothing is known regarding velocity tolerances for species (e.g., A. unicorne, A. bicorne) living under perched stones.
Jowett (2000) gave the mean depth for larvae of an Austrosimulium sp. in a small stream as 0.08 m and the mean velocity as 0.18 m/s, the latter low by our observations. He also quantified that larvae in small streams prefer higher velocity than that of the mean velocity, and we agree fully. This phenomenon allows rapid assessment of a locality for the likely presence of simuliid larvae — substrates in higher velocity are examined first.
Quinn & Hickey (1990a, b) in a New Zealand-wide survey of 88 rivers showed that maximum density of unidentified Austrosimulium larvae was to be found on substrates with sizes up to large cobble (128–256 mm diameter). A rationale for that was the increased complexity and provision of a three-dimensional habitat; aeration is also improved. In an earlier study Pridmore & Roper (1985) examined macroinvertebrates in runs and riffles of three North Island streams. They showed statistically that unidentified Austrosimulium spp. were significantly more abundant in riffles than runs in the Rangitukia Stream. Although in farmland, the stream substrate was andersite and basalt cobble usually favoured by A. tillyardianum, and also within its distributional range.
Boothroyd & Dickie (1991), in an examination of colonisation of artificial substrates by aquatic invertebrates, showed that while chironomids dominated the drift and substrates, A. australense could at times comprise up to 49% of that fauna. Drifting by simuliid larvae is a well known phenomenon (Crosskey 1990) and the initial dominance by simuliids on fresh substrates is in keeping with other studies. Similarly, Death (1996, 2000) dealt with colonisation of aquatic invertebrates after substrate disturbance. Austrosimulium larvae colonised readily, but were eventually replaced by more slowly colonising species. This is in full agreement with Harding et al. (2000) and Suren (2000), with Austrosimulium dominating in streams after disturbances by forestry, and their occurrence in urban milieu. In his study on A. tillyardianum Crosby (1970) constructed a stream deviation having a substrate of small stones, and found that larvae colonised the new substrate in 3–4 days.
Collier (1995) examined some 29 lowland waterways in Northland. While an unidentified Austrosimulium occurred at 19 of those sites, it was not dominant enough to determine habitat requirements. Collier et al. (1998) also examined physical parameters in relation to macroinvertebrate fauna in 20 lowland Waikato streams. Austrosimulium larvae occurred at all sites. In keeping with most other such studies, larvae were found in faster water, and a range of physico-chemical parameters, such as water depth, dissolved oxygen, and temperature accounted for the greater part of larval densities on macrophytes. As usual for most such studies, the simuliid species was not identified, but was assuredly A. australense.
Simuliid larvae are well known to form large concentrations at lake outlets, an explanation for which has been the availability of planktonic food material (Crosskey 1990, Adler et al. 2004). As part of a series of studies on lake outlets, Harding (1992) examined the physico-chemical parameters and invertebrate fauna of three lakes in Westland. There were marked differences between the inlet and outlet faunas. Along with trichopterans, chironomids, and gastropods, simuliids (unidentified) dominated the outflows, but not inlets. Similarly, simuliids dominated farther downstream: these findings mirror those from elsewhere in the world, where it has been suggested food supply and temperature are optimal in these areas. Harding (1994) then examined the benthic fauna for 20 South Island lake outlets; simuliids occurred in 16 of them. He found that natural alpine lake outlets were dominated by mayflies, stoneflies, chironomids, and simuliids. Man-made lake outlets with epilimnetic (surface) discharge also had abundant simuliids, but those with hypolimnetic (subsurface) discharge had few. Regulated epilimnetic and unregulated mid- to low altitude lake outlets were similar with moderate abundances. The percentage of catchment that was forested appeared to be important, as was the amount of regulated flow. As usual with such studies, the simuliid was not identified, but listed as “Simuliidae ungulate”: since the lakes were spread over the latitudinal range of the South Island more than one species may have been involved.
