Chilimalopsis, PARVULA TORO

Rozen, Jerome G., 2011, Immatures of Exomalopsine Bees with Notes on Nesting Biology and a Tribal Key to Mature Larvae of Noncorbiculate, Nonparasitic Apinae (Hymenoptera: Apidae), American Museum Novitates 2011 (3726), pp. 1-52 : 32-36

CHILIMALOPSIS PARVULA TORO

Table 2

Chilimalopsis parvula and Chalepogenus rozeni Roig (Tapinotaspidini) nested in cracks in the ground at 26 km S of Vicuña, Elqui Province, Chile on October 30, 1992, and the site was studied on October 31 and November 1, 1992, and again on November 8 and 10, 1992. Both species were actively foraging and nested in the same cracks. Freshly constructed cells as well as those from previous generations of both were intermixed in the soil next to the cracks. The ground surface, unshaded except for scattered low-growing plants, sloped to the west generally 20° to 30° from horizontal, and at least some Chilimalopsis entered cracks where the surface sloped about 45°. The pollen source of C. parvula was the low-growing Pleurophora polyandra Hook. and Arn. (Lythraceae) , which matted the adjacent ground surface. The cracks, created by drying, were generally vertical and most abundant near the surface where the soil was the least moist. Soil consisted of 48% sand, 17% silt and 35% clay and classified as sandy clay/sandy clay loam. Larger cracks descended to approximately 40 cm, well below the lowest cell depth. Bees avoided excavating through the extremely hard, dry surface soil by descending through the cracks to the moist lower soil, which could be more easily dug with their mandibles. Although a few vespids also nested in these cracks, no other bees were observed using them.

NEST ARCHITECTURE: Nest entrances of Chilimalopsis parvula were scattered as circular, open holes on the vertical or nearly vertical faces of cracks, mostly 12–18 cm below the surface.

Small pellets of excavated soil, uniform in size, adhered to the crack faces below the entrances. Main tunnels penetrated the soil in variable directions, ascending, descending and/ or twisting for up to, but no more than, 2 cm into the surface. The largest nests appeared to consist of only three cells, and some obviously were of single or two cells. Cells in a single nest were often arranged in linear series of two or the tunnel branched with a short, soil-filled lateral leading to another cell. Because cells in a nest were so few in number, nests were probably constructed by single females, and almost certainly females normally constructed more than one nest. Because of the fine soil texture, burrow walls bore distinct impressions of having been tapped repeatedly by the female’s pygidial plate.

Whether nests are normally closed by main tunnels being filled with soil after nest completion remains questionable. Certainly many main tunnels leading to closed cells appeared open. Several nests were discovered where an intercalary cell somewhat smaller than a brood cell appeared in front of the brood cell (fig. 56), and in one of these cases the tunnel in front of the intercalary cell was filled with loose soil particles. A septum forming the front of an intercalary cell was concave and smoothed on the outside, like the cell closure itself; the inside surface was a spiral but rougher than the inner face of brood cell closures. Intercalary cells may simply be a nest closure consisting of two cell closure, the last one being spaced a short distance in front of a cell closure. More completed nests need to be examined to resolve our understanding nest closures.

Cells ranged in depth from 12 to 30 cm, but most seemed to be concentrated between the depths of 13–18 cm. Cell orientation was variable with the long axis ranging from being nearly horizontal to tilting toward the rear as much as 80° as indicated in table 2. Cell entrances were generally somewhat smaller than the burrow diameter. When first observed cells seemed to be symmetrical around their long axis, but on close examination one was clearly slightly flatter on one side (fig. 56); I suspect that all are probably flattened to a similar extent. Several cells were discovered partly penetrating vacated cells that had been packed with fill either by the nesting female or by emerging adults. Cell walls were smooth and lined with a shiny waterproof material from the rear of the cell all of way through the cell entrance. The smooth outer surface of the closure was also waterproof when tested with a droplet, but shortly in front of the closure the lining to the tunnel lost its smooth texture and became water absorbent.

