538 EXCURSUS ON THE NAMES OF SHRIKES 



for the stability of scientific nomeuclature to have some fixed 

 startiug-point, what one being a matter of less consequence. 

 It is also bigbly desirable that the initial point should be the 

 same for both generic and specific terms. In the case of the 

 latter, the ornithologist cannot go back of 1758, which I 

 strongly recommend to be taken as our starting-point, there 

 being various sound reasons why it is more eligible than 17G6. 

 Ornithologists, and, in fact, naturalists at large, are about 

 equally divided on this special point ; the present tendency in 

 this country is to "take Liunoeus at 1758", and this is the 

 course I adopt in my own writings. Moehring's genera are of 

 course thrown out by this limitation of time. 



Supposing, however, that for generic names we may revert to 

 1751 or even to 1735, either of which dates leaves Moehring in, 

 the question arises whether an author who does not conform to 

 the requirements of a system of nomenclature shall be allowed 

 to impose names to the prejudice of the names of others who do 

 conform to such system. With a saving clause for the genera, 

 but not the species, of Brisson (1760), ornithologists are nearly 

 unanimous that those who are not binomenclators have no 

 rights we are bound to respect in nomenclature. By those who 

 hold to this ruling, Moehring is to be thrown out of court on 

 this count also, as he did not deal at all with specific names. 



Discarding Gollyrio of Moehring, therefore, as untenable upon 

 one or both of the foregoing considerations, we take up the 

 next name in order, which is Lanius L., 1758. Admitting the 

 name as of this date, the question arises, For which one of 

 the modern genera of Shrikes must Lanius be retained? It 

 cannot, of course, be abolished, nor is it yet available in its 

 original acceptation, for Linnaeus made it cover not only all 

 the Shrikes he knew, but also the several other birds of differ- 

 ent families which he wrongly considered to be Shrikes ; it 

 must, therefore, be used in a new and much restricted sense 

 for some one of the several valid genera of the family — for 

 which one, however, cannot be easily decided. Linnaeus speci- 

 fied no type, and he, moreover, made his generic characters 

 comprehensive enough to more than apply to the whole family. 

 Failing any criterion in this and other Linnaean cases, by which 

 the author's meaning can be ascertained with the precision 

 requisite for the purposes of modern classification, or the appli- 

 cability of his generic name be narrowed to the required limits, 

 we are driven to an entirely arbitrary course. This is, to elect 



