Comprehensive Description
provided by Smithsonian Contributions to Zoology
Podium rufipes (Fabricius)
A single female of this cockroach-preying wasp stored two nests (33, 32) in 4.8-mm traps on successive days. The traps were suspended on 9 April beneath the limb of a live scrub hickory (Figure 5) at the edge of a fire lane through slash pine-turkey oak association in a wooded area quite open to the sun. The station (#17) was 1.5 m above the ground with the trap entrances facing northeast.
Nesting Behavior
A rufipes female was at this station when I inspected the traps at 1035 on 21 April. She did not enter any of the traps while I was there but crawled over them for a few seconds and then flew off. All of the traps were empty at that time. I remained at the station for 20 minutes expecting that she might return for further inspection or with a cockroach, but she did not reappear. I rechecked the traps at 1310 and again at 1640, but no prey had been brought in during the interim periods.
During my next check of this station at 1030 on 22 April, I found that this or another female had placed a paralyzed cockroach nymph at the inner end of trap 33. The cockroach was lying on its back, head inward, and the wasp egg was attached on the midline behind the fore coxae. There was no temporary plug at the nest entrance. Presumably the cockroach had been brought in earlier on the 22nd, inasmuch as the trap was empty at 1640 the previous day. I remained at this station until 1115, but the Podium did not return to the nest during that 45-minute period. Upon returning at 1345, I found that the wasp had brought in six more paralyzed cockroaches. The prey were all lying head inward, more or less on their venters, and with the anterior end of each prey resting on the abdomen of the preceding individual.
I removed the plain strip of wood from the trap at 1345 to photograph and observe activities through the plastic. At 1348 the wasp flew onto the set of traps, and ran around on the ends of the sticks in a confused manner, searching for her nest. After a few seconds she recognized the nest, entered head first bearing the cockroach venter down and head first clasped in her mandibles, placed it with the head end on the abdomen of the preceding prey, backed out of the nest, and flew off immediately. At 1355 she returned directly to the traps with another cockroach, began to enter the nest, but was frightened off when I approached closely and took a flash picture. She flew around for a few seconds, then returned and took her prey into the nest. She backed out and flew off at 1358. Twelve minutes later she returned with another cockroach, alighted on the end of the traps, crawled in head first, bearing the prey as described above, deposited the prey, and then backed out and flew off.
Half an hour later, at 1428, she flew back without a cockroach, entered the outer end of the boring for a second, and then flew off. I assumed that she might now be ready to plug the nest for she had amassed nine prey. However, at 1449 and 1529 she returned to the nest, carrying neither debris to construct a plug nor additional prey, peered into the nest for a second or two and then flew off. At 1545, after her fourth such visit, I replaced the plain strip of wood and left the area.
Upon returning to this station at 1900, I found that during the preceding three hours the rufipes had brought in two more cockroaches and had constructed the closing plug. Inasmuch as construction of the closing plug is a lengthy process, it is probable that the nest was completed just a short time before 1900. The wasp was not in any of the other traps at this station nor in the immediate vicinity. I removed the completed nest, placed trap 34 in the position occupied by 33, and put an empty trap (32) from an adjacent station in the position occupied earlier by 34.
On the morning of 23 April, I checked all of the other stations before returning to station 17 at 1045. At that time there was a small paralyzed cockroach at the inner end of trap 33, lying on its back, head inward, and bearing the wasp egg behind the right fore coxa. The rufipes female, most probably the same individual that constructed nest 32, alighted on the end of the traps at 1100 without a cockroach. She seemed to have difficulty locating her nest, probably because I had removed the strip of wood covering the plastic. In a few seconds, she found the nest, poked her head inside, and then flew off. She did not return by 1130, so I replaced the strip of wood and left the area.
When I returned at 1311, there was a second cockroach in the nest, venter down, head inward, pushed up onto the abdomen of the small cockroach bearing the egg. At 1318 the wasp flew directly to the trap entrances, walked around in some confusion for a few seconds trying to find her nest because I had again removed the plain strip of wood from the plastic. Finally she entered, carrying a cockroach dorsum up beneath her (Figure 40) and pushed the prey up onto the abdomen of the second prey. I could not ascertain how she held the cockroach while she was in flight with it. When she entered the nest, she was definitely pushing it ahead of her, using only her mandibles to manipulate it. She backed out of the nest immediately, downward onto the end of the trap and then flew off.
