Click here for the original journal page (in Acrobat pdf format).

The text below is grayed out because it is not intended to be read. It is a necessarily imperfect OCR of the original and is only used by a search engine.

1956

The Lepidopterists' News

37

THE HONEY GLAND AND TUBERCLES OF LARVAE OF THE LYC^NID^

by Gowan C. Clark and C. G. C. Dickson

A considerable amount of data having been gathered concerning the honey gland and its attendant tubercles of larvae of the Lycaenidse, it is presented here to help anybody studying this interesting subject. The honey gland is an orifice dorsally placed on the 10th segment, which is capable of exuding a liquid much sought after by ants. It is, in some, a simple transverse slit on the surface, in others the slit is sunken in an elliptical trough, and in others it is raised above the surface on an elongated mole, while another form is an elongated mole sunk in a depression. The gland is placed in the centre of the segment, toward the posterior edge, or is tucked away in a fold between the 10th and 11th segments. It is often protected by spined setae. When the larva exudes a bead of liquid, the gland of most species unfolds into a bulge before the liquid appears.

The tubercles, it they are present, are situated on the 11th segment, one on each side, in the vicinity of the spiracle. In a primitive form, they may be present in the first instar or they may not appear till some later instar, while in a few they are only present in the final instar. The table on pages 38-39 indicates the range of variation in the honey gland and tubercle characters in South African lycaenids.

In the South African larvae studied, there are two forms of tubercles, (a) what may be called warning beacons, and (b) what are definitely whips. Two species at least which possess the former type, namely of Castalius and Syntarucus, use their tubercles as whips. In both of these the spines are long and, with the tubercle half extended, they form a stiff cluster with which the vicinity of the honey gland is dusted. In (b) the tubercles are encased in protruding cylinders the rims of which are, in most cases, provided with hard spines to protect the bristles of the tubercles which are too long to be totally withdrawn. The portion of the rim facing the honey gland usually has no spines or has very small spines so as to allow full freedom of action to the tubercle. Cylinders which rise well above the surface are capable of swaying from side to side to give a bigger dusting range. The tubercles are eversible rods which are forced straight out, then bent over. The spines, some placed on the end and some on the side, sweep a large area as they unfold.

When an ant wishes to "milk" a lycaenid larva, it generally takes a stance on the dorsum and caresses the vicinity of the gland with its antennae. The larva responds by exuding a bead of liquid, which the ant removes. After swallowing this it again caresses the vicinity of the gland until satisfied or driven away by the tubercles.

The warning beacon type of tubercles are generally very sluggish and are normally drawn below the surface of the skin, but when the larva is annoyed by the attentions of insects other than its accustomed ants, the tubercles are fully extended after a few nervous tremblings and half-threatening attempts. The tubercles remain in the fully extended state for about 3 to 4 seconds and are then

38

Clark & Dickson: Honey glands                Vol.10: nos.1-2

withdrawn, but if the interference continues, out come the tubercles again. They can be operated independently. In some cases the tubercles may be made to extend by tickling the vicinity of the honey gland with a paint brush.

The whip type tubercles are all very active and shoot out in rapid determined flashes, one to two flashes per second, some slightly faster. When crawling about, the larva seldom has its tubercles completely dormant, and they are seen to be trembling in and out with occassional flashes as if in a state of nervousness or a warning to insects to keep away.

1956

The Lepidopterists' Neivs

39

SPECIES

Capys alphceus Cram. Deudorix antalus Hoff. Deudorix diodes Hew. Leptomyrina lara Linne Thestor basutus Wallgr. Thestor dicksoni Riley Lachnocnema bibulus Fab. Lachnocnema durbani Trim. Durbania amakosa Trim. Alcena amazoula Bdv. Hypolyccena philippus Fab. Epamera sidus Trim. Epamera cemulus Trim.

Epamera mimosa? Trim. Epamera alienus Trim. Stugeta boivkeri Trim. Argiolaus silas Westw. Myrina ficedula Trim. Alyrina dermaptera Wallgr. Poecilmitis thysbe Linne Spindasis natalensis Dbl. & Hew. Crudaria leroma Wallgr.

The majority of species illustrated have been studied with ants in attendance, and to obtain the detailed drawings larvae have had to be killed in order to force out the tubercles for examination. The illustrations show: a semi-dia-gramatic view of the last four segments of the final instar; a scale drawing of the honey gland as seen from above; if present, the fully developed tubercles extended; and, where possible, the earliest form of the tubercle. All drawings are drawn more or less to one size for ease of comparison.