Life history
Tonnoir (1925) noted that larvae of Austrosimulium are found year round, normally with only larvae occurring during winter months. However, low numbers of female adults of the biting species may be encountered during winter. Crosby (1974a) identified 9 distinct (and two indistinct) cohorts of larvae — a strong indication that A. tillyardianum had a multivoltine life history at Wainui stream, Banks Peninsula. The numbers of larvae ranged from 7 000 per m2 in March, peaked early in April at 15 000, dropped between June and September to 2 000–3 000 and then climbed to 25 000 in November. Crosby (1975) found that A. tillyardianum constituted up to 45% of the aquatic fauna in the Wainui stream. Towns (1981a) in a study of aquatic invertebrates (including A. australense) in the Waitakere River, west of Auckland, found that simuliid larvae overall constituted less than 5% of the aquatic fauna over a full year, but were a major portion of the fauna during spring (September and October), with sharp decreases over summer and into autumn and winter. He considered A. australense to possess the life cycle of rapid growth, high colonisation potential, and to be probably multivoltine.
Scarsbrook (2000) incorrectly stated (as Austrosimulium sp.) that Crosby (1974a, b) found 2 generations for A. tillyardianum. Crosby stated that the species was multivoltine and gave clear evidence of that. Scarsbrook also cited more than 2 generations a year for an Austrosimulium sp., in Otago, but gave no further information.
Number of larval instars
Tonnoir (1925) did not determine the number of larval instars for Austrosimulium, but thought there were 4 or more than. Crosby (1971; 1974a, b) in his detailed study of A. tillyardianum, showed definitely 9 instars, in good agreement with subsequent studies of simuliids, where numbers range from 6 to 11 (Crosskey 1990; Colbo 1989). Towns (1981a) could not distinguish individual instars of A. australense larvae using head length; he instead grouped larvae into 4 size classes, the largest being the final instar with mature pupal respiratory gill histoblasts. Ballard (1991) suggested 7 instars for A. bancrofti.
Pupae
The first accounts of New Zealand Austrosimulium pupae were by Tonnoir (1923a, b, 1925). He recorded that when ready to moult into a pupa a larva usually selected some sheltered spot on the stone they inhabited. He commented on possible negative phototropism, since larvae on leaves normally selected the undersides to pupate. Crosby’s (1974a, b) investigations of A. tillyardianum agreed, with pupation normally on the downstream sides of stones with emergence of adults after some 7 days. There is a variance with Tonnoir (1925) who stated that about 12 days were spent as pupae, however, he did not clarify what species was involved, although probably A. tillyardianum from the Maitai River, Nelson, and there may have been seasonal temperature differences.
Tonnoir (1923a) made detailed observations and illustrated the pupation process, construction of the cocoon, and emergence of A. australense. To settle a disagreement regarding the number of prolegs in larval Diptera, he republished his findings (Tonnoir 1923b) with additional notations. Later he noted (Tonnoir 1925) that one species (probably A. laticorne) completely closed the cocoon except for subsequent openings for the pupal gills — behaviour quite distinct from that of A. australense — and he commented that construction of other forms of cocoons would be interesting to investigate. He also noted that the typical time of day to pupate was midday, with emergence at about the same time.
In a series of studies on cocoon spinning behaviour of Simuliidae, Stuart & Hunter (1995; 1998) demonstrated that components of such behaviour as originally described by Tonnoir (1923a, b) could be used as phylogenetic characters for the family. That work culminated with an examination of A. australense (Stuart 2002). A cladistic analysis using cocoon spinning characters of 7 genera of simuliids placed Austrosimulium as the sister taxon to Eusimulium Roubaud + Simulium s.s. There was strong support for that arrangement. It was also in agreement with Dumbleton (1963b, 1973) who suggested that those taxa shared a recent common ancestry. Stuart (2002) commented that the behaviour required to produce a well formed cocoon, common to and likely homologous within Ectemnia Enderlein, Austrosimulium, and Simulium, allowed the pharate pupae to spin a cocoon on any substrate. This is most noticeable for some species (e.g., A. australense) that produce distinctly different cocoons on a flat surface, such as a leaf, in contrast to the cocoon spun in the fold of a grass leaf.
Colbo & Moorhouse (1979) observed that for A. bancrofti pupae, which normally have the opening downstream, as also noted by Tonnoir (1925), have the openings pointing randomly in turbulent flow. It also appeared that larvae moved to pupation sites in concert, since pupae in groups were often the same age. We have also noticed this in species that have low larval density, such as A. unicorne and A. bicorne. One must assume there is some stimulus to pupate together.
