V. A. TORTILIS

5.1 A. tortilis: Recognition and Relationship

In general, A. tortilis is very distinctive and easily recognised, with the characteristic mixture of long straight spines and shorter hooked ones combined with spirally twisted or contorted pods. Over most of its range the characteristically flattened crown has given it the popular name of Umbrella Thorn; however, this is usually not the shape in the subsp. raddiana. Generally the foliage is smaller than in many acacias, and the whitish flowers in small round heads are also characteristic, though this is a feature shared with other Acacia species.

The only closely related species, but not likely to be confused, is A. campoptila Schweinf. from P.D.R. Yemen. Here the spines are mixed in length but are all straight or nearly so and characteristically pointing upwards. The spirally contorted pods are narrow (3–4 mm wide) and clothed with long whitish spreading hairs up to 3 mm in length; in A. tortilis this is a rare feature only found in the very local var. crinita (see p. 43).

5.2 A. tortilis: Summary of previous taxonomic work

Until the 1950's it was generally assumed that there were two main species in the Acacia tortilis complex in tropical Africa, one with glabrous pods called A. tortilis or A. raddiana, the other with pubescent pods called A. spirocarpa. The mainly South African A. heteracantha was recognised as related but was not well understood. Brenan (1957) revised the complex, pointing out that typical A. tortilis had pubescent not glabrous pods, that the characteristics of A. raddiana were not as clear-cut as had been assumed and that A. heteracantha, while very similar to A. spirocarpa in most ways, had pods glabrous as in A. raddiana. He accordingly recognised only one variable and widespread species, A. tortilis, with four more or less geographically demarcated subspecies.

The sinking of A. raddiana as a subspecies of A. tortilis has not been accepted by everyone (e.g. Karschon, 1961), mainly because typical A. tortilis and subsp. raddiana are generally well defined in the northern part of their range. Nevertheless, while accepting that further evidence and study are needed and that they could cause a change of opinion, it is still likely that, if the species is studied as a whole, raddiana is rightly placed as a subspecies of A. tortilis. For further comments about intermediates between the two see p. 41.

Fig. 16. Acacia tortilis subsp. tortilis. (A) flowering branch; (B) pod; (C) seed; (D) detail of leaf. Acacia tortilis subsp. raddiana. (E) flowering branch; (F) pod. (A-C, E-F approx. x .8; D approx. x 4).

5.3 A. tortilis: Brief description

Usually a tree, sometimes a shrub or bush 1.5–18 m high, occasionally to 21m. Bark on trunk usually rough and fissured, grey to black or dark brown. Crown usually flat and spreading but sometimes (especially in subsp. raddiana) rounded. Young branchlets densely to sparsely pubescent or glabrous to subglabrous. Spines paired; some short and hooked, up to about 5 mm long, others long straight, slender, whitish, 1.2–8(-10) cm long, never enlarged or inflated. Petiole usually with a gland. Pinnae 2–10, occasionally (Ross, 1979) to 14 pairs, on a short rhachis up to 2(-4) cm long, usually glandular between the top 1–3 and lowest 1–2 pairs of pinnae. Leaflets 6–20 pairs per pinna, glabrous to densely pubescent, 0.5–2.5(-6) mm long, 0.2–1(-1.5) mm wide. Flowers white or yellowish-white, scented, in round fluffy heads 0.5–1.1 cm in diameter, on axillary peduncles 0.4–2.5 cm long. Pods variable (see descriptions of subspecies) indehiscent or slowly dehiscent, contorted or spirally twisted, very rarely and abnormally (Kenya) pods straight or nearly so, see p. 43 glabrous, pubescent or tomentellous, rarely with spreading hairs. Seeds oblique or parallel to long axis of pod.

5.4 A. tortilis: Distribution

Widespread in Africa from South Africa northwards to Algeria and Egypt, extending in Asia to Israel and southern Arabia. See also distributions of individual subspecies and varieties.

