1906] Hill, — Perianth of Rynchospora capillacea var. leviseta 187 



^ound one of the smooth bristled forms among the barbed at Browns- 

 town naturally led to expect the reverse^ which may yet be the case. 

 The plants kept for the herbarium by no means comprise all cases 

 of examination, since I have been frequently over ground where some 

 of them grow and have tested them for the purpose of discerning any 

 change in character. I have also collected or examined the barbellate 

 form from three localities in Michigan, all contiguous to Lake Michi- 

 gan; Peteskey and Brownstown in 1878, and Benton Harbor, 1895. 

 Both the typical form and the variety occur in other parts of the state, 

 as recorded in Beal's Michigan Flora. ' 



What now do these collections show with regard to the lack of 

 perianth or its reduction. There are generally from 2 to 6 akenes in a 

 spikelet, mostly 4 or 5. I have met with but one akene without a 

 perianth, and that in a spikelet where the rest were provided with one. 

 The number of bristles when counted was so uniformly six that I 

 ceased at length counting in all cases, but noted the length in both 

 smooth and barbellate forms. The length is apt to vary in the same 

 akene, the difference being covered by the length of the tubercle, a 

 bristle rarely falling short of its base or exceeding its tip. Nor could 

 I discern any difference in this respect in the two forms. The arrest 

 of development was limited to the barbs. Though I am necessarily 

 confined to my herbarium for the data regarding the number and 

 length of bristles, it does not follow with respect to their absence, 

 since in cases examined and not preserved the lack of a perianth 

 would have been noticed, as it would have removed the crucial test 

 for the variety. 



Perhaps so small a matter hardly deserved so full a treatment, but 

 it has a taxonomic value on the constancy of a varietal character, 

 more so than many by which species are delimited. The specimens 

 taken in different years from the same locality and from the various 

 localities themselves prove this for the plants as they occur about the 

 head of Lake Michigan. For the thirty years it has been under 

 observation it comes true to seed, whatever may be its variations else- 

 where. 



Chicago, III. 



