Dibrachys pelos (Grissell) is an occasional gregarious ectoparasitoid of Sceliphron caementarium (Drury). We report the second record of this host association, collected in western Nebraska, and present results of laboratory experiments on host suitability and utilization. When D. pelos was reared alone on prepupae of 6 possible hosts, 4 proved entirely suitable: the mud dauber wasps Sceliphron caementarium and Trypoxylon politum Say, and two of their parasitoids, a velvet ant, Sphaeropthalma pensylvanica (Lepeletier) and a bee fly, Anthraxsp. On these hosts D. pelos completed development in 2-4 weeks, with average clutch sizes of 33-57, of which 24.7% were males. The other two hosts tested, the flesh fly Neobellieria bullata (Parker) and the leaf-cutter bee Megachile rotundata (Say), proved marginal, with very few adult progeny produced. When reared on these same 6 hosts with the addition of a competing parasitoid, Melittobia digitata Dahms, D. pelos fared poorly, being the sole offspring producer in at most 30% of the trials (on Anthrax hosts) and failing to prevail at all on T. politum hosts. Comparative data on host conversion efficiency indicated that M. digitata was more efficient than D. pelos on every host except Anthrax.
Mud dauber wasps (Hymenoptera: Sphecidae) of the widely distributed genera Trypoxylon and Sceliphron share a complex ecological web of inquilines that either parasitize them or use their nests (Matthews 1997). Habits, prey, and inquilines are particularly well known for the organ pipe mud dauber, Trypoxylon politum Say (Barber & Matthews 1979; Brockmann & Grafen 1989; Cross et al. 1975; Molumby 1995; Volkova et al. 1999) and the yellow-and-black mud dauber, Sceliphron caementarium (Drury) (Shafer 1949; Hunt 1993).
In addition to heavy parasitism by Melittobia (Hymenoptera: Eulophidae) wasps and sarcophagid and bombylid flies, both mud dauber species also have other parasitoids that are less commonly encountered (Matthews 1997a). One of the latter is Dibrachys pelos Grissell (Hymenoptera: Pteromalidae) (Fig. 1a), an ectoparasitoid apparently distributed across North America (Grissell 1974) but infrequently collected. The only published record of D. pelos as a member of the mud dauber “community” is that of Grissell (1974). Despite an extensive survey of trap-nesting wasps and bees and their inquilines (mainly from the eastern United States), Krombein (1967) found no associated Dibrachys species. In our own wide-ranging collections of mud dauber nests east of the Mississippi River and particularly in the southeastern US over the last 20 years, we have never before found D. pelos.
Grissell (1974) reared this species on prepupae of S. caementarium and other hosts, but little is known of its natural host preferences or possible competition with other parasitoids. Elsewhere, other Dibrachys species have been reported to parasitize various families of Hymenoptera and Diptera (Floate et al. 1999; Smith & Rutz 1991; Urban & Eardley 1995; Whiteman & Landwer 2000), suggesting that D. pelos may be an opportunistic polyphagous parasitoid capable of attacking a variety of host species.
Field collection of a Sceliphron caementarium nest that was parasitized by D. pelos enabled us to investigate the latter species’ ability to parasitize other potential hosts. In order to better understand its apparent rarity as a parasitoid of mud dauber wasps, we also staged interspecific competition studies with Melittobia digitata Dahms, one of the most common parasitoids of mud dauber wasps.
Materials and Methods
Three cells of a Sceliphron caementarium nest collected by RWM at Lake McConaughy, Keith Co., Nebraska on June 21, 2003 contained pupae and recently emerged adults of Dibrachys pelos. These were brought to our laboratory at the University of Georgia, Athens, GA and reared for one generation on S. caementarium prepupae.
To investigate relative suitability of additional common potential hosts, individual 2-day-old mated female progeny from this D. pelos culture were placed on prepupae of 5 species known to be acceptable hosts for M. digitata: T. politum Say, the leaf-cutter bee, Megachile rotundata Say (Hymenoptera: Megachilidae), the flesh fly Neobellieria bullata Parker (Diptera: Sarcophagidae), the velvet ant Sphaeropthalma pensylvanica (Lepeletier) (Hymenoptera: Mutillidae), and a bee fly Anthrax sp. (Diptera: Bombyliidae). The first 3 species have been routinely used as hosts in other studies on M. digitata (González & Matthews 2002; Silva-Torres & Matthews 2003) and are available readily; the last 2 species are themselves parasitoids of T. politum (Cross et al. 1975; Matthews 1997a; Matthews 1997b). Concurrently, parallel cultures of D. pelos were maintained on S. caementarium. Ten replicates of each host species were used in all experiments except for Sph. pensylvanica, for which only 3 prepupae were available. All cultures were maintained at 25°C, 65% RH. Development time, progeny production, sex ratio, and host use (suitability) were recorded.
