84 THE AMERICAN ROTA XI ST 



ingiy naturalized from Europe in the eastern United States; 

 and var. lanceolata forma Candida, an indigenous form rang- 

 ing from Quebec to Michigan. Doubtless each of the remain- 

 ing varieties has its white-flowered forma; and it has been a 

 matter of some gratification that I am able to add a third one 

 to the list. In the course of last season's collecting, I found 

 in a wet meadow near Salem a Prunella with pure white 

 flowers which I at first supposed to belong to the var. 

 lanceolata. Closer examination, however, showed that the 

 floral bracts lacked the copiously bristly-ciliate margins that 

 distinguish that variety, and that it should therefore be refer- 

 red to the var. calvescens, which has been reported from 

 British Columbia, Washington and Idaho, but seems hitherto 

 to have escaped notice in Oregon. 



Only two specimens were found, growing in tall grass, 

 associated with Centaurium uinbcllatuin. The smaller one, 

 now before me, is 4 dm. in height, with one pair of leaves at 

 the middle of the stem, three pairs close together near the base, 

 and one pair closely subtending the spike. These leaves are 

 rather smaller than in typical var. calvescens ( stated by Pro- 

 fessor Fernald to be 3-7 cm. long and 1-2.5 cm. wide), as the 

 largest one is only 2.5 cm. long and 8 mm. wide. The bracts 

 are green with pale purple margins, and the calyx is very 

 slightly tinged with purple. 



In order to comply with the requirements of the Inter- 

 national Code for valid publication, I append the following- 

 brief diagnosis: 



Prunella vulgaris L. var. calvescens Fernald, forma alba 

 f. nov. Bracteis viridibus vel marginibus purpurascentibus 

 sparse breviter ciliatis; calycibus purpurascentibus sparse 

 setulosis; corolla alba. Oregon: wet meadow one mile north 

 of Salem, July 10, 11)17, J. C. Nelson, no. 101!) (type in Gray 

 Herb). 