At 1331 she flew in with another cockroach, but was frightened away when I arose to take a photograph. She flew back several seconds later but was so disoriented that she placed this prey dorsum up and head inward at the inner end of trap 74 rather than in trap 32. After she flew off, I plugged the entrances to the other three traps containing 4.8- or 6.4-mm borings so that she would not make this error again. She alighted on top of the traps at 1348 but was not carrying a cockroach and flew off immediately. At 1355 she returned with another prey and placed it on its belly, head inward next to the third cockroach brought into 32 (Figure 41). The wasp transported the prey within the nest by clutching it with her long, slender, slightly curved mandibles which fitted into the notch on each side of the thorax between the pronotum and mesonotum.
A period of about an hour now ensued during which the wasp without prey visited the nest momentarily at 1405, 1444, 1510, 1527, and 1548. Usually she flew in and alighted directly on the traps, but on one occasion she alighted first on the limb from which the traps were suspended. Upon each of these visits she either put her head inside the entrance and palpated with her antennae or went a couple of centimeters into the boring and then flew off.
At 1608 she brought in prey again, but was frightened off three times by my attempts at photography. In the quick glimpses I caught of her at this time, it appeared that she might be carrying the cockroach with her mandibles and forelegs. Finally, she took in the cockroach and left it about 2.5 cm away from the preceding prey. At 1624 she brought in another prey which she placed on the abdomen of the previous specimen. And, at 1636 she flew in with another very large cockroach which she had difficulty stuffing into the boring.
After pushing this last cockroach up against the previous prey, rufipes backed out and wandered around on the traps, twice putting her head and antennae into the entrance to nest 32. She flew off after a few seconds and returned at 1639 with a piece of debris which she placed 13 mm inside the entrance. At 1647 she returned without debris, poked her head in for a second and flew off. Then at 1649, 1652, and 1656 she brought in additional loads of debris which she shoved against the first without attempting to compact the mass. At 1659 she alighted on the hickory trunk and in a few seconds flew onto the traps. She did not carry any debris this time and just inspected the plug before flying off again.
At 1702 she returned with a larger load of debris and now she began to compact the various loads. For about half a minute she perched on the end of the trap in various positions, rotating around the entrance and tapping lightly with her mandibles against the mass of debris. At 1705 she was back with another large load which she compacted with the same light pecking movements for 45 seconds. These latter two larger loads of debris consisted of caterpillar frass spun loosely together with silk. Probably she obtained this mixture of frass and silk from the retreat of a leaf-roller or leaf-tier caterpillar.
The shadows were beginning to lengthen as the sun lowered, and the rufipes now began the next step in the construction of the complex closing plug, bringing in her mandibles damp sand which later dried to form a firmly agglutinated layer. The first lump was taken to the nest at 1710 and applied to the surface of the debris carried in earlier. This and successive loads of damp sand were spread over the surface of the plug and tamped in with rapid, light pecking taps with the opened mandibles, the head being held in the normal downward position (Figure 42). To compress the sand evenly the wasp rotated around the entrance. I frightened the wasp away at 1716 when I attempted to photograph her arrival with the next load of sand. She brought in additional lumps of damp sand at 1725, 1729, 1736, 1744, 1751, 1759, and 1804; at 1750 she visited the nest but did not bring in sand. She compacted each load of sand with light blows of the extended mandibles for about a minute, although occasionally the tamping lasted as long as two minutes or as little as 30 seconds. Usually she alighted on the traps and crawled rapidly to the entrance, but occasionally she alighted on the tree trunk and then made a short flight onto the traps.
The final stage in the nest closure began at 1811 when she arrived with a large lump of clear, sticky pine resin in her mandibles. She smeared this over part of the layer of damp sand. She carried in additional lumps of resin at 1815, 1820, 1828, and 1833. These were smeared over the surface of the sand and compacted with more deliberate pecking movements with her opened mandibles for 1–1½ minutes per load. The sun had gone down behind the ridge by 1830, so that it was dusk when she brought in the last few loads of resin. When the wasp did not return by 1841, I picked up the nest, assuming that she had completed the closure.
