170 SEVENTEENTH REPORT. 



Differs from the species in having but one leaf upon the stem. Far- 

 well No. 3629, May 15, 1914, near Rochester, Michigan. 



Apoc)nium Farwellii E. L. Greene, Pittonia. 



A pubescent species that I had collected at Detroit July 7, 1893, and 

 distributed as A. puhescens under the No. 1263a. It is quite common 

 about Detroit. 



Closely related to A. hypericifolium but has differently shaped leaves. 



Apocynum Farwellii. E. L. Greene f . verticillare. N. Form. 



Differs from the species in having the leaves in whorls of threes in- 

 stead of opposite. Farwell No. 3684 June 20, 1914. 



Apocynum Farwellii E. L. Greene var. glaucum, N. Var. 



Differs from the species in having the entire plant glaucous and 

 glabrous. In moist ground on thinly wooded hillsides at Rochester, 

 Farw^ell No. 3815, August 9, 1914. 



Apocynum Farwellii. E. L. Greene var. glaucum f . ternarium, N. Form. 

 Differs from the var. glaucum in having the leaves in threes instead of 

 opposite. Rich, moist grounds on thinly w^ooded hillsides at Rochester. 

 Farwell No. 3724 July 19, 1914. 



Apocj^um Farwellii. E. L. Greene var. glaucum, f. anomalum, N. 

 Form. 



Differs from the var. glaucum in having some of the leaves alternate, 

 some opposite, and some verticillate or subverticillate. Moist grounds 

 on thinly wooded hillsides at Rochester. Farw^ell No. 3803, July 30, 

 1914. 



Apocynum Milleri, Britton var. pauciflorum N. Var. 



Plant, low, six or eight inches in height, bushy, the cj'mes mostly 

 reduced to one or two flowers. Hills near Rochester. Farwell No. 

 3725§, July 19, 1914. 



Eupatorium urticaefolium, Rich. var. trifolium, N. Var. 



Differs from the species in having the leaves in threes instead of 

 opposite. Occasional Avith the ordinary form. FarAvell No. 3843, 

 Aug. 23, 1914. 



Lacinaria scariosa (Lin.) Hill var. trilisioides, N. Var. 



I collected at Rochester, Mich. (3838|) on August 23, 1914, a plant 

 that wnll answer very closely to the variety sphaeroidea (Mx.) Farwell, 

 but it differs in some essentials. Michaux described his Liatris sphae- 

 roidea as with sti]:)itate heads, ])ut in this plant the heads terminate 

 foliolose peduncles two inches long, as in the specific tj'pe but the heads 

 are several times as numerous as in that, making a rather close spike; 

 the plant agrees with Michaux's description in all other respects. What 

 I chiefly wish to bring to your notice, however, is the fact that the 

 plant has a distinct vanilla-like odor as in Trilisa odoratissima Cass, 

 of the southern states, but not so pronounced. So far as I am aware 



