Description & Statistics

Those Meloidae that have been referred to as parasitic have relationships with their hosts that are not a true parasitism, where individual larvae consume the body contents of the host and complete development thereon. In Meloidae, the host egg is consumed, and feeding thereafter is on the food mass stored in the cell by the parent bee. Early reviews on Meloidae behavior may be found in Balduf (1935) and Clausen (1940).

Clausen (1940) believed it important to note that in host preferences, the Meloidae are limited to the two widely separated groups mentioned above. Adaptations required to gain access to the host stages and to develop successfully upon them are markedly different. The lack of transitional forms, possibly on other hosts, is surprising. However, both host groups have in common the fact that stages attacked are most found in soil, but a few species inhabit wood galleries (Clausen 1940/62).

The overall economic value of Meloidae is open to question. Many species that develop in Locustidae egg pods destroy vast numbers of eggs annually, but the actual effect on the host population is not measured in most cases. The numerous species developing in the cells of bees may drastically reduce the size of colonies and thereby result in less complete pollination of surrounding plants. Although they do not attack the immature stages of the honeybee, in Europe the triungulinids sometimes congregate on the adults in such numbers that irritation and death ensues. This activity occurs so frequently that hive populations are depleted. The most serious injury inflicted by the family is by feeding of adult beetles on foliage and blossoms of cultivated crops, which sometimes results in severe injury. Several species of Horia develop at the expense of xylocopid bees, so that these may be considered beneficial. Horia maculata Swed. was introduced to Hawaii from Guatemala in 1934 for control of Xylocopa, but establishment was not confirmed (Clausen 1940/62).

There were over 2,000 species known as of 1993. They are abundant in desert and drier regions worldwide. The characters of these "blister beetles" are a filiform antenna; prothorax as wide as head or narrower; elytra often overlapping at base and separated at apex; tarsal formula 5-5-4. The body is medium-sized, elongate, cylindrical to robust. The color is often dull gray, brown or ochre, sometimes metallic green, blue or purple. They are often lightly sclerotized.

Meloidae adults are phytophagous and often destructive. Some adults apparently do not feed at all. Immature stages are predaceous or parasitic on the provisions and immatures of wild bees and in the egg pods of grasshoppers. The immature stages are not destructive. Predators of Locustidae are found mainly in the genera Zonabris (= Mylabris), Epicauta and a few species of Tetraonyx and Macrobasis. Just a few species are parasitic in the cells of various bees of the families Megachilide and Andrenidae, with the subfamily Nemognathinae being limited entirely to the latter hosts. Common species occurring on these hosts are found in the genera Apalus, Meloe, Nemognatha, Zonitis, Hornia, Sitaris and Tricrania.

Meloids have not been used with much success in biological pest control.

Biology & Behavior

DeGeer first noted the parasitic relationship with Hymenoptera in 1775 (cited by Clausen, 1940). Newport (1845) determined that the triungulinids were carried into bees' nests, later describing the older instars from cells of Anthophora. Newport (1846, 1853) provided more detailed accounts of the biology and behavior of Meloe cicatricosus Leach, parasitoid of Anthophora. Eggs are laid in soil, and transport of young larvae to the nest is on the bodies of adult bees. Fabre (1857) studied Apalus muralis Foerst, developing similarly in the cells of Anthophora. Larvae were found to be carried into the nest by adult bees, and the larvae transferred to the egg of the bee. Valery-Mayet (1875) studied Apalus analis Schaum. (= Sitaris colletis V.-M.), finding similar behavior. Riley (1878b,c) studying Epicauta vittata F. and other species of the genus, discovered the predatory group of Meloidae which feeds on eggs of Locustidae. He gave a full account of the striking metamorphosis undergone by the developing larvae. Extensive early accounts of the biology of Meloidae were those by Beauregard (1890), Cros (1910-1931), Roepke (1917), Milliken (1921), Parker & Böving (1924), Zakvatkine (1931, 1934), Verbeek (1932), and Ingram & Douglas (1932).