Emergence of the adults
Tonnoir (1923a, b) illustrated eclosion of an adult A. australense (as tillyardi) from its pupal cuticle. The adult is surrounded by a bubble of air and floats to the surface. Most simuliids fly away instantly, with a few species known to spend a short time resting before flying away. This phenomenon is well described by Crosskey (1990) and Adler et al. (2004). The adults do not get wet because the body, including wings, has a vestiture of microtrichia and microsculpture (e.g., Hannay & Bond 1971a, b). The sex ratio of Austrosimulium, derived from rearing studies, is close to equal (Tonnoir 1925; Crosby 1974a, b; Colbo & Moorhouse 1979; this study). It is extremely rare to capture males in the wild, and Tonnoir (1925) noted that he never captured one. Neither did Crosby (1974a, b), nor we. Indeed, sweep-netting over water was not part of our standard collection protocol, since, without exception, it produced no adults except when females were biting!
Colbo (1977) investigated A. bancrofti in response to a lack of information on the emergence patterns of any simuliid in Australia. Males and females emerged simultaneously with a small peak mid-morning, one just after mid-day and a major one at sunset. Colbo was of the opinion that emergence was controlled by light with temperature of the water an important factor. A similar level of detail is not known for the New Zealand Austrosimulium, still Tonnoir (1925) noted that emergence tended to be at midday. Similarly, the observation by Henry (1896) that after reduction of adults through trapping, numbers stayed low the rest of the day, but recovered the next day, indicates a daily emergence. That observation, no doubt, was for A. ungulatum.
Ecological importance
Austrosimulium in New Zealand is listed as the tenth most common aquatic invertebrate in 90%, or more, of the country’s running waters examined in the National River Water Quality Network (Boothroyd 2000). We fully agree. Running waters in the North Island that lack simuliids are not common and this is also the case on Stewart Island (Chadderton 1988, 1990). Places that truly lacked simuliids were usually those that had little trailing vegetation, or leaves held against rocks by the current. Similarly, and in addition, those with unstable cobble substrate that rolled frequently in floods tended not to be inhabited. Furthermore, streams and rivers that were stable and had deeply embedded or armoured bed material, and/or had encrusting vegetation were unlikely to harbour simuliid larvae. Habitat substrate preferences of New Zealand aquatic invertebrates have been examined in some detail (e.g., Death 1996; 2000) and Austrosimulium, in general, was deemed characteristic of unstable substrates.
One distinct restriction to higher altitudes is that of A. unicorne. Dumbleton noted that it only occurred above 760 m a s l. In the Arthurs Pass and Mount Cook regions this appears to be related to the presence or absence of perched stones in a stream. With a steep slope, streams in these mountainous areas change rapidly from an erosional pool–riffle profile to a depositional braided form. The former reaches contain large stable boulders, remnants from higher water flows from deglaciation, and provide stable conditions for Blephariceridae and other members of the high altitude aquatic fauna of New Zealand (Stout 1975). The mixture of substrate sizes results in a pool–riffle sequence (Frey & Church 2009) and provides the perched stones required by A. unicorne. The transition from this suitable habitat and the braided profile takes place at about 700 m a s l in the Mount Cook region and somewhat higher in the Arthurs Pass environs. The sharp cut-off in occurrence of A. unicorne is unlikely to be solely the result of preferences for lower temperature. But, recent collection of A. unicorne in the Haast Pass (NZS177) at only 471 m a s l does, however, indicate that temperature may be the major environmental factor, even though the substrate was suitable.
Control of simuliids
The impetus for Tonnoir’s 1925 work on Australasian simuliids was control. He stated in his opening sentences that he was hired by the Cawthron Institute to accumulate information on simuliids that would lead to eventual control. As always, “knowing the enemy” is the first step in such a project, and therefore resulted in that early taxonomy.
Dumbleton (1973) briefly discussed control of Austrosimulium. For reducing effects of biting adults, he suggested repellents. Little has changed since then and basically the same repellents are still effective — formulations, however, have improved. Both Crosby (1992) and Orr (1996) reiterated the obvious and that is to dress appropriately. Colour of clothing is known to make a difference (Henry 1896). Darker colours, supposedly radiating more infrared radiation, are more attractive, lighter clothing less so, but there are exception for some simuliid species elsewhere (Davies 1951 et seq.). Such protective measures are covered well by Adler et al. (2004).
