14 JOURNAL OF THE ARNOLD ARBORETUM /ol. v 



2. Leaves sparsely hairy when mature, deeply lobed on yo* .g but 

 shallowly lobed on mature trees. 



3. Leaves always shallowly lobed and densely hairy. 



4. Leaves always deeply lobed and densely hairy. 



5. Leaves shallowly lobed and sparsely hairy on young, deeply lobed 

 and densely hairy on mature trees. 



By Siebold (Florae Japonicae Familiae naturales, n. 410 [1846] ) the 

 species was described as with "foliis alternis longe petiolatis e basi sub- 

 cordata vel truncata orbicularibus profunde palmato-quinquefidis, lobis 

 oblongis acuminatis argute serratis glabris coriaceis," and further in the 

 detailed notes below he makes the following statement: "lamina basi 

 subcordata, rotundata vel truncata, orbicularis ad medium circiter usque 

 palmato-quinqueloba lobis oblongis acuminatis lateralibus parum brevi- 

 oribus, utrinque glabra subtus dense reticulato-venosa. " 



Thus we readily see that our first and second groups correspond 

 to Siebold 's Panax ricinifolium , though the specimens at Leyden show a 

 variety of forms. Our first and second groups should therefore be named 



var. typicum. 



ifolium var. magnifi 



Acanthopanax acerifolhun Schelle. The fourth and the fifth are Aralia 

 M aximowiczii Van Houtte. The first group is found in Hondo ; the second 

 is distributed through Hondo, Korea, Quelpaert, Manchuria, Ussuri 

 and China; the third is distributed in Yeso, Hondo, Korea and China; 

 the fourth and fifth are in Yeso, Hondo and China. Koehne says, "Blu- 

 hende Zweige mit tief geteilten lila ttern sind noch nicht bekanntgeworden" 

 thus flowers of var. M aximowiczii have not been seen in Germany. But 

 we have a flowering specimen of the variety collected in the province of 

 Shinano and I myself have seen the same type not rarely in the mountains 

 of the province of Kai. Therefore this should be the type of a distinct 

 variety. In the Arnold Arboretum there is growing a tree of this variety 

 which was introduced from France. It is now of good size but has not 

 yet flowered. There are also in this Arboretum three large trees of this 

 species raised from seed collected by Professor Sargent in Yeso. They cor- 

 respond to our third type. Another tree raised from seed collected by 

 Mr. J. G. Jack in Yeso in 1905 is also of the third type, yet it is less 

 hairy. 



The Liukiu type which is just like the figure a of Koehne and as Tafel 

 5, fig. A of Harms has leaves whose main ribs approximately radiate from 

 the base of the leaves, but the leaves are always !)-lobed. In general the 

 leaves of the flowering branches of this species are 5-7 ribbed . The Chinese 

 plants which have very short lobes belong to var. chinense. 



Echinopanax Dene. & Planch. 



Echinopanax Decaisne & Planchon in Rev. Hort., 1854, 105. — Harms 

 in Engler & Prantl, Nat. Pflanzenfam. m. abt. 8, 34 (1894).— Schneider, 

 111. Handb. Laubholzk. n. 429 (1909). 



