Monographs Details: Eugenia tapacumensis Berg
Authority: Maguire, Bassett. 1969. The botany of the Guayana Highland-part VIII. Mem. New York Bot. Gard. 18: 1-290.
Family:Myrtaceae
Description:Distribution and Ecology - BRITISH G U I A N A , "ad lacum Tapakuma," Schomburgk 1352 (K, lectotype);sandhills, Demerara River, 22 M a y 1945 (fl), Fanshawe 2421 (Forest Dept. 5157)( M I C H ) . S U R I N A M E . Maratakka River, 20 k m S of Saparra Creek, 26 May1965 (fl), Maas (L. B. B. 10884) ( M I C H ) . Suriname River, Jodensavanne-MapaneCreek area, in forest, Feb 1961 (fl), Schulz 8665 (MICH, U ) . BRAZIL. Amapa:Serra do Navio, Rio Amapari, 26 Jul 1961 (fr), Pires et al 50250 ( M I C H ) . Para:Belem, 15 Mar 1951 (fl), Pires 3152 ( M I C H ) , 14 Jan 1952 (fl). Black 34133( M I C H ) , 14 M a y 1954 (imm fr). Black 16134 ( M I C H ) ; estrada Monte Alegre aC.A.N.P., 5 M a y 1953 (fr), Lima 53-1326 ( M I C H ) . Maranhao: Without locahty,Don 101 (BR, type of E. doniana). Piauhi: Without locahty, Gardner 2169 (NY,isotype of E. piauhiensis).

Discussion:

Eugenia tapacumensis a angustifolia Berg, Linnaea 27: 222. 1856.

Eugenia tapccumcnsis p latifolia Berg, Linnaea 27: 222. 1856.

Eugenia ochra Berg, Linnaea 27: 216. 1856.

?Eugenia roraimana Berg, Linnaea 27: 219. 1856.

Eugenia piauhirnsis Berg, Mart. F\. Bras. 14(l): 285. 1

For description see Amshoff (Fl. Suriname 3(2): 139. 1951), or see Field INIus Publ. Bot. 13(4) |F1. Peru]: 739. 1958. Among the glabrous'species of Eugenia with few and small flowers in axihary clusters, and blunt elliptic-ovate leaves withthe mid vein impressed or broadly sulcate above, this is the most readhy recognized. The leaves are coriaceous, veiny, usually arcuate and somewhat folded and no more than 1.5-2 times as long as wide; the pedicels are relatively stout and only 5-8 m m long; the bracteoles are coriaceous, blunt and persistent, connate into a conspicuous involucre beneath the bud; the calyx-lobes are broadly rounded, 1-2 m m long, hardly as long as the hypanthium and much surpassed by the dome formed by the petals in bud; the fruit is subglobose, about 1 cm in diam.

The species-complex including Eugenia tapacumensis is widely distributed in northern and eastern South America. Urban (Bot. Jahrb. 19: 649. 1895) reported the species from Martinique and St. Vincent. It is not reported by Wihiams from Trinidad or Tobago, but according to Amshoff a few collections are known from Panama (Ann. Missouri Bot. Gard. 45: 190. 1958). It ranges from British Guiana (the type-region) to eastern Brazil (at least to Piauhi) and far inland in the Amazon basin. It is probable that other names proposed by Berg, and before him by de Candolle, will eventually prove to be synonyms of E. tapacumensis. One such probable synonym E. martiusiana D C , D C . Prodr. 3: 269. 1828; the type, which I examined at M in 1966, is strikingly similar to E. tapacumensis except that the bracteoles are less strongly connate, and the midvein is rather flat or merely concave-channeled (cf Field Mus. neg. 19966).

Berg did not include Eugenia tapacumensis in the Flora Brasiliensis nor did he provide in that work any useful contrasts between E. tapacumensis and such newly described species as E. piauhiensis and E. doniana. The type of E. doniaita, cited below, is in m y judgment indistinguishable from Guiana specimens of E. tapacumensis. The isotype of E. piauhiensis has the pedicels almost imperceptibly puberulent, a feature noted by Berg, but seems otherwise very like other specimens of E. tapacumensis.

The original material of Eugenia tapacumensis seems to have represented a single gathering, which Berg cited under two numbers, Rich. Schomburgk 1352 (i.e., a angustifolia, supposedly from "silvis ad lacum Tapakuma," seen at Berhn), and Rob. Schomburgk 884 (i.e., ß latifolia, from British Guiana without definite locality, seen at Vienna). A Schomburgk specimen at K e w bears the number 884/1352B, indicating that the two numbers were applied to the same gathering (for an explanation of the numbering system used by the Schomburgk brothers, see the introduction to this paper). The type of E. tapacumensis a angustifolia, no. 1352, has presumably been destroyed; a sheet of the same number at Kew, "ex herb. Berol. 1859," named a angustifolia by Berg himself, is designated as lectotype, and as lectotype of the name E. tapacumensis. In m y judgment this represents an ordinary individual of the species; it is also appropriate that the lectotype be chosen from the series of specimens that provided Berg with the specific epithet, namely those he saw at Berlin. A sheet of no. 884 at W , named E. tapacumensis ß latifolia by Berg (cf also Field Mus. neg. 31621), is evidently the type of ß latifolia

In the protologue of Eugenia roraimana. Berg made the statement "Habitat ad Roraima et ad flumen Rupununi in Guiana Anglica, floret Majo (Rob. Schomburgk, coh. II. no. 944; Rich. Schomburgk, coh. no. 1292)." According to his statement, he saw these "florif, in hb. Berol. et Vindob." Probably the two numbers represent parts of the same gathering; a sheet at K e w is numbered 944/1292 (see above). A sheet at Kew, "ex herb. Berol. 1859," Schomburgk 1292, is named by Berg; it is assumed that Berg saw the original specimens of no. 1292 at Berlin, and that these have been destroyed. The sheet of no. 944 at W is hereby designated lectotype

Amshoff, in the Flora of Panama (Ann. Missouri Bot. Gard. 45: 190. 1958), referred Eugenia roraimana to the synonymy of E. tapacumensis. Perhaps this is justifiable, but the type of E. roraimana is not an ordinary specimen of E. tapacumensis. The leaves are thinner, longer and more pointed than usual, 3-5 cm wide and about three times as long as wide, indeed suggesting those of E. lambertiana {E. schomburgkii) except that the midvein is rather broad, biconvex or channeled, like that of E. tapacumensis. The flowers and inflorescence of E. roraimana are so similar to Guiana specimens of E. tapacumensis that I cannot distinguish them. Additional specimens are needed that the status of E. roraimana may be certainly determined.

In the protologue of Eugenia ochra. Berg cited two collections from Roraima, "Rob. Schomburgk, coh. no. 537; Rich. Schomburgk, coh. no. 810"; of these he stated "v. in hb. Berol. et Vindob." The sheet of no. 537 at W is hereby designated lectotype. A specimen at K e w is numbered 537/810B, and it is assumed that Berg saw no. 810 at Berlin, and that the specimen itself has been destroyed (for an explanation of the numbering system used by the Schomburgk brothers, see the introduction to this paper). I should refer the lectotype, and the above specimen at Kew, to E. tapacumensis. Amshoff has already, in the Flora of Suriname and again in the Flora of Panama, relegated E. ochra to the synonymy of E. tapacumensis, and in this I concur.

Distribution:Guyana South America| Suriname South America| Brazil South America|