Managing the size of the population of immatures is a more permanent solution and has been successful in many places worldwide. The first use of DDT on simuliids was in South America by Fairchild & Barreda (1945). This became a method of choice, particularly by WHO (World Health Organisation) for control of Simulium damnosum Theobald and river blindness (onchocerciasis) in Africa (Crosskey 1990), until the discovery that such chlorinated hydrocarbons were concentrated up food chains. A review of early uses of DDT on simuliids is by Chance (1970). As far as is known, DDT was never used to control New Zealand simuliids, but experiments have been carried out (Anon 1961). Hopkins et al. (1966) examined the effects on aquatic invertebrates of DDT prill application on farmland adjacent to connected streams and found there were major effects. There was an “enormous increase” in the number of Austrosimulium larvae, no doubt the result of removal of controlling predator species — Dumbleton (1945) cautioned in his article on the potential usefulness of DDT that there were also potential problems such as the removal of predators in agricultural areas.
Biodegradable organophosphates (e.g., Abate™) were substituted as control agents and are still used in some situations. Again, such measures are discussed well by Adler et al. (2004). In the 1970s a natural toxin produced by Bacillus thuringiensis israelensis (Bti) was discovered to be fairly specific as a poison for simuliid larvae, with little or no effect on other aquatic invertebrates. The alkaline nature of the larval simuliid gut is thought to activate the Bti toxin, which is known to act by lysing the midgut epithelial cells (Soberóne et al. 2007). Since the early 1980s, Bti has become the control agent of choice, and even though it is more expensive than artificial chemicals, is extensively used in the USA (Arbegast 1994) and elsewhere. In New Zealand, Bti has only been used experimentally against simuliids. Chilcott et al. (1983) showed that both A. laticorne and A. multicorne were susceptible. They suggested that A. australense would be a good target species since its larvae occurred in large numbers in open running water. Less possible to control would be the more serious biter, A. ungulatum, the larvae of which tend to inhabit smaller bush-covered streams. Goodwin (1985) also obtained good results with Bti on larvae of A. australense and A. longicorne. In a review of the status of use of Bti in New Zealand, Glare & O’Callaghan (1998), made the point that there is little reason for not using it, were it not for strict regulations in place. No control programmes using Bti are currently in place in New Zealand.
Apart from deterring the biting of female sandflies by using suitable dress, repellents, or insecticides, control of simuliids elsewhere has also been attempted using natural enemies (Laird 1981; Kim & Merritt 1988). An overview of their potential usefulness for controlling New Zealand Austrosimulium was provided by Crosby (1989). Worldwide, there are no examples of successful control of Simuliidae using natural enemies.
Natural enemies and parasites
Predators
Predators of simuliid adults have been well examined in general (Crosskey 1990; Werner & Pont 2003). The latter authors noted that Austrosimulium adults were taken along with those of the South American Lutzsimulium d’Andretta & d’Andretta by the asilid Holocephala oculata (Fabricius) (Carrera & Vulcano 1961). Their Austrosimulium would actually be Paraustrosimulium Wygodzinsky & Coscarón. Dumbleton (1973) briefly mentioned predators, noting the record of Miller (1969) that empidid adults, Thinempis otakouensis (Miller), took Austrosimulium adults. There is no other substantiated record of predation on Austrosimulium adults.
It might be assumed though that the blue duck (Hymenolaimus malacorhynchos (Gmelin)) would take Austrosimulium individuals as part of its diet, but there seems to be no record of that, except perhaps for that of Kear & Burton (1971) who noted dipteran adult and larval remains in the duck’s droppings. Pierce (1986) recorded the occasional Austrosimulium in the diet of stilts (Himantopus spp.), as did Bisset (1976) for the paradise shelduck, Tadorna variegata (Gmelin). Rock wrens (Xenicus gilviventris Pelzeln) are reported to feed on adult black flies, but accounts are anecdotal (e.g., Google Images © , YouTube © ); however, flies and other insects are part of the bird’s usual diet (Troup 2009).
Crosby (1974a, b, 1975) in his in-depth study on A. tillyardianum showed that larvae of the trichopterans Hydrobiosis parumbripennis McFarlane and Hydropsyche colonica McLachlan were the main predators of A. tillyardianum larvae. Hydrobiosis parumbripennis ate the larvae at the same relative frequency that they occurred in the fauna, but H. parumbripennis, when in its later instars, ate proportionally more simuliids. There was no indication of detrimental effects on the population of A. tillyardianum.
