Pyrrhiades anchises anchises (Gerstaecker)

Cock, Matthew J. W., 2010, Observations on the biology of Afro-tropical Hesperiidae (Lepidoptera) principally from Kenya. Part 1. Introduction and Coeliadinae, Zootaxa 2547, pp. 1-63 : 14-23

publication ID

persistent identifier

treatment provided by


scientific name

Pyrrhiades anchises anchises (Gerstaecker)


Pyrrhiades anchises anchises (Gerstaecker) ( Figures 3–11 View FIGURE 3 View FIGURE 4 View FIGURE 5 View FIGURE 6 View FIGURE 7 View FIGURE 8 View FIGURE 9 View FIGURE 10 View FIGURE 11 )

Distribution and incidence in Kenya

This is a sporadically and locally common species, widespread in Kenya, particularly associated in my experience with the drylands from Kiboko to Voi along the Nairobi– Mombasa Road. Along this stretch, it can be common during the short rains (Kibwezi Forest, December 1989 – January 1990), when mass movements may take place. In the long rains (May–June) it can also be common, e.g. Bushwhackers (June 1988), and this may account for the populations which spread as far as Nairobi (May 1988), where I think it may not normally breed. van Someren (1939) reports it plentiful in the Chyulu Hills up to c. 1,800m (6,000 ft.). Although there are records from Kitale, and Neave (1904) records it from Ugaia, South of Kisumu, it is much commoner east of the Rift Valley. Sevastopulo (1974) records it as fairly common around Makadara Forest in the Shimba Hills, and he found it in all months except April.

In Kibwezi Forest, where the caterpillars feed upon Caucanthus auriculatus ( Malpighiaceae ) this species is sometimes common or even abundant. For example, on 23 December 1990, every shoot of C. auriculatus seemed to have shelters on several leaves. Outbreaks of caterpillars at this density would result in large numbers of adult butterflies which could then initiate the mass movements mentioned below.

Adult behaviour

Mass movements are probably a regular occurrence in the Tsavo area , Kenya. Dewhurst (pers. comm. 1987) observed a mass movement in Tsavo, and S.C. Collins (pers. comm. 2009) has also observed mass movements in this area late in the year. My first encounter with P. a. anchises was driving through a migration in Tsavo on 4.xii.1987, over a distance of ten or more kilometres between Makindu Gate and Mtito Andei. The butterflies were flying in a roughly easterly direction, and as I drove up from the coast at 70 m. p.h., I could see these large dark butterflies passing diagonally across my windscreen every few seconds. I was only able to confirm the identity of the passing butterflies when one hit my car windscreen—even this was not enough to kill the butterfly (a female) which was fluttering on the road when I went back to look for it. It is possible that more than one species was involved in this mass movement, but I did not obtain more material.

Adults of P. a. anchises are particularly prone to drink at damp soil, especially as the rains give way to the dry season, and I have seen them come to rain puddles in Kibwezi Forest (17.xii.1989), soil dampened by water sprinklers at Kiboko Ranch, and at Tsavo East Park Headquarters. Urine is also attractive. The adults come readily to flowers, including Lantana camara , Vernonia , Acanthaceae etc. They fly throughout the day (08.00h–18.00h), and remain active at the hottest times ( Figure 3 View FIGURE 3 ).

Food plants

Although food plants of ssp. anchises in Africa are known, the early stages have not been described or illustrated ( Henning et al. 1997). Sevastopulo (unpublished) found caterpillars on a twining creeper, with heads of bright yellow flowers, very common on the cliffs of Mombasa Island and on Tristellateia australasiae (= australis) ( Malpighiaceae ). He subsequently recorded the food plant as Marsdenia sp. ( Apocynaceae ) ( Sevastopulo 1974, 1975), and this may be his yellow flowered creeper. In Kenya I have found caterpillars on two species of Malpighiaceae : once on Acridocarpus zanzibaricus , and frequently on Caucanthus auriculatus ( Figure 5 View FIGURE 5 , left). These and records from the literature are summarised in Table 1 View TABLE 1 .


The hemispherical ovum ( Figure 4 View FIGURE 4 ) is cream in colour, 0.9 x 0.7mm wide x high with 16 ribs; the ribs are joined by a series of parallel lines at right angles to the ribs, and these lines are in turn joined by narrow, closely spaced lines, parallel to the ribs. Those which I have found in the field were laid singly on the underside of leaves of Caucanthus auriculatus . Sevastopulo (unpublished) notes that ova are often laid on a leaf growing near the food plant and not on the food plant itself.

