Annette Aiello.
Adelpha (Nymphalidae): Deception on the wing.
Psyche 91:1-46, 1984.

Full text (searchable PDF, 4660K)
Durable link: http://psyche.entclub.org/91/91-001.html

The following unprocessed text is extracted from the PDF file, and is likely to be both incomplete and full of errors. Please consult the PDF file for the complete article.

PSYCHE
Vol. 91 1984 NO. 1-2
ADELPHA (NYMPHALIDAE):
DECEPTION ON THE WING*
BY ANNETTE AIELLO
Smithsonian Tropical Research Institute
P.O. Box 2072, Balboa, Republic of Panama For the past century, lepidopterists have puzzled over the genus Adelpha Htibner, in an attempt to discover the secret character or combination of characters which might lead to a satisfying classification of the 100 or more butterfly species included in this large neotropical group. Several approaches have been tried (Godman & Salvin, 1884, 1901; Fruhstorfer, 1907; Forbes, un- published manuscript): wing pattern (both upperside and lower- side), wing venation, genitalia, and various combinations of these. If one attempts to coordinate all the information available, the result is a hopeless tangle. As a result, the most obvious set of characters (wing pattern) traditionally has been used in classification; other character groups (genitalia, larvae, pupae) which appear to confuse the situation have been largely ignored. A new strategy is needed, one which would both evaluate the reliability of the different groups of characters already surveyed, and search anew for overlooked sources of information. It was the purpose of my research to review what is known of the immature stages of Adelpha species and, based upon that information plus my own observations made in Panama between 1978 and the present, to make speculations regarding species relationships.
*Manuscript received by the editor January 12, 1984.



================================================================================

Psyche
The first attempt at classification within Adelpha was made by Godman & Salvin (1884, 1901) with their revision of the 32 species reported from Central America. At that time some 70 species were known for the genus. In their treatment, Godman & Salvin discuss Adelpha and its relationship to Limenitis and note that several species (bredowi, populi, carnilla, lorquini) are difficult to assign to either genus. Distribution of such characters as eye pubescence, venational differences, variation in proportions of male leg seg- ments, and peculiarities of male genitalia within Adelpha-Limenitis is surveyed. Classification of the 32 species begins with isolation of A. bredowi (eyes smooth in front) from all others (eyes hairy in front) and continues by arranging the other 31 species using gross features of wing pattern. The result is eleven groups, six of which are represented by single species.
With Fruhstorfer (1907) came the first and only published revision of the entire genus Adelpha. The 90 species treated are assigned to two main groups based upon the length of the forewing discal cell; short = Adelpha group, elongate = Heterochroa group. Upon this division, Fruhstorfer comments, "Anatomically there are also two series [male genitalia with or without clunicula] of species distinguishable. They, however, do not agree with those based upon the structure [discal cell length]." The linear arrangement of species within the largest group (Heterochroa with 82 species) reflects presumed relationships based upon wing pattern features. Fruhstorfer touches upon but does not pursue the possibility of a closer than realized alliance between New and Old World groups. In his introduction he notes that the male clasping organs of AdeZpha ". . .approach those of the [Old World] genus Pantoporia (At-) in such a way that. . .it would be quite impossible to ascertain where organs or photes [sic] of them belong to, which are not denorni- nated." Concerning the male valves, he further observes, ". . . there exist also nearly square ones with 2 or 3 small acicular teeth
(resembling a Limenitis [Moduza]procris from India and Ceylon)." Fruhstorfer's revision included a number of misidentifications, which presumably were corrected by Hall (1938) following exa min- ation of the original material.
At the time of his death (1968) W. T. M. Forbes had made a good deal of progress on a revision of Adelpha, and his manuscript is in