Harding (1997) in a study of coexistence of larvae of two species of Hydropsysche Pictet (Trichoptera) examined gut contents of the main predators in the habitat. The common bully (Gobiomorphus cotidianus McDowall) contained 64% simuliids and a hydrobiosid trichopteran (Costachorema xanthopterum McFarlane) 30%. It is well established that Austrosimulium larvae are used extensively as prey by various fish species. McIntosh (2000) listed proportional occurrence of simuliids in guts of Canterbury galaxias (Galaxias vulgaris Stokell) at ~25%, bluegilled bully (Gobiomorphus hubbsi (Stokell)) at ~12%, longfinned eel (Anguilla dieffenbachii Gray) at ~24%, and shortfinned eel (Anguilla australis Richardson) at ~11%. As usual, the simuliids were not identified. A collection of 38 mature and near mature larvae of A. longicorne extracted in August 1971 from the stomach of the Canterbury mudfish, Neochanna burrowsius (Phillipps) from Mt Somers by P. Cadwallader: these specimens are now in NZAC.
Chadderton (1990), in an examination of aquatic communties of Stewart Island, recorded in some detail predation on simuliid larvae by the isopod Austridotea benhami Nicholls. The examination of gut contents of individual A. benhami from 8 localities showed Austrosimulium sp. larvae occurred at a frequency of 17.5%.
Crosby (1989) analysed in detail attempts to control Austrosimulium larvae with predaceous larvae of the chironomid Cardiocladius australiensis Freeman and up to 4 species of the dragonfly genus Austroaeschna Selys (Anon. 1932). The latter hard-to-find publication in the 6th Annual Report of CSIRO stated on page 23:
”9. Section of Systematic Entomology. — During the year Mr. A. L. Tonnoir, senior entomologist of this Section, and his two assistant entomologists, Miss W. Kent Hughes and Miss L. Graham, carried out a considerable number of economic researches in addition to their systematic work. . .
(iv) New Zealand Problems. — In return for assistance received from New Zealand in work on Oncopera, oak scale, and noxious weeds control, investigations on the New Zealand sandfly problem (Simulium) and on the . . . During the year, 270 live dragonfly larvae and 1,200 live larvae of Cardiocladium, both predatory on Simulium, were successfully transported to New Zealand; . . .”
These attempts to introduce predaceous insects to New Zealand were briefly mentioned by Miller (1969) and Dumbleton (1973), and reported in detail by Crosby (1989). Some 1 200 Cardiocladius australiensis larvae were collected from the Molonglo River in Canberra and sent in 2 consignments to Nelson where they were received in late January and early May 1932. The surviving 777 larvae and 19 pupae were released into the Maitai River, Nelson, at the Council Reserve.
Earlier 6 consignments of Austroaeschna dragonfly larvae, totalling about 500 larvae, were sent in November 1930 and October/November 1931. They were collected in Canberra at the Cotter River below the water supply dam, and possibly also from the Molonglo River. In November 1930 2 consignments were received in Nelson, held in aquaria for about 3 weeks, then the surviving 88 larvae were released at 2 sites in the Maitai River. A consignment was also sent to E. Percival at this time, but there is no information on their survival or release at Cass. In October/November 1931 further specimens were sent to Nelson, probably in 2 consignments, but only 31 survived for release into Poorman Valley Stream. A consignment was also sent to Percival: about half survived the journey and were released at Cass at the beginning of December (Anon. 1931).
Rowe (1987) suggested the Austroaeschna larvae were collected in Tasmania, based on a letter written by E. Percival to J. S. Armstrong on 10 July 1957. On that basis Rowe (1987) considered the species sent to New Zealand was Austroaeschna parvistigma Martin. Crosby (1989) did not regard the letter as reliable evidence, as it contradicted other evidence written at the time of the importations and on DSIR files.
Austroaeschna spp. and Cardiocladius australiensis were sent to New Zealand because of the enthusiasm of Tonnoir, as he thought they were effective in controlling larvae in Canberra where he was then based. David Miller was then in charge of the entomology group at Nelson (which included Dumbleton), and with resources overstretched he was not keen to begin this project (Crosby 1989). The releases at Nelson were therefore done under pressure, and may not have been made in optimal areas for survival. Percival would have needed to make special trips to Cass for the releases, but it is likely they were released in more ideal places. Apparently, these attempts received considerable attention at the time (Anon 1931), but were ultimately unsuccessful.