To investigate potential interspecific competitive interactions, an additional 10 replicates were concurrently established on each host (except Sph. pennsylvanica due to limited availability). For these we simultaneously placed one mated 2-day-old female each of D. pelos and M. digitata on the host, and maintained these under the same conditions as the other cultures. Outcomes of these competition experiments were scored as won (only D. pelos adults emerged), lost (only M. digitata adults emerged), or coexistence (adults of both parasitoids emerged). Number of adult progeny emerging and their sex ratio, were also recorded. We did not conduct a parallel series of intraspecific competition experiments (2 females of D. pelos on each host).
As one indicator of the relative suitability of the various hosts, host conversion efficiency values (analogous to feed conversion efficiencies for poultry or pork) were calculated for both D. pelos and M. digitata. To do this, samples of 10 males and 10 females of each parasitoid species were individually weighed on a Mettler® balance and the average weight of a single female and male of each species was determined. Ten individuals of each of the various hosts were also weighed to obtain an average host weight. The average number of males and females reared from each host when each of the parasitoids were alone was multiplied by the individual wasp’s average weight, this being apportioned according to the average sex ratio obtained when reared alone on the respective hosts. This value was then divided by the average host weight and the result multiplied by 100 to give a percent, the host conversion efficiency.
Results and Discussion
Host Suitability and Development Time
Grissell (1974) reported that D. pelos laid eggs on prepupae of Sceliphron, as well as Ancistrocerus and Euodynerus (Hymenoptera: Vespidae, Eumeninae), and Megachile pacifica, but completed development only in the first 2 hosts. In our experiments, D. pelos oviposited also on at least some of all hosts offered (Table 1).
The most successful development occurred with 4 taxonomically diverse but ecologically related species--the mud daubers S. caementarium and T. politum, and their parasitoids, the velvet ant Sph. pensylvanica and the bee fly, Anthrax sp. (Fig. 1c); all individuals (100%) of these host species were parasitized successfully, as defined by emergence of D. pelos adult progeny. Development times on these 4 preferred hosts were quite similar, requiring 1-3 days for eggs, 7-14 days for larvae, and 7-12 days for pupae, with the total development time ranging from 16-27 days. These ranges for each developmental stage are consistent with data for D. pelos on S. caementarium reported by Grissell (1974).
Although some eggs were laid on Megachile rotundata and N. bullata hosts, most immature D. pelos perished, so that on average fewer than 4 adults eclosed from these 2 hosts (Table 1). Furthermore, the life cycle took significantly longer to complete on these “marginal” hosts. For example, whereas D. pelos started laying eggs on most hosts within 24 hours, oviposition was delayed for up to 4 days on N. bullata. Development was also strikingly slower on N. bullata at every stage with the result that adults emerged only after 24 to 36 days, compared to 16-27 days on the 4 preferred hosts. Development was also somewhat slower on Megachile, requiring from 19-31 days. Grissell (1974) attempted to rear D. pelos on Megachile pacifica, and obtained progeny on 19 of 71 hosts. However, 75% of the eggs laid on M. pacifica prepupae failed to complete development to adults. Similarly, we noted significant larval mortality on M. rotundata hosts and the few adult progeny obtained were on only 3 of the 10 host replicates.
Comparable data for the progeny of M. digitata on the same suite of hosts (except Sphaeropthalma, unpubl. data) showed that all hosts were acceptable with adults of both sexes reared from 100% of the replicates (n = 10 for each host).
Grissell (1974) reported male-biased sex ratios for D. pelos on Sceliphron and Ancistrocerus. In contrast, we obtained female-biased sex ratios in nearly every trial on every host (Table 1). These ratios appeared to vary with the host species. On the 4 most successful host species, D. pelos produced an average of 24.7% males; on the two “marginal” hosts, 43% were male. Overall, the smallest host species (M. rotundata) yielded the highest proportion of males (48%). The 13% on Neobellieria bullata is probably not representative, as it was based on very few individuals.
