368 EAST-ASIATIC CORYLACEJE. 



semipollicari glaberrimis merabranaceo-coriaceis supra lucidis subtus in 

 sicco sublucidis rufulis costulis in utroque latere 8-10 tenuibus subtus 

 parum prominulis sub angulo 55^^ egressis curvulis ante marginem 

 deliquescentibus venularum rete tenuissimo utrinque conspicuo, fruc- 

 tibus 3-8 in ranaulo abbreviato aggregatis sessilibus biennibus, invo- 

 lucre subgloboso nionocarpo diametro circ. sesquipollicari cinereo- 

 sericeo aculeis persistentibus distantibus simplicibus v. 2-3 coalitis 

 robustis compressis 1^ lineam diametro 3 lin. altis apice radiis 

 pungentibus 2-3 lin. longis divaricatis coronatis armato indehiscente, 

 nuce ovoidea extus fulvo-sericea involucro arete adhserente, semine 

 demum libero, cotyledonibus conspicue sinuato-lobatis. 



Ad Amherst, prov. Tenasserim, m. Aprili, 1849, coll. beatus Dr. 

 H. Falconer. Specimen accepi ex herb. hort. bot. Calcuttensis, sub 

 num. 463 communicatum. 



I believe this to be most nearly allied to C. tribuloides, A. DC, but 

 it differs in its less coriaceous entire leaves, its much larger fruit, the 

 structure of the distant aculei, and especially in its very sinuate cotyle- 

 dons, a character hitherto not observed in the group. I should, perhaps, 

 rather say not recorded, for Eoxburgh's plate of his Quercus arniata, 

 copied by Wight,*" contains a transverse section of an acorn, vrith, if I do 

 not altogether misunderstand it, a clear representation of such cotyledons. 

 And this naturally leads to a doubt whether M. Alphonse De Candolle 

 has rightly referred both Q. armata and Q. ferox — which Roxburgh, 

 with specimens before him, always kept distinct — to Castanopsin tribic- 

 loides. Wallich's Castanea martabanica^ also from Amherst, I have 

 not seen, but M. De Candolle says it is similar both in fruit and 

 foliage to Castano2)sis argentea, A. DC, to which he reduces it as a 

 slight variety. Falconer's specimen has foliage not unlike that of the 

 Javanese species, but its fruit is very different from that figured in 

 Blame's plate, f and also from immature ones in my herbarium, 

 gathered on Mount Kendong by Van Gorkom ; and it can in no wise be 

 described as having '* spinas dense acervatce," the term used by M. 

 De Candolle : they are more like those in Roxburgh's plate of 

 Q. armata, just referred to. The young nuts of C. tribuloides are 

 smooth and trigonous, and, like those of all Castampses I have had an 

 opportunity of examining, save the two here described, have plano- 

 convex cotyledons. 



15. Castaxopsis {CallcBocarpus) Lamoxtii, sp. nov. — Arbor 40- 

 pedalis coma ampla pra)ditus, ramulis angulatis purpurascentibus 

 glaberrimis, gemmarum squamis interioribus cinereo-velutinis, foliis 

 lanceolatis v. oblongo-lanceolatis basi ciineatis saepe obliquis apice 

 caudato-acuminatis v. obtusiusculis prajter apicem 2-3 dentatum 

 integerrimus 6-8 poll, longis 2;}-4^ poll, latis petiolo -^-l-poUicari 

 glaberrimo coriaceis supra sublucidis subtus subopacis et non raro rore 

 glaucesccnti obductis costulis in utroque latere 9-12 tenuibus sub 

 aagulo 55'* egressis subtus prominulis curvatis ante marginem delique- 

 scentibus venularum rete tenuissimo satis conspicuo, fructibus 12-20 

 secus ramum validum inter se distantibus plerisque hand maturatis 

 sessilibus biennibus, involucro subgloboso subpentagono H-It poll, 

 diametro crassissimo seriebus verticalibus 5 squamarum oblongo-linea- 



♦ Ic. Plant. Ind. Or., iii., . 770. f Fl. Jav. CupuHf., t. 21. 



