522 NOTKS. 



in Jerdou's own name agricolus, as should demand its rejection 

 in our scientific dog Latin. Linne* himself never exclusively 

 adhered to strictly classically constructed specific names, e.g. 

 " arauna," <l baltimore," " lory" (can any thing be more r^raclassi- 

 cal than this last ending in ay?), and on what grounds therefore 

 are we to insist that Jerdon's adjective " agricolus" must be re- 

 jected as unclassical and changed to the substantive " agricola ?" 



In the second place, I don't think that the editors have quite 

 hit the right nail on the head when the}* - suggest that the 

 non-agreement of genders in the instances cited by them is 

 due to the specific name being a substantive. 



By no possible construction can hypoleucus be a substantive, 

 but Linne gives us Tringa (fem.) hypoleucos. 



At first it may be suggested that he did not decline words 

 derived from the Greek ; but this is not so, he invariably did, 

 e.g., Bradypus tridactylus, Myrmecophaga tridactyla ; Falco 

 leucocephalus, Columba leucocephala, Fringilla erythropthalma, 

 and many others. 



And the instance above given is by no means a solitary one, 

 take, for instance, Motacilla schoenobanus or Motacilla phmnicu- 

 rns, in neither of which cases can the specific name, it seems 

 to me, be construed as other than an adjective. 



Long ago I used to puzzle over this, but I soon found out 

 that in almost every instance of such non-agreement in the 

 whole Systema Natura, specific names not agreeing in gender 

 with the generic name commenced with a capital. 



The few exceptions are all, I think, save one, obvious 

 misprints. I forget the others now, but one I remember is 

 Fringilla bengalus, which, of course, ought to have commenced 

 with a capital. Looking closer I found that in all cases 

 these specific names commencing with a capital were, whether 

 adjectives or substantives, existing appellations, appropriated by 

 Linne from the ancients, or his ornithological predecessors, or 

 lastly from the vulgar tongues. Thus the non-agreement of 

 gender has nothing to do with the specific name being a sub- 

 stantive, but merely indicates that it was some one else's name 

 that Linne, more scrupulous than many of his successors, 

 did not feel himself justified in altering. 



I in no way dispute the editors' dictum that many specific 

 names are substantives originally ; I only contend that all are 

 used as adjectives, and that Linne himself either had recourse 

 to an existing name, which he then reproduced intact, or used 

 adjectives which he declined. 



As to Gracnlus eremitus, that is due to a final correction of 

 my printer's classical reader ; had / intended any change I 

 should have written eremitis; the name eremita is printed 



