222 The Ottawa Naturalist. 



common species in central and western Ontario, delighting in warm 

 sandy soil. 



A common species in rather damp woods and along old woodland 

 roads is R. abortivm which might be taken for the above but it is quite 

 smooth, more branching and has inconspicuous flowers. This has a 

 var. viicranthus i Gray which may be found in our limits. It may 

 be distinguished from the species by being more or less hairy, having a 

 glabrous receptacle, or having some or most of its radical leaves three- 

 parted. 



An annual species R. sceleratus, L. closely related to R. 

 abortivus but with dissected leaves and succulent stems is a common 

 species in boggy places or in the mud of ditches in many parts of the 

 province but more especially west of Kingston. It has been found 

 at Borthwick's Springs in the vicinity of Ottawa. 



Another woodland species R. recurvatus, Poir. has no relatives 

 on this side of the continent and being found in all rich woods is a 

 common species. Easily distinguished by its reflexed sepals and petals, 

 and in fruit by its round head and the long recurved beaks of the carpels. 



Following this are two introduced species R. actis P. and R. 

 bulbosus, L. The former is very common by roadsides and in old damp 

 pastures while the latter is either very rare or seldom distinguished from 

 acris. Only two characters are necessary to distinguish these species. 

 The latter has a globose, solid, bulbous base or corm, the former has not 

 this base ; in the former the sepals are merely spreading, in the latter 

 they are reflexed. 



R. Pennsylvanicus L. is common in boggy places amongst weeds 

 and grass. It is seldom over a foot high but is stout, and branching 

 and has small flowers with reflexed calyx lobes and an oblong or almost 

 cylindrical head. 



Now follows a group of live species that require careful examination 

 in the field, and good fruiting specimens for the herbarium. When 

 Part I of my Catalogue was published, we had little information regard- 

 ing them, but now they arc easily separated. R. repens P., remains as 

 I had it, and my var. hispidus becomes R. Macounii, Hritton., but is 

 still retained in Cray's Manual as R. hispidus, Hook, (page 43.) 



