Flowers: In fertile cluster 8 to 16; stamens and petals generally 5, sometimes 6; stamens 

 in sterile flowers ascending, straight 1/5' to 1/4', in fertile, short recurved and bent laterally 

 upon a visible disc; ovary medium globose, style short conical, stigma small hemispherical. 



Berries: 2 to 8, very large 1/2' to 1', globular, surface roughish, generally dull black, rarely 

 yellowish bronzy in color as in Scuppernong, and with a very slight prunose bloom and numerous 

 brown dots: drops singly as soon as ripe, juicy, without coloring matter in juice, with pleasant 

 but peculiar musky flavor and odor; skin thick and tough; pulpy. 



Seeds: 2 to 3, rarely 1 or 4, very large, 1/4' to 3/8' long, by 1/5' to 1/4' broad, oval-lenticular 

 when 2, ellipsoidal ^when 1; color olive brown when fresh; beak generally no more than the 

 acute termination of the seed,\)r rarely a very short, rounded point; raphe invisible till the 

 groove ceases, then becomes visible though not much larger than a hair, and passing down the 

 prominent narrow ridge between the ventral depressions; chalaza acute-elliptical, small, in center 

 of back of seed, surrounded by a distinct groove which begins at beak, spreads around chalaza 

 where its margins are w r rinkled, extends to top of seed; depressions long, narrow, deep, straight, 

 except at top where they curve outward from raphe, their outer border being distinctly wrinkled, 

 of lighter brown color than body of seed. 



Plantlet: Cotyledons large, ovate, acute, green or pale purple beneath. 



Viticultural Observations and Remarks 



Germination later than in V. Munsoniana and more vigorous; foliation and inflorescence 

 very late; ripens after all other species except the Cinerea and Cordifolia series, and V. Mun- 

 soniana, in late August and through September in Georgia and Texas. 



The vigor, long life and freedom from disease of this species is unsurpassed in Vitis, never 

 attacked by rot or mildew. Cuttings root with the greatest difficulty, but layers of young wood 

 easily if twisted withe-like at one point when covered. This species is said to make fine wine, 

 of which considerable has been manufactured in Southern Georgia, and in the Carolinas. 



V. rotundifolia has never been found naturally hybridized with any other species than 

 V. Munsoniana, so far as is known to the writer, but it has been artificially hybridized with 

 V. -vim/era by Dr. Peter \Vylie (deceased) and with V. rupestris by Prof. Millardet, and this hybrid 

 in the writer's grounds proves sickly, nonproductive and resembles V. rupestris more than 

 V. rotundifolia. The writer has produced a number of vigorous hybrids of the Scuppernong variety 

 with hybrid varieties of the true bunch-grapes. This fact alone, now well demonstrated, that 

 the most widely separated of our species can be hybridized, is one of vast importance to the 

 careful experimenter, as it enables one to make any combination of the species of this wonderful 

 genus, as fancy or utility may suggest. 



V. rotundifolia is found throughout the Gulf and the Atlantic States, from the Potomac 

 River to Central Florida, through Georgia, Southern Tennessee, Alabama, Mississippi Arkansas, 

 Louisiana, and Eastern Texas to the lower Trinity, Brazos and Colorado Rivers to Austin, and 

 up Red River to Grayson County, Texas. It usually grows in high bottoms where the land is 

 well drained, and in moist shady situations. Through all the above extended region it is found 

 quite uniform, while in Southern Florida V. Munsoniana is equally uniform, characteristic and 

 very abundant. 



The difference of seeds alone of these two forms of Muscadinia is as great, or greater than 

 between any two species in any other series and with the numerous other differences pointed 

 out, certainly entitles the series to two species, as designated. 



106 