Oviposition.-- There is a relative short oviposition period of 2-4 weeks in most Meloidae. The gestation period is short, being only a few days, and in some cases eggs are laid on the day of adult emergence. The number of eggs laid varies among species. Species which develop on eggs of Locustidae usually produce a comparatively small number, ranging from <100 to several hundred. These are laid in batches in shallow burrows in the hard, dry soil of the host's breeding area. The burrows are filled and covered when egg-laying is finished. Eggs have an adhesive substance that keeps them together in a mass (Clausen 1940/62).

The number of eggs produced per female in species that attack bees is considerably larger, probably owing to the greater hazards encountered by young larvae before they reach the host cells. Meloe cicatricosus and M. autumnalis var. cribripennis Dej. lays 3,000 or more eggs. Other species approximate this number. A single batch of eggs of M. violaceus Marsh had 3-4,000, and several additional batches, containing smaller numbers, were laid by the same female (Cros 1931). One Meloe sp. was found to contain a total of 4,218 eggs when dissected (Newport 1851). Meloe majalis L. was found to lay several thousand eggs in each batch (Cros 1912, 1913). The first batch usually contained the largest number of eggs, while those following were successively smaller. Each batch represented the entire quantity of mature eggs in the body of the female at the time of egg-laying, and the abdomen consequently shrinks with this activity. A feeding period then follows, and the abdomen gradually becomes distended with more eggs.

There is not as much uniformity in the place of oviposition among species attacking bees as in the case of those attacking Locustidae eggs. Parker & Böving 1924) studied Tricrania sanguinipennis Say, developing in nests of Colletes. The egg batches were laid in the soil in the vicinity of the host nests. Meloe autumnalis cribripennis (Cros 1914), M. majalis, and Apalus muralis (Cros 1910) have similar behavior. However, Hornia nymphoides Esch. lays batches of eggs of several hundred in the gallery of Anthophora or in an old cell, and Horia debeyi Fairm. places them in the galleries of xylocopid bees (Cros 1913). In Nemognatha chrysomelina L., the eggs are found in small batches on the blossoms of Echinops spinosus, and hatching coincides in time with the complete opening of these blossoms (Cros 1912). Apalus rufipes Gory lays eggs upon the blossoms of Ballota hirsuta (Cros 1913). In the Nemognathinae, which are restricted to bees, there is less care for the safety of the egg masses than in other Meloidae (Parker & Böving 1924). Some species of Sitaris, Stenoria and Apalus lay their eggs in uncovered piles in small grooves in the host gallery, while others place them on leaves or stems of various herbaceous plants.

Egg incubation ranges from 4-6 days in the case of Zonitis immaculata Ol. to ca. 4 weeks in other species under normal temperature conditions. Much longer periods of up to 5 months are known for species that oviposit during colder seasons or that overwinter in the egg stage.

Triungulinid Larva.-- In most species the triungulinids disperse almost immediately after hatching, but in certain species there is a tendency for them to remain en masse for some time around the egg shells. Those of apalus muralis remain clustered together throughout the winter (Fabre 1857). Young larvae of most species are able to continue their activities without food for 3-4 weeks in summer. Rau (1930) found some colonies of triungulinids of Hornia minutipennis Riley in the area of Anthophora nests, and these remained intact for 2 weeks, even outdoors in June. However, when disturbed they attached readily to various objects.

Riley (1877) while observing H. minutipennis, found that adult females occurred in the immediate vicinity of the host cells, and that they seldom or never left the host gallery. Rau (1930) found that adult females within host cells may contain many young larvae. On examination, the cells were found to be unbroken except for a tiny hole to the outside, which was too small either for emergence or for entry of adult beetles. Clausen (1940) considered two possible explanations for the occurrence of triungulinids in these cells. Either the female was inseminated through the minute opening mentioned, or reproduction was thelytokous.

The way triungulinids gain access to the host stages on which they develop is variable in Meloidae. Those attacking egg masses of Locustidae are already in the soil in the breeding grounds of the host on emergence, and thus are dependent only on their own searching to find the egg masses. Similarly, species laying eggs in galleries or cells of bees have little trouble finding eggs. Tricrania sanguinipennis lays eggs in the immediate vicinity of a Coletes sp. nest. The young larvae could not be induced to enter the burrow directly, but rather entry was by carriage on the body of the adult bees (Parker & Böving 1924). Male bees play an important role in the economy of Tricrania. They appear in the field ca. 1 month earlier than females, and during this time the parasitoid larvae are active. Many of them attach to the bodies of these males. The latter are much more active than females, and for this reason most of the parasitoid larvae are eventually found on them. It is then necessary for the triungulinids to transfer to the females in order to reach the brood chamber. It is thought that transfer of triungulinids from male to female bees was at the time of mating, this conclusion being supported by their position on the bee body, as they are found primarily on the male venter and on the female dorsum (Clausen 1940/62).