In our staged competition experiments with one female each of D. pelos and M. digitata on a host, the 2 females seldom coexisted successfully. Only in 2 replicates with Sceliphron, 1 replicate with Trypoxylon, and 3 replicates with Anthrax were hosts successfully shared, as defined by the subsequent appearance of adult offspring of both sexes of both species (Table 1, coexistence). Overall, D. pelos was the loser in the competition experiments, producing no adult progeny in 20 of the 30 trials with the three hosts preferred by females alone (Table 1).
These outcomes were not simply related to host size. Despite being the smallest of the 3 preferred hosts, Anthrax was the most likely to be shared (3 replicates), but D. pelos also was the outright competition winner in 3 replicates and the loser in 4 replicates. However, on Sceliphron sharing occurred in 2 of 10 trials; in 7 trials, Melittobia were the sole progeny to emerge as adults, and in 1 trial, only Dibrachys adults emerged. On Trypoxylon, the largest hosts, 9 of the replicates resulted in only Melittobia, and in only 1 trial did adults of both species emerge.
When D. pelos won the competition on Anthrax hosts the number of males emerging was not different than when alone (no competitor), but the number of females emerging was fewer than when alone (Student’s t-test, males P = 0.69, females P = 0.04). When both D. pelos and M. digitata adults emerged after competition for an Anthrax host, the number of D. pelos females was again fewer than when D. pelos was alone (Student’s t-test, P = 0.005), but not when compared to when it won outright (Student’s t-test, P = 0.24). Reduced numbers of progeny in competitive situations is not surprising since the host resource is not unlimited and, when shared, both host quality and quantity decline due to host feeding by each of the female parasitoids.
One straightforward reason why D. pelos suffers most from this competitive interaction was immediately apparent when larvae of M. digitata were observed feeding upon D. pelos larvae (Fig. 1d). Subsequently, emerged M. digitata were observed laying eggs directly upon pupae and even on newly emerged adults (Fig. 1b); in the former case the larvae developed into adult wasps, but larvae perished in the latter case. Depending on the host, M. digitata complete development to adults in 14-24 days (González & Matthews 2002), somewhat more rapid than D. pelos development in this study.
Differences in fecundity on these hosts may provide equally or more important explanations for this disparity. A single M. digitata female on a Trypoxylon host produces an average of 458 females and 13 males (unpubl. data), whereas a single D. pelos female produces the same number of males but about 10 times fewer females (Table 1). Similar disparities exist for the other hosts, although M. digitata is more broadly polyphagous (Dahms 1984) and successfully reproduces large clutches of progeny on both of the hosts that proved only marginally suitable for D. pelos.
In the 5 experiments (total from all hosts) where D. pelos “won” in competition against M. digitata, the proportion of D. pelos males increased substantially from that obtained for a female ovipositing in the absence of competition (Table 1). In the 6 replicates (total from all hosts) where adults of both parasitoids emerged, D. pelos’ sex ratios remained similar to those obtained for D. pelos females alone on hosts, although the proportion of males was elevated for Anthrax hosts. However, small sample sizes and low numbers of progeny in the competition treatments make it difficult to draw definitive conclusions.
So why does D. pelos appear to be relatively rare in field collections of mud dauber nests? In addition to its poor success in our staged competitions for hosts, it may be physiologically less efficient in converting host biomass to parasitoid progeny. To gain a perspective on this possibility, we compared the host conversion efficiency of D. pelos and M. digitata on each of the hosts used in these experiments (Table 2). Melittobia digitata were more efficient on every host tested but Anthrax. This suggests that perhaps the hosts we tested were less suitable for D. pelos development. Perhaps D. pelos is better adapted to twig-nesting wasps or some other unknown host, and its occurrence on S. caementarium was strictly opportunistic and facultative at sites not concurrently colonized by Melittobia. (In our extensive field collections of S. caementarium and T. politum nests over several years in eastern N. America, Melittobia is by far the commonest parasitoid found [unpubl. data].). In support of this it is notable that in the extensive sample of mud dauber nests taken from the same bridge in Nebraska where D. pelos was originally collected failed to turn up any Melittobia.
We thank Eric Grissell, USDA Systematic Entomology Laboratory, for identification of D. pelos. Alan Kamil and Robert Anderson, respectively, director and manager of the University of Nebraska Cedar Point Biological Station, aided us in numerous ways, including use of the station facilities during the field work in western Nebraska. This work was supported in part by a National Science Foundation grant to RWM.