Referring to the illustrations, Cacyreus lingeus has no tubercle; the space is therefore utilized to show the honey gland bulged out and exuding a bead of liquid. The drawings of Castalius melcena include a view of the partially extended tubercle with spines bunched ready to be used as a brush, and the drawing of Syntarucus telicanus shows the method of brushing. The honey gland of Epamera sidus is difficult to detect as it resembles an ill-defined wrinkle between the 10th and 11th segements; consequently a side view of the larva is given with an enlarged view of the open gland. Leptomyrina, Capys, Deudorix, Hypolyccena, and Lepidoc lory sops have no tubercles, and only the honey gland is shown. In Capys alphceus two views are shown at right angles to each other, of the sunken gland, one showing the gland dormant, the other exuding a bead of liquid. The illustration of Castalius hintza shows the dormant tubercle and the spiracle.

In the whip type the tubercles are present in the 1st instar, and drawings of the 1st instar tubercles are shown in all except Phasis thero, Axiocerses bambana,

LYC^ENID^ LARV/E

PLATE 1

Cacyrei/d //'ngeus

Actiiere j/&//<? fa

Lamp ides boeticus z.

Cdjfd//us me/dent

»ttte

Zizeerid knysna

Tdrucus ffiesph

Azonus moriqua

Wa/A

Azanus jesous

o/—\o

^IftffF -i^fr

&

Syntarucus fe/ictinus

Tdrucus Theophrasfus

£ac/?rysops dolorosa

Lepjffochrysopj ptfrjc/d

\Eicochrysop5 messopua

Anthene ofacitici r.

Anthene amara*

/t^Hw, -mi

\Anthene //vida T.

W -#k

41

40

Vol.10: nos.1-2

ind Poecilmitis. The small tubercle in the illustrations of P. felthami is that of ;he second instar.

The protective rim spines of the cylinders of the tubercles are clearly shown, and in order not to foul these when extended, Spindasis natalensis has five spines on one side of the tubercle and only three on the other. In Pcecilmitis                    *

thy she three illustrations are given to show the unfolding of the tubercle and one front view to show the spread of the bristles. The honey gland of the                    f

Alceides is tucked away under a fold of the junction of the 10th and 11th segments and is difficult to find until it functions. Its position in the drawings is marked by an arrow. The Spindasis and Crudaria have, in the centre of the segment, on the dorsum, a saucer-like dishing which suppurates a liquid similar to the honey gland, and ants are seen to take the liquid from these. These patches, called "dew patches" for want of a better name, first appear on the 5th segment in the 3rd instar of C. leroma. In the 4th instar the patch is prominent on the 5 th segment, and there may be one on the 8th, though less developed, and there may be traces of patches on the 6th and 7th segments. In the final instar they are fully developed on segments 5 to 8, as shown in the illustration. In the Spindasis the first appearance of this patch is on the 5th segment of the penultimate instar, but in the final instar there is a fully developed patch on each of segments 5 to 8.

In captivity, the honey gland and "dew patches" of C. leroma suppurate freely, especially in the later instars, and unless ants are in attendance the liquid develops mildew which kills the larva. To prevent this, the liquid can be drawn off by a point of blotting paper. It may be as well to "wash" the glands with a very wet paint brush and thoroughly dry the larva with points of blotting paper.

Museum & Snake Park, Port Elizabeth, SOUTH AFRICA and Cambridge Ave., St. Michael's Est., Cape Town, SOUTH AFRICA

LYOENID/E LARV/E

PLATE 3

So/n da did natdlens/s 2)6/. * Hew.

Crudaria /eromd Wd//^

Woe/des ardnda ,

v m//pr.

V

Poecilm'/t/'s thy 66 e/.

mr-

P/ias/s thero L.

"tHrt

Aloe/des fa/kosdma

WJ//$r,

\/\loeides p/eros

iA             ^ Cram.

Aloe/des thyra l.

h.

Y^xiocerses bambdna

Gr.Sw

Poec/lmifo cbryoaor t\

Poecilmitis pd/nwj

Sj&^eil

Poec/lrnitti pyroe/6 T.

\Poeci/mifoJe/ttm/ T.

Jc

KS

G.CC. +C,G.C.£.

43

LYC^ENID^ LARV/O

PLATE 2

\Anthene iemnoa {

Anthene definite

]£p0mero s/dua z

]£pcT/pera aemuft/j

iP if

\Epdmera mimosdt

Stugefo 6ou//reri

Argio/dud Ji/dJ weAftyrina ficedutd n

%mm* Jt

*^^9 ,

\*^

NW&'.« ^^jL

si-fir ttl ,j\

>y**\4i t&Sffi

x«. . S^di-^L^ I * >5j-i **<

fei

^Ult

<r,ft«-'ol

iM£$*

\Wi/rina dermdptera\

Cdpysd/phdec/iS Cr.

vtfftir'W1          .6 1

Cdpas disjt/n^pd

Weudorix antdtud "'A

V^t^v

XDeudorix dioc/ed

Lepfomyrina lard l

Viypo/ycdend ph/tyoptA

Cddtd/ius fy/nfza T.

m

Sm**

&XJ5KJa®

42