Species that oviposit entirely apart from the host bee nests have triungulinids that frequently show a tendency to climb upward and thus may congregate in large numbers on blossoms. These are frequented by the bees in their search for food, and the larvae attach themselves and are thereby carried to the nest. Egg hatching in Nemognatha chrysomelina coincides with the complete opening of the blossoms and gives the maximum opportunity for the young larvae to attach to Anthidium and other bees that frequent the blossoms of this plant. Usually attachment is to the hairs of the carrier bee, although Cros (1927) found that larvae of Meloe cavensis Pet. cling to the abdominal intersegmental folds. Under the previously mentioned conditions, it is evident that the majority of triungulinids never succeed in reaching host cells. This loss is compensated for by the production of a proportionately larger number of eggs among species facing such hazards (Clausen 1940/62). Obstacles encountered by triungulinids incident to gaining entry into the bee cell in which development will occur are added to by the necessity for transfer from the body of the female bee to the newly laid egg. This is thought to occur at the time of oviposition by the bee. Probably, the triungulinids are brushed off against the cell wall and then reach the egg by their own effort. However, the triungulinid of Tricrania sanguinipennis is able to descend by a silken thread similar to that of many lepidopterous larvae, although in this case the thread originates from the caudal end of the body (Clausen 1940/62).

It is interesting that certain species of bees are immune to parasitization by Meloidae. Parker & Böving (1924) investigated this with T. sanguinipennis, that develops in the cells of Colletes rufithorax Swenk. but does not in cells of Andrena perplexa Smith, even though the nests of the two species are mixed. It was especially puzzling because triungulinids were found on the bodies of Andrena females. It was explained on the basis of different oviposition habits of the two bee species. In Colletes, the egg is attached to the side wall of the cell just above the food mass. The parasitoid larva is thus able to reach it without contacting the food mass. However, the egg of Andrena is placed on end in the center of the food mass, never in contact with the cell walls. It is almost entirely immersed in thin, watery honey, and any larvae that enter the cell are effectively prevented from reaching the egg by the liquid barrier.

After accessing the cell of the host bee, the first act of the triungulinid is to begin feeding on the egg. This furnishes food for the parasitoid and at the same time eliminates the host, leaving the food supply with which the cell is stocked for the later stages of the parasitoid. Normally the first molt occurs right after the host egg has been consumed, and ensuing instars are boat-like which enables them to float freely on the honey while feeding on it. That the host egg is not required for the development of the 1st instar larva is shown by the fact that one individual of T. sanguinipennis was reared to maturity only on pollen and honey. Even the 1st instar larva of several species readily accepted honey as well as host eggs and larvae (Cros 1908-1935). Often several triungulinids access a single cell, but in no case does more than one reach maturity. The youngest individual because of a greater agility and aggressiveness, readily overcomes those which have molted or which have fed sufficiently to render them sluggish.

Larval maturity is usually attained in the cell to which the larva first gained entry. Exceptions were recorded by Cros who found that Meloe autumnalis cribripennis may consume the contents of 6 cells during its feeding period, and that the 4th instar larvae of Nemognatha chrysomelina and Zonitis sp. penetrate into a second cell to complete their feeding. After consuming the food in one cell of Megachile sculpturalis Smith, the larva of Z. pallida F. penetrates the resinous partition of the adjoining one, feeds extensively on the mature larva contained therein, and eventually returns to its original cell, after which it repairs the break in the partition (Iwata 1933).

Species that attack egg masses of Locustidae are much more simple in feeding habits because the food consists only of eggs and a sufficient number are available in each mass to provide for the larva's full requirements. Only 1-2 eggs are consumed before the first molt. Such larvae are wholly entomophagous during their feeding period, in contrast to those species developing at the expense of bees (Clausen 1940/62).