Leaf shelters on Acridocarpus zanzibaricus

I once found a final instar caterpillar ( Figure 6 View FIGURE 6 ) on a small plant of Acridocarpus zanzibaricus which just projected above the surrounding vegetation, in the shade of a tree at the edge of the beach at Diani Beach (Ref 89/96, coll. 11.xii.1989, pupa 20.xii, adult male 7.i.1990). This may not be a normal food plant since I have regularly checked this species for hesperiid caterpillars (it is the food plant for Coeliades kenya and C. sejuncta in the same area), and this was the only occasion when I found P. a. anchises on it. When collected, this caterpillar was in a shelter formed from a single leaf. A wide major cut had been made from each edge of the leaf, just basal to the middle, to close to the midrib; distal to this much of the leaf lamina had been eaten; the basal portion of the leaf had been rolled upwards and the margins drawn towards each other with a few strands of silk—a type 3 two–cut shelter ( Greeney 2009) (Group 8 shelter of Greeney & Jones (2003)). A day later the leaf edges had been pulled more closely together and were 5–10mm apart and held by ten strands of silk. Thus, the caterpillar was readily visible in spite of the shelter—in contrast to the situation described below where Caucanthus auriculatus is the food plant.

A nearby leaf (11.5 cm long) showed the structure of the previous leaf shelter which was a variant on the type 3 two–cut shelter, using the distal 3.5 cm of the leaf; a major cut was made from each margin, one of which partially cut the midrib so that the distal portion of the leaf had turned brown and dead, and hung downwards; each side of the lamina of this part was folded upwards, the margins held together with silk, and the inside lined with silk; access was by the basal and distal open ends of the shelter.

Leaf shelters on Caucanthus auriculatus

The shelters made on C. auriculatus ( Figure 5 View FIGURE 5 , left) are different from those described above on A. zanzibaricus . I have not recorded the first stage shelter on C. auriculatus .

A second shelter as used by second and third instar caterpillars on C. auriculatus is shown in Figure 5 View FIGURE 5 (centre); the apical 10mm of the leaf is unaffected, and the shelter made from the section basal to this. The caterpillar eats two slightly curved major cuts from each edge of the lamina up to the midrib, and folds the resultant flaps of leaf lamina upwards to form a rounded chamber held and lined by silk. In Greeney & Jones’ (2003) terminology, this would be a four–cut shelter, but this group of shelters was not included in their classification or that of Greeney (2009), so I introduce this as a sixth category in Greeney’s (2009) simplified classification. The caterpillar feeds on the flaps that make up the shelter and partially skeletonise them, so that they dry out and turn brown, or become waterlogged and deteriorate, depending upon the weather.

The third stage shelter made by caterpillars of the last two instars on C. auriculatus is shown as Figure 5 View FIGURE 5 (right). This uses nearly the whole leaf, folded together upwards, leaving only the extreme basal portion which is separated from the shelter by a major cut from each margin to the midrib—a type 5 two–cut shelter ( Greeney 2009). The weight of the caterpillar within the shelter causes the affected leaf to hang down as shown. Pupation also occurs in this shelter.


I have found final instar caterpillars of three different colour patterns. Form 1 was the single caterpillar found on A. zanzibaricus at Diani Beach (89/96), Form 2 was the dominant form at Kibwezi Forest on C. auriculatus (on two occasions), and Form 3 was found just once on C. auriculatus at Kibwezi Forest (91/12). I could not associate these caterpillar forms with any differences in the pupae or adults, but did not pursue the question systematically. Based on limited rearing of young caterpillars, P. a. anchises seems to have six caterpillar instars.

Form 1. The following description is based on a single final instar caterpillar (89/96), eight days before pupation, when it measured 40 mm ( Figure 6 View FIGURE 6 ). Head 3.6 x 4.2mm wide x high; rugose; orange-red with black spots; an upper row of four black spots across the face, the central two elongated dorsally; below this a further row of three spots; lateral to the lower row, two further spots, an oval one displaced ventrally, over the stemmata, and a very small one between this and the upper row of spots. Body mat black with white and orange transverse markings; deep dull red ventrally and ventrolaterally, as well as prolegs and claspers. T1 a short, narrow, shiny dorsal plate; black; narrow, double, white band on posterior margin, the anterior band extending from dorsum and ending dorsolaterally; the posterior one ending subdorsally (hidden in a fold of cuticle in Figure 6 View FIGURE 6 ), then starting again, level with where the anterior band finishes, and continuing to the ventrum. T2–A7 the transverse white bands are all interrupted at the dorsum, although sometimes very narrowly. T2 and T3 with double white band on posterior margin, fused and broader laterally, and continuing ventrally. A1–A6 with four white bands on posterior half of each segment, the anterior three narrow and the posterior one thicker; the anterior two fuse dorsolaterally and end laterally; the third band extends slightly lower than the first two, to slightly posterior to the spiracle. The posterior band on A1 broadens below the spiracle and continues ventrally. The posterior band on A2 ends with two white spots just below the level of the spiracle. A3–A5 have the posterior of these two spots only. A5 has the subdorsal part of the first two bands orange-red. A7 as A6, but the third band is shorter dorsally and laterally. A8 and A9 each have a single transverse orange band, that of A9 asymmetrical—white laterally on the right side of the body and orange on the left side. The white stripes continue ventrally on A1–A8. Anal plate shiny, black, with reddish triangle on posterior margin.