Parasitism
Parasitoids are rife in simuliids. Adler et al. (2004), listed 3 families of nematodes (Mermithidae, Onchoceridae, and Robertdollfusidae), 3 major groups of fungi (Classes Chytridiomycetes, Hyphomycetes, Zygomycetes) and 1 of the straminopiles (now Chromista: Oomycota), and also helicosporidia, ichthyosporeans, protists, bacteria, and viruses, not-to-mention rarer organisms such as Gordian worms. One of the most commonly occurring parasitoids, world-wide, is the blastoclad fungus, Coelomycidium simulii Debais. Nematodes are sufficiently pervasive to have been proposed as control agents for pest simuliids (Laird 1981; Kim & Merritt 1988; Poinar 1990). For New Zealand Austrosimulium, such parasitoids as above are relatively uncommon, and with one exception (Fallis et al. 1976) known only from larvae (Dumbleton 1973; Batson 1983; Poinar 1990). Their rarity may be the result of a narrow ecological regime tolerated by the organisms.
In an extensive chapter on pathogens of simuliids, Adler et al. (2004) noted that the classification of some groups of pathogenic microorganisms is undergoing considerable change. One that is relevant here is that Microsporidia is now considered a phylum in the Kingdom Fungi (Keeling & Fast 2002; Lee et al. 2008). More recently McCreadie et al. (2011) reviewed the ecology of symbionts of simuliid larvae.
Dumbleton (1973: 545) noted that a few A. unicorne larvae were parasitised by a probable microsporidian, but it was not mentioned in later discussion on pages 576–577. Examination of material from his collections indicates that he was referring to Coelomycidium Debais (Blastocladiomycota). In an unpublished study Chilcott (1979; and in Glare & O’Callaghan 1998) examined parasitoids of Austrosimulium and the histopathology of Coelomycidium and Thelohania Henneguy. It appears that the Coelomycidium species investigated was not C. simulii. Later Batson (1983) described two parasites of low frequency from immature unidentified Austrosimulium larvae from Deep Stream, west of Dunedin, South Island. One was a new genus and species, Hirsutusporos austrosimulii Batson (then phylum Microspora, family Nosematidae), and the second a member of the family Thelohaniidae for which no further details have ever been provided.
Both DAC and TKC (unpub. obs.) have observed Austrosimulium larvae packed with spherical organisms and assumed this was a result of microsporidial infection. However, by definition microsporidia are intracellular parasites (Sprague et al. 1992). Many of the organisms that we and Dumbleton observed were intercellular and therefore not Microsporida, e.g., Fig. 499, 500.
Kingdom Fungi
Phylum Blastocladiomycota, Class Blastocladiomycetes, Order Blastocladiales
The majority of the parasitised material on hand is with high probability, Coelomycidium simulii (or a closely related species), which is a widespread member of the order formerly known as Chytridiales (Poinar & Thomas 1984, Adler et al. 2004). Crosskey (1990) did not list it as known from Australasia, so the example provided here is a new record for the region and probably represents a new species. Crosskey (1990) gave, however, considerable detail about the known life cycle for Coelomycidium simulii. Typical is that the body of the simuliid larva is closely packed with individual spherical, monocentric thalli, giving a speckled appearance and usually altered colour, meaning that parasitised larvae are easily recognised (Fig. 500). In the head the thalli are less invasive (Fig. 499). At high magnification individual thalli used in this study showed virtually no internal structure and had a refractive body wall, both characters that are indicative of chytrid fungi.
Development of the parasitoid is at the expense of the internal organs of the larva and delays development: therefore, towards the end of a given cohort the frequency of parasitised larvae rises as they remain in the population. Furthermore, development of the pharate pupal gill is compromised and may result in abnormal shapes (Fig. 501, 502). If infected early with C. simulii a larva cannot complete development and dies.
In New Zealand, parasitised larvae have only been found in the South Island and generally at higher altitudes where water is colder. The northernmost record is at Flora Hut, Mount Arthur, where a single parasitised A. ungulatum larva was recovered. Along the Kaikoura Coast, A. multicorne larvae with Coelomycidium were found in Green Burn. The southernmost records for the South Island are those from Dunedin, Tuatapere Scenic Reserve, and Granity Hill.
It does seem unusual, however, for A. stewartense populations in low altitude streams near Dunedin to have Coelomycidium prevalent in some 14% of larvae. No details of water temperature were given with label data for those collections. However, calculations following Mosely (1982) indicate that although at low altitude, the mean temperature at these high latitude localities would be about 9.5°C. This is not outside the temperature range of higher altitude localities with parasitised simuliid larvae — and again indicates that low temperature is needed for the presence of parasites. This is in full agreement with Ezenwa’s (1974) study of mermithid and microsporidian infections of simuliids in Newfoundland, Canada. He showed association of these parasites with cooler water; the majority of stream temperatures were below 10°C. Parasites were more prevalent in the spring and autumn in his study. He found little evidence of pH restriction, most water being slightly acidic, ranging from 5.7–6.8. Again, his findings are in general agreement with the situation in New Zealand.
