CHARACTERS OF P. BRASilLIENSIS FRENATUS. 145 



men above mentioned imperfectly represents tiie species. It 

 is, morover, associated in that locality with an animal which 

 turns perfectly white in winter, and is in other respects insep- 

 arable from northern Ermines. Thus, No. 2839, from Fort 

 Crook (Feilner), is pure white underneath, has the head like 

 the back in color, and both of the usual undressed mahogany- 

 brown : yet it shows the white frontal spot and has a decided 

 trace of the malar strij)e. It is accompanied by 'No. 3872, pure 

 white all over. Still, the white markings of No. 2839 may be 

 remains of a seasonal change, or merely like the similar ap- 

 pearances that some specimens of the European P. erminea 

 present. I refer these two specimens to P, erminea^ but in- 

 clude the other from Fort Crook, No. 3830, in the beginning of 

 the hrasiliensis series. I refer to this species, with some hesita- 

 tion, a very young Stoat from Astoria, Oregon (No. 3520, June 

 19, 1858, J. Wayne). Although the head is not darker than 

 the back, and no head-stripes are apparent, the belly shows 

 strongly the characteristic fawn color of frenatus. 



There remains the discussion of the relationships of the 

 South American forms. Although I have not specimens from 

 Brazil or Ecuador, the sufficient descriptions of authors enable 

 me to speak with confidence resi)ecting them. There is evi- 

 dently but one series of linked forms. We have already seen 

 that frenata begins in Upper California, as xantliogenys^ 

 which is merely the northernmost palest form, between which 

 and true frenata (City of Mexico, &c.) there is no difference 

 requiring recognition by name. In Guatemala, frenata already 

 assumes the rich coloration that culminates further south in 

 hrasiliensis. Gray, indeed, who usually subdivides altogether 

 too much, does not attempt to separate frenata and hrasiliensis 

 except varietally. P. cequatorialis (as I call what Gray named 

 M. aureoventris., a term preoccupied to all intents and purposes 

 by auriventer of Hodgson) was originally described from a 

 very young animal ("length 6 inches, tail 4J" — the adult is after- 

 ward described as length 12, tail 8), without facial markings, 

 hut the adult has the auricular blotch, though the frontal spot 

 appears to be extinguished. It is described as very darkly 

 and richly colored, the under parts and ear-spot " golden-yel- 

 low "j the coloration of the plate is almost precisely that of 

 Guatemalan specimens before me. But that the facial mark- 

 ings may be completely extinguished, as a matter of individ- 

 ual variation, is shown by a specimen before me from Costa 

 10 M 



