1913] Fernald € Wiegand,— Luzula campestris 39 
and C. V. Piper. The characters used by Buchenau and others to 
separate L. comosa from L. campestris are the elongate spikes, fron- 
dose bracts, more ciliate bractlets and prophylla, and larger and more 
denticulate perianth segments. But in many specimens from the 
Northwest which are otherwise good L. comosa the spikes are sub- 
globose; and more or less cylindrical spikes are frequently seen in 
L. campestris, var. multiflora, while they are made the basis of L. 
campestris, var. calabra (Ten.) Buch. In many plants otherwise L. 
comosa the bracts are short and slender while in L. campestris, var. 
frigida, as described by Buchenau, we find “ inflorescentia composita, 
saepe a bracteis 1 vel 2 frondosis rigidis superata"; and similar 
frondose bracts occur occasionally in L. campestris, vars. alpina and 
multiflora. The ciliation of the bractlets and prophylla proves to be 
highly variable in both L. comosa and L. campestris without any clear 
line of demarcation between. Extreme specimens of L. comosa do 
indeed have large flowers, but the examination of a large suite of 
specimens shows that in the two so-called species the measurements 
overlap so frequently that no real line can be drawn between them. 
Extreme L. comosa would seem to be simply a stage larger in size of 
flower just as L. campestris, var. multiflora is a stage larger than L. 
campestris, var. pallescens. 
Although the color of the perianth or capsule has frequently been 
considered of taxonomic importance, it is highly variable and often 
seems to be directly modified by the intensity of the light, being brown- 
ish in the more exposed situations and extremely pale in the woods. 
In the more boreal and alpine habitats, however, the color is, as would 
be expected, very intense, usually dark-chestnut to blackish, and this 
tendency, accompanied by a shortening or suppression of the rays, 
distinguishes such plants as L. campestris, vars. alpina (sudetica), 
frigida and congesta which, having fairly well marked geographical 
ranges, are maintained as varieties. 
The size of the seed and the length of the caruncle have sometimes 
been used in separating plants of the campestris series, but after an 
examination of the seeds of the plants said to have pronounced differ- 
ences in these characters it has seemed to the writers that the differ- 
ences are slight and apparently not constant. The varieties with 
the smallest flowers, L. campestris, var. pallescens for instance, natur- 
ally have their seeds smaller than do the large-flowered plants, but the 
1 Piper, Contrib. U. S. Nat. Herb. xi. 186 (1906). 
