2 2 BOTANICAL GAZETTE [JULY 



similar, scarcely 2^^ long, oblong", obtuse, the slender pedicels 

 intermingled with setae bearing stipitate glands and a few fine 

 slender hairs : filaments linear, sparsely pubescent to above the 

 middle : styles slender, long exserted : ovary glabrous ; mature 

 akene not seen. 



Probably most nearly allied to E. campanulatiim Nutt., though very dif- 

 ferent from that and blossoming much earlier in the season. The proposed 

 species was found in full bloom June 1 1, igoo, on the stony slopes of Orendo 

 butte, in the Red desert. This is very early in the growing season in this 

 range. E. campanulatiim, in this same range, blossoms late in the season, 

 and is a much larger plant, with a compound inflorescence as well as a num- 

 ber of minor floral differences. The type no. is 7143, collected as above. 



Eriogonum ovALiFOLiUM Nutt.— There is perhaps no species 

 in the genus more variable as to color of flowers than this, if the 

 specimens as named in the herbaria are to be considered the 

 only evidence. Nuttall originally distinguished two species in 



the present aggregate and consistently maintained them both. 



E. ov all folium was applied to the form with ** bright sulphur- 

 yellow " perianth, which occurs in greater or less abundance on 

 the arid table-lands about the ** sources of the Missouri/' Of 

 the allied forms, however, this was the less abundant, so that by 

 this time the mention of E, ovalifolium suggests not the original, 

 but the E. purptireuTft of Nuttall. That this is the case is easily 

 verified by referring to Jour. Acad. Philad. 7 : 50, and to Jour. 

 Acad. Philad. new ser, 1:166, in the latter of which these two 

 species appear under the there proposed genus Encycla. 



Whether the two can be maintained as distinct species may 

 well be questioned, but there are the best of reasons for retain- 

 ing the later one as a variety. Not the color of the perianth 

 alone, though in regions in which both grow I have seen 

 nothing to indicate that they intergrade, but the differences 

 (though slight) as given in the publications cited indicate at 

 least varietal rank, as suggested in DC. Prod. 14 : lo. Accord- 

 ingly the white-flowered form, which almost constantly shows 

 traces of purple or may even become of a deep purple hue 

 (never yellowish ?) , may be designated 



