Notes

B. D. Burtt’s specimens form part of an associated series; No. 5786, pubescent throughout, has 12–14 pairs of medium to small leaflets of obviously intermediate shape and venation, on a rhachis with variable stipels unmistakably like those of 1, B. spiciformis; the tomentose mostly simple racemes, also recalling B. spiciformis, have flowers mostly with 4 variable often narrow tepals, sometimes inserted on a short staminal tube; No. 5799 has 8–11 pairs of leaflets, smaller but otherwise similar, glabrous or sparsely ciliate at base, on a subglabrous rhachis otherwise like that of No. 5786; the tomentose depauperate racemes are in young bud; both these trees were reported as showing a flaming claret-red flush of young leaves instead of the pale salmon-pink and green flush of the normal population of 5, B. microphylla in which they were scattered; with these occurred No. 5897, typical B. microphylla with up to 46 pairs of small leaflets, and No. 5798 with 30–35 pairs of larger leaflets of which some have irregular bases, probably indicating a back-cross, (sp. 5 × sp. 1) × sp. 5; No. 5773, “resembling B. spiciformis but with unusually small foliage”, could easily be a back-cross (sp. 5 × sp. 1) × sp. 1. Procter 1559, otherwise similar in all respects to B. D. Burtt 5786, has 10–14 pairs of somewhat larger leaflets. Greenway 7389A (originally mixed with 7389B under 7389 but recorded as a different form) has 7–9 pairs of small broad leaflets and is manifestly intermediate in all characters between the associated No. 7387 (B. spiciformis with 2–4 pairs) and No. 7389B with 17–20 pairs of narrow leaflets; the latter has fewer and larger leaflets than more normal B. microphylla from the same district (B. D. Burtt 6452 attaining 31 pairs) and is therefore probably another back-cross, (sp. 5 × sp. 1) × sp. 5. Anderson 933 (Newala escarpment, 23 Sept. 1953) with 11–12 pairs of leaflets, is a rare glabrous intermediate between B. microphylla (Eggeling 6734, same locality) and B. spiciformis (observed here by Hoyle and always collected with glabrous leaves in this area). Other more or less obvious intermediates are: B. D. Burtt 1894 (in BM !, = 1296 in K, partly !) & 1388 & 660 (in Game Dept. 5660); 132 km. Iringa-Mbeya, Procter 2070; Lindi District, Nachingwea, Anderson 1053; Handeni District, Swagilo, Semsei in F.H. 2909; 88 km. Njombe-Songea, Boaler 552.It is important to distinguish the evident manifestations of hybridity from the results of pathological or other growth-stimuli. The former usually show most obviously in differences between leaflets or flowers on the same leaf or inflorescence, or between leaves or inflorescences on different branchlets diverging at the same level; the latter usually show in sudden changes, in the number and size only, of leaflets on regrowth continuing the same branchlet. It is noticeable that many putative hybrid specimens show a surprising constancy in the number of leaflets per leaf throughout one season’s growth, even more so than their associated putative parents.