No parasitised Austrosimulium larvae have been recovered from Stewart Island. Coelomycidium is now known for Campbell Island, but not yet the Auckland Islands.
Phylum Microsporidia, Class Microsporea, Order Microsporida, Family Thelohaniidae(?)
At the Homer Tunnel location (NZS32a) we recovered in A. tonnoiri larvae, at low frequency (<10%), a parasitic organism that packs the posterior abdominal fat bodies tightly and imparts a light pink coloration. When separated into individual organisms these are slightly dumbbell-shaped and up to 0.003 mm long (Fig. 504). Some appear to possess filaments (poorly shown in the figure).
The size and shape of these organisms agrees well with those illustrated for a hyphomycete Tolypocladium cylindrosporum W. Gams (e.g., Samson & Soares 1984), known from New Zealand and infecting mosquito larvae. A similar shaped organism, tentatively identified as Tolypocladium, isolated from Simulium piperi Dyar & Shannon in California was illustrated by Adler et al. (2004) although no size was given. Nadeau & Boisvert (1994) experimentally infected Simulium vittatum Zetterstedt with T. cylindrosporum, but achieved low mortality. However, Tolypocladium W. Gams is not intracellular and its hyphal bodies are free in the haemocoel, so it is most unlikely the organism we have is Tolypocladium.
Batson (1983), however, when describing the microsporidian Hirsutusporos austrosimulii, noted the presence of another microsporidian of the family Thelohaniidae, but he never gave further details. However, Thelohania in Austrosimulium had been noted by Chilcott (1979). Thelohaniids are known from mosquitoes (Andreadis & Vossbrinck 2002) where diplokaryotic meronts pack the larval fat bodies. The size and shape of the organisms we have is consistent with such stages. The filaments observed may be polar filaments known for other stages of Thelohaniidae. A major problem, though, might be that thelohaniids have a second host, often a cyclopoid copepod that is unlikely to be present in the stream in which we encountered the parasitised A. tonnoiri. More detailed examination will be needed to identify this organism.
Phylum Zygomycota, Subphylum Kickxellomycotina, Order Harpellales
Crosby (1974a, b) noted that the trichomycete, Harpella melusinae L. Léger & Duboscq was common in the alimentary tract of A. tillyardianum larvae, but appeared not to have any effect on populations. He also recovered Smittium sp. Williams & Lichtwardt (1990) studied the trichomycete gut fungi found in the New Zealand aquatic insects and found that 4 species were present in simuliids: Harpella melusinae, Pennella asymmetrica M. C. Williams & Lichtw., Smittium simulii Lichtw., and Smittium culicis Tuzet & Manier ex Kobayasi. Trichomycetes are worldwide in distribution (Lichtwardt & Williams 1987; Beard et al. 2003) and Stachylina litoralis Lichtw., White & Colbo and Smittium culicisoides Lichtw. have even been recovered from Crozetia seguyi Beaucournu-Saguez & Vernon from the isolated Crozet Islands, South Indian Ocean (Reeves et al. 2004), and H. melusinae from A. vexans on the Auckland Islands (Crosby 1974d). The instar at which trichomycetes colonise larvae was investigated by Crosby (1974d) who showed that 50% of 2nd instar A. tillyardianum larvae were infested, with later instars almost completely. While trichomycetes can be detrimental to female simuliids, replacing the eggs with fungal cysts, in larvae the relationship appears to be more that of a commensual (Adler et al. 2004).
In a wide ranging examination of trichomycete gut fungi in Australian aquatic insects, Lichtwardt & Williams (1990) recorded Harpella melusinae from larvae of Austrosimulium bancrofti, A. furiosum, A. mirabile, A. torrentium, and A. victoriae, as well as from larvae of Paracnephia and Simulium species. Smittium aciculare Lichtw. was recovered from larvae of Austrosimulium mirabile as was Smittium simuliiLichtw. from those of A. bancrofti and A. furiosum. Other harpellids were taken from larvae of numbers of other simuliid species. An unusual find was an amoeboid Paramoebidium sp. L. Léger & Duboscq in larvae of A. furiosum and A. bancrofti, including those of some Simulium species. Overall, the Australian simuliid larvae had a much more diverse gut fauna than found in larvae of New Zealand Austrosimulium (Williams & Lichtwardt 1990).
