CAREX AND EPILOBIUM IN THE LINNEAN HERBARIUM 365 



gines fossarum," C. contigua being usually a plant of rather dry 

 situations ; but, no doubt, like G. divulsa, it occurs in both. 



11. C. canescens. Two very different species are thus named 

 in Herb. Linn. The first (under which is written " Norfolc"), 

 annotated by Smith as "divulsa — Good. & Fl. Brit.," is certainly 

 that. The second bears this inscription: — "11 canescens [queried, 

 in pencil — by Smith?] spiculis subrotundis in summo culmo se 

 contingentibus, brevissime pediculatis," and is G. fusca All. (Bux- 

 baumii Wahl.) ; as is an attached sheet, subscribed "Lapp S.," 

 with the addition " canescens vera, J.E.S." The description 

 quoted in Spec. Plant, ed. 1, from Flora Lapponica 332 does not 

 fit C. curta Good, at all ; on the other hand, a tracing kindly sent 

 to me by Mr. Britten from Loeselius, Flora Prussica no. 32, ad pag. 

 117, also cited, and named by Loeselius " Gramen Cyperoides 

 spicis curtis divisis," represents a remarkably remote-flowered and 

 subdistichous curta. As tending further to prove that Linn6 did 

 not really regard curta as his canescens, may be mentioned the 

 reference to Micheli, Gen. 69, t. 23, f. 10, in ed. 2 : — " Carex nemo- 

 rosa, fibrosa radice, caule exquisite triangulari, spica longa divulsa 

 seu interrupta, capitulis solitariis praeterquam ultimo." C. curta 

 is not (or very rarely) a woodland plant, and the above description 

 fits C. divulsa admirably. 



Additional light is afforded by the specimen in Herb. Linn., 

 given as "10 brizoides," and underwritten " Lapp. S." Against it 

 are pencilled the following remarks of Smith: — "curta Good h "; 

 "canescens Herb. Leche. J.E.S." ; "A Herb, Lights. Dr. G." It 

 has narrow leaves and pale-brown glumes, thus somewhat resem- 

 bling the slender Scottish Alpine form of curta which has been 

 named by Kukenthal C. canescens L. var. fallax F. Kurtz; the 

 fruit is not fully formed, but I am inclined to consider it as either 

 C. Gebhardi Hoppe, 1826 (C. Persoonii Sieb., 1842, C. canescens 

 var. alpicola Wahl., C. curta var. brmvirscens Pers.) or C. vitilis 

 Fr. It should be remarked, en passant, that vitilis is not known 

 to be a British plant ; the original specimen from Lochnagar was 

 almost certainly C. helvola Blytt (curta x Lachenalii), the other 

 alleged localities belonging to G. canescens var. robustior Blytt 

 or var. fallax F. Kurtz. This so-called brizoides, evidently 

 mistaken by Linn6 for the true species of central and southern 

 Europe, is so very near curta that he would not have separated it 

 from his canescens, had he understood by canescens what is com- 

 monly supposed. 



Although the matter is by no means free from difficulty, I con- 

 sider that C. divulsa has far better claims to be reckoned as C. cane- 

 scens L. (the name is, indeed, quite descriptive of it) than has C. 

 curta Good. Considering the presence of G. fusca under the same 

 name, and the conflicting testimony of the citations, the best 

 course will be to drop the Linnean name altogether, as too un- 

 certain and ambiguous, and to adopt G. curta Good. 



12. C. flava. This is represented by a well-grown plant, bear- 

 ing three culms. Female spikelets three, approximate, the male 

 spikelet sessile, springing from between the two uppermost ; fruit 



