nomos XLIII

Organic Evolution in terms of the Implicate and Explicate Orders.

Part XXXIX

Hymenoptera (wasps, bees, ants) (Sequel)

The evolutionary diversification in the Order Hymenoptera in terms of Strategies (Sequel).

Delayed-parasitic (metaparasitic) phase

The development of the terebrant-egg-eaters at the expense of single eggs or of small clutches of the host led, as we already saw, to a diminution, sometimes extraordinarily, of the size of the representatives of whole families or even superfamilies such as the Chalcidoidea, (part of the) Cynipoidea, and Serphoidea. But, as is known, in the suborder Terebrantia there also exists a series of other large systematic units of which the representatives fall into the categories of medium or even large sizes. Already this one fact shows that the development of the terebrants of the latter group took place under other conditions and followed other paths than those of the egg-eaters. Meanwhile, in the archaic inquilinoid phase where the development of the terebrant-inquiline basically took place at the expense of the tissue of the original gall [taken over by the inquiline], and also in the case of feeding by the solitary larvae of the egg-eaters on large egg-clutches of the host, sufficient favorable conditions were created for the preservation of normal [i.e. complete] growth of the terebrants, which is visible for example in the Evanoidea.
Closer inspection of the life habits of the terebrants of this second group -- which are not egg-eaters -- indicates that the preservation by them of sufficiently large sizes, and sometimes even an increase of their growth, has been realized along different paths. One of the basic and, apparently, most widely distributed way to reach a complete satiation of the larva of the terebrant was the laying of the egg by the terebrant into the egg of the host or into its very young larva such that the latter did not die but continued developing normally. Now, the parasitic larva, finding itself inside the prey, successfully emerged from its egg, and was in the position to develop later at the expense of the more or less grown host-larva and sometimes even of the pupa of the host. Thanks to that the terebrant could attain proper growth. Not themselves being thus egg-eaters, the terebrants of the present group originated, as we must assume, basically from precisely those primitive egg-eaters that layed their egg not on the surface, but inside the egg of the host and after that began to develop at the expense of its more or less grown larva.
This strategy of delayed parasitism is very widely distributed. It is encountered also among the most primitive terebrants from widely separated systematic groups, as such pointing to their common origin from a single basic kernel of terebrants. Here are some examples.
The relict cynipoid Ibalia leucospoides Hoch, from the subfamily Ibaliinae, being a giant among the other cynipoids, lays its eggs in the egg-clutches of the blue horn-tail Sirex cyaneus F., which arranges them [i.e. its egg-clutches] in special deepenings in the bark of a tree. Having lowered its ovipositor into the egg-tunnel of the horn-tail, the Ibalia carries its egg over into the egg of the horn-tail or into its newly-born larva. In a single lowering of the ovipositor several eggs or larvae of the prey may be infected. The length of the period of the egg-stage of the Ibalia is very variable, namely from 6 weeks until almost a year. The third-instar larva of the Ibalia leaves the body of the horn-tail larva and finishes feeding outside it, and thus in its fourth instar the larva of the Ibalia does not eat anymore and may remain in this state for about a year. The overall length of the period of development of the Ibalia from egg to adult is not ascertained precisely, but one can safely assume that it takes no less than three years. All developmental stages of the Ibalia can be found in nature the whole year around. Thus, the connection of this primitive cynipoid with egg-clutches of the host is very indicative.
As regards another remarkable relict, Orussus of the superfamily Orussoidea, and of the family Orussidae, see next Figures, sometimes, on the basis of morphological features placed into the suborder Symphyta [Saw-flies and allies], but on the basis of behavior into the suborder Terebrantia, there are up to now [1966] only very fragmentary data.

Figure 1 : Orussus abietinus Scop. Length 7-15 mm. Family Orussidae.
(From BERLAND, 1951, in MALYSHEV, 1966)

Figure 2 : Orussus abietinus Scop. Family Orussidae.
(After CHINERY, in Elseviers Insektengids voor West-Europa, 1983 )

From a series of observations, reported by Burke, 1917, it follows that the species Orussus hopkinsi Rohw. and O. occidentalis Cress., living in America, are solitary parasites of larvae of beetles of the family Buprestidae (Buprestes), that live in the trunks of poplar, pine-tree, and other trees. As it was shown, the larvae of the mentioned Orussidae are encountered at different times of the year, but in each case the orussid larva was already fully-grown and was found in some cases next to a large larva of the beetle, and in other cases already next to its remains. So in spite of the different seasons only fully-grown orussid larvae were found and thereby next to their preys. This leads us to think of parasitism of young orussid larvae inside the larvae of the host, to which also Burke himself was partly inclined to think. This state of affairs is possible, however, only when the egg-laying by the parasite takes place in a corresponding young stage of development of the host, similar to what was noted above with respect to Ibalia. This fact, consequently, points to a significant biological affinity between both relicts, which do not seem to display any genealogic affinity.