Sevastopulo (unpublished) includes photographs of similar caterpillars, although others match Form 2, with a dorsolateral streak in the anterior black half of each segment. Sevastopulo notes that this feature is variable.

Form 2. This caterpillar form was not uncommon at Kibwezi Forest 31 Dec 1989 (89/109) and a year later it was common on 23 Dec 1990 (90/121)—on the first occasion, as second and third instars together with similar sized caterpillars of Coeliades pisistratus (Fabricius) , and on the second occasion as fifth instar caterpillars, and no other species found. The following description is based primarily on one caterpillar from the first collection (89/109E). Collection number 89/109 included both P. a. anchises and C. pisistratus species under the same reference number, as I didn’t realise for a few days that I had two species mixed together.

Final instar: Individual 89/109E moulted to the final instar on 10 Jan 1990, was a prepupa on 30 Jan and pupated on 31 Jan, i.e. the instar lasted 20 days. Around the middle of the instar, on 16 Jan, it measured 29mm, head 4.1 x 3.9mm wide x high (average of nine specimens: 4.4 x 4.3mm). I noted ( Figure 7 View FIGURE 7 ) that it was similar to the penultimate instar (below) and differed from the final instar of Form 1 as follows: (1), the black band on T2 –A8 has a white dorsolateral transverse dash; (2), the transverse white bands on the posterior margin on T2 –A9 extend only as far as dorsolaterally, where they stop short; (3), only A7 (and not A6) has orange dorsally, anterior to the four white transverse bands; and (4) T2 and T3 have a white transverse dash laterally, below and parallel to the dorsal bands .

Penultimate instar: Individual 89/109E moulted to the penultimate instar on 1 Jan and to the final instar on 10 Jan. On 4 Jan it measured 17mm, and I prepared the following description. Head 2.6 x 2.6mm wide x high (average of two: 2.9 x 3.0mm); light orange-brown; upper row of six spots, the outermost slightly ventral to the others; lower row of five spots, again the outermost slightly more ventral. T1 black with posterior white band, divided by a narrow black band. T2 as T1. T3 as T2 but white dorsolateral transverse dash in black band; anterior white band slightly narrowed at dorsum. A1–A6 black; narrow white band on posterior margin; anterior to this three narrow white lines ending ventrolaterally, dark at dorsum; narrow white transverse dash dorsolaterally in anterior black part of segment. A7 black; narrow white transverse band on posterior margin; anterior to this a narrow white band with 1mm gap at dorsum; anterior to this two partially adjacent bands, dorsally orange and laterally white; white transverse dash dorsolaterally in anterior black part of segment. A8 black; orange transverse band on posterior margin (broader than white bands on previous segments); short white transverse dash dorsolaterally in anterior black part of segment. A9 black; transverse line bowed posteriorly at dorsum, orange dorsally, white laterally. A10 black with white transverse band arced anteriorly at dorsum. Legs dark; prolegs concolorous. Spiracles dark, inconspicuous.

Earlier instars: The smaller caterpillars from the third instar onwards resemble those of the penultimate and final instars.

Second instar: Head 0.8 x 0.8mm wide x high; pale brown with weak, diffuse dark spots arranged similar to later instars. Body markings similar to later instars but the colouring is not as intense ( Figure 8 View FIGURE 8 ).

The abdomen of Form 2 has distinct white subdorsal streaks in the black anterior half of each segment ( Figure 7 View FIGURE 7 ) which are absent in Form 1 ( Figure 6 View FIGURE 6 ). The strength of these white streaks varies so these two forms intergrade. Sevastopulo (unpublished) also noticed this range of variation in caterpillars at the coast.

Form 3. A third distinctive form is based on one caterpillar collected at Kibwezi on Caucanthus auriculatus , 7 Mar 1991 (91/12). No other caterpillars of P. a. anchises were collected on this occasion. It was collected as a penultimate instar caterpillar, in a leaf shelter made by folding under longitudinally about 1/3 of a leaf. Overall, Form 3 is more reminiscent of Coeliades f. forestan than the other forms of P. a. anchises .

