1918] Variation in Cerodonta 65 



dark also. So it seemed completely demonstrated that in late 

 fall the eastern specimens may become as dark as the western — 

 undoubted lineal descendants of the pale midsummer broods, 

 among which dark forms never occur. 



In the summer of 1916 I continued sweeping more actively 

 than before, but not very late in the fall. The only additional 

 observations of any significance on this matter were the collec- 

 tion of one somewhat dark specimen at Aberdeen, S. D., on 

 May 29th, and of several dark ones mixed with a much larger 

 number of light ones at Sioux City, la., on May 23d and 26th 

 (Ainslie). This showed that the first spring brood is also 

 affected by the tendency of cold to produce dark colors. The 

 total number of specimens examined in the season was 513, 

 although the number of sweepings was much larger than in the 

 preceding year, indicating that the species was much less 

 abundant in 1916. Except as noted already, all Indiana 

 specimens were decidedly of the pale form, except a single one 

 taken May 10, 1915, which was intermediate. 



In 1916 I received from T. D. Urbahns, then located at 

 Pasadena, two pale specimens taken at Yuma, Ariz., which 

 were the first of this form that I had ever seen from the region 

 in or beyond the Rockies. 



In the summer of 1917 I was enabled to extend my observa- 

 tions into the Southwest in June, making stops for collection at 

 Marfa, Texas; Las Cruces, N. M.; Tucson, Tempe and Yuma, 

 Ariz. Sweepings at all of these places gave the pale form only, 

 and the prevailing tempertaure seemed an ample explanation 

 of the phenomenon. Continuing my trip, I collected at San 

 Diego, Cal., in late June, and in July at Santa Barbara, Berke- 

 ley, Palo Alto, Martinez, and Fallen Leaf, in California; in 

 Utah at Salt Lake City and in Emigration Canyon at an eleva- 

 tion of 7000 feet ; and in Colorado at Tennessee Pass, elevation 

 10,290 feet. At all of these places I collected only the dark 

 form, and it was abundant wherever fresh grass occurred. 



Evidently dorsalis is no more than a pale variety of fem- 

 oralis; but when we turn to the European literature we find a 

 whole series of names that have been proposed on color char- 

 acters that are mostly the same as the ones just discussed in our 

 species. Hence there is some doubt as to whether femoralis is 

 not itself a variety of denticornis, an older name ; and also as to 