I revisited this station at 0820 on 24 April to set out a replacement trap for nest 32. The wasp apparently had finished the plug on the previous evening, for there was no resin at the entrance to any of the other traps. I checked the traps at this station several times daily through 26 April, but the rufipes never returned to begin another nest.
Nest Closures
Nests 33 and 32 were examined and photographed in the laboratory on the evenings of 22 and 23 April respectively. The closing plug of nest 33 (Figure 43) was 10.5 mm thick, the inner debris being capped by a layer of agglutinated sand 0.5 mm thick and an outer layer of clear pine resin 1 mm thick. The debris at the inner end consisted of some dry hollowed-out spider parts, bits of spider web, particles of rotten wood, some fine brown fibrous material and then more rotten wood.
The plug of nest 32 (Figure 44) was 12.5 mm thick, the inner debris being capped by a layer of agglutinated sand 0.5 mm thick and by an outer layer of clear pine resin of the same thickness. The debris at the inner end consisted of some particles of rotten wood, then large clumps of caterpillar frass fastened together with silk, followed by a few more fragments of rotten wood.
Prey
The placement of prey in both nests was identical. The first cockroach brought into each trap was placed on its back, head inward at the inner end of the boring, and the wasp egg was laid on it before additional prey were stored (Figures 45–46). Subsequent cockroaches were brought into the nest head first and venter down and stored in that position. Usually the subsequent prey were shingled, so that the head end of one cockroach rested upon the abdominal dorsum of the preceding cockroach (Figure 47). The legs of almost all of the cockroaches were complete, indicating that rufipes seldom or never practices amputation.
I did not observe storing of the first cockroach and oviposition. Consequently, it must still be ascertained whether the wasp carries the first cockroach into the nest venter up, or whether the wasp turns this first prey over after placing it at the inner end of the boring. In either case, before ovipositing, the wasp would have to back out of the nest, turn around, and back into it, so that the egg could be placed in the proper position.
Three or possibly four species of prey were stored in the nests as follows: nymphs of Latiblatella rehni Hebard (Figure 50), adults of Chorisoneura texensis Saussure and Zehntner (Figure 51), nymphs of Chorisoneura sp. probably texensis (Figure 52), and a nymph of Eurycotis floridana (Walker) (Figure 53). I did not observe hunting behavior, but presumably the wasp flushes the cockroaches from their daytime retreats, most likely among fallen leaves or other vegetation on the ground.
Table 3 presents information on the prey stored in these nests: their identity, their length, and their fresh weight. The specimens are listed in the order in which they were placed in the nests, 33–1 being the first cockroach taken into nest 33 and 33–11 the last. The weights of 33–1 and 32–1 include the very infinitesimal weight of the wasp egg which could not be removed without danger of injury. Cockroach 32–4 is the one which was mistakenly placed in trap 74. The mass weight of cockroaches in nest 33 was 347 mg, with an average weight of 32 mg. In nest 32 similar figures were 310 and 34 mg respectively. A male wasp developed in each of these nests.
All of the cockroaches were thoroughly paralyzed and exhibited only relatively weak reflex actions of their antennae, mouthparts, and legs. The prey bearing the egg was no more heavily paralyzed than the other cockroaches. The cockroaches also continued to void feces as may be noted in Figure 54.
Prey
number Prey identity Length
in mm Fresh
weight
in mg
33-1 Latiblatella nymph 10.5 52
33-2 Latiblatella nymph 8.0 24
33-3 Latiblatella nymph 10.5 46
33-4 Chorisoneura nymph 7.0 17
33-5 Chorisoneura adult 9.0 21
33-6 Latiblatella nymph 7.0 19
33-7 Latiblatella nymph 10.5 46
33-8 Eurycotis nymph 6.0 17
33-9 Latiblatella nymph 7.5 25
33-10 Chorisoneura adult 9.0 27
33-11 Latiblatella nymph 11.0 56
32-1 Chorisoneura nymph 6.0 13
32-2 Chorisoneura nymph 5.0 8
32-3 Latiblatella nymph 11.0 59
32-4 Latibtatella nymph 10.5 57
32-5 Chorisoneura nymph 8.0 20
32-6 Chorisoneura nymph 7.5 20
32-7 Chorisoneura nymph 7.0 15
32-8 Latiblatella nymph 10.0 59
32-9 Latiblatella nymph 11.0 60
Life History
The eggs were sausage shaped, 2.67 mm long and 0.67 mm wide. The head end was glued firmly to the intersegmental membrane just to the left of the midline and behind and somewhat beneath the left fore coxa. The left leg was displaced slightly in a larger cockroach (Figures 45, 48) and rather noticeably in a smaller specimen (Figures 46, 49). The eggs extended backward to the apices of the mid or hind coxae.