The first 4 larval instars represent the feeding period in most species, with the following two being inactive and serving as a resting period or to carry the species through winter or through periods of adverse temperature and moisture conditions. On completion of the scarabaeoid stage, the larva may either leave the egg capsule and transform to the following instar in a cell in the soil near by or it may go through its following transformations while still within the capsule. In some species the coarctate larvae is found in a vertical position, with the exuviae forming a pad about the abdomen's tip. The larva remains in this instar until suitable conditions for further development occur. Milliken (1921) noted the change of Epicauta sericans Lec. from the coarctate to the colytoid stage within a few minutes after the application of water to the body. The colytoid larva exhibits some degree of activity for several days.

Coarctate larvae represent a resting stage in which the species is able to withstand adverse conditions to a great extent. The integument is heavy, providing protection from winter conditions in some species and from extremely arid conditions in others. In other species a portion of these larvae consistently persist until the 2nd year, and in E. erythrocephala Pall. the diapause may extend over many years (Zackvatkine 1931, 1934). These larvae may, under certain conditions, also revert to the coarctate form several times. Milliken also noted this reversion in E. sericans Lec., and the occurrence of definite supplementary molts was verified. Locust egg predators among Meloidae show a perfect adaptation, unequaled in any other group of similar food habit, for life under the most adverse conditions (Clausen 1940/62).

Among bee-attacking meloids, the 6th instar larva is enveloped by the unbroken 4th and 5th exuviae. As no food is taken after the 4th molt, the body volume decreases and the exuviae are thus not distended. In Nemognatha and Zonitis, the 6th exuviae are also retained complete, the pupa thus being enveloped in 3 larval skins, whereas in other genera of Nemognathinae the last larval skin is found as a pad attached to the tip of the abdomen. Retaining exuviae in these species is correlated with the less specialized forms of the last two larval instars, which are less adapted to adverse conditions than are those of the species subjected to free life in the soil (Clausen 1940/62).

Life Cycle

The duration of the life cycle varies considerably among species of Meloidae. In E. lemniscata in Louisiana, development from egg to adult is in 35-50 days and 3 generations may be produced annually (Ingram & Douglas 1932). Meloe majalis, Hornia nymphoides and E. vittata have 2 generations annually. Cerocoma vahli F. requires ca. 3 yrs. for its cycle (Cros 1919, 1924). Most species seem to have only one generation per year, which is correlated with the host cycle, with adults appearing during the oviposition period of the host. Apalus muralis has only one generation in 2 years, and in a number of species with an annual cycle, a high percentage of the larvae persist until the 2nd year or longer.

The various immature stages have variable durations, especially the nonfeeding periods. Incubation may range from 4-5 days to 6 weeks, and the triungulinid, after a free-living phase that may take 3-4 weeks, spends an additional week or more feeding before it molts. Three to 5 weeks seems a normal figure for the feeding stage. The caraboid stage takes about one week, and the two scarabaeoid stages total about the same. The coarctate or pseudopupal stages are those in which hibernation or diapause occurs most frequently. Thus, they vary greatly in length, even within a single species. Cros determined that in H. nymphoides a portion of the eggs of the 2nd generation may carry over to the following spring. Fabre found that the triungulinids of A. muralis persist in dense clusters through winter, while a few species hibernate in the adult stage in the host cells. The scolytoid stage of E. vittata takes only a few days, while in E. lemniscata it takes 11-21 days and is extended in Zonabris pustulata Thbg. to 10-60 days (Verbeek 1932). In Z. zebraea the pupal stage is 2 weeks and in Tricrania sanguinipennis it is 24 days.

Ingram & Douglas (1932) commented on an unusual variation in habit of E. lemniscata, which is a predator of grasshopper eggs. In the summer generation, development through the immature stages is rapid and the scarabaeoid larva changes directly to the pupa, omitting the coarctate and scolytoid stages. These two stages appear in the overwintering generation or under unfavorable conditions. These two instars are lacking entirely in Macrobasis immaculata.

For detailed descriptions of immature stages of Meloidae, please see Clausen (1940/62).

References: Please refer to <biology.ref.htm>, [Additional references may be found at: MELVYL Library]

Ingram, J. W. & W. A. Douglas. 1932. J. Econ. Ent. 25: 71-4.

van Dyke, E. C. 1928. Univ. Calif. Publ. Ent. 4: 395-474.