ROSE—MEXICAN AND CENTRAL AMERICAN PLANTS. 53 
CORNACEAE. 
. THE MEXICAN SPECIES OF CORNUS. 
Hemsley cites five species of Cornus from Mexico. Of these the 
two United States species, dorida and stricta, are probably to be 
excluded.  C. ercelsa and C. tolucens/s are very similar to each other 
and difficult to distinguish. Cornus disciflora embraces two species, 
one with leaves very pale beneath and with appressed hairs, answering 
to (. grandis, and a second with woolly pubescence, which may or 
may not be C. désc{fora, [have also distinguished below a beautiful 
species of the C. florida type. A species from north Mexico collected 
by E. W. Nelson I have also described as new. The following key is 
simply suggestive: 
KEY TO SPECIES. 
Flowers in dense head-like clusters. 
Bracts of involucre very large, petal-like .....-..+---+++++-++-+---- 1. CL urbiniana. 
Braets of involucre small. 
Pubescence on under surface of leaves woolly... .2202----200225--: 2. C. disciflora. 
Pubescence on under surface of leaves not woolly ....-.----+-+-+---+-- 3. C. grandis. 
Flowers in more or less open cymes. 
Leaf margins undulate; lateral veins 5 or 6 pairs....----+-+-+-++-++---- 4. CL nelsoni. 
Leaf margins entire; lateral veins 2 or 4 pairs. 
Pubescence on under surface of leaves soft and somewhat spreading and woolly, 
5. C. excelsa. 
Pubescence on under surface of leaves harsh and closely appressed. 
Leaves narrowly lanceolite......220----2--222--22 222 eee eee 6. C. tolucensis. 
Leaves broadly lanceolate ......22+-00++--- eee eee eee eee ee 7. C. lanceolata. 
1, Cornus urbiniana Rose, sp. nov. 
Branches glabrous; leaves lanceolate, narrowed at base, acuminate, densely strigose 
on both sides, sometimes almost lanate beneath; involucral bracts pinkish, 5 cm. 
long, 2 cm. or more broad, tapering at base. 
Collected by Dr. Manuel Urbina on Cerro de San Cristobal near Orizaba, April, 
1891. Type specimen in National Herbarium, duplicate in herbarium Museo 
Nacional, City of Mexico. 
This species is of the forida type, but has much larger and comparatively narrow 
bracts, ete. Here probably belongs the Cornus florida referred to by Mr. Hemsley 
in the Biologia. 
This beautiful species is named for my good friend Dr. Urbina, acting director of 
the National Museum of Mexico, by whom it was collected, and to whom Iam under 
many obligations. 
2. Cornus disciflora DC. Prod. 4: 273. 1830. 
I have not been able to distinguish by the descriptions C. disciflora from C. grandis, 
Both were described in 1830, and so far I have not been able to make out which 
name should be given the preference. While it is very likely that the two descrip- 
tions refer to the same species, as Mr. Hemsley believed, yet I find two species in 
our so-called C. disciflora, the name usually adopted for this group. C. grandis has 
been fully characterized, and this name can be satisfactorily given to one of these 
species, and will be retained for it in case C. disciflora is not an older name or in case 
it belongs to a different species. (. disciflora is not so easily made out, as it is based 