From the obtained data Cooper, 1953, drew an unexpected conclusion : Because Burke had actually not seen young larvae of Orussidae in the act of feeding, their being carnivorous could not be taken as proved. According to Cooper it is more probable that the larvae of Orussus feed on wormhole left behind by the larvae of Buprestidae in their channels. But Cooper himself had not seen young, nor fully-grown larvae of Orussus, and had only seen two females of O. sayii Westw., having lowered their ovipositor into a dry log, lying in a storage depot, while eggs were not observed at the place where the Orussus did have their boring activity. Moreover, feeding on wormhole (which is in fact coprophagia [= feeding on excrements] ) is not at all typical to larvae of Hymenoptera.

In a similar way, evidently, we can see things also in a relict from the group of evanoids, namely from the third family of them, the Aulacidae. See next Figure.

Figure 3 : Aulacus striatus Jur. Family Aulacidae.
(After CHINERY, in Elseviers Insektengids voor West-Europa, 1983 )

As in the previous cases, we might think, here the small female Aulacus, not possessing a corresponding organization, cannot lay its eggs into larvae of horn-tails and beetles living deeply hidden in wood, but can infect only their egg-clutches placed in the bark of trees. This assumption becomes even more probable by the fact that the relatives of the Aulacidae -- the other evanoids, Gasteruptiidae and Evaniidae, are in their development, as was already mentioned, connected with the egg-stage of their hosts. It is therefore interesting to note that the European species Aulacus striatus Jur., having a length of 7-9 mm, is encountered on logs and trunks of trees, containing eggs of the horn-tail Xiphydria prolongata Goef., at the expense of which also this aulacid develops. But until now it is not known whether it does so as an internal parasite of the larva or as an external one.

Anatomic enquiry has shown that the ovaries of Aulacus striatus contained about 200 very small (having a length of 0.14 mm) eggs. These latter have a spindle-like shape and are provided with a short stalk.

In connection with this we must note that the representatives of the small, structurally primitive, family Stephanidae, parasitizing on larvae living in wood, are, with respect to their structure, especially of their head, extraordinarily similar to the Orussidae, as was pointed to earlier when the parasitism of the latter became known. Together with this the connection of the Stephanidae with certain representatives of the archaic group Evanoidea became clear, especially with the just mentioned Aulacidae.
This fact, consequently, also confirms the above mentioned connection of the Orussidae with the Terebrants, and absolutely not with the phytophags-horntails (Symphyta).
So it is very probable that also the Stephanidae, as to their method of egg-laying and as to their development at the expense of larvae living in wood of trees, belong to that same group of "delayed-parasitic" forms, as do the Ibalia, the Aulacidae, and the Orussidae. According to Berland, 1951, the Stephanidae are very abundant in the tropics where they parasitize on larvae of xylophags (wood-eaters), and in the same way, apparently (in Oceania) also on solitary bees. This latter habit brings them close to the evanoid terebrants of the type Gasteruptiidae.
Similar cases of development, not at the expense of the host-egg, into which the egg of the terebrant is laid, but at the expense of a later stage of the development of the host, are observed also in other families of terebrants represented by comparably large forms and also solitarily parasitizing. Thus, the braconids Chelonus (see next Figure) and Ascogaster -- solitary parasites of Lepidoptera [butterflies and moths] -- lay their eggs in the eggs of butterflies, but the development of their larvae only begins when the caterpillars, emerged from the infected eggs, have become almost fully grown.

Figure 4 : Chelonus annulipes Wesm. Family Braconidae.
(From CLAUSEN, 1940, in MALYSHEV, 1966)

