Biology, images, analysis, design...
|"It has long been an axiom of mine that the little things are infinitely the most important" |
Mites parasitizing and predating aphids:
Some less well known aphid natural enemiesOn this page: Larval velvet mites Host-range of velvet mites Site of attachment Pattern of dispersion Effect of mites on aphids Are these mites or eggs? Silver fly larvae (Chamaemyiidae) Whirligig mites (Anystidae)
It should come as no surprise that aphids have their own parasites - as the adage goes: "big fleas have little fleas, upon their backs to bite 'em, and little fleas have lesser fleas, and so, ad infinitum". The tiny Aphidius wasps which parasitize aphids and mummify them are well known, and are marketed commercially to control pest aphids. But these are correctly known as parasitoids, since they invariably kill the host. Some mites, on the other hand, are true aphid ectoparasites. Larval velvety mites suck the haemolymph of aphids, only killing them when there are several feeding on the same aphid. These mites are much less well known, but in Britain we have found them parasitizing many aphid species. Still less well known to aphidologists are several types of predatory mite, such as Anystis - a voracious predatory mite which seems to specialise in the aphids other predators don't want.
Larval velvet mites (Trombidiidae) - a neglected biological control agent?
Let us begin with a familiar picture: an adult red velvet mite (family Trombidiidae). Trombidiid mites have a dense coat of setae (which give them a velvety appearance), short chelicerae and a single set of prodorsal sensilli. Red velvet mites are usually red, relatively large (up to 2.5 mm - some of their tropical kin are rather larger) and found in soil litter.
Adult velvet mites, such as the one above, predate a variety of soil organisms; their larvae are parasitic on various arthropods. Three genera (Allothrombium, Podothrombium and Monothrombium), are aphid specialists. The large sowthistle aphid Uroleucon sonchi (below) is unusually heavily infested with larval velvet mites.
There has been interest for many years in using this mite for biological control (Howard, 1918 ). Another use of velvet mites - as a means to increase sexual desire (see Indian Viagra) - but we cannot recommend it...
Velvet mites' host range
We find trombidiid mites parasitizing many different species of aphids, including the black foxglove aphid Aphis armata, black bean aphid Aphis fabae (see below first), ivy aphid Aphis hederae, small bramble aphid Aphis ruborum, large blackberry aphid Amphorophora rubi (see below second), plum-thistle aphid Brachycaudus cardui, willow-parsnip aphid Cavariella theobaldi, meadow-sweet aphid Macrosiphum cholodkovskyi, rose aphid Macrosiphum rosae, blackberry-grass aphid Sitobion fragariae, large cat's ear aphid Uroleucon hypochoeridis, and Uroleucon cichorii.
We have also found trombidiid mites on poplar shoot aphid Chaitophorus populeti (see below first), poplar leaf aphid Chaitophorus populialbae, and rufous willow bark aphid Pterocomma rufipes (see below second). In fact, trombidiid mites probably occur on every type of aphid. Although, interestingly, we have yet to find any on large conifer aphids Cinara species - despite having photographed aphids of that genus many hundreds of times.
Whilst the Chinese aphid mites are relatively well known, thanks to the pioneering work of Dr ZhiQuiang Zhang, very little has been published on trombidiid mites on aphids elsewhere in the world. In Europe Goldarazena & Zhang (1999) recorded the host range of two species of Allothrombium (Allothrombium pulvinum and Allothrombium monochaetum) in a meadow in northern Spain. Both species were present on the black bean aphid Aphis fabae. Allothrombium monochaetum also occurred on the peach-potato aphid Myzus persicae, potato aphid Macrosiphum euphorbiae, and English grain aphid Sitobion avenae. Allothrombium pulvinum also occurred on the alder buckthorn-willowherb aphid Aphis frangulae, peach-potato aphid Myzus persicae, Therioaphis luteola and a Brachycaudus aphid species.
So far in England only two species of Allothrombium seem to have been recorded (Allothrombium pulvinum and Allothrombium fuliginosum) each of which probably parasitize a range of aphid species.
Where mites attach themselves to aphids
The pictures above give a reasonably good idea of the most frequent type of attachment sites of trombidiid mites - namely laterally or ventrally on the thorax. Zhang (1991) found that the ventral surface was most preferred by mites, whereas the dorsal surface was least preferred. The majority of the mites (61.5%) chose to attach to the thorax, whereas only 1% of the mites attached to the head. The choice of the thorax may be because the aphid hosts are unable to defend this site. Presumably they can knock mites off the head and abdomen using their legs.
On one of the rare occasions when we found a mite dorsally, there was a mite scar on the aphid (see picture below) - possibly the result of predation of a previous mite in that position.
This mite was identified as the newly-described species Podothrombium proti (Haitlinger, 1994 ).
Mites' pattern of dispersion
We can look at the pattern of dispersion in two ways - firstly per unit of plant surface area and secondly per host unit. When first hatched, trombidiid mite larvae do not disperse immediately, but remain aggregated on the plant near where the eggs were laid (Robaux, 1974 ). This probably explains why one encounters marked spatial aggregations of such mites, as shown below. The larvae of two species that parasitize aphids (Allothrombium pulvinum and Allothrombium ovatum) are known to congregate for about a day before dispersing.