================================================================================

19841 Aiello - Genus Adelpha 3
the Archives of the Museum of Comparative Zoology, Harvard University. About half of this manuscript is in a nearly illegible hand; half is typewritten. The typed portion includes part of an introduction, notes on each species, a key (based upon wing pattern) to species and species groups, and descriptions of four new species. Included also are more than 70 genitalic illustrations (inside of right valve) by Howarth. Forbes had analyzed these latter and was in the process of constructing a key to genitalia. His two approaches, of wing pattern and genitalia, yielded different species groupings, and it is not clear whether he favored one of these, or intended to use a combination of the two in his final classification. Of the immature stages of Adelpha, very little has been said, and what has been said has been largely ignored for the simple reason that this new information appears to confuse matters rather than clarify them.' Forbes (uncompleted manuscript) had read Moss's (1933) paper on Adelpha larvae and pupae, and had examined several pupal skins of Adelpha and Old World Limenitini in the collections of the British Museum (Natural History) when he commented, "The larvae and pupae are highly varied and unless a high percent are misdetermined, show characters wholly incongru- ent with adult structures and patterns." The fact that, within Adelpha, the study of adult characters results in species-alignments different from those obtained by consideration of the immature stages and/or genitalia, suggests that at least one set of characters is unreliable or perhaps even deliberately deceptive.
Due to taxonomic confusion within Adelpha, it is impossible to know how many species actually are represented by the 34 or so life history accounts published for this genus. My estimate is that at least 24 species are illustrated (including illustrations in this paper) as final instar larva, or pupa, or both, many of these by more than one author. Most accounts, whether illustrated or not, include a foodplant record.
'This author has located only four publications (Moss, 1933; Muller, 1886; Young, 1974; Cornstock & Vasquez, 1960) which figure any immature stages of Adelpha; several publications present descriptions only.



================================================================================

4 Psyche [vo~. 91
In the present paper, all illustrated reports are reviewed to see whether it is possible to distinguish natural species groups within the genus Adelpha.
Judging from the few scattered reports (Table 1) of oviposition in the Limenitidini, eggs are laid singly and are usually placed at the tips of leaves. While many accounts do not specify whether placement is on the upper or lower surface, the majority that do specify, report upper surface oviposition. Adelpha iphicla females alight upon a leaf and walk backwards while searching for the desired oviposition site with the tip of the abdomen; the egg is then placed and the butterfly takes flight before her next oviposition, which may be on the same leaf. In the case of lower surface oviposition, the female merely bends her abdomen around the edge of the leaf and touches its tip to the leaf. Among the six species of Adelpha whose eggs were collected in Panama, four (Table 1) place their eggs along damaged portions of leaves, especially on jagged points, as well as at the leaf tip, and one of these (A. iphicla) is about as likely to lay eggs along an intact leaf margin as at the tip. As well, a female A. iphicla may return to the same leaf several times and place as many as four eggs on one leaf, often on its undersurface.
Eggs of the seven Adelpha species (basiloides, cocala, cytherea, marcia, iphicla, melanthe, phylaca aethalia, and salmoneus) exam- ined by this author, were all similarly sculptured (figure 1) with pits (hexagonal due to packing) and seta-like projections (one from each junction of three pits). This same sculpture type is figured for Limenitis by both 2Scudder (1889) and Eltringham (1923). Young's (1974) description of the egg of Adelpha leucophthalma as having the seta-like projections "aris[ing] from the facets," is doubtful.
Also doubtful is the account by Comstock and Vazquez (1960) which describes and illustrates the egg of A. celerio as having convex rather then concave hexagons. Such an error is easily made, as these *As Basilarchia.




================================================================================

19841 Aiello - Genus Adelpha 5
Table 1. Placement of egg on leaf by various members of the Limenitidini (Nymphalidae)
EGG
PLACE-
MENT REFERENCE
-- -
Adelpha basiloides Bates TI, Dl Aiello
Adelpha celerio diademata Fruh. T Comstock & Vazquez (1960, pg. 407)
Adelpha cocala Cr. Ti, (Dl) Aiello
Adelpha cytherea (?) t Muller (1886, pg. 484) Adelpha iphicla L. TI(!), MQt) Aiello
Adelpha iphicla t Muller (1886, pg. 484) Adelpha isis Dru. Tl, Dl Miiller (1886, pg. 48 1) Adelpha leucophthalma tegeata Fruh. Dl Young (1974, fig. 2) Adelpha melanthe Bates Dl Aiello
Adelpha phylaca aethalia Feld. Tl Aiello Adelpha plesaure Tl Muller (1886, pg. 484) Adelpha salmoneus Butl. TI, (Dl) Aiello
Adelpha serpa Tl Muller (1886, pg. 484)
Adelpha syma Godt. T Hoffmann (1937, pg. 212) Athyma nefte Cr.
(as Parathyma nefte Cr.)
TI, (Mt )
Morrell (1954, pg. 160)
Afhyma opalina Kollar t Robson (1894, pg. 338) Lasippa tiga Moore
(as Neptis heliodore Fruh.) T Morrell (1954, pg. 160) Limenitis archippus
(as Basilarchia archippus) TI Scudder (1 889, fig. 16) Limenitis lorquini burrisonii Maynard T Dornfeld (1980, pg. 6 1) Moduza procris Cr. T Morrell (1954, pg. 157) Neptis nata Moore T Morrell (1954, pg. 162) T = leaf tip (tooth, in case of A syma), M = leaf margin (intact), D = damaged portion of leaf, I = upper surface of leaf, T = lower surface of leaf, () = less commonly Figure 1. Eggs of Adelpha (left) and Doxocopa (right).