Of the ichneumonoid terebrants, Oocenteter tomostethi Cush., of the family Tryphonidae, lays its eggs into eggs of the saw-fly Tomostethus, and develops only in the fully-grown pseudo-caterpillar of the latter. Also the ichneumonid tribes Bassini and Exochini develop at the expense of more or less fully-grown larvae of the host, although the first (Bassini) lay their eggs into eggs of flies, while the second (Exochini) lay them into the eggs of Tortricidae (butterflies).
These cases of infection of the prey through its egg-stage led Telenga (1952) be inclined to interpret as pointing to the overall evolutionary path of the formation of the egg-eaters from an initiatory parasitism by terebrants on host-larvae. Even apart from the non-correspondence of such an interpetation with the overall path of the origin and evolution of the terebrants, about which we have spoken earlier, the interpretation of Telenga contradicts a multitude of perfectly common facts where whole families of typical egg-eaters develop only in the egg-stage of the hosts and absolutely not infect their later stages. This interpretation (of Telenga's) is in fact untenable also when applied to the just brought forward individual cases. It is not understandable why the terebrant, of old, as is supposed, having developed at the expense of the larval stage of the host, should have abandoned its usual way of laying its egg into this stage (of the host), and would now begin to lay its eggs into the eggs of the host, at the expense of which it could not develop, being now in need, with the help of new special adaptations, of penetrating into precisely that stage of the host (the larval stage) in which it developed (evolutionarily) already earlier. For what, then, this series of new subtle adaptations was needed, when the effect, that had been achieved successfully without them, was left as mere something of old times? Al this, naturally, points to the absolute non-correspondence of this interpretation with reality.
It is remarkable that delayed development is observed sometimes also in cases of external feeding of the terebrant larva, whereby its egg may be laid outside the prey or even on the mere egg of the prey. Thus, the ichneumonid Habrocryptus graenicheri Vier. (of the subfamily Cryptinae) lays, similar to the above mentioned Grotea, its egg on the egg of the bee Ceratina dupla Say. But the feeding of the newly-born larva of this terebrant proceeds differently. It crawls over the egg, and later over the larva, of the bee without inflicting any harm neither to the egg, nor to the larva. Evidently it now enjoys only weak feeding which shows itself in its very slow growth. The parasite touches the surface of the egg or young larva with its mouthparts and sucks for a few moments and after this makes an interrupted upward movement with its head and crawls a certain distance to repeat all this again. As was found out, on the egg there are no traces of damage, it yields a normal bee-larva. The latter feeds on the honey supply and grows as do the other larvae of Ceratina [in the other cells]. The parasitic larva grows very slowly especially during the first 4-5 days of its life, but at about the eighth day it energetically attacts the half-grown larva of the ceratine and sucks its contents. This results in a very rapid growth of the parasitic larva. It soon penetrates into the neighboring cell, and in it destroys the host larva, and usually does the same with the third and fourth cells. Having reached an age of 13 days it begins making a cocoon.
In the European ichneumonid Hoplocryptus mediterraneus Tschek., also of the Cryptinae, according to observations of the author, the delayed method of development is shown still clearer. The elongated stick-like egg of this cryptin is laid next to the egg of the ceratine bee (Ceratina cyanea Kby., more rarely, C. callosa F.). The "campodeiform" larva [= a larva looking like Campodea, a primitive wingless insect (Apterygota, Diplura) of which the appendages (antennae, legs, cerci [tail-threads] are relatively long], of the terebrant, emerging from the egg, crawls under the egg of the ceratine but does not grab it. When from it emerges the larva of the bee, the parasitic larva crawls on it. It lies now motionless, with the head pointing forwardly, at one and the same place : On the dorsal side of the first segments -- above the heart of the prey. The larva of the ceratine feeds on the honey bun, grows, and attains a normal size, wheras the parasitic larva remains as it was before, and, evidently, does not disturb its prey. Having finished feeding the ceratine larva becomes restless. It begins to make violent movements with the head to all sides, trying to free itself from the parasite. But all its effords are in vain : It cannot reach the parasite with its jaws at precisely that time when the latter is in favorable position. And not until the prey, transforming into a prepupa, finds itself in a deep resting state, the parasitic larva goes over into the second instar and quickly destroys it, and after that yet two or three other ceratine larvae in their respective cells. Having finished feeding, the cryptine larva makes a parchment-like cocoon, two times longer than its own length, and in it finishes its development.
Cases are known in which in large preys, in such delayed parasitism, small braconids develop or ichneumonids, that is, terebrants from groups that are represented usually by medium and large forms. This is observed either in the case of the laying of several eggs by the mentioned terebrants into the young stage of the host, or in individual special cases. As an example of the first type of relationship may serve the reproduction as it takes place in a series of species of the genus Apanteles (family Braconidae), parasitizing in various caterpillars. [As an example of an Apanteles we here reproduce a picture of one of them, not belonging, however, to the species under discussion here] :

Figure 5 : Apanteles glomeratus, 4 mm.
(After SEVERA, in ZAHRADNIK, Thieme's insektengids voor West- en Midden-Europa)

