This article was published in the Yearbook of the South African Orchid Council - ORCHIDS South Africa, 2003, pp.59 - 65.

For many years I have promised myself that I would take a look into what must be a fascinating relationship between African orchids and the hawk moths of this continent. The result of this short study is presented here and will hopefully stimulate others to start asking the questions and making the observations that will help to fill in the extensive gaps in our knowledge.
     Africa and South America appear to have been the continents least affected by the vicissitudes of Earth's traumatic history of the last 100 000 years. These continents appear to have borne the smallest brunt of ice-ages, the proposed Mt Toba eruption and the suggested multiple impacts by cometary fragments which resulted in Noah's flood. The result of this luck has been a good rate of species survival. Man's long time presence in Africa, together with his use of fire, must have affected its ecology and may have had some effect on distribution and survival in such groups as the epiphytic orchids. Hawk moths have probably been more fortunate, having wings as they do, and in the Ethiopian Faunal Region, (Africa south of the Sahara), well over 300 species of this family survive today, many of them narrow endemics. Compare this to the mere 65 species to be found in the Western Palaearctic region (Europe and western Asia).
     Africa's fairly limited orchid flora includes the white-flowered, spurred, evening-scented Angrecoids, as well as terrestrials of the genera Habenaria and Bonatea, which show characteristics one can associate with hawk moth pollination.
     It is only in the last two decades that orchid botany in Africa has started developing an ecological slant. Prior to this it seems that most work in this field, with a few exceptions, revolved around the discovery of new species and the study and classification of the resultant pressed specimens. Nowadays, much work is being done on the pollination biology and other interactive systems and we seem well on our way to our heydays of orchidology, even if a little belatedly.
     Resulting from this work, it was reported in 1997 that the species Bonatea speciosa was pollinated by hawk moths (Johnson S.D. & Liltved W.R., Nordic Journal of Botany, 17: 5). This Bonatea is a common species occurring throughout the moister summer rainfall areas of South Africa and additionally there are populations that occur in the winter rainfall areas of the south western Cape. It is not a very homogenous species, with growth habit and flowering time varying considerably from area to area. It is probable that a number of different hawk moth species are its chief pollinators in the various populations in South Africa, contributing towards this diversity.
     My own plant of this species, of unknown provenance (I received it some 12 years ago from a deceased estate) flowers in autumn and it has the most delicious honeysuckle scent in the evening. It brings in the hawk moths and as a result, quite a few pods develop each year. I am loath to trap and identify the moths that arrive. As a child, I often collected eggs or young larvae of various Lepidoptera, especially hawk moths and raised them to adults, which I then liberated. From this association, I see hawk moths as much more than "bugs". I have not yet formulated a method of identifying the species involved which does not offend my strange sensibilities.

Bonatea speciosa

     Pollination by Hawk moths, also known as sphyngophily, marks such flowers with a very specific set of characteristics. The flowers are usually white to cream in colour, which maximally reflects the small amount of light available at night, such as that from the moon. Occasionally, the flowers are light green, to aid in camouflage during daylight. They are usually strongly evening-scented, producing a sweet fragrance. The fragrance is usually the long-distance lure which attracts the hawk moth, the colour and shape being the short-distance cue which aids in the correct alignment of the moth to gather nectar and thereby effect pollination. Douglas McMurtry & Shane Burns, who reviewed this article, comment: "Look at how species like Habenaria laurentii and H. holubii, and even H. cornuta have evolved to align the approaching moth."
      Whether small day-flying hawk moths like the clear-winged Bee Hawk Moth Cephanodes hylas and the African Humming Bird Hawk Moth Macroglossum trochilus are involved in the pollination of orchids is unknown. The former is often seen on an autumn afternoon here in my garden visiting mauve and purple flowers like Barleria repens (Acanthaceae).
     As hawk moths are strong flyers with excellent hovering abilities (similar to hummingbirds), no landing platform need be supplied by the flowers visited by these insects. Orchid lips, which are usually presented as landing platforms for insects, need to be differently designed in those orchids pollinated by hawk moths. Such lips may be minute, greatly divided, or the flower may be non-resupinate or even hyper-resupinate, causing the lip to be displayed as the topmost segment. Hyper-resupination may be seen in Angraecum eburneum, where the flower twists through 180° as in most other orchids, but then continues to twist through a further 180° so that the lip returns to the dorsal position. In such "flag-lip" flowers, the lip's main function is as an attractant.