The egg in nest 33 hatched between 1930, 24 April, and 0800, 25 April, and that in nest 32 between 1700 and 2030, 25 April. Assuming that the latter egg may have been laid about 1000 on 23 April, we can estimate that egg hatch occurs in 55–56½ hours when it is held at about 80°–88° for 9 hours and at 72°–74° for the remaining time. Judged from the hatching data recorded for the egg in nest 32 and the fact that the egg in nest 33 was laid just about 24 hours earlier, we can postulate that egg hatch in nest 33 probably occurred shortly after 1930 on 24 April.
The anterior end of the egg was glued to the intersegmental membrane between the fore and mid coxae. It appeared that the chorion may not be shed until the larva makes some growth, because slight rhythmic contractions were visible posteriorly in the egg in nest 32 at 2030 on 25 April, indicating that feeding had begun. After the larva hatched, it penetrated the cockroach body between the fore and mid coxae and began to suck blood.
The egg-bearing cockroach in nest 32 still exhibited some reflex movements at 0845 on 26 April, a little more than 12 hours after the larva commenced feeding, but it was quiescent and probably dead by 1915 on that same date. One cockroach in nest 33 exhibited reflex movements as late as 0900 on 28 April, before the wasp larva fed on it and almost 6 days after it was paralyzed.
The newly hatched larvae developed slowly at first (Figures 54–56). If the assumption as to the probable hatching time of the larva in nest 33 is correct, then Figure 54 was taken about 16 hours after hatch; note that the larva is noticeably thicker than the egg (Figure 48), but no longer. Figures 55 and 56 were taken of this same larva 24 and 32 hours after Figure 54; note that the larva has increased in girth during these periods but has not become longer. The larva in nest 33 continued to feed with its head inside the first cockroach until 28 April. By 0900 on that date (Figure 57) it had consumed all of the first cockroach except for the exoskeleton.
Development was now much more rapid. By 0700 on 29 April the larva in nest 33 had gutted the second cockroach in the nest (Figure 58). A day later there were only four whole cockroaches left out of the original eleven (Figure 59). On 1 May at 0645 there was only one untouched cockroach (Figure 60). At 0720 on 2 May the larva was feeding on just exoskeletal remains (Figure 61), which it continued to do until that evening when only very small fragments remained.
Larval feeding in nest 32 followed a similar pattern except that the larva gutted the small first cockroach more quickly and was feeding internally on the second by 0900 on 28 April. There were only two large and one small prey untouched at 0640 on 1 May, and only one large cockroach at 0715 on 2 May. By 1945 that evening there were only exoskeletons remaining, and, except for tiny fragments, these had all been consumed by 0825 on 4 May.
Thus, the larval feeding period was about 8 days in nest 33 and 8½ in nest 32, although 37 mg more of food (the equivalent of one moderate-size cockroach) was provided for the former. The discrepancy is undoubtedly due to the larva in nest 32 being subjected to the cooler Arlington temperatures a day earlier in its developmental cycle.
Regrettably, I was unable to obtain information on the number of larval instars. The manner of feeding during the early stages, with the head buried inside the cockroaches, precluded periodic measurements of the head capsule. I was never fortunate enough to witness the shedding of the cuticle at a molt.
Construction of the cocoon in nest 33 required about 2½ days. By 1800 on 3 May the larva had pulled some of the looser debris from the inner end of the closing plug about 25 mm further into the boring. It had also begun to construct a network of silken threads in which to suspend the cocoon at the inner end of the boring against the plastic and sides of the boring. On the following morning at 0825 there was an opaque, fusiform white cocoon at the inner end; no feces had been voided. Most of the loose debris was now at the extreme inner end of the boring and some was attached to the anterior end of the cocoon. By 1925 the larva had coated the cocoon walls with light-tan liquid feces, and it was still moving inside the cocoon. The next morning at 0705 the cocoon was darker brown, still soft and the larva was still working inside. By 0630 on 6 May the cocoon walls were dry and brittle, and the wasp larva was lying motionless inside. Cocoon construction in nest 32 was done in the same manner and required the same amount of time.
