58 Proceedings of the Biological Society of Washington. 



later, Biddulph* adopted canadensis, citing it from Shaw 1804 

 and claiming priority for it over cervina which he believed 

 to date from 1818. He obtained the date 1804 not from 

 the title page but from an estimate based upon assumed 

 regularity of issue of the parts of the Naturalists' Miscellany, 

 the work in which the name appeared. He was thus the 

 first modern author to use any of the three names upon 

 the basis of its real date as we now know it. The burden 

 of proof, therefore, rested upon names subsequently brought 

 forward. Apparently in this belief, Merriam in 1890t and 

 L891, + adopted canadensis. In April, 1895, Sherborn § pub- 

 lished a careful collation of the parts of the Naturalists' Mb- 

 cellany in which he more than corroborated the date obtained 

 by Biddulph for canadensis since he made it appear that in all 

 probability it was published in December, 1808. In spite of 

 this, Allen II in June of the same year rejected canadensis in 

 favor of cervina which lie cited from its original source bearing 

 the date 1804 on its title-page. He based this action upon his 

 personal doubt of tin' date December, 1803. He readily ad- 

 mitted that the name must in any case have appeared early in 

 L804, thus making it of even date with cervina, but he chose 

 cervina in preference to canadensis because he regarded a title- 

 page date more reliable than one ascertained from other sources. 

 Dr. Allen's views were followed in some quarters but in others 

 his mere expression of doubt was not accepted as evidence. 

 Those who continued to use canadensis did so on the ground 

 that to the best of their knowledge it was published in Decem- 

 ber, 1803, and therefore antedated 0. cervina, which lacked even 

 a pretension to publication prior to 1804. A title-page date 

 had no sanctity to them for they knew it might be called in 

 question as well as one determined by investigation. That is, 

 Sherborn's determination of 1803 as the date of 0. canadensis 

 stood accepted in the absence of proof to the contrary and in 

 the last analysis no more could be said of the later title-page 

 date of 0. cervina. It was evident, moreover, that a subse- 

 quently discovered error in the collation of Shaw's work could 



* Proc. Zool. Soc. Lond., pp. 681-684, 1885. 



t N. Amer. Fauna Xo. ::, p. 78, Sept, 11, 1890. 



tN. Amor. Fauna. No. 5, p. 81, July, 189i. 



? Ann. & Mai.'. Nat. Hist. (6), XV, |>. 376, 1895. 



|| Bull. Am. Mus. Nat. Hist., VII, p. 258, footnote, June 29, IS',)5. 