Inner surfaces of cell closures were 1.5–1.8 mm (N = 5) in diameter, uniformly flat, and consisted of a spiral of 3–4 coils to the radius (fig. 57). When tested with a droplet, the surface absorbed water in several seconds, suggesting that soil compactness rather than a waterproof secretion was responsible for slowness of absorption. Closures were uniformly thick, about 0.7–0.8 mm in the middle. The coil was constructed first, and then soil of the outer surface was deposited, smoothed, and waterproofed by the female. Septa separating cells in series were also uniform in thickness, about 0.5–0.8 mm thick in the middle, and the inner surface of the closure of the rear cell was also a coil identical to that of the first cell.

PROVISIONING AND DEVELOPMENT: In provisioning, female Chilimalopsis parvula first deposit the reddish-purple pollen of Pleurophora polyandra as irregular masses on the cell floor. After importing the total allotment for a cell, she shapes it into a freestanding, loaflike mass that varied considerably in contour and degree of surface smoothness; internally it was mealymoist. In all cases it was longer than wide or high, and its top surface was somewhat flattened. Its undersurface conformed to the curvature of the cell floor. Dimensions are given in table 2.

Eggs (table 1), curved, white, and with shiny, smooth chorions, rested with their anterior and posterior ends touching the top surface of the provisions, their long axis in the sagittal plane of the cell, and their more rounded anterior end directed toward the cell closure. On hatching, young larvae were elongate and moved over the surface of the provisions, making shallow channels roughly the width of their body as they progressed. As the larvae grew, the provisions became smaller and more ovoid. Older larvae cradled ovoid provisions holding them away from the cell surface in that only the dorsal surface of the larva touched the cell wall. When the provisions were reduced to a small semitransparent irregular mass, the larva discharged the contents of its Malpighian tubules, which up to that time were yellow, clearly visible through the transparent body wall because of the dark reddish color of the filling mid intestine. The discharge, an opaque yellow material, was applied by the anus onto the cell wall as a series of blotches. Afterward the vacated Malpighian tubules were transparent, no longer visible. Defecation commenced immediately thereafter, before the food mass was entirely consumed. Because the tubule discharge was so limited, the discharge obviously was not instrumental in modifying the absorption quality of the cell wall, as has been suggested for some bees ( Rozen, 1987).

Feces were deposited first as elongate reddish pellets applied to the cell wall. They eventually covered all surfaces of the cell including the closure as an even layer before cocoon construction started; they were not embedded in cocoons as characteristic of Anthophorula nitens and A. chionura . Cocoons were a distinct white or slightly tan, soft, tissuelike, semiopaque sheet of silk strands that covered the entire cell surface and fecal layer. In some cases they were light brown, not much darker than the surrounding soil. Although the fabric appeared to be a single layer, when it was pulled away from the fecal lining to the cell, a layer of silk strands remained glued to the fecal material although the part pulled away maintained its integrity. Under a light microscope, the front end of the cocoon was of the same consistency and structure as the rest of the cocoon.

From the appearance of vacated cocoons, emerging adults chew away the front end of the cocoon at egress. Those emerging from the rear cell in a series chew not only a hole in the front of their cells but also obviously a hole in the rear of the cell in front. Vacated cocoons were loosely packed with soil no doubt from the cell closure and perhaps from the closure of intercalary cells and burrow fill if present. Cocoons from the present as well as from previous generations appear uniform in texture indicating that all individuals still spin cocoons (in some multivoltine bees, nondiapausing generations do not always make cocoons or the cocoon produced by the earlier generation is different in construction from that of the overwintering generation ( Rozen, 1984)). Although the gas exchange aperture was not found on the remnants of any cocoons, there is no reason to think that it is absent. Cocoon fabric (fig. 64) is similar to that of other exomalopsines treated here viewed with an SEM.

ADULT ACTIVITY: Adults were on the flowers and flying over the ground in considerable abundance. Mating was not studied, but numerous small masses of adults (as many as 5 or 6) tumbling on the ground suggested that groups of males aggressively attempt to mate with a female, with some individuals departing and others joining the mass. Males were also noted gathered around cracks, entering and departing, suggesting that they may have been searching for emerging females in the cracks. Adults were most active during midday, both with respect to mate searching and foraging.

There appears to be only one generation a year. Only one pupa was encountered, presumably an individual that was slow in emerging in the spring. All larvae became quiescent after cocoon spinning.

PARASITISM: No parasitic bees of appropriate small size were discovered at this site, and no cells contained cleptoparasitic bee larvae. Bombyliids and meloids, either as adults or immatures, were not detected at the site.