205 



C.flocosus Rydb. riicre seems to he scmic douhl as to just what 

 C. canescens is, nor am I certain just what form is C. oblanceolatus . 

 I have collected at Glenwood Sprinjjjs a form with rather small 

 heads, and small bracts, and having the leaves ahnost entire, 

 which I have thought may be this; but the leaves of C.flocosus 

 are often quite entire and it is possible that the forms I speak of 

 may be referred to that. C. Traceyi is not very well known and 

 it is not likely that it has often been referred to C. undulalus. 



Probably the only one of these which has been generally re- 

 ferred to C. undulatus, is C. flocosus. This is quite widely dis- 

 tributed in middle elevations, and I have collected it not only at 

 the type locality, Wolcott, Eagle Co., but at Hayden in Routt 

 Co., Glenwood Springs in Garfield Co., Delta in Delta Co., and 

 Ouray in Ouray Co. I should say that it differs quite widely 

 from C. undulatus Nutt. Instead of being a foot or two high it is 

 three or even four feet high, and the involucral bracts instead of 

 being comparatively narrow, are comparatively wide, resembling 

 rather those of C. megaceplialus ; and in some of the plants the 

 leaves resemble the latter, and sometimes the heads approach it 

 in size. 



I can not say, of course for a certainty, that Carduus undulatus 

 Nutt. is not found in Colorado, but after a number of years of 

 collecting, I am quite sure I have not found it. 



Geo. E. Osterhout 



Windsor, Colorado 



REVIEWS 



The western g-ulde-books of theU. S. Geolog:ical Survey 



Last year, when many more Easterners than usual were 

 visiting the Pacific coast on account of the two California expo- 

 sitions, several national conventions, and the liberal concessions 

 made by the transcontinental railroads in the way of rates, 

 routes, stopovers, etc., the U. S. Geological Survey published a 

 "Guidebook of the Western United States" in four parts (A to 

 D), constituting Bulletins 611-614 of the Survey. The set com- 

 prises 794 pages, 95 maps, 152 plates, and 113 text-figures. 

 Unlike other bulletins of the same series, which are distributed 