================================================================================

6 Psyche [vo~. 91
eggs can present an optical illusion, however, an edge view of a partly eaten egg shell shows clearly that the hexagons are indeed pits.
Comstock and Vazquez (1960) also describe and figure a differently sculptured egg, which they call A. iphicia with the reservation that it might actually be an egg of 3Doxocopa. Because their larvae died soon after hatching, the authors never knew that they had indeed been fooled by the ovipositing Doxocopa female, a mimic of Adelpha. My own rearing of Doxocopa laure (LOT 83-6) was from two eggs with pattern (figure 1) identical to that in figure 33
of their paper. My reared Adelpha iphicla were from eggs patterned as those of all the other Adelpha species examined so far. Possibly this egg sculpture pattern will be found throughout the Limenitini and may prove useful in defining tribal limits. Oviposition was observed only for Adelpha iphicla, and each of the three fresh eggs collected required 5 days development before the first instars emerged. Of five other species collected as eggs, of unknown age, three had longer minimum egg-development times than did A. iphicla: A. basiloides (LOT 82-65), one egg hatched after 6 days; A. cytherea (LOT 83-3), one egg hatched after 7 days; and A. salmoneus (LOT 83-14), two eggs hatched after 6 days. Because individuals were collected at various stages of devel- opment, total time from hatching to eclosion could be determined for only a few individuals of ten species. In spite of the fact that the resulting data (Table 2) is not uniform, it is clear that development time is variable, and often more so within a species than between species. The extreme example of wide variation is in A. basiloides which has a longer development time than other species, and which sometimes passes through six instars instead of the usual five. Five 6-instar individuals of A. basiloides were encountered, and these were from four separate rearing lots, each of which included 5-instar individuals as well. Each of these lots was collected on Amaioua corymbosa; larvae collected on Alibertia edulis and Bertiera guianensis did not produce extra instars. In spite of the additional instar, 6-instar individuals did not require longer development times 3As Chlorripe.




================================================================================

19841 Aiello - Genus Adelpha 7
than 5-instar individuals, however, larvae on Amaioua, regard less of number of instars, were slower to develop (41-48 days) than were larvae on Alibertia or Bertiera (32-44 days). Miles Moss reared twelve species of Adelpha from Para, Brasil, and illustrated (1933) the final instar larva and the pupa for each of them. In spite of the lack of detail in his illustrations and descriptions, and the fact that his scheme for numbering body segments omits abdominal segment-9, his little paper remains tile masterpiece on Adelpha immatures. It was Moss who realized (page 15) that:
". . .systematists,. . . by a careful examination of certain hither- to unsuspected points of likeness or dissimilarity between the species in their early stages, may perhaps be led to modify tile existing order and grouping of the butterflies of this difficult genus. It is just possible that a few unexpectedly close relationships may thus be established, while others at present confused, or regarded as near of kin, may be found to be more distantly related than was supposed."
He concluded that A. cytherea, and pseudococala are closely related, also delphicola and mesentina, and serpa and paraw. My analysis of published illustrations and live material lends support to Moss's groupings and adds several more.
FIRST INSTAR LARVAE
First instar larvae, of all species which I have seen, appear identical in form, and are some shade of brown or grey. The head bears setae, but none of the4chalazae of later instars; pale bumps on the body are found where the future ^coli will be, and the body is covered with tiny pale spots each centered with a minute seta. After their first meal, larvae take on the green color of their foodplant, although the head remains brown.
SECOND INSTAR LARVAE
From the second instar on, the head is ornamented with chalazae which give it a spiny appearance. The face has numerous round,



================================================================================

Psyche
[Vol. 91




================================================================================

Table 2. (Continued) Duration (in days) of larval instars and pupal stage for twelve Adelpha species. leucophthalma
melanthe
nr. paraha -
phylaca aethalia m4
111
salmoneus (41, 5
PI
Unless otherwise noted, the above data was collected in Panama by the author. [l=n
() = commonest number of days in cases of strong bias towards one end of range; otherwise, ranges are normal curves. m = minimum number of days in cases where full time for an instar is not known.