Zhang et al. (1993) looked at the dispersion patterns of Allothrombium pulvinum in terms of the number of individuals per host unit. On Aphis gossypii (on cotton, in Jiangsu Province, China) the dispersion indices (variance-to-mean ratios) of larval mites per host were greater than 1: indicating that the mite parasites were overdispersed (aggregated) among their aphid hosts. The overdispersion appears to result from the mites preferring already parasitized aphids to unparasitized ones (Zhang, 1991 ).
Effect of velvet mites upon aphids
The impact that mites have on their aphid hosts is reviewed by Zhang (1998). The effect of larval trombidium mites on an individual aphid depends on the parasitic mite load and the age/size of the aphid. The larvae of Allothrombium pulvinum can kill an adult black bean aphid (Aphis fabae) in 3 days when the mite load is two or more. With only one mite, the reproductive rate of adult aphids is decreased and the development of nymphs is arrested. For larger hosts such as the pea aphid Acyrthosiphon pisum, five larvae of Allothrombium pulvinum kill about 50% of adult hosts in 4 days.
Allothrombium larvae have been shown to be important early-season natural enemies of the cotton aphid (Aphis gossypii) in cotton fields. They limit aphid population growth early in the growing season, when other natural enemies are still rare in cotton fields. Other natural enemies only increase in abundance later in the season. Recent studies on another species of velvet mite, Allothrombium ovatum, in China have shown similar results. A study in Germany found that numbers of Allothrombium fuliginosum larvae were too low to have any effect on rose-grain aphid Metopolophium dirhodum populations.
Are these mites - or are they midge eggs laid on aphids?
Identifying aphid predators from photographs is not always easy. When we first observed the orange 'capsules' (below) stuck to the back of aphids, we thought they were probably predatory midge eggs: They seem to have a very regular pattern of distribution on the individual aphids - one finds that each aphid usually has just one 'capsule'. Also they are nearly always positioned dorsally, on the aphids abdomen.
However, most of the literature on Aphidoletes states explicitly that these predatory midges lay their eggs on the plant in the vicinity of aphid colonies (as above ) - not directly on to the aphid. We then considered whether these orange 'capsules' could be mites: one suggestion was that they were tarsonemid mites. However, close examination of our images revealed no indication of any legs at all, so we were back to eggs (assuming they were not mite prelarvae). Eventually we found a reference to cecidomyiids sometimes laying eggs directly on aphids (Mamaev & Krivoshena 1993 ) so, until we find evidence to the contrary, we surmise those orange capsules are cecidomyiid eggs.
Which cecidomyiid species remains unknown. Perhaps this is previously unrecorded behaviour by one of the three commoner European species - Aphidoletes aphidimyza, Aphidoletes urticaria or Monobremia subterranea. We will just have to find some more, rear them out, and identify the adults.
Silver fly larvae (Chamaemyiidae) - little known dipteran predators of aphids
We did debate briefly whether the eggs laid on aphids shown above could be eggs of another relatively 'unresearched' predator of aphids - namely a silver fly. The best known genus is Leucopsis. The picture below shows a larval chamaemyiid sitting quietly in a colony of the mealy plum aphid Hyalopterus pruni.
Once again, the limited literature on silver flies suggest that, like other aphid predators, chamaemyiids lay their eggs on the plant amongst aphid colonies, not directly on individual aphids. Albeit a digression from this page's main topic, we included the picture of this larva partly because there seem to be very few pictures of chamaemyiid larvae on the web, and partly because they are a fascinating group.
Much like Aphidoletes larvae, chamaemyiid larvae use a what Fréchette et al. 2008 described as "furtive predation strategy". They sit around within the aphid colony occasionally consuming an aphid, but cause little defensive reaction among the aphids - or disruption of the colony.
So what about predatory mites? - whirligig mites (Anystidae)
All the mites we have looked at above are parasitic in their larval stages, but those below are predatory both as larvae and adults. This mite was found predating an ivy aphid Aphis hederae. Note the exudations from its siphunculi.
This is Anystis mite looks very much like Anystis baccarum, popularly known as the "whirligig mite" owing to its spiral-like running fashion. Anystis baccarum has been found feeding on the apple-grass aphid Rhopalosiphum oxyacanthae, and is considered a potential biocontrol agent in UK apple orchards (Cuthbertson & Murchie, 2008 ).
We have also found Anystis mites predating grey waxy pine needle aphids Schizolachnus pineti (see picture below). In Canada, Grobler 1962 found that Anystis agilis was an important predator of Schizolachnus pini-radiatae, causing a 30-50% mortality among overwintering eggs.
As with the ivy aphid, there are exudations from the siphunculi of this Schizolachnus pineti. Such exudations would normally indicate release of an alarm pheromone, but in this case there was no indication of any response from the other aphids in its colony. Waxy pine needle aphids show what is called "contact clustered aggregation" - in which the bodies of individuals are in mutual contact in rows along the pine needles. Schizolachnus pineti use touch to communicate the presence of a predator, and there is no evidence for an alarm pheromone (Kidd, 1982 ).
Interestingly, other predators are seldom observed attacking either the ivy aphid or the waxy pine needle aphid. The former is protected by toxins from its foodplant, and the latter is protected by its wax covering. Perhaps whirligig mites are more successful predators when they are not having to compete with a host of other predators, which may well predate the whirligig mites along with their normal prey.