The depauperate gut flora in larvae of New Zealand Austrosimulium spp. appears to be a good example of the ‘enemy release hypothesis (ERH)’, where a colonising organism leaves behind its parasite load (e.g., Torchin et al. 2003; Moran & Krasnov 2010). Can this be regarded as another indication that New Zealand Austrosimulium dispersed from Australia?
Aspects of endemism of gut symbionts and species of host simuliid are discussed by Nelder et al. (2005).
Kingdom Chromista
Phylum Myzozoa, Class Aconoidasida, Order Haemospororida, Family Plasmodiidae
Leucocytozoon tawaki is a well-studied blood parasite of the Fiordland crested penguin (Eudyptes pachyrhynchus) (Fallis et al. 1976; Allison et al. 1978; also see Craig & Crosby 2008) and is vectored by A. australense, A. dumbletoni, and A. ungulatum. It is known (Allison et al. 1978) that L. tawaki can be vectored to blue penguins (Eudyptula minor (Forster)) and domestic chickens (Gallus gallus domesticus). Allison et al. (1978) reported that A. ungulatum with heavy infections of L. tawaki had shorter life spans than those with lighter infection, so the organism does appear to have an effect on its host.
On the Anglem coast of Stewart Island and nearby Codfish Island, a leucocytozoon, unrelated to L. tawaki, is known (Hill et al. 2010) to cause considerable mortality of chicks of the endangered yellow-eyed penguin Megadyptes antipodes (Hombron & Jacquinet). From the moment they hatch, chicks of the penguin are badly bitten by simuliids. Identity of the flies is, however, unknown as is any definitive information about transmission of the disease.
Kingdom Animalia
Phylum Nematoda, Class Adenophorea, Family Mermithidae
There is only one literature record of a mermithid nematode parasitising New Zealand Austrosimulium, that by Poinar (1990). In Australia, similarly, there has also been only one report — that by Hunter & Moorhouse (1976a) on the feminisation of A. bancrofti males associated with mermithisation, a condition well known elsewhere in simuliids (Crosskey 1990; Stanfield & Hunter 2010). Although mermithids were reported from larvae, pupae, and adults of A. bancrofti, no details of the nematode were given.
Poinar’s (1990) account was for 2 new genera of parasitic mermithids, Austromermis namis Poinar from larvae of Austrosimulium multicorne, and Blepharomermis craigi Poinar from larvae and pupae of Neocurupira hudsoni Lamb (Blephariceridae). Dumbleton had, however, collected larvae of A. ungulatum containing nematodes (Fig. 503) at the summit of Porters Pass (most likely from the Foggy Peak stream) on 4 March 1958. A tube with 7 larvae, of intermediate instars, with a label in Dumbleton’s handwriting, was overlooked since he made no mention of it in his 1973 work. The larvae are currently in the ethanol collection of NZAC, Auckland. Although badly bleached, an examination of 2 of the larvae showed that the mermithid is that described three decades later by Poinar. So far, this collection by Dumbleton and that by Poinar are the only 2 records of nematodes in New Zealand simuliids. We found none during this study.
Poinar’s mermithids were from Cave Stream, a shallow, fast, cold-water, stream in the Craigieburn Forest region, west of Christchurch. He tracked Austromermis development in the larvae of A. multicorne (then as A. multicorne multicorne). The parasitised simuliid larvae could be easily seen in the stream because of their white abdomens, an uncommon appearance. Infection rates in mid-January varied from 3–30%, depending on where the sample was taken. Because mature nematodes were found in smaller larvae in mid-January, Poinar concluded that there was a single generation of Austromermis namis per year. He further noted that the early infections by A. namis in November were in the Malpighian tubules of the simuliid larvae and that indicated a probable “per os” (by mouth) route of infection.
At the Craigieburn locality the mermithids had completed their development by the end of January when water temperature was 10–12 °C. At the higher altitude (ca 950 m a s l) at Porters Pass mean water temperature would be closer to 9°C (Mosely 1982), hence it is not unexpected that the mermithids collected by Dumbleton in 1958 were still present in A. ungulatum larvae in early March.