64 Rhodora [March 



Barnes and Heald's Keys, in the fact that several forms tliere 

 treated as varieties have been elevated to specific rank. In the 

 treatment of species I have differed in the following points: S. suh- 

 bicolor Hampe (S. iutrrmedium Russ., S. centraJe C. Jens.) 1 have 

 not been able to separate satisfactorily from the forms of S. cymbi- 

 folium. It is placed between tiiis species and S. papillosum, having 

 walls of hyaline cells of branch leaves smooth as in 5. n/mbifolium, 

 but differing slightly in shape and ])osition of chlorophyllose cells, a 

 matter however of considerable variability in this group as Roll has 

 pointed out. I have seen New England specimens referred to it. S. 

 parvi folium (Sendt.) Warnst. I have retained in its former plac>e as a 

 variety of S. recurvum. S. Torret/aninn Sull. and its variety Mique- 

 lonense I have in the same way retained with S. nt.spidafum. With 

 reference to the varieties which have been named in such numbers in 

 this genus I have included such as appear to show a distinct and read- 

 ily recognized variational tendency, upon which various authors show 

 general agreement. In so doing I have omitted varieties based upon 

 the presence of a greater or less amount of coloring ])igment, relative 

 slenderness or compactness of the plant, direction of the branches, 

 and other characters possibly explainable as due to immediate envi- 

 ronment of the plants, calling for an arbitrary delimitation, and of ques- 

 tionable systematic value. In case a species has been exhausted by 

 varieties so that no form ap])cars to renuiin under the head of the 

 species, as in S. imbricaium, I liave omitted the variety corres])onding 

 with the original conception of the species, in this case var. cristatum 

 Warnst. In the same way I have included var. viiirronntum Russ. 

 with its species S. recurvum. 



If the list emphasizes the fact that the New England species, or for 

 that matter the American species of Sphagnum, are in need of exten- 

 sive collection and study, it will have served its purpose. A very little 

 collecting in Vermont should serve to fill many gaps in the list, while 

 the same is true in hardly less degree of the other states. Especially 

 do the questionable aquatic forms need particular attention. The 

 status of the species, especially in the subsecunda group, can hardly 

 be regarded as settled, and also demands the attention of the collector. 



In addition to the species included in the list the following Euro- 

 pean forms may be looked for in New England with a pros{)ect of 

 success; S. Ang.sfroemii Hartm., a very distinct species of high lati- 

 tudes, may possibly be found on the mountains of the northern states. 

 It has recently been collected in America, a specimen kindly com- 