A. tortilis is cultivated in India (subsp. raddiana) and Pakistan (Ali, 1973) but here the subspecies is not given.

Fig. 17 Acacia tortilis. Map showing approximate distribution of species and subspecies.

5.5 A. tortilis: Variation

(i) Introduction

A. tortilis is a widespread and variable species within which four subspecies are recognised here. Their geographical ranges are generally distinctive, though there is some overlap in subsp. tortilis and subsp. raddiana. The latter is sometimes treated as a distinct species but is here regarded as a subspecies for reasons given on p.42.

The subspecies are separated by the presence or absence of pubescence on the pods, their width and differences in the pubescence of the branchlets. The shape of the crown is distinctive in subsp. raddiana.

(ii) Key to subspecies and variaties

(iii) Descriptions, synonymy and distributions of subspecies and varieties

1. A. tortilis (Forsk.) Hayne subsp. tortilis Synonyms:-

   Mimosa tortilis Forsk. (1775)

   A. spirocarpa var. minor Schweinf. (1867–8)

   Shrub or small tree 2–6 m high, with flattened crown. Young branchlets densely and shortly pubescent. Petioles and rhachides of leaves densely and shortly pubescent. Pods 3–5 mm wide, shortly pubescent, eglandular.

   Subsp. tortilis occurs in Somalia, Ethiopia?, and the Sudan, northwards to Egypt and Israel and extending to the Yemen Arab Republic, P.D.R. Yemen, Oman, Saudi Arabia, Abu Dhabi and Qatar.

   See comment under var. crinita (p.43) about the possible occurrence of long spreading hairs on the pods of subsp. tortilis.

2. A. tortilis subsp. raddiana (Savi) Brenan (1957) var. raddiana

   Synonyms:-

   A. raddiana Savi (1830)

   A. fasciculata Guill. & Perr. (1832), non H.B.K.

   A. tortilis var. lenticellosa Chiov. (1932)

   A. tortilis forma raddiana (Savi) Roberty (1948)

   Tree 1.2–10 m high, with more or less rounded crown. Young branchlets and leaves glabrous or subglabrous. Pods 6–9 mm wide, glabrous, eglandular.

   Subsp. raddiana var. raddiana occurs in northern Africa from the Senegal eastwards to the Sudan, Somalia and Kenya. Its status (native or introduced) in Kenya is uncertain. It also extends through Egypt to Israel, Jordan and Saudi Arabia. It is cultivated in India (Rajasthan).

3. A. tortilis subsp. raddiana var. pubescens A. Chev. (1927) Synonyms:-

   A. fasciculata var. pubescens (A. Chev.) A. Chev. (1928)

   A. tortilis var. pubescens Aylmer ex Burtt Davy (1930), illegitimate name

   A. raddiana var. pubescens (A. Chev.) A.F. Hill (1940)

   Similar to var. raddiana but young branchlets shortly more or less pubescent and pods appressed-puberulous.

   The var. pubescens has been found in Mali, Algeria and the Sudan. It will be noted that the features differentiating var. pubescens from var. raddiana are those that might be expected from hybridisation between subsp. raddiana and subsp. tortilis. Whether this is the explanation of var. pubescens must await further evidence. It is noteworthy that Karschon (1961) mentions trees in Israel “intermediate in some characters between A. raddiana and A. tortilis” and considers them hybrids. Monod (1974) discusses similar trees as “tortilis (raddiana X spirocarpa” in Tunisia near Bou Hedma. El Amin (1973) considered that the distinctions between spirocarpa and raddiana were not clear, though he suggested (without reasons) that raddiana might be better reinstated as a species.

   It is considerations such as these, together with the occurrence of glabrous pods in subsp. heteracantha that lead me to consider that raddiana is nevertheless best placed as a subspecies of A. tortilis.

