By Matthew Cobb

No Kipling this time, but some science from the 1920s. After the Beetle Handle post went round Twitter, ‘Wikispecies Editor’ @stho002 (aka S. E . Thorpe) who’s based in Auckland NZ, tweeted about this amazing and very tiny parasitoid wasp, Inostemma boscii, which has a stupendous ‘handle’ arching above its thorax. As can be seen in this photo taken by Thorpe and posted on Wikispecies, the structure is so long that the wasp even has a groove in its head to make space for the handle:

A clue as to what on earth this outgrowth might be – sadly, it’s not a handle – is given by the observation that this is a female wasp. The ‘handle’ in fact contains the ovipositor, which is curled up in the handle when it’s not in use.

Here’s a rather nice drawing of an I. boscii wasp in flight, taken from here. Notice the lack of veins in the wings, with the exception of that rather odd short ‘strut’ in the forewings:

The Wikispecies page for I. boscii  includes a reference to a 1927 paper in the Bulletin of Entomological Research by  Dr J G Myers of the steampunk-sounding ‘Imperial Bureau of Entomology’. (The Bureau was apparently later transformed into CABI, the Commonwealth Agricultural Bureaux International (yes I know it’s not grammatical).)

Myers begins his excellent paper thus:

In the summer of 1925 I was commissioned by the New Zealand Government Department of Agriculture to attend the Second Imperial Entomological Conference and afterwards to enquire in Europe into the possibilities of natural control for some nine primary pests of the Dominion. As the pear leaf-curling midge presented at the same time the most pressing and the most promising of these problems it was decided to make a personal study of this insect, as intensive as time would permit. The various natural enemies discovered having now been determined by the kindness of Dr. Ch. Ferriere, through the Imperial Bureau of Entomology, the biological data acquired are here published for the first time.

In other words, he had to explore potential predators on the pear leaf-curling midge in Europe, which might be introduced to deal with the (apparent) problems created by the Perrisia midge for New Zealand’s pear production. However, it turned out that the midge was not much present in Europe, and certainly wasn’t a pest for pear growers. (It appears that Inostemma boscii is currently particularly useful in controlling pests of oil-seed rape, whose bright yellow flowers mark the European countryside in springtime.)

Nevertheless, Myers went to France in the summer and was able to observe both pest and parasitoids of the Platygasterid family (Inostemma and Misocyclops). His description of the behaviour of Inostemma boscii is a good bit of natural history:

This energetic little black wasp was first observed at Versailles on 16th July, when it was ovipositing during hot and brilliant sunshine in the eggs of Perrisia. Misocyclops was busy at the same time, as were also the midges themselves. The two Platygasterids were in the proportion of perhaps one to every eight midges, while among themselves the two species seemed in about equal numbers. On most subsequent occasions, however, and especially towards the middle of August, the Misocyclops was much the more abundant. Both species were observed on such of the intervening dates as the weather permitted. Inostemma takes from half a minute to more usually a minute for each oviposition. This is much slower than Misocyclops, but much quicker than I. piricola of the pearfruit midge, which Marchal (1907) found to require a quarter or half an hour or even longer. The long period is probably associated with the much longer ovipositor in Inostemma than in the other genus, and this in its turn with the extraordinary dorsal horn which sheathes the ovipositor and renders Inostemma  unique among insects.

“Unique among insects”! I’ll say! Some close observation revealed how the female Inostemma gets to work:

Before beginning oviposition Inostemma boscii  explores assiduously, burrowing in between the leaf buds and under the matted hairs which clothe them. While both Inostemma and Misocyclops  exploit the eggs in the young buds—the usual laying-place of the midge — Inostemma  is often seen also on the older leaves, which have begun to unroll along the midrib, but are curling more tightly on the margins as a result of midge attack. In such situations I never saw Misocyclops.

When Inostemma oviposits, as soon as the tip of the abdomen touches the point of application, the dorsal horn, which has been lying closely appressed to the dorsum of the thorax, so as to be almost invisible to the naked eye, and to the lens appearing as a mere ridge, jerks violently upwards and retains this position during the act, with a wide space between its apex and the pronotum. It is then easily visible to the unaided eye.

The next bit is particularly intriguing, as the parasitoid targets the midge maggot’s brain, and either lays only one egg, or only one survives:

this Inostemma apparently always oviposits in the embryonic tissue which will later form the brain of the larval midge. But in spite of repeated examinations I was not able to discover the egg until considerable development had taken place. Even then it was by no means easy to distinguish, but by dint of examining many hundreds of larvae I eventually found eggs in one or other of the lobes of the supra-oesophageal ganglion. This examination was carried out day after day, but in no case was the Inostemma embryo found to have passed the blastula stage. All stages up to this were encysted in the brain, the egg having increased greatly in size (fig. 1). In the case of I. piricola Marchal found several more often than one in the same brain, but in the present species not one larva, of the hundreds examined, showed more than one such encysted egg. I think this is a constant difference; but the final result is the same in the two species, since only one I. piricola survives in each host larva.

Here is Figure 1 – the maggot’s noses (my favourite subject) are the two horn-like structures on the right. The shaded comma-shaped blobs to the left are its mouth hooks. The encysted egg is circular structure on the left, I think.

The final bit is pretty gruesome, in classic parasitoid fashion, and is reminiscent of “Alien,” for the larva hatches from the brain and makes its way through the body of the midge maggot, using its ‘huge curved mandibles’ (yikes!) as it goes:

At last, on 12th August, a midge larva was obtained with the cerebral cyst of Inostemma greatly enlarged and containing a larva at the primary or cyclopoid stage rendered familiar by Marchal’s work. The next day the cyst burst and the cyclopoid larva wandered into the body cavity. Even before the larva emerged from the cyst the huge curved mandibles were seen to work vigorously. Occasionally one would be opened completely so that its apex was outside the outline of the body, while the other was shut like the blade of a clasp knife. By the time the larva was mounted, 24 hours after discovery, it appeared to be ensconced in the gut of the host.

Here is Figure 2, showing the mandibles of I. boscii:

Can nothing stop this terrifying parasitoid? Myers concludes that I boscii seems pretty dominant, with no parasitoids attacking it (‘hyperparasites’), and its only predator being a ‘small spider’:

To sum up the life-history data, it seems that oviposition may occur at almost any time from the middle of July to the middle of August, but that development has apparently proceeded no further than the primary larva by the latter date. This would suggest that the later stages are passed in the wintering larvae of the host; but such an explanation is open to many objections. On the other hand it is almost impossible that I could have missed seeing the older larvae of Inostemma had they been present in my material, since I had learned to detect with facility the much less conspicuous early stages.

No signs of hyperparasites were found, nor do I think it probable that any exist. Possibly, as in the case of the related species attacking Contarinia, Inostemma sometimes lays in a midge ovum containing already a Misocyclops (Platygaster) egg or vice versa ; but I saw no instance of this. Cases where Torymus abbreviatus parasitised larvae already attacked by Inostemma were several times observed. The only direct enemy of Inostemma was a small spider which built its web on the pear shoots and seemed frequently to catch it. Inostemma apparently needs for ovipository activity hotter weather and brighter sunshine than any of the other Perrisia parasitoids.

And that, O Best Beloved, is how the wasp got her handle.

____________________

Reference:

Myers, J.G. 1927. Natural enemies of the pear leaf-curling midge, Perrisia pyri, Bouche (Dipt. Cecidom.) Bulletin of Entomological Research, 18(2): 129-138.doi: 10.1017/S0007485300019799