315 



intermediate place between those of F. spatulata and F. canescens, by 

 their form, as well as by the form, number and size of their heads, 

 and by the form of the scales and length of the bracts ; but they are 

 frequently as cottony as those of F. canescens, and the points of the 

 scales are straightish. 



" Filago canescens (Jord.) has whitish tomentum, and the points of 

 the scales pale not red. Its leaves are remarkable by their wavy 

 margin, usually much revolute ; they are oblong or lanceolate, like 

 those of F. lutescens, but always acute. The branches are little 

 spreading, repeatedly dichotomous, almost as in F. lutescens. The 

 clusters are usually very round, very cottony, and furnished with 

 short bracts; they are composed of numerous heads, straightish, with 

 inconspicuous angles, and with nearly straight scales, very distinct 

 from those of F. spatulata. 



" Filago eriocephala (Guss.) is remarkable by the abundance of 

 grayish cottony down, which covers all parts and assumes often a yel- 

 lowish green colour at the summit of the plant. The leaves are very 

 numerous and imbricated ; their margins are less revolute and less un- 

 dulating than in F. canescens ; their form differs little from that of the 

 leaves of F. lutescens. The stems differ in habit from those of these 

 two species, slightly ascending from their base, only once or rarely 

 twice dichotomous at the summit, with more curved branches. The 

 clusters are very round or subellipsoidal, and consist of heads straighter 

 and more numerous than in the other three species. The achenia are 

 remarkably small ; beiDg half the size of those of F. canescens. This 

 last character is decisive, since in the other three species, which are 

 otherwise very distinct, the achenia do not present such appreciable 

 differences of size." 



Such is Jordan's account of these four alleged species. The first, 

 F. spatulata, he shows to be identical with F. Jussiaei of Cosson and 

 Germain ; and he retains the name of F. spatulata (Presl) in accordance 

 with the recognized rule of priority. For a similar reason Jordan's own 

 name of F. lutescens, given to the second species, must yield before 

 the earlier one of the Rev. G. E. Smith, whose F. apiculata is un- 

 doubtedly the same thing ; although it seems probable that Mr. Smith 

 may have had some specimens of F. spatulata in view while drawing 

 up his character for F. apiculata, and may have thus laid stress upon 

 the " spathulate" leaves of the latter, while that character more truly 

 belongs to the former, as is well indicated by the name. 



The third species, F. canescens, apparently intends the plant which 

 ordinarily represents the Linnean F. germanica in herbaria. But in 



