142 1". W. C. Areschoug. 



form, according to Neumax, grows close by. Also the outside of the sepals 

 are often süpplied with stipitate glandules. The following form, from Bohuslän, 

 wliicli Mr. Lixdebeug has described and divided into no less than two species, 

 is of greater importance, but according to my opinion is only a subspecies of 

 R. insularis. 



H. *confinis Lindeb. Herb. Rub. Scand. n. 12; R. umbraticus (P. J. 

 Müll.) Lindeb. Herb. Rub. Scand. n. 11 (forma umbrosa); turionibus pilosis, 

 aneulatis, sursum sulcatis vel Iateribus excavatis, aculeis crebris, angulis maxi- 

 mani partein insidentibus, subina^qualibus, validis, e basi valde compressa subu- 

 latis. rectis & curvatis, glandulisque instructis; foliis 5-nato-pedatis, supra satu- 

 rate viridibue, pilosis, crassiusculis, subtus canescentibus vel pallide viridibus, 

 tomento lovi & pilis longioribus obsitis; petiolis aculeis sparsis, falcatis munitis; 

 stipulis subfoliaceis, lanceolatis; foliolis parvis, insequaliter & simpliciter serratis 

 vel subduplicato-serratis, terminali rotundato-ovato vel ovato, ad basin cordato 

 vel rotundato, acuminato; ramis floriferis villosis, aculeis mediocribus, versus 

 apicem accrescentibus, subinsequalibus, rectis & plus vel minus curvatis instruc- 

 tis; foliis ternatis, foliolis simpliciter vel prsecipue versus basin duplicato-serratis, 

 terminali obovato vel ovato, acuto; inflorescentia racemosa, contracta, subsimplici 

 & aphylla, vel composita, ramis inferioribus longe distantibus & axillaribus, 

 1 — ;!-ftoris vel racemos gerentibus; pedicellis sparse glandulosis, glandulis villo 

 obtectis, aculeis tenuibus, subulatis, curvatis munitis; floribus parvis; sepalis 

 extua canescentibus quam pedicelli sgepe vix brevioribus, in fructu reflexis; petalis 

 albis, oblongo-ovatis — late ovalibus, anguste unguiculatis, extus pubescentibus; 

 filamentis albis, stylos virescentes superantibus; germinibus glabris; receptaculo 

 piloso. 



Thie subspecies appears as well in open places, in brushwood, or rocky 

 clefts, as in woodlands (R: umbraticus Lindeb., haud Müller), in the central 

 part of Bohuslän, on the west coast of Sweden, in many places in the parish 

 of Bro. It, however. never grows, as far as I know, on places which are 

 dircctlv exposed to the seawind. 



Although R. confinis is a natne already used by P. J. Müller for 

 ånother species, I yct thought it could be retained as the name of a form, 

 which is a subspecies of a quite different species. It seems to nie that R. 

 confinis sliuuM rightly be considered as a northern form of R. insularis, and 

 ii als., bears appearance of being influenced by unfavourable outward circum- 

 stances. 1t is mucb smaller and weaker than the typical species, has consider- 



