206 JOURNAL OF THE ARNOLD ARBORETUM [vol. iv 



que ovato-elliptica vel ovato-oblonga acutiora, pleraque grossius distanter 

 serrata, ad 5 : 2.2 cm. magna, petiolo vix 1 cm. longo incluso. Inflores- 

 centiae subspicatae, elongato-racemosae, densiflorae, iis B. nummnlariae 

 similes, 4-0 cm. longae pedunculo nudo ad 1.5 cm. longo incluso; pedicelli 

 fructiferi vix ultra 7 mm. longi, bracteis lanceolatis acuminatis iis 3-plo 

 brevioribus suffulti; flores flavi?, aperti 7-8 mm. diametientes; sepala 

 externa parva prophylliformia ovato-triangularia, media iis multo majora, 

 ovato-elliptica vel obovato-oblonga, circiter 5.5 mm. longa, intcgerrima 

 vel apice leviter emarginata, basi vix contracta, glandulis 2 oblongis 

 mediocribus aurantiacis separatis praedita; stamina petalis fere duplo 

 breviora, apice vix apiculata; ovarium ovato-oblongum, substylare, stig- 

 mate mediocri, ovulis 2 subsessilibus instructum. Fructus elliptico- 

 oblongi vel ovato-elliptici, pallide rubri, circiter 10-11 mm. longi, 5-6 mm. 

 crassi, estylares vel stylo brevissimo coronati; semina 1-2, purpurascentia, 

 circiter 7 mm. longa, minutissime punctulata. 



Unfortunately I have not been able to study this very remarkable 

 hybrid in a living state. It originated in 1888 from seeds of a plant which 

 had been raised in this Arboretum from seeds received in 1874 from the 

 Jardin des Plantes at Paris under the name B. cretica. According to the 

 material preserved in the herbarium this tirst plant was B. crataegina DC, 

 or a form of it. Of B. bella I saw flowering specimens collected June 7, 

 1900, and fruiting ones of October 31, 1898 (no. 62-1). In the shape of 

 the leaves, their bluish green color, the strong spines and the color of the 

 twigs, as well as in its dense inflorescences B. bella much resembles B. nam- 

 mularia var. pyrocarpa Schneid. but the fruits are of about the same size 

 as those of B. vulgaris. We do not at present know enough of B. crataegina 

 and its behaviour in cultivation to make a positive statement as to the 

 origin of B. bella. I am not even sure whether B. vulgaris is one of the 

 parents. There is, however, in the herbarium material of another form 

 raised in 1S98 from seeds of B. bella consisting of flowering branches col- 

 lected May 25, 1906 (no. 62-2). It is almost identical with B. vulgaris 

 bul the leaves are very narrow, In the old shrub-collection of the Ar- 

 boretum B. bella had its place between B. oblonga (no. 671-2) and a form of 

 B. vulgaris (no. 1417; received in 1880 from Spaeth as B. laxiflora). This 

 no. 62-2 looks to me like a cross between B. bella and no. 1417, but not 

 having seen young shoots and fruits I am unable to make a definite state- 

 ment about it. 



X Berberis durobrivensis, hybr. nov. 



Frutex habitu ignoto, ramuli annotini rubro-brunnei vel novelli olivacei, 

 sulcato-angulati, pruinosi, laeves, minute lenticellosi, biennes subteret- 

 iusculi, vetustiores nigro-cinerei; internodia 1-1.8 cm. longa ;spinae debiles, 

 simplices vel 3-fidae, flavae vel flavo-brunneae, mediae 3-8, vix ad 10 mm. 

 longae. Folia ad 8-fasciculata, inaequalia, matura crasse papyracea, 

 ramulorum fertilium oblanceolata vel anguste obovato-oblonga, apice 



