EUROPE AND NORTH AMERICA. 



27 



4. Apex entire or indistinctly denticulate. 



6". mollitscntn, riibelltim, rnspidatum. 



This, it will be seen, is readily applicable, but it has the dis- 

 advantage of bringing together species not otherwise the least allied. 



LiNDiiERG, in Ofvers. K. Vet. Ak. Forhar.^i,. for 1862, after 

 separating S. macrophyllum as a genus IsocladuSy arranges the 

 rest as follows : — 



1. Homophylla. Plants glossy, shining; hyaline cells quite 



free from spiral fibres. 



S. sericenm, Holleanunt. 



2. Heterophylla. Plants opake ; hyaline cells furnished with 



spiral fibres. 



A. S. cuspidata. 



S. cuspidahon, Lindbergit, recurvum^ fimbriatuniy 

 acitii/olitim, teres, squarrosum. 



B. S. rigida. 



o 



S. rigidjim, Miilleri, Angstromii. 



C. S. subsecunda. 



S. subseamdum, rubelhctn, tcnellum. 



D. S. cymbifolia. 



S. cymbi folium. 



These groups are natural, have several characters in common, 

 and are also adopted by Russow and Milde, except that they alter 

 rigida to trnncata, a more appropriate term, since it is applicable 

 to all three species. 



I have also followed this arrangement, but at Professor Lind- 

 berg's suggestion have reversed the order, so as to commence with 

 what we must consider the most highly developed species of the 

 genus. 



ScHLiEniACKE, in the paper already referred to, proceeds on 

 the same plan, but increases the number of groups to seven, as 

 follows : — 



1. acutifolia. 



6*. rnbellurd, acuti/olium, finibriatum, Wulfiamim. 



2. cuspidata. 



iS". cuspidattcm, laxi/olium, Lindbergii. 



3. squarrosa. 



^. teres, squarrosum. 



4. rigida. 



o 



6*. rigidmn, Mulleri, Angstromii. 