Angraecum eburneum

     Hawk moths are not stupid fellows, and as a reward for their visit they require nectar. This nectar is presented at the bottom of a long, descending spur. Such nectar is fairly viscous in order to detain the moth for the critical period required for the viscidial glue to set. The "air-conditioning" supplied by the rapidly beating wings also helps to dry the glue (all things are connected, if you look closely enough). Whereas other, more stupid insects can be deceived with spurious offers by dry flowers, hawk moths will not visit such flowers repeatedly, so attempts at deception in sphyngophilous flowers would lead to extinction of the plant. Robert Dressler points out in his 1981 book The Orchids - Natural History and Classification (Cambridge MA./London: Harvard Univ. Press), that the hairy bodies of hawk moths are poor sites for the deposition of sticky orchid viscidia. The only smooth areas which could serve as stable platforms for viscidial attachment are the proboscis, such attachment being seen with pollinaria of the angrecoid orchids; the eyes (Bonatea) and possibly the legs.
     As hawk moths would not fly into dense vegetation, flowers pollinated by them need to be exposed. Terrestrial sphyngophilous orchids must produce flower spikes long enough to lift the flowers above the surrounding vegetation. Some of the sphyngophilous terrestrial orchids may grow in shady places or other special localities, where lack of light limits the competition from other plants. Epiphytes pollinated by hawk moths usually have pendulous spikes and if the spikes are more upright, they are long in relation to the leaves, in order to give the hawk moths clear access to the flowers, without the chance of collision with foliage.
     In his 1954 book, BlütenbiologischeTypen als Elemente der Sippengliederung (Jena: Fischer), S Vogel is reported to have discussed the pollination of some South African orchids by hawk moths, including various plants of the genus Habenaria, such as H. epipactidea and suggests hawk moth pollination for Bonatea, Holothrix and Bartholina.
     The Angrecoids are a large group of principally African and Madagascan, epiphytic species with usually white or cream to greenish, spurred flowers which includes genera such as Angraecum, Aerangis, Mystacidium, Cyrtorchis, and Plectrelminthus, and must be principally pollinated by hawk moths.
     Checking through the list of South African orchids makes me suspect sphyngophily may occur in the following groups, based on flower colour, morphology, fragrance, etc.:

1. Longer spurred species of Holothrix like H. filicornis and H. randii

2. Both species of Bartholina

3. Many of the South African species of Habenaria

4. Most species of Bonatea

5. The related Centrostigma occultans

6. Perhaps some of the Disa (Monadenia) species like D. conferta, D. comosa and D. bolusiana

Other disas worth investigating include D. intermedia, D. patula, D. salteri, and D. fragrans

7. Some plants of the genus Satyrium in the white and light greenish colour-range

8. Cynorchis kassneriana ?

9. Amongst the Angrecoids;

and Tridactyle.

     McMurtry comments: "I am not happy that Bartholina are moth pollinated, as where I saw them flowering, they were in thickets of Elytropappus rhinocerotis."
Recently I have seen a colony of plants flowering in restio thickets, with all the spikes coming out of the tufts at an angle to be presented in clear areas. Additionally, it is said that this species blooms best after fire, when the restio masses have been severely reduced. Greig

