194 JOURNAL OF THE ARNOLD ARBORETUM [vol. iv 



longis utrinque circiter 6-12 serrulata, 1.3-4 cm. longa et 0.5-1.5 cm. lata, 

 superne saturate viridia, subtus pallidiora, pruinosa, margines cellularum 

 epidermidis utriusque faciei undulatae (subtus distinctius) ; folia ramulos 

 novellos fulcientia surculorumque saepius majora, ovato- vel obovato- 

 elliptica, minus dense sed interdum grossius serrata, ad 4 : 2.1 cm. magna 

 vel versus apicem surculorum minora sed satis latiora circiter 2 : 1.8 cm. 

 magna, subintegra. Inflorescentiae pleraeque breviter racemosae, apicem 

 ramulorum versus fasciculato-racemosae, densiflorae, circiter 5-16-florae, 

 1-3 cm. longae, nutantes; pedicelli 5-8 mm. longi, graciles, apice leviter 

 incrassatae bracteis lanceolatis acuminatis 2-3 mm. longis suffulti; flores 

 lutei, 8-10 mm. diametientes ; sepala externa vel prophylla bracteiformia, 

 media ovato-oblonga vel elliptica, interna iis Y$ majora, latiora, obovato- 

 elliptica, 5-5.5 mm. longa; petala obovato-oblonga, 4.5-5 mm. longa, 

 apice distincte inciso-emarginata, basi subunguiculata, glandulis 2 irre- 

 gulariter ovoideis satis magnis paullo distantibus aurantiacis praedita; 

 stamina petalis ]^ breviora, subapiculata; ovarium ellipticum, stigmate 

 lato, ovulis ut videtur saepissime 3, interdum 2 vel 4 subsessilibus. Fructus 

 maturi elliptici, circiter 9-11 mm. longi, 4.5-6 mm. crassi, stylo brevissimo 

 vel indistincto excluso, saturate coccinei, leviter pruinosi; semen plerum- 

 que 1, circ. 7 mm. longum, purpureo-brunneum, minutissime punctulatum. 



Of this form I have seen the author's type preserved at Berlin. Will- 

 denow states that it came from "Siberia" but there is no wild species or 

 variety known from Asia that agrees with it. In 1811, Willdenow himself 

 says: "Ein drei bis vier Fuss hoher Strauch, derbeim ersten Anblick der 

 vorhergehenden Art (B. vulgaris) ahnlich zu sein scheint, aber in der 

 Blattform und im feineren Ansehn der sibirischen Berberitze viel naher 

 verwandt ist." Therefore Willdenow himself indicates that B. emarginata 

 is intermediate between B. vulgaris and B. sibirica, and there can not be 

 the least doubt that it is a hybrid between these two species. The form 

 described by me is almost identical with the one figured by Guimpel. In 

 the Arnold Arboretum I saw plants that came from Highland Park at 

 Rochester, N. Y., under no. 16 as B. sibirica which approach more nearly 

 B. vulgaris. 



K. Koch, Dendr. I. 398 (1869), declares that B. emarginata is "Ohne 

 Zweifel" a cross between B. canadensis and J5. sibirica, but no trace of the 

 first species can be detected in any plant of B. emarginata which I have 

 seen. Regel says of his B. vulgaris var. microphylla: " Videtur forma 

 hybrida hortensis inter B. vulgarem et B. sibiricam" The real B. emar- 

 ginata he confuses with some oriental forms of very different relationship. 



Being a hybrid, somewhat different forms of B. emarginata may be met 

 with in gardens but there seems no need to give them varietal names. I 

 wish, however, to call attention to the following form which looks like a 

 distinct variety or may have had its origin in a cross of B. eviarginata 

 with a form of B. amurensis. 



