NOTE ON COFFEA ENGLERI 199 



Matopo Hills, Miss Gibbs, 67» do not belong to that species, but to 

 T. pachy stigma K. Schum. The error arose from specimens of the 

 two species being mixed up together in the Kew Herbarium, and 

 my inadvertence, due partly to the special appropriateness of the 

 trivial to the Ehodesian plant, in naming my specimens as I did. 

 In my recent note, then, I was really comparing C. Engleri with 

 T. pachystigma, and I confess myself still unable to see any dis- 

 tinction between them except in two points. One of these relates 

 to the calyx, which, as Dr. Krause tells us, is entire in the case of 

 Cm Engleri, whereas in T. pachystigma, as in T. jasminiflora, ex- 

 pansion of the bud causes the calyx to split down on either side, 

 although in early states of flowering the calyx is entire. As regards 

 the second point, namely, the ovules, I would remark that close to 

 the top of each ovary- cell of T. pachystigma is attached a large 

 fleshy placenta upon which, near its middle, are inserted two small 

 collateral ovules, the placenta projecting beyond them in the form 

 of a tongue. In each of the ovarian cells of C. Engleri Dr. Krause 

 finds one ovule "attached to the common septem," which is just 

 what one would conclude is the case in 1\ pachystigma, unless one 

 discovered the two small ovular knobs projecting from the placenta. 

 Of course I am not in a position to say that Dr. Krause has over- 

 looked these ovules ; but seeing that we have to do with two plants 

 growing in the same locality, a locality carefully explored by 

 museum correspondents, the two, calyx apart, indistinguishable 

 except when their ovaries are dissected ; seeing, too, that one of 

 these plants is said to belong to a genus hitherto unknown from 

 that part of Africa, and that, as the oversights of skilled botanists 

 have shown, it is often an easy matter to mistake a mass of closely 

 pressed ovules or a number of ovules more or less immersed in a 

 prominent placenta for a single ovule, it is a matter for regret 

 that, as Dr. Krause admits, his material should be in a somewhat 

 immature condition. I would remark further that T. pachystigma 

 has a double calyculus unsplit during flowering ; it is to be pre- 

 sumed that what Dr. Krause regards as a calyx is not the upper- 

 most of these whorls, the calyx having been already shed. 



Before dismissing the matter, one would like to know whether 

 T. pachystigma is in the collection made at Bulawayo by Professor 

 Engler. If it be, and dissection reveals its difference from C. 

 Engleri, then I must concede the point at issue, and conclude that 

 we have to do with a remarkable case of homoplastic resemblance. 

 If it be not, I venture to think there is only one inference to be 

 drawn from its absence. 



As regards the plant I supposed to be Oldenlandia Wiedenmannii 

 K. Schum., although naturalists of repute have been known to give 

 two names to identical species, I naturally saw the difficulty Dr. 

 Krause mentions, and not that only, but a further one, seeing that 

 O. Wiedenmannii was mentioned in Dr. Krause's memoir, where it 

 is compared with another new species. Ex descriptione identifica- 

 tions must always be to some extent unsatisfactory ; and I am 

 much obliged to Dr. Krause for so promptly responding to my 

 enquiry about the proper naming of Kassner, 653. The name, it 