On 8 May at 0645 I removed part of the anterior end of the cocoons in nests 33 and 32. The resting larva in the former was quiescent, but that in the latter wriggled forward out of the cocoon; a little later it wriggled back into the cocoon. Pupation occurred in nest 33 between 0630 and 1830 on 21 May, and between 2200, 22 May, and 0600 on 23 May in nest 32. Eclosion of the adult male wasp occurred 15 June in 33; this adult left the nest 18–19 June. In nest 32 the adult male wasp left the nest 19–20 June.
Discussion
The evidence offered by these observations does not permit a determination as to how a female rufipes knows how much prey to store in a nest to bring a larva to maturity. Judged from these data alone, one might speculate that about 9–11 cockroaches are required to bring one larva to maturity. In my earlier lengthy account of this species (Krombein, 1967, pp. 251–255, figs. 62–63), however, I mentioned that nests from North Carolina and Florida contained 3–16 cockroaches. Consequently, it does not appear that the number of prey stored is likely to be the determining factor. An alternative possibility is that the wasp has some sort of internal calculating device to gauge the amount of prey stored by the cumulative weight of the individuals brought into the nest. It will be recalled that these two nests each contained about a third of a gram fresh weight of cockroaches. Many more observations of freshly stored nests will be needed to determine if provisioning by weight is a plausible hypothesis.
There is another possible explanation concerning the number or weight of prey stored; namely, that the wasp calculates neither the mass nor the number, but that the number of prey stored is entirely dependent upon the success of the wasp in finding cockroaches during a single day. It should be recalled that both nests were begun relatively early in the morning, perhaps about 1000, and that the complex closing plug was begun late in the afternoon as the light intensity began to decrease rapidly. One must also remember that the wasp apparently hunted fruitlessly for cockroaches for extended periods during the day, judged from the numerous preyless visits she made to the nest. Can it be that rufipes begins each nest early in the day, stores it with as many cockroaches as she can find during the day, and then constructs the closure, regardless of how many prey have been stored, as soon as the decreasing light intensity imposes a requirement for closing the nest? Podium rufipes does not spend the night at the nesting site as do many other wasps; this could very well dictate that the nest must be plugged at the end of the day. Again, many more observations will be needed to ascertain the validity of this theory.
Figure 62 in my earlier account of rufipes (1967) is a photograph of the prey lying in a nest from North Carolina, showing all of the cockroaches, including the one bearing the wasp egg, lying on their backs head inward. In contrast, in the two Florida nests reported here, only the cockroach bearing the egg was on its back, all of the rest were venter down. Additional field observations are needed to ascertain whether the former method of prey storage is characteristic of the North Carolina population, or whether either manner of prey storage is an individual idiosyncrasy.
I saw no parasites during my lengthy observations at the nest site, and no infestation developed in the nests after they were brought into the laboratory. Actually, rufipes is very subject to parasitism by the cuckoo wasp Neochrysis panamensis (Cameron). In my account of the latter species (1967, pp. 473–475), I mentioned that it was found in 21 rufipes nests from the Archbold Biological Station in Florida, a parasitism rate of 36 percent. Access to the rufipes nests is very simple because there is no temporary closure, such as that made by Isodontia auripes (Fernaid), q.v., and the open, partially stored nest is left unguarded for nearly 8 hours during the day while the wasp hunts for prey.
All of the cockroach species found in these two nests were reported (Krombein, 1967) from earlier nests from the Archbold Biological Station. In addition, Parcoblatta (?) and Cariblatta minima Hebard were found in one or more of the earlier series of nests.
- bibliographic citation
- Krombein, Karl V. 1970. "Behavioral and Life-history Notes on Three Floridian Solitary Wasps (Hymenoptera: Sphecidae)." Smithsonian Contributions to Zoology. 1-26. https://doi.org/10.5479/si.00810282.46
