ORCHIDACE.E 



distinguished by fragrant flowers. Fragrance is not uncommon H. dilatata 

 in Habenaria dilatata. It has been detected frequently enough 

 to be regarded an attribute of the species. I have detected it in 

 all the material I have collected, and in the plants from Mt. Al- 

 bert, Quebec, which resemble the type, fragrance was noted by 

 Collins and Fernald. 



Platanthera graminea Lindl., which Rydberg reinstated in his 

 genus Limnorchis, under the specific name graminifolia, was de- 

 scribed originally by Lindley from Western specimens collected 

 by INIenzies. I examined Lindley's type with extreme care in 

 1905, and compared it with a rich series of dilatata forms from 

 my herbarium. The results of my examination absolutely con- 

 vince me that P. gi'aviinea is nothing more than a slender, nar- 

 row-leaved condition of Habenaria dilatata. 



In 0?xhidacearwn Genera et Species Dr. Kranzlin reduces 

 Habenaria dilatata to a variety of H. hyperborea. This treat- 

 ment of the subject, while it may appear very radical to Ameri- 

 can botanists, is more comprehensible than the course which 

 Rydberg has pursued in his monograph of the genus Limnorchis. 

 These two authors stand at the extremes in their views of the 

 systematic arrangement of the Dolichostachyas. Kranzlin re- 

 duces species because he finds it a fruitless task to try to dif- 

 ferentiate them; Rydberg multiplies species, using characters 

 which Kranzlin finds inconstant. In my opinion both of these 

 writers have gone too far, yet, notwithstanding an extended 

 study of numerous specimens, many of them constituting the 

 types of the species examined, I feel that finality is hardly to be 

 expected where the Dolichostachyas group is concerned until 

 material from numerous localities has been observed under cul- 

 tivation, or at least while fresh and equally developed. Stout- 

 ness and thinness of spur, comparative length of spurs and lips, 



[61 ] 