================================================================================

10 Psyche [vo~. 91
flatbottomed pits which first appear in small numbers in instar two, three, or four depending upon the species. These may be the same color as the rest of the head capsule or may be a contrasting color. Facial stripes appear in the fourth instar of some species as well. The body now bears stubby scoli (each with 3-5 radiating apical spines). These scoli are arranged in three rows (subdorsal, supra- spiracular, and subspiracular), thus, each body segment has three pairs of scoli. The subdorsal scoli on thoracic segments 2 & 3 and abdominal segments 2,7, & 8 are very slightly larger than the others, but all are similar in form.
Beginning with the second instar there are color and pattern changes in some species, but most are fairly uniformly colored (brown, green, or black) and have paler scoli and tiny spots, much as the first instar. In Adelpha basiloides, the larval color depends upon the foodplant; larvae are light brown or reddish brown on Amaioua, and dark brown on Alibertia and Bertiera. THIRD AND FOURTH INSTAR LARVAE
From instar three on, the face is framed by two distinct rings of chalazae, and body scoli show further development, especially those scoli which will be the largest or most distinctive in the final instar. By late fourth instar, body scoli have become swollen at the bases due to developing final instar scoli inside. In these two instars, the scolus spines are pale, and as before, the body is speckled with tiny pale spots.
A dorsal, paler patch ("saddle") appears in the third or fourth instar of many species. In A. salmoneus and cytherea the "saddle" is faint and extends from abdominal segment 2 or 3 through segment 6. A. cocala and leucophthalma may have a "saddle" on abdominal segments 5-6. A. justina is pale dorsally from thoracic segment 2 through abdominal segment 8. A. basiloides, nr. param and celerio have a triangular "saddle" with its base on the posterior portion of abdominal segment 4, and its apex at mid abdominal segment 6. In the latter two species the "saddle" is poorly defined; in A. basiloides it is sharply demarcated.
4Terminology of Peterson (1962). cHALAZAE(AE): a distinctly elevated cone-shaped area, bearing 1-3 simple setae; (SCOLUS(I): an elongated projection, bearing 4 or more setae or spines.




================================================================================

Aiedo - Genus Adelpha
FINAL INSTAR LARVAE
Final nstar larvae (Figures 4 & 5) of all species studied have several tz mings in common.
The head (Figure 2) has a spiny
appearam e e owing to the numerous chalazae which frame the face. These s e c m to be fairly constant in number and position (Figure 3) and to v a Ty from species to species mainly in their relative size. A. iphicla f o example has a relatively smooth face because it lacks several D the chalazae found in other species. A, phylaca and especially mehihe are the spiniest in appearance due to numerous Wae. color and pattern also vary: A. celerio has a ~triped face, A. basiloides and A. cocala patterned ones. The pits of A. cytherea are darker than the rest of the face; the head of A. salmoneuS is reddish and constrasts strongly with its green body. Each b o d y segment bears three pairs of scoli (subdorsal, supra- ~~iracular- and subspiracular); there are no dorsal scoli. While the scoli are variable in form and length, in all species studied, those of the proth o Tax and abdomen-1 are either very short or are reduced to a few ~p5nes. Usually the longest are the subdorsal scoli of meso- and meta-zhorax and abdomen-2, -7 & -8, and the supraspiracular scoli of t h e meso-thorax. In many species the subdorsal scoli of abdomen-z are the most distinctive in form and often are curved backward S -
BO~Y scoli are diverse but the various forms can be grouped into two main types: those which are terete (round in cross-section), and those which are flattened.
The terete SCO~US, in its simplest form, is a short stalk with 3-5 spines radiating star-like from its apex (e.g., 5scoli A3-6 of A, celerio, cyzAerea, sa~moneus, and justina). More complex scoli are slender and longer with spines at intervals along their full length, and either with one to a few ascending spines towards the apex (thoracic scdi of many species, e.g., A. phylaca, melanthe, cocala, basiloide~, cytherea, justina, salmonem), or with 3-5 radiating apical spines (e.g., scoli of A3-6 in A. phylaca, melanthe). Scoli also may be short and thick with a dense covering of spines (e.g., A2 of A. cocala, Zeucophthalma, basiloides), or be club-shaped and more sparsely spined (e.g., A8 of basiloides). s~hroughout the remainder of this paper, "sco1us(i)*' refers to the subdorsal set unless otherwise noted; T = thorax, thoracic segment; A = abdomen, abdominal segment.



