58 
JOSEPH F. ROCK 
Cyrtandia Oliveri Rock n. name. 
Cyrtandra Hillehrandi Oliver in Hillebr. Fl. Haw. Isl. 331. 1888. Not 
C. Hillehrandii C. B. Clarke. 
Leaves opposite, coriaceous with stout ribs and veins, elliptical or ovate- 
oblong, 7.5-12 cm. long, 3-5 cm. wide, on petioles of 2.5 cm., acute, dentic- 
ulate, pubescent above, faintly ferruginous or glabrate beneath; flowers 
3-5, cymosely umbellate on a peduncle of 12-24 mm., the pedicels as long 
or longer, the bracts ovate-lanceolate, 10-14 mm.; calyx glabrate, thin, 
12-18 mm., cleft beyond the middle (deeper on one side), into broad-lan- 
ceolate, long-acuminate lobes; corolla little exserted, 18-22 mm. long, 
pubescent, somewhat curved, with ampliate throat and spreading lobes; 
ovary glabrous; berry ovoid, elongate, 18 mm. long, enclosed in the calyx. 
Oahu: From Nuuanu to Palolo, Hillebrand, in herb. Berlin* and part 
of type (Nuuanu Valley specimen) in herb. College of Hawaii; Kalihi Valley, 
flower buds, A. S. Hitchcock no. 14105, in U. S. National Herbarium and 
part in herb. College of Hawaii. 
Cyrtandra Oliveri must be classed with Section Schizocalyces, owing to 
the thin, deeply lobed calyx. The specific name Hillehrandi must be 
changed owing to Clarke's C. Hillehrandii which antedates that of Oliver 
in Hillebrand 's Flora. That C. Oliveri, C. Pickeringii, C. honolulensis, 
and C. Hillehrandii are closely related there is no doubt. The latter has 
already been classed as a synonym of C. Pickeringii. The writer has a 
large amount of material at hand but the plants are so variable as to thick- 
ness, shape, pilosity of leaf, etc., that it is next to impossible to determine 
them properly. The species mentioned above are in the same state of 
evolution as are C. platyphylla and its numerous varieties on Hawaii. 
Certain species of Cyrtandra are exceedingly local while others range over 
the mountains and valleys in various forms, the extremes of which one 
would be tempted to describe as new species were not the intermediates 
available. Unfortunately, the writer has not seen the type of Clarke's 
C Hillehrandii, which is in the Kew Herbarium (Hillebrand no. 329). 
Drake Del Castillo cites it as a synonym of C. Pickeringii. Whether he 
has compared it with other specimens or has determined the synonymy on 
the strength of the description is doubtful, though the latter is more prob- 
able because his synonymy of other species, for example that of C. latehrosa — 
C. hawaiiensis — is quite faulty. The description of C. Hillehrandii does, 
however, differ very little from that of C. Pickeringii as understood by 
Hillebrand. The specimen referred by Hillebrand to C. Pickeringii 
A. Gray has nothing to do with that species, the type of which the writer 
has only recently been able to examine, but seems to be C. Hillehrandii 
C. B. Clarke. The true Cyrtandra Pickeringii A. Gray has apparently not 
been re-collected ; at least the writer did not find it in any of the collections 
examined by him either in Europe or America. The type of C. Pickeringii 
A. Gray consists of a single leaf and an undeveloped inflorescence; the latter 
