428 MISCELLANEOUS NOTES ON VITIS. 
There is generally a hairy (rarely cottony, as in Labrusca and dstivalis) pubescence on the 
under side of young leaves, especially along the ribs; more so in Cord?folia, less in the other species ; 
in the former this pubescence sometimes remains throughout the season, and rarely even verges to 
the arachnoid down of dstivalis. 
In this neighborhood Riparia matures its fruit in July and August, further north in September, 
earlier than even #stivalis ; Cordifolia not before October. 
Now, having distinguished the species, let us see about their geographical distribution. Riparia 
is the northern and western, Cordifolia the southern and eastern form; in the Middle or Central 
States they both occur together. I have found Riparia on the Great Lakes, on Niagara, on Lake 
George, and have it from Vermont; it is common in Missouri and Illinois, and extends to the , 
Rocky Mountains of Wyoming, Colorado, and New Mexico. How far south of the Ohio it is found 
I have now no means of ascertaining. V. cordifolia is common throughout the Middle and South- 
ern States, but I have seen no specimen north of New York, nor west of Missouri. 
I may add that V. @stivalis extends through the whole Vitis region of eastern North 
America, from New England to Texas, and from the Atlantic to the great plains, but not to [234] 
the mountains beyond. V. Zabrusca is our most local species, being confined to the Alleghany 
Mountains and the region between them and the Atlantic, unknown in the Mississippi Valley or 
beyond. Whatever has been called so there, or in Louisiana or Texas, is a large and downy-leaved 
form of dstivalis, always readily distinguished by its “intermittent” tendrils, while Labrusca has 
more or less “ continuous” tendrils. 
Will local botanists assist me to more accurately define the geographical limits of our species of 
Vitis? And may I request them to collect fertile flowers as well as sterile, the only ones found in 
most herbaria ?— [June, 1878.] 
In the June number of last year I have enumerated some of the characters which distin- [310] 
guish Vitis riparia from Cordifolia. I can now confirm all I have said there. Our present 
spring being later than the very precocious one of last year, Riparia bloomed about May 10th and 
Cordifolia begins now, May 27, to open. 
Another, and a very valuable, character to distinguish Riparia not only from Cordifolia, but 
from all other species of Vitis, has been indicated by Prof. Millardet of Bordeaux, and is fully 
confirmed by my observations made on specimens from all parts of their geographical area. [511] 
The dissepiments or diaphragms, as they are called, which at each node interrupt the 
medullary tissue, and which are best studied in vines of the previous year, are in Riparia very thin, 
only 4 to } of a line in thickness, while in Cordifolia they are 3-1 line thick, and in Astivalis a 
little thicker yet. . 
Pursuing these investigations through all the species of Vitis attainable, I find that the Rocky 
Mountain Vitis and that from Lake Superior have been correctly referred to Riparia, while V. Ari- 
zonica, about the relationship of which I had some doubts, is certainly distinct from Riparia. 
All true Vitis have such diaphragms at each node, while all the species of Ampelopsis and of 
Cissus are destitute of them. But the startling fact appears that V. vulpina of the South in this 
character is different from all other Vitis species and affiliates with Cissus, its pith being continuous 
and not interrupted. 
. cordifolia, thus completely separated from Riparia, approaches, strange as it may seem, close 
to Hstivalis in the character of the diaphragm, in its period of flowering, and even in its seeds, and 
the connection seems to be made by that western entire-leaved and small and black-fruited form of 
4istivalis, which I have distinguished as Cinerea, to which downy-leaved forms of Cordifolia 
approach almost too closely. 2 
All the species of Vitis contain in their foliage more or less of a fragrant principle, most proba 