   

   Fig. 18 Acacia tortilis subsp.spirocarpa Habit of tree. Tanzania, Iringa District (Milne-Redhead & Taylor 11224)

4. A. tortilis subsp. spirocarpa (Hochst. ex A. Rich) Brenan var. spirocarpa

   Synonyms:-

   A. spirocarpa Hochst. ex A. Rich. (1847)

   A. petersiana Bolle (1861)

   A. spirocarpa var. major Schweinf. (1867–8)

   A. spirocarpa forma pubescens Terracc. (1893) probably

   A. pappii Gandoger (1913)

   A. tortilis forma spirocarpa (Hochst. ex A. Rich.) Roberty (1948)

   Tree 2–21 m high, with flattened spreading crown. Young branchlets densely pubescent. Petioles and rhachides of leaves densely pubescent. Pods 6–13 mm wide, densely tomentellous or pubescent with short spreading or recurved hairs mostly up to 0.75 mm long; scattered longer hairs sometimes also present. Numerous dark reddish glands present among the hair and visible with a hand lens.

   Subsp. spirocarpa var. spirocarpa is restricted to eastern Africa, occurring in the Sudan, Ethiopia, Somalia, Uganda, Kenya, Tanzania, Malawi, Mozambique, Zimbabwe and Botswana. For the occurrence of subsp. spirocarpa in Namibia/S.W.Africa, see under var. crinita. (See below).

   In general, var. spirocarpa is easily recognised and not very variable. However, a few specimens at Kew from Kenya and Tanzania have glandular but very sparsely hairy pods, thus showing same approach to subsp. heteracantha. In addition, some Kenya specimens (Pratt s.n., Pratt K 559) are remarkable in showing pods almost straight, not spiral or contorted. Further information is needed about such variation.

5. A. tortilis subsp. spirocarpa var. crinita Chiov. (1916)

   Similar to var. spirocarpa but pods more or less densely clothed with long spreading whitish hairs 0.75–3 mm long, in addition to shorter hairs and glands.

   The var. crinita has been found in Somalia, Kenya and Tanzania. Material of subsp. spirocarpa from Namibia/S.W. Africa is probably also referable to var. crinita. (See Ross, 1979).

   Some specimens at the Royal Botanic Gardens, Kew, from the Yemen Arab Republic (Wood 3090, 3134, 3248) are very similar to var. crinita in their densely spreading-hairy pods, but the pods lack glands. The evidence is insufficient for a certain opinion, but it is possible that these specimens are subsp. tortilis showing long hair on the pods similar to that of var. crinita, and thus possibly an additional variety of subsp. tortilis.

6. A. tortilis subsp. heteracantha (Burch.) Brenan (1957)

   Synonyms:-

   A. heteracantha Burch. (1822)

   A. litakunensis Burch. (1824)

   A. spirocarpoides Engl. (1888)

   A. maras Engl. (1888)

   Tree up to 15 m high, with flattened spreading crown. Young branchlets pubescent. Petioles and rhachides of leaves pubescent. Pods 4–9 mm wide, glabrous or almost so, and eglandular.

   The subsp. heteracantha occurs in southern Angola, Namibia/S.W.Africa, Botswana, Zimbabwe, Mozambique, South Africa and Swaziland.

   Ross (1979) refers to the occurrence of occasional intermediates between subsp. heteracantha and subsp. spirocarpa, but remarks that the vast majority of specimens can be placed without difficulty in one or the other subspecies.

5.6 A. tortilis: Further work needed

The wide concept of A. tortilis (p.37) has been generally accepted, except as stated, for subsp. raddiana. Further information is needed about the occurrence and status of subsp. raddiana var. pubescens in particular, and of the frequency and taxonomic nature of the intermediates between tortilis and raddiana noted by various authors (see p. 41). Cultivation would be helpful. In addition, more material and field evidence is needed for var. crinita (see p. 43) before a final decision can be reached about its taxonomic validity. A similar comment applies to the straight podded variant mentioned on p. 43.

5.7 A. tortilis: Literature consulted and cited