BOTANY: W. TRELEASE 
31 
had qmte prepared me for an apparently inordinate increase in the 
number of differentiable species in the genus, I was not a Httle surprised 
to find, when casting my results then into classified form, that on an 
average nearly two new named forms appeared for each one already 
admitted to our flora. 
Notwithstanding an intention to limit my investigation to the species 
of continental North America, the temptation to learn the characters 
of the South American species proved irresistible when, at Brussels, I 
examined the specimens in the personal herbarium of von Martins, 
whose collections have done so much to make known the flora of Brazil; 
and it was not long before the genus as a whole engaged my attention, 
though West Indian material was given less care thB,n the other until 
at Dahlem I reached the collection of Professor Urban, who in 1897 had 
published a revision of all of the West Indian Loranthaceae. To my 
keen satisfaction, I then found that for the Antillean region very few 
forms were to be differentiated from those admitted by Urban, confirm- 
ing my judgment that the large increase in our own flora rested upon 
previous neglect of application to them of characters which appear to 
be really differential, rather than to excessive optimism concerning their 
separabiHty. The thorough study of tropical forms by Eichler in his 
revision of Loranthaceae for the Flora Brasiliensis, in 1868, supplemented 
by a reelaboration of available material when Urban monographed the 
West Indian forms, has also prevented an increase in the number of 
South American species at all comparable with that within our own flora, 
though the number of names added is relatively greater than for the 
Antilles. 
The essential characters of Phoradendon in its group of Loranthaceous 
genera are chiefly its axillary spikes of sessile unisexual and monochla- 
mydeous small flowers often sunken in hollows of the frequently swollen 
internodes of the rachis, normally trimerous, with 2-celled longitudinally 
dehiscent anthers. A very few species, like P. cymosum, present the 
phenomenon of a terminal spike corresponding to the 1- or few-flowered 
cyme of the old world Viscum, but in addition to axillary spikes. Except 
in the species taken by Hooker for falcatum, the receptacular cups, which 
range from so shallow as hardly to surround the base of the flower to a 
depth covering a noticeable part of the mature fruit, are essentially even 
on their margin; but in this species the cup is parted so as to present 
sometimes the appearance of a deeply divided calyx. The flowers, 
with a small vestigial nectar-gland and apparently adapted to polli- 
nation by such short tongued insects as flies and small bees, are usually 
yellowish green when expanded, but in P. Brittonianum and some of 
