This species has been found in only a few of our northern counties in tamarack bogs. Van Gorder found it in a birch marsh in Noble County and Hill reported it from La Porte County. I have had this species in cultivation in the open in garden soil for about 10 years and it thrives.
Stems 3-6 dm from a short, stout rhizome, sparsely hirsutulous; basal lvs to 3 dm, with 3(5) principal lfls, the terminal one broadly obovate or subrotund, serrate, ±3-lobed, the others adjacent, narrowly obovate; additional lateral lfls remote and irregular; cauline lvs much smaller, reduced above, variously toothed to divided; fls several, campanulate, nodding, the pedicels eventually elongate; sep purple, ascending or erect, 7-10 mm; bractlets linear, 3-4 mm; pet yellowish, suffused with purple and purple-veined, usually a little shorter than the sep; style at anthesis jointed near the middle, the distal segment plumose, to 8 mm at maturity; achenes spreading, with decurved styles; 2n=42. Swamps and wet meadows; Nf. and Que. to Alta., s. to N.J., Pa., Ind., Mich., and Calif. May- July. G. pulchrum Fernald, with clear yellow pet conspicuously exceeding the sep, has been thought to be a hybrid with G. macrophyllum, but may be merely a form of G. rivale.
Gleason, Henry A. & Cronquist, Arthur J. 1991. Manual of vascular plants of northeastern United States and adjacent Canada. lxxv + 910 pp.