The caterpillar moulted to the final instar four days later on 11 Mar, and by 27 Mar it was a prepupa. The following description was made 21 Mar ( Figure 9 View FIGURE 9 ), when the caterpillar measured 37mm. Head 4.2 x 4.4mm wide x high; light brown with upper row of four black spots and lower row of five smaller black spots; surface sculpture patchy ( Figure 9 View FIGURE 9 , below left). T1 anterior margin down to the level of the spiracle black, then continuing ventrally brown; rest of segment white; spiracle black. T2 anterior half reddish brown; a subdorsal white streak within anterior half; posterior half white. T3 anterior half reddish brown, black dorsally on anterior margin; subdorsal and lateral white streaks in anterior half; posterior half white. A1–5 similar to T3 but no black on dorsal part of anterior margin, pale ventrally to spiracle; lateral spot in anterior half white, rounded; 2–3 transverse dark creases in posterior half; brown in dorsal part of posterior half, divided by pale dorsal line; spiracle black with white surround, at posterior margin of brown anterior half. A6 similar to A5, but a short broad transverse orange mark dorsally on anterior margin of white posterior half. A7 similar to A6 but orange mark stronger and longer. A8 anterior 2/3 reddish brown, with single white streak subdorsally to laterally; posterior 1/3 orange with posterior margin white. A9 anterior to anal plate: black, white, reddish brown, orange and a narrow reddish brown transverse dorsal band. Anal plate white. True legs black; prolegs light brown. The pupa was formed between 29 Mar and 1 Apr, and a female adult emerged 15 Apr.

Form 3 is generally quite similar to Form 2, but the ground colour is reddish grey-brown as opposed to black; in addition to the subdorsal streaks there are lateral ones; and the head is light brown rather than orange brown, with smaller black spots.


The pupa, but not the shelter, is covered with a white waxy bloom, except for two brown, subdorsal, round, bare, subequal spots on the anterior portion of T2, as shown in Figure 10 View FIGURE 10 . One reared by Sevastopulo (unpublished) at Mombasa took 11 days to emerge.

Form 1. The pupa associated with the Form 1 caterpillar ( Figure 10 View FIGURE 10 ) was covered with a dense white waxy bloom, except on the anterior part of the mesothorax which had a round, bare subdorsal brown spot (slightly different on each side of segment). Black spots on wing (end cell), T1 leg (near base, and just beyond middle), T2 leg (near middle), and tip of cremaster. Spiracle T1 a protruding spike of about 1mm, black and conspicuous; abdominal spiracles light brown; inconspicuous.

Form 2: Based on 89/109E: 21 mm; shape as C. forestan ; black marks on projecting spiracle T1, tip of frontal spike, anteroventral margin of eye, end of forewing cell, T1 leg at base and 2/3, T2 leg at 2/3 with row of bars distal to this; a small double arc just anterior to cremaster ventrally. A male emerged after 16 days.

Form 3 ( Figure 11 View FIGURE 11 ): 21 mm; greenish, covered with white waxy bloom; projecting spiracle T1 black; two black spots on T1 leg; one on T2 leg, with series of dark bars distally; hollow dark quadrate cell in wing cell. A female emerged after about 15 days.

Natural enemies

The caterpillars are attacked by a gregarious braconid ectoparasitoid as described under C. f. forestan below. Five ectoparasitoid larvae attacking a penultimate instar caterpillar of P. a. anchises (about to moult) on C. auriculatus at Kibwezi Forest (31.xii.1989), spun cocoons on 4.i.1990 and emerged on 12.i.1990.


Bearing in mind the demonstration of cryptic species of Hesperiidae in Costa Rica when life history information and barcoding are combined with traditional taxonomy (see introduction), how should the diversity of larval forms described above be interpreted? Forms 1 and 2, particularly when combined with Sevastopulo’s (unpublished) observations, most likely represent a range of variation. It would be a reasonable expectation that this range of variation represents typical ssp. anchises , given the type locality is Zanzibar. Form 3, however, seems radically different from the other two, and so could be interpreted as either representing an extreme variation, a cryptic species or a hybrid.

TABLE 1. Recorded food plants of Pyrrhiades anchises anchises. Localities preceded by a question mark reflect a food plant record from a national or regional work, which may or may not be based on observations in that country or region.

Family Food plant Locality References and comment
Apocynaceae Marsdenia angolensis East Africa van Someren 1974; Kielland 1990; Henning et al. 1997
Marsdenia Kenya Sevastopulo 1974, 1975
Malpighiaceae Acridocarpus zanzibaricus Kenya (coast) This work
Caucanthus auriculatus Kenya (Kibwezi) This work
Triaspis glaucophylla (= leendertziae) South Africa Gifford 1965; Dickson & Kroon 1978; Pringle et al. 1994; Henning et al. 1997
  ?Tanzania Kielland 1990
Moraceae Ficus sp. Socotra Larsen (1991) based on an erroneous record of P. a. jucunda from Socotra (Larsen pers. comm. 2009)