      The authors of the chapter on pollination in Linder & Kurtzweil's great work Orchids of Southern Africa (Rotterdam/Brookfield: AA Balkema), Steve Johnson and Kim Steiner, state that in addition to the above-mentioned Bonatea speciosa and Habenaria epipactidea, the following two South African species are hawk moth pollinated :
Satyrium hallackii subsp. ocellatum (Johnson S.D. and van Ede G., SA Orchid Journal, 24: 124 (1993)). This plant is lighter coloured and longer spurred than the typical variety. As this flower still makes some floral anthocyanins, which are metabolically expensive and unnecessary in a hawk moth pollinated flower, one can hypothesize that sphyngophily here is of recent advent.
Disa cooperi (Johnson S.D., Nordic Journal of Botany, 15: 121 (1995)). This species has most of the typical characteristics associated with hawk moth pollination. The most interesting aspect here is the spur, which rises from the flower at a 45º angle; not the usual presentation.
     Two very special cases of African hawk moth - orchid associations need to be addressed. Firstly, the famous pollination biology of Angraecum sesquipedale by the Madagascan moth Xanthopan (Macrosila) morgani subsp. praedicta. The prediction which gave rise to the subspecies name having been variously attributed to Alfred Russel Wallace (mostly by moth people) and Charles Darwin (by the rest of the world). The reason for this has long been of interest to me and after much probing, I find myself no nearer an answer to the question; Wallace or Darwin? It is really all a matter of semantics. In 1862 in the book On the various contrivances by which British and foreign orchids are fertilised. (London: John Murray), (pp.197-8), Darwin wrote;

"I fear that the reader will be wearied, but I must say a few words on the Angræcum sesquipedale, of which the large six-rayed flowers, like stars formed of snow-white wax, have excited the admiration of travellers in Madagascar. A whip-like green nectary of astonishing length hangs down beneath the labellum. In several flowers sent me by Mr. Bateman I found the nectaries eleven and a half inches long, with only the lower inch and a half filled with very sweet nectar. What can be the use, it may be asked, of a nectary of such disproportional length? We shall, I think, see that the fertilisation of the plant depends on this length and on nectar being contained only within the lower and attenuated extremity. It is, however, surprising that any insect should be able to reach the nectar: our English sphinxes have probosces as long as their bodies: but in Madagascar there must be moths with probosces capable of extension to a length of between ten and eleven inches!"

(One can read this wonderful book on the Web at The book has been digitised by John van Wyhe of the Dept. of the History and Philosophy of Science at Cambridge University, UK, with some contributions by myself.)

     By anyone's criteria, that ranks as an amazing prediction. And the case should be closed there, but the authors of the subspecific name praedicta, W Rothschild & K Jordan, in their 1903 "A revision of the Family Sphyngidae" (Nov. Zool. 19, Suppl Parts 1-2), attribute the prediction to Alfred Russel Wallace. Hence the differing view of moth people and the need to take intention into consideration.
     Alfred Russel Wallace is a hero of mine. Darwin's ideas of Natural Selection arose after visiting the fairly young and simple, 7 million-year-old ecosystems of the Galapagos, and then digesting what he had seen over some 20 years with infinite care, patience and attention to detail. Alfred Russel Wallace developed similar concepts, completely independantly, while still in the jungle, after spending about eight years in the much older, mature ecosystems of the Amazon and South East Asian rainforest, amidst such a cacophony of sensory white-noise. What a guy, it appears that he could think in broad brushstrokes.
     Wallace was a great essayist and in his 1867 essay Creation by Law which appeared in the October issue of the Quarterly Journal of Science (it can be read on the Web at Charles H Smith's The Alfred Russel Wallace Page at,) a footnote can be found which reads:

"I have carefully measured the proboscis of a specimen of Macrosila cluentius from South America in the collection of the British Museum, and find it to be nine inches and a quarter long! One from tropical Africa (Macrosila morganii) is seven inches and a half. A species having a proboscis two or three inches longer could reach the nectar in the largest flowers of Angræcum sesquipedale, whose nectaries vary in length from ten to fourteen inches. That such a moth exists in Madagascar may be safely predicted; and naturalists who visit that island should search for it with as much confidence as astronomers searched for the planet Neptune,--and they will be equally successful!"