================================================================================

Psyche
Figure 2.
Faces of final instar larvae of Adelpha species reared in Panamh: B A S = basiloides, CEL = celerio, COC = cocala, CYT = cytherea marcia, IPH = iphicia, MEL = melanthe9 PHY = phylaca aethalia, SAL = salmoneus.



================================================================================

Aiello - Genus Adelpha
Figure 3.
Location of head chalazae 1-5, referred to in text. Flattened scoli have a plumose or leaf-like appearance (e.g., A. celerio, serpa, paraena) which is due not only to the arrangement of spines in two opposite rows along the scolus, but also to the flattening and widening of those spines which may take on spatulate, elliptical, or lanceolate forms, and may be so crowded that they overlap one another. In the case of extremely condensed scoli, the spines appear to fuse with one another (see A. serpa in Muller, 1886).
Upon hatching, a larva eats some or all of its egg shell and then feeds on the leaf tip leaving the midrib intact. Larvae rest out on the midribs which they have exposed, and eventually extend them by addition of fecal pellets held in place with silk. Most larvae fashion several such supports during each instar, often using lateral veins instead of the midrib. When not feeding, Adelpha larvae rest out on their supports, facing away from the leac molting takes place on the support as well. After they molt to the final instar, larvae abandon their supports and thereafter rest on the upper surface of a leaf.



================================================================================

14 Psyche [vo~. 91
In addition to the above behavior, typical of many nymphalid butterfly species, first through fourth instar Adelpha larvae engage in an odd practice not known for other New World butterflies; they accumulate their fecal pellets, fastening them in place with silk, to form a mass which either surrounds the base of the support or is suspended beneath it. This mass may include bits of leaf in A. iphicla, or consist of more leaf than feces in A. phylaca and melanthe. The work of A. basiloides is the most distinctive because, in addition to the mass just described, this species consistently constructs a small, usually curved, larva-form mass on the leaf surface, several mm away from either the leaf edge or the usual mass. Upon viewing this artistry, one cannot help imagining that it serves as a decoy larva to discourage would-be predators. A single reared individual of A. iphicla also engaged in this behavior, but excepting that, I have observed it only in A. basiloides. Resting larvae position their bodies in one of several ways: (1) Straight Position (figures 4 & 5). Larvae in any instar may use this position, and typically when out on their supports they rest this way.
(2) Front-Curved Position, as shown in Young (1974, fig. 2. B, C). In this position, the larva grasps its support with the prolegs only, and raises and curves its anterior portion (head through A2) so that the head is somewhat inclined, and the toracic scoli are directed forwards. Larvae about to molt use this position; the second instar in Young's (1974, fig. 2C) photograph has a swollen pro-thorax and is probably preparing to molt to third instar. Final instar larvae of A. justina use a raised but uncurved version of this position instead of the usual final instar stance, described next, (3) Front-Arched-Rear-Up Position.
This position is typical of
final instar larvae but occurs in earlier instars as well. Involved are the raising and arching of the anterior pdrtion of the body (head through A2), ,plus elevation of the posterior portion (A7-10). In addition, the thoracic scoli are directed forwards, and those of A2 are held backwards. In this position, the face and the area at the top of the arch (T3 through A2) are parallel to each other and to the substrate. This and the next position are assumed when the larva is disturbed.
(4) Curled Position. The larva curls to one side into a "C" or "J" shape on the upper surface of the leac the rear portion (A7-10) may or may not be elevated. A. celerio often rests in this position,