MEXICAN AND CENTRAL AMERICAN UMBELLIFER^ 1 43 



19. Arracacia nelsoni C. & R. Contr. Nat. Herb. 3: 296. 1895. 



Oaxaca : oak forests, Sierra de San Felipe, altitude 2850 meters, 

 Pringle 5955, December 10, 1S95. Type (no. 4S1) in Nat. 

 Herb. 



20. Arracacia ovata C. & R. Contr. Nat. Herb. 3 : 296. 1895. 

 Type (no. 448) in Nat. Herb. 



21. Arracacia pringlei C. & R. Contr. Nat. Herb. 3: 297. 1895. 

 Type (no. 450) in Nat. Herb. 



22. Arracacia rigida C, & R., sp. nov. 



Stems several from a common root, 1-2 meters high, branching 

 above, glabrous; lower leaves few, on petioles 30-60 cm. long; leaves 

 large, 3 or 4 times ternate ; upper ones once to twice ternate, petiolate 

 or sessile, ovate, rounded at base, acute, regularly and sharply serrate, 

 glabrous on both sides; umbels few, usually terminal; involucre 

 wanting or a single ovate bract with a long acumination ; involucel of 

 several filiform bractlets ; rays and pedicels puberulent ; fruiting 

 peduncle stiff and stout, 12 cm. long (in our specimen); rays stiff, 

 subequal, 2 to 4 cm. long; pedicels 4 mm. long, shorter than fruit; 

 fruit oblong-ovate, 8 mm. long, rounded at base, glabrous; carpel 

 nearly terete in section ; ribs very obtuse ; oil-tubes solitary in the in- 

 tervals ; flowers purplish. 



Hidalgo: by a tumbling brook in a rocky gorge of the Sierra de 

 Pachuca, altitude 2850 meters, Pringle 6953, August 3, 189S. 



The whole plant has a strong odor of musk. This species has 

 much the habit and foliage of A. ovata ^ but with sharper toothed 

 leaves. 



23. Arracacia tolucensis (H.B.K.) Hemsl. Biol. Centr.-Am. Bot. i: 



564. 1880. Plates XII, XIII. 



Lignsticum tolucensis H.B.K. Nov. Gen. et Spec. : 19. //. 422. 1821. 

 Velaa tolucensis DC. Prod. 4 : 231. 1830. 

 Cnidium tolucense Spreng. Syst. i : 888. 1825. 



This species has long been a puzzle both as to genus and species. 

 No material could be found in the recent rich collections from Mexico, 

 while the illustration and description are silent as to the essential gen- 

 eric characters of the group. At the request of Dr. Rose, Mr. 

 Pringle visited the type locality, but found it planted in corn and no 

 specimens could be obtained. Through the kindness of Dr. Karl 

 Schumann photographs of the type of Velcea tolucensis were ob- 

 tained (reproduced herewith), which show that Mr. Hemsley has 

 properly referred the species to Arracacia. As this species is also the 

 type of Velcea that genus must be merged under Arracacia. 

 Proc. Wash. Acad. Sci., December, 1899. 