     Such a defined prediction so close to the name Macrosila morganii must also be seen as remarkable in its own right. This prediction was made five years after Darwin's, obviously based on that of Darwin, and 36 years before praedicta was described.
     Those people interested in fabled creatures, proposed and supposed animals and other as yet unseen in the flesh organisms, the cryptozoologists, like to use Xanthopan morgani subsp. praedicta as an example of a cryptozoological success story. The Institut Virtuel de Cryptozoologie in France (or are they so "virtual" that they are nowhere in particular ?) have published a lovely web-page in English (also available in French) at, which is well worth a visit. They also mention Gene Kritsky of Ohio College in Cincinnati, who in 1991 predicted the existence of yet another unknown, large hawk moth in Madagascar, with a proboscis of 38 cm, the pollinator of another Madagascar orchid, Angraecum longicalcar, a species with a spur about 40 cm long. (American Entomologist, 37(4): 206-210).

Angraecum longicalcar

     The species Xanthopan morgani as it occurs on the African mainland is probably responsible for the pollination of several Angrecoid species. It is the hawk moth species with the longest proboscis (in the region of 19 cm) on the mainland and occurs widely in tropical and equatorial coastal areas and less frequently in the tropical interior. Orchid species that this hawk moth services probably include the west African Angraecum infundibulare which has a 20 - 23 cm spur, the west African Plectrelminthus caudatus (spur of 17 - 25 cm), Aerangis brachycarpa from east and central Africa (spur 12 - 20 cm) and the south-east African Summerhayesia zambesiaca (spur up to 20 cm).

Summerhayesia zambesiaca

     This hawk moth is large, its size is only surpassed by very few of the 122 species covered in Elliot Pinhey's 1962 book, Hawk Moths of Central and Southern Africa (Cape Town: Longmans), a useful, if somewhat dated reference. The length of the forewing varies from 53 - 65 mm and is light brown, marbled with dark streaks. The hindwing is dark brown with yellow patches basally. The moth's body is light brown with rows of yellow spots down the sides. This plump moth apparently makes good bat food, as on Zanzibar, where the moth is quite common, the ground in bat-frequented areas can become confettied with their wings.
      The larvae of Xanthopan morgani feed on the leaves of trees of the genera Annona (custard apple genus) and Uvaria. The first part of the genus name Xanthopan means yellow, the meaning of the "pan" part is obscure. The species epithet commemorates Rev D F Morgan, a missionary who worked in West Africa in the mid-nineteenth century.
      The other African hawk moth of particular interest with regard to orchids is one of the smaller hawks (with a forewing length of 20 - 25 mm) called Theretra orpheus. Having the common name of Lined Hawk Moth, this brown moth has darker brown lines along the body and forewings. It is widespread in the Ethiopian region, but occurs only in specific localities. The reason for this is that its larval foodplants of record are epiphytic orchids, that are themselves only locally common. The genera Ansellia, Angraecum and Polystachya are on record as dietary items. Unlike most other hawk moths which pupate under the soil, this species pupates in trees at the base of their foodplant.
      Knowing the voracious appetites of hawk moth larvae, one can only be thankful that Theretra orpheus is of small size. Bearing in mind the difficulty a caterpillar would experience in transfering to another plant in the event that it had completely consumed its natal plant, large size in this species would probably have lead to extinction.
      I am not aware whether this species represents a problem as a pest in orchid collections in its area of distribution, including Zimbabwe and the Natal-KwaZulu, Mpumalanga and Limpopo provinces of South Africa.

Suggested additional reading includes the delightful Darwin and the Star Orchid of Madagascar by Carl T Ramsey (AOS Bulletin, 34: 1056-62 (Dec. 1965)) and Darwin's Orchid Obsession by Eric Hsu (Orchids (AOS), 71: 428 - 35 (May 2002)).
Acknowledgements: I wish to thank Douglas McMurtry and Shane Burns from the Lowveld for their careful and considered reading of this article, as well as their constructive comments and corrections. Many thanks also go to Lourens Grobler for the use of his photographs in the printed article, as well as this Webpage.
All photographs on this page © 2003 Lourens Grobler


This page was last updated on the 9th January 2011.

© 2003-2011 Greig Russell