— 62 



A great part of our present material is provided with larger or smaller 

 portions of the rhizome, which gives me cause to consider it as an 

 irrefutable axiom, that all the species have a quite similar rhizome, naked, 

 thick, fleshy, creeping, tortuous, the younger parts often covering the older 

 ones, in living state green and glaucous, when dried yellow-brown, grey 

 or black (differences probably due to habitat, age or manner of drying), 

 provided with semiglobose or conical protuberances, at length forming a 

 cavernous crust on and finally surrounding the trunk or the branches of 

 trees, inhabited by ants. 



The stipes are placed on the protuberances, at the apex; they are articulated 

 at the base. Fronds oblong or linear-oblong, linear-lanceolate or oblanceolate, 

 deeply pinnatifid, pinnatipartite or pinnate; segments (lobes or leaflets) oblong, 

 linear, ligular-obovate or lanceolate, several to many below an often caudiform 

 terminal one, when barren entire to repando-dentate or repando-subsinuate, 

 when fertile more or less undulated, deeply and obliquely lobed or more 

 shallowly and obliquely repando-sinuate and then bearing on the anterior 

 side of the lobules a smaller or larger lobe- or tooth-like outgrowth, one 

 to each; veins anastomosing irregularly; areolae with free veinlets spreading 

 in various directions. Sori solitary on the ultimate lobules or teeth. 



Diels gives in Engl. & Prantl, Nat. Pfl.fam., fig. 169 for L. carnosa 

 Bl. a picture representing the habit of the plant (being a reproduction of 

 the habit-drawing of Burck's plate in Ann. Btz., IV, tab. Vli) and detail-drawings 

 of a pinna and a sorus (taken from original specimens). This combination 

 of drawings of different artists and Burck's enumeration of synonyms have 

 led to the conclusion that Polypodium patelliferiim Burck should be identic 

 with L. carnosa Bl. This is only partly correct. 



Comparing Burck's plate with that of Blume (Flor. Jav,, II, tab. XCIV), 

 we see that in the former the pinnae are blunt with twisted soriferous lobules, 

 so that the soral cavities are directed with the mouth obliquely towards 

 the margin of the segments; in the latter, however, the pinnae are acuminate 

 with the soriferous lobules not twisted, so that the retroflexed lobules have 

 the mouth directed upwards. We can observe the same differences when 

 comparing Teysmann's Karimata plant, on which Burck partly based his 

 diagnosis of P. patelliferum, with our specimens of the true L. carnosa 

 gathered in Celebes, which absolutely agree with Blume's plate. The only 

 resemblance in the material quoted is, that the soriferous lobules of both 

 forms are retroflexed, i.e. turned back on the upper surface of the pinnae. 



As to Burck's diagnosis and plate I still observe that Burck used for 

 both as well Teysmann's material as some living plants from Western 

 Java, formerly cultivated' in the Buitenzorg Gardens, which might have 

 been specimens of L. carnosa BL, L. pumila Bl. or L. Curtisii Bk. — 

 These three species or forms occur in Westert) Java and Burckjs habit- 