Thus, Apanteles militaris Walsh. can drive 70 eggs a second into its prey. And -- [presumably as another example] -- from a dead head of a caterpillar can exit more than a thousand (1200) adult individuals of A. acherontiae Cam. Finishing its development, the larva of the Apanteles tears up the skin of the host and exits into the open where it builds its cocoon from spinnings, and sometimes, in addition, together [i.e. several individuals] encapsulate themselves with spinnings, forming cotton wadding-like clods, reminding of egg-cocoons of spiders.
It is interesting that the phenomenon of delayed parasitism is also observed among Chalcididae and Proctotrupidae, that is, terbrants generally represented by small or very small forms. Thus, the chalcidid Tetrastichus asparagi Cwf. lays its eggs into the eggs of the asparagus leaf-eater (beetles, Chrysomelidae) (Criocerus asparagi L.), but its larva attains maturity not until the larva of the beetle makes a cocoon in the earth and changes into a prepupa. From a single cocoon of the beetle 5 or 6 adult chalcids emerge.
According to not-published data of Clansy, communicated by Clausen, 1940, the proctotrupoid Helorus paradoxus Prov. (Proctotrupoidea, family Heloridae), see next Figure, lays its eggs into different-staged larvae of the lace-wing Chrysopa majescula Banus (Neuroptera, Chrysopidae).

Figure 6 : Helorus paradoxus Prov. Family Heloridae.
(From CLAUSEN, 1940, in MALYSHEV, 1966)

In this egg-laying the ovipositor is lowered from below or from the side of the abdominal segments of the prey, after which the female Helorus folds her legs together, and then the agitated prey may carry her for some time. The female Helorus only lays one egg in the larva of the lace-wing, but during the course of her life, lasting 4-6 weeks, she may lay about 50 eggs. The laid egg of the terebrant freely floats in the cavity fluid of the prey, but then, after minimally two days, from it emerges the peculiar "many-legged" larva, reminding us of the larva of Ibalia. The length of the period of the first instar of the larva of Helorus is very variable. Thus, when it happens to be in a larva of Chrysopa (lace-wing) that is hibernating, then also the Helorus-larva hibernates in it without undergoing further changes. In summer conditions, on the other hand, and where an almost fully-grown Chrysopa-larva is infected, the first stage of the Helorus-larva may last all in all 3-6 days. However, in this latter case, as well as in the first case, further development only takes place when the Chrysopa-larva finishes feeding and makes a cocoon. Soon after the first moult of the parasite, in the body of the Chrysopa-larva appear opaque white round particles, and with the increase of their number the fat-body and necessary tissues of the prey disappear. One must assume that the parasitic larva gives off some secretion that causes the disintegration of the host's tissues. About in the middle of the second instar of the Helorus-larva, lasting all in all 2.5-3 days, all movements of the prey stop, and it soon dies. The parasitic larva of the third instar eats the whole content of the prey in two days, and after that it exits from it into the open, leaving of the prey only its 4-5 posterior segments. In this way several generations of Helorus paradoxus develop in one year.
In clarifying the evolutionary history of the Heloridae we must emphasize that, on the one hand, their development proceeds basically at the expense of the fully-grown host-larva having already concluded feeding and having prepared for pupation, and, on the other hand, that the Helorus-larva, having emerged from its egg, is, when necessary, (when the transformation of the host larva has been delayed till the next season), able not to moult [i.e. delaying it], remaining in that same instar during a series of months. From all this we must assume that originally the Heloridae laid their eggs only into very young host-larvae (as in Ibalia) [that is, not from time to time into different stages of the host larva], and precisely then when they had to [evolutionarily] work out the ability of slow development and of feeding on a fully-grown prey [In the case of the host hibernating, the Helorus must be able to delay its development (as well)].
Among the relatives of the Heloridae -- other Proctotrupoid-like forms (Serphoidea) there exists, as already mentioned, a whole family of egg-eaters, Scelionidae, infecting eggs of various insects, including also Neuroptera -- Chrysopidae. But the Heloridae, evidently, never have been true egg-eaters. This is supported by the fact that they have preserved a rather archaic look, prompting Tillyard to reckon the fossil representatives of the Heloridae to be the starting forms of the remaining families of the proctotrupoids, and with them the gall-wasps and relatives and the Chalcids.
From what has been expounded it is, I think, clear that the delayed parasitism of the Heloridae gave them (as also the Ibalia's and, with respect to the way of development, relatives) the possibility of preserving a series of primitive features. From this viewpint the Heloridae show themselves as a side-branch of the serphoid terebrants, having been sepated early on from the basic trunk, and still living today. All this indicates how widely the discussed way of parasitism was applied, although the basic lines of its development originated from a common kernel -- from the delayed parasitism in its archaic inquilinoid stage.

With all this we have concluded the exposition of the Delayed-parasitic (metaparasitic) Phase of hymenopterous evolution.
In the next document we will expound the Vagabondish-parasitic (planidial) Phase.

e-mail :