Epicharis (Epicharoides) albofasciata, SMITH, 1874

Rozen, Jerome G., 2017, Larval Anatomies, Eggs, and Developmental Biologies of Centris bicornuta and Epicharis albofasciata (Apoidea: Apidae), American Museum Novitates 2017 (3879), pp. 1-20 : 16-19

EGG OF EPICHARIS (EPICHAROIDES) ALBOFASCIATA SMITH

Figures 27, 30

DIAGNOSIS: See Diagnosis of egg of C. bicornuta , above.

DEScRIPTION: Specimens uniformly pale cream color. Chorion minutely reticulate at least at ends; under SEM, chorion strongly reticulate with pronounced elevated boundaries (fig. 30). Overall shape of egg elongate, gently curved, tending to be parallel sided (fig. 27), but some slightly swollen at one end; rounded at both ends. Length 2.8–3.5 mm; maximum diameter 0.7–0.9 mm (sample of 10 of 31). Micropyle not evident without SEM examination, but with SEM micropylar area identified with small mound surrounded by converging reticular boundaries (fig. 30); micropyle openings obscure; egg polarity not evident without SEM examination or embryo orientation.

MATERIAL STUDIED: 31 eggs: Trinidad: Hollis Reservoir near Valencia, III-2-1968 (J.G. R. and B.L. Rozen) .

orient for defecation and/or for adult emergence. One wonders whether traction against the smooth, waxlike surface to the cell wall of E. albofasciata ( Rozen, 2016: fig. 14) will eventually be the explanation. More difficult to evaluate are differences in certain structures related to perception. Antennal papillae project in the case of C. bicornuta 3 as do all of its palpi, in contrast to these structures of E. albofasciata and its congeners whose larvae have been described ( Rozen, 1965). The antennal papilla projects on none of its instars, and the palpi scarcely project at all. Perhaps observation of feeding larval representatives of the two genera will be revealing.

As indicated in the Introduction, a recent paper by Martins and Melo (2016) concluded that the tribe Centridini as used by Michener (2007) was paraphyletic as evaluated through a detailed molecular study. One of the external reviewers of the current manuscript requested an evaluation that might support one or the other interpretation of the monopoly of the tribe. While it should be understood the purpose of this study was to explore the developmental larval anatomy of these two genera, the request of the reviewer seems appropriate:

(1) Is there evidence supporting the monophyly of the tribe Centridini sensu Michener (2007) ?

(2) Are there characters that support the combined genera Centris and Epicharis as a distinct lineage, i.e., distinct at the level of tribe?

The answer to the second question is brief: ability and anatomical modifications to spin cocoons could be a good tribal characteristic, but obviously is not necessarily so (e.g., in the Rophitinae ( Halictidae ) species of Dufourea spine cocoons, those of Conanthalictus do not); in Anthophorula (Apidae) some species of winter generations spine cocoon while those of the summer generation of the same species do not spin cocoons ( Michener, 2007).

The response to the first question is longer, in part because it introduces an anatomical character of corbiculate larvae that has gone unreported. Representatives of all corbiculate tribes (Euglossini, e.g., Rozen, 2016: figs. 1, 6, 17, 27; Bombini, e.g., Michener, 1953: fig. 248; Meliponini, e.g., Michener, 1953: fig. 266; and Apini, e.g., Michener, 1953: fig. 275) as fifth larval instars exhibit mostly small, paired, usually pigmented, elevated dorsal tubercles bearing fine setae on the three thoracic segments (as well as of the first abdominal segment of Euglossa ). Such tubercles are lacking in both Centris and Epicharis , thereby separating the corbiculate taxa from the monophyletic tribe Centridini . These tubercles tend to become more pigmented the longer the specimen has been in the last instar, so that those of newly eclosed fifth instars tend to be only slightly pigmented. All tubercles are found on the posterior dorsolateral part in the same relative position on all thoracic segments on all corbiculate specimens, attesting to their homology.

Apis mellifera (listed above as Michener, 1953: fig. 275) is a notable exception in some ways. Its tubercles, though bearing scattered fine setae, are never naturally pigmented, although on specimens cleared and stained with Chlorazol Black E, tubercles are clearly identified. Those

3 Although the antennal papilla of the mature larva of Centris bicornuta is conical and projecting (though small), the projecting shape is not a consistent characteristic for the genus as observed in various species whose larvae have been treated ( Rozen, 1965; Rozen and Buchmann, 1990). However, projecting palpi and galea are persistent features of all known mature Centris larvae. of the pronotum are large and strongly transverse while those of the following thoracic segment are somewhat smaller than those of the pronotum ( Michener, 1953: fig. 275).

Because these tubercles have been identified in all recognized tribes of corbiculate bees for more than 50 years, they are a well-established, if little known, autapomorphy of mature corbiculate bee larvae. This character has not been detected in any other bee larva including those of Centris and Epicharis and thereby supports the recognition of the tribal status of these two genera, assuming that mature larvae of all species of these two genera will eventually be found to lack elevated thoracic tubercles.