Tas. 8857. 
BERBERIS arTrocarpa. 
Western Szechuan. 
BEerRBERIDACEAE. Tribe BERBERIDEAE. 
Berseris, Linn.; Benth. et Hook. f. Gen. Plant. vol. i. p. 43. 
Berberis atrocarpa, Schneider in Plant. Wils. vol. iii. p. 437; species a 
B. levi, Franch., haud procul distans, ab ea tamen ramulis angulatis et 
suleatis, foliis tenuioribus, ambitu longioribus et pro rata angustioribus, 
margine minus arcte serratis, facillime distinguenda, 
Frutex erectus, 2-metralis; ramuli cinerei, glabri, sulcati, stricti, spinis rigidis 
8-partitis, pallide brunneis, 2-4 cm. longis armati; internodi 2°5-6 cm, 
longi. Folia sempervirentia, apud nodos 3-8-phyllos fasciculata, coriacea, 
glabra, elliptico-lanceolata, 4-8°5 em. longa, 0°5-1°5 cm. lata, apice 
acuta, basi gradatim cuneata, margine dentibus gracillimis sursum versis 
1°5. mm. longis remotius obsita, supra subnitentia atroviridia, subtus 
pallidiora, utrinque levissima; nervi valde obscuri; petiolus 4 mm. 
longus nunc subobsoletus. Flores vernales, ad nodos in fasciculos 6-12- 
floros aggregati, lutei, 8-12 mm. diametientes; pedicelli graciles, teretes, 
glabri, 6-12 mm. longi. Sepala ovata. Petala oblonga, apice 2-fida. 
Fructus oyoideus vel ovalis, primum coeruleo-pruinosus, demum aterrimus, 
nitens, 9 mm. longus, stylo persistente coronatus, auctumno peracto 
maturus.—B. levis, Schneider in Plant. Wils. vol. i. p. 360; Bean, Trees 
& Shrubs Brit. Isles, vol. i. p. 244; non. Franch.—W. J. Bran, 
The plate of the Barbery here figured was prepared 
from a plant raised from seed collected in Western 
Szechuan in November, 1908, at about 5000-6500 feet 
above sea-level, by Mr. E. H. Wilson. A share of this 
seed had been presented to Kew by Professor C. 8. 
Sargent, Arnold Arboretum, and was received under the 
‘Wilson’? number 1284. The first of the resulting 
plants to flower at Kew did so in April, 1914. In his 
original elaboration of the barberies collected by Wilson, 
Mr. C. K. Schneider (Plant. Wils. vol. i. p. 360) 
identified the plant now figured with the species 
described by the late Mr. Franchet as B. levis. Further 
study, however, led Schneider to reconsider this deter- 
mination, and in a later section of the same work 
(vol. iii. p. 437) he made Wilson’s no. 1284 the type of 
a distinct species, B. atrocarpa, basing his differentiation 
JULY-SEPTEMBER, 1920. 
